Dinosaurs and their Descendents 4 - How Feathers Develop

` In my previous Dinosaurs and their Descendents articles (starting here), I've described a bit about how birds fit into the maniraptorian clade of dinosaurs, why people have thought they had feathers, and the species that bear out those predictions.
` Now that we know why dinosaurs would have feathers, it's time to look down their developmental path - or as Ascitu-saurus would say:

... ,--=@ + ......< "How could feathers have evolved?"...
...."'`**") # ............................................................................
.............` &. "..------====_-_-_ - ------==<,=<:<:<:<+=~...
............... zz . >>>? 7) 5 ...................................``````` ``` ``...
.................~$j> ^^/%y ..........................................................
................../# ........:4 .............................................................
.................^M..........9 ...........................................................
........................~^JfZ .............................................................

` I admit I'm not exactly an authority, but what I do know for sure can be described as both fascinating and bizarre nonetheless...

` I have already explained why feathers could have evolved, but I haven't said from what existing structures they could have evolved from. So how can we find out? One field of science which scientists can more or less 'see' how various features could have evolved is called developmental biology.
` When you look at how an animal grows - especially before it is born and when it is young - in order to grow into its current state, the easiest route for various parts of the anatomy seems to be to grow precursors of them and shape these into the modern state. That's why human embryos have gill-like structures, tails, and internal organs somewhat like those of the animals we've descended from, and these 'evolve' toward a state uniquely associated with our species until it fully resembles what you'd call a 'human'.
` In whales and snakes, tiny hind legs are visible in the embryonic state, as they had evolved before in the lineage, but these degenerate. On the same note, birds are born with three fingers like other coelurosaurs, though before they hatch, one can observe five distinct digits like earlier dinosaurs had! (More bizarreness about finger development later...)
` Of course, the three fingers of birds fuse together before they are born anyway - except for the hoatzin (HWAT-zin) of South America. When the hoatzin is born (hatched, anyway), it has two hooked claws on its hands - which resemble those of primitive birds. From what I already know about similar things like this, I would say that this is because the development of hoatzin hands is slower than other birds; as it grows into an adult, its hands 'evolve' into 'modern' bird wingtips.

` Of course, if you screw with genes while an embryo is developing, parts of it might not 'evolve' all the way, similar to the natural development of wings in hoatzins.
` Now, when feathers begin developing, they start out as a scale-like structure, which develops into a follicle before sprouting barbs and a rachis.
` In addition to feathers, of course, birds have scales. On the feet, specifically, are three types; scutes on the top; scutellae on the back; and the pebbly 'reptillian scales' called reticulae on the bottom.
` While injecting embryonic animals with a virus to block the development of specific proteins in one of their limbs (to see what happened), Hongyan Zou and Lee Niswander found that they could keep a chicken's toes from separating, meaning the webbing between them did not commit 'cell-icide' (called apoptosis). This also resulted in the scutes becoming feathers!

` In one way, this is not terribly surprising, as Alan Brush has shown that bird scutes, scutellae, claw sheaths, beak sheaths, and scales around the eyes are of the same chemical composition as feathers and are controlled by the same genes!
` The reticulae have been shown to be identical to those of crocodiles, both in composition and their location in the DNA. However, crocodiles also have scutes, which have almost the same chemical composition as bird scutes.
` Each time they infected a chick with the inhibitor virus at 15-18 days, at least some of the scutes developed into feathers in varying degrees, ranging from a thickening of the edge of the scute (like a follicle, I gather), to short, fat feathers, to long, thin feather filaments.

` The feathers developed the barbs characteristic of normal feathers, though there were a lot more of them.
` Unsurprisingly, the pebbly reticulae did not develop into feathers, as these scales are, according to Brush, more reptillian in various ways. Though Dhouaily, Hardy and Sengel (1980) were able to convert reticulae into feathers after treating them with retinoic acid, they were much more successful with the other two types of scales - keep in mind, I don't think this conversion had that much to do with development.
` Clearly, scutes and feathers are much more strongly linked than the reptilian scales, reticulae being chemically and genetically different. Now, dinosaurs such as Tyrannosaurus also had scutes - in fact armored dinosaurs were actually covered in modified scutes. And we know that more primitive tyrannosaurs and related dinosaurs had feathers.
` You know, before I'd learned these things, I might have asked; 'So were feathers modified scutes?'

` But you know what's really weird about that question? If you haven't caught this by now from all I've written so far, generally, when one character develops from another, new signals (from proteins, chemicals, or genes) are required to keep the old one from forming. Accordingly, when scientists block the evolutionarily newer signals in birds, it allows things from the ancestors of birds to break through!
` For example, the fused tail-stump of modern birds becomes a longer, less modified tail, or a bird's beak will grow teeth similar to those of primitive birds! Neat, huh?
` But when Zou and Niswander blocked signals in chick embryos... scutes became feathers! (Instead of feathers only developing as far as scutes!) Very counterintuitive!

` I think that there may be more going on than meets the eye. Because I don't know any other option, I'm tempted to wonder if bird-foot-scutes evolved from feathers in earlier dinosaurs or other archosaurs, which themselves evolved from other scutes.
` Or... did scutes evolve from feathers? Did the ancestors of crocodiles have feathers? Well, I just saw an article headline in Nature that new research shows that crocodiles may have evolved from warm-blooded ancestors, because their hearts - though different than those of warm-blooded animals - have four chambers and other structures that are similar.
` Whatever it says beyond the introduction, I have no idea, because I can't access most full-text articles out of my reluctance to pay for them. I'd sure like to know everything biologists know about the evolution of archosaurs! The only way to do that is to ask several of them.

` I can't say I know many... so unfortunately, for now, I'm stopped at saying; "I know that scutes and feathers are very similar (as opposed to reticulae), but I don't know how they are linked."

` Moving on, another disconnected bit of information (which is interesting nonetheless), shows that feathers do not develop in the most intuitive way, to say the least. It had been thought that the rachis was more like an extension of the scale, with the barbs developing second. But Cheng-Ming Chuong et al dared to screw with feathers themselves and were richly rewarded.
` The found that when you pluck feathers from chickens (ouch!) and observe their regrowth while screwing around with the expression of three genes important to feather formation, interesting things happen!
` When the team increased the expression of the gene noggin, the rachis began to split into several, small, thin rachides, with more barbs. Increasing the expression of Bmp 4 - with which noggin is antagonistic - caused the rachis to come out much bigger, with the barbs merged together around it. When they suppressed sonic hedgehog, the barbs turned out webbed instead of separate where the cells failed to self-destruct (like in the chicken's feet).
` Through such genetic tinkering and careful observation, they found that barbs form first around the inside of the calamus (quill) and later fuse to make a rachis. The discovery makes sense from a developmental perspective, as baby birds have down feathers, which have barbs but not rachides.
` Later on, they develop the more 'advanced' feathers as the barbs fuse into a rachis. Since down is the most simple and first-formed on birds, it is probably the first type of feather, which could have easily evolved for insulation.
` For a visual reference, these animations of feather growth might help.

` That's all for now - hopefully I'll add more to this entry when I think of it... My next DatD entry is about how flight could have evolved.