Fine Feathered Friends?

[augwhite@neosoft.com (Toby White)]

Pete Buchholz wrote:

>SCHWEITZER gave an amazing talk on protein presserved with the flying raptor
>and _Mononykus_.  She recovered proteins that were homologous in structure
>and chemistry to the beta keritan claw sheeths of modern birds.
> Additionally, she showed quite conclusively, that _Mononykus_ had some sort
>of hollow beta keritan tube like structure, probably homologous to the
>central rachis of bird feathers.  The protein data is very good and indicates
>that the structures were entirely beta keritan, which is only seen in
>feathers.  The SEM photo she showed of one of these structures was VERY
>similar to a photo of a turkey vulture neck feather.

BOREDOM ALERT: what follows is technical and tedious in the extreme.

Keratin is one of the most common proteins on earth.  Skin is rich in
keratin and leather is, in fact, almost pure keratin.  Looking for traces of
keratin residues invites artifacts because the stuff could have come from
anywhere, including the skin of a preparator or the scales or skin of a
completely featherless dinosaur.  In fact, keratin is a major structural
component of hair, wool, skin, scales, nails, hooves and horns as well as
feathers. 

<omitted daydream about a flying Jurassic sheep>  

Schweitzer et al. note that what they saw was beta keratin.  The difference
between alpha and beta keratin is largely conformational, not chemical.  Are
they really maintaining that the protein, not only survived, but maintained
its three-dimensional structure for a hundred million years?  

Their second point is that only beta keratin was found.  Assuming this is
not a contaminant (a major assumption), it is not at all clear that the
conditions under which the fossil was preserved would permit alpha keratin
to exist in that form.  Moist heat, in particular can break up the helical
conformation of alpha keratin.   In any case, if their premise is that beta
keratin is only found in feathers, they're simply wrong.

Third, they apparently stated that only beta keratin was found.  That is,
that no other protein species was recovered in measurable amounts.  This is
unsurprising whether or not the keratin was a contaminant.  Again assuming
no contamination, keratin is very common, very insoluble, and individual
keratin fibres are very sticky with respect to each other.  Thus it is hard
to digest and will be preferentially preserved when everything else is
either dissolved or degraded.  The fact that nothing else was recovered may
only mean that (a) it was very abundant to begin with and (b) everything
else got eaten up or washed away (either naturally or by the prep method).  

Finally, the tubes.  Keratin molecules are sticky in the beta conformation
because adjacent strands hydrogen bond to each other.  This conformation,
the "beta sheet," is energeticly favored strongly whether the strands are
parallel or antiparallel.  If the number of strands falls within a fairly
broad range of sizes, there should be a strong energetic reason for the
strands to curl around to form a tube-like structure.  In this conformation,
all of the hydrogen bonding is with other keratin strands rather than the
less-favored interaction with water.  (The SEM prep method could affect the
result as well.)

I'd be real careful about citing this one.  Its an interesting result, and
may be correct, but there's an awful lot that could have gone wrong.

  --Toby White