TO: Professor-Pauling FROM: W. A. Schroeder DATE: MARCH 18, 1955
SUBJECT: PEPTIDES FROM GELATIN AND COLLAGEN
Recently Kroner, of United Shoe Machinery Corporation has isolated: 28 new peptides
from partial hydrolysates of collagen (information now in press). His data are especially
useful because he has identified many tripeptides. His isolation of gly-pro-hypro
and gly-(pro, hypro)-gly is a gratifying confirmation of our suggestion that the pro-hypro
sequence may be important in collagen and gelatin especially because gly-pro-hypro
was isolated in an amount somewhat greater than calculated for purely random distribution
and the gly-(pro, hypro)-gly in an amount 3 times the calculated statistical frequency.
In attempting to make use of this rather considerable collection of sequences in gelatin
and collagen, I have felt that it might be suggestive to arrange the information in
the way in which it is done on the attached pages. On these pages, the amino acids
are abbreviated in the accepted way. A hyphen between names indicates that the sequence
is definitely known; if the names are enclosed in parentheses and separated by commas
the sequence of the amino acids within the parentheses is unknown. However, the arrangement
of names within parentheses is that which is considered to be the most probable sequence.
Question marks indicate that the peptide was only tentatively identified. Because
not all of the peptides were isolated from a single hydrolysate, the totals have only
questionable significance.
On the following pages, the peptides of a given amino acid have been arranged so that
the given amino acid is in a vertical column. In this way, it is easier to determine
by inspection whether the amino or carboxyl group of the amino acid tends to be associated
with a specific amino acid or type of amino acid. The most interesting conclusion
from this consideration of the data is that hydroxyproline tends to be associated
with X-hypro-gly -where X is any amino acid except glycine and proline to be associated
with gly-pro-X although here there is a somewhat lesser restriction. I might note
that I have become somewhat skeptical of Grassmann's identification of lys-pro-gly.
Neither Kroner nor I find any indication of -it and yet the quantity should be appreciable.
In view of Lindley's ideas about bending α-helixes through 180°, the identification
of such sequences as glu-glu, asp-asp, asp-glu, asp-arg, and glu-arg takes on added
significance.
An arrangement of peptides in the order of decreasing amount isolated is also given.
Gly-pro and hypro-gly are predominant. This predominance as well as other considerations
once led me to suggest that the sequence gly-(pro, hypro)-gly might be of much importance.
The available data now appear to lend much more credence to this suggestion than my
indirect reasoning. Apparently, in the sequences gly-pro-X and X2 -hypro-gly, X1 frequently
is hydroxyproline and X2 is proline.
Walter Schroeder
[Ten additional pages of data]
