To: Professor Pauling From: H. Segall Date: 7-11-55 E
SUBJECT: The Average of Amino Acid Residues Contained in the Unit Cell of Collagen.
A survey of the recent literature was made to collect consistent on the unit cell
dimensions, density and average residue weight for collagen. The most complete work
on dry and wet untreated kangaroo tail tendon was reported by Rougvie and Bear (J.
Amer. Leather Chemists Assn., 48, 735 (1953)). A less comprehensive investigation
was made by Dr. R. A. Pasternak of this Institute and was contained in a report to
Professor Pauling on March 11, 1954.
According to the report of Dr. Schroeder of Nov. 6, 1952 the existing chemical analyses
of the amino acid content of mammalian collagen are sufficiently consistent not to
warrant any further determinations at present. An average residue weight of 92.6 was
taken from the work of Bowes and Kenton (Biochem. J., 43, 358 (1948))
(1) Rougvie and Bear reported the equatorial spacing of exhaustively dried kangaroo
tail tendon as 10.6 A. The sample was dried for one week over P2O5 at room temperature. Although this value is slightly higher than other reported values
it is consistent with their measurements of the equatorial spacing as a function of
the water content of collagen from dry to water saturated specimens. The usual meridional
distance of 2.86 Å was also reported. The volume of the hexagonal unit cell is accordingly
V = 2.86 x 10.62/cos 30 = 371.1 A3.
The density of dry unpurified collagen was determined by flotation in bromheptane
and carbon tetrachloride by Rougvie and Bear. Their reported value of 1.34 is close
to other reported values of dry collagen and gelatin. The weight for a mole of units
is then
M = V x d x N = 371.1 x 10-24 x 1.34 x 6.023 x 1023 = 299.5 gms
By taking 92.6 as the average residue weight the average number of amino acid residues
in the unit cell, n, is found to be
[Data Equation]
The uncertainties in this value of n are the average residue weight (1 or 2%) and
the density. The presence of non-collagenous material in the unpurified sample may
affect the observed density.
(2) The experimental values as determined by Dr. Pasternak for hydrated collagen containing
16.3% water are as follows:
meridional spacing = 2.88 Å
equatorial spacing = 11.54 Å
density = 1.32
By taking an average residue weight of 93.3 and adding 18.6 for the water associated
with it he obtained an apparent molecular weight of 111.9. The calculated value of
n was 3.15.
Rougvie and Bear have shown that from 0 to about 20 g water/100 g collagen the sorbed
water was effective in increasing the volume of the unit cell according to its normal
density of 1.00. Greater hydration increases the volume at a considerable smaller
rate. Thus it can be said that Dr. Pasternak’s value of n is fortuitously close to
Bear’s value since his water of hydration falls within the 0 to 20 range.
(3) The value of 3.23 as the average number of amino acid residues in the unit cell
of collagen calculated from the data of Rougvie and Bear is probably accurate to 3
or 4 per cent. The reasonably exhaustive work of Rougvie is sufficient for the present.
Herbert Segall
