The Adsorption of Water Vapor and Other Gases by Proteins
Oetoder 5, 194k

Yor many years investigators have been interested in the amount of
water held by protein molecules, and many studies have been made of the
hydration of proteins, These studies have, however, led to little contri-
bution to the fundamental questions of protein structure. I believe that
the mensurement: of the amount of water vapor held by proteins at different
values of the partixl pressure of weter, and of the amount of other gases
similarly adsorbed by rreteins may be an extremely useful method of obtaining
information cbout the structure of proteins, Interesting expo rimental data
on the hydration of proteins held in atmospheres of varying relative humidity
have been reported in the September issue of the Journal of the Amoricen
Chemical Society by Henry Dull and in the September iceue of the Journel
of Chemical Physics by T. M. Shaw. These authors interpreted their data
by the application of the Srunauer-Snmett-~Teller theory of multi-layer
adsorption, obtaining in this way a velue for each protein of the surface
area on which water vapor may be adsorbed. ‘these effective surface areas
wero found to vary between ten percent and forty rercent of Kua total
surfsce area of tue proteins when spread in thin filus. Wo very significant
interpretation of the values of the surface arcas wns made.

‘It cceurred tO mé tant a comparison of tue number of witer wolecules
held by primary adsorption and the nuzber of amino acid residues of a polar
muture, that lea, capsble cf forming hydrogen bonds with the water nolecuiesn,

might be interesting. This has indeed been found to ve se.
the nuaber of witer molecules held strongly by silk, as given ty the
Saterpretation of Bull's date, is just equal to the sum of the numbers of
serine and tyrosine residues, these being the only pelar residues in the
protein. Fir other proteins there is similar evidence that primary attraction
for w.ter vapor results in the attachment of one molecule of water to each
polar residue,

A very interesting conclusion may be sade from the data for gelatin,
It is found that the number of weter molequles held strongly by gelatin is
equal to the total number of amino acid residues normally considered to be
polar, plus the nupber of proline residues, In most proteins the imino snd
carbonyl groups of the peptide links seem not to have very strong attraction
for water. The result for gelatin suggests thet the proline residue does
attract a water molecule, presumbly because the tend between the hydrecerbon
side chain and the nitrogen atom liberates a cgrbdonyl group, which otherwise
would be used in hydrogen bend formation with an N-# group of a pactids tend,
and the liberated carbonyl attracts and holds a v-ter molecule.

The protein salmin shows a unique behavior in ite attraction for water.
The adsorption ieotherm may be interpreted by 4 thecry which rapresents an
extension of the Brunsuer-Smeett-Teller theory, *han thie Intervretation te
made it is yvound that ine first prosees is tne addition of one water molecule
for every two arginine residuas, this acleaule yresusc bly being bonded to both
residuee by ceans of its teo hydrogen atoms. The next atop consists in the
addition of an equal amount of water, presumably rmsulting in « citustion in
which each arginine residue holds ons mater molecule, than ther: are intro}
duced successive layara of water moloqales, the amount of ister in each laver

being just the auount nevesenry te produce s unimolecular layer between iayers
of the protein of the thickness precent in monomolecular films; the amount
ef water in ench of these layers if 3.2 x that held by the primary process.

I think that it is very probable that similarly detailed information
can be obteined about other preteins with ue of experimental data on the
eieorction of water. It is not unlikely that careful meaqurements sade at
very low relative huuidities would indiomte the aidition of water succassively
to the groups with diminishing attraction for the water molecules, perhaps
permitting an amino seid analysis of a protain to be mde by this technique,
The stuay of schor guises. auch as amtonin, hydracine, hydroxylamine, hydrogen
poroxide, and aydrogen sulfide, awlgut vrevide add4¢ional vahuable information,
Jt ds chear tht the job sf collecting exp rivental outa would be o large one,

wut that the results crasumabdly would justify it.