12. IMMUNOLOGY OF FISH
473
A combining site area appears to involve N-terminal portions of
both the H and L chains (cf. Pressman et al., Cold Spring Harbor Symposium, 1968). The specific nature of any one site is apparently determined by variants in the primary amino acid sequence of the regions
involved. These sequences in turn determine the ultimate three-dimensional tertiary structure of the site which must “fit” intimately into contact with the antigenic determinant for which it is contrived. (Note that
the growth of polypeptide chains starts at the N-terminal end and
finishes at the C-terminal.) From 40 to 73 of the 100 or so amino acid
positions along the N-terminal side of L chains appear to be variable
with respect to the amino acids that can occupy them, and the Nterminal fragment appears to be equally variable in H chains. In contrast, the remaining positions ( C terminating) along both chains
appear to be invariant except as stated below.
The invariant lengths of the chains, while interesting in their own
right, do not appear to be particularly remarkable, but rather they seem
to be synthesized according to the relations known to occur among
genes and the polypeptides such as those of hemoglobin. These relations
include the demonstration that the invariant portions of antibody molecules are much more alike than different among the individuals of several
mammalian species so far investigated. Differences that do occur seem
readily explainable as the result of mutations that substitute one amino
acid for another and that are reflected in the antigenic variants discussed earlier. Considering the “invariant” regions, the story seems to
be that a good case can be made for their common genetic ancestory,
modified through evolution by gene duplication and the relatively minor
evolutionary disturbances introduced by codon mutation ( Jukes, 1966).
Not so, however, for the “variable” regions of antibody molecules.
Here, as shown above, it is found that while some positions along the
chain are comparable to those of the invariant areas and remain constant
from individual to individual and species to species many other positions are extremely variable with respect to the kind of amino acid that
occurs in them, The magnitude of this variability is indicated by the
statement of Haurowitz (1968, p. 105) that in human L chains sequenced
at the time of his writing, approximately 22 of the 105 N-terminal amino
acids have been observed to be replaced by other amino acids. As he
points out, if it is assumed that each amino acid is replaced by only one
other amino acid, the number of combinations possible could produce
four million different L chains. Considering the same degree of variability in the H chains and that L and H chains are synthesized separately
and then assembled, there would seem to be a physical basis for providing for the “infinite” heterogeneity that is observed at Ig combining
473
A combining site area appears to involve N-terminal portions of
both the H and L chains (cf. Pressman et al., Cold Spring Harbor Symposium, 1968). The specific nature of any one site is apparently determined by variants in the primary amino acid sequence of the regions
involved. These sequences in turn determine the ultimate three-dimensional tertiary structure of the site which must “fit” intimately into contact with the antigenic determinant for which it is contrived. (Note that
the growth of polypeptide chains starts at the N-terminal end and
finishes at the C-terminal.) From 40 to 73 of the 100 or so amino acid
positions along the N-terminal side of L chains appear to be variable
with respect to the amino acids that can occupy them, and the Nterminal fragment appears to be equally variable in H chains. In contrast, the remaining positions ( C terminating) along both chains
appear to be invariant except as stated below.
The invariant lengths of the chains, while interesting in their own
right, do not appear to be particularly remarkable, but rather they seem
to be synthesized according to the relations known to occur among
genes and the polypeptides such as those of hemoglobin. These relations
include the demonstration that the invariant portions of antibody molecules are much more alike than different among the individuals of several
mammalian species so far investigated. Differences that do occur seem
readily explainable as the result of mutations that substitute one amino
acid for another and that are reflected in the antigenic variants discussed earlier. Considering the “invariant” regions, the story seems to
be that a good case can be made for their common genetic ancestory,
modified through evolution by gene duplication and the relatively minor
evolutionary disturbances introduced by codon mutation ( Jukes, 1966).
Not so, however, for the “variable” regions of antibody molecules.
Here, as shown above, it is found that while some positions along the
chain are comparable to those of the invariant areas and remain constant
from individual to individual and species to species many other positions are extremely variable with respect to the kind of amino acid that
occurs in them, The magnitude of this variability is indicated by the
statement of Haurowitz (1968, p. 105) that in human L chains sequenced
at the time of his writing, approximately 22 of the 105 N-terminal amino
acids have been observed to be replaced by other amino acids. As he
points out, if it is assumed that each amino acid is replaced by only one
other amino acid, the number of combinations possible could produce
four million different L chains. Considering the same degree of variability in the H chains and that L and H chains are synthesized separately
and then assembled, there would seem to be a physical basis for providing for the “infinite” heterogeneity that is observed at Ig combining