330 Chapter 9. Cooperative transitions in macromolecules[[Student version, January 17, 2003]]
simple Mass Action (solid dots in Figure 9.9).
A. Hill found a more successful model for hemoglobin’s oxygen binding in 1913: If we assume
that hemoglobin can bindseveraloxygen molecules, and does so in an all-or-nothing fashion, then
the binding reaction becomes Hb+nO 2 Hb(O 2 )n.Hill’s proposal is very similar to the cooperative
model of micelle formation (see Section 8.4.2 on page 279).
Your Turn 9n
a. Find the fractional bindingY in Hill’s model.
b. For what values ofnandKeqwill this model give an inflection point in the curve ofY versus
[O 2 ]?Fitting the data to bothKeqandn,Hill found the best fit for myoglobin gaven=1,asexpected,
butn≈ 3 for hemoglobin.
These observations began to make structural sense after G. Adair established that hemoglobin
is atetramer,that is, it consists of four subunits, each resembling a single myoglobin molecule,
and in particular each with its own oxygen binding site. Hill’s result implies that the binding of
oxygen to hemoglobin is highly cooperative. The cooperativity is not really all-or-none, since the
effective value ofnis less than the number of binding sites (four). Nevertheless,the binding of one
oxygen molecule leaves hemoglobin predisposed to bind more.After all, if each binding site operated
independently, we would have foundn=1,since in that case the sites may as well have been on
completely separate molecules.
Cooperativity is certainly a good thing for hemoglobin’s function as an oxygen carrier. Coop-
erativity lets hemoglobin switch readily between accepting and releasing oxygen. Figure 9.9 shows
that a noncooperative carrier would either have too high a saturation in tissue (like myoglobin), and
so fail to release enough oxygen, or else too low a saturation in the lungs (like the imaginary carrier
shown as the dashed line), and so fail to accept enough oxygen. Moreover, hemoglobin’s affinity for
oxygen can be modulated by other chemical signals besides the level of oxygen itself. For example,
Bohr also discovered that the presence of dissolved carbon dioxide or other acids (produced in the
blood by actively contracting muscle), promotes the release of oxygen from hemoglobin (delivering
more oxygen when it is most needed). ThisBohr effectfits with what we have already seen: Once
again, binding of a molecule (CO 2 )atone site on the hemoglobin molecule affects the binding of
oxygen at another site, a phenomenon calledallostery.More broadly, “allosteric control” is crucial
to the feedback mechanisms regulating many biochemical pathways (Figure 9.10).
The puzzling aspect of all these interactions is simply that the binding sites for the four oxygen
molecules (and for other regulatory molecules such as CO 2 )arenot closeto each other. Indeed
M. Perutz’s epochal analysis of the shape of the hemoglobin molecule in 1959 showed that the closest
twoiron atoms in hemoglobin, which bind the oxygens, are 2. 5 nmapart. (The full elaboration
of this structure, using X-ray crystallography, took Perutz twenty-three years!) Quite generally,
interactions between spatially distant binding sites on a macromolecule are calledallosteric.For
some time it was difficult to imagine how such interactions could be possible at all. After all, we
have seen that the main interactions responsible for molecular recognition and binding are of very
short range. How, then, can one binding site communicate its occupancy to another one?
9.6.2 Allostery often involves relative motion of molecular subunits
Akey clue to the allostery puzzle came in 1938, when F. Haurowitz found that crystals of hemoglobin
had totally different morphologies when prepared with or without oxygen: The deoxygenated form