15.5 Proteins

233

6. Crystallize the protein (often unusual salt conditions are required) and record

the X-ray diffractogram,¹⁵ or carry out nuclear magnetic resonance spectroscopy

(one or more ofSuperscript 1¹H,Superscript 13¹³C,Superscript 15¹⁵N) with a fairly concentrated solution of the protein to

yield an adjacency matrix (cf. Sect. 12.2) from which the pattern of through-bond

and through-space couplings can be derived.

7. Calculate the atomic coördinates.

8. Refine the structure by minimizing interatomic potentials, or use Ramachandran

plots.

Under favourable conditions, X-ray diffraction and nuclear magnetic resonance spec-

troscopy (n.m.r.) can yield structures at a resolution of 1 Å. Some of the difficulties

in these procedures are as follows:

1. The protein may not crystallize. Membrane proteins are especially problematical,

but their structures may be obtainable from high-resolution electron diffraction

of two-dimensional arrays, or by crystallizing them in a cubic-phase lipid.

2. Hydrogen atoms are insufficiently electron dense to be registered in the

X-ray diffractogram (they are detectable in experimentally more onerous neu-

tron diffraction).

3. Energy refinement will yield the majority structure. Most proteins have two or

more stable structures, which may be present simultaneously, although in unequal

proportions.

4. The crystal structure, or the structure in concentrated solution, may not be repre-

sentative of the native structure(s).

5. Nuclear magnetic resonance cannot cope with large proteins (the spectra become

too complicated, and the assignment of peaks to the individual amino acids along

the sequence becomes problematical).

6. Nuclear magnetic resonance yields a set of distance constraints, but there are

usually so many that the problem is overdetermined, and no physically possible

structure can satisfy all of them.

Protein stability can be assessed by determining the structure of a protein at different

temperatures. Since thermal denaturation is accompanied by a large change in spe-

cific heat, whose midpoint provides a quantitative parameter characterizing stability,

microcalorimetry is a useful technique for assessing stability.

15.5.4

Protein Structure Overview

The techniques described in the previous subsection revealed that proteins have a

compact structure akin to a ribbon folded back and forth. Drop a piece of thick

string about a metre long on a table, pick it up, and push it together between one’s

15 Multiple isomorphous replacement—MIR—whereby a few heavy atoms are introduced into the

protein, which is then remeasured, and is used to determine the diffraction phases. The heavy atoms

should not, of course, induce any changes in the protein structure.