by Isaac Asimov
Graham called materials that could pass through parchment (and that happened to be easily obtained in crystalline form) crystalloids. Those that did not, such as glue (in Greek, kolla), he called colloids. The study of giant molecules (or giant aggregates of atoms, even where these do not form distinct molecules) thus came to be known as colloid chemistry. Because proteins and other key molecules in living tissue are of giant size, colloid chemistry is of particular importance to biochemistry (the study of the chemical reactions proceeding in living tissue).
Advantage can be taken of the giant size of protein molecules in a number of ways. Suppose that pure water is on one side of a sheet of parchment and a colloidal solution of protein on the other. The protein molecules cannot pass through the parchment; moreover, they block the passage of some of the water molecules, which might otherwise move through. For this reason, water moves more readily into the colloidal portion of the system than out of it. Fluid builds up on the side of the protein solution and sets up an osmotic pressure.
In 1877, the German botanist Wilhelm Pfeffer showed how one could measure this osmotic pressure and from it determine the molecular weight of a giant molecule. It was the first reasonably good method for estimating the size of such molecules.
Again, protein solutions could be placed in bags made of semipermeable membranes (membranes with pores large enough to permit the passage of small, but not large, molecules). If these were placed in running water, small molecules and ions would pass through the membrane and be washed away, while the large protein molecule would remain behind. This process of dialysis is the simplest method of purifying protein solutions.
Molecules of colloidal size are large enough to scatter light; small molecules cannot. Furthermore, light of short wavelength is more efficiently scattered than that of long wavelength. The first to note this effect, in 1869, was the Irish physicist John Tyndall; in consequence, it is called the Tyndall effect. The blue of the sky is explained now by the scattering effect of dust particles in the atmosphere upon the short-wave sunlight. At sunset, when light passes through a greater thickness of atmosphere rendered particularly dusty by the activity of the day, enough light is scattered to leave chiefly the red and the orange, thus accounting for the beautiful ruddy color of sunsets.
Light passing through a colloidal solution is scattered so that it can be seen as a visible cone of illumination when viewed from the side. Solutions of crystalloidal substances do not show such a visible cone of light when illuminated, and are optically clear. In 1902, the Austro-German chemist Richard Adolf Zsigmondy took advantage of this observation to devise an ultramicroscope, which viewed a colloidal solution at right angles, with individual particles (too small to be seen in an ordinary microscope) showing up as bright dots of light. For his endeavor, he received the Nobel Prize for chemistry in 1925.
The protein chemists naturally were eager to synthesize long, polypeptide chains, with the hope of producing proteins. But the methods of Fischet and Bergmann allowed only one amino acid to be added at a time—a procedure that seemed then to be completely impractical. What was needed was a procedure that would cause amino acids to join up in a kind of chain reaction, such as Baekeland had used in forming his high-polymer plastics. In 1947, both the Israeli chemist E. Katchalski and the Harvard chemist Robert Woodward (who had synthesized quinine) reported success in producing polypeptides through chain-reaction polymerization. Their starting material was a slightly modified amino acid. (The modification eliminated itself neatly during the reaction.) From this beginning, they built up synthetic polypeptides consisting of as many as a hundred or even a thousand amino acids.
These chains are usually composed of only one kind of amino acid, such as glycine or tyrosine, and are therefore called polyglycine or polytyrosine. It is also possible, by beginning with a mixture of two modified amino acids, to form a polypeptide containing two different amino acids in the chain. But these synthetic constructions resemble only the simplest kind of protein-for example, fibroin, the protein in silk.
THE POLYPEPTIDE CHAINS
Some proteins are as fibrous and crystalline as cellulose or nylon: for example, fibroin; keratin, the protein in hair and skin; and collagen, the protein in tendons and in connective tissue. The German physicist R. O. Herzog proved the crystallinity of these substances by showing that they diffract X rays. Another German physicist, Rudolf Brill, analyzed the pattern of the diffraction and determined the spacing of the atoms in the polypeptide chain. The British biochemist William Thomas Astbury and others in the 1930s obtained further information about the structure of the chain by means of X-ray diffraction. They were able to calculate with reasonable precision the distances between adjacent atoms and the angles at which adjacent bonds are set. And they learned that the chain of fibroin is fully extended: that is, the atoms are in as nearly a straight line as the angles of the bonds between them permit.
This full extension of the polypeptide chain is the simplest possible arrangement. It is called the beta configuration. When hair is stretched, its keratin molecule, like that of fibroin, takes up this configuration. (If hair is moistened, it can be stretched up to three times its original length.) But in its ordinary, unstretched state, keratin shows a more complicated arrangement, called the alpha configuration.
In 1951, Linus Pauling and Robert Brainard Corey of the California Institute of Technology suggested that, in the alpha configuration, polypeptide chains take a helical shape (like a spiral staircase). After building various models to see how the structure would arrange itself if all the bonds between atoms lay in their natural directions without strain, they decided that each turn of the helix would have the length of 3.6 amino acids, or 5.4 angstrom units.
What enables a helix to hold its structure? Pauling suggested that the agent is the so-called hydrogen bond. As we have seen, when a hydrogen atom is attached to an oxygen or a nitrogen atom, the latter holds the major share of the bonding electrons, so that the hydrogen atom has a slight positive charge and the oxygen or nitrogen a slight negative charge. In the helix, it appears, a hydrogen atom periodically occurs close to an oxygen or a nitrogen atom on the turn of the helix immediately above or below it. The slightly positive hydrogen atom is attracted to its slightly negative neighbor. This attraction has only 1/20 of the force of an ordinary chemical bond, but it is strong enough to hold the helix in place. However, a pull on the fiber easily uncoils the helix and thereby stretches the fiber.
We have considered so far only the “backbone” of the protein molecule—the chain that runs …CCNCCNCCNCCN… But the various side chains of the amino acids also play an important part in protein structure.
All the amino acids except glycine have at least one asymmetric carbon atom—the one between the carboxyl group and the amine group. Thus each could exist in two optically active isomers. The general formulas of the two isomers are:
However, it seems quite certain, from both chemical and X-ray analysis, that polypeptide chains are made up only of L-amino acids. In this situation, the side chains stick out alternately on one side of the backbone and then the other. A chain composed of a mixture of both isomers would not be stable, because, whenever an L-amino and a D-amino acid were next to each other, two side chains would be sticking out on the same side, which would crowd them and strain the bonds.
The side chains are important factors in holding neighboring peptide chains together. Wherever a negatively charged side chain on one chain is near a positively charged side chain on its neighbor, they will form an electrostatic link. The side chains also provide hydrogen bonds that can serve as links. And the double-headed amino acid cystine can insert one of its amine-carboxyl sequences in one chain and the other in the next. The two chains are then tied together by the two sulfur atoms in the side chain (the disulfide link). The binding together of polypeptide chains accounts for the strength of protein fibers. It explains the remarkable toughness of the apparently fragile spider web and the fact that keratin can form s
tructures as hard as fingernails, tiger claws, alligator scales, and rhinoceros horns.
PROTEINS IN SOLUTION
All this nicely describes the structure of protein fibers. What about proteins in solution? What sort of structure do they have?
They certainly possess a definite structure, but it is extremely delicate.
Gentle heating or stirring of a solution or the addition of a bit of acid or alkali or any of a number of other environmental stresses will denature a dissolved protein: that is, the protein loses its ability to perform its natural functions, and many of its properties change. Furthermore, denaturation usually is irreversible: for instance, a hard-boiled egg can never be un-hard-boiled again.
It seems certain that denaturation involves the loss of some specific configuration of the polypeptide backbone. Just what feature of the structure is destroyed? X-ray diffraction will not help us when proteins are in solution, but other techniques are available.
In 1928, for instance, the Indian physicist Chandrasekhara Venkata Raman found that light scattered by molecules in solution was, to some extent, altered in wavelength. From the nature of the alteration, deductions could be made about the structure of the molecule. For this discovery of the Raman effect, Raman received the 1930 Nobel Prize for physics. (The altered wavelengths of light are usually referred to as the Raman spectrum of the molecule doing the scattering.)
Another delicate technique was developed twenty years later, one based on the fact that atomic nuclei possess magnetic properties. Molecules exposed to a high intensity magnetic field will absorb certain frequencies of radio waves.
From such absorption, referred to as nuclear magnetic resonance and frequently abbreviated NMR, information concerning the bonds between atoms can be deduced. In particular, NMR techniques can locate the position of the small hydrogen atoms within molecules, as X-ray diffraction cannot do. NMR techniques were worked out in 1946 by two teams, working independently: one under E. M. Purcell (later to be the first to detect the radio waves emitted by the neutral hydrogen atom in space; see chapter 2); and the other under the Swiss-American physicist Felix Bloch. Purcell and Bloch shared the Nobel Prize for physics in 1952 for this feat.
To return, then, to the question of the denaturation of proteins in solution. The American chemists Paul Mead Doty and Elkan Rogers Blout used lightscattering techniques on solutions of synthetic polypeptides and found them to have a helical structure. By changing the acidity of the solution, Doty and Blout could break down the helices into randomly curved coils; by readjusting the acidity, they could restore the helices. And they showed that the conversion of the helices to random coils reduced the amount of the solution’s optical activity. It was even possible to show which way a protein helix is twisted: it runs in the direction of a right-handed screw thread.
All these findings suggest that the denaturation of a protein involves the destruction of its helical structure.
BREAKING DOWN A PROTEIN MOLECULE
So far I have taken an over-all look at the structure of the protein molecule—the general shape of the chain. What about the details of its construction? For instance, how many amino acids of each kind are there in a given protein molecule?
We might break down a protein molecule into its amino acids (by heating it with acid) and then determine how much of each amino acid is present in the mixture. Unfortunately, some of the amino acids resemble each other chemically so closely that it is almost impossible to get clear-cut separations by ordinary chemical methods. The amino acids can, however, be separated neatly by chromatography (see chapter 6). In 1941, the British biochemists Archer John Porter Martin and Richard Laurence Millington Synge pioneered the application of chromatography to this purpose. They introduced the use of starch as the packing material in the column. In 1948, the American biochemists Stanford Moore and William Howard Stein brought the starch chromatography of amino acids to a high pitch of efficiency and, as a result, shared the 1972 Nobel Prize in chemistry.
After the mixture of amino acids has been poured into the starch column, and all the amino acids have attached themselves to the starch particles, they are slowly washed down the column with fresh solvent. Each amino acid moves down the column at its own characteristic rate. As each emerges at the bottom separately, the drops of solution of that amino acid are caught in a container. The solution in each container is then treated with a chemical that turns the amino acid into a colored product. The intensity of the color is a measure of the amount of the particular amino acid present. This color intensity is measured by an instrument called a spectrophotometer, which indicates the intensity by means of the amount of light of the particular wavelength that is absorbed (figure 12.2).
Figure 12.2. A spectrophotometer. The beam of light is split into two, so that one beam passes through the specimen being analyzed and the other goes directly to the photocell. Since the weakened beam that has passed through the specimen liberates fewer electrons in the photocell than the unabsorbed beam does, the two beams create a difference in potential that measures the amount of absorption of the light by the specimen.
(Spectrophotometers can, by the way, be used for other kinds of chemical analysis. If light of successively increased wavelength is sent through a solution, the amount of absorption changes smoothly, rising to maxima at some wavelengths and falling to minima at others. The result is an absorption spectrum. A given atomic group has its own characteristic absorption peak or peaks. This is especially true in the region of the infrared, as was first shown by the American physicist William Weber Coblentz shortly after 1900. His instruments were too crude to make the technique practical then; but since the Second World War, the infrared spectrophotometer, designed to scan, automatically, the spectrum from 2 to 40 microns, and to record the results, has come into increasing use for analysis of the structure of complex compounds. Optical methods of chemical analysis, involving radio-wave absorption, light absorption, light scattering, and so on, are extremely delicate and nondestructive—the sample survives the inspection, in other words—and are completely replacing the classical analytical methods of Liebig, Dumas, and Pregl that were mentioned in the previous chapter.)
The measurement of amino acids with starch chromatography is quite satisfactory; but by the time this procedure was developed, Martin and Synge had worked out a simpler method of chromatography. It is called paper chromatography (figure 12.3). The amino acids are separated on a sheet of filter paper (an absorbent paper made of particularly pure cellulose). A drop or two of a mixture of amino acids is deposited near a corner of the sheet, and this edge of the sheet is then dipped into a solvent, such as butyl alcohol. The solvent slowly creeps up the paper through capillary action. (Dip the corner of a blotter into water and see it happen yourself.) The solvent picks up the molecules in the deposited drop and sweeps them along the paper. As in column chromatography, each amino acid moves up the paper at a characteristic rate. After a while the amino acids in the mixture become separated in a series of spots on the sheet. Some of the spots may contain two or three amino acids. To separate these, the filter paper, after being dried, is turned around ninety degrees from its first position, and the new edge is now dipped into a second solvent which will deposit the components in separate spots. Finally, the whole sheet, after once again being dried, is washed with chemicals that cause the patches of amino acids to show up as colored or darkened spots. It is a dramatic sight: all the amino acids, originally mixed in a single solution, are now spread out over the length and breadth o(the paper in a mosaic of colorful spots. Experienced biochemists can identify each amino acid-by the spot it occupies, and thus can read the composition of the original protein almost at a glance. By dissolving a spot, they can even measure how much of a particular amino acid was present in the protein. For their development of this technique, Martin and Synge received the 1952 Nobel Prize in chemistry.
Figure 12.3. Paper chromatography.
(Martin, along with A. T. James, applied the princ
iples of this technique to the separation of gases in 1952. Mixtures of gases or vapors may be passed through a liquid solvent or over an adsorbing solid by means of a current of inert carrier gas, such as nitrogen or helium. The mixture is pushed through and emerges at the other end separated. Such gas chromatography is particularly useful because of the speed of its separations arid the great delicacy with which it can detect trace impurities.)
Chromatographic analysis yielded accurate estimates of the amino-acid contents of various proteins. For instance, the molecule of a blood protein called serum albumin was found to contain 15 glycines, 45 valines, 58 leucines, 9 isoleucines, 31 prolines, 33 phenylalanines, 18 tyrosines, 1 tryptophan, 22 serines, 27 threonines, 16 cystines, 4 cysteines, 6 methionines, 25 arginines, 16 histidines, 58 lysines, 46 aspartic acids, and 80 glutamic acids—a total of 526 amino acids of 18 different types built into a protein with a molecular weight of about 69,000. (In addition to these 18, there is one other common amino acid—alanine.)
The German-American biochemist Erwin Brand suggested a system of symbols for the amino acids which is now in general use. To avoid confusion with the symbols of the elements, he designated each amino acid by the first three letters of its name, instead of just the initial. There are a few special variations: cystine is symbolized CyS, to show that its two halves are usually incorporated in two different chains; cysteine is CySH, to distinguish it from cystine; and isoleucine is Ileu rather than Iso, for iso is the prefix of many chemical names.
