by Isaac Asimov
This proposal seemed to make sense. The material in chromosomes and elsewhere was thought of as nucleoprotein, which in turn consisted of a large protein molecule to which were attached one or more of these tetranucleotide groups, which served some unknown but, presumably, subsidiary purpose.
It turned out, however, that what Levene had isolated were not nucleic-acid molecules but pieces of them; and by the middle 1950s, biochemists found that the molecular weights of nucleic acids ran as high as 6 million. Nucleic acids are thus certainly equal and very likely superior to proteins in molecular size.
The exact manner in which nucleotides are built up and interconnected was confirmed by the British biochemist Alexander Robertus Todd, who built up a variety of nucleotides out of simpler fragments and carefully bound nucleotides together under conditions that allowed only one variety of bonding. He received the Nobel Prize in chemistry in 1957 for this work.
As a result, the general structure of the nucleic acid could be seen to be somewhat like the general structure of protein. The protein molecule is made up of a polypeptide backbone out of which jut the side chains of the individual amino acids. In nucleic acids, the sugar portion of one nucleotide is bonded to the sugar portion of the next by means of a phosphate group attached to both. Thus, a sugar-phosphate backbone runs the length of the molecule, and from it extend purines and pyrimidines, one to each nucleotide.
Nucleoproteins, it became clear, consist of two parts that are each large macromolecules. The question of the function of the nucleic-acid portion became more urgent.
DNA
By the use of cell-staining techniques, investigators began to pin down the location of nucleic acids in the cell. The German chemist Robert Feulgen, employing a red dye that stained DNA but not RNA, found DNA located in the cell nucleus, specifically in the chromosomes. He detected it not only in animal cells but also in plant cells. In addition, by staining RNA, he showed that this nucleic acid, too, occurs in both plant and animal cells. In short, the nucleic acids are universal materials existing in all living cells.
The Swedish biochemist Torbiorn Caspersson studied the subject further by removing one of the nucleic acids (by means of an enzyme that reduced it to soluble fragments that could be washed out of the cell) and concentrating on the other. He would photograph the cell in ultraviolet light; since a nucleic acid absorbs ultraviolet much more strongly than do other cell materials, the location of the DNA or the RNA—whichever he had left in the cell—showed up clearly. By this technique, DNA showed up only in the chromosomes. RNA made its appearance mainly in certain particles in the cytoplasm. Some RNA also showed up in the nucleolus, a structure within the nucleus. (In 1948, the Rockefeller Institute biochemist Alfred Ezra Mirsky showed that small quantities of RNA are present even in the chromosomes, while Ruth Sager showed that DNA can occur in the cytoplasm, notably in the chloroplasts of plants. In 1966, DNA was located in the mitochondria, too.)
Caspersson’s pictures disclosed that the DNA lies in localized bands in the chromosomes. Was it possible that DNA molecules are none other than the genes, which up to this time had had a rather vague and formless existence?
Through the 1940s, biochemists pursued this lead with growing excitement. They found it particularly significant that the amount of DNA in the cells of an organism was always rigidly constant, except that the sperm and egg cells had only half this amount-as expected, since they had only half the chromosome supply of normal cells. The amount of RNA and of the protein in chromosomes might vary all over the lot, but the quantity of DNA remained fixed. This certainly seemed to indicate a close connection between DNA and genes.
The nucleic acid tail was beginning to wag the protein dog, and then some remarkable observations were reported that seemed to show that the tail was the dog.
Bacteriologists had long studied two different strains of pneumococci grown in the laboratory: one with a smooth coat made of a complex carbohydrate; the other lacking this coat and therefore rough in appearance. Apparently the rough strain lacked some enzyme needed to make the carbohydrate capsule. But an English bacteriologist named Fred Griffith had discovered that, if killed bacteria of the smooth variety were mixed with live ones of the rough strain and then injected into a mouse, the tissues of the infected mouse would eventually contain live pneumococci of the smooth variety! How could this happen? The dead pneumococci had certainly not been brought to life. Something must have transformed the rough pneumococci so that they were now capable of making the smooth coat. What was that something? Evidently it was some factor contributed by the dead bacteria of the smooth strain.
In 1944, three American biochemists—Oswald Theodore Avery, Colin Munro Macleod, and Maclyn McCarty—identified the transforming principle. It was DNA. When they isolated pure DNA from the smooth strain and gave it to rough pneumococci, that alone sufficed to transform the rough strain to a smooth.
Investigators went on to isolate other transforming principles, involving other bacteria and other properties, and in every case the principle turned out to be a variety of DNA. The only plausible conclusion was that DNA could act like a gene. In fact, various lines of research, particularly with viruses (see chapter 14), showed that the protein associated with DNA is almost superfluous from a genetic point of view: DNA can produce genetic effects all by itself, either in the chromosome or—in the case of nonchromosomal inheritance—in cytoplasmic bodies such as the chloroplasts and mitochondria.
THE DOUBLE HELIX
If DNA is the key to heredity, it must have a complex structure, because it has to carry an elaborate pattern, or code of instructions (the genetic code), for the synthesis of specific enzymes. If it is made up of the four kinds of nucleotide, they cannot be strung in a regular arrangement, such as 1, 2, 3, 4, 1, 2, 3, 4, 1, 2, 3, 4… Such a molecule would be far too simple to carry a blueprint for enzymes. In fact, the American biochemist Erwin Chargaff and his co-workers found definite evidence, in 1948, that the composition of nucleic acids was more complicated than had been thought. Their analysis showed that the various purines and pyrimidines are not present in equal amounts, and that the proportions vary in different nucleic acids.
Everything seemed to show that the four purines and pyrimidines were distributed along the DNA backbone as randomly as the amino acid side chains were distributed along the peptide backbone. Yet some regularities did seem to exist. In any given DNA molecule, the total number of purines seemed always to be equal to the total number of pyrimidines. In addition, the number of adenines (one purine) was always equal to the number of thymines (one pyrimidine), while the number of guanines (the other purine) was always equal to the number of cytosines (the other pyrimidine).
We could symbolize adenine as A, guanine as G, thymine as T, and cytosine as C. The purines would then be A + G and the pyrimidines T + C. The findings concerning any given DNA molecule could then be summarized as:
A = T
G = C
A + G = T + C
More general regularities also emerged. As far back as 1938, Astbury had pointed out that nucleic acids scatter X rays in diffraction patterns, a good sign of the existence of structural regularities in the molecule. The New Zealand-born British biochemist Maurice Hugh Frederick Wilkins calculated that these regularities repeat themselves at intervals considerably greater than the distance from nucleotide to nucleotide. One logical conclusion was that the nucleic-acid molecule takes the form of a helix, with the coils of the helix forming the repetitive unit noted by the X rays. This thought seemed the more attractive because Linus Pauling was at that time demonstrating the helical structure of certain protein molecules.
Wilkins’s conclusions were based largely on the X-ray diffraction work of his associate, Rosalind Elsie Franklin, whose role in the studies was consistently underplayed in part because of the anti-feminist attitudes of the British scientific establishment.
In 1953, the English physicist Francis Harry Compton Crick and his co-worker, the American biochem
ist (and one-time Quiz Kid) James Dewey Watson, put all the information together—making use of a key photograph taken by Franklin, apparently without her permission—and came up with a revolutionary model of the nucleic-acid molecule. This model represented it not merely as a helix but (and this was the key point) as a double helix—two sugar-phosphate backbones winding like a double-railed spiral staircase up the same vertical axis (figure 13.6). From each sugar-phosphate chain, purines and peptides extended inward toward each other, meeting as though to form the steps of this double-railed spiral staircase.
Figure 13.6. Model of the nucleic-acid molecule. The drawing at the left shows the double helix; in the center, a portion of it is shown in detail (omitting the hydrogen atoms); at the right is a detail of the nucleotide combinations.
Just how might the purines and pyrimidines be arrayed along these parallel chains? To make a good uniform fit, a double-ring purine on one side should always face a single-ring pyrimidine on the other, to make a three-ring width altogether. Two pyrimidines could not stretch far enough to cover the space; while two purines would be too crowded. Furthermore, an adenine from one chain would always face a thymine on the other, and a guanine on one chain would always face a cytosine on the other. In this way, one could explain the finding that A = T, G = C, and A + G = T + C.
This Watson-Crick model of nucleic-acid structure has proved to be extraordinarily fruitful; and Wilkins, Crick, and Watson shared the 1962 Nobel Prize in medicine and physiology as a result. (Franklin had died in 1958, so the question of her contribution did not arise.)
The Watson-Crick model makes it possible, for instance, to explain just how a chromosome may duplicate itself in the process of cell division. Consider the chromosome as a string of DNA molecules. The molecules can first divide by a separation of the two helices making up the double helix; the two chains unwind themselves fro n each other, so to speak. This can be done because opposing purines and pyrimidines are held by hydrogen bonds, weak enough to be easily broken. Each chain is a half-molecule that can bring about the synthesis of its own missing complement. Where it has a thymine, it attaches an adenine; where it has a cytosine, it attaches a guanine; and so on. All the raw materials for making the units, and the necessary enzymes, are on hand in the cell. The half-molecule simply plays the role of a template, or mold, for putting the units together in the proper order. The units eventually will fall into the appropriate places and stay there because that is the most stable arrangement.
To summarize, then, each half-molecule guides the formation of its own complement, held to itself by hydrogen bonds. In this way, it rebuilds the complete, double-helix DNA molecule, and the two half-molecules into which the original molecule divided thus form two molecules where only one existed before. Such a process, carried out by all the DNAs down the length of a chromosome, will create two chromosomes that are exactly alike and perfect copies of the original mother chromosome. Occasionally something may go wrong: the impact of a subatomic particle or of energetic radiation, or the intervention of certain chemicals, may introduce an imperfection somewhere or other in the new chromosome. The result is a mutation.
Evidence in favor of this mechanism of replication has been piling up. Tracer studies, employing heavy nitrogen to label chromosomes and following the fate of the labeled material during cell division, have tended to bear out the theory. In addition, some of the important enzymes involved in replication have been identified.
In 1955, the Spanish-American biochemist Severo Ochoa isolated from a bacterium (Aztobacter vinelandii) an enzyme that proved capable of catalyzing the formation of RNA from nucleotides. In 1956, a former pupil of Ochoa’s, Arthur Kornberg, isolated another enzyme (from the bacterium Escherichia coli), which could catalyze the formation of DNA from nucleotides. Ochoa proceeded to synthesize RNA-like molecules from nucleotides, and Kornberg did the same for DNA. (The two men shared the Nobel Prize in medicine and physiology in 1959.) Kornberg also showed that his enzyme, given a bit of natural DNA to serve as a template, could catalyze the formation of a molecule that seemed to be identical with natural DNA. In 1965, Sol Spiegelman of the University of Illinois used RNA from a living virus (the simplest class of living things) and produced additional molecules of that sort. Since these additional molecules showed the essential properties of the virus, this was the closest approach yet to producing test-tube life. In 1967, Kornberg and others did the same, using DNA from a living virus as template.
The amount of DNA associated with the simplest manifestations of life is small—a single molecule in a virus—and can be made smaller. In 1967,
Spiegelman allowed the nucleic acid of a virus to replicate and selected samples after increasingly shorter intervals for further replication. In this way, he selected molecules that completed the job unusually quickly—because they were smaller than average. In the end, he had reduced the virus to one-sixth its normal size and multiplied replication speed fifteenfold.
Although it is DNA that replicates in cells, many of the simpler viruses.contain RNA only. RNA molecules in double strands replicate in such viruses. The RNA in cells is single-stranded and does not replicate.
Nevertheless, a single-stranded structure and replication are not mutually exclusive. The American biophysicist Robert Louis Sinsheimer discovered a strain of virus that contained DNA made up of a single strand. That DNA molecule had to replicate itself; but how could that be done with but a single strand? The answer was not difficult. The single strand brought about the production of its own complement, and the complement then brought about the production of the “complement to the complement”—that is, a replica of the original strand.
It is clear that the single-strand arrangement is less efficient than the double-strand arrangement (which is probably why the former exists only in certain very simple viruses and the latter in all other living creatures). For one thing, a single strand must replicate itself in two successive steps, whereas the double strand does so in a single step. Second, it now seems that only one strand of the DNA molecule is the important working structure—the cutting edge of the molecule, so to speak. Its complement may be thought of as a protecting scabbard for that cutting edge. The double strand represents the cutting edge protected within the scabbard except when actually in use; the single strand is the cutting edge always exposed and continually subjected to blunting by accident.
GENE ACTIVITY
Replication, however, merely keeps a DNA molecule in being. How does it accomplish its work of bringing about the synthesis of a specific enzyme—that is, of a specific protein molecule? To form a protein, the DNA molecule has to direct the placement of amino acids in a certain specific order in a molecule made up of hundreds or thousands of units. For each position it must choose the correct amino acid from some twenty different amino acids. If there were twenty corresponding units in the DNA molecule, it would be easy. But DNA is made up of only four different building blocks—the four nucleotides. Thinking about this, the astronomer George Gamow suggested in 1954 that the nucleotides, in various combinations, might be used as what we now call a genetic code (just as the dot and dash of the Morse code can be combined in various ways to represent the letters of the alphabet, numerals, and so on).
If you take the four different nucleotides (A, G, C, T), two at a time, there are 4 × 4, or 16 possible combinations (AA, AG, AC, AT, CA, GC, GC, GT, CA, CG, CC, CT, TA, TG, TC, and TT)—still not enough If you take them three at a time, there are 4 × 4 × 4, or 64 different combinations—more than enough. (You may amuse yourself trying to list the different combinations and see if you can find a sixty-fifth.)
It seemed as though each different nucleotide triplet or codon represented a particular amino acid. In view of the great number of different codons possible, it could well be that two or even three different codons represented one particular amino acid In this case, the genetic code would be what cryptographers call degenerate.
This left two chief questions: Which codon (or codon
s) correspond to which amino acid? And how does the codon information (which is securely locked in the nucleus where the DNA is to be found) reach the sites of enzyme formation in the cytoplasm?
To take the second problem first, suspicion soon fell upon RNA as the substance serving as go-between—as the French biochemists Francois Jacob and Jacques Lucien Monod were the first to suggest. The structure of such RNA would have to be very like DNA with such differences as existed not affecting the genetic code. RNA had ribose in place of deoxyribose (one extra oxygen atom per nucleotide) and uracil in place of thymine (one missing methyl group, CH3, per nucleotide). Furthermore, RNA was present chiefly in the cytoplasm, but also, to a small extent, in the chromosomes themselves.
It was not hard to see, and then demonstrate, what was happening. Every once in a while, when the two coiled strands of the DNA molecule unwound, one of those strands (always the same one, the cutting edge) replicates its structure, not on nucleotides that form a DNA molecule, but on nucleotides ’ that form an RNA molecule. In this case, the adenine of the DNA strand attaches not thymine nucleotides to itself but uracil nucleotides instead. The resulting RNA molecule, carrying the genetic code imprinted on its nucleotide pattern, can then leave the nucleus and enter the cytoplasm.
Since it carries the DNA message, it has been named messenger-RNA, or more simply, mRNA.
The Rumanian-American biochemist George Emil Palade, thanks to careful work with the electron microscope, demonstrated, in 1956, the site of enzyme manufacture in the cytoplasm to be tiny particles, about 2 millionths of a centimeter in diameter. They were rich in RNA and were therefore named ribosomes. There are as many as 15,000 ribosomes in a bacterial cell, perhaps ten times as many in a mammalian cell. They are the smallest of the subcellular particles or organelles. It was soon determined that the messenger-RNA—carrying the genetic code on its structure—makes its way to the ribosomes and layers itself onto one or more of them, and that the ribosomes are the site of protein synthesis.
