But what was genetic code? How did four bases in a molecular string of DNA—A, C, G, and T (or A,C,G,U in RNA)—determine the consistency of hair, the color of an eye, or the quality of the coat of a bacterium (or, for that matter, the propensity for mental illness or a deadly bleeding disease in a family)? How did Mendel’s abstract “unit of heredity” become manifest as a physical trait?
In 1941, three years before Avery’s landmark experiment, two scientists, George Beadle and Edward Tatum, working in a basement tunnel at Stanford University, discovered the missing link between genes and physical traits. Beadle—or “Beets,” as his colleagues liked to call him—had been a student of Thomas Morgan’s at Caltech. The red-eyed flies and the white-eyed mutants puzzled Beadle. A “gene for redness,” Beets understood, is a unit of hereditary information, and it is carried from a parent to its children in an indivisible form in DNA—in genes, in chromosomes. “Redness,” the physical trait, in contrast, was the consequence of a chemical pigment in the eye. But how did a hereditary particle transmute into an eye pigment? What was the link between a “gene for redness” and “redness” itself—between information and its physical or anatomical form?
Fruit flies had transformed genetics by virtue of rare mutants. Precisely because they were rare, the mutants had acted like lamps in the darkness, allowing biologists to track “the action of a gene,” as Morgan had described it, across generations. But the “action” of a gene—still a vague, mystical concept—intrigued Beadle. In the late 1930s, Beadle and Tatum reasoned that isolating the actual eye pigment of a fruit fly might solve the riddle of gene action. But the work stalled; the connection between genes and pigments was far too complicated to yield a workable hypothesis. In 1937, at Stanford University, Beadle and Tatum switched to an even simpler organism called Neurospora crassa, a bread mold originally found as a contaminant in a Paris bakery, to try to solve the gene-to-trait connection.
Bread molds are scrappy, fierce creatures. They can be grown in petri dishes layered with nutrient-rich broth—but, in fact, they do not need much to survive. By systematically depleting nearly all the nutrients from the broth, Beadle found that the mold strains could still grow on a minimal broth containing nothing more than a sugar and a vitamin called biotin. Evidently, the cells of the mold could build all the molecules needed for survival from basic chemicals—lipids from glucose, DNA and RNA from precursor chemicals, and complex carbohydrates out of simple sugars: wonder from Wonder Bread.
This capacity, Beadle understood, was due to the presence of enzymes within the cell—proteins that acted as master builders and could synthesize complex biological macromolecules out of basic precursor chemicals. For a bread mold to grow successfully in minimal media, then, it needed all its metabolic, molecule-building functions to be intact. If a mutation inactivated even one function, the mold would be unable to grow—unless the missing ingredient was supplied back into the broth. Beadle and Tatum could thus use this technique to track the missing metabolic function in every mutant: if a mutant needed the substance X, say, to grow in minimal media, then it must lack the enzyme to synthesize that substance, X, from scratch. This approach was intensely laborious—but patience was a virtue that Beadle possessed in abundance: he had once spent an entire afternoon teaching a graduate student how to marinate a steak, adding one spice at a time, over precisely timed intervals.
The “missing ingredient” experiment propelled Beadle and Tatum toward a new understanding of genes. Every mutant, they noted, was missing a single metabolic function, corresponding to the activity of a single protein enzyme. And genetic crosses revealed that every mutant was defective in only one gene.
But if a mutation disrupts the function of an enzyme, then the normal gene must specify the information to make the normal enzyme. A unit of heredity must carry the code to build a metabolic or cellular function specified by a protein. “A gene,” Beadle wrote in 1945, “can be visualized as directing the final configuration of a protein molecule.” This was the “action of the gene” that a generation of biologists had been trying to comprehend: a gene “acts” by encoding information to build a protein, and the protein actualizes the form or function of the organism.
Or, in terms of information flow:
Beadle and Tatum shared a Nobel Prize in 1958 for their discovery, but the Beadle/Tatum experiment raised a crucial question that remained unanswered: How did a gene “encode” information to build a protein? A protein is created from twenty simple chemicals named amino acids—Methionine, Glycine, Leucine, and so forth—strung together in a chain. Unlike a chain of DNA, which exists primarily in the form of a double helix, a protein chain can twist and turn in space idiosyncratically, like a wire that has been sculpted into a unique shape. This shape-acquiring ability allows proteins to execute diverse functions in cells. They can exist as long, stretchable fibers in muscle (myosin). They can become globular in shape and enable chemical reactions—i.e., enzymes (DNA polymerase). They can bind colored chemicals and become pigments in the eye, or in a flower. Twisted into saddle clasps, they can act as transporters for other molecules (hemoglobin). They can specify how a nerve cell communicates with another nerve cell and thus become the arbiters of normal cognition and neural development.
But how could a sequence of DNA—ATGCCCC . . . etc.—carry instructions to build a protein? Watson had always suspected that DNA was first converted into an intermediate message. It was this “messenger molecule,” as he called it, that carried the instructions to build a protein based on a gene’s code. “For over a year,” he wrote in 1953, “I had been telling Francis [Crick] that the genetic information in DNA chains must be first copied into that of complementary RNA molecules,” and the RNA molecules must be used as “messages” to build proteins.
In 1954, the Russian-born physicist-turned-biologist George Gamow teamed with Watson to form a “club” of scientists to decipher the mechanism of protein synthesis. “Dear Pauling,” Gamow wrote to Linus Pauling in 1954, with his characteristically liberal interpretation of grammar and spelling, “I am playing with complex organic molecules (what I never did before!) and geting [sic] some amusing results and would like your opinnion [sic] about it.”
Gamow called it the RNA Tie Club. “The Club never met as a whole,” Crick recalled: “It always had a rather ethereal existence.” There were no formal conferences or rules or even basic principles of organization. Rather, the Tie Club was loosely clustered around informal conversations. Meetings happened by chance, or not at all. Letters proposing madcap, unpublished ideas, often accompanied by hand-scribbled figures, were circulated among the members; it was a blog before blogs. Watson got a tailor in Los Angeles to embroider green woolen ties with a golden strand of RNA, and Gamow sent a tie, and a pin, to each in the group of friends that he had handpicked as club members. He printed a letterhead and added his own motto: “Do or die, or don’t try.”
In the mid-1950s, a pair of bacterial geneticists working in Paris, Jacques Monod and François Jacob, had also performed experiments that had dimly suggested that an intermediate molecule—a messenger—was required for the translation of DNA into proteins. Genes, they proposed, did not specify instructions for proteins directly. Rather, genetic information in DNA was first converted into a soft copy—a draft form—and it was this copy, not the DNA original, that was translated into a protein.
In April 1960, Francis Crick and Jacob met at Sydney Brenner’s cramped apartment in Cambridge to discuss the identity of this mysterious intermediate. The son of a cobbler from South Africa, Brenner had come to England to study biology on a scholarship; like Watson and Crick, he too had become entranced by Watson’s “religion of genes” and DNA. Over a barely digested lunch, the three scientists realized that this intermediate molecule had to shuttle from the cell’s nucleus, where genes were stored, to the cytoplasm, where proteins were synthesized.
But what was the chemical identity of the “message” that was built from a gene? Was it a protein
or a nucleic acid or some other kind of molecule? What was its relationship to the sequence of the gene? Although they still lacked concrete evidence, Brenner and Crick also suspected that it was RNA—DNA’s molecular cousin. In 1959, Crick wrote a poem to the Tie Club, although he never sent it out:
What are the properties of Genetic RNA
Is he in heaven, is he in hell?
That damned, elusive Pimpernel.
In the early spring of 1960, Jacob flew to Caltech to work with Matthew Meselson to trap the “damned, elusive Pimpernel.” Brenner arrived in early June, a few weeks later.
Proteins, Brenner and Jacob knew, are synthesized within a cell by a specialized cellular component called the ribosome. The surest means to purify the messenger intermediate was to halt protein synthesis abruptly—using a biochemical equivalent of a cold shower—and purify the shivering molecules associated with the ribosomes, thereby trapping the elusive Pimpernel.
The principle seemed obvious, but the actual experiment proved mysteriously daunting. At first, Brenner reported, all he could see in the experiment was the chemical equivalent of thick “California fog—wet, cold, silent.” The fussy biochemical setup had taken weeks to perfect—except each time the ribosomes were caught, they crumbled and fell apart. Inside cells, ribosomes seemed to stay glued together with absolute equanimity. Why, then, did they degenerate outside cells, like fog slipping through fingers?
The answer appeared out of the fog—literally. Brenner and Jacob were sitting on the beach one morning when Brenner, ruminating on his basic biochemistry lessons, realized a profoundly simple fact: their solutions must be missing an essential chemical factor that kept ribosomes intact within cells. But what factor? It had to be something small, common, and ubiquitous—a tiny dab of molecular glue. He shot up from the sand, his hair flying, sand dribbling from his pockets, screaming, “It’s the magnesium. It’s the magnesium.”
It was the magnesium. The addition of the ion was critical: with the solution supplemented with magnesium, the ribosome remained glued together, and Brenner and Jacob finally purified a minuscule amount of the messenger molecule out of bacterial cells. It was RNA, as expected—but RNA of a special kind.I The messenger was generated afresh when a gene was translated. Like DNA, these RNA molecules were built by stringing together four bases—A, G, C, and U (in the RNA copy of a gene, remember, the T found in DNA is substituted for U). Notably, Brenner and Jacob later discovered the messenger RNA was a facsimile of the DNA chain—a copy made from the original. The RNA copy of a gene then moved from the nucleus to the cytosol, where its message was decoded to build a protein. The messenger RNA was neither an inhabitant of heaven nor of hell—but a professional go-between. The generation of an RNA copy of a gene was termed transcription—referring to the rewriting of a word or sentence in a language close to the original. A gene’s code (ATGGGCC . . .) was transcribed into an RNA code (AUGGGCC . . .).
The process was akin to a library of rare books that is accessed for translation. The master copy of information—i.e., the gene—was stored permanently in a deep repository or vault. When a “translation request” was generated by a cell, a photocopy of the original was summoned from the vault of the nucleus. This facsimile of a gene (i.e., RNA) was used as a working source for translation into a protein. The process allowed multiple copies of a gene to be in circulation at the same time, and for the RNA copies to be increased or decreased on demand—facts that would soon prove to be crucial to the understanding of a gene’s activity and function.
But transcription solved only half the problem of protein synthesis. The other half remained: How was the RNA “message” decoded into a protein? To make an RNA copy of a gene, the cell used a rather simple transposition: every A,C,T, and G in a gene was copied to an A, C, U, and G in the messenger RNA (i.e., ACT CCT GGG→ACU CCU GGG). The only difference in code between the gene’s original and the RNA copy was the substitution of the thymine to a uracil (T→U). But once transposed into RNA, how was a gene’s “message” decoded into a protein?
To Watson and Crick, it was immediately clear that no single base—A, C, T, or G—could carry sufficient genetic message to build any part of a protein. There are twenty amino acids in all, and four letters could not specify twenty alternative states by themselves. The secret had to be in the combination of bases. “It seems likely,” they wrote, “that the precise sequence of the bases is the code that carries the genetical information.”
An analogy to natural language illustrates the point. The letters A, C, and T convey very little meaning by themselves, but can be combined in ways to produce substantially different messages. It is, once again, the sequence that carries the message: the words act, tac, and cat, for instance, arise from the same letters, yet signal vastly different meanings. The key to solving the actual genetic code was to map the elements of a sequence in an RNA chain to the sequence of a protein chain. It was like deciphering the Rosetta Stone of genetics: Which combination of letters (in RNA) specified which combination of letters (in a protein)? Or, conceptually:
Through a series of ingenious experiments, Crick and Brenner realized that the genetic code had to occur in a “triplet” form—i.e., three bases of DNA (e.g., ACT) had to specify one amino acid in a protein.II
But which triplet specified which amino acid? By 1961, several laboratories around the world had joined the race to decipher the genetic code. At the National Institutes of Health in Bethesda, Marshall Nirenberg, Heinrich Matthaei, and Philip Leder used a biochemical approach to try to crack the cipher. An Indian-born chemist, Har Khorana, supplied crucial chemical reagents that made code breaking possible. And a Spanish biochemist in New York, Severo Ochoa, launched a parallel effort to map the triplet code to corresponding amino acids.
As with all code breaking, the work proceeded misstep by misstep. At first, one triplet seemed to overlap with another—making the prospect of a simple code impossible. Then, for a while, it seemed that some triplets did not work at all. But by 1965, all of these studies had successfully mapped every DNA triplet to a corresponding amino acid. ACT, for instance, specified the amino acid Threonine. CAT, in contrast, specified a different amino acid—Histidine. CGT specified Arginine. A particular sequence of DNA—ACT-GAC-CAC-GTG—was therefore used to build an RNA chain, and the RNA chain was translated into a chain of amino acids, ultimately leading to the construction of a protein. One triplet (ATG) was the code to start the building of a protein, and three triplets (TAA, TAG, TGA) represented codes to stop it. The basic alphabet of the genetic code was complete.
The flow of information could be visualized simply:
Or, at a conceptual level:
Or:
Francis Crick called this flow of information “the central dogma” of biological information. The word dogma was an odd choice (Crick later admitted that he never understood the linguistic implications of dogma, which implies a fixed, immutable belief)—but the central was an accurate description. Crick was referring to the striking universality of the flow of genetic information throughout biology. From bacteria to elephants—from red-eyed flies to blue-blooded princes—biological information flowed through living systems in a systematic, archetypal manner: DNA provided instructions to build RNA. RNA provided instructions to build proteins. Proteins ultimately enabled structure and function—bringing genes to life.
Perhaps no illness illustrates the nature of this information flow, and its penetrating effects on human physiology, as powerfully as sickle-cell anemia. As early as the sixth century BC, ayurvedic practitioners in India had recognized the general symptoms of anemia—the absence of adequate red cells in blood—by the characteristic pallor of the lips, skin, and fingers. Termed pandu roga in Sanskrit, anemias were further subdivided into categories. Some variants of the illness were known to be caused by nutritional deficiencies. Others were thought to be precipitated by episodes of blood loss. But sickle-cell anemia must have seemed the strangest—for it was hereditary, often appeared
in fits and starts, and was accompanied by sudden, wrenching bouts of pain in the bones, joints, and chest. The Ga tribe in West Africa called the pain chwechweechwe (body beating). The Ewe named it nuiduidui (body twisting)—onomatopoeic words whose very sounds seemed to capture the relentless nature of a pain that felt like corkscrews driven into the marrow.
In 1904, a single image captured under a microscope provided a unifying cause for all these seemingly disparate symptoms. That year, a young dentistry student named Walter Noel presented to his doctor in Chicago with an acute anemic crisis, accompanied by the characteristic chest and bone pain. Noel was from the Caribbean, of West African descent, and had suffered several such episodes over the prior years. Having ruled out a heart attack, the cardiologist, James Herrick, assigned the case rather casually to a medical resident named Ernest Irons. Acting on a whim, Irons decided to look at Noel’s blood under the microscope.
Irons found a bewildering alteration. Normal red blood cells are shaped like flattened disks—a shape that allows them to be stacked atop each other, and thus move smoothly through networks of arteries and capillaries and veins, bringing oxygen to the liver, heart, and brain. In Noel’s blood, the cells had morphed, mysteriously, into shriveled, scythe-shaped crescents—“sickle cells,” as Irons later described them.
But what made a red blood cell acquire a sickle shape? And why was the illness hereditary? The natural culprit was an abnormality in the gene for hemoglobin—the protein that carries oxygen and is present abundantly in red cells. In 1951, working with Harvey Itano at Caltech, Linus Pauling demonstrated that the variant of hemoglobin found in sickle cells was different from the hemoglobin in normal cells. Five years later, scientists in Cambridge pinpointed the difference between the protein chain of normal hemoglobin and “sickled” hemoglobin to a change in a single amino acid.III