by Thomas Hager
Their office became an unofficial communication center between Cambridge and Pasadena. Peter and Donohue were both in correspondence with Pauling, and their office talk provided Crick and Watson with at least a small idea of what Pauling was up to. He wrote his son, for instance, that he was working hard on a structure for natural keratin made of helix ropes. When Peter told Crick, Crick's first thought was that Pauling had stolen his idea after he had let it slip during their car ride. He immediately refocused his own efforts. Within a month he had solved the last mathematical problems and on October 22 sent a short note to Nature describing the general outlines of his idea. It arrived just a few days after Pauling's longer manuscript on the same subject reached the same journal.
Crick's piece, however, arrived with a cover note requesting expeditious publication and, since notes are generally published more quickly than full papers in any case, reached print well before Pauling's. There was grumbling on both sides of the Atlantic for a few weeks about who had priority until the Cavendish crew conceded that Pauling's set of ideas went significantly beyond Crick's. Pauling, unsurprised by Crick's work when it appeared, was only irritated that Nature did not publish the two papers together. Finally, a gentleman's agreement was reached that the two men had come to the same point independently.
That was minor. The important fact was that the final major hurdle to acceptance of the alpha helix had been cleared away. By late fall Pauling was certain that his structure, with its various coiled-coil permutations, was the basic substance in hair, horn, and fingernail. It gave him a new idea about the structure of feathers as well. Evidence that the alpha helix existed in a number of globular proteins continued to mount. By late November, Pauling felt that his belief in the alpha helix had been fully confirmed. The alpha helix had been shown to be a principal structural feature in hemoglobin, serum albumin, insulin, pepsin, lysozyme, and a dozen other globular proteins, he happily wrote a colleague. "In fact, all of the globular proteins that have been investigated have been found to contain the alpha helix as their principal structural feature."
A Beautiful Structure
But the alpha helix was no longer the prize it once seemed. It was a breakthrough in methodology, a vindication of Pauling's stochastic approach, an important part of bigger things, but in the end it was only, as Pauling said, "a structural feature." The alpha helix appeared to be a stable arrangement that polypeptide chains assumed to get from point A to point B. As the molecular biologist Gunther Stent later put it, "No matter how great Pauling's triumph was, the discovery of the alpha helix did not immediately suggest to anyone very many new ideas about proteins, about how they work or are made. It did not seem to lead to very many new experiments, or to open new vistas to the imagination."
The real prize, the true secret of life, Pauling now knew, was DNA, and it was here that he next turned his attention.
On November 25, 1952, three months after returning from England, Pauling attended a Caltech biology seminar given by Robley Williams, a Berkeley professor who had done some amazing work with an electron microscope. Through a complicated technique he was able to get images of incredibly small biological structures. Pauling was spellbound. One of Williams's photos showed long, tangled strands of sodium ribonucleate, the salt of a form of nucleic acid, shaded so that three-dimensional details could be seen. What caught Pauling's attention was the obvious cylindricality of the strands: They were not flat ribbons; they were long, skinny tubes. He guessed then, looking at these black-and-white slides in the darkened seminar room, that DNA was likely to be a helix. No other conformation would fit both Astbury's x-ray patterns of the molecule and the photos he was seeing. Even better, Williams was able to estimate the sizes of structures on his photos, and his work showed that each strand was about 15 angstroms across. Pauling was interested enough to ask him to repeat the figure, which Williams qualified by noting the difficulty he had in making precise measurements. The molecule Williams was showing was not DNA, but it was a molecular cousin—and it started Pauling thinking.
The next day, Pauling sat at his desk with a pencil, a sheaf of paper, and a slide rule. New data that summer from Alexander Todd's laboratory had confirmed the linkage points between the sugars and phosphates in DNA; other work showed where they connected to the bases. Pauling was already convinced from his earlier work that the various sized bases had to be on the outside of the molecule; the phosphates, on the inside. Now he knew that the molecule was probably helical. These were his starting points for a preliminary look at DNA. He did not know how far he would get with this first attempt at a structure, especially because he still had no firm structural data on the precise sizes and bonding angles of the base-sugar-phosphate building blocks of DNA, but it was worth a look.
Pauling quickly made some calculations to determine DNA's molecular volume and the expected length of each repeating unit along its axis. Astbury's photos showed a strong reflection at 3.4 angstroms—according to Pauling's calculations, about three times his estimated length of a single nucleotide unit along the fiber. Repeating groups of three different nucleotides seemed unlikely; a threefold chain structure would explain the repeat more easily. His density calculations indicated that three chains would need to pack together tightly to fit the observed volume, but that was all right. In crystallography, the tighter the packing, the better. After five lines of simple calculations on the first page of his attack on DNA, Pauling wrote, "Perhaps we have a triple-chain structure!"
He was immediately captivated by the idea: three chains wound around one another with the phosphates in the middle. Sketching and calculating, he quickly saw that there was no way for hydrogen bonds to form along the long fiber axis, holding the windings of the chain in place, as in the alpha helix. Without them, what held the molecule in shape? One place that hydrogen bonds could form, he saw, was across the middle of the molecule, from phosphate to phosphate. That was a surprise, but everything else seemed to be working out. After six pages of calculations, he wrote, "Note that each chain has . . . roughly three residues per turn. There are three chains closely intertwined, and held together by hydrogen bonds between P04's." The only problem was that there did not seem to be quite enough space in the center of the molecule, where the phosphates came into closest contact. He put down his pencil for the night.
Three days later, he came back to the problem. According to Astbury's figures, DNA was a relatively dense molecule, which implied tight packing at the core. But trying to jam three chains' worth of phosphates into Astbury's space restrictions was like trying to fit the stepsisters' feet into Cinderella's glass slipper. No matter how he twisted and turned the phosphates, they wouldn't fit. "Why are the P04 in a column so close together?" he wrote in frustration. If Astbury's estimates on distances could be relaxed a bit, everything would fit, but Pauling could not do that without deviating too far from Astbury's x-ray data. Pauling next tried deforming the phosphate tetrahedra to make them fit, shortening some sides and lengthening others. It looked better, but still not right. He stopped again.
Next, he had an assistant go back through the literature in the chemistry library and pick up everything he could find on the x-ray crystallography of nucleic acids. There was not much to go on besides Astbury's work and that of Sven Furberg, a Norwegian crystallographer who had studied under Bernal and had found that the bases in DNA were oriented at right angles to the sugars. There was not one detailed structure of any purine or pyrimidine, much less a nucleotide.
On December 2 he made another assault, filling nine pages with drawings and calculations. And, he thought, he came up with something that looked plausible. "I have put the phosphates as close together as possible, and have distorted them as much as possible," he noted. Even though some phosphate oxygens were jammed uncomfortably close in the molecule's center, not only did it all just fit, but Pauling saw that the innermost oxygens packed together in the form of an almost perfect octahedron, one of the most basic shapes in crystallography. It was very
tight, but things were lining up nicely. It had to be right. It had been less than a week since he first sat down with the problem.
The next day, Pauling excitedly wrote a colleague, "I think now we have found the complete molecular structure of the nucleic acids." During the next several weeks he ran downstairs every morning from his first-story office in Crellin to Verner Schomaker's office in the first basement, "very enthusiastic," Schomaker remembered, bouncing ideas off the younger man, thinking aloud as he checked and refined his model. He began working with Corey to pinpoint the fine structure.
Then came trouble. Corey's detailed calculation of atomic positions showed that the core oxygens were, in fact, too close to fit. In early December, Pauling went back to twisting and squeezing the phosphate tetrahedra. Someone brought up the question of how his model allowed for the creation of a sodium salt of DNA, in which the positive sodium ions supposedly adhered to the negative phosphates. There was no room for sodium ions in his tightly packed core, was there? Pauling had to admit he could find no good way to fit the ions. But that would sort itself out later. The other results were positive. Running the proposed structure through Crick's mathematical formula indicated that his model helix would fit most of the x-ray data, although not all of it. Schomaker played with some models on his own and found a way to twist the phosphate tetrahedra so that they were not quite so jammed, but for the moment Pauling saw no reason to change his ideas. The core phosphates were too neatly close-packed not to be true.
And this was what the central problem had reduced itself to in his mind: a question of phosphate structural chemistry. The biological significance of DNA would be worked out later, he thought; if the structure was right, the biological importance would fall out of it naturally in some way. At this point it was his business to get the structure, not the function. So he ignored the larger context surrounding the molecule and focused single-mindedly on one thing: finding a way to fit those phosphates into the core so that the resulting helixes fit the available data.
His faith in that approach had been justified by his success with the alpha helix. He had built his protein spiral from strict chemical principles, published it in the face of contradictory data, and later found the facts he needed to answer his critics. He was confident now about his ability to jump ahead of the pack, to use his intuitive grasp of chemistry to tease out a structure that felt right. If you waited for every doubt to be answered first, you would never get credit for any discovery. And his DNA triple helix felt right.
A week before Christmas, he wrote Alex Todd at Cambridge, "We have, we believe, discovered the structure of nucleic acids. I have practically no doubt. . . . The structure really is a beautiful one." Pauling knew that Todd had been working with purified nucleotides and asked him to send samples for x-ray analysis. "Dr. Corey and I are much disturbed that there has been no precise structure determination reported as yet for any nucleotide. We have decided that it is necessary that some of the structure determinations be made in our laboratory. I know that the Cavendish people are working in this field, but it is such a big field that it cannot be expected that they will do the whole job." He then wrote his son Peter and Jerry Donohue that he was hoping soon to complete a short paper on nucleic acids.
But the structure still was not quite right. Everything would seem to fall into place when Corey came up with another set of calculations showing that the phosphates were packed just a little too tightly, their atoms jostling each other a little too closely to be reasonable. Pauling would readjust and tinker, bend and squash, so close to the answer yet unable to make it all fit perfectly.
- - -
He was becoming frustrated with it when another distraction cropped up: On December 23, professional FBI informer and darling of the congressional investigating committees Louis Budenz testified publicly, before a House special committee investigating charitable foundations, that Pauling, a member of the advisory board of the Guggenheim Foundation, was a concealed Communist. Budenz outdid himself, pouring out the names of twenty-three grantees of various organizations and three other officials, most of whom had no more to do with communism than did Pauling. His testimony would enrage a number of influential people associated with powerful foundations and eventually help spur a backlash against McCarthyism, but in the short term the timing of the announcement—two days before Christmas, at a time when the news media would be hungry for headlines but without the staff to do follow-up—did maximum damage to those named with little chance for response.
Pauling, who had for the most part abided by his decision a year before to pull out of active politics, felt as if he had been sucker punched. His response was characteristically straightforward. "That statement is a lie," he told the press. "If Budenz is not prosecuted for perjury, we must conclude that our courts and Congressional committees are not interested in learning and disclosing the truth." When he discovered that Budenz was not liable for perjury because his testimony was protected by congressional privilege, Pauling tried another tack to get his accuser into court, calling Budenz a "professional liar" in the press in hopes that Budenz would sue him. Budenz did not take the bait.
- - -
Depressed about this unexpected political attack, Pauling took the unusual step of inviting some colleagues into his laboratory on Christmas Day to have a look at his work on DNA. He was tired of the niggling problems with his model and ready for some good news. He got it from his small audience, who expressed enthusiasm for his ideas. Much cheered, Pauling spent the last week of the year working with Corey on the finalization of a manuscript.
On the last day of December 1952, Pauling and Corey sent in their paper, "A Proposed Structure for the Nucleic Acids," to the Proceedings of the National Academy of Sciences. This was, they stressed, "the first precisely described structure for the nucleic acids that has been suggested by any investigator"—thus positioning the work as the nucleic acid equivalent to the alpha helix. He went through his reasoning for the core structure. Most of the paper concentrated on precisely stacking phosphate tetrahedra, but there was a little biology, too. In Pauling's model, the bases, the message-carrying portion of nucleic acids, were directed outward, like leaves along a stalk, with room enough to be put into any order, providing maximum variability in the molecule and thus maximum specificity in the message. Astbury had already noted that the 3.4-angstrom repeat in nucleic acid was about the same as the distance per amino acid along an extended polypeptide chain, raising the idea that new proteins might be struck directly off a nucleic acid mold. Pauling noted that his model allowed the same thing to happen, with the sides of four adjacent bases along his chains forming a space just right for fitting an amino acid.
There was, however, an uncharacteristic tentativeness in the piece. This was "a promising structure," Pauling wrote, but "an extraordinarily tight one"; it accounted only "moderately well" for the x-ray data and gave only "reasonably satisfactory agreement" with the theoretical values obtained by the Crick formula; the atomic positions, he wrote, were "probably capable of further refinement."
- - -
It was, in fact, a rush job. Pauling knew that DNA was important; he knew that Wilkins and Franklin were after it and that Bragg's group had already made at least one stab at it. He knew that it was a relatively simple structure compared to proteins. And he knew that whoever got out a roughly correct structure first—even if it was not quite right in all its details—would establish priority. That is what he was aiming for, not the last word on DNA but the first, the initial publication that would be cited by all following. It did not have to be precise. He wanted credit for the discovery.
The hurried haphazardness of the nucleic-acid paper can best be understood by comparison to Pauling's protein work. Pauling's alpha helix was the result of more than a decade of off-and-on analysis and thousands of man-hours of meticulous crystallographic work. Before he published his model, his lab pinned down the structure of the amino-acid subunits to a fraction of a degree and a hundred
th of an angstrom. There was an abundance of clean x-ray work available on the subject proteins, allowing Pauling to scrutinize and eliminate dozens of alternative structures. Two years passed between the time he came up with the rough idea for his helix and the time he published it. Much of that interval was spent with Corey, overseeing and refining the precise construction of a series of elaborate three-dimensional models.
None of that went into DNA.
"The only doubt I have …"
Crick and Watson were downcast by the news from Peter in late December that Pauling had solved DNA. Alternating between bouts of despair and denial—trying to figure out how he could have beaten them and then deciding that he certainly could not have without seeing Wilkins and Franklin's x-ray work and then thinking, well, of course, he is Pauling, so anything is possible—they continued working on the problem themselves. If they could come up with something independently before Pauling's paper appeared, at least they might share credit.
The previous spring, a few months after they had been warned off DNA and a few months before Pauling's visit to the Cavendish, Crick and Watson had been introduced to Erwin Chargaff, the acerbic and opinionated Austrian-born biochemist who had been using chromatography to analyze the chemical composition of nucleic acids. Chargaff was not impressed. "I never met two men who knew so little and aspired to so much," he said. "They told me they wanted to construct a helix, a polynucleotide to rival Pauling's alpha helix. They talked so much about 'pitch' that I remember I wrote it down afterwards, 'Two pitchmen in search of a helix.'" But this conversation was critical to Crick and Watson. Chargaff told them that there was a simple relationship between the occurrence of different bases in DNA, that adenine and thymine were present in roughly the same amounts and so were guanine and cytosine. One of each pair was a larger purine; the other, a smaller pyrimidine. It was the same relationship that he had told Pauling about during their Atlantic crossing in 1947 and that Pauling had ignored.