by James Watson
Noon seminars from September 1967
Frank Stahl (back to camera), Gerald Seher, Mark Ptashne, Wally Gilbert, and Celia Gilbert at the 1968 Cold Spring Harbor symposium
Wally Gilbert and Max Gottesman at Cold Spring Harbor in 1968
By then Benno had returned to Germany as a professor at the Genetics Institute of the University of Cologne. Before going, he again made elegant use of E. coli genetics to generate a mutant that overproduced the lac repressor. Putting this mutant gene on a λ-like phage led to hundredfold amounts of the typical lac repressor output. Just before Benno left Harvard, Jacques Monod was visiting and popped his head in to say hello and comment on Benno's identification of the lac repressor, remarking, “After all, Benno, it was pedestrian.” Then he turned around and left. It was Jacques's sour grapes way of admitting that clever experiments, not a clever idea, had beaten him and his friends at the Institut Pasteur. With Benno, Wally, and Mark's success, the center of gene regulation was now not Paris but Harvard.
An unanticipated by-product of Wally's lac repressor purification effort was helping my graduate student Dick Burgess to optimize his RNA polymerase purification methodologies. More than ever, our lab wanted further insight into how RNA is made from DNA templates. In 1963, I'd had my graduate student John Richardson focus his Ph.D. research on RNA polymerase, the enzyme responsible for making RNA off DNA. RNA polymerase was discovered in 1959, but its molecular form was not easy to pin down. John found that the molecular weight decreases by half when the molecule is suspended in a highly ionized solution. But even this smaller form was larger than any known single polypeptide chain, so it seemed likely that RNA polymerase was constructed from a number of smaller polypeptides. No deep insights, however, came from electron microscopy done at MIT. It revealed approximately spherical particles 120 Ä in diameter, compatible with a molecular weight of almost a million atomic mass units. Just before leaving for Paris to become a postdoc at the Institut Pasteur, John used phage T7 templates to demonstrate the synthesis of RNA products containing up to ten thousand bases. Conceivably their long lengths reflected failure of the purified enzyme to recognize normal T7 stop signals for RNA synthesis.
After John left for France, Dick Burgess, who had just arrived from Caltech, took up the challenge presented by RNA polymerase's large size. Every week Wally's technician Chris Weiss separated the proteins from thirty liters of E. coli cells into several fractions. The one containing the lactose repressor went to Wally and Benno, while Dick took the fraction containing RNA polymerase. The latter still contained many, many other proteins, requiring him to work out a glycerol gradient centrifugation procedure that yielded a highly active RNA polymerase preparation that he called GG. Starting late in 1967, he added a phos-phocellulose column step to his recipe, leading to even purer RNA polymerase, called PC RNA polymerase.
In April 1968, Dick gave a talk at the Federation Society meeting in Atlantic City revealing that PC RNA polymerase particles were made up of small subunits, called a, and large subunits, called ß. Two months later, at the Gordon Conference on Nucleic Acids in New Hampshire, he gave out details of his GG and PC purification methods, later sending written protocols to some ten interested labs, including that of Ekke Bautz at Rutgers. By then Wally's graduate student Jeff Roberts was also studying RNA polymerase. Initially using Dick's GG purified enzyme, he found it very active in transcribing λ DNA. But later, using Dick's even more purified PC preps, he inexplicably found no transcription of λ DNA.
Jeff's unexpected negative result made Dick wonder whether his PC purification procedure had removed an RNA polymerase component necessary for transcription of phage DNA. To test this possibility, he passed a fresh batch of GG enzyme over a PC column to see whether this again led to a loss of phage DNA's transcription ability. After getting the same negative result as Jeff, he showed that the PC enzyme could be activated by adding back a protein that had been extracted in purification. Experiments done the following week by our English postdoc Andrew Travers demonstrated that a single polypeptide chain, soon to be called σ factor, was the GG-enzyme-containing ingredient lacking in PC preps.
Less than a week passed before Dick heard that Ekke Bautz at Rutgers and his student John Dunn had obtained the same results using GG and PC enzymes following the protocol he had sent them in July. At my suggestion, Dick immediately contacted Bautz to see whether they would like their work included in a joint lab paper on the σ factor to be submitted to Nature. Liking the idea, they came up to Harvard on November 6. For me their presence was also a distraction from the unbearable cliffhanger of the presidential elections that day. Richard Nixon's victory over Hubert Humphrey would become clear only late that night. Even the discovery of σ factor did not alleviate the gloom I felt the rest of the week at the prospect of a Nixon presidency.
At the 1970 Cold Spring Harbor symposium: Jeff Roberts with Ann Burgess (left) and John Richardson with Dick Burgess (right)
On Monday morning, November 11, a silver lining briefly presented itself. In that day's Crimson I read the supersize headline “Pusey to Quit Harvard.” The accompanying article reported that Presidentelect Nixon had paid an unannounced visit to Harvard, coming secretively at 9:00 P.M. the night before, to ask Pusey to become his Postmaster General. After a frank and comradely conversation, during which tea and vanilla wafers were served, Pusey accepted later declared to be “the greatest moment of my life.” All too soon, however, the happy glow permeating my being dissipated. Lower on the page was another story reporting that the proposed JFK Library was to be moved from Cambridge to Bayonne, New Jersey, making me realize I was reading one of the hoax issues Crimson editors occasionally created, perhaps to assure themselves that no want of wit had consigned them to the Crimson rather than the Lampoon. After this monumental letdown nothing in the Crimson ever seemed funny.
When the manuscript announcing σ factor was submitted to Nature on December 2, it contained only data obtained at Harvard. The equivalent experiments from Rutgers had been done fewer times and were less complete. “Factor Stimulating Transcription by RNA Polymerase,” by Burgess, Travers, Dunn, and Bautz, speedily appeared as a full article in January 1969. Before publication, I announced our lab's breakthrough in a lecture to Arthur Kornberg's perennially self-congratulatory Biochemistry Department at Stanford. It was a moment I never dreamed would come: our Harvard Biolabs demonstrating the biochemical competence to take on and best mighty Stanford.
In mid-December, Dick and Andrew had heard from Ekke Bautz and John Dunn that σ molecules disappear from cells infected with T4 phage, explaining why host bacterial RNA synthesis stops soon after phage infection. Phage DNA molecules likely coded for phage-specific σ factors that directed core RNA polymerase molecules to specific signals on their respective DNA. Over the next six months, Andrew Travers confirmed this conjecture and sent a paper to Nature in August called “Bacteriophage Sigma Factors for RNA Polymerase.” In its conclusion, he speculated that σ-like transcription factors might be responsible for the massive shifts in RNA transcription patterns underlying the development of higher organisms. Sensing again a major paper on their hands, Nature editors rushed it into print in just over a month.
By early February, Dick had defended his thesis and started a Helen Hay Whitney postdoctoral fellowship that would let him remain at Harvard until his intelligent, blond wife, Ann, finished her Ph.D. thesis. She hailed from a prosperous, hardworking Wisconsin family and did her experiments on bacteriophage ΦX174 down the hall in Dave Denhardt's lab. By then, Dick knew that not one but two ß chains existed in the core (PC) RNA polymerase particle whose structure was α2ββ He and Andrew, furthermore, had preliminary evidence pointing to how σ functioned in the initiation of RNA synthesis. Its role was to direct the core α2ββ1 complex to appropriate starting sites to the DNA for RNA synthesis. RNA polymerase's enzymatic property is solely owing to its α2;ββ1 core component.
The relative importance in controlling gene regulation of repr
essore and operators versus σ factor shifts still remained open. Bernie Davis at Harvard Medical School bet me a case of wine that my group would not discover a second bacterial σ factor over the next two years. Andrew then had let Bernie know of his tentative evidence for a σ factor controlling ribosomal RNA synthesis. Here time was not on his side, costing me a case of cabernet. But Richard Losick, a newly appointed junior fellow, had begun experiments in the Biolabs on how Bacillus subtilis forms spores, soon getting hints of a possible σ factor specific to bacterial sporulation.
The first visible recognition of σ factor's importance came in November 1969 at a meeting in Florence, Italy, on RNA polymerase and transcription. Dick Burgess gave the opening talk, coming from Geneva, where he was now a postdoc in Alfred Tissières's lab. Sponsoring the meeting was Lepetit, the Milan drug company whose rifamycin and rifampicin antibiotics had been shown to inhibit bacteria through binding to the ß subunit of RNA polymerase. At this meeting, Jeff Roberts announced his recent discovery of a protein called p that halts RNA synthesis at specific stop signals on DNA molecules. Just before coming to Italy, he sent off his manuscript “Termination Factors for RNA Polymerase” to Nature, which published it in December 1969.
Though the low sun was not optimal for picture viewing in the Uffizi Gallery, the Palazzo Vecchio provided a reception site equal in its satisfactions to the science of the three-day gathering. Even being forced to listen to a minister up from Rome could not diminish the pleasure of knowing that the Biolabs were still at the center of how genes are regulated.
Remembered Lessons
1. Two obsessions are one too many
Experiments, like many speculative enterprises, are likely to require at least five times more effort than you initially guess. Being a really good anything—be it university president, violinist, securities lawyer, or a scientist—requires a virtually obsessive devotion to one's objectives. Dividing one's attention will give the edge to competitors who have the same talent but greater focus. For this reason, highly successful bankers who also claim to be accomplished cellists are often neither. Their banking reputation likely rests on the labors of talented associates working day and night, and their cello playing as likely suffers from the time lost to even the pretense of being a banker.
2. Don't take up golf
The moment golf clubs are first spotted in your trunk, you will be subject to constant ribbing. Only the rare few content to play occasionally with no fantasy of breaking 90 should even consider hitting the links. Once you become obsessive about bettering your personal best—say, now 94—your weekend science experiments cease. You have become a thank-God-it's-Friday scientist, always fighting not to fall too embarrassingly behind those peers who have sensibly chosen the less Zen but more aerobic thrill of hitting tennis balls.
3. Races within the same building bring on heartburn
A serious competitor aiming for the same objective inherently creates anxiety. It is emotionally draining to wish ill upon anyone constantly nearby, yet few, if any, human psyches are capable of turning off the visceral wish for an opponent to stumble. In science, the journey is not the destination; the destination is the destination. And so it is better that one's competitors be in a different city, if not country. Having them in the same building is a small model of hell, and not even an efficient one. Once Jacques Monod had two research teams working in adjacent labs attempt to make ß-galactosidase in the test tube. Maybe he knew their respective leaders already disliked each other. Anyway, in that spring 1962 race, no one crossed the finish line.
4. Close competitors should publish simultaneously
Science works better when the winners do not take all. The agony of losing a very close race may break the spirit of a competitor who may again bring out the best in you. And so when you beat someone across the line by only a nose, offer to publish at the same time, if not back to back, in the same journal. Those scorekeepers in the know usually are aware what happened and will think more highly of you. Doing unto others may also yield reciprocal benefits the next time you are oh-so-close.
5. Share valuable research tools
Do not hog powerful new research tools or reagents. If Dick Burgess had not shared his GG and PC RNA polymerase preps with his lab neighbor Jeff Roberts, he unlikely would have been the first to discover σ factor. One hand washes the other. Though it was sharing his RNA polymerase protocols with Ekke Bautz and John Dunn that brought them into his game, they later immediately shared their discovery that σ factor disappears following phage T4 infection. This openness further helped Andrew Travers get an early start in the hunt for phage-specific σ factors.
14. MANNERS FOR HOLDING DOWN TWO JOBS
IN THE FALL of 1967, Harvard gave me permission to become the director of the Cold Spring Harbor Laboratory while remaining a full-time member of the faculty. They did so upon realizing that Cold Spring Harbor's precious research and educational resources were on the brink of disaster and would likely disappear unless someone stepped in to make this unique Long Island institution financially viable. I was already a member of its board of trustees and knew its shaky state through my close friendship with its sharp-minded director, John Cairns. Born and bred in North Oxford, John exuded an ironic intelligence as well as an inability to seek help from individuals who had more power than warranted by the agility of their brains.
Since arriving from Australia in July 1963, John increasingly had come to see the biochemist Ed Tatum, chairman of his Board of Trustees, as a personal nemesis. On paper he seemed an asset. Tatum, then a professor at Rockefeller, had done research at Stanford in the 1940s on gene-protein relationships that led to his sharing the 1958 Nobel Prize in Physiology or Medicine with the Nebraska-born geneticist George Beadle. But Tatum was a polite plodder who would have gone nowhere but for Beadle, and later at Yale he was propped up by his graduate student and protege Joshua Lederberg. The sharp intellectual crossfire of the geneticists and molecular biologists who spent summers at Cold Spring Harbor was not Tatum's cup of tea. Prior to his becoming chairman, he had attended only one summer symposium. Much to John Cairns's annoyance, Tatum arranged for all trustee gatherings to be held in New York City at Rockefeller University. Avoiding the thirty-mile trip east, the chairman spared himself and the other trustees having to see the decrepit state of the some twenty-five buildings on the Lab's almost one-hundred-acre campus. At board meetings, Tatum's demeanor reminded me ofthat of Nathan Pusey Neither knew how to handle individuals who presented unwanted facts.
John Cairns (center) at the 1968 symposium
By the time I joined the board, John operated in day-to-day uncertainty, even though he had converted the lab's $50,000 negative cash balance to a surplus of some $100,000. Three years of hard physical and mental effort had been required, including his oft cutting the Lab's many green lawns himself. At the same time, the Lab's endowment remained effectively zero, and its survival depended upon the success of its small number of scientists each obtaining one to several significant research grants, which supported not only their science but also the budget for administration, facilities maintenance, and the like. Apart from this grant income, the only other major barriers to insolvency were a few corporate sponsorships and the ever-increasing sales of the lab's annual symposium report, a must-have volume for anyone in molecular biology.
By mid-1966, John began to talk about quitting, and such talk only increased as Tatum showed himself entirely unperturbed by it. But several key junior scientists took note and in turn began to seek jobs elsewhere. In early 1967, John submitted his resignation, effective upon the selection of a successor. By then it was cold comfort to John that Ed Tatum would be gone even sooner. Replacing him as chairman was an H. J. Muller-trained geneticist from Texas, Bentley Glass, whose association with the lab went back to the late 1930s. Recently he had moved from Johns Hopkins to become provost of the new Long Island campus of the State University of New York, at Stony Brook.
Bentley knew that he alone
could not dramatically improve the Lab's chances for survival. Though he was connected to innumerable government funding sources, no new monies would flow to Cold Spring Harbor until an appropriate new director was found. When I flew to New York City for the Lab's late October trustee meeting, I feared they might choose the German phage geneticist Carsten Bresch, then looking for an escape from an insecure job in Dallas. If Bresch were to come, he would continue the Lab's historically strong focus on molecular genetics. But I feared that he would see the job as a way station for him until a permanent, well-financed position was created in Germany.
Still, this was not an objection likely to sway my fellow trustees. Better a temporary leader than none, the others would counter. So I threw my hat into the ring, saying I would accept the directorship if I could simultaneously remain at Harvard as a full-time member of the faculty. In this way, the Lab trustees would be spared the need to find a source from which to pay the new director. Since his arrival, a five-year grant from the Rockefeller Foundation had covered John Cairns's $15,000 annual salary. But one could not count on an extension of this grant. As soon as I raised the possibility of my leading Cold Spring Harbor, further discussion about Bresch subsided. I sensed that if Harvard said yes, the job was mine.