The Tangled Tree

by David Quammen

  “My work had sort of come to a climax,” he said later, and he didn’t care to morph into an administrator. He declined a knighthood, having no desire to be addressed as “Sir Fred” by friends and strangers. “A knighthood makes you different, doesn’t it,” he said, “and I don’t want to be different.” But that Cincinnatus retirement lay long in the future when Carl Woese, in his 1969 letter to Crick, daydreamed of getting a Sanger protégé to help him.

  One of Sanger’s grad students had already come to Urbana, in fact, as a postdoc in the lab of another scientist within Woese’s department. That postdoc was David Bishop, brought over to assist Sol Spiegelman in sequencing viral RNA. Spiegelman had recruited Woese to the University of Illinois, rescuing him from obscurity at General Electric, in 1964. One year after Bishop’s arrival, Spiegelman left Illinois, returning to Columbia University in New York City, where his career had begun, and eventually taking Bishop with him. That might have yanked the Sanger techniques beyond Woese’s grasp. But in the interim months, Woese found a promising doctoral student named Mitchell Sogin and assigned him to learn what he could from Bishop before Bishop left. Molecular biology was in its formative phase, and though results could be announced in journal papers, the gritty details of lab methodology were often passed person to person, like the gift of stone tools or fire.

  Mitch Sogin was a bright Chicago kid who had come down to the University of Illinois as an undergraduate on a swimming scholarship, planning to do premed. The swimming ended, the allure of medicine faded, but Sogin stuck around to earn a master’s degree in industrial microbiology within the Department of Food Science, part of the College of Agriculture. He worked on bacteria—specifically, the germination of bacterial spores, a matter of some practical interest to the food industry, given the implications for human health. Carl Woese, inhabiting a different department, almost a different universe, happened to have a lingering interest in spore germination from studies earlier in his career. For that slim reason, someone sent young Sogin to meet him. They clicked.

  “And so I would go down and talk to him,” Mitch Sogin told me, almost fifty years later. “I liked him.”

  Sogin was seventy at the time of our conversation, with a face that looked youthful but was now framed by thick, white hair. Behind his glasses, with his diffident smile, he resembled a professorial Paul Simon. We sat in his third-floor office in an old redbrick building on Water Street in Woods Hole, Massachusetts, headquarters of the Marine Biological Laboratory, a venerable research institution, where Sogin held the position of senior scientist and director of a center for comparative molecular biology and evolution. He seemed slightly bemused to have ended up there at Woods Hole, studying microbial communities of the oceans, microbial communities of the human gut, and microbial stowaways on space vehicles bound for Mars, as I nudged him to recall his early encounters with Carl Woese, back in 1968.

  At that dicey moment in history, Sogin found himself, by age and geography, at the top of the rolls for his local Selective Service board. He hadn’t been drafted yet, but it seemed imminent, and this was before the first lottery made draft boards less arbitrary. “I had to make a sudden decision whether to stay in school or whether to go to Vietnam.” The war was at its ugliest; the Tet offensive in February that year had curdled the thinking of many young American males (including Mitch Sogin and me), and, unfair as it was, you could still get a deferment for graduate school. “Decided to stay in school,” Sogin told me. “It was simple.” He began work toward a doctorate under the mentoring of Woese. His topic was ribosomal RNA.

  Woese had noticed something about Mitch Sogin during their early interactions: the kid was not just smart but also handy around equipment. Some combination of talents—dexterity, mechanical aptitude, precision, patience, a bit of the plumber, a bit of the electrician—made him good not just at experimental work but also at creating the tools for such work. Sol Spiegelman had ordered and paid for a collection of apparatus to be used for RNA sequencing by the Sanger method; but now Spiegelman was off to Columbia, leaving behind the tools.

  “So Carl inherited that equipment. But he had no one that knew how to use it.” No one, that is, until Sogin joined his lab. “I was essentially responsible for importing all the technology”—importing it from Spiegelman’s lab, and other sources, into the Woese operation. Sogin learned as much as possible from Bishop about Fred Sanger’s techniques before Bishop decamped to New York, and then Sogin became Woese’s handyman as well as his doctoral student, assembling and maintaining an array of hardware to enable the sequencing of ribosomal RNA.

  Woese himself was not an experimentalist. He was a theorist, a thinker, like Francis Crick. “He never used any of the equipment in his own lab,” Sogin said. None of it—unless you count the light boxes for reading films. Sogin himself had built these fluorescent light boxes, on which the film images of RNA fragments, cast by radioactive phosphorus onto large X-ray negatives, could be examined. He had converted an entire wall of bookshelves, using translucent plastic sheeting and more fluorescent bulbs, into a single big, vertical light box, like a bulletin board. They called it the light board. Viewed over a box or taped up on the light board, every new film would show a pattern of dark ovals, like a herd of giant amoebae racing across a bright plain. This was the fingerprint of an RNA molecule. Recollections from his lab members at the time, as well as a few old photographs, portray Carl Woese gazing intently at those fingerprints, hour upon hour.

  “It was routine work, boring, but demanding full concentration,” Woese himself recalled later. Each spot represented a small string of bases, usually at least three letters but no more than about twenty. Each film, each fingerprint, represented ribosomal RNA from a different creature. The sum of the patterns, taking form in Carl Woese’s brain, represented a new draft of the tree of life.

  14

  The mechanics of this effort in Woese’s lab, during Mitch Sogin’s time and for much of the next decade, were intricate, laborious, and a little spooky. They involved explosive liquids, high voltages, radioactive phosphorus, at least one form of pathogenic bacteria, and a loosely improvised set of safety procedures. Every boy’s dream. Courageous young grad students, postdocs, and technical assistants, under a driven leader, were pushing their science toward points where no one, not even Fred Sanger or Linus Pauling, had gone before. The US Occupational Safety and Health Administration (OSHA), though recently founded, was none the wiser.

  The fundamental goal was to sequence variants of a molecule from the deepest core of all cellular life, compare those variants, and deduce the history of evolutionary relationships since the beginning. Woese had already settled on that one universal element of cellular anatomy, the ribosome, the machine that turns genetic information into proteins, but there remained a crucial decision: Which ribosomal molecule should he study? Ribosomes comprise two subunits, as I’ve mentioned—a small one snuggled beside a larger one, like an auricle and a ventricle of the heart, each constructed of both RNA and proteins. The RNA fractions include several distinct molecules of different lengths. At first, Woese targeted a short RNA molecule from the large subunit, known as 5S (“five-S”) for obscure reasons that I don’t ask you to contemplate. Just remember 5, a smallish number. That molecule proved unsatisfactory because its very shortness limited the amount of information it contained. The alphabet of nucleotides composing RNA is slightly different from that of DNA—it’s A, C, G, and U (for uracil) in place of T (for thymine)—and there was just not enough of the A-C-G-U alphabet in any little 5S sequence to distinguish different creatures from one another. So he switched to a longer molecule in the small subunit, and at the risk of causing your eyes to roll back in your head, I’m going to tell you its name. Why? Because it’s important, and once you’ve got it, you own it: 16S rRNA. There. Not so bad?

  In English we say: “sixteen-S ribosomal RNA.” It’s a structural component of every bacterium on Earth, and bacteria were what Woese studied initially.

&
nbsp; There’s a close variant, 18S rRNA, in the ribosomes of more complex creatures, such as animals and plants and fungi. This 16S molecule and its 18S variant, therefore, could serve as the reference standard, the great clue, for deducing divergence and relatedness among all cellular organisms. It was, arguably, the single most reliable piece of evidence, molecular or otherwise, for drawing a tree of life. And that recognition, though it never made the front page of the New York Times, was Carl Woese’s single greatest contribution to biology in the twentieth and twenty-first centuries.

  The immediate goal for Woese, back in the early 1970s, was to extract ribosomal RNA from different organisms, to learn as much as possible about the genetic sequence of the chosen rRNA molecule from each organism, and to make comparisons from which he could gauge degrees of relatedness. He started with bacteria, because many kinds of bacteria are easy to grow in a lab, and their collective history is very ancient. Looking at bacteria from numerous different families allowed him the prospect of seeing contrasts, even in such a slowly evolving molecule as 16S rRNA. He and his team proceeded by extracting ribosomal RNA from the bacterial cells, purifying samples of the 16S molecules in each, and cutting those molecules into variously sized fragments with enzymes. Then they separated the fragments by electrophoresis, using an electrical field and a racetrack of soaked paper or gel.

  In electrophoresis, a solution of mixed fragments is added to the racetrack, the power is turned on, and the electrical force pulls small fragments along faster than large ones, causing them to separate as distinct bands or ovals along the track. In Woese’s effort, each fragment comprised just a few of those A, C, G, U bases—maybe three, maybe five, maybe eight, maybe as many as twenty, but always a minuscule fraction of the full molecule. Those small fragments could then be pulled again, this time in a sideways direction, and their exact sequence would begin to come clear, based on the chemical and electrical differences among A, C, G, and U. Small fragments were easier to sequence by this method than one mammoth chain. AAG was easier to discern, as you might imagine, than AAUUUUUCAUUCG.

  There were several stages of work. The primary run began the process of separating the fragments from one another. The secondary run, in a sideways dimension, revealed more about each fragment, which grew discretely recognizable as it raced not just down the racetrack but also now across. Those fragments, because of their radioactive content, showed as ovals burned onto the X-ray films. The oval-marked films would let an expert interpreter such as Woese infer the sequences—that is, to sort the As, Cs, Gs, and Us from one another and determine their order in each fragment. Once illuminated that way, a fragment became more like a word than like a shadowy amoeba. It had its own spelling. What was the spelling of this little word, this fragment, or that one? Was it CAAG? Or was it CAUG? Was it something a little longer and quite different—maybe CUAUGG? The answers were important because from those words, added up into paragraphs, Woese would deduce the degree of relatedness of the creatures from which they had come.

  If the sequences were still ambiguous after a secondary run, as they often were, at least for longer fragments, then those were cut further, using other enzymes, and a third run was made. Rarely there might be a fourth run, but that was usually impracticable (as well as unnecessary) because the short half-life of the radioactive phosphorus that had been fed into these bacteria meant that its radiation faded quickly, and, after two weeks, the bits wouldn’t burn their images onto film. With experience, Woese developed a good sense of how to cut the fragments and get it all done in three runs at most.

  Mitch Sogin and his successors did the culturing of microbes, the extraction of RNA, the cutting, and the electrophoresis. They added improvements to the methodology—different enzymes for cutting, modifications of the electrophoresis—and by 1973, the Woese lab had become the foremost user of Sanger-type RNA-sequencing technology in the world. While the grad students and technicians produced fingerprints, Woese spent his time staring at the spots. Was this effort tedious in practice as well as profound in its potential results? Yes. “There were days,” he wrote later, “when I would walk home from work saying to myself, ‘Woese, you have destroyed your mind again today.’ ” The years between 1968 and 1977 were lonely and long. Today sequencing is a snap, but Woese was ahead of his time, gathering data like a man crawling across desert gravel on his hands and knees. He couldn’t have done it without a strong sense of purpose.

  Being his assistant or his student called on a certain gravelly fortitude too. Mitch Sogin described the deliveries of radioactive phosphorus (an isotope designated as P-32, with a half-life of fourteen days), which by 1972 amounted to a sizable quantity arriving every other Monday. The P-32 came as liquid within a lead “pig,” a shipping container designed to protect the shipper, though not whoever opened it. Sogin would draw out a measured amount of the liquid and add it to whatever bacterial culture he intended to process next. “I was growing stuff with P-32. It was crazy,” he said, tossing that off as a casual memory. “I don’t know why I’m alive today.” Because the bacteria were cultured in growth media lacking other phosphorus, a vital nutrient, they would avidly seize the P-32 and incorporate it into their own molecules. Sogin would then extract and purify the ribosomal RNA, “all the while not contaminating the laboratory.” That was the hope, anyway. For separating 16S from the other ribosomal fractions, he used “home-built electrophoresis units,” cylinders of acrylamide gel through which the different molecular fragments would migrate at different speeds. (Acrylamide is a water-soluble thickener, sometimes used in industry as well as in science.) Then he would freeze the gel and attempt to slice it, like bologna, with a very precise knife. The slicing was difficult: slices would fall off when they shouldn’t, he had to work the material at just the right temperature, and “this was pretty radioactive stuff.” Sogin then cut the 16S molecules into fragments with an enzyme, and those fragments would run a race of their own, not through cylinders of gel but along a racetrack of special, absorbent paper.

  One end of the long paper strip went into a receptacle known as a Sanger tank (as developed by Fred Sanger), containing a liquid buffer. The strip passed over a rack, beyond which its far end dropped into another Sanger tank, and both tanks were wired to an apparatus that provided the electrical pull. At the bottom of the tanks were high-voltage platinum electrodes, covered by three inches of liquid buffer and then at least fifteen inches of Varsol, a solvent not unlike paint thinner, intended to cool the paper strip. “Varsol is both volatile and explosive,” Sogin said. The power source delivered around 3,500 volts and plenty of amps, he recalled—“certainly enough to kill you.” Also enough, with an errant spark into the Varsol, to blow you up.

  This whole panoply of dangerous, intricate machinery dwelt within a shielding hood that could be closed behind large sliding doors, floor to ceiling, in a nook off the main lab known as the electrophoresis room. Set up the system, close the doors, turn on the juice, hope for the best. “I was too stupid to be afraid of anything,” Sogin told me. “Too naïve. Too young. Immortal.” He was also lucky. Nobody got hurt.

  Around the time Sogin finished his doctorate and prepared to leave, Woese hired a young woman named Linda Bonen, a walk-in from a different building, to take on some of the technical work. Raised in rural Ontario, she had come down to the University of Illinois and gotten a master’s degree in biophysics. Woese trained her for this new lab work himself—how to chop the RNA into fragments, how to run the electrophoresis in two dimensions, how to prepare the films, even a bit about how to interpret them, deducing which spot on a film represented which fragment, which little blurt of letters. Was it UCUCG, or was it UUUCG? Tricky to tell. But here’s GAAGU, obviously different. Woese coached her patiently on the tasks and their meaning.

  “He was very good about bringing me along,” Bonen recalled four decades later, when I visited her at the University of Ottawa, where she was by then a biology professor herself, gray haired, deeply expert in molecular ge
netics, gentle mannered as a schoolteacher. “The end product would be a ‘catalog’ for microbe X,” she said, meaning simply a list of the different fragments found within the 16S rRNA molecules of that creature. A catalog. If the fragments resembled words, these catalogs were the paragraphs. Comparing one catalog with another revealed the degree of similarity between any two organisms, by a very precise standard, and more dissimilarity could be taken to reflect more distance in evolutionary time. Where had the great limbs diverged from the trunk, the big branches from the limbs—and why there, and why then, and leading to what creatures? Beyond the mind-numbing methodology of data collection, those were the questions Woese hoped to answer.

  What was he like, I asked Bonen, as a boss and a teacher?

  “Well, he never came across as a boss,” she said. “He was very soft-spoken and quiet, reserved. I’m sure you’ve . . .” She hesitated. “Did you know him yourself? Did you ever meet him?”