The Tangled Tree

by David Quammen

  Another memorable Goldenfeld-Woese collaboration appeared in Nature a year later, under the title “Biology’s Next Revolution.” This one ran just a page. Its purpose, the two authors announced, was to explain why a radical transformation would soon sweep through the science, catalyzed by fresh thinking and “the coming avalanche of genomic data,” and probably forcing biologists to revise some of their fundamental tenets, including the concepts of species, organism, and evolution itself. Their explanation started with horizontal gene transfer.

  “Among microbes, HGT is pervasive and powerful,” wrote Goldenfeld and Woese, acknowledging (as Woese earlier had been disinclined to acknowledge) that it wasn’t just a thing of the dim past. “The available studies strongly indicate that microbes absorb and discard genes as needed, in response to their environment.” Because of that genetic fluidity, the two men argued, the concept of “species” is useless among bacteria and archaea. With genes flowing sideways, information moving across boundaries, and energy flowing upward from cells through communities and environments, the concept of an “organism”—an isolated creature, a discrete individual—seemed less valid too.

  And then there was “evolution” in its familiar Darwinian sense. That also seemed obsolete. Their newer idea—that evolutionary innovation might occur by means other than incremental mutation, and spread by means other than vertical inheritance—called the Darwinian model into question, they claimed. By this time, Woese had been reading some brilliantly unconventional scientific thinkers, such as the biologist Stuart Kauffman and the physicist Ilya Prigogine, associated with the swirl of ideas known as chaos and complexity theory, who proposed that certain “emergent properties,” unpredictable and wondrously elaborate, could arise spontaneously within complex interactive systems. Kauffman particularly, in books such as The Origins of Order and At Home in the Universe, had suggested the possibility of “self-organization” emerging from biological systems. To some biologists, these ideas would seem dangerously metaphysical, steps toward rejecting Darwinian theory that, if misunderstood, as they surely would be, again might give aid and comfort to creationists. But to Woese, in his hunger for more, they appealed.

  Somewhere amid the rich chaos of the RNA-world, Goldenfeld and Woese wrote, an “operating system” might have spontaneously taken form, by which the more promising innovations arising from random mistakes in RNA self-replication could be communicated and applied. They were alluding to the translation mechanism as seen eventually in cells, by which DNA information is turned into working proteins. At the core of that mechanism sits the ribosome, and at its core, Woese’s beloved 16S rRNA molecule. That thought led to another: that early life evolved in what Goldenfeld and Woese called “a lamarckian way,” meaning the inheritance of acquired characteristics, with vertical inheritance less important than horizontal gene transfer. “Thus, we regard as regrettable the conventional concatenation of Darwin’s name with evolution, because other modalities must be considered.” It was a fancy way of saying: Let’s pull Darwin down from his pillar. He may not have been wrong entirely, but his theory failed to cover the first two billion years.

  74

  The other change in Woese’s work life began earlier, back in 1997, when he arrived late to a meeting with the provost, several deans, some other faculty, and representatives from a charitable foundation that was considering a big grant to the University of Illinois. The grant, if it happened, would launch a center for comparative genome studies. The grant proposal came from a biologist studying mammal genomes and immunology, a young man who was friendly with Woese. The mammal fellow, Harris Lewin, had invented a fancy name for their enterprise: phylogenomics. Lewin would be lead researcher on this early effort. Woese would lend it star power, and the foundation’s program officer badly wanted to hear from him before committing the money. But Woese had fallen from a ladder the day before, injuring his neck, and for that reason, among others (including his inherent crankiness), his colleagues didn’t expect him to attend this meeting. Finally, he did show up, an “unearthly figure in a full neck brace,” as Harris Lewin recalled later, and sat quietly at the end of the conference table. He looked uncomfortable. Probably he was in pain. The program officer asked him, “So, Professor Woese, what does phylogenomics mean to you?”

  He took a deep breath, closed his eyes, and began to riff. Lewin remembered it as “an unrehearsed statement, a stream of scientific consciousness that blows the intellectual socks off everyone in the room. We all sit in stunned silence for a moment, knowing that Carl had just done something extraordinary.” He had laid out a rationale for the whole, ambitious program: unraveling the true relationships among living organisms, using molecular evidence, and tracing those relationships backward in time to illuminate the true history of all life on the planet. Oh, only that? “I wish I had recorded it,” Lewin said later.

  Woese’s performance especially surprised Lewin because he knew that Woese didn’t even like the word phylogenomics. It seemed clever but undefined. But it was good enough, if one word be required, at least for the grant makers. It could pass as a label for Woese’s own work, plus the work of Lewin and others among the interdisciplinary team they were assembling there in Urbana, and distinguish their enterprise from the efforts of others elsewhere. Woese hit his mark and made his speech, Lewin thought, out of loyalty to the university that had harbored him for a lifetime. “He did it for Illinois.”

  Whatever his motivation, whatever his unvoiced reservations, it worked. The foundation’s check arrived within two months—seed money for a much bigger effort—and the project began. Ten years and $75 million later, the university opened its Institute for Genomic Biology, with Lewin as founding director and Woese as resident sage.

  Harris Lewin considered himself an “improbable friend” to Carl Woese, because of wide differences in their scientific backgrounds, among other reasons. Lewin came from the field called animal sciences, meaning livestock and factors affecting their health, including genetics and immunology, and he studied topics such as bovine leukemia virus, a retrovirus that invades the genomes of cows. Woese didn’t care much about animals, not scientifically, because the evolutionary questions that engaged him go back far earlier in time. Animals represent a small twig in the canopy of the tree of life, and he concerned himself with the big, deep limbs. Lewin had heard “some scary stuff” about Woese’s unapproachability, but after a decade or so in Urbana, as his own work led toward comparative mammal genomics, he got curious enough to try to meet him anyway.

  “He was perceived very much as a loner and a guy with a chip on his shoulder,” Lewin told me during our conversation at the University of California, Davis, where he has lately finished a stint as vice chancellor for research. “I found a person that was quite the opposite of everything everybody said.” He walked into Woese’s lab, where yellowing sheets of paper showing the early trees, drawn with George Fox and others two decades earlier, still hung on the walls. Woese was seated, as usual, in his old swivel chair with his feet up on the lab bench. He stood, greeted Harris warmly, gave him a tour, and they talked for hours. Their friendship grew so trusting that, when the Institute for Genomic Biology (IGB) took shape, Woese lobbied university officials to appoint Lewin as director. And he lobbied Lewin, against a reluctance about giving up his own research, to accept. When the IGB opened its doors in 2007, in a new building just across from the university observatory, a sleek thing with big windows overlooking a stone plaza, they all moved in.

  Woese left his old lab reluctantly. He had been up there on the third floor of Morrill Hall—site of his greatest discoveries, his hardest work, and many good times—for more than forty years. Lewin persuaded him to accept a nice office (he declined a grand one) in the IGB building, with a view of the plaza. That ended his “isolation” in Morrill, as Lewin saw it, and put him among “fresh young troops,” Nigel Goldenfeld and others, “as well as his able and beloved assistant Debra Piper.” Woese met Piper when he moved into the bu
ilding, where she worked for the program that Goldenfeld called Biocomplexity, and she became a protector of Woese, as well as a helper and a friend. All seemed nicely hospitable to Woese, except that the office arrangement turned sour, and a small shadow came over his friendship with Harris Lewin, when Lewin commissioned for the plaza a certain three-piece sculpture meant to celebrate Woese’s most famous achievement: the illumination of a third major domain of life.

  “I wanted to do something to honor Carl’s discoveries,” Lewin wrote later, “with symbolic trees.” A committee deliberated, and a statewide competition led to the choice of an “irreverent” artist from Chicago. The tree concept disappeared. The sculptor produced three abstract blobs, large things, molded from polyurethane but looking somewhat like mounds of cookie dough scraped off a spoon, all identical in their irregular shape but differing in scale and color. The biggest one was chartreuse. The middle-sized one was deep orange. The smallest was yellow. You can see them on your next trip to Urbana. Lewin accepted the blobs and renamed them, along with the plaza, Darwin’s Playground. It struck Woese as an affront (to Lewin’s enduring regret), and he wouldn’t look at them. He started entering and leaving the institute by a side door, away from the plaza. He moved to an interior office with no windows. “Eventually, he forgave me,” Lewin wrote, but it didn’t come easily.

  This marks the Late Woese period, when he was angry at Charles Darwin, disenchanted with molecular biology as practiced during the twentieth century, frustrated by resistance to what he considered his best ideas, annoyed by the very word prokaryote, embittered by what he considered his inadequate acclaim, disappointed in particular at not having won a Nobel Prize, and disinclined to appreciate whimsical modern art, outside his window, bearing Darwin’s name. He had five years to live.

  75

  With the sequencing of the human genome, as a rough draft in 2000 and a better (but still provisional) version in 2003, by way of the competitive “collaboration” between Craig Venter’s group and the International Consortium, scrutiny of the full sequence began to yield some surprises.

  Those surprises related not just to the makeup of human DNA and how parts of it function, but also to the history of its assemblage and the sources from which some bits may have arrived. The first big revelation turned out to be a false alarm: the Consortium’s announcement in 2001 that 223 human genes seemed “likely to have resulted from horizontal transfer from bacteria.” The team had made some hasty assumptions, on insufficient data—their mistakes noted quickly by Steven Salzberg and other critics, as I described earlier—and that claim didn’t stand. Further sequencing of genomes of nonhuman creatures put the human genome in better perspective. Meanwhile, another revelation from the sequenced human genome, also mentioned by the Consortium when it published its 2001 analysis, but given less public attention, was the vast amount of seemingly pointless repetition in the three-billion-letter sequence. There it sat, within the genome, like a mammoth landfill of what had been called “junk DNA.”

  So much redundant blather in the fundamental human blueprint seemed almost embarrassing. Most of it occurred in the form of relatively short bursts of code, each just a few hundred to a few thousand bases long, constituting units that reappeared thousands or hundreds of thousands of times. In total, those repetitive sequences accounted for half of the total genome. (Coding sequences that produced actual proteins comprised only 5 percent of the genome.) The repeats had been noticed years earlier, by other methods even before DNA sequencing became possible, and dismissed by some biologists with that “junk” label. Other scientists knew them, more descriptively, as “transposable elements.” They were transposable in the sense that they seemed not just to have copied themselves many times, but also to have jumped around into different parts of the genome. These leaping, repetitive sequences weren’t, in fact, useless; they were clues, and in some cases more. The Consortium scientists recognized them as potentially instructive—“an extraordinary trove of information” constituting “a rich paleontological record” about human evolution. What did that record reveal? Harder to say.

  There’s an ironic prehistory to the study of these transposable elements. They were first detected by the visionary plant geneticist Barbara McClintock back in the 1940s as she studied the genetics of maize (corn). At that time, McClintock worked at Cold Spring Harbor Laboratory, on Long Island, where she grew and tended her own maize, raising a few hundred plants on an acre or so of ground every summer. She looked for mutations, induced artificially by X-raying the kernels, and traced those mutation from chromosome to chromosome, from cornstalk to cornstalk, among genetic crosses she created by pollinating the plants by hand. Maize served well as a study organism for geneticists in the era before molecular biology, because many of its mutations show themselves clearly in the color of kernels within the variegated cobs. McClintock discovered that some of her induced changes took form as mobile entities, which could somehow bounce from one chromosome site to another in the course of plant development. She focused on two mutations in particular, observing the way they interacted to cause breakage in a chromosome. She called them “controlling elements” because they seemed to play a role in gene expression. What she had discovered, besides a gene-regulatory relationship, were the first transposable elements ever recognized. For that, almost forty years later, she received a Nobel Prize.

  But the irony of her story isn’t obscurity followed by acclaim. That’s the mythic version, satisfying but inaccurate, and preferred by some tellers, who have made McClintock a great feminist hero. She was heroic, certainly, and she preferred the mythic version herself, though feminism was never her flag. The real irony is that McClintock always considered the controlling aspect of her elements far more significant than their jumping from place to place in the genome. By some testimony, she wasn’t even much interested in transposition, at least later in her career. But the Nobel Committee was, and it gave her the prize “for her discovery of mobile genetic elements.”

  After McClintock’s early work, as genetic research moved into the molecular dimension, other transposable elements turned up in the genomes of other creatures: bacteria, fruit flies, yeast, humans. They acquired a shortened label, transposons. And they acquired catchy names, some of them, just as genes are given names. There’s a group of transposons known as mariner, which have sailed widely from place to place for millions of years and can be found in the genomes of fruit flies and many other animals, including humans. The original two mariner elements in humans arrived (from somewhere) during early primate evolution and copied themselves throughout our ancestors’ genomes roughly fourteen thousand times. The most abundant human transposon is Alu. It’s only about three hundred bases long, but that three-hundred-letter nonsense word recurs in the human genome more than a million times. Nature is wildly various, we know that, but nature under Darwinian selection is also thought to be sternly economical, and such redundancy has enticed some biologists to wonder what the devil has been going on. One of those wondering is Cedric Feschotte, a Frenchman from Toulouse.

  Feschotte worked on transposable elements in insects for his doctorate at the University of Paris. He was hired as a postdoc at the University of Georgia, in Athens, to help study transposable elements again, this time in rice. As a side project, he worked on maize, crossing corn plants in a greenhouse, much as Barbara McClintock had done. Part of what makes maize still so apt for such research is that—unsuspected even by McClintock back in the 1940s and 1950s—transposons comprise 85 percent of its genome, and they jump around that genome frequently. From the Peach State, Feschotte went to the University of Texas at Arlington; and from cereals, he went to vertebrate animals. The constant again was these crazily mobile elements, bouncing and copying throughout genomes. By the time I caught up with him, he was a professor at the University of Utah School of Medicine, focused on transposons in humans and other vertebrates. On a shelf above his desk sat two variegated cobs of corn—mementoes of his sweaty days in
Georgia, no doubt, and tokens of homage to Barbara McClintock.

  Feschotte’s first graduate student, in Texas, was a local fellow named John K. Pace II (no relation to Norman Pace), slightly older than his fellow students. Married, with kids, and ten years’ experience as a computer programmer, John Pace just wanted to get a master’s degree so he could teach biology at a community college somewhere. But then, under Feschotte’s guidance, he made a major discovery. Putting his computer skills to work, scanning genomes in search of transposons, he found one in an East African primate known as a bush baby. The element ran to almost three thousand letters and repeated itself in the bush baby genome more than seven thousand times. That was notable enough—but what seemed more odd to Pace was finding virtually the same element in the genome of a very different animal: the little brown bat, native to North America. This time, nearly three thousand copies.

  Pace and Feschotte, along with others from the lab, scanned genomes more widely, finding a close variant of the same transposon in a tenrec (a small mammal roughly resembling a miniature porcupine) from Madagascar. Recognizable parts of it occurred also in an opossum from South America, a frog from West Africa, and a lizard from the southeastern United States. Clearly, this thing—this fiercely assertive and agile stretch of DNA—had gotten around, both within and between creatures, within and between continents. But the fact of its complete absence from the genomes of many other vertebrate animals (including nineteen kinds of mammal) suggested strongly that it had gotten around horizontally, not by vertical descent throughout vertebrate ancestry. And once passed to a new genome, it replicated prolifically. The tenrec genome contained 13,963 complete copies. The bush baby held 7,145. Each version within the different animals was at least 96 percent identical to the other versions, giving full confidence that they shared a single source and a recent invasion history. Any set of transposons so invasive and strange deserved a vivid moniker, so Feschotte’s team called it Space Invaders.