News Focus

When and how did the command and control center of the eukaryotic cell arise? The Birth of the Nucleus

LES TREILLES,FRANCE—What stands between tion missing.” proteins that are modified and shipped out of us and Escherichia coli is the nucleus. Biologists have long considered the nu- the nucleus to build ribosomes. Eukaryotic cells—the building blocks of peo- cleus the driving force behind the complex- The picture is far different in bacteria, in ple, plants, and amoebae—have these special- ity of eukaryotic cells. The Scottish which DNA, RNA, ribosomes, and proteins ized, DNA-filled command centers. Bacteria botanist Robert Brown discovered it 180 operate together within the main cell com- and archaea, the prokaryotes, don’t. The years ago while studying orchids under a partment. It’s a free-for-all in that as soon as nucleus’s arrival on the scene may have paved microscope. In his original paper, Brown the DNA code is transcribed into RNA, the way to the great diversity of multicellular called the novel cellular structure both an nearby proteins begin to translate that RNA life seen today, so the membrane-bound or- areola and a nucleus, but the latter name into a new protein. In eukaryotes, “the dou- ganelle fascinates scientists probing the evolu- stuck. Now, as then, the organelle’s com- ble membrane [of the nucleus] uncoupled tion of modern organ- plexity inspires awe. The nucleus is a “huge transcription and translation” and resulted in isms. “The question of evolutionary novelty,” says Eugene better quality control, says John Fuerst, a the origin of the cell nu- Koonin of the National Center for microbiologist at the University of Queens- cleus is intimately linked Biotechnology Information in land, Australia. As a result, RNA is modified to the question of our Bethesda, Maryland. as needed before it comes into contact with own origin,” says Patrick a ribosome outside the nucleus. Forterre, a molecular The nuclear distinction between prokary- biologist at the Univer- otes and eukaryotes shaped early specula- on December 19, 2008 sity of Paris-Sud in tion about the development of complex life. Orsay, France. Until the 1970s, two competing theories Last month, Forterre dominated the debate over early eukaryotic and two dozen micro- evolution. According to one, a subset of bac- biologists, evolutionary teria slowly developed eukaryotic features, biologists, cell biologists, and others such as the nucleus. In the other, eukaryotes met* here to hash out leading theo- came first, but over time, some of them lost ries about the origin of the nucleus. the nucleus and evolved a cell wall, spawn- One camp holds that the organelle ing modern-looking bacteria. is the result of a microbial merger. Then the Woesean revolution struck. By www.sciencemag.org Another contends that residual nu- looking at DNA sequence differences in the clei hidden away in some bacteria same gene across hundreds of microorgan- indicate that the crucial innovation isms, Carl Woese, a microbiologist at the is far older than commonly thought. University of Illinois, Urbana-Champaign, Perhaps the most radical theory of showed that “bacteria” were actually two all puts viruses at the center of this kingdoms, the bacteria proper and the ar-

cellular development. chaea, which apparently arose some 2 bil- Downloaded from At the meeting’s end, the dis- lion years ago, millions of years before eu- cussions of the origin of the nucle- karyotes. The initial genetic analyses indi- us had left biologists with a key in- Precocious prokaryote. Bacteria aren’t supposed to have cated that archaea were more closely related sight: They had underestimated the nuclei, but Gemmata obscuriglobus does. A closer look to eukaryotes than were bacteria. This kin- complexity of the eukaryotic cell’s shows DNA (N, blue) inside a proper nuclear envelope (E, ship hinted that eukaryotes came from the 1.5–billion-year-old precursor. The green), as well as a cytoplasmic membrane (CM, red). seemingly simple archaeal stock. data presented indicated that this Recent comparisons of fully sequenced ancestral cell had more genes, more struc- Each nucleus in a eukaryotic cell consists microbial genomes have, however, added a tures, and more diverse biochemical of a double lipid-based membrane punctuated twist to this story: Eukaryotes contain both processes than previously imagined. by thousands of sophisticated protein com- archaeal and bacterial genes. Archaeal genes But when it came to accounting for how plexes called nuclear pores, which control tend to run processes involving DNA and the nucleus was born, no single hypothesis molecular traffic in and out of the organelle. RNA, so-called information functions; the bubbled to the top. “It’s like a puzzle,” says Inside, polymerases and other specialized bacterial genes are responsible for metabolic Forterre. “People try to put all the pieces to- enzymes transfer DNA’s protein-coding mes- and housekeeping chores. From the jumble gether, but we don’t know who is right or if sage to RNA. Other proteins modify the of genes, some evolutionary biologists have there is still some crucial piece of informa- strands of RNA to ensure that they bring an concluded that this division of labor arose accurate message to the ribosomes outside from the ancient symbiotic partnership be- * “The Origin of the Nucleus” was held in Les the nucleus. The nucleus also contains a nu- tween bacteria and archaea, a partnership

Treilles, France, from 7 to 13 July. cleolus, a tightly packed jumble of RNA and that gave rise to eukaryotes. CREDIT: J. FUERST

766 6 AUGUST 2004 VOL 305 SCIENCE www.sciencemag.org Friendly mergers signaling molecules, such as kinases and G very little in it. A second sits in the center of Such a partnership may have been enough to proteins, in common with eukaryotes. the microbe and holds a dense collection of create the nucleus, according to Purificación genetic material—RNA and DNA mixed López-García and David Moreira of the Self-starters with DNA- and RNA-processing proteins. University of Paris-Sud. The two evolution- López-García and Moreira’s proposal as- The stuff in between—the cytoplasm—is full ary biologists speculate that the original sumes that bacteria and archaea appear ear- of proteins, ribosomes, and RNA. union between bacteria and archaea grew lier on the tree of life than eukaryotes, but At least one planctomycete has a double from metabolic requirements. The nucleus, Fuerst holds that the reverse is true. He is internal membrane around its DNA instead they further argue, arose as a way for these of the more typical single membrane. The endosymbionts to keep their metabolic membrane is not continuous but consists of chemistries from interfering with one anoth- pieces of folded membranes linked together. er. “You needed the [nuclear] membrane be- The gaps between the folds could indicate cause you have two competing pathways,” how nuclear pores got their start, says Fuerst. López-García explains. Explaining these structures has always In 1998, she and Moreira proposed that posed a sticking point for nuclear evolution. in life’s earliest days, methane-making ar- Without pores, the nucleus can’t function. chaea sometimes lived within bacteria that But nothing similar to these complex chan- depended on fermentation for sustenance: nels had been seen in bacteria before. At the the so-called syntrophic model. The rela- meeting, however, Fuerst showed dramatic tionship worked for the archaea because fer- electron micrographs of craterlike spots in mentation yielded a resource they needed, the internal membranes of planctomycetes. namely hydrogen. The bacterium may have These depressions closely resemble nuclear benefited because fermentation requires that pores, he says. Although nuclear pore genes hydrogen concentrations remain low. are hard to compare, Fuerst is encouraged López-García and Moreira hypothesize that a preliminary look at a planctomycete that Earth’s changing environmental condi- genome hints that the bacteria have primi- tions ultimately prompted a shift in the sym- tive versions of eukaryotic genes for some biosis. The archaeum gradually lost its ap- key nuclear pore proteins. petite for hydrogen, ceased making methane, “If you combine all the evidence, it on December 19, 2008 and instead relied more on the bacterial host Fruitful partnership. A bacterium akin to this makes a consistent picture,” he asserts. for other nutrients. The archaeum’s mem- myxobacterium may have paired off with an “Gemmata is a valid model for a non- brane, which had been critical for methano- archaeum, eventually evolving a nucleus. symbiotic origin of the eukaryotic nucleus.” genesis, became superfluous. At the same It may not be alone. There’s a recently dis- time, the outer bacterial membrane invaginat- convinced that eukaryote-like cells were covered phylum of sponge-dwelling bacteria ed the cellular compartment, eventually sur- around before bacteria and archaea or that also seem to have nuclei, says Fuerst, and rounding the archaeal DNA but excluding the emerged right at the time when these there are likely more, yet-to-be-discovered ribosomes. The change was advantageous to prokaryotes split off to form separate king- microbes with similar features. Bacteria with the bacteria, because in separating ribosomes doms of their own. Fuerst points to an un- nuclear pores and internal membranes, fea- www.sciencemag.org from the microbial chromosomes, it helped usual group of bacteria that he’s studied for tures typically considered eukaryote-specific, ensure more accurate conveyance of the the past decade. These remarkable microbes suggest that the nucleus was born much earli- DNA’s message. This set-up persisted and ul- have nuclei, or something akin to them, and er than traditionally thought. If Fuerst’s sce- timately evolved into the eukaryotic nucleus, may resemble the early cells that evolved in- nario is correct, “then the nucleus actually says López-García. And what remained of the to modern eukaryotes, according to Fuerst. precedes eukaryotes,” says Koonin. archaeal cytoplasm became the nucleolus. Found in soil and fresh water, these mi- In fact, this compartment could date back The researchers suggest that modern to the last universal common ancestor crobes, called planctomycetes, have cell walls Downloaded from methanogenic archaea bearing a resem- that are not quite as rigid as those of other (LUCA), a putative organism from which eu- blance to eukaryotes are possible descen- bacteria. As early as 1984, researchers had karyotes, bacteria, and archaea eventually dants of the ancient methanogens that en- suggested that some planctomycetes also emerged, says Fuerst. If that’s the case, certain tered into the nucleus-generating symbiosis have internal membranes. In 2001, Fuerst and LUCA features, such as the nucleus, were re- with bacteria. These archaea and eukaryotes his colleagues, using sophisticated electron tained in eukaryotes but lost to some degree in have similar genes encoding proteins in- microscopy techniques, confirmed the exis- most archaea and bacteria. Indeed, that seems volved with DNA and RNA. For example, tence of these membranes, even revealing to be the case, as eukaryotic cells possess fea- they share genes for histones, proteins that double ones like those of a nucleus. Those tures now seen in each of these groups. help stabilize chromosomes. In contrast, observations “turn the dogma that ‘prokary- bacteria don’t have histones. otes have no internal membranes’ upside Hostile takeover 70 (1996) , 60

. Another modern microbe, the myxo- down,” says Philip Bell, a yeast biologist at A third option for the origin of the nucleus bacterium, may resemble the ancient bacter- Macquarie University in Sydney, Australia. revolves around viruses. “Viruses predated ial host in which the nucleus evolved. Like Using sophisticated electron microscopy the divergence between the three domains of eukaryotic cells, myxobacteria communicate techniques, Fuerst and his colleagues life,” says David Prangishvili, a virologist at MICROBIOL. REVS with other cells, move, and can form multi- have now verified that there are discrete the University of Regensberg, Germany. He cellular complexes. Myxobacteria “have membrane-bound compartments within two argues that viruses were already quite com- complex structures that are very striking” planctomycetes, Gemmata obscuriglobus mon in the primordial soup and only later be- and reminiscent of eukaryotic cells, López- and Pirellula marina. One compartment, came dependent on cells to survive. When

CREDIT: M. DWORKIN, García notes. These bacteria also have cell- pushed up along the periphery, seems to have these early cells came along, “viruses played

www.sciencemag.org SCIENCE VOL 305 6 AUGUST 2004 767 N EWS F OCUS Did a virus provide the first nucleus? Or was it something an early bacter- Viruses Bacteria ial cell evolved, either on its own or in partnership with an archaeum? To re- solve the origin of the nu- Eukaryotes cleus, evolutionary biolo- Last universal gists are exploring new common ancestor Viral intervention. Persistent viral infec- techniques that enable tions could have paved the way for the them to determine rela- nucleus at different points in early cellu- tionships of microorgan- Archaea lar evolution. isms that go much further back in time. And as new a critical role in the evolution of the complex quences found at the ends of eukaryotic genome sequences become available, such [eukaryotic] system,” adds Forterre. chromosomes. as those of several planctomycetes, Fuerst Viruses do have the ability to set up per- Bell speculates that a virus living in an ar- and others plan to search for more genetic manent residency in a cell, infecting but not chaeum set the stage for the nucleus. Ulti- similarities between these bacteria and eu- killing the host. Thus they and their genes mately, viral DNA and archaeal DNA karyotes. Meanwhile, García-López anx- can stay around and influence a cell’s evolu- merged inside the virus, and the new genome iously awaits sequenced genomes of tion. Bell, Forterre, Prangishvili, and Luis later shed genetic material from both. In the myxobacteria and plans to compare them Villarreal, a virologist at the University of end, “the unique genetic architecture of the with the genes of eukaryotes. California, Irvine, each have a different pro- eukaryote is a result of superimposing a viral Overall, says Forterre, it’s “a really ex- posal for how viruses were important to the genetic architecture on an archaeal genetic citing time to tackle questions which were evolution of the nucleus. Their supporting architecture,” Bell argues. previously only considered seriously by a data are provocative, but circumstantial and “If this is true, then we are all basically few theoritists.” controversial. “I do not believe [it],” says Ja- descended from viruses,” remarks Forterre. –ELIZABETH PENNISI comine Krijnse-Locker of the European Molecular Biology Laboratory in Heidel- berg, Germany. “The idea of the viruses ‘in- on December 19, 2008 venting’ [eukaryotic cells] from scratch is Profile Edward Hammond hard for me to conceive.” When viruses persist in cells instead of killing them, cells “can acquire a whole new set of genes in one event,” counters Villarreal. Activist Throws a Bright Light on While in residence over millions of years, the new viral genes could have supplanted bacte- Institutes’ Biosafety Panels rial or archaeal genes, replacing, for instance, proteins that process DNA. These extra genes Edward Hammond’s aggressive sleuthing has triggered a debate on the oversight of www.sciencemag.org could also evolve to play new roles in the cell. the growing field of biodefense research Villarreal points out that there are intrigu- ing similarities between nuclei and viruses, AUSTIN,TEXAS—In late January, Edward redacted information is private, proprietary, which are basically packets of DNA sur- Hammond sent out a blizzard of faxes to al- or security-sensitive. More important, Ham- rounded by a protein coat—and often by a most 400 research institutes from Honolulu mond has concluded that the IBC system, membrane. In red algae, for example, a nu- to New York. His request was straight- designed in the 1970s to review recombinant

cleus can move from cell to cell, much like forward enough: He asked for the minutes of DNA research, is in disarray. He claims that Downloaded from an infectious virus. And in general, cell nu- the last two meetings of each organization’s dozens of IBCs, many of them at the nation’s clei and viruses lack protein- and lipid- Institutional Biosafety Committee (IBC). research powerhouses, aren’t staffed proper- producing pathways within their borders. Hammond, who directs the Sunshine ly, don’t seriously review proposals, or never Both contain linear chromosomes, whereas Project, a small nonprofit organization based meet at all. Outraged, he has filed com- most bacterial chromosomes are circular. in Austin, wondered whether the IBCs fulfill plaints with NIH, asking it to cut off funding Both disassemble their “membrane” during their oversight role for certain types of biol- retroactively to 19 institutions. Dozens more replication. Both transcribe DNA but don’t ogy experiments as prescribed by guidelines complaints are on the way. translate mRNA within their boundaries. As from the National Institutes of Health (NIH). NIH officials are investigating the they replicate within a cell, some poxviruses In particular, he questioned whether they charges, but there’s no reason to assume that even make a membrane around their DNA would publicly share their deliberations. the entire system is broken, says Allan Shipp using the endoplasmic reticulum of the in- Such openness, he says, is vital to prevent of NIH’s Office of Biotechnology Activities fected cell. The eukaryotic cell uses this biodefense research from going astray. (OBA), which oversees IBCs. Most IBCs same material to build its nucleus. Today, Hammond is fighting testy e-mail are “very earnest in their attempts and desire SOURCE: P. FORTERRE Large, complex DNA viruses, which in- battles with his targets over their tardy re- to fulfill their responsibilities,” he says. ; clude poxviruses and the African swine fever sponses. How to answer his query has be- Some researchers who have followed virus, likely bear the closest resemblance to come a hot topic among biosafety officers Hammond’s quest—he posts alleged viola- SCIENCE the putative viral ancestor of the nucleus, and university lawyers. Some universities tions frequently on his Web site—disagree. Bell suggests. DNA strands in these viruses have sent him minutes, but with almost “Frankly, I’ve been surprised by the number

have primitive telomeres, protective DNA se- every detail blanked out, arguing that the and magnitude of the deviations from the CREDITS: P. HUEY/

768 6 AUGUST 2004 VOL 305 SCIENCE www.sciencemag.org