Downloaded from http://cshperspectives.cshlp.org/ on September 29, 2021 - Published by Cold Spring Harbor Laboratory Press The Persistent Contributions of RNA to Eukaryotic Gen(om)e Architecture and Cellular Function Ju¨rgen Brosius Institute of Experimental Pathology (ZMBE), University of Mu¨nster, D-48149 Mu¨nster, Germany Correspondence: [email protected] Currently, the best scenario for earliest forms of life is based on RNA molecules as they have the proven ability to catalyze enzymatic reactions and harbor genetic information. Evolu- tionary principles valid today become apparent in such models already. Furthermore, many features of eukaryotic genome architecture might have their origins in an RNA or RNA/ protein (RNP) world, including the onset of a further transition, when DNA replaced RNA as the genetic bookkeeper of the cell. Chromosome maintenance, splicing, and regulatory function via RNA may be deeply rooted in the RNA/RNP worlds. Mostly in eukaryotes, conversion from RNA to DNA is still ongoing, which greatly impacts the plasticity of extant genomes. Raw material for novel genes encoding protein or RNA, or parts of genes including regulatory elements that selection can act on, continues to enter the evolutionary lottery. Everything has been said already, but not yet by every- Kruger et al. 1982; Guerrier-Takada et al. 1983; one. Noller et al. 1992) leveled a major hurdle in —Karl Valentin understanding the origin of life. The salient Sturgeon’s Revelation: Ninety percent of science fiction discoveries eliminated the virtually impossible is crud, but then, ninety percent of everything is crud. prerequisite for two to three different classes of —Theodore Sturgeon macromolecules to converge as an evolving unit. At the same time, RNA provides a required con- They think that intelligence is about noticing things tinuity in the path of evolution (Yarus 2011) that are relevant (detecting patterns); in a complex world, intelligence consists in ignoring things that are during various genetic takeovers or evolution- irrelevant (avoiding false patterns). ary transitions (Cairns-Smith 1982; Szathma´ry —Nassim Nicholas Taleb (Taleb 2010) and Smith 1995). In a remarkably insightful ar- ticle dating back half a century, Alex Rich fore- f all extant cellular macromolecules, RNA saw much of what now is becoming main- Ois the most ancient, persisting as much as 4 stream, for example, that RNA was ancestral to  109 years in our planet’s life-forms. The abil- protein and DNA (Rich 1962). This landmark ity to combine genotype with phenotype such publication received little attention over the as catalytic activity (Noller and Chaires 1972; years; even early proponents of an RNA world Editors: Patrick J. Keeling and Eugene V. Koonin Additional Perspectives on The Origin and Evolution of Eukaryotes available at www.cshperspectives.org Copyright # 2014 Cold Spring Harbor Laboratory Press; all rights reserved; doi: 10.1101/cshperspect.a016089 Cite this article as Cold Spring Harb Perspect Biol 2014;6:a016089 1 Downloaded from http://cshperspectives.cshlp.org/ on September 29, 2021 - Published by Cold Spring Harbor Laboratory Press J. Brosius did not refer to this article (Woese 1967; Crick could replicate themselves or each other. Should 1968; Orgel 1968; Gilbert 1986), although at the threshold be set at the transition when the least one of the investigators must have had molecules involved could change during repli- knowledge about the article, as it was cited in cation and the variants are subjected to selec- a different context concerning the stereochem- tion—the initial Darwinian ancestor (IDA) ical possibility of six distinct base pairs (Crick (Szathma´ry 2006; Yarus 2011)? The first self- 1968). The origin of the DNA genome from replicating macromolecules must not necessar- RNA and that “DNA may be regarded as a de- ily have been RNA. Derivatives of RNA, espe- rivative molecule which has evolved in the form cially with altered backbones, have been sug- that it only carries out part of the primitive nu- gested as predecessors of RNA owing to more cleic acid function” is anothercorrect prediction favorable chemistries/stabilities for spontane- (Rich 1962). Furthermore, the investigator pre- ous generation and persistence of oligomeriza- saged mechanisms such as antisense RNA con- tion at the presumed planetary conditions trol of gene expression, short interfering RNAs (Joyce et al. 1987; Scho¨ning et al. 2000; Zhang (siRNAs), and perhaps microRNAs (miRNAs): et al. 2005; Powner et al. 2009; Robertson and “If both strands are active, then the DNAwould Joyce 2012; Neveu et al. 2013). produce two RNA strands which are comple- An interesting question is, if it is that “sim- mentary to each other. Only one of these might ple,” why did life not evolve multiple times? be active in protein synthesis, and the other There are a number of explanations. The early strand might be a component of the control or environment of the planet with conditions fa- regulatory signal” (Rich 1962). voring the necessary chemical reactions differed In this article, I shall present the rise and from the more temperate conditions now (Rob- persistence of RNA from the dawn of an RNA ertson and Joyce 2012). Perhaps life did evolve world and discuss current evolutionary princi- before LUCA numerous times independently, ples already apparent in an RNAworld. In com- but the descendants of LUCA are the only sur- parison to Archaea and Bacteria, the eukaryotic vivors. Perhaps primitive forms of life, for ex- genome is a better vantage point, as archaeal and ample, in the form of IDAs, still do arise, but we bacterial genomes are more derived and, thus, are not aware of them, in part, because we have lost many of the RNA signatures that eukaryotes not searched for such simple and different life- still show. It is likely that eukaryotic DNA ge- forms. Another reason is that a nascent form of nomes not only kept much more of their RNA/ life would easily be outcompeted by the estab- RNP world heritage than previously anticipat- lished ones, as the latter had a great chronolog- ed, but also continue to evolve novel RNAs in ical advantage adapting to current conditions. various functional roles. New forms of life might have a chance only if their metabolism is sufficiently different and, thus, not useful prey to LUCA-related life-forms WHEN DOES LIFE BEGIN? or if they happened to evolve in an unoccupied An excellent treatise of possible scenarios lead- niche so as not to succumb to immediate pre- ing to and continuing in an RNA world to the dation by the fitter “incumbents.”1 last universal common ancestor (LUCA) is available (Atkins et al. 2011). Can the beginning of life be defined along the transitions from RETRACING THE PATH physicochemical to biological reactions? Like In any event, by applying in vitro synthesis and almost everything in biology, clear boundaries selection procedures (Ellington and Szostak are difficult to demarcate and thus the defini- tion of the first life-form rather occupies a 1 bandwidth on a continuum. One of several pos- An afterthought worthy of note is the dichotomy with re- spect to life’s fragility considering individuals, even species sible thresholds to consider would be the fortu- versus the resilience of life as a whole, over an 4-billion- itous generation of one or two molecules that year timescale. 2 Cite this article as Cold Spring Harb Perspect Biol 2014;6:a016089 Downloaded from http://cshperspectives.cshlp.org/ on September 29, 2021 - Published by Cold Spring Harbor Laboratory Press RNA and Eukaryotic Gene/Genome Architecture 1990; Tuerk and Gold 1990; Gold et al. 2012), vide, fuse while shuffling their contents and di- several laboratories are making great strides to- vide again, in other words, performing sexual ward generating RNA molecules with the ability acts. Another “forecast” of the mechanisms gen- to self-replicate (Doudna and Szostak 1989; erating genomic diversity would be recombina- Johnston et al. 2001; Zaher and Unrau 2007; tion not only between cells, but also between Lincoln and Joyce 2009; Shechner et al. 2009; different RNA molecules, a mechanism that ac- Wochner et al. 2011; Attwater et al. 2013; Mast tually had been observed in a two-component et al. 2013), although these RNA polymerase ribozyme system (Lincoln and Joyce 2009). The ribozymes still fail to completely self-replicate origin of viruses could date to this early stage of (Deamer 2005). Cooperation of two or more cellular evolution as well. RNA molecules could, RNA enzymes in hypercycles (Eigen and Schus- perhaps protected by a lipid envelope, move ter 1977) may be a solution to this problem from cell to cell, blurring the line between infec- (Vaidya et al. 2012) (see also below). tion and horizontal transfer. Once a self-replicating ribozyme (mono- or multimeric) arose with the further potential to MOST EVOLUTIONARY PRINCIPLES ARE AT evolve into a replicator not restricted to only LEAST AS ANCIENT AS THE RNA WORLD self-copy but to copy other RNA templates as well, a prerequisite for a metabolically self-suf- Lessons from the RNA world apply remarkably ficient RNA conglomerate, further challenges well to extant organisms and their genomes. are apparent. First, the replicator indiscrimi- In a primitive RNA cell, conflict and coopera- nately copying any “junk RNA”in the mix hard- tion, selfishness and altruism had to coexist and ly would be able to persist. Second, if further establish a fine balance. Importantly, the suc- RNA molecules would arise by copying with cess of individual ribozymes also depended to errors—just like in extant organisms, new genes a large degree on functional interactions with still arise by duplication and variation—to other cellular RNAs (today: gene products), eventually take over metabolic functions other namely, the (genetic) background of the proto- than replication (e.g., activated compounds, in- cell (Brosius 2003c).