Noncoding RNAs: Persistent Viral Agents as Modular Tools ... · life style; nucleic acid language...

NATURAL GENETIC ENGINEERING AND NATURAL GENOME EDITING

Noncoding RNAs: Persistent Viral Agentsas Modular Tools for Cellular Needs

Gunther Witzany

Telos-Philosophische Praxis, A-5111-Burmoos, Austria

It appears that all the detailed steps of evolution stored in DNA that are read, tran-scribed, and translated in every developmental and growth process of each individualcell depend on RNA-mediated processes, in most cases interconnected with other RNAsand their associated protein complexes and functions in a strict hierarchy of temporaland spatial steps. Life could not function without the key agents of DNA replication,namely mRNA, tRNA, and rRNA. Not only rRNA, but also tRNA and the processing ofthe primary transcript into the pre-mRNA and the mature mRNA are clearly descendedfrom retro-“elements” with obvious retroviral ancestry. They seem to be remnants ofviral infection events that did not kill their host but transferred phenotypic competencesto their host and changed both the genetic identity of the host organism and the identityof the former infectious viral swarms. In this respect, noncoding RNAs may representa great variety of modular tools for cellular needs that are derived from persistentnonlytic viral settlers.

Key words: noncoding RNAs; regulatory networks; addiction modules; persistent virallife style; nucleic acid language

Introduction

Current knowledge indicates that DNA isa stable information storage medium. As pro-posed in an article by Vetsigian, Woese, andGoldenfeld, it serves as an “evolutionary pro-tocol” for evolutionary novelties and selectedproperties. A wide variety of small RNAsregulate key cellular processes of replicationas well as genetic arrangements, rearrange-ments, recombination and repair, and eveninventions. In most cases they act after be-ing transcribed from the stable DNA storagemedium into a pretranscriptional or transcrip-tional modus, with the advantage of being ac-tive prior to all translated proteins. Elementssuch as micro-RNAs, small nuclear and smallnucleolar RNAs, tRNA, and rRNA as well asthe assemblies of the spliceosome and ribosomeare vital to all life processes. In this respect I re-

Address for correspondence: Gunther Witzany, Telos-PhilosophischePraxis, Vogelsangstraße 18c, A-5111-Burmoos, Austria. Voice/fax: 004362746805. [email protected]

member a quote from the Austrian philosopherLudwig Wittgenstein: “The meaning of a wordis its use within a language and to understand asentence means to understand a language. Tounderstand a language means to be the masterof a technique.”

What are the “masters” of the techniqueused to edit nucleotide sequences of the ge-netic text according to combinatorial (syntac-tic), context-sensitive (pragmatic), and content-specific (semantic) rules, according to CharlesMorris the obligate and nonreducible levels ofrules that are inherent to any kind of languageor language-like codes?

Why Editing Needs Editors

For a long time the textbook conviction heldsway that the genetic content arrangementsof DNA sequences are evolutionary results ofrandom mutations and their selection. Thenit was noticed that DNA regions which codefor proteins are decreasing with organismic

Natural Genetic Engineering and Natural Genome Editing: Ann. N.Y. Acad. Sci. 1178: 244–267 (2009).doi: 10.1111/j.1749-6632.2009.04989.x c© 2009 New York Academy of Sciences.

244

Witzany: Noncoding RNAs: Persistent Viral Agents 245

complexity (in humans 1.5%) and former“junk DNA” plays increasing important rolesin gene regulation. Additionally, it was as-sumed that this “junk DNA” is a kind ofselfish genetic element, with an inherent ten-dency to replicate themselves, in contrast withits helpful and symbiotic functions within hostgenomes.

For many decades it was common practiceto speak about the “genetic code” with its in-herent language-like features. Long before, andeven after, Manfred Eigen’s suggestion that thenucleic acid sequences are comparable to andfunction like any real language coherent with a(molecular) syntax, linguistic, and communica-tive vocabulary was commonly used in genet-ics, cell biology, and molecular biology: geneticcode, code without commas, misreading of thegenetic code, coding, genetic storage mediumDNA, genetic information, genetic alphabet,genetic expression, messenger RNA, cell-to-cellcommunication, immune response, transcrip-tion, translation, nucleic acid language, aminoacid language, recognition sequences, recog-nition sites, protein coding sequences, repeatsequences, etc.

In contrast with the evolutionary paradigmof random assemblies of nucleic acids thatconstitute the genetic text we do not knowany real-life languages or codes whichemerged as a randomly derived mixture ofcharacters.

If Manfred Eigen’s suggestion is still valid,and the description of nucleic acid sequenceorder in terms of linguistics (molecular syntax)makes sense for the future, we should look atthe current scientific knowledge of “language”and “communication.”

Every language is based on signs, whetherthey are signals or symbols. In humans andother animals they are transported auditively,visually, or tactilely. In nonhuman living be-ings they are transported by small moleculesin crystallized, fluid, and gaseous form. Ad-ditionally these signs can be combined coher-ently with combinatorial rules (syntax). Signsare not generated and used by themselves, but

in real-life languages by living beings. Thesesign-generating and sign-using agents live in vivo

in continued changing interactions and envi-ronmental circumstances. This is the context(pragmatics) in which a living being is inter-woven. This context determines the meaning(semantics) of the signs in messages that areused to communicate and to coordinate singleas well as group behavior.

Therefore, we can understand that the samesentence, or the same syntactic sequence or-der, of any language or code can have different,and in extreme cases, opposite meanings andtherefore transport different messages. The im-portant consequence of this fact is that it is notpossible to extract the meaning of an informa-tional content solely out of the syntactic struc-ture, but someone has to identify the contextwithin which the living being uses this syntacticstructure.

The primary agents are not the sequences ofsigns, not the rules which determine sequenceorders, but the living agents. Without livingagents there are no signs, no semiotic rules, nosignalling, and no communication. Paradoxi-cally, without signs, semiotic rules, signalling,and communication, no living agents could co-ordinate growth and development.

If we assume the genetic code to functionlanguage-like, knowing that no language whichhas been observed functions by itself, then wehave to postulate living agents that are compe-tent to use signs coherent with syntactic, prag-matic, and semantic rules. Adapted to the ge-netic code, this means that there must be livingagents competent in generation and integrationof meaningful nucleotide sequences, and mean-ingful nucleotide sequences are not a randomlyderived mixture of nucleotides.

Natural genome editing from a biocommu-nicative perspective means competent agent-driven generation and integration of mean-ingful nucleotide sequences into preexistinggenomic content arrangements and the abilityto (re-)combine and (re-)regulate them accord-ing to the context-dependent (i.e., adaptational)purposes of the host organism.

246 Annals of the New York Academy of Sciences

DNA: Information Storage Mediumand Life Habitat of Endogenous

Viruses

There is increasing evidence that all cellularlife is colonized by exogenous and/or endoge-nous viruses in a nonlytic but persistent lifestyle.A persistent lifestyle in cellular life-forms mostoften seems to derive from an equilibrium statusreached by at least two competing genetic set-tlers and the immune function of the host thatkeeps them in balance. Persistent settlement ofhost genomes means that if we postulate agent-driven genetic text editing then we have to lookat their in vivo life strategies to understand theirhabits and the situational contexts that deter-mine their content arrangements. Then we canreconstruct nucleic acid sequences that func-tion as a code, not as a statistically random-like mixture of nucleotides, but as informationalcontent in a syntactic order that is coherent withthe whole sequence space generated by agentsthat are linguistically competent in nucleic acidlanguage, that is, the genetic code. As in everylanguage each character, word, and sentencetogether with starts, stops, commas, and spacesin-between has content and a text-formattingfunction and is generated by competentagents.

If we imagine that humans and one of thesimplest animals, Caenorhabditis elegans, share anearly equal number of genes (ca. 20,000) itbecomes obvious that the elements that cre-ate the enormous diversity are not the protein-coding genes but their higher order regulatorynetwork that is processed by the mobile geneticelements, such as transposons and retroposonsand the noncoding RNAs1 (Juergen Brosius cre-ated the appropriate image of “genes floatingin a sea of retroposons”). If we consider theimportant role of the highly structured and or-dered regulatory network of noncoding RNAsas not being randomly derived, one of the mostfavorable models with explanatory power is thevirus-first thesis. This means that the evolutionof the noncoding RNA world is the result ofpersistent viral life strategies.

The whole range of mobile genetic agentsthat are competent to edit the geneticcode/nucleic acid language not only edit, butalso regulate key cellular processes of replica-tion as well as transcription, translation, re-combination, repair, and even inventions viaa wide variety of small RNAs. In this respect,DNA is not only an information-storing archivebut a life habitat for linguistically-competentRNA agents, most of them seemingly of viralor subviral descent. To understand their nat-ural genome editing competence we have tolook not only at their linguistic competence inediting and regulating correct nucleotide se-quences but at their communicative compe-tence too, that is, how they interact with eachother, how they compete within host organ-isms, how they symbiotically interact with hostorganisms to ward off competing parasites, andwhat life strategies they share. Persistent infec-tion lifestyles that do not harm hosts and sym-biotic, cooperating viral swarms may be moresuccessful in evolutionary terms for integratingadvantageous phenotypes into host organismsthan “selfish” agents.

A Viral Progenitor of theEukaryotic Nucleus?

If we look at persistent viral life strategieswe should first look at the role of the eukary-otic nucleus. The eukaryotic cell most prob-ably evolved by a symbiogenetic integrationevent of former free-living bacteria. This in-tegration, however, cannot explain the progen-itor of the eukaryotic nucleus because its keyfeatures could not have derived from prokary-otes.2,3 The eukaryotic nucleus has numerouskey features, proteins, and RNAs that are notfound in any prok, aryote. Interestingly, thesekey features are present in certain prokaryoteviruses.4–6 These viruses use linear chromo-somes, telomere repeats, multiple membranes,histone-packaged chromosomes with mark-ing effect for self/nonself identification, andnuclear pores.


No single virus encompasses all of these keyfeatures, but every key feature of the eukary-otic nucleus is present in some large dsDNAviruses. This requires consideration of a pro-cess in which different viral competences wereintegrated into a single dsDNA virus that wasthe progenitor of the eukaryotic nucleus. Alter-natively, a large dsDNA virus functioned as asimple eukaryotic nucleus and later integratedother viral competences. On examination ofthe key features of several candidates for thisintegration, the focus is primarily on prokary-otic, eukaryotic, and archaeal phages.

Prokaryotic phages such as cyanophageshave double-stranded DNA, DNA and RNApolymerases similar to eukaryotes. Eubacterialphages possess linear double-stranded DNA,telomeres, DNA and RNA polymerases, chro-matin, and internal membranes. Archaealphages with linear double-stranded DNA havetelomere repeats similar to eukaryotes. Theyalso possess chromatin and an internal lipid ten-dency to nonlytic, persistent, (and often mixed)infections.4

Other DNA viruses share similar featuresthat are characteristic of the eukaryotic nucleusbut are not found in prokaryotes. An exampleis the vaccinia virus (poxvirus).7 These viruseshave a membrane-bound segregation of tran-scription and translation, multiple membranes,and their DNA synthesis combines membraneloss and a cell-cycle-dependent restoration aswell as an actin/tubulin-bound transport sys-tem4,8 and, interestingly, nuclear pores.9 Cyto-plasmic DNA viruses (African swine fever virus)have chromatin and linear chromosomes withtelomeres. PhycoDNA viruses have mRNAcapping, introns, and diverse DNA replicationproteins. Torque Teno Virus (TTV) 1-4 havelinear double-stranded DNA genomes with amolecular basis for the evolution of eukaryoticchromatin; they also have capsids which inte-grate internal and external lipid proteins.4

In addition, all these viruses have the capabil-ity for self and nonself identification. All virusesmark their genomes, RNAs, and proteins bydifferent kinds of chemical modifications, for

example, methylation. This marking allows thedifferentiation between self and nonself. Non-self may be other viruses, the host genome, orhost-related transcripts.4

Evolutionary Roles of Viruses asNatural Genome Editors

To understand the evolutionary emergenceof the eukaryotic nucleus it could be useful toreconstruct the natural genome-editing compe-tences of viruses.10 Recent research in micro-biology, based on comparative genomics andphylogenetic analyses, has demonstrated thatlife must be viewed from the perspective of thecrucial role played by viruses.4,11–15

This contradicts former concepts thatfocused on viruses in the framework of (i)escape theories, that is, viruses are intact ordeformed genetic parasites that escaped fromcellular life, or that viruses (ii) evolved from cel-lular ancestors or (iii) that they are not livingbeings because they cannot live without cellu-lar life. From these perspectives, viruses couldnot play crucial roles in the evolution of cel-lular life. Interestingly, phylogenetic analysesdo not support the former concept of RNA-and DNA-viruses descending from cellular life.These analyses also show that DNA and RNAviruses most probably did not have a commonancestor but evolved independently. Virusesprobably have to be placed at the very begin-ning of life, long before cellular life evolved.4

Persistent Viral Life Strategiesare Beneficial for Their Hosts

Acute viruses that exhibit lytic action inducedisease and even death. In contrast, a persistentlifestyle implies compatible interactions withthe host, either by being integrated into thehost genome or within its cell plasma.16 Theresult is nondestructive symbiosis during mostlife stages of the host. The persistent lifestyleallows the virus to transmit complex viral


phenotypes to the host organism. This process,which changes both the genetic identity of thehost and the identity of its persistent settler,enables the host to broaden its evolutionaryadaptational potential and may promote theformation of new species.4

The persistent lifestyle of viruses is typicallytissue specific, that is, host tissues are colo-nized by different nonlytic viruses which in-tegrate themselves into the host cytoplasm, forexample, as plasmids or into the host genomes,and co-evolve with them. A common habit ofpersistent viral settlers is that during host cellreplication they function in a tissue-specific,replication-cycle dependent manner. Interest-ingly, micro-RNAs in eukaryotic cells have sim-ilar tissue-specific or developmental expressionpatterns.17 Micro-RNAs play important rolesin Dicer- and Risc-mediated mRNA degrada-tion or mRNA translation inhibition.18 Thisimplies an RNAi immune function. Becausemicro-RNAs act on mRNAs, not on proteins,they are probably encoded by persistent nuclearDNA viruses.19 We will look at these compe-tences later.

Persistent Status throughAddiction Modules

The persistent status is the result of multi-ple colonization events into a host. This neu-tralizes former antagonistic and incompatiblefeatures of competing viral agents withoutharming the host.20–22 Most of the endogenousor exogenous inhabitants inherent to bacteria,protozoa, plants, animals, and fungi are a com-plementary mix of formerly antagonistic viralfeatures. They can still be identified today astoxin/antitoxin, restriction/modification-, in-sertion/deletion modules, that is, complemen-tary counterpart regulatory functions that donot harm the host.4,23,24 As symbiotic neutral-ization and counterpart regulation, they rep-resent new phenotypic features. One featureis regulated exactly by the antagonist accord-ing to developmental stages in the cell cycle,

replication, tissue growth, or similar develop-mental contexts. Should this suppressor func-tion become unbalanced, then the normallydownregulated part may become lytic with po-tentially lethal consequences, as documentedfor symbiodinium and its major role in coralbleaching.25

Retroviral Competences in a PersistentSymbiotic Lifestyle: Endogenization

Endogenous retroviral competences in thepersistent status are often characterized by fea-tures expressed only in the strict time win-dow of a developmental process, such as axisformation, trophectoplast formation, or the Sphase of the cell cycle. In these highly special-ized contexts they are replicated through sig-nalling, which blocks the suppression of thereplication process. After the function is ful-filled, a signal once again initiates suppressorfunction. Retroelements, with their (i) higher-order regulatory functions, (ii) capability forgenetic creativity, and (iii) capacity for inno-vation of new regulatory patterns and com-binations descended from retroviruses, whichcan easily be identified by their three essentialparts gag, pol, and env.26–28 Most endogenousretroviruses have been degraded into formerlyconnected domains, but they can still be rec-ognized by retroposons or one of these threegenes,21,29–31 which means their formerly con-nected genomic content may be used by host or-ganisms as single or networking modular toolsfor a variety of new regulatory functions. Thegag gene encodes structural proteins, pol en-codes enzymes such as reverse transcriptase,protease, ribonuclease and integrase functions,and env encodes envelope proteins, surface andtransmembrane proteins and proteins causinghost cell fusion and immunosuppression.22

Reconstructing the Highways that alsoPlay Important Roles in Persistence

Interestingly we also find small DNAviruses as genetically stabilized and co-evolved


persistent viruses that do not trigger and arenot part of an immune response of the hostorganism. Their active role is regulated anddepends on the cell cycle of the host in whichthey are transcribed, replicated, and silencedagain during a certain phase of the cell cycle.This means their persistent status is changedinto an active role only during a strictly de-fined phase of the cell cycle in which theyare needed for a specialized function and arestill competent, most likely being adapted espe-cially for this function. After fullfilling this func-tion they disappear again. To ensure this func-tion and its fine-tuned regulation, they needa highly conserved regulative viral protein do-main with characteristic host interactions to use(manipulate) the host replication for their spe-cial needs. Their behavioral pattern is adaptedto the host and circumvents the acute lyticphase.32

We can find similar reproduction patterns inRNA viruses. Replication of retroviruses in eu-karyotes depends on successful entry into themembrane of host cells.33 Some retrovirusescircumvent this active entry of the cellularmembrane in that they wait until the start ofthe cell-cycle phase-dependent dissolution ofthe cell membrane during replication. Oncethe direct path to the nucleus is free, retroviralRNA is transported to the eukaryotic nucleusand integrates into the host genome. Throughtranscription of the host genome a complete vi-ral RNA genome is processed and transportedout of the nucleus through the cytoplasm tothe cell membrane. Here the reproduced viralgenomes assemble and are encapsulated by vi-ral gag encoded structural proteins and leavethe host cell. Retroviral life integrates dozensof retroviral competences such as replication,transcription, translation, repair, trafficking toand from the nucleus, splicing, alternative splic-ing, and 3′ end processing. Transport to thenucleus and afterwards from the nucleus tothe membrane again gives an overview aboutagent-driven evolution of eukaryotic cells if wethink about the eukaryotic nucleus being anancient dsDNA virus.

In contrast to these retroviral and often lyticreplication cycles, the overwhelming majorityare nonlytic but persistent retroviruses. Theyinfect the host organism and integrate their gag,

pol, env functional parts into the host genomeand adapt to the replication cycle of the hostorganism without leaving the host cells. In be-coming part of the identity of the host genotypethey change the genome formation and trans-fer a phenotype to the host that noninfectedhost genomes do not possess. We are still atthe beginning of imagining how the persistentlifestyle of viruses plays a role in the evolution,development, and genomic regulatory ratio ofeukaryotes.4

This advantageous behavior of not leavingthe host genome again, but reaching a persis-tent status within the host genome (most prob-ably by becoming part of an addiction mod-ule, i.e., a competing genetic settler that createsan equilibrium status balanced by the host im-mune system), can be better understood if welook at the patterns of retroviral movement dur-ing infection events.

Patterns of Retroviral Movement: TheKinesin/Dynein Addiction Module

It is the prokaryotic viruses that tend to usepores during cell destruction and exit. Their en-try pores are their tail plates, which can also betoxins. In eukaryotes, the most relevant poresare on the mitochondria, associated with apop-tosis. However, eukaryotic DNA viruses (ade-novirus) do often bind to nuclear pores andthese are clearly also associated with micro-tubules that transport virus to the pore. Thispore-building ability plays an evolutionary rolein all tubular structures that connect multicel-lular tissues of eukaryotic organisms.

The alternative way of most eukaryoticviruses, including retroviruses, to pass throughthe cell membrane of the host is by endocy-tosis followed by receptor binding.34 A well-known behavioral motif then occurs: the re-lease of its RNA or DNA into the host cell andits immediate spread into smaller parts, suchas reverse transcriptases and preintegration


complexes. This is an important behavioral mo-tif in relation to noncoding RNA transcript pro-cessing into small noncoding RNA species, asdiscussed later.

These smaller parts move on the “highway”of the actin-based cytoskeleton and its micro-tubules in the direction of the nucleus35 by us-ing the kinesin-motor protein superfamily.36 Af-ter reaching the nuclear membrane these viralparts pass through the membrane through nu-clear pores by using importin proteins (whichconsists of two subunits with complementaryfunctions: an indicator for an ancient addictionmodule itself) which bind to a special recog-nition sequence. Afterwards, integrase, whichis also produced and used by DNA viruses forthe same purpose, integrates these viral partsinto the genome. The integration process isnot random but involves strict coherence tothe syntax of the nucleotide sequences of thehost.

If the host cell replicates, these RNA viralparts are transcribed into DNA. As DNA se-quences they pass through the membrane of thenucleus again to move towards the cell mem-brane. Again they use the “highway” of micro-tubules, but unlike earlier, they use dynein asthe motor protein.36 Interestingly, the changein direction of the kinesin/dynein transporterproteins depends on suppression of antagonistsof the dynein or kinesin. This could be an indi-cator that both are part of an addiction moduleof former competing genetic parasites. Prior tobecoming an internal part of cellular transportit could have been an external system for move-ment in and out of cells. And indeed there is aconnection between these two motor proteinsand retroviral (gag)-parts.28,37

When they reach the cell membrane the vi-ral parts form a patch, via protein assemblyon internal membranes, that is divided fromthe cell by exocytosis and produces its owncapsule in which the RNA genome matures.In other contexts, similar proteins are involvedin cytokinesis,38 and most interestingly, in neu-ral and immunological synaptic communica-tion.39,40 Other retroviruses build their cap-

sule, not at the plasma membrane, but at thecentrioles.30 Both sites of capsid building andtransport depend on intact env and gag codingsand the recycling of membrane parts.37 The gag

parts bind the kinesin motor proteins that areneeded for mictrotubulin transport. The exacttransport of the retroviral RNA through the cy-toplasm depends on interactions with numer-ous host proteins that build the so-called RNA-Transport-Granulat (RTG).37 This RTG is alsopresent in nearly all cells, such as fibroblasts,T-cells, and epithelial cells.41,42

Retroviral RNA editing for processing of ex-tracellular viral parts is a very complex pro-cess: retroviral RNA editing functions in a sim-ilar way to cellular mRNA processing but ismuch more regulated by cis- and transactivemechanisms, which seem to be a former retro-viral competence. Export of retroviral RNA outof the nucleus requires several RNA helicases,such as RNA helicase A, DEAD box proteins,DDX1 and DDX2.43–46

After retroviral RNA processing in the nu-cleus by (i) alternative splicing, (ii) 3′end pro-cessing, and (iii) RNA transport from thenucleus through the cytoplasm on the micro-tubule highways, they reach the membranewhere they assemble. Transport depends onintact micro-tubules.47 One part of the retro-viral RNA is not spliced but is translated intostructural proteins to form the capsules.48–51

The addiction module of the antagonistic mo-tor proteins kinesin/dynein (see note above),which drive retroviral RNA transport, plays animportant role in the cell division of eukary-otic cells.52,53 Different kinesin proteins regu-late the movement (Kip2p) and also the direc-tion/orientation of this movement (Kip3p).54

Interestingly, mitotic spindle processingwithout kinesin/dynein transcription does notfunction as well as positioning of the twopoles and the segregation of the chromo-somes in the anaphase.55–58 Both motor pro-tein families have genetic similarities witharchaea and eubacteria, indicating their im-portant roles in prokaryotes as well.59–61 Motorproteins and cytoskeleton interactions are very


specific and also interconnect with the Golgiapparatus.62,63

Agents of Natural Genome Editing:Genetic Settlers in a Comfortable

DNA Habitat

Recent research shows extensive dynamicDNA remodelling by mobile agents, suchas transposons, retroposons, and noncodingRNAs, which are able to cause a wide va-riety of DNA arrangements, rearrangements,and recombinations.64–67 Some authors referto these as agents of genomic creativity,21 mo-bile or regulatory elements68–70 or entities,71

while others refer to transposable elements,72

noncoding RNA populations,73 endogenousmutators,74 and still others to mobile DNAspecies or genetic parasites.4,75 Together, theseagents enable complex organisms to integrateseveral temporal steps and a great variety of co-ordinated signalling processes in eukaryotic cellreplication, fix them in a conserved DNA stor-age medium, and if necessary, resolve conserva-tion, change, rearrange, or newly construct thewhole genomic content and sequence order.76

The DNA information storage medium isand has to be edited. I predict a future discus-sion on how to refer to these editing agents,for example, as interactions of more or lesschemical molecules or as “non-random ge-netic change operators” (Shapiro 2007, per-sonal communication) or as natural genomeeditors.

From a bio-communicative perspective—which investigates combinatorial (syntactic),content-specific (semantic), and contextual(pragmatic) rules of genetic text processing—itis important to note that there can be no edit-ing without competent agents that edit, that is,an editor or most likely a swarm of editors.77

For example, the spliceosome or even the ribo-some works as an integrated network of severalsmall nuclear RNAs and their associated pro-teins.65 They are clearly authorities in compe-

tently acting upon the molecular syntax of theDNA language.

Life could not function without the keyagents of DNA replication, namely mRNA,tRNA, and rRNA. Not only rRNA, butalso tRNA and the processing of the pri-mary transcript into the pre-mRNA and themature mRNA, are clearly descended fromretro-“elements”78–82 with obvious retroviralancestry.

It is now possible to appreciate how so-phisticatedly the competent agents act in thecase of endogenous retroviral swarms, thathave reached a persistent and nonlytic lifestyle.We also know that all related retro-elementsshare a common genome-editing capacity sim-ilar to transposable elements. Nonetheless, itremains difficult to reconstruct how all theseDNA-encoded RNA agents reached a persis-tent status in hundreds, thousands, and tensof thousands of elements. We only know thatthey act in a precisely coordinated mannerthat would be impossible without competentsignalling. This includes a strict capacity forself/nonself identification, which is a major as-set of RNAs in general and of small nucleolarRNAs in particular.83

Persistent endogenous agents that are com-petent in natural genome editing apparentlyprefer a special kind of habitat characterizedas noncoding DNA sectors. They use a syn-tax mainly consisting of direct or inverted re-peats. They colonized DNA genomes by insert-ing their sites beside coding elements; then theyuse these coding elements for different needs.So we have to look at sequence orders thatconsist of noncoding elements, such as repeatsand coding elements. In the human genome,only 3% of coding regions remain. The remain-ing 97% serve as a habitat for persistent vi-ral operators that orchestrate a highly sophisti-cated division of labor. From these regions theycan actively regulate coding sequences becausethey are able to change specific DNA contentthroughout the genome. All eukaryotic DNAreplication processes share a cut-and-paste pro-cess in which noncoding elements, that is,


introns, are spliced out; later the remaining ex-ons that code for proteins are combined intoa coherent protein-coding content ready fortranslation.

As opposed to persistent endogenous agentsof natural genome editing in eukaryotes, wefind persistent exogenous agents in prokaryotesthat are competent in natural genome editingof the prokaryotic gene pool. This process haslong been visualized as horizontal gene transferand is now recognized as occurring via plas-mids, phages, and transposons, all with viralancestors.84

It is difficult to perceive mere molecules ormolecule buildings as being “competent” toprocess the sophisticated DNA language. It isless difficult to think of viruses as these subject-like agents.

Large Noncoding RNAs: CompetentRegulators of Gene Expression

Noncoding RNAs that function in gene regu-lation coordinate and organize various actions,such as chromatin modification and epigeneticmemory, transcriptional regulation, control ofalternative splicing, RNA modification andRNA editing, control of mRNA turnover, con-trol of translation, and signal transduction.85

In contrast to former opinions about the ex-pression levels of genomes it is now increasinglyclear that most eukaryotic genomes are highlyexpressed and a great abundance of noncod-ing RNAs with regulatory functions are tran-scribed. Most of these noncoding RNAs arealternatively spliced and divided into smallerRNAs that are integral parts of ribonucleopro-tein (RNP) complexes. They regulate nearly allaspects of gene regulation. Small RNA speciesinclude micro-RNAs, small interfering RNAs,small nuclear RNAs, and small nucleolar andtransfer RNAs.86

Although recent research has tried to evalu-ate the enormous regulatory networks of smallRNAs, the role of thousands of longer tran-scripts is not yet clear. We know that theyplay important roles in histone modification,

methylation, that is, epigenetic control of de-velopmental processes such as the mammalianHOX clusters,85 and also transcriptional in-terference, promoter inactivation, and effectson enzymatic pathways. Interestingly, theselarge noncoding RNAs are found as interlac-ing and overlapping sense and antisense tran-scripts derived from introns or intergenic re-gions, which means that they stem from thepreferred life habitat of persistent viral set-tlers. That they may be of viral descent isindicated by their developmental stage- andtissue-specific expression patterns, which aretypical habits of persistent viral settlers. Sim-ilar to their smaller relatives they are involvedin the formation of ribonucleoprotein (RNP)complexes.

Micro-RNAs and Their AssociatedProteins are RNPs

Small noncoding RNAs also share a spe-cial competence for epigenetic regulation ofgene expression87 and are derived from repet-itive genomic sequences.88 Repetitive genomicsequences indicate descent from retroviral in-fection events.89 The capacity for epigeneticregulation of gene expression includes the“recognition” (identification) of specific se-quences in other nucleic acids and is com-mon to RNAs,83 especially small nuclear RNAsand tRNAs that (i) identify splice junctions inboth pre-mRNAs and codons, and (ii) processboth the subunits of the spliceosome and theribosome.17 This implicates their capacity forself/nonself distinction as well as for identify-ing the molecular syntax. If one of these doesnot function, that is, error or damage occurs,the regulation or structural features of smallnoncoding RNAs do not function.

Maybe a key feature of noncoding RNAsis that they share an analog/digital language-competence,90 such that in their secondarymolecular structure they can act as molecularadaptors to the protein world, whereas theirnucleotide word order seems to be digitallystructured information. Their ability to edit


the molecular syntax of genetic texts accordingto different needs is exemplified by recent re-search on the functions of learning and memoryin mammalian neuronal networks.91,92 This ispossible in that RNA editing alters transcriptsfrom loci encoding proteins involved in neuralcell identity, resulting in DNA recoding.93

The action of endogenous RNA species suchas micro-RNAs, small interfering RNAs, andpiwi RNAs in animals and plants is a mediat-ing process in that they guide the binding ofprotein complexes to specific nucleic acid se-quences.94 There action starts both as a regula-tory process at the transcriptional level (e.g., byendogenous siRNAs)95 when the action poten-tial is activated out of the “evolutionary proto-cols” fixed in the DNA storage medium, or atthe post-transcriptional level in that they stabi-lize messenger RNA and translation into pro-teins.96 This means that small noncoding RNAsdo not solely mediate the transfer of genetic in-formation from DNA to protein but also act assequence-specific regulators in the expressionof other RNA transcripts and, interestingly, insilencing specific transposons.18,87

Control patterns include mRNA degrada-tion (siRNAs,), translational repression (miR-NAs), heterochromatin formation, and trans-poson control (piwiRNAs).87 EndogenousmiRNAs and siRNAs share biogenesis and canperform interchangeable functions.97

They cannot be distinguished by their chem-ical composition or their action. But they dif-fer in their production pathways: (i) miRNAsderive from genetic sequences that are differ-ent to known genes, siRNAS derive from mR-NAs, transposons, viruses, (ii) miRNAs are pro-cessed out of transcripts that can form RNAhairpins, siRNAs are processed from long bi-molecular RNA duplexes, (iii) miRNAs arealways conserved in related organisms, endoge-nous siRNAs are rarely conserved, (iv) miR-NAs are produced from genes that are special-ists in silencing of different genes, siRNAs aretypically auto silencing, such as viruses, trans-posons, and repeats of centromeres.18 SiRNAsare expressed from extended double-stranded

regions of long inverted repeats and can in-hibit the expression of nearly any target genein response to double-stranded RNA and havea very efficient and ancient immune functionagainst genetic parasites.98,99 This functions byidentifying foreign RNA sequences and inhibit-ing their replication. The ancient RNAi im-mune function is based on self/nonself identi-fication competence.98 Many of these elementsare retroposons or transposons and are encodedin the repetitive sequences of the genome99 andtherefore are clearly of viral origin.

Micro-RNAs are single-stranded RNAs of19–25 nucleotides in length and are generatedfrom endogenous hairpin transcripts of 70 nu-cleotide precursor miRNAs.100 The transcrip-tion of this pre-miRNA is processed by RNApolymerases pol II and pol III. Whereas polII produces the messenger RNA, small nucleo-lar and small nuclear RNAs of the spliceosomepol III produce shorter noncoding RNAs, suchas tRNAs, some rRNAs, and a nuclear RNAthat is part of the spliceosome.18 They controlnot only developmental timing, hematopoiesis,organogenesis, apoptosis, and cell proliferationbut also fat metabolism in flies, neuronal pat-terning in nematodes, and control of leaf andflower development in plants.18 Most of themare processed out of introns!

It is predicted that every metazoan cell typeat each developmental stage has a distinctmiRNA expression profile.18 The most charac-teristic differences of miRNAs acting in plantsand animals are found in the stem loop.

Micro-RNAs, as well as small interferingRNAs, seem to have descended from transpos-able elements with an inherent regulatory ratioon gene regulation that is fulfilled by a vari-ety of small interfering RNAs or micro-RNAsthat act in a coordinated manner in that theyshare a division of labor in hierarchical stepsof suppression and amplification. This is in-dicated in transposable elements that encodeboth siRNAs and miRNAs.101 They can befound in intronic regions and build stem loopstructures (hairpins) as a common feature ofactive RNA species, such as ribozymes. The


defense mechanism of host genomes againsttransposable element invaders through siRNAevolved into miRNAs with a new regulatorycomplexity and a new phenotype. First evolvingas an immune function, it was later co-opted asa tool for complex regulatory pathways for hostgene expression.101 This co-option of identicalcompetences for different purposes seems to bea common evolutionary pattern for regulatorycontrols that can be flexibly altered and rear-ranged to cause phenotypic variation withoutaltering basic components.102

Micro-RNAs are acknowledged as key reg-ulators of gene expression. This means thatdysfunctions of these regulatory mechanismsmay lead to dysregulated genes with a cas-cade of disease-causing consequences.103 Thismeans that the smallest RNAs are the basisof gene regulation, although in their originalfunction they had an immune function againstviruses and similar agents. Later on their func-tion was adapted for eukaryotic gene expres-sion. Sometimes these new functions over-lap with old features when virus infection isacute104 and the balanced (persistent) statusis disturbed. The eukaryotic signalling path-ways that act during gene expression or DNAreplication and the role of transport from thecytoplasm through nuclear pores into the nu-cleoplasm, genetic expression, and retransportinto the cytoplasm and the role of membranes,show that viruses had, and still have, ma-nipulatory abilities in host genomes and hostcells.105 In particular, the ability of persistentinvaders to self-splice out before export to thecytoplasm—an ability derived from the divi-sion of transcription and translation throughnuclear membranes with pores—produces ananalog protein-coding data set ready for trans-lation and determines the identity of the hostwithout harming its reproductive cycle.

RNPs and Their Functions

The complex network of all RNA inter-actions and the great variety of functions aspreconditions of eukaryotic complexity include

gene-gene interactions as well as the integra-tion and regulation of most of the gene activi-ties on different levels, such as chromatin struc-ture, DNA-methylation, transcription, RNAsplicing, RNA translation, RNA stability, andRNA signalling. This means that most of thefunctions of gene control that we currentlyknow17,66,67,73 descended from retro-elements.One important feature is that noncoding RNAsfunction as RNPs and not as naked RNAs.106

Some of them are small nuclear and smallnucleolar RNAs (see below). Interestingly, thenucleolus plays an important role with simi-lar performance to other endogenous retro-elements during a small time window in thecell cycle. Nucleoli appear during the inter-phase of mammalian cells and are the locus ofribosome processing. Nucleoli also control reg-ulation of the cell cycle.107 In general, noncod-ing RNAs play important roles as intermediateproducts in the function and structure of thehost genome, for example, in the reproductivecycle of host cells from the start to the end ofthe S-phase. This is a common feature of per-sistent endogenous retroviruses that adapted tothe host genome as an addiction module. Theynow represent a phenotypic function of thehost organism as neutralized (balanced) formercompeting viral agents. If one of these agentsis damaged or its regulation becomes unstablethe counterpart may still become virulent withpotential disease-causing consequences for thehost organism.89

Regulatory functions are restricted to spe-cific phases of the cell cycle. One antago-nist is suppressed and therefore the other canbe produced at increasing rates according tothe phase of the cell cycle, for example, pro-duction of tubulin and/or nanotubulin struc-tures108 or production of proteins with a va-riety of functions in cell division. At the endof this specific phase of the cell cycle the sup-pression of the antagonist ceases, mediated byspecific signalling processes, and the antagonistthen suppresses the increased production of itscounterpart. This is the precondition for termi-nating its function in this phase of the cell cycle.


These interactions represent a kind of universalmodule with key regulatory function. Noncod-ing RNAs are involved in nearly all of these keyfunctions of cellular activities in all domains oflife (bacteria, eukarya, archaea).

Small Nuclear and Small Nucleolar RNPs

Most of these functions are fulfilled bynoncoding RNAs, which act as binding part-ners to ensure the correct position of the nucleicacid target molecule for its enzymatic functions.Normally this process works as Crick-Watsonbase-pairing which includes a lot of proteins.These interconnected networks between non-coding RNAs and proteins are termed theRNPs.106

Some of these noncoding RNPs are small nu-clear (sn) and small nucleolar (sno) RNPs. Theyare competent in multiple functions, many ofthem for intracellular transport and motility.In human cells they are encoded within in-trons. Both act in a complementary fashion:snRNPs retrieve the snoRNAs from the in-trons. On the other hand, snoRNAs are re-quired by the snRNPs for posttranscriptionalmodifications. Their interdependency seems tobe typical for viral derived addiction modules.Both snRNPs and snoRNPs depend on stablebut inactive pre-RNPs, which only can matureif they are located far from their later activeposition.106

In most vertebrates, the snoRNAs comefrom introns of pre-RNA transcripts. ThesesnoRNAs are integrated into a complex struc-ture by endonucleases, exonucleases, and he-licases.109,110 Interestingly, in the yeast Saccha-

romyces cerevisiae, most snoRNAs are not encodedin introns.83 This could be an indicator of anearlier evolutionary phase in which snoRNAshad not yet become endogenous by an infectionevent by retro-agents.

Most of the snoRNAs of vertebrates thatare encoded in their introns are transcribedby polymerase II and produced after splic-ing through exonucleolytic trimming. They are

highly lineage specific and form a separate fam-ily of mobile genetic elements.111

Genes coding for snoRNAs switch on hostgenes by retroposition. They also play impor-tant roles in excision and integration of spe-cific sequences. They are as important as Alu,SINEs, and LINEs.111 Interestingly, the best-known mobile genetic elements are integratedinto snoRetroposons. The insertion is char-acterized by their individually different tar-get site duplication-repeats, for example, full-length LINEs, that is, ALUs in primates. snoRe-verse Transcriptases (snoRTs) are also insertedinto other mobile elements as well as intoDNA transposons.111 Retro-elements seem tobe compatible with each other and are inte-grated into small nucleolar reverse transcrip-tase sequences, each of them with its indi-vidual target-site duplication repeat.111 Thiscould be an indicator of identification processesand identity modules. Many of the snoRNAgenes are retroposons with retroviral ances-try.111 snoRTs that are part of the introns ofhost genes can be used as RNAs with novelfunctions. After a retroposition event they cansilence the snoRNA copy of their parents andlead to a new function of the new snoRNA.111

This seems to be a common feature of en-dogenous retroposons and transposons in thatthese genetic invaders are driving forces of ge-nomic creativity: the same competences in dif-ferent contexts may lead to new functions andphenotypes.

Transport of small nucleolar RNPs ischaracerized by both nuclear and cytoplasmiccycles. Cajal bodies play important roles in in-tranuclear transport as centers for RNP as-sembly, transport, modification, and editing.Their production is a complex highly coordi-nated and conserved process. When the smallnuclear RNPs arrive in the nucleoplasm theyspread over the whole interchromatin space.Newly generated RNPs are assembled in theCajal bodies before they are integrated into thenucleolar subdomains fibrallin and interchro-matin clusters. This seems to be an indicatorthat they are processed and modified within


the Cajal bodies and specialized (informed) forspecific functions.106 In addition to their func-tion as de novo assembling factors of RNPs, mul-ticompetent Cajal bodies play an importantrole in modification and recycling processes ofU4/U6 snRNP complexes that are remnantsof splicing processes.106 Cajal bodies are nu-clear organelles with high motility and impor-tant roles in splicing, ribosome processing, andtranscription.112 It has been observed that Ca-jal bodies move from one end of the nucleus tothe other, assemble with other Cajal bodies, ordivide themselves into smaller ones.113 Theseendogenous retro-elements are still importantagents in natural genetic engineering and nat-ural genome editing.

Important Roles of Small Nucleolar RNPsin Eukaryotic Genome Editing

Small nucleolar RNAs seem to derive froma very ancient group of endogenized RNAviruses with a wide variety of functions such as2′-O-methylation, pseudouridylation of manyclasses of RNAs, rRNA processing, and synthe-sis of telomeric DNA.114

Today we know two families of snoRNAs:the C/D and the H/ACA RNAs. They areproduced in the nucleolus and have a variety offunctions within the nucleolus but also functionout of the nucleolus as a recreation center forspecial substances. Eukaryotic cells have severaldozen species of snRNAs and 200 known snoR-NAs (C/D and H/ACA RNAs). These RNAsare one of the most diverse transacting RNAscurrently known. They are available not only ineukaryotes but also in archaea. These snoRNPs(C/D H/ACA) share important functions suchas: protein translation, mRNA splicing, genomestability, ribosome function, and modificationsin snRNAs of eukaryotes, tRNAs in archaea,and neuronal mRNAs in mammals. One kindof H/ACA RNA, telomerase RNA, is neededfor telomere production.115

Most of the known snoRNPs guide modifica-tions of other ncRNAs. Both motifs are simpleand very ancient and are part of the telomerase

RNA.116,117 Depending on their function, C/Dand H/ACA finally orientate to nucleoli, Cajalbodies, or telomeres.106

We know that one H/ACA RNP functionsin vertebrate telomere synthesis. TelomeraseRNA and its protein partner, telomerase re-verse transcriptase, are both target orientatedand strictly regulated. Telomerase enzymes areactive in telomere elongation only in the S-phase of the cell cycle. Interestingly, telomerasedoes not elongate telomeres continuously butat different times with interruptions in betweencellular subcycles, especially in the shorter ones.The 3′ region of the endogenous human telom-erase RNA possesses all the structural fea-tures of H/ACA boxes of snoRNAs. Only inthe S-phase of the cell cycle do both telom-erase RNA and telomerase reverse transcrip-tase move toward telomeres, otherwise they arein different (waiting?) positions.118 The accu-mulation of human telomerase RNA in nucle-oplasmic Cajal bodies in certain cancer cellsoccurs only during the S-phase of the cell cy-cle when telomerases are generated.119 Thismeans that the accumulation process is cell cy-cle dependent and time limited. Telomerasetransport in the S-phase of the cell cycle is ad-vantageous because telomerase activity in eu-karyotes is limited to chromosome replicationand suppresses destructive functions of telom-erases at nontelomere elongation locations suchas chromsome repair and repair of double-strand breaks.113

Cajal bodies in particular play importantroles in the generation and function of telom-erase RNPs. Human telomerase RNA has anH/ACA motif that it shares with small Cajalbody RNAs and a wide variety of snoRNAs.The snoRNA family regulates modificationsand cleavage of ribosomal RNAs in the nu-cleolus, and the small Cajal body RNA familyregulates modifications of small nuclear RNAswithin the Cajal body.120

Besides the nucleolus, the Cajal bodiesare the best investigated nucleoplasmatic or-ganelles. They are enriched by spliceoso-mal and nucleolar RNPs. Especially in the


interphase, many nucleolar intermediatesemerge and play important roles in trans-port and modification of a variety of cellularRNAs.121 C/D and H/ACA boxes guide theposttranscriptional pol-II specific spliceosomalsnRNAs via pseudouridylation, which is char-acteristic of posttranscriptional RNA modifi-cations in eukarya and archaea and generallyplays an important role in the correct func-tioning of cellular RNAs. These guide RNAsshare a Cajal body-specific localization signal,the CAB Box.122 It is suggested that pseu-douridylation guide RNPs play important rolesin the processing of rRNA and in function-control/regulation of telomerases in eukary-otes. Ribosomal RNAs play important rolesand therefore are DNA encoded in hundredsof transcription units. These units are orga-nized in large tandem arrays. In active synthe-sis these rRNA loci form nucleoli with impor-tant roles in evolution and development. TheserRNA loci are habitats for mobile elements likeretroposons.123 Pseudouridylation pockets alsoseem to direct snoRNAs to the nucleolus, whichmeans that they have identity functions.124

In yeast, telomerase is not associated withH/ACA snoRNPs but with SM snRNPs, whichdemonstrates that telomerase is generated andregulated in different organisms in differentways.106

The Enigma of tRNA

Parallel to the ribosomal rRNAs, tRNA isone of the most important RNAs and is essentialfor the replication of both RNA genomes andDNA genomes. It seems to be a very ancientcompetence because it functions in a wide vari-ety of contexts in quite different ways. Investiga-tions of the variety of functions of tRNA showthat it seems to be an addiction module-likeassociation of different predecessors dating tothe RNA world. The 3′ end of the tRNA struc-ture could be the start of an ancient replicase.Replicase is the CCA-adding enzyme that isnecessary to complete nucleotides that were lostbecause of incorrect starts. This CCA-adding

activity has been the first telomerase functionwith its function of completing lost nucleic acidend sequences during replication. If it is reallyas old as suggested it must be present in all threedomains of life: archaea, bacteria, eukarya.These CCA-adding enzymes are found in allthree domains of life, are part of the same nu-cleotidyltransferase superfamily, and are verysimilar in their function.79,80

The genomic tag hypothesis is another partof the puzzle over the role of noncoding RNAabilities. It is suggested that the one half oftRNA evolved to mark single-stranded RNAgenomes for replication in the early RNAworld. The second half of the tRNA evolvedseparately as a primer of templated proteinsynthesis and started the RNP world. Bothparts derived from independent RNA agents,which together built a kind of addiction mod-ule. Then this module became involved intranslation from RNA template into protein.That tRNA plays important roles in the repli-cation of single-stranded RNA viruses of bacte-ria, plants, and mammals, replication of duplexDNA plasmids of fungal mitochondria, retrovi-ral replication, and also replication of presentchromosomal telomeres, is less noticed nowa-days.79,80

Ancient and Prominent: ReverseTranscriptases

Last but not least let’s have a look at one of themost prominent and ancient natural genomeediting agents: reverse transcriptases, such astelomerases that function in telomere mainte-nance are noncoding retro-elements. This in-dicates that their ancestry is of retroviral ori-gin.118 The vast majority of retro-elements usetRNAs or RNAs with strong secondary struc-tures to process reverse transcription. Interest-ingly, there is a similarity between the functionof tRNA in the production of proteins fromRNA information, and reverse transcriptase,an enzyme that is crucial for turning RNA intoDNA.91


Retroposition—a billions of years-oldprocess—still plays important roles in buildingthe structure and function of genomes in acontinuing interaction between host genomesand the colonizing life strategies of mobile ge-netic invaders, an everlasting evolution-drivingprocess of rearrangement, renovation, andinnovation.125

Copying from RNA into DNA generally in-volves reverse transcriptases. Recent researchhas demonstrated that overlapping epigeneticmarking in eukaryotic cells is an importantevolutionary feature to silence the expressionof mobility of these mobile elements.72 Mo-bile elements can silence single genes as wellas larger chromosomal regions and, therefore,play an important role in the evolution of di-versity. They share the ability to recombine,rearrange, repair, and insert into genomic con-tent with other retro-elements.126,127 They in-fluence neighboring genes through alternativesplicing and are active agents as enhancers andpromoters or act by polyadenylation patterns.72

Reverse transcriptases play key roles in mo-bile elements, such as transposons and retro-posons, both of viral origin. One type of retro-poson has direct repeats at its ends (LTR),whereas others do not (non-LTRs). Interest-ingly, the number of retroposons increases withevery transposition (transposition duplication)so that they can expand genomes: LINE-1is 20% of the human genome. In contrast,transposons contain a code for the transposaseprotein. This enzyme identifies the terminal in-verted repeats that flank mobile elements, ex-cises them and integrates itself instead. Thegap at the donor site is repaired in a cut-and-paste transposition or is filled up with a copyof the transposon by a gap repair technique.72

Transposons can also integrate themselves inphages and plasmids and are transferred withthem into other cells.84 This is evidence forself/nonself differentiation.

In contrast to nonmobile telomeres and cen-tromeres, mobile sequences, such as trans-posons and retroposons, 128 and noncodingrepetitive elements, such as LTRs, SINEs, and

LINEs, enable far-reaching DNA rearrange-ment and reorganization.64,129,130 Together,they play a decisive role in the evolution ofnew genomic structures.130–132 The repetitivesequences are highly species-specific and aremore suitable for determining the identity ofspecies than the coding sequences.4

This does not mean that only mobile se-quences represent ancient genetic settlers. Thesimilarity of telomeres and centromeres—nonmobile repeat elements—in descent andtheir relatively poor loci of inverted repeats orretro-elements could indicate an ancient im-mune function that protects both from massiveinvasions by genetic parasites.133,134

Major Roles of Reverse Transcriptases inNatural Genome Editing

In addition, reverse transcriptases play keyroles in altering genomic structures135 andtherefore, in evolutionary processes facilitatedby natural genome editing. Reverse transcrip-tases are used to generate (i) copies of mRNAsthat they need for integration into a genome,and (ii) copies of non-mRNAs, such as smallnucleolar RNAs, one of the largest classes ofnoncoding RNAs,109 which, like DNA copies,are SINEs. SINEs can initiate new genes thatcode for small RNAs with regulatory abilitiesin existing genes.

One further key feature of reverse tran-scriptases is that they are a primer for retro-posons such as LTRs (copia, gypsy, Ty1, IAPs,HERVs). Non-LTRs (Het-A/TART, SINEs,LINEs) act like telomerases in several arthro-pods and plants. Moreover, reverse transcrip-tases are encoded and used by open read-ing frames (ORF), ORF1 (an RNA-bindingand shuttling protein), ORF2 (endonucle-ase, reverse transcriptase activities), as wellas ALUs (manipulation of LINE-1 functionfor mobilization), group II self-splicing in-trons and snoRNAs (type 1–3 retroposons),all of which have important regulatory func-tions.106,111,113,136 Reverse transcriptases arealso found in retroviruses of mammals andbirds, in the hepadnavirus of mammals and


birds, and the caulimovirus of plants, in LTRretroposons of animals, plants, fungi, and pro-tozoa, in non-LTR retroposons of animals,plants, fungi, and protozoa, in group II intronsof bacteria, fungi, plant mitochondria, chloro-plasts, and plastids, in mitochondrial plasmidsof Neurospora mitochondria, and in multiplesingle-stranded DNAs.4,137 Although many re-searches believe that non-LTRs evolved be-fore retroviruses, there is recent evidence thatviruses are their ancestors.142

Reverse transcriptases (DNA polymerases)together with RNA-dependent RNA poly-merases replicate positive-strand RNA viruses,double-stranded RNA viruses, negative-strand RNA viruses, and retroviruses. RNA-dependent RNA polymerases produce dsR-NAs in that they copy single-stranded RNAtemplates into dsRNAs. The RNA-dependentRNA polymerases are initiated by two or morecomplementary micro-RNA sites72 that couldindicate an addiction module because of itscounterpart regulation. These polymerases arealso involved in the coupling of heterochro-matin for the production of siRNAs.138 TheRNAi system is competent in posttranscrip-tional gene silencing and is, therefore, a cru-cial instrument in keeping the balance be-tween the need for expression and the need forsilencing.139

As mentioned above, ORFs also code forreverse transcriptase. Many organisms haveORFs that code for proteins with sequencesvery similar to retroviral reverse transcrip-tases.140,141 RNA-dependent DNA polymerase(reverse transcriptase) is related to RNA-dependent RNA polymerase. Rooting theselines of descent in RNA-dependent RNA poly-merases yields two groups: (i) group 1 containsLTR retroposons, RNA viruses, DNA viruses;(ii) group 2 contains non-LTR retroposons,bacterial and other organelle parts.73

The telomerase function is cell-cycle reg-ulated. It functions only when its suppres-sion is removed. Once the telomerase func-tion in telomere replication is fulfilled, a sig-nal initiates its suppression again. A disturbed

signalling process may lead to uncontrolledcell replication. Telomerase has to be trans-ported to telomere repeats for its elongationduring the S-phase of the cell cycle. The de-livery agents are, again, Cajal bodies, smallnucleolus-like organelles that are competent in(i) splicing, (ii) ribosome production, and (iii)transcription.112,119

Natural Genome Editors:Communal Interacting Agents

with Persistent Status

In contrast to DNA with its stable featuresand enormous information storage potential,RNA is involved in the active parts of copy-ing and coding processes, as demonstrated innew sequence generation, replicative processes,gene invention, and higher order regulations inall key processes of life. Prior to the evolutionof cellular life it is proposed that very simplestructured RNA (pre-RNAs) started by grow-ing through base-pairing mechanisms withoutcoding features. The selection of an RNA popu-lation, a direct product of error prone uneditedRNA replication, is known as the quasispeciestheory.

Growth by base-pairing mechanisms is dif-ferent to the growth of primitive secondarystructures of single-stranded RNAs, which canstabilize and replicate themselves as hair-pin/stem loop structures with inherent codingcapabilities. When coding began, catalyticfunctions connected with syntactic rule-ordered information enabled these simplemolecules to act as semiotic agents: they be-came capable of generating nucleic acid se-quences with a functional meaning that hadto be recognized, identified, and interpretedcorrectly in the situational context, with acombinatorial pattern of the base pairs thatdiffered from the diverse features inherentin nonself agents of the same structure, thatis, self/nonself identification.142 Biotic compe-tences differ from abiotic interactions, becausein contrast to the latter, biotic competences


which became active may even fail. These abili-ties are still evident in the t-loop structure of tR-NAs, a variety of ribozymes, and self-enforcingRNAi loops, which couple heterochromatin as-sembly to siRNA production.138

The high density of early RNA life led tocompeting situations in which it was an advan-tage to escape into DNA informational storageand protein-based cellular life as outlined inanother article.143 Protein and DNA inventionwere a prerequisite for the evolution of diver-gence and the variety of life because all evolu-tionary inventions could be stored as evolution-ary protocols77 in this stable storage medium.The information content of the human genomeis comparable to an archive of 5000 books with300 pages each.

It appears that all the detailed steps of evolu-tion stored in DNA that are read, transcribed,and translated in every developmental andgrowth process of each individual cell dependon RNA-mediated processes, in most cases in-terconnected with other RNAs and their as-sociated protein complexes and functions in astrict hierarchy of temporal and spatial steps.It is clear that this regulatory order could notevolve by chance or that it represents solely arandomly derived mixture of nucleotides, butthat it is composed of individual functions andintegration into one developmental target, instrict coherence to the syntax of the nucleicacid language.

Today we are beginning to realize the de-gree of abundance and variety of RNA specieswith their different, sometimes complementaryand competing roles in all key processes of life.RNAs play complementary roles in informa-tion processing and regulation.144 In most casesthey have an inheritable status, being integratedin the genome of organisms, and are termed en-dogenous. In other cases they are ancient indi-viduals living in the cytoplasm of cells as persis-tent nonlytic parasites similar to DNA settlers,with important endosymbiotic roles. Their rela-tion to viruses is close and some virologists con-sider an evolutionary tree of RNA species andRNA viruses. Interestingly, some DNA viruses

have other features (linear chromosome, telom-ere ends, intron-like structures) that indicate adifferent origin with ancient roots comparableto RNA viruses. These features connect the evo-lutionary roots of archaea and eukaryotes, be-cause ancient dsDNA viruses have similar fea-tures in viruses of archaea and the eukaryoticnucleus. A special feature is the RNA proof-reading and repair ability of RNA polymerases,which would be the precondition of an RNAgenome in the early RNA world because of therelatively unstable RNA structures.145–147 Onthe other hand, this instability is a necessaryprecondition for the high productivity of differ-ent RNA sequences with the rapidly adaptingfeatures necessary in the early RNA world, thatis, a large variety of different RNA identities.

In particular, the mode of replication of eu-karyotes shows a wide variety of hierachicalordered processes that each depend on sig-nalling processes to indicate the successful ter-mination of the preceding process. A tempo-ral order exists, with time windows processedby different process design connected with thecell cycle, along with transport systems of sig-nals and complex messages, agents, co-agentsand helpers, such as ancillary proteins anda network of interwoven regulatory elementswhich suppress or amplify the start and stops ofsemioses, production, regulation, and the wholetoolbox of natural genetic engineering. We nowknow that all these processes involve RNAs,which become active via transcription out ofthe DNA storage medium prior to translationinto proteins.

During translation from digital DNA stor-age into the analog code of protein languageit is interesting that DNA information containsmultiple RNA- and protein meanings, that is,from the same genetic data set it is possible totranscribe multiple RNA species or translate avariety of proteins according to the higher or-der regulations inherent in epigenetic controland/or transcriptional, pre- or posttranscrip-tional modification targets. Additionally, in eu-karyotes we find some noncoding RNA species,such as small nuclear, small nucleolar, small


Cajal body, and small interfering RNAs. Thereare some indications that noncoding DNAplays a role in the de novo generation ofgenes.148

Nearly all genetic and genome editing pro-cesses involve RNAs, which in most casesfunction as a network. Most of the functionsare performed, conducted, and regulated bynoncoding RNAs, which are encoded in in-tronic DNA in most cases with repetitive syn-tax. Most of their functions are active only in anintermediate stage of RNA processing, whichis regulated by strict starts and stops that aresignal mediated. Error in these regulations isinterconnected with organismic diseases.

Noncoding RNAs are similar to all kinds ofmobile genetic elements such as LTRs, non-LTRs, SINEs, LINEs, snoRNAs, and snR-NAs, all of persistent viral origin. They areintegrated via addiction modules, that is, anagent/antagonist relationship between com-peting genetic settlers that are neutralized andbalanced by their antagonism and the immunesystem of the host. This is an advantage forboth the host, which attains a new genetic phe-notype that noninfecte relatives do not possess,and the viral settlers, which both survive andco-evolve within a new genomic habitat.

Transfer RNA (tRNA) and ribosomal RNA(rRNAs), with their key functions in proteintranslation, represent such addiction modules,containing several subunits, each of them nec-essary for function. The whole cell cycle of eu-karyotes is regulated by these noncoding RNAs.

Obviously DNA functions as both a relativelystable information storage medium, an evolu-tionary protocol to fix advantageous innova-tions, and as a comfortable habitat for persis-tent genetic settlers. If all the RNA capabilitiesderive from viruses or similar agents that com-pete in the available global pool of organismalgenomes—viruses and their relatives are tentimes more abundant than cellular genomes—then only those that give their hosts an advan-tageous genomic identity that is able to wardoff an abundance of competing genetic settlerswill survive. They must be able to build addic-

tion modules (genetic and genomic innovations)together with the host immune system, each ofthem a unique culture-dependent habitat. Onlythen will the survival of both the genetic settlersand their host populations be likely.

Conclusion

The biocommunicative approach investi-gates both communication processes withinand among cells, tissues, organs and organ-isms, as sign-mediated interactions, and nu-cleotide sequences as code, that is, language-like text, which follows in parallel three kindsof rules: combinatorial (syntactic), context-sensitive (pragmatic), and content-specific(semantic). From the bio-communicativeperspective editing genetic text sequencesrequires—similar to signalling codes betweencells, tissues and organs—biotic agents that arecompetent in correct sign use according tocombinatorial, context-sensitive, and content-specific rules. Otherwise, neither communica-tion processes nor nucleotide sequence genera-tion or recombination can function. Even if theprocess of following these rules is very conser-vative, variations, alterations, and change mayoccur. The agents rule following identities mayfail, and nucleic acid sequence grammar can bedamaged or deformed and may become syntac-tically incorrect.

At the genomic level we can identify a re-markable process of change from a mechanisticview to the perspective of nonmechanistic ge-netic content processing. If we look at the cur-rent knowledge of hierarchical and temporalorder of single steps and substeps in replicationand transcription processes there must be nat-ural genome editing agents that are competentboth in generation of meaningful nucleotidesequences and in the use of these sequencesaccording to different needs, such as integra-tion, modification, recombination, and extrac-tion into preexisting genetic texts. As we willsee, in this respect DNA is not only an informa-tion storing archive but a life habitat for nucleic


acid language competent RNA agents of viralor subviral descent. These agents are compe-tent in almost error-free editing of nucleotidesequences according to combinatorial, context-sensitive, and content-specific rules. They evengenerate nucleotide sequences de novo. Theyare also able to generate new rules of use fornucleic acid sequence modules by rearrange-ments in the higher order regulatory network ofnoncoding domains. Thus ancient sequencesof the DNA storage medium may be used asmodular tools in a wide variety of different con-texts for new functions, made possible throughthe different meaning of syntactically identicalsequences.

Competent agents in nucleic acid languageare not solus ipse agents but are competentas mutual or parasitic “swarms” or “clouds,”most of them RNA-based communities thatshare these competences. Their competence isa communal one, each of them being capableof self and nonself identification. The interac-tive competence of a community enables eachindividual to be competent. If we look at in-teracting communities, such as ribosomes andspliceosomes (each containing subunits withoutwhich they cannot function), we see their com-munal competence. If we look at the hierarchi-cal processes of gene expression, transcription,RNA processing, mRNA and tRNA transportfor translation we can also see communally act-ing agents. From the virus-first perspective theyare now mutually interacting but may derivefrom formerly competing agents. mRNA andtRNA maturation in eukaryotes in particularalso seem to reflect communal processing.

Formerly competing agents have reached anequilibrium status balanced by the immune re-sponse of the infected host to achieve a persis-tent lifestyle in the host genome, for example,toxin/antitoxin-modules. The number of com-munal interacting agents represented by ribo-somes, spliceosomes, or even the consortiumthat cooperates as the adaptive immune sys-tem139 ranges from a few to hundreds and thou-sands. In the latter case communal agents in-teract in DNA rearrangements with enormous

consequences for many protein-based productsthat play important roles in immune functions.

This view could change the construction ofresearch projects, that is, shifting the focus frommutational (random) changes of nucleotide se-quences to investigating nucleotide sequencesfrom the perspective of viral-derived sequencesthat now play important roles in the regulationof cellular functions. Their status within oneof many addiction modules can be changed bynonbeneficial circumstances for the cell (e.g.,stress) and they may become lytic again, result-ing in a wide variety of diseases.

Acknowledgments

I am grateful to Luis P. Villarreal, Director ofthe Center of Virus Research at the Universityof California–Irvine for critical comments andhelpful suggestions.

Conflicts of Interest

The author declares no conflicts of interest.

References

1. Claverie, J.M. 2005. Fewer genes, more noncodingRNA. Science 309: 1529–1530.

2. Bell, P.J.L. 2001. Viral eukaryogenesis: Was the an-cestor of the nucleus a complex DNA virus? J. Mol.

Evol. 53: 251–256.3. Bell, P.J.L. 2006. Sex and the eukaryotioc cell cycle is

consistent with a viral ancestry for the eukaryotic nu-cleus. J. Theor. Biol. doi: 10.1016/j.jtbi.2006.05.015.

4. Villarreal, L.P. 2005. Viruses and the Evolution of Life.ASM Press. Washington.

5. Forterre, P. 2006. The origin of viruses and theirpossible roles in major evolutionary transitions. Vir.

Res. 117: 5–16.6. Forterre, P. 2006. Three RNA cells for ribosomal

lineages and three DNA viruses to replicate theirgenomes: A hypothesis for the origin of cellular do-main. Proc. Natl. Acad. Sci. USA 103: 3669–3674.

7. Takemura, M. 2001. Poxviruses and the origin ofthe eukaryotic nucleus. J. Mol. Evol. 52: 419–425.

8. Van Lent, J.W.M. & C. Schmitt-Keichinger. 2006.Viral movement proteins induce tubule formationin plant and insect cells. In Cell-Cell Channels. F.Baluska, D. Volkmann & P. Barlow, Eds.: 1–13.


Landesbioscience-Springer Verlag, Berlin, Heidel-berg, New York.

9. Bapteste, E., R.L. Charlebois, et al. 2005. The twotempos of nuclear pore complex evolution: Highlyadapting proteins in an ancient frozen structure.Gen. Biol. 6: R85.

10. Witzany, G. 2007. The agents of genomic creativity.In Biosemiotics in Transdisciplinary Contexts. Proceedings

of the Gathering 6 in Biosemiotics. G. Witzany, Ed.: 95–103. Umweb. Helsinki.

11. Forterre, P. 2001. Genomics and early cellular evo-lution. The origin of the DNA world. Comptes rendus

de l’Academie des sciences. Serie 3, Sciences de la vie 324:1067– 1076.

12. Forterre, P. 2002. The origin of DNA genomes andDNA replication proteins. Curr. Opin. Microbiol. 5:525–532.

13. Forterre, P. 2005. The two ages of the RNA world,and the transition to the DNA world: A story ofviruses and cells. Biochimie 87: 793–803.

14. Koonin E.V., T.G. Senkevich & V.V. Dolja, 2006.The ancient virus world and evolution of cells. Biol.

Dir. 1: 29.15. Tran, E., J. Brown & E.S. Maxwell. 2004. Evolu-

tionary origins of the RNA-guided nucleotide mod-ification complexes: From the primitive translationapparatus? Tren. Biochem. Sci. 29: 343–350.

16. Gorinsek, B., F. Gubensek & D. Kordis. 2004. Evo-lutionary genomics of chromovirus in eukaryotes.Mol. Biol. Evol. 21: 781–798.

17. Mattick, J.S. 2003. Challenging the dogma: Thehidden layer of non-protein-coding RNAs in com-plex organisms. Bioessays 25: 930–939.

18. Bartel, D.P. 2004. MicroRNAs: Genomics, biogen-esis, mechanism and function. Cell 116: 281–297.

19. Cullen, B.R. 2006. Viruses and microRNAs. Nat.

Gen. 38: 25–30.20. Ryan, F.P. 2004. Human endogenous retroviruses in

health and disease: a symbiotic perspective. J. Roy.

Soc. Med. 97: 560–565.21. Ryan, F.P. 2006. Genomic creativity and natural

selection: A modern synthesis. Biol. J. Linn. Soc. 88:655–672.

22. Ryan, F.P. 2007. Viruses as symbionts. Symbiosis 44:11–21.

23. Gerdes, K. 2000. Toxin-antitoxin modules may reg-ulate synthesis of macromolecules during nutritionalstress. J. Bac. 182: 561–572.

24. Makarova, K.S., N.V. Grishin & E.V. Koonin.2006. The HicAB cassette, a putative novel, RNA-targeting toxin-antitoxin system in archaea and bac-teria. Bioinform 22: 2581–2584.

25. Witzany, G. 2007. The Logos of the Bios 2. Bio-

Communication. Umweb. Helsinki.26. Rashkova, S., S.E. Karam, R. Kellum et al. 2002.

Gag proteins of the two Drosophila telomeric retro-transposons are targeted to chromosome ends. J.

Cell Biol. 159: 397–402.27. Peterson-Burch, B.D. & D.F. Voytas. 2002. Genes

of the Pseudoviridae (Ty1/copia retrotransposons).Mol. Biol. Evol. 19: 1832–1845.

28. Tang, Y., U. Winkler, E.O. Fredd, et al. 1999. Cel-lular motor protein KIF-4 associates with retroviralgag. J. Virol. 73: 10508–10513.

29. Gao, X., E.R. Havecker, P.V. Baranov, et al. 2003.Translational recoding signals between gag and polin diverse LTR retrotransposons. RNA 9: 1422–1430.

30. Sfakianos, J.N. & E. Hunter. 2003. M-PMV capsidtransport is mediated by env/gag interactions at thepericentriolar recycling endosome. Traffic 4: 671–680.

31. Ivanyi-Nagy, R. & J.L. Darlix. 2008. Ancient retro-transposons as possible remnants of the prim-itive RNP World. In Retrotransposition, Diversity

and the Brain. F.H. Gage & Y. Christen, Eds.Springer. Berlin/Heidelberg, doi: 10.1007/978-3-540-74966-0_4.

32. Shadan, F.F. & L.P. Villarreal. 1996. The evolutionof small DNA viruses of eukaryotes: Past and presentconsiderations. Vir. Gen. 11: 239–257.

33. Nisole, S. & A. Saib. 2004. Early Steps of retrovirusreplicative cycle. Retrovir. 1: 9.

34. Greber, U.F. 2002. Signalling in viral entry. Cell Mol.

Life Sci. 59: 608–626.35. Greber, U.F. 2005. Viral trafficking violations in ax-

ons: The herpesvirus case. Proc. Natl. Acad. Sci. USA

102: 5639–5640.36. Vale, R.D. 2003. The molecular motor toolbox for

intracellular transport. Cell 112: 467–480.37. Cochrane, A.W., M.T. McNally & A.J. Mouland.

2006. The retrovirus RNA trafficking granule: Frombirth to maturity. Retrovir. 3: 18.

38. Carlton, J.G. & J. Martin-Serrano. 2007. Paral-lels between cytokinesis and retroviral budding: Arole for the ESCRT machinery. Science 316: 1908–1912.

39. Dustin, M.L. & D.R. Colman. 2002. Neural andimmunological synaptic relations. Science 298: 785.

40. Piguet, V. & Q. Sattenaut. 2004. Dangerous liaisonsat the virological synapse. J. Clin. Invest. 114: 605.

41. Shav-Tal, Y. & R.H. Singer. 2005. RNA localisation.J. Cell Sci. 118: 4077–4081.

42. Mingle, L.A., N.N. Okuhama, J. Shi, et al. 2005.Localization of all seven messenger RNAs for theactin-polymerazitation nucleator arp2/3 complexin the protrusions of fibroblasts. J. Cell Sci. 118:2425–2433.

43. Yedavalli, V.S., C. Neuveut, Y.H. Chi, et al. 2004.Requirement of DDX3 dead box RNA helicase for


HIV – I Rev-RRE export function. Cell 119: 381–392.

44. Reddy, T.R., W. Xu & F. Wong-Staal. 2000. Generaleffect of Sam68 on rev/Rex regulated expression ofcomplex retroviruses. Oncogene 19: 4071–4074.

45. Li, J., H. Tang, T.M. Mullen, et al. 1999. A role forRNA helicase A in posttranscriptional regulation ofHIV type 1. Proc. Natl. Acad. Sci. USA 96: 709–714.

46. Tang, P.Z., C.H. Tsai-Morris & M.L. Dufau. 1999.A novel gonadotropin-regulated testicular RNA he-licase. A new member of the dead-box family. J. Biol.

Chem. 274: 37932–37940.47. Basyuk, E., T. Galli, M. Mougel, et al. 2003. Retro-

viral genomic RNAs are transported to the plasmamembrane by endosomal vesicles. Develop. Cell 5:161–174.

48. Butsch, M. & K. Boris-Lawrie. 2002. Destiny ofunspliced retroviral RNA: Ribosome and/or virion?J. Virol. 76: 3089–3094.

49. Kaye, J.F. & A.M.Lever. 1999. Human immunode-ficiency virus types I and 2 differ in the predominantmechanism used for selection of genomic RNA forencapsidation. J. Virol. 73: 3023–3031.

50. Poon, D.T., E.N. Chertova & D.E. Ott. 2002.Human immunodeficiency virus type 1 prefer-entially encapsidates genomic RNAs that encodePr55(Gag): Functional linkage between translationand RNA packaging. Virol. 293: 368–378.

51. Butsch, M. & K. Boris-Lawrie. 2000. Translationis not required to generate virion precursor RNAin human immunodeficiency virus type 1-infectedT-cells. J. Virol. 74: 11531–11537.

52. Cottingham, F.R. & M.A. Hoyt. 1997. Mitotic spin-dle positioning in Saccharomyces cerevisiae is accom-plished by antagonistically acting microtubule mo-tor proteins. J. Cell. Biol. 138: 1041–1053.

53. Terada, S. & N. Hirokawa. 2000. Moving on to thecargo problem of microtubule-dependent motors inneurons. Curr. Opin. Neurobiol. 10: 566–573.

54. Miller, R.K., K.K. Heller, L. Frisen, et al. 1998. Thekinesin-related proteins Kip2p and Kip3p, functiondifferently in nuclear migration in yeast. Mol. Biol.

Cell. 9: 2051–2068.55. Hoyt, M.A., L. He, K.K. Loo, et al. 1992. Two

Saccharomyces cerevisieae kinesin-related gene productsrequired for mitotic spindle assembly. J. Cell. Biol.

118: 109–120.56. Saunders, W.S., D. Koshland, D. Eshel, et al. 1995.

Saccharomyces cerevisieae kinesin- and dynein-relatedproteins required for anaphase chromosome segre-gation. J. Cell. Biol. 128: 617–624.

57. Hoyt, M.A., A.A. Hyman & M. Bahler. 1997. Mo-tor proteins of the eukaryotic cytoskeleton. Proc. Natl.

Acad. Sci. USA 94: 12747–12748.58. Haraguchi, K., T. Hayashi, T. Jimbo, et al. 2006.

Role of the kinesin 2 family protein, KIF3, duringmitosis. J. Biol. Chem. 281: 4094–4099.

59. Defeu-Soufo, J.H. & P.L.Graumann. 2004. Dy-namic movement of actin-like proteins within bac-terial cells. EMBO Rep. 5: 789–794.

60. Graumann, P.L. 2004. Cytoskeletal elements in bac-teria. Curr. Opin. Microbiol. 7: 565–571.

61. Graumann, P.L. & H.J. Defeu-Soufo. 2004. An in-tracellular actin motor in bacteria? Bioessays 26:1209–1216.

62. Preuss, M.L., D.R. Kovar, Y.R.J. Lee, et al. 2004. Aplant-specific kinesin binds to actin microfilamentsand interacts with cortical microtubules in cottonfibers. Plant Physiol. 136: 3945–3955.

63. Lu, L., Y.R.J. Lee, R. Pan, et al. 2005. An internalmotor kinesin is associated with the Golgi appara-tus and plays a role in trichome morphogenesis inArabidopsis. Mol. Biol. Cell. 16: 811–823.

64. Shapiro, J.A. 2002. Genome organization and re-organization in evolution. Ann. N. Y. Acad. Sci. 981:111–134.

65. Vaughn, M.W. & R. Martienssen. 2005. It’s a smallRNA world, after all. Science 309: 1525–1526.

66. Mattick, J.S. 2001. Non-coding RNAs: The archi-tects of eukaryotic complexity. EMBO Rep. 2: 986–991.

67. Mattick, J.S. 2006. The underworld of RNA. Nat.

Gen. 38: 393.68. Eickbush, T. 1999. Mobile introns: Retrohoming by

complete reverse splicing. Curr. Biol. 9: 11–14.69. Brosius, J. 1999. RNAs from all categories generate

retrosequences that may be exapted as novel genesor regulatory elements. Gene 238: 115–134.

70. Brosius, J. 2003. The contribution of RNAs andretroposition to evolutionary novelties. Genet. 118:99–115.

71. Daubin, V. & H. Ochman, 2004. Start-up entitiesin the origin of new genes. Curr. Opin.Gen. Dev. 14:616–619.

72. Slotkin, R.K. & R. Martienssen. 2007. Transpos-able elements and the epigenetic regulation of thegenome. Nat. Rev. Genet. 8: 272–285.

73. Mattick, J. 2007. A new paradigm for developmen-tal biology. J. Exp. Biol. 210: 1526–1547.

74. Kidwell, M.G. & D.R. Lisch. 2001. Perspective:Transposable elements, parasitic DNA and genomeevolution. Evol. 55: 1–24.

75. Nakamura, T.M. & T.R. Cech. 1998. Reversingtime: The origin of telomerase. Cell 92: 587–590.

76. Shapiro, J.A. 2006. Genome informatics: The roleof DNA in cellular computations. Biol. Theory 1:288–301.

77. Vetsigian, K., C. Woese & N. Goldenfeld. 2006.Collective evolution and the genetic code. Proc. Natl.

Acad. Sci. USA 1003: 10996–10701.


78. Rao, A.L.N., T.W. Dreher, L.E. Marsh, et al. 1989.Telomeric function of the tRNA-like structure ofbrome mosaic virus RNA. Proc. Natl. Acad. Sci. USA

86: 5335–5339.79. Maizels, A. & A.M. Weiner. 1993. The genomic

tag hypothesis: Modern viruses as molecular fossilsof ancient stretegies for genomic replication. In The

RNA World. R.F. Gesteland & J.F. Atkins, Eds.: 577–602. Cold Spring Harbor Laboratory Press. ColdSpring Harbor, NY.

80. Maizels, N., A.M. Weiner, D. Yue, et al. 1999. Newevidence for the genomic tag hypothesis: ArchaealCCA-adding enzymes and tRNA substrates. Biol.

Bull. 196: 331–334.81. Flavell, A.J. 1995. Retroelements, reverse transcrip-

tase and evolution. Comp. Biochem. Physiol. B. Biochem.

Mol. Biol. 110: 3–15.82. Eickbush, T.H. & D.G. Eickbush. 2007. Finely or-

chestrated movements: Evolution of the ribosomalRNA genes. Genet. 175: 477–485.

83. Filipowicz, W. 2000. Imprinted expression of smallnucleolar RNAs in brain: Time for RNomics. Proc.

Natl. Acad. Sci. USA 97: 14035–140357.84. Frost, L.S., R. Leplae, A.O. Summers, et al. 2005.

Mobile genetic elements: The agents of open sourceevolution. Nat. Rev. Microbiol. 3: 722–732.

85. Amaral, P.P., M.E. Dinger, T.R. Mercer, et al. 2008.The eukaryotic genome as an RNA machine. Science

319: 1787–1789.86. Yazgan, O. & J.E. Krebs. 2007. Noncoding but

nonexpendable: Transcriptional regulation by largenoncoding RNA in eukaryotes. Biochem. Cell. Biol.

85: 484–496.87. Chu, C.Y. & T.M. Rana. 2007. Small RNAs: Regu-

lators and guardians of the genome. J. Cell. Physiol.

213: 412–419.88. Farazi, T.A., S.A. Juranek & T. Tuschl. 2008. The

growing catalogue of small RNAs. Exiqon. The2008 Collection Booklet 35–69.

89. Witzany, G. 2006. Natural genome editing com-petences of viruses. Acta Biotheor. 54: 235–253.

90. St.Laurent III, G. & C. Wahlestedt. 2007. Noncod-ing RNAs: Couplers of analog and digital informa-tion in nervous system function. Trends Neurosci. 30:612–631.

91. Presutti, C., J. Rosati, S. Vincenti, et al. 2006. Noncoding RNA and brain. BMC Neurosci. 7: S5.

92. Mehler, M.F. & J.S. Mattick. 2007. NoncodingRNAs and RNA editing in brain development, func-tional diversification, and neurological disease. Phys-

iol. Rev. 87: 799–823.93. Mattick, J. & M.F. Mehler. 2008. RNA editing, DNA

recoding and the evolution of human cognition.Trends Neurosci. 31: 227–233.

94. Ambros, V. & X. Chen. 2007. The regulation ofgenes and genomes by small RNAs. Develop. 134:1635–1641.

95. Watanabe, T., Y. Totoki, A. Toyoda, et al. 2008. En-dogenous siRNAs from naturally formed dsRNAsregulate transcripts in mouse oocytes. Nature 453:539–543.

96. Chen, K. & N. Rajewsky. 2007. The evolution ofgene regulation by transcription factors and mi-croRNAs. Nat. Rev. Gen. 8: 93–103.

97. Doench, J.G., C.P. Petersen & P.A. Sharp. 2003.siRNAs can function as miRNAs. Gen. Develop. 17:438–442.

98. Fire, A. 2005. Nucleic acid structure and intracellu-lar immunity: Some recent ideas from the world ofRNAi. Quart. Rev. Biophys. 38: 303–309.

99. Sontheimer, E.J. & R.W. Carthew. 2005. Silencefrom within: Endogenous siRNAs and miRNAs. Cell

122: 9–12.100. Kim, V.N. 2005. MicroRNA biogenesis: Coordi-

nated cropping and dicing. Nat. Rev. Mol. Cell Biol.

6: 376–385.101. Piriyapongsa, J. & I.K. Jordan. 2008. Dual coding

of siRNAs and miRNAs by plant transposable ele-ments. RNA 14: 1–8.

102. Mattick, J. & Gagen. 2001. The evolution of con-trolled multitasked gene networks: The role of in-trons and other noncoding RNAs in the develop-ment of complex organisms. Mol. Biol. Evol. 18:1611–1630.

103. Calin, G.A. & C.M. Croce. 2006. MicroRNA sig-natures in human cancers. Nat. Rev. Can. 6: 857–866.

104. Quellet, D.L., M.P. Perron, L.A. Gobeil, et al. 2006.Micro RNAs in gene regulation: When the smallestgovern it all. J. Biomed. Biotech. 69616: 1–20.

105. Miller, S. & J. Krijnse-Locker. 2008. Modificationof intracellular membrane structures for virus repli-cation. Nat. Rev. Microbiol. 6: 363–374.

106. Matera, A.G., R.M. Terns & M.P. Terns. 2007.Non-coding RNAs: Lessons from the small nuclearand small nucleolar RNAs. Nat. Rev. Mol. Cell. Biol.

8: 209–220.107. Hiscox, J.A. 2002. The nucleolus – a gateway to

viral infection. Arch. Virol. 147: 1077–1089.108. Rustom, A., R. Saffrich, I. Markovic, et al. 2004.

Nanotubular highways for intercellular organelletransport. Science 303: 1007–1010.

109. Zemann, A., A. Beckke, M. Kiefmann, et al. 2006.Evolution of small nucleolar RNAs in nematodes.Nuc. Ac. Res. 34: 2676–2685.

110. Richard, P., A.M. Kiss, X. Darzacq, et al. 2006. Co-transcriptional recognition of human intronic boxH/ACA snoRNAs occurs in a splicing-independentmanner. Mol. Cell. Biol. 26: 2540–2549.


111. Weber, M.J. 2006. Mammalian small nucleolarRNAs are mobile genetic elements. PLoS Genet. 2:e205.

112. Platini, M., I.Goldberg, A.I. Lamond, et al. 2002.Cajal Body dynamics and association with chro-matin are ATP dependent. Nat. Cell. Biol. 4: 502–508.

113. Tomlinson, R.L., T.A. Ziegler, T. Supakorndej, et

al. 2006. Cell cycle-regulated trafficking of humantelomerase to telomeres. Mol. Biol. Cell. 17: 955–965.

114. Kiss, T. 2002. Small nucleolar RNAs: An abun-dant group of noncoding RNAs with diverse cellularfunctions. Cell 109: 145–148.

115. Kiss, A.M., B.E. Jady, X. Darzacq, et al. 2002. ACajal body-specific pseudouridylation guide RNAis composed of two box H/ACA snoRNA-like do-mains. Nuc. Ac. Res. 30: 4643–4649.

116. Clouet d’Orval, B., M.L. Bortolin, C. Gaspin, et al.2001. Box C/D RNA guides for the ribose methyla-tion of archaeal tRNAs. The tRNATrp intron guidesthe formation of two ribose-methylated nucleosidesin the mature tRNATrp. Nuc. Ac. Res. 29: 4518–4529.

117. Wang, C. & U.T. Meier. 2004. Architecture andassembly of mammalian H/ACA small nucleolarand telomerase ribonucleoproteins. EMBO J. 23:1857–1867.

118. Witzany, G. 2008. The viral origins of telomeres,telomerases and their important role in eukaryoge-nesis and genome maintenance. Biosem. 2: 191–206.

119. Jady, B.E., E. Bertrand & T. Kiss. 2004. Humantelomerase RNA and box H/ACA scaRNAs sharea common Cajal body-specific localization signal. J.

Cell. Biol. 164: 647–652.120. Fu, D. & K. Collins. 2006. Human telomerase and

Cajal body ribonucleoproteins share a unique speci-ficity of Sm protein association. Gen. Dev. 20: 531–536.

121. Darzacq, X., B.E. Jady, C. Verheggen, et al. 2002.Cajal body-specific small nuclear RNAs: A novelclass of 2′-O-methylation and pseudouridylationguide RNAs. EMBO J. 21: 2746–2756.

122. Richard, P., X. Darzacq, E. Bertrand, et al. 2003.A common sequence motif determines the Cajalbody-specific localization of box H/ACA scaRNAs.EMBO J. 22: 4283–4293.

123. Eickbush, T.H. & D.G. Eickbush. 2007. Finely or-chestrated movements: Evolution of the ribosomalRNA genes. Genet. 175: 477–485.

124. Narayanan, A., A. Lukowiak, B.E. Jady, et al. 1999.Nucleolar localization signals of box H/ACA smallnucleolar RNAs. EMBO J. 18: 5120–5130.

125. Brosius, J. 2005. Echoes from the past – are we stillin an RNP world? Cyt. Gen. Res. 110: 1–4.

126. Teng, S.C., B. Kim & A. Gabriel. 1996. Retro-transposon reverse transcriptase-mediated repair ofchromosomal breaks. Nature (London) 383: 641–644.

127. Coffin, J.M., A.H. Hughes & H.E. Varmus. 1997.Retroviruses. Cold Spring Harbor Laboratory Press.New York.

128. Volff, J.N. 2006. Turning junk into gold: Domesti-cation of transposable elements and the creation ofnew genes in eukaryotes. BioEssays 28: 913–922.

129. Sternberg, R. 2002. On the roles of repetitiveDNA elements in the context of a unified genomic-epigenetic system. Ann. N. Y. Acad. Sci. 981: 154–188.

130. Shapiro, J.A. & R. Sternberg. 2005. Why repetitiveDNA is essential to genome function. Biol. Rev. 80:1–24.

131. Shabalina, S.A. & N.A. Spiridonov. 2004. Themammalian transcriptome and the function of non-coding DNA sequences. Gen. Biol. 5: 105e.

132. Sternberg, R. & J.A. Shapiro. 2005. How repeatedretroelements format genome function. Cyt. Gen. Res.

110: 108–116.133. Nosek, J., P. Kosa & L. Tomaska. 2006. On the

origin of telomeres: A glimpse at the pre-telomeraseworld. Bioessays 28: 182–190.

134. Villasante, A., J.P. Abad, R. Planello, et al. 2007.Drosophila telomeric retrotransposons derived froman ancestral element that was recruited to replacetelomerase. Gen. Res. 17: 1909–1918.

135. Haoudi, A. & J.M. Mason. 2000. Reversetrannscriptase can stabilize or destabilize thegenome. Genome 43: 949–956.

136. Batzer, M.A. & D.L. Deininger. 2002. ALU repeatsand human genomic diversity. Nat. Rev. Gen. 3: 370–380.

137. Temin, H.M. 1985. Reverse transcription in theeukaryotic genome: Retroviruses, pararetroviruses,retrotransposons and retrotranscripts. Mol. Biol.

Evol. 2: 455–468.138. Sugiyama, T., H. Cam, A. Verdel, et al. 2005. RNA-

dependent RNA polymerase is an essential compo-nent of a self-enforcing loop coupling heterochro-matin assembly to siRNA production. Proc. Natl.

Acad. Sci. 102: 152–157.139. Grewal, S.I. & S.C. Elgin. 2007. Transcription and

RNA interference in the formation of heterochro-matin. Nature 447: 399–406.

140. Xiong, Y. & T.H. Eickbush. 1990. Origin and evolu-tion of retroelements based upon their reverse tran-scriptase sequences. EMBO J. 9: 3353–3362.

141. Mesnard, J.M. & G. Lebeurier. 1991. How doviral reverse transcriptases recognize their RNAgenome? FEBS Lett. 287: 1–4.

142. Villarreal, L.P. 2009. Origin of Group Identity: Viruses,

Addiction and Cooperation. Springer Verlag, New York.


143. Witzany, G. 2008. Bio-communication of bacteriaand their evolutionary roots in natural genome edit-ing competences of viruses. Open Evol. J. 2: 44–54.

144. Kenzelmann, M., K. Rippe & J.S. Mattick. 2006.RNA: Networks & imaging. Mol. Sys. Biol. 2: 44.

145. Jeffares, D.C., A.M. Poole & D. Penny. 1998. Relicsfrom the RNA world. J. Mol. Evol. 46: 18–36.

146. Poole, A.M., D.C. Jeffares & D. Penny. 1998. Thepath from the RNA world. J. Mol. Evol. 46: 1–17.

147. Poole, A.M. & D.T. Logan. 2005. ModernmRNA proofreading and repair: Proof that thelast universal common ancestor (LUCA) hadan RNA genome? Mol. Biol. Evol. 22: 1444–1455.

148. Levine, M.T., C.D. Jones, A.D. Kern, et al. 2006.Novel genes derived from noncoding DNA inDrosophila melanogaster are frequently X-linked andexhibit testis-biased expression. Proc. Natl. Acad. Sci.

USA 103: 9935–9939.

Noncoding RNAs: Persistent Viral Agents as Modular Tools ... · life style; nucleic acid language...

Documents

Transcript of Noncoding RNAs: Persistent Viral Agents as Modular Tools ... · life style; nucleic acid language...