Evolutionary deviations from the universal genetic code in ciliates

ISSN 1022-7954, Russian Journal of Genetics, 2009, Vol. 45, No. 4, pp. 379–388. © Pleiades Publishing, Inc., 2009.Original Russian Text © N.P. Lukashenko, 2009, published in Genetika, 2009, Vol. 45, No. 4, pp. 437–448.

379

INTRODUCTION

An unusual feature of ciliates is their genetic code,which is among the few codes deviating from the uni-versal one. Such deviations are known for the nucleargenomes of yeast

Candida

cylindracea

[1], green algae

Acetabularia

[2], and some species of animals andprokaryotes [3].

Very recently the universal character of genetic codewas considered as a very important argument for mono-phyletic origin of all existing living organisms. It wasthought that genetic code, once arisen, has been stabi-lized by strict elimination of deviations from the stan-dard, which results in its universality. However, thisview was considered doubtful after the discovery ofdeviations in codon assignment in various genetic sys-tems [3, 4].

Furthermore, of key importance were discoveries oftwo novel, genetically encoded, naturally occurringamino acids selenocystein (21st amino acid) [5–14] andpyrrolysine (22nd amino acid) [15–22]. These discov-eries have changed our understanding of genetic code,because the latter expanded from 20 standard aminoacids to 22 amino acids (table).

However, genetic code remains an enigma, in spiteof the fact that the complete codon catalog was com-piled several decades ago. Although we know whichtriplets correspond to each amino acid and even under-stand how triplet assignments change among taxa, wedo not know the reason for the specific correspondencebetween triplets and codon assignments.

It has been hypothesized that genetic code evolvedin two stages [23]. At the first stage, the canonical codeappeared. This event preceded the appearance of thelast universal common ancestor of the existing organ-isms. Later, this code diverged in numerous nuclear and

mitochondrial lineages. The evidence reported in recentworks [23–25] suggests explanation of secondary devi-ations from the canonical genetic code. This means thatmost codes deviating from the original one, appear as aresult of changes in tRNA generated by posttransla-tional modifications. These are caused by modificationsof bases or editing the precursor RNAs, rather than bychanges in tRNA anticodons [24, 25].

Traditionally, changes in the two groups are consid-ered separately: alterations are recorded more fre-quently in the mitochondrial code than in the nucleargenome. However, all codons that were “lost” or reas-signed in the nuclear lineages were also lost or reas-signed in the mitochondrial lineages.

DISCOVERY OF DEVIATIONSFROM THE UNIVERSAL GENETIC CODE

IN CILIATES

Until recently, information on deviations from thecodon assignment in ciliates were restricted to the stud-ies of classes Oligohymenophorea (genera

Parame-cium

and

Tetrahymena

) and Hypotrichea (genera

Euplotes

,

Stylonychia

, and

Oxytricha

). In these ciliates,deviations from the assignment of standard codons arestudied in most detail.

Deviations in the Mitochondrial Genome

A deviation from the universal genetic code in cili-ates was first reported for the mitochondrial genome of

Paramecium

tetraurelia

[26]. In the mtDNA of thisorganism, the standard termination codon UGAencodes tryptophane (Trp). Similarly, in the mito-chondrial genome of

Tetrahymena

thermophila

UGA

Evolutionary Deviations from the Universal Genetic Code in Ciliates

N. P. Lukashenko

Vavilov Institute of General Genetics, Russian Academy of Sciences, Moscow, 119991 Russia

Received February 4, 2008

Abstract

—The review surveys the information, including the most recent data, on the evolution of genetic codein ciliates, which is among the few codes deviating from the universal one. We discuss the cases of recurrent,independently arising deviations from the assignments of standard codons of polypeptide chain termination inthe mitochondrial and nuclear genomes of ciliates and some other protozoans. Possible molecular mechanismsare considered, which underlie deviations from standard termination code to coding glutamine (codon UAA andUAG) and cystein or tryptophane (codon UGA) in the nuclear genome. Critical analysis of the main hypotheseson the evolution of secondary deviations from the universal code in ciliates is presented.

DOI:

10.1134/S1022795409040012

REVIEWS AND THEORETICAL ARTICLES

380

RUSSIAN JOURNAL OF GENETICS

Vol. 45

No. 4

2009

LUKASHENKO

in open reading frames is also translated as tryp-tophane [26].

Of interest is the fact that in mtDNA of mammals,fungi, and ciliates of the genus

Paramecium

, UGA alsocodes for tryptophane, whereas in the nuclear genomeof these organisms, UGA serves as a standard termina-tion codon [26]. It was shown that sequences of mito-chondrial tRNA

Trp

in

P

.

aurelia

,

Neurospora

fungi, and

E

scherichia

coli

exhibit high similarity (63–70%),while human tRNA

Trp

sequences are more differentfrom them (47% homology).

These results are not unexpected, because tRNA

Trp

,13S rRNA, and incomplete sequences of 20S rRNAgenes of

Paramecium

are strikingly similar to the cor-responding sequences in

E

.

coli

. Consequently, thoughtRNA

Trp

of

Paramecium

is probably a typical mito-chondrial tRNA (an altered anticodon, the absence ofthe 3'-terminal CCA in the gene, etc.), it is at the sametime clearly similar to the “unchanged” tRNA sequenceof

E

.

coli

[26]. The importance of these findings is hardto evaluate, but it may well be that

Paramecium

hasdiverged in evolution from any hypothetical commonancestor to a lesser extent than fungi and mammals andthus the former show closer similarity to more primitiveorganisms.

Deviations leading to reassignment of UGA as a ter-mination codon to coding tryptophane among organ-isms as different as

Paramecium

, mammals, and fungi,may have different explanations. For instance, mito-chondria may have evolved from one protomitochon-drion, in which such alteration, having occurred, wasnot eliminated by subsequent natural selection. How-

ever, it cannot be excluded that the UGA deviation fromthe standard stop code is caused by some mechanismsthat support the corresponding functions of mitochon-dria or provide interaction between the nucleus andcytoplasm. Yet, if this assumption is true, it seemsstrange that this modification is absent in plant mito-chondria [27] and in chloroplasts (the genetic code ofthese cell structures is apparently universal).

Analysis of mitochondrial genes in two members ofthe family Trypanosomae,

Trypanosoma

brucei

[28, 29]and

Leishmania

tarentolae

[30] has also shown that inthese organisms UGA encodes tryptophane rather thanstop. There are many examples of reassignment of thiscodon; here, we will only note that the deviation ofUGA from the standard assignment for termination tocoding tryptophane was observed in mitochondria ofhuman [31–35] and invertebrates [36–43]. Moreover, inmitochondria of budding yeast

Saccharomyces

cerevi-siae

, UGA does not terminate the polypeptide chain,but is used as a sense codon, probably coding for tryp-tophane [44].

Thus, the reassignment of UGA from standard ter-mination codon to tryptophane in mitochondria iswidely spread in members of both lower and highereukaryotic taxa. Higher plants, whose genetic codeapparently is universal, are an exception to this rule.

Deviations in the Nuclear Genome

Deviations in the assignment of the standard termi-nation codons UAA and UAG in the nuclear genomewere first reported for two related ciliate species,

Parame-

Universal genetic code, 2007

First position (5' end)Second codon position

Third position (3' end)U C A G

U UUU-Phe UUC-Phe UUA-Leu UUG-Leu

UCU-Ser UCC-Ser UCA-Ser UCG-Ser

UAU-TyrUAC-Tyr UAA-Ter UAG-Pyl (Ter)

UGU-Cys UGC-Cys UGA-Sec (Ter)UGG-Trp

U C A G

C CUU-Leu CUC-Leu CUA-Leu CUG-Leu

CCU-Pro CCC-Pro CCA-Pro CCG-Pro

CAU-His CAC-His CAA-Gln CAG-Gln

CGU-Arg CGC-Arg CGA-Arg CGG-Arg

U C A G

A AUU-Ile AUC-Ile AUA-Ile

*AUG-Met

ACU-Thr ACC-Thr ACA-Thr ACG-Thr

AAU-Asn AAC-Asn AAA-Lys AAG-Lys

AGU-Ser AGC-Ser AGA-Arg AGG-Arg

U C A G

G GUU-Val GUC-Val GUA-Val GUG-Val

GCU-Ala GCC-Ala GCA-Ala GCG-Ala

GAU-Asp GAC-Asp GAA-Glu GAG-Glu

GGU-Gly GGC-Gly GGA-Gly GGG-Gly

U C A G

Notes: * Initiation codon; Ter, codon of termination of the polypeptide chain; Sec, selenocysteine (21st amino acid in the universal geneticcode [9, 10]); Pyl, pyrrholysine (22nd amino acid in the code [15, 17, 19]); UGA encodes Sec in some protozoans, lower plants,mammals, and fungi [13]; UAG is the Pyl codon in the code of archaebacterium

Methanosarcina barkeri

[18].

RUSSIAN JOURNAL OF GENETICS

Vol. 45

No. 4

2009

EVOLUTIONARY DEVIATIONS FROM THE UNIVERSAL GENETIC CODE IN CILIATES 381

cium

primaurelia

and

Paramecium

tetraurelia

[45–47].Both codons (UAA and UAG) in these species aretranslated as glutamine codons. Horowitz and Gor-ovsky [48] were the first to show that in

Tetrahymenathermophila

UAA is not a termination codon, butencodes glutamine. Later, Hanyu et al. [49] experimen-tally demonstrated that UAA and UAG, which are stan-dard termination codons, code for glutamine in

T

.

ther-mophila

thus making UGA a single termination codonin this ciliate.

The use of UAA and UAG as glutamine codons wasfound also in two closely related ciliates of the classHypotrichea. In one of them,

Stylonychia

lemnae

onlyUAA is used as a glutamine codon. In the other species,

Oxytricha

nova

both UAA and UAG are translated asglutamine codons [50], although standard glutaminecodons CAA and CAG are also preserved [3, 27].

However, nonstandard glutamine codons UAA andUAG were found not in all hypotrichous ciliates. Evenwithin one genus,

Euplotes

codons that act as stopcodons in the case of the universal code are used differ-ently. For example, in

Euplotes vannus

,

E. octocarina-tus

, and

E. crassus

, glutamine is coded only by the uni-versal codons CAA and CAG, while UAA and UAG areused as stop codons. It was found that UAA is used asa stop codon in translated sequences of actin,

β

-tubulin,and histone H4 genes in

Euplotes

crassus

[51] and thepheromone

Er-1

gene in

E. raikovi

[52]. A similar useof UAA as a stop codon was found in heterotrich ciliate

Blepharisma

[53].

Unexpectedly, UGA was found to be translated ascystein in the nuclear genome of

Euplotes

octocarina-tus

[54

−

56] and

E

.

crassus

[57].Baroin Tourancheau et al. [58] extended analysis of

genetic code by studying it in six classes of ciliates:seven species from four classes that had not been exam-ined earlier (Karyorelictea, Heterotrichea, Nassopho-rea, and Litostomatea) and nine examined earlier spe-cies from the classes Oligohymenophorea and Hypot-richea (figure). They sequenced

α

-tubulin genes foreight species and phosphoglycerate kinase (PGK)genes for five species.

It was found that glutamine codons were used in the

α

-tubulin and PGK genes in a similar fashion. In theciliate species whose

α

-tubulin and PGK genes weresequences, nonstandard glutamine codons UAA andUAG were used at the same frequency. The only excep-tion is

Tetrahymena

pyriformis,

in which the terminationcodons UAA and UAG of the universal genetic codewere not found in the

α

-tubulin gene as coding ones,while in the genes for PGA [58],

β

-tubulin [59], andactin [60], these codons are reassigned.

Comparison of various classes of ciliates showedthat the use of nonstandard glutamine codons (mainlyUAA) is a common phenomenon. For instance, in

Oxytricha

granulifera

and

Stylonychia

lemnae

(Hypot-richea), UAA in the

α

-tubulin gene codes forglutamine, in

Oxytricha

nova

both UAA and UAG inthe PGK gene are translated as glutamine codons,whereas in four

Euplotes

(

E

.

aediculatus

,

E

.

vannus

,

Spathidium

sp.

Epidinium

sp.

Entodinium

sp.

Didinium nasutumChaenea vorax

Tetrahymena thermophila

(UGA)

Tetrahymena pyriformis

(UGA)

Paramecium primaurelia

(UGA)

Paramecium caudatumZosterograptus

sp.

Colpoda

sp.

Stylonychia lemnae (UGA)Oxytricha sp.

Halteria grandinellaUrostyla sp.

Euplotes aediculatusBlepharisma japonicum (UAA)

Stentor coeruleusCondylostoma magnum

Loxodes striatusSaccharomyces cerevisiae

Litostomatea

Oligohymenophorea

NassophoreaColpodea

Hypotrichea

Heterotrichea

Karyorelictea

97

99

10062

71

59

8887

58

9289

77

100

100

Dendrogram reflecting features of the genetic code in ciliates. The phylogenetic tree of 28S rRNA in ciliates is constructed usingthe neighbor-joining method. A part of the 5'-end of 28S rRNA was used for analysis. Species from eight classes of ciliates are pre-sented. Saccharomyces cerevisiae represents an outgroup. The names of the species in which glutamine residues are encoded byboth universal codons and additional codons UAA and (or) UAG are underlined. For some species, termination codons UGA andUAA are given at right (after [58] with simplifications).

382

RUSSIAN JOURNAL OF GENETICS Vol. 45 No. 4 2009

LUKASHENKO

E. octocarinatus, E. crassus)), no reassignments of thestandard termination codons UAA and UAG wasobserved in the α-tubulin and PGK genes.

Nonstandard glutamine codons are more often usedin Oligohymenophorea than in Hypotrichea. In thePGK gene sequences, reassignment of the terminationcodons was observed in all members of the former class(more often in Paramecium; UAA reassignment ismost common). This was also shown for genes for cellsurface proteins (antigens)—calmodulin and tubulin[61–64].

Nonstandard UAA use (in the α-tubulin gene) as theglutamine codon was found in members of three otherclasses: Zosterograptus sp. (Nassophorea), Condylos-toma magnum (Heterotrichea), and Loxodes striatus(Karyorelictea). In C. magnum, both UAA and UAG inthe PGK gene are translated as glutamine codons. How-ever, in Stentor coeruleus and Blepharisma japonicum(Heterotrichea), as well as in three Lytostomea species,UAA and UAG seem to be used exclusively as stopcodons [58]. The listed facts [58] were confirmed assupplemented by Lozupone et al. [65], who showedthat UAA and UAG were used as glutamine codons inKentrophoros sp. (Karyorelictea) and Naxella sp. (Nas-sophorea), which had not been examined earlier, aswell as in two species of Oligohymenophorea and anumber of Spyrotrichea species. Lozupone et al. [65]were the first to show the use of UGA as the tryp-tophane codon in the eukaryotic nuclear genome of twodistantly related lineages of Blepharisma americanumand Colpoda.

The data presented above show that the genetic codehas deviations from the universal code in six classes ofciliates, except probably three Litostomatea species(Spathidium sp., Epidinium sp., and Entodinium sp.), inwhich no deviations in the use of terminations codonswere recorded for the α-tubulin and PGK genes [58]. Itseems that the frequency of using codons in nonstand-ard fashion differs even among the species within aclass. The class Litostomatea is an only ciliate groupthat apparently uses the universal genetic code. Themembers of the class Nassophorea employ both theuniversal genetic code and a code deviating from it,which may have a sound explanation in view of the factthat there is no reliable molecular genetic data support-ing a monophyletic origin of this class [65].

A dendrogram of genetic code (figure), which wasconstructed [58] on the basis of the data of combinedcode analysis and the phylogenetic tree of ciliates basedon incomplete 28S rRNA coding sequences, shows theabsence of unambiguous divergence between the spe-cies group with deviating code and the group using theuniversal code; rather, the two code types are mixedwithin monophyletic groups. This is particularly clearlyseen in Heterotrichea. The member of this group Ble-pharisma japonicum employs UAA as a standard ter-mination codon [53], while in the closely related spe-cies Stentor coeruleus, the assignment of codon UAA

and UAG is not determined. Yet, in a more distant spe-cies Condylostome magnum, both UAA and UAG areused as glutamine codons.

Thus, within the monophyletic group Heterotrichea,two types of using codons UAA and UAG are observed.A similar situation was described for hypotrichous cil-iates [54, 55].

The above data on nonstandard use of terminationcodons of the standard genetic code seem to indicate apolyphyletic origin of the assignments of these codons(mainly UAA and UAG) in the nuclear genome of cili-ates. Alteration of the genetic code in ciliates was prob-ably a recurrent event rather than a single and ancientone, as thought earlier [66].

Secondary Deviations in Genetic Code of Ciliates

A number of authors [41, 42] have shown that thedeviating genetic code appeared in ciliates in a second-ary manner rather than being a remainder of the primor-dial genetic code. Based on the results of their analysisof genetic code and phylogeny of complete sequencesof 18S rRNA in three groups of ciliates, Greenwood et al.[67] have considered two possible alternative scenariosof the origin of the altered genetic code in these proto-zoans.

According to the first scenario, the latest commonancestor of infisorians had an altered genetic code andciliates from the genus Euplotes (Hypotrichea)reversed to the state, in which their genetic codeapproached the universal code. This hypothesis impliesthat Hypotrichea and possibly other groups have alteredgenetic code (which was confirmed later). The secondscenario, proposed earlier by Harper and Jahn [51],assumed that the latest common ciliate ancestor had theuniversal genetic code, whereas the altered codeappeared later, in one or several lineages.

The key proof of the fact that many groups of cili-ates have altered genetic code was obtained in studiesof ribosomal RNA phylogeny [68, 69]. The results ofthese studies have shown for the first time that ciliatesappeared on the common eukaryotic tree as a late-split-ting group, while the groups that diverged earlier pos-sessed the universal genetic code.

Let us consider the main component of the alteredgenetic code using the genus Tetrahymena as an exam-ple. These ciliates, in addition to the ordinary glutamine

tRNA (tRN reading the universal codons

CAA and CAG, were shown to have two unusual

glutamine-specific tRNAs (tRN and

tRN Based on the sequence similarity of these

three tRNAs, it was concluded that two novel glutaminetRNAs, which recognize both UAA and UAG, origi-nated from the ordinary glutamine tRNA via duplica-tion of the corresponding gene followed by divergenceof the copies [49]. This finding has shown that devia-

AUmUGGln),

AUUAGln

ACUAGln).



tions from the universal genetic code in ciliates mightbe based on alterations in tRNA.

Since altered genetic code has probably a secondaryorigin, a question arises as to how often have changedthe use of codons UAA, UAG, and UGA in the evolu-tionary history of ciliates? A number of authors pre-sumed that deviations in the genetic code recurrentlyappeared in the evolution of both prokaryotes andeukaryotes [3, 41–43, 70]. Some evidence suggests thatstop codons in ciliates were reassigned at least thrice[71, 72].

Interestingly, ciliates that show deviations from theuniversal genetic code are bacterial carnivores (phag-otrophs). All of them, except very distant forms have aspecialized organelle (called oral apparatus) that servesfor capturing and active digesting bacteria. Based onthis, Ninio [73] suggested that deviating code is prefer-able for ciliates, because it represents a mechanism pre-venting the expression of foreign DNA. Likewise, Pres-cott [74] noted that all studied ciliates with nonstandardgenetic code are free from viruses, in spite of the factthat bacterial symbionts are widespread in these organ-isms [75].

However, Ninio’s suggestions seem to be somewhatunfounded, because a number of ciliates were shown tohave mobile elements, at least some of which areancient. For instance, transposon-like elements Tel-1were found in the micronucleus of T. thermophyla [76, 77]and the macronucleus of Oxytricha fallax [78]. Appar-ently, in transposon-like elements, as in other genes ofciliates, codons were reassigned during the evolution.This assumption was indirectly supported by the resultsof studying the structure of elements Tec1 and Tec2 inthe ciliate E. crassus [57]. Jahn et al. [57] have identi-fied the regions of these elements that presumablyencode transposase, the enzyme catalyzing transposi-tion of some transposon-like elements. Thus, membersof the transposase gene family were found not only inprokaryotes and multicellular eukaryotes, but also inciliates [79].

The widespread occurrence of transposons amongciliates suggests horizontal gene transfer and possibleinvolvement of these elements in the evolution ofgenetic code in this protozoan group. However, suc-cessful gene transfer between species in nature seemsto be a rare event. As noted by Wilson [80], if this pro-cess occurred often, the genome of any species wouldbe mosaic and included genes from other species.Dubinin [81] doubted the possibility of natural directgene exchange between species.

However, the interest to horizontal (lateral) genetransfer (HGT) has been considerably growing in thelast 15 years, owing to sequencing of whole genomes inmembers of various taxa [82–84]. Horizontal transferof genes coding for catalase–peroxidase enzymes wasfound between archaebacteria and pathogenic bacteria[85]. Recently, widespread HST from intracellular bac-teria to multicellular eukaryotic hosts has been shown

[86]. Based on the recent studies of the early divergedeukaryotic diplomonade Giardia lamblia (a humanpathogenic intestine parasite), Morrison et al. [87] sug-gested that the G. lamblia genome was formed withparticipation of lateral gene transfer from bacterial andarcheae donors.

Some authors [83, 88] believe that HGT is con-stantly affecting the evolution of eukaryotes. However,this view has been doubted by Kurland et al. [82].

It is possible that HGT was a potent evolutionaryforce in the progenotic period [89–91]. It cannot beexcluded that occasional gene transfers may result inHGT fixation in the global populations of modern cells[92]. For instance, α-protobacterial sequences, codingfor mitochondrial proteins, are more often found in thenuclear genomes than in the mitochondrial ones [92].There are other well-documented examples of horizon-tal gene transfer, for instance, HGT among aminoacyl-tRNA synthetases [92]. Notwithstanding, testing of anample number of genomic events revealed fewinstances of horizontal gene transfer [83, 84]. Appar-ently, these were fixed in the global populations underunusual circumstances [93].

Only two incomplete transfers of rRNA are knownamong thousands of rRNA sequences of studied organ-isms [82]. This means that rRNA may serve as a verystable phylogenetic marker. Thus, there is no docu-mented evidence suggesting that transgenic exchangesof rRNA domains are common in nature, as proposedGogarten et al. [83].

The above data on the presence of transposons in thegenomes of members of three biological kingdoms [57, 79]indicate that these elements could be incorporated intothe ciliate evolutionary lineage, employing the univer-sal genetic code, and in some way be involved in theevolution of the genetic code in these unicellulareukaryotes.

MAIN HYPOTHESES EXPLAINING THE MECHANISMS OF CODON

REASSIGNMENT AND THE EVOLUTION OF GENETIC CODE IN CILIATES

Three major hypotheses were advanced to explainthe many facts of deviation of the genetic code from theuniversal one. Osawa and Jukes [41–43, 70] haveadvanced the codon capture hypothesis, according towhich mutations can affect the AT–GC composition ofthe genome and lead (due to mismatches C · A and G · Ucaused by mutations in the tRNA D-loop) to completeelimination of the codon from the whole genome withsubsequent loss of tRNA responsible for its translation.In view of Osawa and Jukes [41, 42], these codonscould later be restored (i.e., reappear via neutral pro-cesses, without selection) together with duplicated andmutated tRNA, resulting in reassignment of the corre-sponding codon.

384


LUKASHENKO

Jukes et al. [94], the authors of stop-codon capturehypothesis, have suggested that the factors of polypep-tide chain release (RFs) could become specific only toUGA, so that UAA and UAG could not be recognizedand disappeared from the genome. At the same time,some of the standard glutamine codons CAA and CAGmutated into UAA and UAG, while the glutamine

tRNA gene (tRN duplicated and mutated.

This generated two novel tRNAs (tRN and

tRN with anticodons UUA and CUA, which

could pair with codons UAA and UAG, respectively. Inthe Euplotes lineage, another scenario took place: therelease factors were specific to UAA and UAG, thusliberating the stop codon UGA and making possible itsreassignment to the cystein codon. Thus, UAA andUAG once again became standard termination codons.

In opposition to the hypothesis of neutral codonelimination [70, 95, 96], Andersson and Kurland [97,98] stated that the evolution of the translation systemcould occur gradually, by mutation of the correspond-ing tRNAs. This suggestion was supported by Schultzand Yarus [99], the authors of the tRNA-mediatedcodon reassignment hypothesis. This hypothesis postu-lated that the key link in the code evolution is a muta-tion in tRNA that permits tRNA to recognize asglutamine codons both standard codons CAA and CAGand codons UAA and UAG.

According to this model, the codon reassignmentoccurs during the “ambiguous translation periods,”when the codons acquire two different assignments.This codon exchange mechanism [100], in contrast tothe Osawa–Jukes codon capture [42], does not requirea complete elimination of the codon from the genomebefore its reassignment.

In the case of glutamine codons, UAA and UAG canact both as stop codons and codon encoding glutamine.All of the three isoacceptor tRNAsGln T. thermophylarecognize nonstandard glutamine codons (UAA andUAG). These tRNAsGln, as suggest the authors of thetRNA-mediated codon reassignment model [101, 102],may have diverged from the mutated ancestor thatcould translate UAA and UAG as glutamine codons inaddition to the standard glutamine codons CAA andCAG. If these events indeed occurred in tRNAGln

genes of the ancestral ciliate that used the universalgenetic code, they could create a situation that impliedmultiple, independent changes in the genetic code ofciliates [102].

In support of their earlier model, Schultz and Yarus[103] advanced the hypothesis of “ambiguous transla-tional intermediate,” postulating that genetic codechanged at an intermediate stage, when translationbecame ambiguous and thus some codons had morethan one assignment. According to this hypothesis, theprocess of codon reassignment is governed by selec-tion. If the new assignment is advantageous under given

AUmUGGln)

AUUAGln

ACUAGln)

conditions, selection may favor the new codon assign-ment by increasing the proportion of one amino acidover the proportion of another one in ambiguous sites.Ultimately, tRNA that determined the previous codonassignment is eliminated by mutation or deletion, thuscreating a prerequisite for the appearance of a new, non-ambiguous codon assignment [104].

Thus, selection accelerates each stage in the processof codon reassignment. There is at least one known casein which selection for one of the variants upon ambig-uous codon assignment promoted the appearance of analternative genetic code (in Candida sp.) [105], whichindicates that genetic code is evolving [106].

These hypotheses and other assumptions on themechanisms of code change are not mutually exclusive.Different codon reassignments could be explained bydifferent causes, both immediate and remote. Based onnumerous facts testifying to reassignment of termina-tion codons of the universal genetic code in ciliates, aquestion arises as to why this group of unicellulareukaryotes manifested such predisposition to the devel-opment of deviations simultaneously in several lin-eages.

An answer to this question can only partially satisfyus. If we accept the “simplified genome” hypothesis[107, 108], according to which departures from the uni-versal code in the mitochondrial genome conferred anadvantage to the organism, then there are many waysfor substantially reducing the genome in size.

Other question posed by modern genetics remain,for instance, why and how ciliates started using UAAand UAG as glutamine codons.

According to some authors [48, 49], who notedmany unique properties specific to ciliates, this groupof organisms split of the common eukaryotic tree veryearly in evolution. Since all eubacteria and eukaryotesemploy UAA, UAG, or UGA as termination codons(except mitochondria and micoplasms), it may well bethat ciliates had lost two of these codons (which begancoding for glutamine) earlier than these codonsappeared in other organisms via convergent evolution.

It was also suggested [49] that stop codons UAAand UAG were rarely used during the evolution of thecommon ancestor of proinfusorians and other organ-isms and that suppressor tRNAs for these codons wererather weak. If later these early suppressor tRNAs infact acquired mutated anticodons UUA and CUA forrecognition of UAA and UAG, respectively, then usingthese codons as termination ones could become fixed(after the splitting of ciliates from the common tree).

Then, a question arises: were (are) there primitiveorganisms that lack any termination codons, i.e., organ-isms that terminate translation of their proteins via theirexcessive synthesis?

According to Caron [66, 109], the unusual biologyof ciliates with their unique nuclear dimorphismreflects a very early stage in the evolution of livingorganisms. This author conjectures that initially, each



protein was encoded by a separate RNA fragment,without any termination codon. Then genes writtein onthe RNA were converted into distinct DNA genes. Theevent of deep structural organization of the genome inthe developing macronucleus, in which most DNA isrepresented by regions equal to one gene may shed lighton this stage of genome reorganization in the evolution.Thus, in Caron’s view [66], termination codonsappeared at late stages of the ciliate evolution.

It is tempting to presume that the appearance ofnuclear dimorphism, which is a distinguishing featureof all ciliates and unique for these organisms, mighthave promoted an important event in evolution—driftto departure from the universal genetic code.

CONCLUSIONS

To date, it is quite clear that reassignment of stan-dard codons of polypeptide chain termination in cili-ates, as well as in some other protist lineages, occursduring evolution. These departures from the universalgenetic code are evident in the mitochondrial genomeand were also shown for the nuclear system of unicel-lular eukaryotes. It is also clear that the deviatinggenetic code in ciliates has the secondary nature.

The evidence presented in this review clarifies twofeatures that are inherent characteristics of the unusualgenetic code in ciliates. First, code deviations are char-acteristic not only of species that appeared late in theevolution, but also of many early diverging lineages.Seconds, different reassignments of termination codonsin the universal genetic code are observed amongclosely related species, which testifies to a special flex-ibility of the process of codon reassignment.

Generally, the results discussed in the reviews indi-cate that genetic code is far from being ‘frozen” and hasbeen developing in many lineages. Genetic code of theliving world is still full of enigmas.

ACKNOWLEDGMENTS

I thank V.S. Artamonova and A.A. Makhrov forassistance in preparing the manuscript and the staff mem-bers of the library of the Vavilov Institute of GeneralGenetics, Russian Academy of Sciences, A.N. Kuz’minaand M.N. Perfil’eva for help in compiling the list of ref-erences, and the anonymous reviewer for carefully con-sidering the manuscript.

REFERENCES

1. Kawaguchi, Y., Honda, H., Tanuguchi-Morimura, J.,and Iwasaki, S., The Codon CUG Is Read as Serine inan Asporogenic Yeast Candida cylindracea, Nature,1989, vol. 341, pp. 164–166.

2. Schneider, S.U., Leible, M.B., and Yang, X., StrongHomology between the Small Subunit of Ribulose-1,5-Biphosphate Carboxylase Oxygenase of Two Species of

Acetabularia, Mol. Gen. Genet., 1989, vol. 218,pp. 445–452.

3. Fox, T.D, Natural Variation in the Genetic Code, Ann.Rev. Genet., A. Compbell, B.S. Barker, I. Herskawitz,L.A. Sandler, Eds., Polo Alto, California, 1987, vol. 21,pp. 67–91.

4. Lukashenko, N.P. and Rybakova, Z.I., Struktura ifunktsiya genomov prosteishikh (Structure and Functionof Protozoan Genomes), Moscow: Nauka, 1991.

5. Leinfelder, W., Zehelein, E., Mandrand-Berthelot, M.A.,and Böck, A., Gene for a Novel tRNA Species ThatAccepts L-Serine and Contranslationally Inserts Seleno-cysteine, Nature, 1988, vol. 331, pp. 723–725.

6. Leinfelder, W., Stadtman, T.C., and Böck, A., Occur-rence in vivo of Selenocysteyl-tRNASer in Escherichiacoli, J. Biol. Chem., 1989, vol. 264, pp. 9720–9723.

7. Leinfelder, W., Forchhammer, K., Veprek, B., et al.,In vitro Syntesis of Selenocysteinyl-tRNAUGA fromSeryl-tRNAUGA: Involvement and Characterization ofthe sel D Gene Product, Proc. Natl. Acad. Sci. USA,1990, vol. 87, pp. 543–547.

8. Zinoni, F., Birkmann, A., Stadtman, T., and Böck, A.,Nucleotide Sequence and Expression of the Selenocys-teine-Containing Polypeptide of Formate Dehydroge-nase (Formate-Hydrogen-Lyase-Linked) from Escheri-chia coli, Proc. Natl. Acad. Sci. USA, 1986, vol. 83,pp. 4650–4654.

9. Zinoni, F., Birkmann, A., Leinfelder, W., and Böck, A.,Cotranslational Insertion of Selenocysteine into For-mate Dehydrogenase from Escherichia coli Directed bya UGA Codon, Proc. Natl. Acad. Sci. USA, 1987,vol. 84, pp. 3156–3160.

10. Böck, A., Forchhammer, K., Heider, J., and Baron, C.,Selenoprotein Synthesis: An Expansion of the GeneticCode, Trends Biochem. Sci., 1991, vol. 16, pp. 463–467.

11. Böck, A., Forchhammer, K., Heider, J., et al., Seleno-cysteine: The 21st Aminoacid, Mol. Microbiol., 1991,vol. 5, pp. 515–520.

12. Hatfield, D., Choi, I.S., Mischke, S., and Owens, L.D.,Selenocysteyl-tRNA Recognize UGA in Beta vulgaris,a Higher Plant and in Glyocladiun virens, a FilamentousFungus, Biochem. Biophys. Res. Commun., 1992,vol. 184, pp. 254–259.

13. Hatfield, D. and Diamond, A.M., UGA: A Split Person-ality in the Universal Genetic Code, Trends Genet. Lett.,1993, vol. 9, pp. 69–70.

14. Söll, D., Enter a New Amino Acid, Nature, 1988,vol. 331, pp. 662–663.

15. Srinivasan, G., James, C.M., and Krzycki, J.A., Pyrrol-ysine Encoded by UAG in Archaea: Charging of aUAG-Decoding Specialized tRNA, Science, 2002,vol. 296, pp. 1459–1462.

16. Srinivasan, G., Translation of the Amber Codon inMethylamine Methyltransferase Genes of a Methano-genic Archaeon, PhD. Thesis, Ohio: State Univ. Micro-biol. Columbus, 2003, p. 163.

17. Alkins, J.F. and Gesteland, R.F., The 22nd Amino Acid,Science, 2002, vol. 296, pp. 1409–1410.

18. James, C.M., Ferguson, T.K., Leykam, J.F., and Krzy-cki, J.A., The Amber Codon in the Gene EncodingMonomethylamine Methyltransferase Isolated fromMethanosarcina barkeri Is Translated as a Sense

386


LUKASHENKO

Codon, J. Biol. Chem., 2001, vol. 276, pp. 34252–34258.

19. Hao, B., Gong, W., Ferguson, T.K., et al., A New UAG-Encoded Residue in the Structure of a MethanogenMethyltransferase, Science, 2002, vol. 296, pp. 1462–1465.

20. Ibba, M. and Söll, D., Genetic Code: Introducing Pyr-rolysine, Curr. Biol., 2002, vol. 12, pp. R464–R466.

21. Polycarpo, C., Ambrogelly, A., Bérubé, A., et al., AnAminoacyl-tRNA Synthetase That Specifically Acti-vates Pyrrolysine, Proc. Natl. Acad. Sci. USA, 2004,vol. 101, pp. 12450–12454.

22. Théobald-Dietrich, A., Frugier, M., Giegé, R., andRudinger-Thirion, J., Atypical Archaeal tRNA Pyrrol-ysine Transcript Behaves Towards EF-Tu as a TypicalElongator tRNA, Nucleic Acids Res., 2004, vol. 32,pp. 1091–1096.

23. Knight, R.D., Freeland, S.J., and Landweber, L.F.,Selection, History and Chemistry: The Three Faces ofthe Genetic Code, Trends Biochem. Sci., 1999, vol. 24,pp. 241–247.

24. Knight, R.D. and Landweber, L.F., The Early Evolutionof the Genetic Code, Cell, 2000, vol. 101, pp. 569–572.

25. Knight, R.D., Freeland, S.J., and Landweber, L.F.,Rewiring the Keyboard: Evolvability of the GeneticCode, Genetics, 2001, vol. 2, pp. 49–58.

26. Seilhamer, J.J. and Cummings, D.J., Altered GeneticCode in Paramecium tetraurelia Mitochondria: Possi-ble Evolutionary Trends, Mol. Gen. Genet., 1982,vol. 187, pp. 236–239.

27. Fox, D.T., Diverged Genetic Codes in Protozoans andBacterium, Nature, 1985, vol. 314, pp. 132–133.

28. Benne, R., De Vries, B.F., Van den Burg, J., et al., TheNucleotide Sequence of a Segment of Trypanosomabrucei Mitochondrial Maxicircle DNA That Containsthe Gene for Apocytochrome b and Some UnusualUnassigned Reading Frames, Nucleic Acids Res., 1983,vol. 11, pp. 6925–6941.

29. Johnson, B.J.B., Hill, G.C., and Donelson, J.E., TheMaxicircle of Trypanosoma brucei Kinetoplast DNAEncodes Apocytochrome b, Mol. Biochem. Parasitol.,1984, vol. 13, pp. 135–146.

30. de la Cruz, V.F., Neckelmann N., and Simpson, S.,Sequences of 6 Genes and Several Open ReadingFrames in the Kinetoplast Maxicircle DNA of Leishma-nia tarentolae, J. Biol. Chem., 1984, vol. 259,pp. 15136–15147.

31. Barrell, B.G., Bankier, A.T., and Drouin, J.A., A Differ-ent Genetic Code in Human Mitochondria, Nature,1979, vol. 282, pp. 189–194.

32. Anderson, S., Bankier, A.T., Barrell, B.G., et al.,Sequence and Organization of the Human Mitochon-drial Genome, Nature, 1981, vol. 290, pp. 457–474.

33. Anderson, S., de Bruijn, M.H.L., Coulson, A.R., et al.,Complete Sequence of Bovine Mitochondrial DNA:Conserved Features of the Mammalian MitochondrialGenome, J. Mol. Biol., 1982, vol. 156, pp. 683–717.

34. Bibb, M.J., Van Etten, R.A., Wright, C.T., et al.,Sequence and Gene Organization of Mouse Mitochon-drial DNA, Cell, 1981, vol. 26, pp. 167–180.

35. Roe, B.A., Ma, D.P., Wilson, R.K., and Wong, J.F.H.,The Complete Nucleotide Sequence of the Xenopus lae-vis Mitochondrial Genome, J. Biol. Chem., 1985,vol. 260, pp. 9759–9774.

36. de Bruijn, M.H.L., Drosophila melanogaster Mito-chondrial DNA: A Novel Organization and the GeneticCode, Nature, 1983, vol. 304, pp. 234–241.

37. Benne, R., Van Den Burg, J., Brakenhoff, J.P.J., et al.,Major Transcript of the Frameshifted cox II Gene fromTrypanosome Mitochondria Contains Four NucleotidesThat Are not Encoded in the DNA, Cell, 1986, vol. 46,pp. 819–826.

38. Wolstenholme, D.R., Clary, D.O., Macfarlane, J.L.,et al., Organization and Evolution of Invertebrate Mito-chondrial Genomes, in Archievements and Perspectivesof Mitochondrial Research: Biogenesis, Amsterdam:Elsevier, 1985, vol. 2, pp. 61–69.

39. Ohama, T., Osawa, S., Watanabe, K., and Jukes, T.H.,Evolution of the Mitochondrial Genetic Code: VI. AAAas an Asparagine Codon in Some Animal Mitochondria,J. Mol. Evol., 1990, vol. 30, pp. 329–332.

40. Fox, T.D. and Staempfli, S., Supressor of Yeast Mito-chondrial Ochre Mutations That Maps in or near the15S Ribosomal RNA Gene of mtDNA, Proc. Natl.Acad. Sci. USA, 1982, vol. 79, pp. 1583–1587.

41. Osawa, S. and Jukes, T.H., Evolution of the GeneticCode as Affected by Anticodon Content, Trends Genet.,1988, vol. 4, pp. 191–198.

42. Osawa, S. and Jukes, T.H., Codon Reassignment(Codon Capture) in Evolution, J. Mol. Evol., 1989,vol. 28, pp. 271–278.

43. Jukes, T.H. and Osawa, S., The Genetic Code in Mito-chondria and Chloroplasts, Experientia, 1990, vol. 46,pp. 1117–1126.

44. Macino, G., Coruzzi, G., Nobrega, F., et al., Use of theUGA Terminator as a Tryptophan Codon in Yeast Mito-chondria, Proc. Natl. Acad. Sci. USA, 1979, vol. 76,pp. 3784–3785.

45. Caron, F. and Meyer, E., Does Paramecium primaureliaUse a Different Genetic Code in Its Macronucleus?,Nature, 1985, vol. 314, pp. 185–188.

46. Meyer, E., Caron, F., and Guiard, B., Blocking ofin vitro Translation of Paramecium Messenger RNAs Isdue to Messenger RNA Primary Structure, Biochimie,1984, vol. 66, pp. 403–412.

47. Preer, J.R., Jr., Preer, L.B., Rudman, B.M., and Barnett, A.J.,Deviation from the Universal Code Shown by the Genefor Surface Protein 51A in Paramecium, Nature, 1985,vol. 314, pp. 188–190.

48. Horowitz, S. and Gorovsky, M.A., An Unusual GeneticCode in Nuclear Genes of Tetrahymena, Proc. Natl.Acad. Sci. USA, 1985, vol. 82, pp. 2452–2455.

49. Hanyu, N., Kuchino, Y., Nishimura, S., and Beier, H.,Dramatic Events in Ciliate Evolution: Alteration ofUAA and UAG Termination Codons to GlutamineCodons Due to Anticodon Mutations in Two Tetrahy-

mena tRN , EMBO J., 1986, vol. 5, pp. 1307–1311.

50. Herrick, G., Hunter, D., Williams, K., and Kotter, K.,Alternate Processing during Development of a Micro-nuclear Chromosome Family in Oxytricha fallax,Genes Dev., 1987, vol. 1 P, pp. 1047–1058.

ASGln



51. Harper, D.S. and Jahn, C.L., Differential Use of Termi-nation Codons in Ciliated Protozoa, Proc. Natl. Acad.Sci. USA, 1989, vol. 86, pp. 191–201.

52. Miceli, C., La Terza A., Melli M. Isolation and Struc-tural Characterization of cDNA Clones Encoding theMating Pheromone Er-1 Secreted by the CiliateEuplotes raikovi, Proc. Natl. Acad. Sci. USA, 1989,vol. 86, pp. 3016–3020.

53. Liang, A. and Heckmann, K., Blepharisma Uses UAAas a Termination Codon, Naturwissenschaften, 1993,vol. 80, pp. 225–226.

54. Meyer, F., Schmidt, H.J., Plumper, E., et al., UGA IsTranslated as Cysteine in Pheromone 3 of Euplotesoctocarinatus, Proc. Natl. Acad. Sci. USA, 1991,vol. 88, pp. 3758–3761.

55. Meyer, F., Schmidt, H.J., and Heckmann, K., Feromone4 Gene of Euplotes octocarinatus, Dev. Genet., 1992,vol. 13, pp. 16–25.

56. Liang, A., Brünen-Nieweler, C., Muramatsu, T., et al.,The Ciliate Euplotes octocarinatus Expresses TwoPolypeptide Release Factors of the Type eRF1, Gene,2001, vol. 262, pp. 161–168.

57. Jahn, C.L., Doctor, S.Z., Frels, J.S., et al., Structure ofthe Euplotes crassus Tec 1 and Tec 2 Elements: Identi-fication of Putative Transposase Coding Regions, Gene,1993, vol. 133, pp. 71–78.

58. Baroin Tourancheau, A., Tsao, N., Klobutcher, L.A.,et al., Genetic Code Deviation in the Ciliates: Evidencefor Multiple and Independent Events, EMBO J., 1995vol. 14, pp. 3262–3267.

59. Barahone, I., Soares, H., Cyrne, I., et al., Sequence ofOne α- and β-Tubulin Genes of Tetrahymena pyrifor-mis, J. Mol. Biol., 1988, vol. 202, pp. 365–382.

60. Hirono, M., Erdoh, H., Okada, N., et al., TetrahymenaActin: Cloning and Sequencing of the TetrahymenaActin Gene and Identification of Its Gene Product, J.Mol. Biol., 1987, vol. 194, pp. 181–192.

61. Prat, A., Katinka, M., Caron, F., and Meyer, E., Nucle-otide Sequence of the Paramecium primaurelia G Sur-face Protein: A Huge Protein with a Highly PeriodicStructure, J. Mol. Biol., 1986, vol. 189, pp. 47–60.

62. Preer, J.R., Preer, L.B., Rudman, B.U., and Barnett, A.J.,Deviation from the Universal Code Shown by the Genefor Surface Protein 51A in Paramecium, Nature, 1985,vol. 311, pp. 188–190.

63. Kink, J.A., Maley, M.E., Preston, R.R., et al., Mutationsin Paramecium Calmodulin Indicate Functional Differ-ences between the C-Terminal and N-Terminal Lobesin vivo, Cell, 1990, vol. 62, pp. 165–174.

64. Dupuis, P., Structure, organization et expression desgenes de tubuline chez la Paramecie, PhD Thesis, Paris:Univ. Paris, 1992, p. 230.

65. Lozupone, C.A., Knight, R.D., and Landweber, L.F.,The Molecular Basis of Nuclear Genetic Code Changein Ciliates, Curr. Biol., 2001, vol. 11, pp. 65–74.

66. Caron, F., Eukaryotic Codes, Experientia, 1990, vol. 46,pp. 1106–1117.

67. Greenwood, S.J., Sogin, M.L., and Linn, D.H., Phylo-genetic Relationship within the Class Oligohymeno-phorea, Phylum Ciliophora, Inferred from the CompleteSmall Subunit rRNA Gene Sequences of Colpidum

campylum, Glaucoma chattoni, and Opistonecta hen-neguyi, J. Mol. Evol., 1991, vol. 33, pp. 163–174.

68. Baroin, A., Perasso, R., Qu, L.-H., et al., Partial Phylog-eny of the Unicellular Eukaryotes Based on RapidSequencing of a Portion of 28S Ribosomal RNA, Proc.Natl. Acad. Sci. USA, 1988, vol. 85, pp. 3474–3478.

69. Baroin-Tourancheau, A., Delgado, P., Perasso, R., andAdoutte, A., Broad Molecular Phylogeny of Ciliates:Identification of Major Evolutionary Trends and Radia-tions within the Phylum, Proc. Natl. Acad. Sci. USA,1992, vol. 89, pp. 9764–9768.

70. Osawa, S., Jukes, T.H., Watanabe, K., and Muto, A.,Recent Evidence for Evolution of the Genetic Code,Microbiol. Rev., 1992, vol. 56, pp. 229–264.

71. Doolittle, R.F., Convergent Evolution: The Need to BeExplicit, Trends Biochem. Sci., 1994, vol. 19, pp. 15–18.

72. Inagaki, Y. and Doolittle, W.F., Class I Release Factorsin Ciliates with Variant Genetic Codes, Nucleic AcidsRes., 2001, vol. 29, pp. 921–927.

73. Ninio, J., Divergence in the Genetic Code, Biochem.System. Ecol., 1986, vol. 14, pp. 455–457.

74. Prescott, D.M., The DNA of Ciliated Protozoa, Micro-biol. Rev., 1994, vol. 58, pp. 233–267.

75. Lukashenko, N.P. and Rybakova, Z.I., Genetika infu-zorii: Tetrahymena i Paramecium (Genetics of Infuso-rians: Tetrahymena and Paramecium), Moscow: Nauka,1986.

76. Cherry, J.M. and Blackburn, E.H., The InternallyLocated Telomeric Sequences in the Germ-Line Chro-mosomes of Tetrahymena Are at the Ends of Transpo-son-Like Elements, Cell, 1985, vol. 43, pp. 747–758.

77. Wyman, C. and Blackburn, E.H., Tel-1 Transposon-Like Elements of Tetrahymena thermophila Are Asso-ciated with Micronuclear Genome Rearrangements,Genetics, 1991, vol. 128, pp. 57–67.

78. Hunter, D., Williams, K., Gantinhour, S., and Herrick, G.,Precise Excision of Telomere-Bearing Transposonsduring Macronuclear Development in Oxytricha fallax,Genes Dev., 1989, vol. 3, pp. 2101–2112.

79. Doak, T.G., Doerder, F.P., Jahn, C.L., and Herric, G., AFamily of Transposase Genes in Transposons Found inProkaryotes, Multilocular Eukaryotes and Ciliated Pro-tozoans, Proc. Natl. Acad. Sci. USA, 1993, vol. 91,pp. 942–946.

80. Wilson, A.K., Molecular Bases of Evolution, Sci. Am.,1985, no. 12, pp. 122–132.

81. Dubinin, N.P., Potential Change in DNA and Mutations, inMolekulyarnaya tsitogenetika (Molecular Cytogenetics),Moscow: Nauka, 1985, vol. 99, issue 1, pp. 3–21.

82. Kurland, C.G., Canback, B., and Berg, O.G., HorizontalGene Transfer: A Critical View, Proc. Natl. Acad. Sci.USA, 2003, vol. 100, pp. 9658–9662.

83. Gogarten, J.P., Doolittle, W.F., and Lawrence, J.G.,Prokaryotic Evolution in Light of Gene Transfer, Mol.Biol. Evol., 2002, vol. 19, pp. 2226–2238.

84. Kurland, C.G., Something for Everyone: HorizontalGene Transfer in Evolution, EMBO J., 2000, vol. 1,pp. 92–95.

85. Faguy, D.M. and Doolittle, W.F., Horizontal Transfer ofCatalase-Peroxidase Genes between Arhaea and Patho-genic Bacteria, Trends Genet., 2000, vol. 16, pp. 196–197.

388


LUKASHENKO

86. Hotopp, J.C.D., Clark, M.E., Oliveira, D.C.S.G., et al.,Widespread Lateral Gene Transfer from IntracellularBacteria to Multicellular Eukaryotes, Science, 2007,vol. 317, pp. 1753–1756.

87. Morrison, H.G., McArthur, A.G., Gillin, F.D., et al.,Genomic Minimalism in the Early Diverging IntestinalParasite Giardia lamblia, Science, 2007, vol. 317,pp. 1921–1926.

88. Doolittle, W.F., Phylogenetic Classification and theUniversal Tree, Science, 1999, vol. 284, pp. 2124–2129.

89. Woese, C.R., The Universal Ancestor, Proc. Natl. Acad.Sci. USA, 1998, vol. 95, pp. 6854–6859.

90. Woese, C.R., Interpreting the Universal PhylogeneticTree, Proc. Natl. Acad. Sci. USA, 2000, vol. 97,pp. 8392–8396.

91. Woese, C.R., On the Evolution of Cells, Proc. Natl.Acad. Sci. USA, 2002, vol. 99, pp. 8742–8747.

92. Karlberg, O., Canback, B., Kurland, C.G., and Anders-son, S.G.E., The Dual Origin of the Yeast MitochondrialProteome, Yeast, 2000, vol. 17, pp. 170–187.

93. Canback, B., Andersson, S.G.E., and Kurland, C.G.,The Global Phylogeny of Glycolytic Enzymes, Proc.Natl. Acad. Sci. USA, 2002, vol. 99, pp. 6097–6102.

94. Jukes, T.H., Bessho, Y., Ohama, T., and Osawa, S.,Release Factors and Genetic Code, Nature, 1991,vol. 352, p. 575.

95. Osawa, S., Evolution of the Genetic Code, Oxford:Oxford Univ. Press, 1995.

96. Jukes, T.H., Neutral Changes and Modifications of theGenetic Code, Theor. Popul. Biol., 1996, vol. 49,pp. 143–145.

97. Andersson, S.G.E. and Kurland, C.G., Codon Prefer-ences in Free-Living Microorganisms, Microbiol. Rev.,1990, vol. 54, pp. 198–210.

98. Andersson, S.G.E. and Kurland, C.G., An ExtremeCodon Preference Strategy: Codon Reassignment, Mol.Biol. Evol., 1991, vol. 8, pp. 530–544.

99. Schultz, D.W. and Yarus, M., A Simple and Sensitivein vivo Luciferase Assay for tRNA-Mediated NonsenceSuppression, J. Bacteriol., 1990, vol. 172, pp. 595–602.

100. Schultz, D.W. and Yarus, M., tRNA Structure and Ribo-somal Function: II. Interaction between AnticodonHelix and Other tRNA Mutations, J. Mol. Biol., 1994,vol. 235, pp. 1395–1405.

101. Schultz, D.W. and Yarus, M., tRNA Structure and Ribo-somal Function: I. tRNA Nucleotide 27–43 MutationsEnhance First Position Wobble, J. Mol. Biol., 1994,vol. 235, pp. 1381–1394.

102. Schultz, D.W. and Yarus, M., Transfer RNA Mutationand Malleability of the Genetic Code, J. Mol. Biol.,1994, vol. 235, pp. 1377–1380.

103. Schultz, D.W. and Yarus, M., On Malleability in theGenetic Code, J. Mol. Evol., 1996, vol. 42, pp. 597–601.

104. Yarus, M. and Schultz, D.W., Response: Further Com-ment on Codon Reassignment, J. Mol. Evol., 1997,vol. 45, pp. 1–8.

105. Santos, M.A., Chusman, C., Costa, V., et al., SelectiveAdvantages Created by Codon Ambiguity Allowed forthe Evolution of an Alternative Genetic Code in Can-dida spp., Mol. Microbiol., 1999, vol. 31, pp. 937–947.

106. Wagner, G.P. and Altenberg, L., Complex Adaptationsand the Evolution of Evoluability, Evolution, 1996,vol. 50, pp. 967–976.

107. Andersson, S.G. and Kurland, C.G., Genomic Evolu-tion Drives the Evolution of the Translation System,Biochem. Cell Biol., 1995, vol. 73, pp. 775–787.

108. Andersson, S.G. and Kurland, C.G., Reductive Evolu-tion of Resident Genomes, Trends Microbiol., 1998,vol. 6, pp. 263–268.

109. Caron, F., Deviations from the ‘Universal’ GeneticCode, Microbiol. Sci., 1986, vol. 3, pp. 36–40.

Documents

Evolutionary deviations from the universal genetic code in ciliates