"Botany, Department of"@en . "Science, Faculty of"@en . "Non UBC"@en . "DSpace"@en . "BMC Evolutionary Biology. 2006 Jun 14;6(1):46"@en . "Patron et al."@en . "Patron, Nicola J."@en . "Rogers, Matthew B."@en . "Keeling, Patrick J. (Patrick John), 1969-"@en . "2015-11-04T19:53:43Z"@* . "2006-06-14"@en . "Background:\r\n The nucleomorphs associated with secondary plastids of cryptomonads and chlorarachniophytes are the sole examples of organelles with eukaryotic nuclear genomes. Although not as widespread as their prokaryotic equivalents in mitochondria and plastids, nucleomorph genomes share similarities in terms of reduction and compaction. They also differ in several aspects, not least in that they encode proteins that target to the plastid, and so function in a different compartment from that in which they are encoded.\r\n \r\n \r\n Results\r\n Here, we test whether the phylogenetically distinct nucleomorph genomes of the cryptomonad, Guillardia theta, and the chlorarachniophyte, Bigelowiella natans, have experienced similar evolutionary pressures during their transformation to reduced organelles. We compared the evolutionary rates of genes from nuclear, nucleomorph, and plastid genomes, all of which encode proteins that function in the same cellular compartment, the plastid, and are thus subject to similar selection pressures. Furthermore, we investigated the divergence of nucleomorphs within cryptomonads by comparing G. theta and Rhodomonas salina.\r\n \r\n \r\n Conclusion\r\n Chlorarachniophyte nucleomorph genes have accumulated errors at a faster rate than other genomes within the same cell, regardless of the compartment where the gene product functions. In contrast, most nucleomorph genes in cryptomonads have evolved faster than genes in other genomes on average, but genes for plastid-targeted proteins are not overly divergent, and it appears that cryptomonad nucleomorphs are not presently evolving rapidly and have therefore stabilized. Overall, these analyses suggest that the forces at work in the two lineages are different, despite the similarities between the structures of their genomes."@en . "https://circle.library.ubc.ca/rest/handle/2429/55257?expand=metadata"@en . "ralssBioMed CentBMC Evolutionary BiologyOpen AcceResearch articleComparative rates of evolution in endosymbiotic nuclear genomesNicola J Patron*, Matthew B Rogers and Patrick J KeelingAddress: Canadian Institute for Advanced Research, Department of Botany, University of British Columbia, Vancouver, V6T 1Z4, CanadaEmail: Nicola J Patron* - nicolapatron@mac.com; Matthew B Rogers - mbrogers@interchange.ubc.ca; Patrick J Keeling - pkeeling@interchange.ubc.ca* Corresponding author AbstractBackground: The nucleomorphs associated with secondary plastids of cryptomonads andchlorarachniophytes are the sole examples of organelles with eukaryotic nuclear genomes.Although not as widespread as their prokaryotic equivalents in mitochondria and plastids,nucleomorph genomes share similarities in terms of reduction and compaction. They also differ inseveral aspects, not least in that they encode proteins that target to the plastid, and so function ina different compartment from that in which they are encoded.Results: Here, we test whether the phylogenetically distinct nucleomorph genomes of thecryptomonad, Guillardia theta, and the chlorarachniophyte, Bigelowiella natans, have experiencedsimilar evolutionary pressures during their transformation to reduced organelles. We comparedthe evolutionary rates of genes from nuclear, nucleomorph, and plastid genomes, all of whichencode proteins that function in the same cellular compartment, the plastid, and are thus subjectto similar selection pressures. Furthermore, we investigated the divergence of nucleomorphswithin cryptomonads by comparing G. theta and Rhodomonas salina.Conclusion: Chlorarachniophyte nucleomorph genes have accumulated errors at a faster ratethan other genomes within the same cell, regardless of the compartment where the gene productfunctions. In contrast, most nucleomorph genes in cryptomonads have evolved faster than genes inother genomes on average, but genes for plastid-targeted proteins are not overly divergent, and itappears that cryptomonad nucleomorphs are not presently evolving rapidly and have thereforestabilized. Overall, these analyses suggest that the forces at work in the two lineages are different,despite the similarities between the structures of their genomes.BackgroundWhile the primary acquisition of the plastid from a free-living cyanobacterium is believed to have occurred onlyonce [1], plastids have continued to spread througheukaryotes by means of secondary and tertiary endosym-biosis. This is the process whereby a plastid-containing,(exemplified by those of plants) have two membranes,while secondary plastids have additional membranes cor-responding to the outer membrane of the engulfedeukaryote and the phageosomal membrane of the host, aswell as the original membranes of the primary plastid[2,3], although in some lineages membranes have subse-Published: 14 June 2006BMC Evolutionary Biology 2006, 6:46 doi:10.1186/1471-2148-6-46Received: 16 March 2006Accepted: 14 June 2006This article is available from: http://www.biomedcentral.com/1471-2148/6/46\u00C2\u00A9 2006 Patron et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 12(page number not for citation purposes)free-living eukaryote is consumed by another eukaryoticcell and becomes an organelle itself. Primary plastidsquently been lost. The nucleus of the engulfed cell is, in allbut two described cases, absent, and the genes encodingBMC Evolutionary Biology 2006, 6:46 http://www.biomedcentral.com/1471-2148/6/46plastid-targeted proteins having been relocated to the hostnucleus [4-6]. The exceptions are the cryptomonads andchlorarachniophytes, which contain nucleomorphs, theremnant nuclei of the plastid-containing algae that wereengulfed in the secondary endosymbioses that gave rise tothese lineages (Figure 1). The cryptomonad endosymbi-ont is derived from a red alga, while that of chlorarachni-ophytes is derived from a green alga. Their genomesencode very few genes, and most of them are housekeep-ing genes for replication, transcription and protein fold-ing and degradation [7,8]. A handful of proteins related toplastid function have also been retained, however, theyare relatively few [7,9,10]. The periplastidial space (equiv-alent to the cytosol of the engulfed alga) itself has specificmetabolic processes, such as starch synthesis in crypto-monads, but most of the proteins for these pathways aremissing from the nucleomorph genome [7] and are antic-ipated to be found in the nuclear genome, as has beenshown for a few examples [11].The nucleomorph is often thought of as an anomaly, arare occurrence, since it is known only in cryptomonadsand chlorarachniophytes, but if one considers 'loss orgain' rather than 'presence or absence' then it is perhapsnot so anomalous. All lineages that are known to containsecondary plastids (haptophytes, heterokonts, crypto-monads, dinoflagellates, apicomplexans, euglenids andchlorarachniophytes) have ancestors that contained anucleomorph. Depending on the number of secondaryendosymbiotic events that took place, which is still con-tentious [3,12-14], the number of nucleomorph lossesand gains differs. The balance of molecular evidencepoints to two events involving green algae [15,16] andone involving a red alga [17-19]. With respect to greenalgae this means one lineage lost its nucleomorph andone retained it. With respect to red algae, this means a sin-gle nucleomorph gain (if one accepts the chromalveolatehypothesis [20]) and perhaps only one loss, if cryptomon-ads are the deepest branch of chromalveolates, or perhapstwo if they diverged later. Overall, lineages retainingnucleomorphs may be as common as lineages that lostthem, or at least the proportions are very similar. What-ever the case, nucleomorphs existed in the commonancestors of a great deal of algal diversity, so the study ofthe lineages in which they remain may help us understandthe process of secondary (and higher order) endosymbi-otic events, especially the reduction and subsequent lossof the enslaved genome.Cryptomonads and chlorarachniophytes arose from sepa-rate endosymbiotic events, and neither host cell nor endo-symbiont are very closely related. Yet the nucleomorphgenomes of the cryptomonad, Guillardia theta [7] and theEndosymbiotic events that gave rise to cryptomonads and chlorarachniophytesFigure 1Endosymbiotic events that gave rise to cryptomonads and chlorarachniophytes.- gene transfer to nucleus- differentiation to red and green alage and glaucophtes- gene transfer to nucleus- loss of algal nucleus in some lineages- compaction to nucleomorph in cryptophytes and chlororachneophytesAlgae with two-membraned plastidsPrimary Endosymbiosis (single event)ChloroachneophyteFour-membraned plastid. Nucleomorph between second and third membraneCryptophyteFour-membraned plastid. Nucleomorph between second and third membrane. Outer membrance contiguouswith endoplasmic reticulum.Secondary Endosymbiosis(at least once involving a red alga, at least twice involving green algae)Page 2 of 12(page number not for citation purposes)chlorarachniophyte, Bigelowiella natans [8-10] share sev-eral characteristics. Both nucleomorph genomes haveBMC Evolutionary Biology 2006, 6:46 http://www.biomedcentral.com/1471-2148/6/46undergone substantial gene loss and are ultra-compactcompared to their free-living relatives in the red and greenalgae. Some of these features, such as overlapping genes,short intergenic regions, a reduction in elements liketransposons, and the presence of multigene transcriptshave been found in other compact eukaryotic genomessuch as microsporidia [21,22]. Compact genomes andmany of these features are common to endosymbionts ingeneral, however, until the sequences of the G. theta andB. natans, nucleomorph genomes were completed, allknown endosymbiont genomes have been of prokaryoticorigin. The best examples of prokaryotic endosymbiontgenomes are those of the mitochondrion, once a free-liv-ing alpha-proteobacterium, and the chloroplast, once afree-living cyanobacterium [1]. Also well described,although not organellar, are the bacterial endosymbiontsof insects, of these there are several complete genomes; forexample, Wolbachia [23-25], Buchnera [26], Wigglesworthia[27] and Blochmania [28], the features of which have beencompared and defined [29-31]. These bacteria residewithin a range of diverse insects but, while they retain cer-tain distinct genes that can be linked to the physiology oftheir host, they show similar patterns of genome reduc-tion, strong mutational AT bias and strict amino acid biasat high expression genes [32]; an effect of selection againstmutation driven amino acid changes [31,33]. The ATmutational pressure in endosymbionts, is sometimes veryextreme; estimated to be a remarkable 90% GC->AT inBuchnera [34]. A universal AT mutational bias, has beensuggested because many types of spontaneous mutations(e.g. the deamination of cytosine) cause GC to AT changes[35]. The effects of this mutational bias may be more pro-nounced and gene loss more rapid in small, endosymbi-ont genomes because they are deficient in at least oneDNA repair mechanism, experience strong genetic driftand have experienced a relaxation of selection in the intra-cellular environment in comparison to free-living exist-ence [31,33].There is less chromosomal information for eukaryoticobligate intracellular parasites, however certain alveolateand microsporidian genomes show some similar charac-teristics such as genome compaction [22], AT bias[7,36,37], codon bias [38,39] and extreme divergence. Asummary of the features of organelle-, obligate-intracellu-lar- and nucleomorph-genomes is given in Table 1. Thesefeatures are important to consider as measure of how unu-sual, or not, nucleomorph genomes are.With the recent availability of red algal [40] and greenalgal [41] genomic data we are for the first time in a posi-tion to do comparative genomics between nucleomorphsof both cryptomonads and chlorarachniophytes andwhether the phylogenetically distinct nucleomorphgenomes of G. theta and B. natans have experienced simi-lar evolutionary pressures that influenced genome-widevariation in predictable ways and with the same severityand whether these effects are in common to thosedescribed in other enslaved nuclei. Proteins from bothnucleomorph genomes have been observed to reside onlong branches of phylogenetic trees indicating that theyare poorly conserved [42-45], however this has never beeninvestigated at the genomic level. It is also assumed thatnucleomorph genes are highly derived because the pro-teins function within a sub-cellular compartment, theperiplastidial space, where selection is relaxed due toreduced interactions with other proteins. However, boththe G. theta and B. natans nucleomorphs encode proteinsthat are directed to the plastid. Proteins that function inthe plastid are presumably subject to similar selectionpressures in organisms with nucleomorphs as they are inother algae. We have therefore used plastid proteinsencoded in the plastid genome, the nucleomorph, or thenucleus, to examine differences in rates of evolution in thedifferent genomes to determine whether the nucleo-morph is evolving at a dissimilar rate to the plastid andnuclear genomes. We also investigate the overall variabil-ity of evolutionary rates of nucleomorph-encoded pro-teins and their homologues in other species to determineif the proteins still encoded within these genomes are gen-erally well conserved, and whether this can shed light ontheir retention in the nucleomorph. By comparing pro-teins from the nucleomorph of two cryptomonads, G.theta and Rhodomonas salina, we also investigate whethercryptomonad nucleomorph genomes are diverging at thesame rate as their nuclear genomes.Results and discussionPlastid-encoded proteins are less divergent than nuclear-encoded plastid-targeted proteinsThe plastids of both G. theta and B. natans use proteinsencoded in the nuclear genome, the nucleomorphgenome and the plastid itself. Of the 147 proteinsencoded in the G. theta plastid genome [46] 45 are alsopresent in the plastid genomes of the red alga C. merolaeand the green plant A. thaliana. Of the 57 proteinsencoded in the B. natans plastid genome, 53 are alsopresent in the plastid genome of the green alga C. rein-hardtii and A. thaliana. One of these proteins, YCF1proved to be unalignable and was excluded from the anal-ysis. Since the genomes of all plastids are descendents ofthe cyanobacterial primary plastid ancestor, these pro-teins are homologues (although some gene duplicationshave occurred in certain plastid lineages).The average distances, calculated by all methods (with orPage 3 of 12(page number not for citation purposes)examples of their free-living relatives, with the plant Ara-bidopsis thaliana serving as an outgroup. Here we testwithout substitution matrices, see methods) between theplastid-encoded proteins of G. theta, C. merolae and A.BMC Evolutionary Biology 2006, 6:46 http://www.biomedcentral.com/1471-2148/6/46thaliana are smaller than the average distances betweenthe nuclear-encoded proteins (Figure 2a). The distancefrom G. theta to A. thaliana and the distance from C. mero-lae to A. thaliana is slightly greater than between the G.theta and C. merolae, indicating that red and green plastidsare more distant than primary and secondary red plastids(red, Figure 2a), however the difference is not substantial.The average distances between plastid-encoded proteinsof B. natans, A. thaliana and C. reinhardtii, the plastids ofwhich are all of the green lineage, are also smaller thannucleus-encoded proteins (Figure 2b). However, the threeplastids are roughly equidistant indicating that secondaryendosymbiosis did not affect the speed of divergence ofplastid genes in B. natans (red, Figure 2b).Nuclear-encoded plastid genes have been transferred fromthe plastid genome during endosymbiosis resulting inreduced organelle genomes. Nucleus-encoded plastid-tar-geted proteins of G. theta and B. natans [16,47] were iden-tified from ongoing expressed sequence tag (EST)sequencing projects (see methods) by similarity to knownplastid proteins and, where present, the characteristics oftargeting N-terminal presequences that direct these pro-teins to their secondary plastid; a signal peptide flowed bya transit peptide [48]. In G. theta transit peptides have thecharacteristics of red algal transit peptides [49], and in B.natans, of green algal transit peptides [47]. For each ofgenomes of C. merolae (for G. theta) or C. reinhardtii (forB. natans). Twenty-four nucleus-encoded plastid-targetedproteins were found in EST data from G. theta for whichthere were identifiable homologues in C. merolae and A.thaliana, and 45 plastid proteins were identified from B.natans for which there were identifiable homologues fromC. reinhardtii and A. thaliana. A G. theta gene encoding anisoform of glycogen (starch) synthase was excluded fromthe analysis since starch is accumulated in the periplastid-ial space in this species, but its homologue in C. merolae isactive in the cytosol and the its homologues in green algaeand plants are active in the plastid [50]. Also excludedfrom the analysis was a nuclear copy of the tha4 gene alsofound in the G. theta nucleomorph. The protein encodedby this gene was longer than the nucleomorph protein,which, in comparison to isoforms from other speciesappears truncated. It is possible that the nuclear tha4 geneis a recent transfer that has assumed the function of thenucleomorph-encoded protein and that the truncated,nucleomorph copy is in the process of being lost.The average distances between nuclear-encoded plastid-targeted proteins of G. theta, A. thaliana and C. merolae arelarger than the plastid-encoded proteins, and are almostidentical between the three species (blue, Figure 2a). Sim-ilarly the average distances between nuclear-encoded plas-tid-targeted proteins of B. natans, A. thaliana and C.Table 1: Features of endosymbiont and organelle genomes. '*' \u00E2\u0080\u0093 no genome of free-living relative, '?' \u00E2\u0080\u0093 not determined.Organism Genome Compacted (compared to free-living relative)AT bias Codon BiasExpression bias (highly expressed genes are less divergent and GC rich)Divergent (compared to free-living relative)Organelle genomes of prokaryotic originMitochondria Y Y Y Y YPlastids Y Y Y Y YProkaryotic obligate intracellular symbiontsWolbachia Y Y Y Y YWigglesworthia Y Y Y Y YBuchnera Y Y Y Y YOrganelle genomes of eukaryotic originGuillardia theta nm Y Y ? ? YBigelowiella natans nmY Y ? ? YEukaryotic obligate Intracellular parasitesPlasmodium -* Y Y Y -*Toxoplasma -* Y ? ? -*Cryptosporidium -* No ? ? -*E. cuniculi Y No ? ? YPage 4 of 12(page number not for citation purposes)these proteins homologues were identified from thenuclear genomes of A. thaliana and from the nuclearreinhardtii arealmost equal (blue, Figure 2b), but largerthan distances for plastid-encoded proteins from the sameBMC Evolutionary Biology 2006, 6:46 http://www.biomedcentral.com/1471-2148/6/46taxa. Distances calculated for both plastid and nuclear-encoded proteins using the Dayhoff and VT substitutionmatrices were larger than the average number of substitu-tions (i.e. calculated without substitution matrix), whichshows that amino acids were most often substituted withsimilar residues, suggesting functional conservation.Overall, these analyses show that nucleus-encoded plas-tid-targeted proteins are on average more divergent thanproteins encoded in the plastid genome. Two possiblecauses for this observation are 1) the rates of general sub-stitution are higher in nuclear genomes, or 2) the genesretained in the plastid genome are those under the greatestselection. A combination of both factors may occur. Thesetive distances between the species for which the rates ofdivergence of the nucleomorph genomes can be com-pared.Nucleomorph encoded, non-plastid proteinsPrevious phylogenetic observations of nucleomorph-encoded proteins, have led to speculation that the nucle-omorph genomes are extraordinarily divergent, howeverthese studies have been made of proteins that do not tar-get to the plastid. The nucleomorph genomes of G. thetaand B. natans each only encode a handful of plastid pro-teins, and even fewer for periplastidial metabolism. Therest of the genes encode proteins to support the nucleo-morph; proteins for transcription, translation, proteinRadar graphs of average distance of plastid-, nucleus-, and nucleomorph-encoded plastid proteins, and nucleomorph-encoded non-plastid pr teins of: (A) the cryptophyte G. th ta (GT), the red algae C. merolae (CM) and the pl t A. thaliana (AT); and (B) the chlororachniophyte B. natans (BN), he green algae C. reinhartii (CR) a d A. thaliana (AT) w thout a d with substitutionmatrice (Day off, VT)Figu 2Radar graphs of average distance of plastid-, nucleus-, and nucleomorph-encoded plastid proteins, and nucleomorph-encoded non-plastid proteins of: (A) the cryptophyte G. theta (GT), the red algae C. merolae (CM) and the plant A. thaliana (AT); and (B) the chlororachniophyte B. natans (BN), the green algae C. reinhartii (CR) and A. thaliana (AT) without and with substitution matrices (Dayhoff, VT). In each case the secondary endosymbiont-containing organism is compared to a free living example of its symbiont (red or green algae for A and B, respectively) and the plant A. thaliana as an outgroup. Note, scale is different for graphs either without or with substitution matrices.00.20.40.60.8GT:CMGT:ATCM:ATC. merolaeG. thetaA. thaliana00.40.81.21.62GT:CMGT:ATCM:ATC. merolaeG. thetaA. thaliana00.40.81.21.62GT:CMGT:ATCM:ATC. merolaeG. thetaA. thaliana00.20.40.60.8BN:CRAT:BNCR:AT00.40.81.21.62BN:CRAT:BNCR:AT00.40.81.21.62BN:CRAT:BNCR:ATC. reinhardtii B. natansA. thalianaC. reinhardtii B. natansA. thalianaC. reinhardtii B. natansA. thaliananucleomorph-encoded plastid-targetedplastid nucleus-encoded plastid-targeted nucleomorph-encoded non-plastidDayoff VTABPage 5 of 12(page number not for citation purposes)results for plastid-encoded and nucleus-encoded plastid-targeted proteins are an important indication of the rela-folding and degradation and RNA metabolism [7,8].These proteins are active within this discrete and reducedBMC Evolutionary Biology 2006, 6:46 http://www.biomedcentral.com/1471-2148/6/46cellular space and do not interact with very many otherproteins, and therefore selection pressure is hypothesizedto be relaxed resulting in proteins of greater relative diver-gence.To test this, we selected nucleomorph-encoded genes forproteins that function in the periplastidal space, and com-pared the rates of evolution of these genes with homo-logues from nuclear genomes (black, Figure 2a). Averagedistances between nucleomorph-encoded proteins of G.theta and nuclear-encoded homologues from A. thaliana&C. merolae are larger than the distances between proteinsthat are plastid-encoded in all species (red), whereas thereis less difference between these protein distances andthose of proteins that are nucleus-encoded in all species(blue). However, significantly, the relative distancesbetween taxa are not equal. The distance to G. theta fromboth A. thaliana and C. merolae is greater than the differ-ence between A. thaliana and C. merolae (black, Figure 2a).This is consistent with relaxed selective pressure for pro-teins in the periplastidal space. This trend is even morepronounced in the chlorarachniophyte. Average distancesbetween nucleomorph-encoded proteins of B. natans, andnuclear-encoded homologues from A. thaliana &C. rein-hardtii (black, Figure 2b) are larger than either plastid(red) or nucleus-encoded plastid-targeted proteins (blue),and the distances are also not equal. The distance to B.natans from both A. thaliana and C. reinhardtii is muchgreater than the difference between A. thaliana and C. rein-hardtii (black, Figure 2b).Overall, this confirms expectations that protein-codinggenes encoded and active in the nucleomorph and peri-plastidal space are accumulating mutations faster thannuclear or plastid-encoded proteins. By themselves, how-ever, these observations do not allow us to distinguishbetween rapid mutation rates in the nucleomorphgenomes as opposed to relaxed selective pressures on pro-teins active within the periplastidal space.The rate of divergence of nucleomorph-encoded plastid-targeted proteins is restrained in cryptomonads but not in chlorarachniophytesThe nucleomorph of G. theta contains 19 genes thatencode plastid-targeted proteins of known function [7].Of these, only two isoforms of Clp protease, and Cpn60are also represented in the nucleomorph of B. natans, (theother 16 genes are not common to B. natans), which con-tains 14 further genes encoding proteins targeted to theplastid [8].Why these plastid-targeted proteins remain encoded inthe nucleomorph may be the key to the existence of theof the genome. A variety of biological explanations havebeen suggested for the retention of certain core proteins inmost chloroplast and mitochondrial genomes [51,52],however, given that the nucleomorph is itself a remnantnucleus none of these apply to nucleomorphs. It remainsa possibility that, despite there being almost no overlap inplastid-protein content, these proteins are retained ineach genome for biological reasons specific to each sys-tem, as hypothesized for core genes of the mitochondrialand plastid genomes. Alternatively, they may be genesthat simply have not yet been successfully transferred tothe nucleus. Indeed, in this study we identified a nuclearcopy of a nucleomorph gene, tha4, which may have led tothe demise of the nucleomorph-encoded gene relativelyrecently showing the ongoing nature of the process. Byextension, it is possible that only the few genes whose pro-teins are more permissive to mutation can tolerate thehigh mutation rate of nucleomorph genomes. Selectionpressure favouring the successful transfer of genes for pro-teins under tighter selection for sequence conservationwould be stronger. This would suggest that the genes forplastid-targeted proteins remaining in the nucleomorphswould be divergent compared with homologues in othereukaryotes, perhaps as divergent as other nucleomorphproteins on average.To test these hypotheses, we first compared the relativedistances of nucleomorph-encoded plastid-targeted pro-teins to nucleus-encoded plastid-targeted and plastid-encoded proteins (Figure 2). Fifteen nucleomorph-encoded plastid-targeted proteins of G. theta had identifi-able homologues in the nuclear genomes of C. merolaeand A. thaliana and 17 nucleomorph-encoded plastid-tar-geted proteins of B. natans had identifiable homologuesin the nuclear genomes of C. reinhardtii and A. thaliana.Average distances between nucleomorph-encoded plas-tid-targeted proteins from G. theta and nuclear-encodedhomologues from A. thaliana and C. merolae are largerthan plastid-encoded proteins, but similar to nucleus-encoded plastid-targeted proteins. The distances betweenthe three species are not equal. As for the plastid-encodedproteins, the distance to A. thaliana from both G. theta andC. merolae is much greater than the difference between G.theta and C. merolae (green, Figure 2a). Again, this indi-cates that red and green plastids are more distant than pri-mary and secondary red plastids (as expected). Howeverthis result is interesting because it is contrary to the resultsobtained for nucleomorph-encoded non-plastid proteins,which suggested that nucleomorph proteins were evolv-ing at a faster rate. In the case of the chlorarachniophyte,average distances between nucleomorph-encoded plastid-targeted proteins from B. natans, and nucleus-encodedPage 6 of 12(page number not for citation purposes)genome itself, since almost all other nucleomorph-encoded proteins are for self-maintenance and expressionhomologues from A. thaliana and C. reinhardtii are alsogreater than plastid-encoded proteins. In this case, how-BMC Evolutionary Biology 2006, 6:46 http://www.biomedcentral.com/1471-2148/6/46ever, the results contrast sharply with G. theta because thedistance to B. natans from both A. thaliana and C. rein-hardtii is much greater than the difference between A. thal-iana and C. reinhardtii (green, Figure 2b), showing that inthis case both types of nucleomorph-encoded proteins(plastid and periplastidal) have experienced acceleratedevolution.Relative rate tests can be used to measure the degree ofdivergence of two genes from an equally distant outgroup[53,54]. Relative rate tests were performed to determinedifferences in rates of evolution of individual genesencoding plastid-targeted proteins from the threegenomes of both B. natans and G. theta and their homo-logues in the green alga C. reinhardtii and the red algal C.merolae. A. thaliana was used as an outgroup for both theB. natans and G. theta datasets. Relative rates were calcu-lated using RRTree [55] and were tested at a 95% confi-dence interval (Table 2). Nucleomorph-encoded plastidproteins in B. natans fail the relative rate test at a 95% con-fidence level at a far high frequency than plastid proteinsencoded in either the chloroplast or nuclear genomes. Ofthe plastid proteins encoded in the B. natans nucleo-morph genome, 82% fail the relative rate test, in each casethe peptide is evolving more rapidly in B. natans. Similarproportions of nuclear-encoded plastid-targeted proteins(33%) and plastid-encoded proteins (37%) fail the rela-tive rate test in B. natans in which cases B. natans is typi-cally the most rapidly evolving peptide. In G. theta,nucleomorph-encoded plastid-targeted proteins fail therelative rate test more frequently than those encoded inthe plastid or nucleus, but the difference is not nearly aspronounced as in B. natans. In fact, nucleomorph encodedplastid-targeted proteins in G. theta only fail the relativerate test 11% more frequently than nuclear-encoded plas-tid-targeted proteins in which G. theta is the most rapidlyevolving taxon. Interestingly, of the 17% of the plastid-encoded peptides that fail the relative rate test, G. theta isnot the most rapidly evolving ingroup. This may indicatethat the plastid of C. merolae is evolving at an acceleratedrate compared to that of G. theta.Overall, the rate of evolution of plastid proteins encodedin the nucleomorph of cryptomonads is in line with thoseencoded in the nucleus, despite the fact that other nucle-omorph-encoded proteins are generally evolving at ahigher rate. In chlorarachniophytes, however, the nucleo-morph-encoded plastid-targeted proteins are evolvingmuch faster than those encoded in the nucleus (as wasalso seen for non-plastid nucleomorph-encoded pro-teins), which provides one of the first indications that themode of evolution in these two genomes is fundamentallydifferent.The proteins retained in nucleomorph genomes are not fast-evolving in other organismsTo further test if the genes retained in the nucleomorphgenome are present because the proteins they encode aretolerant of high mutation rates, we compared the evolu-tionary rates of these proteins in other organisms to theaverage rates of other plastid-targeted proteins in theirnuclear genomes as well as genes retained in the plastidgenome. This would reveal if the proteins encoded in thenucleomorph genomes were generally more divergent inall species or not. Since these are proteins of plastid origin,the complete genomes of photosynthetic eukaryotes wereused, including the diatom Thalassiosira pseudonana, andthe distance of these proteins compared to an extant free-living plastid relative; the cyanobacterium SynechocystisPCC 6803. This analysis showed that plastid proteins thatare encoded in the nucleomorph of either G. theta or B.natans (green bars, Figure 3) are not significantly moredivergent in any other species than plastid-targeted pro-teins are in general (Figure 3). We should point out thatdetecting any differences now may be hampered by thefact that all nucleus-encoded plastid-targeted proteinsmay have existed for some time in a nucleomorph-likegenome that has since been lost. This analysis also showsthat plastid-encoded proteins are generally less divergent(red bars, Figure 3), as shown in Figure 1, however in thisanalysis the range of error was large because of the greatdistance to the cyanobacterium.Table 2: Percentage relative rates rest (calculated by RRTree) failures (P < 0.05; 95% confidence) of plastid proteins encoded in three genomesOrganism Genome % Failure % failure when the B. natans or G. theta encoded protein is evolving fasterG. theta plastid 17 0nucleus 26 22nucleomorph 33 33B. natans plastid 37 31nucleus 33 33Page 7 of 12(page number not for citation purposes)nucleomorph 82 82BMC Evolutionary Biology 2006, 6:46 http://www.biomedcentral.com/1471-2148/6/46Are cryptomonad nucleomorphs still diverging rapidly?One of the important observations of prokaryoticenslaved genomes is that, despite the divergence fromtheir free-living relatives, enslaved genomes themselvesare generally closely related. For example on a phyloge-netic tree of gamma-proteobacteria there is a long branchleading to the Buchnera aphidicola clade, but strains of B.aphidicola from many aphid species are separated by rela-tively short distances[56]. This is important because itshows that there are large changes after enslavement, esti-mated to be 200\u00E2\u0080\u0093250 million years ago [57], but then thegenomes become stable [31]. This has been shown inother systems, see table 1. So, these genomes, while highlyderived, are apparently stable in this derived condition. Inthe case of bacterial endosymbionts of invertebrates thereSimilarly, while there may still be some ongoing genetransfer from plastid and mitochondrial genomes [58-61],it seems that a core genome is relatively stable [51,52]. Toextrapolate to endosymbiont nuclear genomes, it is criti-cal to know if the rate of divergence between two nucleo-morph genomes is similar or different than the rate ofdivergence between their hosts. If they are behaving asother enslaved genomes do, then the distance will besmaller and perhaps this is one indication of havingreached stability. If the forces driving the divergent natureof nucleomorphs are still active, then they will be moredivergent than their hosts.The average distances between nucleus and nucleomorph-encoded plastid-targeted proteins, and nucleomorph pro-teins active in the periplastidial space were calculated fortwo cryptomonads, G. theta, and R. salina, and comparedto their homologues in C. merolae. This analysis was madewith homologues of six nucleomorph-encoded plastid-targeted protein, six nucleus-encoded plastid-targetedproteins, and nine nucleomorph-encoded non-plastidproteins. The distances between nucleomorph-encodedproteins (both plastid and non-plastid) from G. theta andR. salina are actually less than the distances betweennucleus-encoded proteins (Figure 4). Moreover, for bothsets of nucleomorph-encoded proteins and for thenucleus-encoded proteins, the distance to C. merolae fromboth G. theta and R. salina is greater than the distancebetween G. theta and R. salina. The distance between R.salina and G. theta for nucleomorph non-plastidproteinsis slightly greater than for plastid-targeted proteins. Takentogether, these results suggest that the nucleomorph pro-teins of cryptomonads are not diverging rapidly but, liketheir plastid genomes, are evolving at a slower rate thantheir nuclear genomes. However, the proteins not targetedto the plastid are slightly less constrained than those pro-teins targeted to the plastid.ConclusionOur analyses show that nucleus-encoded plastid-targetedproteins are, on average, more divergent than proteinsencoded in the plastid genome. Although the results can-not explain the reason for this difference, because the pro-teins encoded in both genomes are active in the samecellular compartment, the plastid, we assume that they areunder similar selection pressures and so the difference ismore likely to be attributed to a higher rate of substitutionin the nuclear genome than to differences in selectionpressure. Similarly we confirmed the expectations thatprotein-coding genes encoded and active in the nucleo-morph have accumulated more mutations than nuclear orplastid-encoded proteins but again cannot distinguishbetween rapid mutation rates in the nucleomorphAverage distances of homologues from four taxa of plastid proteins encoded in t e nucleus (blue), nucleomorph (green) and pla tid (red) in (A) G. theta and (B) B. natans f om th cyanobacterium Sy echocystis spFigure 3Average distances of homologues from four taxa of plastid proteins encoded in the nucleus (blue), nucleomorph (green) and plastid (red) in (A) G. theta and (B) B. natans from the cyanobacterium Synechocystis sp. PCC 6803.C.merolaeA.thaliana C. reinhardtii T.pseudonana00.10.20.30.40.50.60.70.80.900.10.20.30.40.50.60.70.80.9C.merolaeA.thaliana C. reinhardtii T.pseudonananucleus-encoded plastid-targetednucleomorph-encoded plastid-targetedplastid-encodedABPage 8 of 12(page number not for citation purposes)is little evidence to suggest that they are becomingorganelles and losing genetic information to the host.genomes as opposed to relaxed selective pressures on pro-teins active within the periplastidal space.BMC Evolutionary Biology 2006, 6:46 http://www.biomedcentral.com/1471-2148/6/46Two more significant results, however, come from thenucleomorph genomes. First, nucleomorph-encodedplastid-proteins reveal differences in the evolution ofcryptomonad and chlorarachniophyte nucleomorphs. InG. theta, the nucleomorph-encoded plastid-proteins areevolving, on average, at about the same rate as nuclear-encoded plastid proteins. In contrast, B. natans nucleo-morph-encoded plastid-targeted proteins are evolvingmuch faster than those encoded in the nucleus, andindeed evolve at about the same rate as other nucleo-morph proteins. Second, the nucleomorphs of two cryp-tomonads are diverging less rapidly than their nucleargenomes. The nucleomorph-encoded proteins active inthe periplastidial space are somewhat more divergentthan plastid-targeted proteins, but still less than nuclearproteins and this may reflect relaxed selection pressure inthis compartment. Together with evidence from Lane et al[62], which shows that cryptomonad nucleomorphgenomes differ in size but have conserved other propertiessuch as gene order, our results suggest that the nucleo-morph genomes of cryptomonad species are not rapidlyevolving and are likely relatively conserved. This is com-parable to other enslaved genomes such as bacterial endo-symbionts and many plastid and mitochondrial genomes.Unfortunately, there is no data from other species of chlo-rarachniophytes with which to make a similar compari-son. From this single species it is difficult to determinewhether the nucleomorph genome is stable or not, but bycomparison to cryptomonads it seems that the nucleo-morph-encoded proteins in B. natans are more weaklyconstrained. It is possible that differences exist betweenthe biology of these two compartments that promote ahigher degree of sequence conservation in one lineagewhat is currently known about nucleomorphs, but furtherinformation from a greater diversity of chlorarachnio-phyte nucleomorphs may resolve whether the nucleo-morph of B. natans is itself evolving rapidly, or whetherthe ancestor of chlorarachniophyte nucleomorphs under-went a rapid burst of sequence evolution subsequent tothe endosymbiotic event that gave rise to the chlorar-achniophyte endosymbiont.MethodsIdentification of plastid-proteinsProteins representing known plastid functions from othereukaryotes and cyanobacteria, were used to search ongo-ing EST projects from the cryptomonads Guillardia theta(CCMP 327) and Rhodomonas salina (CCMP 1319) andalso previously published data from B. natans [16,47],resulting in a set of putative nucleus-encoded plastid-tar-geted protein genes. In the cases of B. natans where severallateral gene transfers have been identified [16], onlynuclear encoded plastid proteins of chlorophyte originwere used. ESTs were completely sequenced on bothstrands from over-lapping cDNA clones for each cluster.New sequences analysed here have been deposited inGenBank under accession numbers DQ383756-DQ383799. Proteins were also identified from the codingsequences of the ongoing sequencing project of the plas-tid genomes of Bigelowiella natans and the plastid genomesof G. theta [46], Arabidopsis thaliana [63], Cyanidioschyzonmerolae [64] and Odontella sinesis [65]. Homologues ofplastid-proteins were identified from the nuclear genomesof Thalassiosira pseudonana [66], A. thaliana [67], C. mero-lae [40]. Proteins sequences were also used from the com-plete genome of the cyanobacterium Synechocystis sp. PCCRadar graphs of average distance of nucleus- and nucleomorph-encoded plastid proteins and nucleomorph-encoded non-plas-tid proteins fr m the two cryptomo ads R. s lina (RS), G. theta (GT) and the free living red algae C. erolae (CM)Figu e 4Radar graphs of average distance of nucleus- and nucleomorph-encoded plastid proteins and nucleomorph-encoded non-plas-tid proteins from the two cryptomonads R. salina (RS), G. theta (GT) and the free living red algae C. merolae (CM).00.20.40.60.8GT:RSGT:CMRS:CM00.40.81.21.62GT:RSGT:CMRS:CM00.40.81.21.62GT:RSGT:CMRS:CMR. salinaG. thetaC. merolaeR. salinaG. thetaC. merolaeR. salinaG. thetaC. merolaenucleomorph-encoded plastid-targetednucleus-encoded plastid-targeted nucleomorph-encoded, non-plastidDayoff VTPage 9 of 12(page number not for citation purposes)than in the other. Just what the underlying causes of suchdifferent rates of evolution may be is not obvious, given6803, and the nucleomorph genomes of G. theta [7] andB. natans (DQ158856 \u00E2\u0080\u0093 DQ158858). When multiple iso-BMC Evolutionary Biology 2006, 6:46 http://www.biomedcentral.com/1471-2148/6/46forms existed in the algal or plant nucleus and it was notobvious which isoform was the orthologue, the distancesfor all isoforms were calculated and the isoforms with theclosest distance to the cryptomonad or chlorarachnio-phyte was used, providing that the same isoforms fromthe algae and plant were also closest to each other. Alter-natively, in a few cases, a neighbour-joining phylogenetictree was constructed to determine groups of isoforms. In aminority of cases for nucleomorph-encoded plastid pro-teins in B. natans where there were multiple paralogues inboth A. thaliana and C. reinhardtii, the nearest Arabidopsisparalogue to B. natans was not nearest to the C. reinhardtiiparalogue closest to B. natans. In these cases the paralogueclosest to B. natans in pair-wise distance (using Dayhoff)was chosen. If it was not possible to determine which iso-form was the likely original paralogue then that proteinwas excluded from the analysis. For analyses with nucleo-morph-encoded non-plastid proteins a subset of proteinsinvolved in transcription, translation (ribosomal subunitsexcepted) and protein folding for which homologuescould be identified in A. thaliana and C. reinhardtii or C.merolae, was used.Identification of R. salina nucleomorph transcriptsProteins encoded in the nucleomorph genome of Guillar-dia theta were used to search a database of Rhodomonassalina (CCMP 1319) ESTs using tBLASTn. The GC contentof the transcripts was calculated and compared to the GCcontent of the G. theta nucleomorph and nuclear genomeand also to R. salina proteins identified as being nuclear-encoded, plastid-targeted. R. salina transcripts with homo-logues in the G. theta nucleomorph with coding regions of28% GC content or less were determined to be nucleo-morph encoded.Calculation of distancesProtein alignments were made using Clustal X [68] andrefined in MacClade (Sinauer Associates, MA. USA). Dis-tances were calculated using PAUP 4.0b10 (Sinauer Asso-ciates, MA. USA) and TREE-PUZZLE 5.2 [69] with eitherthe Dayhoff or VT substitution matrix. For comparisons toG. theta distances were also calculated with the Dayhoffsubstutution matrix and nine rates catagories (eight varia-ble and one invariable), to test for saturation [see Addi-tional file 1].Relative ratesRelative rate tests were performed using the RRTREE pro-gram [55] using C. reinhardtii as an ingroup and A. thal-iana as an outgroup for B. natans datasets. C. merolae wasused as an ingroup and A. thaliana as an outgroup for G.theta datasets. The test was used to compare the evolution-ary rate of individual genes from each of the threemerolae. Since a failure of a relative rate test does not indi-cate which taxon is evolving more rapidly, we comparefailures where G. theta or B. natans is the most rapidlyevolving ingroup.AbbreviationsEST, Expressed Sequence TagAuthors' contributionsAll authors contributed to the experimental concept anddesign, the generation of EST data, data acquisition, anal-ysis, statistics, and manuscript preparation. All authorsread and approved the final manuscript.Additional materialAcknowledgementsWe would like to thank Paul Gilson, Vanessa Mollard, Claudio Slamovits and Geoff Mcfadden for use of the Bigelowiella natans nucleomorph genome. We also thank Ross Waller for helpful commentary on the manuscript. This work was supported by a grant (MOP-42517) from the Canadian Institutes for Health Research (CIHR) and the Protist EST Program of Genome Can-ada/Genome Atlantic. PJK is a Fellow of the Canadian Institute for Advanced Research (CIAR) and a CIHR and MSFHR new investigator.References1. Gray MW, Spencer DF: Organellar Evolution. In Evolution of Micro-bial Life Volume 54. Edited by: Roberts DM, Sharp P, Alderson G andCollins MA. Cambridge, U.K., Cambridge University Press;1996:109-126. 2. Cavalier-Smith T: Genomic reduction and evolution of novelgenetic membranes and protein-targeting machinery ineukaryote-eukaryote chimaeras (meta-algae). Philos Trans RSoc Lond B Biol Sci 2003, 358:109-33; discussion 133-4.3. Archibald JM, Keeling PJ: Recycled plastids: a 'green movement'in eukaryotic evolution. Trends Genet 2002, 18:577-584.4. Martin W, Herrmann RG: Gene transfer from organelles to thenucleus: How much, what happens, and why? Plant Physiology1998, 118:9-17.5. Archibald JM, Keeling PJ: Comparative genomics. Plantgenomes: cyanobacterial genes revealed. Heredity 2003,90:2-3.6. Nozaki H, Matsuzaki M, Misumi O, Kuroiwa H, Hasegawa M, Higash-iyama T, Shin T, Kohara Y, Ogasawara N, Kuroiwa T: Cyanobacte-rial genes transmitted to the nucleus before divergence ofred algae in the chromista. Journal of Molecular Evolution 2004,59:103-113.7. Douglas S, Zauner S, Fraunholz M, Beaton M, Penny S, Deng LT, WuAdditional File 1Supplementary Figure 1. Radar graph of average distances of plastid-, nucleus-, and nucleomorph-encoded plastid proteins, and nucleomorph-encoded non-plastid proteins of the cryptophyte G. theta (GT), the red algae C. merolae (CM) and the plant A. thaliana (AT) calculated with the Dayhoff substitution matrix and nine catagories of gamma corrections (eight variable +1 invariable).Click here for file[http://www.biomedcentral.com/content/supplementary/1471-2148-6-46-S1.EPS]Page 10 of 12(page number not for citation purposes)genomes of B. natans and G. theta to its compartment spe-cific homologue in the genomes of C. reinhardtii and C.X, Reith M, Cavalier-Smith T, Maier UG: The highly reducedgenome of an enslaved algal nucleus. Nature 2001,410:1091-1096.BMC Evolutionary Biology 2006, 6:46 http://www.biomedcentral.com/1471-2148/6/468. Gilson P, Mollard V, Slamovits CH, Reith M, Keeling PJ, McFadden G:Complete nucleotide sequence of the chlorarachniophytenucleomorph. Proc Natl Acad Sci U S A 2006, in press:.9. Gilson PR, McFadden GI: Good things in small packages: the tinygenomes of chlorarachniophyte endosymbionts. Bioessays1997, 19:167-173.10. Gilson PR, McFadden GI: Jam packed genomes--a preliminary,comparative analysis of nucleomorphs. Genetica 2002,115:13-28.11. Gould SV, Sommer MS, Hadfi K, Zauner S, Kroth P, Maier UG: Pro-tein Targeting into the Complex Plastid of Cryptophytes.Journal of Molecular Evolution 2006, in press:.12. Keeling PJ, Archibald JM, Fast NM, Palmer JD: Comment on \"Theevolution of modern eukaryotic phytoplankton\". Science2004, 306:2191; author reply 2191.13. Keeling P: A brief history of plastids and their hosts. Protist2004, 155:3-7.14. Falkowski PG, Katz ME, Knoll AH, Quigg A, Raven JA, Schofield O,Taylor FJ: The evolution of modern eukaryotic phytoplankton.Science 2004, 305:354-360.15. Sulli C, Fang Z, Muchhal U, Schwartzbach SD: Topology of Euglenachloroplast protein precursors within endoplasmic reticu-lum to Golgi to chloroplast transport vesicles. J Biol Chem1999, 274:457-463.16. Archibald JM, Rogers MB, Toop M, Ishida K, Keeling PJ: Lateral genetransfer and the evolution of plastid-targeted proteins in thesecondary plastid-containing alga Bigelowiella natans. Pro-ceedings of the National Academy of Sciences of the United States of Amer-ica 2003, 100:7678-7683.17. Fast NM, Kissinger JC, Roos DS, Keeling PJ: Nuclear-encoded,plastid-targeted genes suggest a single common origin forapicomplexan and dinoflagellate plastids. Molecular Biology andEvolution 2001, 18:418-426.18. Harper JT, Keeling PJ: Nucleus-encoded, plastid-targeted glyc-eraldehyde-3-phosphate dehydrogenase (GAPDH) indicatesa single origin for chromalveolate plastids. Mol Biol Evol 2003,20:1730-1735.19. Patron NJ, Rogers MB, Keeling PJ: Gene replacement of fructose-1,6-bisphosphate aldolase supports the hypothesis of a singlephotosynthetic ancestor of chromalveolates. Eukaryotic Cell2004, 3:1169-1175.20. Cavalier-Smith T: A revised six-kingdom system of life. BiologicalReviews 1998, 73:203-266.21. Katinka MD, Duprat S, Cornillot E, Metenier G, Thomarat F, PrensierG, Barbe V, Peyretaillade E, Brottier P, Wincker P, Delbac F, El AlaouiH, Peyret P, Saurin W, Gouy M, Weissenbach J, Vivares CP: Genomesequence and gene compaction of the eukaryote parasiteEncephalitozoon cuniculi. Nature 2001, 414:450-453.22. Williams BA, Slamovits CH, Patron NJ, Fast NM, Keeling PJ: A highfrequency of overlapping gene expression in compactedeukaryotic genomes. Proc Natl Acad Sci U S A 2005,102:10936-10941.23. Foster J, Ganatra M, Kamal I, Ware J, Makarova K, Ivanova N, Bhatta-charyya A, Kapatral V, Kumar S, Posfai J, Vincze T, Ingram J, Moran L,Lapidus A, Omelchenko M, Kyrpides N, Ghedin E, Wang S, GoltsmanE, Joukov V, Ostrovskaya O, Tsukerman K, Mazur M, Comb D,Koonin E, Slatko B: The Wolbachia genome of Brugia malayi:endosymbiont evolution within a human pathogenic nema-tode. PLoS Biol 2005, 3:e121.24. Pfarr K, Hoerauf A: The annotated genome of Wolbachia fromthe filarial nematode Brugia malayi: what it means forprogress in antifilarial medicine. PLoS Med 2005, 2:e110.25. Genome Sequence of the Intracellular Bacterium Wol-bachia. PLoS Biol 2004, 2:E76.26. Shigenobu S, Watanabe H, Hattori M, Sakaki Y, Ishikawa H: Genomesequence of the endocellular bacterial symbiont of aphidsBuchnera sp. APS. Nature 2000, 407:81-86.27. Akman L, Yamashita A, Watanabe H, Oshima K, Shiba T, Hattori M,Aksoy S: Genome sequence of the endocellular obligate sym-biont of tsetse flies, Wigglesworthia glossinidia. Nat Genet2002, 32:402-407.28. Wernegreen JJ, Lazarus AB, Degnan PH: Small genome of Candi-datus Blochmannia, the bacterial endosymbiont of Cam-ponotus, implies irreversible specialization to an29. Ochman H: Genomes on the shrink. Proceedings of the NationalAcademy of Sciences of the United States of America 2005,102:11959-11960.30. Wernegreen JJ: Endosymbiosis: Lessons in conflict resolution.Plos Biology 2004, 2:307-311.31. Silva FJ, Latorre A, Moya A: Why are the genomes of endosym-biotic bacteria so stable? Trends in Genetics 2003, 19:176-180.32. Schaber J, Rispe C, Wernegreen J, Buness A, Delmotte F, Silva FJ,Moya A: Gene expression levels influence amino acid usageand evolutionary rates in endosymbiotic bacteria. Gene 2005,352:109-117.33. Rispe C, Delmotte F, van Ham R, Moya A: Mutational and selec-tive pressures on codon and amino acid usage in Buchnera,endosymbiotic bacteria of aphids. Genome Research 2004,14:44-53.34. Wernegreen JJ, Moran NA: Evidence for genetic drift in endo-symbionts (Buchnera): Analyses of protein-coding genes.Molecular Biology and Evolution 1999, 16:83-97.35. Birdsell JA: Integrating genomics, bioinformatics, and classicalgenetics to study the effects of recombination on genomeevolution. Molecular Biology and Evolution 2002, 19:1181-1197.36. Wilson RJ, Denny PW, Preiser PR, Rangachari K, Roberts K, Roy A,Whyte A, Strath M, Moore DJ, Moore PW, Williamson DH: Com-plete gene map of the plastid-like DNA of the malaria para-site Plasmodium falciparum. J Mol Biol 1996, 261:155-172.37. Gardner MJ, Hall N, Fung E, White O, Berriman M, Hyman RW, Carl-ton JM, Pain A, Nelson KE, Bowman S, Paulsen IT, James K, Eisen JA,Rutherford K, Salzberg SL, Craig A, Kyes S, Chan MS, Nene V, Shal-lom SJ, Suh B, Peterson J, Angiuoli S, Pertea M, Allen J, Selengut J, HaftD, Mather MW, Vaidya AB, Martin DM, Fairlamb AH, Fraunholz MJ,Roos DS, Ralph SA, McFadden GI, Cummings LM, Subramanian GM,Mungall C, Venter JC, Carucci DJ, Hoffman SL, Newbold C, DavisRW, Fraser CM, Barrell B: Genome sequence of the humanmalaria parasite Plasmodium falciparum. Nature 2002,419:498-511.38. Chanda I, Pan A, Dutta C: Proteome composition in Plasmo-dium falciparum: Higher usage of GC-rich nonsynonymouscodons in highly expressed genes. Journal of Molecular Evolution2005, 61:513-523.39. Peixoto L, Fernandez V, Musto H: The effect of expression levelson codon usage in Plasmodium falciparum. Parasitology 2004,128:245-251.40. Matsuzaki M, Misumi O, Shin-I T, Maruyama S, Takahara M, Miyag-ishima SY, Mori T, Nishida K, Yagisawa F, Yoshida Y, Nishimura Y,Nakao S, Kobayashi T, Momoyama Y, Higashiyama T, Minoda A, SanoM, Nomoto H, Oishi K, Hayashi H, Ohta F, Nishizaka S, Haga S, MiuraS, Morishita T, Kabeya Y, Terasawa K, Suzuki Y, Ishii Y, Asakawa S,Takano H, Ohta N, Kuroiwa H, Tanaka K, Shimizu N, Sugano S, SatoN, Nozaki H, Ogasawara N, Kohara Y, Kuroiwa T: Genomesequence of the ultrasmall unicellular red alga Cyanidio-schyzon merolae 10D. Nature 2004, 428:653-657.41. The Chlamydomonas reinhartdii genome sequencingproject [http://genome.jgi-psf.org/Chlre3/Chlre3.home.html]42. CavalierSmith T, Couch JA, Thorsteinsen KE, Gilson P, Deane JA, HillDRA, McFadden GI: Cryptomonad nuclear and nucleomorph18S rRNA phylogeny. European Journal of Phycology 1996,31:315-328.43. Archibald JM, Cavalier-Smith T, Maier U, Douglas S: Molecularchaperones encoded by a reduced nucleus: the cryptomonadnucleomorph. J Mol Evol 2001, 52:490-501.44. Keeling PJ, Deane JA, Hink-Schauer C, Douglas SE, Maier UG, McFad-den GI: The secondary endosymbiont of the cryptomonadGuillardia theta contains alpha-, beta-, and gamma-tubulingenes. Mol Biol Evol 1999, 16:1308-1313.45. VanDePeer Y, Rensing SA, Maier UG, DeWachter R: Substitutionrate calibration of small subunit ribosomal RNA identifieschlorarachniophyte endosymbionts as remnants of greenalgae. Proceedings of the National Academy of Sciences of the UnitedStates of America 1996, 93:7732-7736.46. Douglas SE, Penny SL: The plastid genome of the cryptophytealga, Guillardia theta: complete sequence and conservedsynteny groups confirm its common ancestry with red algae.J Mol Evol 1999, 48:236-244.47. Rogers MB, Archibald JM, Field MA, Li C, Striepen B, Keeling PJ: Plas-Page 11 of 12(page number not for citation purposes)intracellular lifestyle. Microbiology 2002, 148:2551-2556. tid-targeting peptides from the chlorarachniophyte Bigelow-iella natans. Journal of Eukaryotic Microbiology 2004, 51:529-535.Publish with BioMed Central and every scientist can read your work free of charge\"BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime.\"Sir Paul Nurse, Cancer Research UKYour research papers will be:available free of charge to the entire biomedical communitypeer reviewed and published immediately upon acceptancecited in PubMed and archived on PubMed Central BMC Evolutionary Biology 2006, 6:46 http://www.biomedcentral.com/1471-2148/6/4648. McFadden GI: Plastids and protein targeting. Journal of EukaryoticMicrobiology 1999, 46:339-346.49. Patron NJ, Waller RF, Archibald JM, Keeling PJ: Complex proteintargeting to dinoflagellate plastids. J Mol Biol 2005,348:1015-1024.50. Patron NJ, Keeling PJ: Common evolutionary origin of starchbiosynthetic enzymes in green and red algae. Journal of Phycol-ogy 2005, 41:1131-1141.51. Race HL, Herrmann RG, Martin W: Why have organellesretained genomes? Trends in Genetics 1999, 15:364-370.52. Daley DO, Whelan J: Why genes persist in organelle genomes.Genome Biology 2005, 6:.53. Sarich VM, Wilson AC: Generation time and genomic evolutionin primates. Science 1973, 179:1144-1147.54. Wu CI, Li WH: Evidence for higher rates of nucleotide substi-tution in rodents than in man. Proc Natl Acad Sci U S A 1985,82:1741-1745.55. Robinson-Rechavi M, Huchon D: RRTree: Relative-Rate Testsbetween groups of sequences on a phylogenetic tree. Bioinfor-matics 2000, 16:296-297.56. Herbeck JT, Degnan PH, Wernegreen JJ: Nonhomogeneousmodel of sequence evolution indicates independent originsof primary endosymbionts within the enterobacteriales(gamma-Proteobacteria). Mol Biol Evol 2005, 22:520-532.57. Baumann P, Baumann L, Lai CY, Roubakhsh D, Moran NA, Clark MA:Genetics, Physiology, and Evolutionary Relationships of theGenus Buchnera - Intracellular Symbionts of Aphids. AnnualReview of Microbiology 1995, 49:55-94.58. Ayliffe MA, Scott NS, Timmis JN: Analysis of plastid DNA-likesequences within the nuclear genomes of higher plants.Molecular Biology and Evolution 1998, 15:738-745.59. Noutsos C, Richly E, Leister D: Generation and evolutionary fateof insertions of organelle DNA in the nuclear genomes offlowering plants. Genome Research 2005, 15:616-628.60. Richly E, Leister D: NUPTs in sequenced eukaryotes and theirgenomic organization in relation to NUMTs. Molecular Biologyand Evolution 2004, 21:1972-1980.61. Richly E, Leister D: NUMTs in sequenced eukaryotic genomes.Molecular Biology and Evolution 2004, 21:1081-1084.62. Lane CE, Khan H, Mackinnon M, Fong A, Theophilou S, Archibald JM:Insight into the Diversity and Evolution of the CryptomonadNucleomorph Genome. Mol Biol Evol 2005.63. Sato S, Nakamura Y, Kaneko T, Asamizu E, Tabata S: Completestructure of the chloroplast genome of Arabidopsis thaliana.DNA Res 1999, 6:283-290.64. Ohta N, Matsuzaki M, Misumi O, Miyagishima SY, Nozaki H, TanakaK, Shin IT, Kohara Y, Kuroiwa T: Complete sequence and analy-sis of the plastid genome of the unicellular red alga Cyanidi-oschyzon merolae. DNA Res 2003, 10:67-77.65. Kowallik KV, Stoebe B, Schaffran I, KrothPancic P, Freier U: Thechloroplast genome of a chlorophyll a+c-containing alga,Odontella sinensis. Plant Molecular Biology Reporter 1995,13:336-342.66. Armbrust EV, Berges JA, Bowler C, Green BR, Martinez D, PutnamNH, Zhou SG, Allen AE, Apt KE, Bechner M, Brzezinski MA, ChaalBK, Chiovitti A, Davis AK, Demarest MS, Detter JC, Glavina T, Good-stein D, Hadi MZ, Hellsten U, Hildebrand M, Jenkins BD, Jurka J, Kap-itonov VV, Kroger N, Lau WWY, Lane TW, Larimer FW, LippmeierJC, Lucas S, Medina M, Montsant A, Obornik M, Parker MS, Palenik B,Pazour GJ, Richardson PM, Rynearson TA, Saito MA, Schwartz DC,Thamatrakoln K, Valentin K, Vardi A, Wilkerson FP, Rokhsar DS:The genome of the diatom Thalassiosira pseudonana: Ecol-ogy, evolution, and metabolism. Science 2004, 306:79-86.67. Analysis of the genome sequence of the flowering plant Ara-bidopsis thaliana. Nature 2000, 408:796-815.68. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG: TheCLUSTAL_X windows interface: flexible strategies for mul-tiple sequence alignment aided by quality analysis tools.Nucleic Acids Research 1997, 25:4876-4882.69. Schmidt HA, Hasseler A: Maximum-Likelihood Analysis UsingTREE_PUZZLE. In Current Protocols in Bioinformatics Volume Unit6.6. Edited by: Baxevanis DB, RDM D, G P, G S and L S. New York,Wiley and Sons; 2003. yours \u00E2\u0080\u0094 you keep the copyrightSubmit your manuscript here:http://www.biomedcentral.com/info/publishing_adv.aspBioMedcentralPage 12 of 12(page number not for citation purposes)"@en . "Article"@en . "10.14288/1.0220542"@en . "eng"@en . "Reviewed"@en . "Vancouver : University of British Columbia Library"@en . "BioMed Central"@en . "10.1186/1471-2148-6-46"@en . "Attribution 4.0 International (CC BY 4.0)"@en . "http://creativecommons.org/licenses/by/4.0/"@en . "Faculty"@en . "Comparative rates of evolution in endosymbiotic nuclear genomes"@en . "Text"@en . "http://hdl.handle.net/2429/55257"@en .