MBE Advance Access originally published online on December 20, 2005
Molecular Biology and Evolution 2006 23(3):701-712; doi:10.1093/molbev/msj080
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Article |
Comparative Analysis of Bacterial-Origin Genes for Plant Mitochondrial Ribosomal Proteins
Biology Department, University of Ottawa, Ottawa, Canada
E-mail: lbonen{at}science.uottawa.ca.
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResults and DiscussionConcluding RemarksAcknowledgementsReferences |
|---|
Mitochondrial ribosomes contain bacterial-type proteins reflecting their endosymbiotic heritage, and a subset of these genes is retained within the mitochondrion in land plants. Variation in gene location is observed, however, because migration to the nucleus is still an ongoing evolutionary process in plants. To gain insights into adaptation events related to successful gene transfer, we have compiled data for bacterial-origin mitochondrial-type ribosomal protein genes from the completely sequenced Arabidopsis and rice genomes. Approximately 75% of such nuclear-located genes encode amino-terminal extensions relative to their Escherichia coli counterparts, and of that set, only about 30% have introns at (or near) the junction in support of an exon shuffling–type recruitment of upstream expression/targeting signals. We find that genes that were transferred to the nucleus early in eukaryotic evolution have, on average, about twofold higher density of introns within the core ribosomal protein sequences than do those that moved to the nucleus more recently. About 20% of such introns are at positions identical to those in human orthologs, consistent with their ancestral presence. Plant mitochondrial-type ribosomal protein genes have dispersed chromosomal locations in the nucleus, and about 20% of them are present in multiple unlinked copies. This study provides new insights into the evolutionary history of endosymbiotic bacterial-type genes that have been transferred from the mitochondrion to the nucleus.
Key Words: ribosomal protein • mitochondria • introns • evolution • gene transfer • plant • human
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionConcluding RemarksAcknowledgementsReferences |
|---|
Mitochondria have their own distinctive translation systems to direct the synthesis of proteins encoded by the organelle genome, and central to this process is the ribosome. It comprises two subunits (denoted as LSU and SSU) made up of ribosomal RNAs and proteins, and traces of its bacterial endosymbiotic ancestry are evident from sequence comparisons (reviewed in Lang and Gray 1999
). Proteomic analysis of mammalian mitoribosomes (Koc et al. 2001a, 2001b) has established the presence of 78 different proteins (49 LSU and 29 SSU), which include homologs for all but 12 of the 54 proteins in Escherichia coli ribosomes (33 LSU and 21 SSU) (reviewed in T. W. O'Brien, B. J. O'Brien, and Norman 2005). Little is known about the role of the 36 additional novel proteins acquired during eukaryotic evolution, although it has been suggested that some may have taken over certain rRNA functions because animal mitochondrial 16S LSU and 12S SSU rRNAs are much shorter than in bacteria (cf., 23S and 16S rRNAs). In contrast, plant mitochondrial rRNAs are quite long (26S/18S), but as yet the protein composition of mitoribosomes in plants has not been fully elucidated (reviewed in Bonen 2004). It does appear, however, that they too contain about 80 proteins based on early electrophoretic analysis of potato and broad bean mitochondrial ribosomes (Pinel, Douce, and Mache 1986; Maffey, Degand, Boutry 1997), and the number listed in plant mitochondrial proteomic databases continues to grow (cf., Arabidopsis mitochondrial protein database, Heazlewood and Millar 2005).
During eukaryotic evolution, many of the bacterial-type genes, including ribosomal protein ones, have been transferred from the mitochondrion to the nucleus (reviewed in Lang, Gray, and Burger 1999; Timmis et al. 2004). Indeed in animals, all ribosomal protein genes are nuclear located, while in plants and certain protists, some are still in the mitochondrion and in certain cases have retained remnants of ancestral bacterial operon gene order (reviewed in Lang, Gray, and Burger 1999). The mitochondrial genome of the protist Reclinomonas americana, which has been described as resembling a "eubacterial genome in miniature" (Lang et al. 1997), has 27 bacterial-type ribosomal protein genes, and in virtually all other eukaryotes examined to date, their mitochondrial-encoded ones are a subset of the Reclinomonas ones (Gray, Lang, and Burger 2004). Sixteen such ribosomal protein genes are present in the mitochondrion of the bryophyte, Marchantia polymorpha (Takemura et al. 1992), and in flowering plants, the mitochondrial-encoded subset drops to a maximum of 14, namely, S1–S4, S7, S10–S14, S16, L2, L5, and L16 (reviewed in Adams et al. 2002b; Adams and Palmer 2003). Such comparisons have provided insights into the timing of ribosomal protein gene transfer events, and flowering plants are particularly interesting because functional gene transfer to the nucleus is still occurring (Adams et al. 2002b). Plant mitochondrial genes (unlike those in yeast or animals) use the standard genetic code so that present-day transfer is not precluded. However, because their coding regions typically undergo amino acid–altering RNA editing and some genes contain group II–type introns, successful transfer presumably involves processed RNA intermediates. The acquisition of appropriate gene expression signals and targeting information to guide the protein back into the mitochondrion has been documented to be achieved in various ways (reviewed in Adams and Palmer 2003; Timmis et al. 2004), for example, by recruiting duplicated copies of amino-terminal presequences from preexisting mitochondrial-type nuclear genes (cf., S10, Adams et al. 2000) or hitchhiking via alternative splicing when located within the intron of a "host" nuclear gene (cf., S14, Figueroa et al. 1999; Kubo et al. 1999). After a transition stage when potentially functional copies coexist in both the mitochondrion and the nucleus, one copy presumably becomes superfluous and is lost. Indeed, remnant ribosomal protein pseudogenes (or their complete absence) in various plant mitochondrial genomes have been correlated with recently relocated functional nuclear copies. Interestingly, there is as yet just one characterized case of a ribosomal protein gene in a transition stage, namely, L5 in wheat (Sandoval et al. 2004). If a nuclear gene copy is not successfully established after a period of coexistence, the functional mitochondrial copy will be retained, as is the case for S19 in rice in contrast to other cereals where the nuclear copy has "won out" (Fallahi et al. 2005). Furthermore, ribosomal protein genes can have surprisingly complex evolutionary histories. For example, in some plant lineages (such as Arabidopsis), the L2 gene has been fractured into two segments, with one encoded in the nucleus and the other in the mitochondrion (Adams, Ong, and Palmer 2001), and there are intriguing cases of the recapture of alien S2 and S11 ribosomal protein genes after earlier loss from the mitochondrion (Bergthorsson et al. 2003).
Although the absence of an intact ribosomal protein gene from the mitochondrion is usually correlated with its movement to the nucleus, other possibilities include the protein no longer being needed in the ribosome or its function having been taken over by another protein. In plants, obvious candidates for the latter are chloroplast-type nuclear genes, and indeed, the recruitment of a duplicated chloroplast-origin S13 homolog has been experimentally documented for Arabidopsis (Adams et al. 2002a; Mollier et al. 2002). In contrast, there is still a "native" functional S13 gene copy in the rice mitochondrion (Notsu et al. 2002). It should be noted that the chloroplast ribosome contains 58 proteins (33 LSU and 25 SSU) based on spinach proteomic analysis (Yamaguchi and Subramanian 2000; Yamaguchi, von Knoblauch, and Subramanian 2000), of which only 22 are chloroplast encoded. Six of the nuclear-encoded chloroplast ribosomal proteins have been categorized as "plastid specific" (Yamaguchi and Subramanian 2003), and orthologs for bacterial L25 and L30 are absent. A second group of candidates for being recruited for a role in mitoribosomes are cytosol-type ribosomal proteins, and S8, a duplicated cytosol-origin homolog, appears to have replaced the native mitochondrial counterpart (Adams et al. 2002a). The timing of this event is placed early in land plant evolution because a bacterial-type gene is still present in Marchantia mitochondria but has not been observed in the mitochondria of flowering plants.
We are interested in evolutionary events that accompany successful transfer from the mitochondrion to the nucleus, in particular, how gene structure may be influenced as the formerly mitochondrial gene adapts to its new environment. To this end, we have compiled data for the bacterial-origin mitochondrial-type ribosomal protein genes from Arabidopsis thaliana (a eudicot) and Oryza sativa (rice, a monocot), which have been separated from a common ancestor for at least 150 Myr. These plants were selected because complete sequence data are available for the mitochondrion (Unseld et al. 1997; Notsu et al. 2002), chloroplast (Hiratsuka et al. 1989; Sato et al. 1999), and nucleus (AGI 2000; Goff et al. 2002; Yu et al. 2002). In our analysis, we focus on a comparison between genes that were transferred to the nucleus early in eukaryotic evolution with those that migrated more recently.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionConcluding RemarksAcknowledgementsReferences |
|---|
To identify potential plant mitochondrial ribosomal protein (MRP) genes, TBlastN searches were conducted using E. coli ribosomal protein sequences (Blattner et al. 1997
) to query the Arabidopsis and rice nuclear genomes. This query set of 54 proteins did not include E. coli S22 (AAC74553
[GenBank]
), a stationary phase-induced ribosome-associated protein that is restricted to a few bacterial lineages (not including 
-proteobacteria). Note that in the chloroplast nomenclature S22 is a different protein (Yamaguchi, von Knoblauch, and Subramanian 2000). We also conducted TBlastN and BlastP searches using mitochondrial proteins from Marchantia (Takemura et al. 1992) Reclinomonas (Lang et al. 1997), humans (Koc et al. 2001a, 2001b), and yeast (Gan et al. 2002), the latter two sets being nuclear encoded. Proteomic data from spinach chloroplast (Yamaguchi and Subramanian 2000, 2003; Yamaguchi, von Knoblauch, and Subramanian 2000; ) were also useful in distinguishing chloroplast homologs from potential mitochondrial ones. In the case of rice, certain ribosomal protein sequences lacked protein data bank entries, and tBLASTx searches of expressed sequence tag (EST) and nr data banks using Arabidopsis proteins as query enabled the identification of putative rice homologs. Alignments of these ribosomal proteins from Arabidopsis, rice, E. coli, and humans were carried out using ClustalW, with adjustments by visual inspection in certain cases. In this way, the junctions between "core" (ancestral) and "noncore" (acquired) coding regions were also evaluated. Intron positions within the coding regions of the ribosomal protein genes were determined by comparisons of messenger RNA (mRNA) (or EST) data with genomic sequences. In the case of Arabidopsis, annotated information was available at National Center for Biotechnology Information, although some genes were listed as unknown function or were misannotated as putative chloroplast ribosomal proteins. Similarly, using the annotated human genome entries for MRPs, the positions of introns were compared to those in plant homologs.
The potential mitochondrial targeting properties of the Arabidopsis nuclear-encoded mitochondrial-type ribosomal proteins were assessed using the algorithms Predotar v.1.03 http://www.inra.fr/predotar/ (Small et al. 2004), TargetP1.1 http://www.cbs.dtu.dk/services/TargetP/ (Emmanuelsson et al. 2000), and PSort http://psort.nibb.ac.jp/ (Bannai et al. 2002), with default settings and "plant" being selected as the organism group.
|
Results and Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionConcluding RemarksAcknowledgementsReferences |
|---|
Identification of Bacterial-Origin Mitochondrial-Type Ribosomal Protein Genes in Arabidopsis and Rice Genomes
To identify candidate genes for plant MRPs derived from the ancestral
-proteobacterial–type endosymbiont (cf., Esser et al. 2004), we searched the Arabidopsis and rice nuclear genomes for homologs to the 54 E. coli ribosomal proteins. This was accompanied by comparisons with plant chloroplast proteomic data to eliminate any known chloroplast homologs. Similarly, proteomic data for mitochondrial ribosomes from mammals and yeast were useful in corroborating designation of mitochondrial-specific ones. In addition, it is known from the Arabidopsis and rice mitochondrial genomic sequencing (Unseld et al. 1997; Notsu et al. 2002) that 7 and 11 ribosomal proteins, respectively, are still encoded within their mitochondria (fig. 1). These data are summarized in Table 1.
|
|
For the 54 E. coli ribosomal proteins, we identified 48 mitochondrial-type counterparts in rice and 46 in Arabidopsis, the difference being due to the presence of S1 and S13 gene orthologs in the mitochondrial genome of rice but in neither the mitochondrial nor nuclear genomes in Arabidopsis. However, a chloroplast-type S13 protein encoded by a divergent duplicated nuclear gene has been demonstrated to be targeted to the mitochondrion in Arabidopsis (Adams et al. 2002a
; Mollier et al. 2002), and in an analogous fashion, it is possible that one of the several S1 chloroplast-origin nuclear genes (cf., Table 1) may have been recruited for mitochondrial use in the Arabidopsis lineage. Similarly, for S8, there is experimental evidence that the mitochondrial type has been functionally replaced by a divergent cytosol-origin S8 (denoted S15a in cytosol nomenclature) in flowering plants (Adams et al. 2002a). It is possible that the five remaining ones (S20, L25, L31, L34, and L35; fig. 2, gray bars) escaped detection because of sequence divergence or that they are absent from the plant mitoribosome. An alternative possibility for S20, L31, L34, and L35 is that nuclear-encoded chloroplast-type homologs are dual targeted to both compartments. The sole exception is L25, and this protein is also absent from animal/yeast mitoribosomes as well as chloroplast ribosomes.
|
The core bacterial-origin coding regions of these plant mitochondrial-type ribosomal proteins were determined by alignment with E. coli homologs (Supplementary data available upon request) and are shown as black bars in figure 2, whereas noncore extensions are shown in white. In several cases, the core sequences are slightly shorter than the length of the E. coli protein, but in only two cases do bacterial lengths greatly exceed the mitochondrial ones. The E. coli L6 protein has an amino-terminal extension of
80 amino acids relative to the Marchantia and flowering plant mitochondrial counterparts, and the E. coli S1 is approximately 340 amino acids longer than in rice mitochondria, with a long carboxy-terminal extension. Several of the mitochondrial-encoded ribosomal proteins (notably S2, S3, S4, and L2) have extra domains relative to E. coli, and in the case of S2, the mature protein is known to lack the carboxy-terminal extension (Perotta,Grienenberger, and Gualberto 2002).
The plant mitochondrial-type ribosomal proteins were divided into four categories designated as A–D (fig. 1 and Table 1) based on their evolutionary history with respect to timing of gene transfer from the bacterial-type endosymbiont to the nucleus during eukaryotic evolution. This was inferred by the absence of functional ribosomal protein genes in the mitochondrial genome. There are 26 genes placed in category A ("early transfer"), that is, ones that have not been found in the mitochondrial genome of any eukaryote to date, based on the jakobid protists R. americana and Malawimonas jakobiformis of which the former has the largest known set of ribosomal protein genes (Gray, Lang, and Burger 2004). Second, there are 14 ribosomal proteins in category B ("midera transfer") based on genes still present in the Reclinomonas mitochondrion but not in those of land plants, as indicated by the absence in Marchantia (although it is not excluded that there have been certain Marchantia lineage–specific or Reclinomonas lineage–specific losses). Among angiosperms, there are 14 ribosomal protein genes that have been identified, albeit with lineage-specific variation in their number (Adams et al. 2002b), and in our analysis of rice and Arabidopsis, they have been divided into category C ("late transfer") when the gene is nuclear located in one or both of these plants or category D ("native mitochondrial") when the gene is still in the mitochondrion of both these plants. The particularly mobile nature of these genes in flowering plants is illustrated by the presence of 11 in the mitochondrion of rice compared to seven for Arabidopsis, and for several of them, namely, S10, S11, and S14 (Wischmann and Schuster 1995; Kadowaki et al. 1996; Figueroa et al. 1999; Kubo et al. 2000), there is a compelling support for independent transfer events having occurred in these two plant lineages (fig. 1).
Intron Distribution and Density in Nuclear-Located Mitochondrial-Type Ribosomal Protein Genes
Using amino acid sequence data in conjunction with genomic and EST information from Arabidopsis and rice, we have determined the positions of introns within ribosomal protein coding regions (fig. 2), and this is summarized in Table 1. A total of 109 introns were identified within the coding sequences of the genes for 40 of the 47 different ribosomal protein genes that are located in the nucleus in Arabidopsis and/or rice (fig. 2, categories A–C, white triangles) as well as two group II–type introns within mitochondrial genes (fig. 2, category D, gray triangles). Notably seven genes (three in category A and four in category B) contain no introns within the coding regions, although several of these do have an intron within the nontranslated sequences (cf., Table 1, numbers in parentheses). Any intron located within bacterial-origin core sequences (fig. 2, black bars) presumably were acquired after transfer to the nucleus, whereas ones within noncore regions might instead have been obtained at the time of integration in the nuclear genome. For category A (early transfer) genes, the possibility that the endosymbiont had preexisting (ribozymic) introns, which evolved into spliceosomal-type introns in the nucleus, is not excluded (cf., Cavalier-Smith 1991).
With respect to introns within core regions, the Arabidopsis and rice gene homologs show virtually identical profiles for intron positions, the one exception being L9, which has an extra intron in rice (fig. 2, dotted oval). Incidentally, this intron shows no sequence similarity with other (neighboring) introns or transposable elements (cf., Coghlan and Wolfe 2004), and may reflect either recent gain in the rice lineage or loss from the Arabidopsis one. Overall, intron positions appear relatively stable during flowering plant evolution, although variation in intron length is seen between the two plants, with Arabidopsis counterparts typically being shorter (fig. 2, triangle size). For example, Arabidopsis has only one ribosomal protein intron longer than 1 kb (within S6), whereas rice has over 20 introns greater than that length. When considering intron positions within noncore coding sequences for categories A and B, several differences are seen between Arabidopsis and rice (fig. 2, dotted ovals). As expected for category C (recent transfer), there is even greater difference in gene structure between rice and Arabidopsis counterparts, reflecting lineage-specific independent transfer events.
To address the question as to whether transferred genes tend to accumulate introns over evolutionary time, we compared the intron density within the core bacterial-origin regions (i.e., coding sequences that migrated from the mitochondrion rather than being acquired in the nucleus, as is the case for noncore coding sequences) of early-transferred genes (category A) with the more recently transferred gene set (category B). Notably in category C, no introns are located within the core regions. Of the 73 core-type introns, the density was 1.7 introns/100 codons for category A compared to a value of 0.8 for category B (fig. 3). Thus, within the core, the early-transferred genes have about twofold higher intron density than those transferred more recently. In contrast, within the noncore coding regions (which include introns close to the junction), intron density is lower and less biased, showing values of 0.7 and 1.0 introns/100 codons, respectively, for the early and more recently transferred gene sets. Incidentally, for the noncore regions, there is also a higher number of amino-terminal–located introns than carboxy-terminal ones (fig. 2).
|
To extend our investigation of intron gain/loss to a longer evolutionary time period, we compared the plant intron-exon organization with human mitochondrial homologs. Accurate alignment was in some cases difficult because of low amino acid similarity or indels, so we restricted our analysis to those exhibiting strong amino acid conservation immediately flanking the intron position. For this set of 73 introns (omitting genes that have no orthologs in humans), 15 are located at identical positions (fig. 2, black stars; fig. 4A) and an additional seven are located within 6 nt of each other (fig. 2, white stars; fig. 4B). The latter group shows features that have been referred to as "intron sliding," and all but one of them are within the same genes as those having identical intron positions. They represent 11 different ribosomal protein genes, all except one being in category A, consistent with presence in the common ancestor of plants and animals, so dating back over one billion years or so. This set includes the L9 gene that has an extra intron in rice compared to Arabidopsis (fig. 2, dotted oval), and that intron is located 12 nt away from the position of one in human MRPL9 (fig. 4C). Interestingly, in both plants and humans, the L9 gene also has an intron near the amino-terminal core/noncore junction, and the similarity in amino acid sequence (fig. 4C) raises the possibility of a shared history.
|
Typically, the human genes have a greater number and much longer introns than their plant counterparts (cf., one example being MRPS6-intron 1, which is 51.6 kb long). In addition, there are numerous examples of alternatively spliced transcripts for the human ribosomal protein genes but relatively few for plant ones. Interestingly, one of the Arabidopsis duplicated L14 genes (designated as HLP, Skinner et al. 2001
) has CAG triplet repeats at the intron 2/exon 3 junction (fig. 4D, underlined), and alternatively spliced transcripts have been reported. Moreover, a nonsense mutation in the HLL (HUELLENLOS) coding sequence (CAA to TAA, fig. 4D in black) has been found to be correlated with defective ovule growth and development (Skinner et al. 1996). None of the chloroplast-type ribosomal protein genes have introns at positions identical to the mitochondrial-origin ones, with the exception of L21, which originated from a duplicated mitochondrial-type L21 gene (Gallois et al. 2001). The chloroplast-type genes also have fewer introns, perhaps reflecting their shorter evolutionary history.
Acquisition of Mitochondrial Targeting Signals for Nuclear-Encoded Ribosomal Proteins
Because the signals that target nuclear-encoded proteins to the mitochondrion are in many cases located at the amino-terminus, we examined their potential targeting capacity. About 75% of the ribosomal proteins possess amino-terminal extensions relative to their E. coli counterparts, and for the rest, presumably the native sequences possess the required features (either at the amino-terminus or elsewhere along the protein) to enable import into the mitochondrion. The lengths of noncore amino-terminal extensions vary markedly among these ribosomal proteins, and in several cases, they are two- to threefold longer than the core ribosomal protein region (fig. 2, white vs. black bars) so may well encode other important information in addition to targeting signals. The long amino-terminal extension of S5 had been postulated to compensate for the absence of S4 and S8 in human mitoribosomes (Koc et al. 2001a), however, that is brought into question because S4 and S8 counterparts are still present in plant mitoribosomes. In several cases, such as L7/L12, L15, and L32, homology was detected in noncore regions with mitochondrial counterparts from other eukaryotes such as animals and fungi, indicating that these sequences were acquired early in eukaryotic evolution (or perhaps were even present in the ancestral -proteobacteria).
Because the acquisition of noncore sequences could occur through exon shuffling, we scored cases in which introns are located at the core/noncore junction. In some cases, the precise position of the break point in homology between the bacterial and mitochondrial homologs was difficult to map, and we ranked as potentially coincident if it was located within 50 bp of the junction. By these criteria, for 14 of the 50 ribosomal protein gene transfer events (i.e., approximately 30%), there is an intron at the core/noncore amino-terminal junction (fig. 2).
A number of computer algorithms have been developed for predicting the targeting locations of nuclear-encoded proteins, and for plants, there is the added complexity (compared to animals or fungi) of distinguishing between routing to the chloroplast and the mitochondrion. When we subjected our data set to Predotar (Small et al. 2004), TargetP (Emmanuelsson et al. 2000), and PSort (Bannai et al. 2002), we observed that while the majority of the ribosomal proteins showed strong prediction for targeting to the mitochondrion (Table 1, + symbols), not all did, and the results for the three programs differed somewhat in certain cases (cf., discussion in Heazlewood et al. 2004). Among those failing to yield high predictive values for targeting to the mitochondrion are ones for which only chloroplast-origin homologs could be found in the nuclear genome (Table 1, asterisks).
When we compared the amino-terminal sequences of the Arabidopsis and rice ribosomal proteins in categories A and B (i.e., genes transferred prior to divergence of these two plant lineages), we observed a wide variation in the degree of amino acid similarity (Supplementary data available upon request). Certain proteins exhibited high sequence identity within this region (e.g., L29, L14) consistent with a common origin, whereas others (e.g., L15, L10) appeared unrelated. The latter might reflect divergence due to relatively low constraint at the amino acid level, or alternatively, there may have been lineage-specific shuffling events that generated nonhomologous targeting sequences. Indeed, this is further supported in cases where intron position differs between these two plants within the noncore regions (e.g., S18, L19). As sequence data from additional plants become available, it will be of interest to pursue such issues in greater detail.
Duplicated Genes for Mitochondrial-Type Ribosomal Protein Genes and Dispersed Chromosomal Locations in the Arabidopsis Nuclear Genome
The mitochondrial-type ribosomal protein genes are dispersed in the nuclear genome, and their locations on the five Arabidopsis chromosomes are shown in figure 5. The closest two ribosomal protein genes in Arabidopsis are L15 and L17 (about 13 kb apart on chromosome 5); however, in rice they are on chromosomes 1 and 8. On the other hand, the duplicated L27 genes are in tandem in rice, being separated by only 4.4 kb on chromosome 8, whereas in Arabidopsis, they are on separate chromosomes.
|
Among the 48 ribosomal protein genes that are nuclear located in Arabidopsis and/or rice, 11 are present in multiple copies (ranging from two to four) in one or both plants (Table 1). For example, there are single copies of S10 and S11 in the nucleus of Arabidopsis, whereas both are present in duplicate copies in rice. The evolutionary history of the rice S11 duplication event, which has been characterized in detail (Kadowaki et al. 1996
), includes the acquisition of different targeting sequences. Some of the duplicated copies are very similar within the core (e.g., 97%–100% for L17, L19, L22, and L27), whereas others are quite divergent (e.g., 62%–74% for S16, L7/L12, and L14). All have retained the same intron positions within core sequences, although several differ within untranslated regions (cf., L7/L12 copies, Table 1). The duplicated genes include representatives from categories A–C (fig. 2, asterisks) in rather similar proportions (namely, six, three, and two cases out of 25, 14, and 8 genes, respectively), so it does not appear that the early-transferred genes have undergone a greater number of duplication events, at least over this evolutionary time period. |
Concluding Remarks |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionConcluding RemarksAcknowledgementsReferences |
|---|
For the vast majority of the 54 proteins present in
-proteobacterial ribosomes, candidate genes for counterparts in plant mitochondrial ribosomes (as represented by Arabidopsis and rice) can be identified in either the nuclear or mitochondrial genome. The latter comprise a relatively small subset reflecting ancestral endosymbiont-type genes still retained in the organelle, with the others having been transferred to the nucleus over evolutionary time. For the remaining bacterial-origin type, it is possible that sequence divergence precluded their detection; however, there are nuclear-located chloroplast-origin genes that might have been co-opted to substitute for mitochondrial ones, and indeed there is experimental evidence for this having occurred (Adams et al. 2002a; Mollier et al. 2002) as well as for the recruitment of a cytosol-type homolog (Adams et al. 2002a). The sole exception is L25, which notably is absent from other organellar ribosomes, such as chloroplast or mammalian mitochondrial ones based on proteomic analysis (Yamaguchi and Subramanian 2000; Koc et al. 2001b). Thus, plant mitoribosomes appear more bacterial-like than mammalian ones that lack 12 of the 54 proteins found in present-day bacterial ribosomes, including eight (namely, S3, S4, S8, S13, S19, L5, L6, and L29) that have been considered "universal" among all life-forms (Lecompte et al. 2002). Interestingly, of those, all but the L6 and L29 genes are deduced to have been still present within the mitochondrion at the time of the plant-animal lineage split (fig. 1). This illustrates different evolutionary pathways that mitochondrial ribosomes have followed in various eukaryotes, and it is worth noting that as in mammals, the plant mitoribosomes are estimated to contain a total of about 80 proteins due to the acquisition of additional novel ones during eukaryotic evolution. Even though plant MRPs (and rRNAs) have retained stronger bacterial-origin features than in mammals, certain relatively recent events within the land plant lineage illustrate plasticity in the makeup of the ribosome. In addition to native mitochondrial proteins being replaced by chloroplast (or cytosol) homologs, there are reports of lateral transfer of MRP genes among distantly related plant species.
About 75% of the nuclear-encoded plant MRPs have amino-terminal extensions compared to their bacterial orthologs, and some are much longer than typical amino-terminal targeting signals. This raises the possibility that such domains (which also include internal and carboxy-terminal ones) perform extra functions. They might have structural roles (e.g., related to the longer rRNAs in plant mitochondria or ribosomal assembly) or regulatory ones (such as translational control analogous to yeast mitochondria, cf., Williams, Perez-Martinez, and Fox 2004). Alternatively, they may be involved in extraribosomal functions (Wool 1996), such as RNA-level events like editing or splicing. Indeed, RNA-binding motifs have been identified, for example, the RRM domain in Arabidopsis S19 (Sanchez et al. 1996). It will also be of interest to learn more about not only the function of these extra domains but also their origin and evolutionary history.
With respect to the gene structure of these nuclear-located genes, their intron/exon organization is for the most part conserved not only among plants (as exemplified by rice and Arabidopsis) but also in some cases with human ribosomal protein counterparts. About 20% of the introns are at identical positions (or approximately 30% if near-identical locations are included) in the human and plant genes, suggesting that they have been present for at least 1000 Myr. In other studies of intron positions in eukaryotic genes, it has similarly been concluded that about 14% (Fedorov, Merican, and Gilbert 2002) and about 24% (Rogozin et al. 2003) are at identical positions in plant-animal orthologs. There are more introns within core bacterial-origin regions than in acquired coding extensions, and this appears to increase over evolutionary time based on comparisons of intron density in genes transferred early in eukaryotic evolution with those migrating to the nucleus later on. Only about 30% show support for intron-mediated acquisition of upstream sequences (i.e., an exon shuffling model).
Moreover, the complete absence of amino-terminal extensions in a subset of both the early and recent transfer categories raises questions about the nature of targeting information. For very recently transferred genes, it has sometimes been possible to trace the origin of acquired sequences, for example, through the recruitment of duplicated sequences from other mitochondrial-targeted genes (reviewed in Adams and Palmer 2003). In addition, there is growing evidence that targeting of mRNAs for translation on the surface of the mitochondrion may be important for protein import (cf., Sylvestre et al. 2003). Proteomic analysis of plant mitochondrial ribosomes will also undoubtedly yield new insights into the mechanistic and coevolutionary processes that enable the native and "acquired" ribosomal components to perform their essential roles in translation.
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionConcluding RemarksAcknowledgementsReferences |
|---|
We thank O. Antoniuk and G. Voros for their valuable assistance in compiling data from the rice and Arabidopsis nuclear genomes, respectively. Financial support from the Natural Sciences and Engineering Research Council of Canada is also gratefully acknowledged.
|
Footnotes |
|---|
Jennifer Wernegreen, Associate Editor
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionConcluding RemarksAcknowledgementsReferences |
|---|
Adams, K. L., D. O. Daley, Y. L. Qiu, Whelan, J., and J. D. Palmer. 2000. Repeated, recent and diverse transfers of a mitochondrial gene to the nucleus in flowering plants. Nature 408:354–357.[CrossRef][Medline]
Adams, K. L., D. O. Daley, J. Whelan, and J. D. Palmer. 2002a. Genes for two mitochondrial ribosomal proteins in flowering plants are derived from their chloroplast or cytosolic counterparts. Plant Cell 14:931–943.
Adams, K. L., H. C. Ong, and J. D. Palmer. 2001. Mitochondrial gene transfer in pieces: fission of the ribosomal protein gene rpl2 and partial or complete gene transfer to the nucleus. Mol. Biol. Evol. 18:2289–2297.
Adams, K. L., and J. D. Palmer. 2003. Evolution of mitochondrial gene content: gene loss and transfer to the nucleus. Mol. Phylogenet. Evol. 29:380–395.[CrossRef][ISI][Medline]
Adams, K. L., Y. L. Qiu, M. Stoutemyer, and J. D. Palmer. 2002b. Punctuated evolution of mitochondrial gene content: high and variable rates of mitochondrial gene loss and transfer to the nucleus during angiosperm evolution. Proc. Natl. Acad. Sci. USA 99:9905–9912.
[AGI] Arabidopsis Genome Initiative. 2000. Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408:796–815.[CrossRef][Medline]
Bannai, H., Y. Tamada, O. Maruyama, K. Nakai, and S. Miyano. 2002. Extensive feature detection of N-terminal protein sorting signals. Bioinformatics 18:298–305.
Bergthorsson, U., K. L. Adams, B. Thomason, and J. D. Palmer. 2003. Widespread horizontal transfer of mitochondrial genes in flowering plants. Nature 424:197–201.[CrossRef][Medline]
Blattner, F. R., G. Plunkett, C. A. Bloch et al. 1997. The complete genome sequence of Escherichia coli K-12. Science 277:1453–1474.
Bonen, L. 2004. Translational machinery in plant organelles. Pp. 323–345 in H. Daniell and C. Chase, eds. Molecular biology and biotechnology of plant organelles. Springer, the Netherlands.
Cavalier-Smith, T. 1991. Intron phylogeny: a new hypothesis. Trends Genet. 7:145–148.[ISI][Medline]
Coghlan, A., and K. H. Wolfe. 2004. Origins of recently gained introns in Caenorhabditis. Proc. Natl. Acad. Sci. USA 101:11362–11367.
Emmanuelsson, O., H. Nielsen, S. Brunak, and G. von Heijne. 2000. Predicting subcellular localization of proteins based on their N-terminal amino acid sequence. J. Mol. Biol. 300:1005–1016.[CrossRef][ISI][Medline]
Esser, C., N. Ahmadinejad, C. Wiegand et al. 2004. A genome phylogeny for mitochondria among -proteobacteria and a predominantly eubacterial ancestry of yeast nuclear genes. Mol. Biol. Evol. 21:1643–1660.
Fallahi, M., J. Crosthwait, S. Calixte, and L. Bonen. 2005. Fate of mitochondrially located S19 ribosomal protein genes after transfer of functional copy to the nucleus in cereals. Mol. Genet. Genomics 273:76–83.[Medline]
Fedorov, A., A. F. Merican, and W. Gilbert. 2002. Large-scale comparison of intron positions among animal, plant, and fungal genes. Proc. Natl. Acad. Sci. USA 99:16128–16133.
Figueroa, P., I. Gomez, R. Carmona, L. Holuigue, A. Araya, and X. Jordana. 1999a. The gene for mitochondrial ribosomal protein S14 has been transferred to the nucleus in Arabidopsis thaliana. Mol. Gen. Genet. 262:139–144.[CrossRef][ISI][Medline]
Figueroa, P., I. Gomez, L. Holuigue, A. Araya, and X. Jordana. 1999b. Transfer of rps14 from the mitochondrion to the nucleus in maize implied integration within a gene encoding the iron-sulphur subunit of succinate dehydrogenase and expression by alternative splicing. Plant J. 18:601–609.[CrossRef][ISI][Medline]
Gallois, J. L., P. Achard, G. Green, and R. Mache. 2001. The Arabidopsis chloroplast ribosomal L21 is encoded by a nuclear gene of mitochondrial origin. Gene 274:179–185.[CrossRef][ISI][Medline]
Gan, X., M. Kitakawa, K. Yoshino, N. Oshiro, K. Yonezawa, and K. Isono. 2002. Tag-mediated isolation of yeast mitochondrial ribosome and mass spectrometric identification of its new components. Eur. J. Biochem. 269:5203–5214.[ISI][Medline]
Goff, S. A., D. Ricke, T. H. Lan et al. 2002. A draft sequence of the rice genome (Oryza sativa L. ssp. japonica). Science 296:92–100.
Gray, M. W., B. F. Lang, and G. Burger. 2004. Mitochondria of protists. Annu. Rev. Genet. 38:477–524.[CrossRef][ISI][Medline]
Heazlewood, J. L., and A. H. Millar. 2005. AMPDB: the Arabidopsis mitochondrial protein database. Nucleic Acids Res. 13:D605–D610.
Heazlewood, J. L, J. S. Tonti-Filippini, A. M. Gout, D. A. Day, J. Whelan, and A. H. Millar. 2004. Experimental analysis of the Arabidopsis mitochondrial proteome highlights signaling and regulatory components, provides assessment of targeting prediction programs, and indicates plant-specific mitochondrial proteins. Plant Cell 16:241–256.
Hiratsuka, J., H. Shimada, R. Whittier et al. 1989. The complete sequence of the rice (Oryza sativa) chloroplast genome: intermolecular recombination between distinct tRNA genes accounts for a major plastid DNA inversion during the evolution of the cereals. Mol. Gen. Genet. 217:185–194.[CrossRef][ISI][Medline]
International Rice Genome Sequencing Project. 2005. The map-sequence of the rice genome. Nature 436:793–800.[CrossRef][Medline]
Kadowaki, K., N. Kubo, K. Ozawa, and A. Hirai. 1996. Targeting presequence acquisition after mitochondrial gene transfer to the nucleus occurs by duplication of existing targeting signals. EMBO J. 15:6652–6661.[ISI][Medline]
Koc, E. C., W. Burkhardt, K. Blackburn, A. Moseley, and L. L. Spremulli. 2001a. The small subunit of the mammalian mitochondrial ribosome: identification of the full complement of ribosomal proteins present. J. Biol. Chem. 276:19363–19374.
Koc, E. C., W. Burkhart, K. Blackburn, M. B. Moyer, D. M. Schlatzer, A. Moseley, and L. L. Spremulli. 2001b. The large subunit of the mammalian mitochondrial ribosome: analysis of the complement of ribosomal proteins present. J. Biol. Chem. 276:43958–43969.
Kubo, N., K. Harada, A. Hirai, and K. Kadowaki. 1999. A single nuclear transcript encoding mitochondrial RPS14 and SDHB of rice is processed by alternative splicing: common use of the same mitochondrial targeting signal for different proteins. Proc. Natl. Acad. Sci. USA 96:9207–9211.
Kubo, N., X. Jordana, K. Ozawa, S. Zanlungo, K. Harada, T. Sasaki, and K. Kadowaki. 2000. Transfer of the mitochondrial rps10 gene to the nucleus in rice: acquisition of the 5' untranslated region followed by gene duplication. Mol. Gen. Genet. 263:733–739.[CrossRef][ISI][Medline]
Lang, B. F., G. Burger, C. J. O'Kelly, R. Cedergren, G. B. Golding, C. Lemieux, D. Sankoff, M. Turmel, and M. W. Gray. 1997. An ancestral mitochondrial DNA resembling a eubacterial genome in miniature. Nature 387:493–497.[CrossRef][Medline]
Lang, B. F., M. W. Gray, and G. Burger. 1999. Mitochondrial genome evolution and the origin of eukaryotes. Annu. Rev. Genet. 33:351–397.[CrossRef][ISI][Medline]
Lecompte, O., R. Ripp, J. C. Thierry, D. Moras, and O. Poch. 2002. Comparative analysis of ribosomal proteins in complete genomes: an example of reductive evolution at the domain scale. Nucleic Acids Res. 30:5382–5390.
Maffey, L., H. Degand, and M. Boutry. 1997. Partial purification of mitochondrial ribosomes from broad bean and identification of proteins encoded by the mitochondrial genome. Mol. Gen. Genet. 254:365–371.[CrossRef][Medline]
Mollier, P., B. Hoffmann, C. Debast, and I. Small. 2002. The gene encoding Arabidopsis thaliana mitochondrial ribosomal protein S13 is a recent duplication of the gene encoding plastid S13. Curr. Genet. 40:405–409.[CrossRef][Medline]
Notsu, Y., S. Masood, T. Nishikawa, N. Kubo, G. Akiduki, M. Nakazono, A. Hirai, and K. Kadowaki. 2002. The complete sequence of the rice (Oryza sativa L.) mitochondrial genome: frequent DNA sequence acquisition and loss during the evolution of flowering plants. Mol. Genet. Genomics 268:434–445.[CrossRef][ISI][Medline]
O'Brien, T. W., B. J. O'Brien, and R. A. Norman. 2005. Nuclear MRP genes and mitochondrial disease. Gene 354:147–151.[CrossRef][ISI][Medline]
Perrotta, G., J. M. Grienenberger, and J. M. Gualberto. 2002. Plant mitochondrial rps2 genes code for proteins with a C-termimal extension that is processed. Plant Mol. Biol. 50:523–533.[CrossRef][ISI][Medline]
Pinel, C., R. Douce, and R. Mache. 1986. A study of mitochondrial ribosomes from the higher plant Solanum tuberosum L. Mol. Biol. Rep. 11:93–97.[Medline]
Rogozin, I. B., Y. I. Wolf, A. V. Sorokin, B. G. Mirkin, and E. V. Koonin. 2003. Remarkable interkingdom conservation of intron positions and massive, lineage-specific intron loss and gain in eukaryotic evolution. Curr. Biol. 13:1512–1517.[CrossRef][ISI][Medline]
Sanchez, H., T. Fester, S. Kloska, W. Schroder, and W. Schuster. 1996. Transfer of rps19 to the nucleus involves the gain of an RNP-binding motif which may functionally replace RPS13 in Arabidopsis mitochondria. EMBO J. 15:2138–2149.[ISI][Medline]
Sandoval, P., G. Leon, I. Gomez, R. Carmona, P. Figueroa, L. Holuigue, A. Araya, and X. Jordana. 2004. Transfer of RPS14 and RPL5 from the mitochondrion to the nucleus in grasses. Gene 324:139–147.[CrossRef][ISI][Medline]
Sato, S., Y. Nakamure, T. Kaneko, E. Asamizu, and S. Tabata. 1999. Complete structure of the chloroplast genome of Arabidopsis thaliana. DNA Res. 6:283–290.[Abstract]
Skinner, D. J., S. C. Baker, R. J. Meister, J. Broadhvest, K. Schneitz, and C. S. Gasser. 2001. The Arabidopsis HUELLENLOS gene, which is essential for normal ovule development, encodes a mitochondrial ribosomal protein. Plant Cell 13:2719–2730.
Small, I., N. Peeters, F. Legeai, and C. Lurin. 2004. Predotar: a tool for rapidly screening proteomes for N-terminal targeting sequences. Proteomics 4:1581–1590.[CrossRef][ISI][Medline]
Sylvestre, J., A. Margeot, C. Jacq, G. Dujardin, and M. Corral-Debrinski. 2003. The role of the 3' untranslated region in mRNA sorting to the vicinity of mitochondria is conserved from yeast to human cells. Mol. Biol. Cell 14:3848–3856.
Takemura, M., K. Oda, K. Yamato, E. Ohta, Y. Nakamura, N. Nozato, K. Akashi, and K. Ohyama. 1992. Gene clusters for ribosomal proteins in the mitochondrial genome of a liverwort, Marchantia polymorpha. Nucleic Acids Res. 20:3199–3205.
Timmis, J. N., M. A. Ayliffe, C. Y. Huang, and W. Martin. 2004. Endosymbiotic gene transfer: organelle genomes forge eukaryotic chromosomes. Nat. Rev. Genet. 5:123–135.[CrossRef][ISI][Medline]
Unseld, M., J. R. Marienfeld, P. Brandt, and A. Brennicke. 1997. The mitochondrial genome of Arabidopsis thaliana contains 57 genes in 366,924 nucleotides. Nat. Genet. 15:57–61.[CrossRef][ISI][Medline]
Williams, E. H., X. Perez-Martinez, and T. D. Fox. 2004. MrpL36p, a highly diverged L31 ribosomal protein homolog with additional functional domains in Saccharomyces cerevisiae mitochondria. Genetics 167:65–75.
Wischmann, C., and W. Schuster. 1995. Transfer of rps10 from the mitochondrion to the nucleus in Arabidopsis thaliana: evidence for RNA-mediated transfer and exon shuffling at the intergration site. FEBS Lett. 374:152–156.[CrossRef][ISI][Medline]
Wool, I. G. 1996. Extraribosomal functions of ribosomal proteins. Trends Biochem. Sci. 21:164–165.[CrossRef][ISI][Medline]
Yamaguchi, K., and A. R. Subramanian. 2000. The plastid ribosomal proteins: identification of all the proteins in the 50S subunit of an organelle ribosome (chloroplast). J. Biol. Chem. 275:28466–28482.
———. 2003. Proteomic identification of all plastid-specific ribosomal proteins in higher plant chloroplast 30S ribosomal subunit. Eur. J. Biochem. 270:190–205.[ISI][Medline]
Yamaguchi, K., K. von Knoblauch, and A. R. Subramanian. 2000. The plastid ribosomal proteins: identification of all the proteins in the 30S subunit of an organelle ribosome (chloroplast). J. Biol. Chem. 275:28455–28465.
Yu, J., S. Hu, J. Wang et al. 2002. A draft sequence of the rice genome (Oryza sativa L. ssp. indica). Science 296:79–92.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Smits, J. A. M. Smeitink, L. P. van den Heuvel, M. A. Huynen, and T. J. G. Ettema Reconstructing the evolution of the mitochondrial ribosomal proteome Nucleic Acids Res., July 9, 2007; 35(14): 4686 - 4703. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||