- Email: [email protected]
A deviant genetic code in the green alga-derived plastid in the dinoflagellate Lepidodinium chlorophorum
Molecular Phylogenetics and Evolution 60 (2011) 68–72
Contents lists available at ScienceDirect
Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev
A deviant genetic code in the green alga-derived plastid in the dinoflagellate Lepidodinium chlorophorum Takuya Matsumoto a, Sohta A. Ishikawa a, Tetsuo Hashimoto b, Yuji Inagaki b,⇑ a b
Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba, Ibaraki 305-8572, Japan Center for Computational Sciences and Institute for Biological Sciences, University of Tsukuba, Tsukuba, Ibaraki 305-8577, Japan
a r t i c l e
i n f o
Article history: Received 10 October 2010 Revised 6 April 2011 Accepted 6 April 2011 Available online 22 April 2011 Keywords: Genetic code Codon reassignment tRNA Plastid Secondary endosymbiosis Dinoflagellates
a b s t r a c t We here report a deviant genetic code, in which AUA is read as methionine (Met) instead of isoleucine (Ile), in the green alga-derived plastid in the dinoflagellate Lepidodinium chlorophorum. Although L. chlorophorum cDNA sequences of 11 plastid-encoded genes were deposited in the GenBank database, the non-canonical usage of AUA in this dinoflagellate plastid has been overlooked prior to this study. We compared 11 plastid-encoded genes of L. chlorophorum with the corresponding genes of 17 green algal plastids. Intriguingly, AUA often occurred in the L. chlorophorum sequences at codon positions that are predominantly occupied by Met amongst the green algal sequences. Coincidentally, the L. chlorophorum sequences utilized few AUA codons at the positions predominantly occupied by Ile amongst the green algal sequences. These observations clearly indicated that both AUA and AUG encode Met, while AUU and AUC encode Ile, in the L. chlorophorum plastid. Despite the rapidly-evolving nature of L. chlorophorum plastid-encoded genes, our statistical tests incorporating the deviant code suggest no significant difference in amino acid composition among the L. chlorophorum plastid and the green algal plastids considered in this study. Finally, the possible evolutionary events required for the reassignment of AUA from Ile to Met in Lepitodinium plastids were discussed. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction The genetic code is one of the most fundamental rules for the flow of genetic information. Twenty amino acids and a signal for translational termination are specified by the code, comprising 64 types of three consecutive nucleotides (codons), and each codon specifies a single amino acid or termination signal in mRNA. Each tRNA carries three consecutive nucleotides (anticodon) complementary to a codon or codons, and charges a specific amino acid. In the ribosome, mRNA and tRNA interact with each other via codon-anticodon pairing, and tRNAs work as adaptor molecules, which convert a codon sequence in mRNA to the corresponding amino acid sequence (Chapeville et al., 1962; Osawa et al., 1992). The vast majority of living organisms share the same genetic code, and thus this code was most likely established in the last universal common ancestor (henceforth we designate this code as ‘standard’) (Crick, 1968). Nevertheless, codons in the standard code are re-assignable to some extent, as minor alternations in the code (i.e. deviant genetic codes) have been reported in several bacterial and eukaryotic lineages. For instance, AUA, which is read as isoleu⇑ Corresponding author. Address: Center for Computational Sciences, University of Tsukuba, 1-1-1 Tennoudai, Tsukuba, Ibaraki 305-8577, Japan. Fax: +81 29 853 6406. E-mail address: [email protected] (Y. Inagaki). 1055-7903/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2011.04.010
cine (Ile) in the standard code, was re-assigned to methionine (Met) in mammal mitochondria (Barrell et al., 1979). Transition from the standard to this deviant code, the ‘AUA = Met’ code, should have been associated with (i) secondary loss of the tRNA that used to translate AUA as Ile, and (ii) emergence of a new tRNA that recognizes AUA as Met (see review for Osawa et al. (1992)). Mitochondrial genomes have been extensively investigated to date (2228 complete genomes are listed in GenBank Entrez Genome as of August 2010). Mitochondrial sequence data have largely contributed to understanding the evolution of the genetic code, since various types of deviant codes have been identified in mitochondria, particularly those in metazoans (e.g., Yokobori et al., 2001). On the other hand, sequencing of plastid genomes is less advanced than that of mitochondrial genomes (185 complete plastid genomes are listed in GenBank Entrez Genome as of August 2010). Within plastid sequence data, only a single deviant code has been found to date. In the standard code, three codons, UAA, UAG, and UGA, are recognized by a protein factor (release factor) in ribosomes, and promote translation termination (Capecchi and Klein, 1969; Caskey et al., 1969; Caskey, 1980). Meanwhile, the plastids of apicomplexan parasites and their close relatives utilize UGA as tryptophan (Trp) (Lang-Unnasch and Aiello, 1999; Cai et al., 2003; Moore et al., 2008). However, it is unclear whether this ‘UGA = Trp’ code is the sole deviant code that has emerged in the whole of plastid evolution.
T. Matsumoto et al. / Molecular Phylogenetics and Evolution 60 (2011) 68–72
Plastid evolution in the dinoflagellate genus Lepidodinium is unique. Ancestral dinoflagellates most likely possessed red alga-derived plastids containing chlorophylls a and c (Chl-a + c) and the carotenoid peridinin. In sharp contrast, members of the genus Lepidodinium possess green-colored plastids containing chlorophylls a and b (Chl-a + b). The pigment composition (Watanabe et al., 1987), phylogenetic analyses of plastid-encoded genes (Takishita et al., 2008; Matsumoto et al., 2011), and a survey of nuclear genes encoding plastid-targeted proteins in Lepidodinium chlorophorum (Minge et al., 2010) clearly indicated that the ancestral Lepidodinium cells replaced the original plastids containing Chl-a + c with the Chl-a + b-containing plastid of an endosymbiotic green alga. We here report an AUA = Met code in the L. chlorophorum plastid. Our close assessment revealed that L. chlorophorum plastid-encoded genes utilize AUA, which assigns Ile in the standard code, at the codon positions predominantly occupied by Met amongst the 17 green algal plastid sequences considered in this study. As anticipated from the AUA = Met code, the L. chlorophorum plastid rarely uses AUA for the positions predominantly occupied by Ile amongst the green algal sequences. We reassessed the amino acid composition in the L. chlorophorum plastid in light of the AUA = Met code, and noticed that the biased amino acid frequency in the L. chlorophorum sequences was attributed to the misassignment of AUA. Finally, the putative evolution of the genetic code in the L. chlorophorum plastid was discussed by invoking the data from metazoan mitochondria with the same deviant code (Hoffmann et al., 1992; Moriya et al., 1994; Watanabe et al., 1994).
2. Materials and methods 2.1. Codon and amino acid alignments For L. chlorophorum, both cDNA and genomic sequences of 11 plastid-encoded genes (GenBank accession Nos. AB367934AB367941, AB561045, AB589330-AB589340, AB591240 and AB591241): (i) psaA (encoding photosystem I P700 apoprotein A1), (ii) psaB (encoding photosystem I P700 apoprotein A2), (iii) psbA (encoding photosystem II D1 apoprotein), (iv) psbB (encoding photosystem II CP47 apoprotein), (v) psbC (encoding photosystem II CP43 apoprotein), (vi) psbD (encoding photosystem II D2 apoprotein), (vii) atpA (encoding ATP synthase CF1 alpha subunit), (viii) atpB (encoding ATP synthase CF1 beta subunit), (ix) petD (encoding cytochrome b6/f complex subunit IV), (x) rbcL (encoding ribulose1,5-bisphosphate carboxylase/oxygenase large subunit), and (xi) tufA (encoding translation elongation factor Tu). Since the genomic sequence and the corresponding cDNA sequence were identical, we concluded that no RNA editing occurred in any of the L. chlolophorum plastid-encoded genes. We generated 11 plastid-encoded gene alignments including L. chlorophorum and 17 green algae closely related to the endosymbiont that gave rise to the Lepidodinium plastids (Matsumoto et al., 2011). The green algae considered in this study are three ulvophytes (Pseudendoclonium akinetum, Ulva arasakii, and Bryopsis hypnoides), four chlorophytes (Chlamydomonas reinhardtii, Scenedesmus obliquus, Oedogonium cardiacum, and Stigeoclonium helveticum), four trebouxiophytes (Chlorella vulgaris, Parachlorella kessleri, Oocystis solitaria, and Leptosira terrestris), two pedinophytes (strain YPF 701 and Pedinomans minor), three prasinophytes (Tetraselmis spp., Picocystis salinarum, and strain CCMP 1205), and Oltmannsiellopsis viridis (undesignated class). We manually assembled codon alignments of the 11 plastid-encoded genes. After exclusion of all ambiguously aligned codon positions and those including any gaps, the final psaA, psaB, psbA, psbB, psbC, psbD, atpA, atpB, rbcL, tufA, and petD alignments contained 351, 422, 198, 326, 335, 297, 205,
69
277, 339, 296, and 117 codon positions, respectively. Subsequently the codon sequences were converted into amino acid sequences based on the standard genetic code in which AUA assigns Ile (Fig. S1). The 11 amino acid alignments were concatenated into a single alignment (11-proteinSTD alignment). We also made a second 11-protein alignment by applying a deviant genetic code, in which AUA assigns methionine (Met), to the L. chlorophorum plastid-encoded genes (11-proteinDEV alignment). Note that the standard code was applied to the 17 green algal plastid-encoded genes in the second alignment. 2.2. v2 and bootstrap tests for amino acid frequencies The 11-proteinSTD and 11-proteinDEV alignments were subjected to the v2 test assessing whether the overall amino acid frequency in the L. chlorophorum sequence significantly departs from the average amino acid frequencies in the 17 green algal sequences. We also subjected the two alignments to bootstrap tests comparing the frequencies of Met and Ile in the L. chlorophorum sequence to those in the green algal sequences. By excluding phylogenetically redundant green algal taxa in the 11-proteinSTD alignment, we selected seven representative taxa, specifically C. reinhardtii (Chlorophyceae), C. vulgaris (Trebouxiophyceae), P. minor (Pedinophyceae), P. akinetum (Ulvophyceae), O. viridis (undesignated class), Tetraselmis (Prasinophyceae), and P. salinarum (Prasinophyceae). Although not shown here, no significant bias in amino acid frequency was detected in the seven green algal sequences. For each green algal sequence, we generated 10,000 bootstrap replicates and counted the frequencies of Ile and Met in each replicate. Finally, the observed frequency of Ile (or Met) in the L. chlorophorum sequence was compared to the distribution of the frequency of Ile (or Met) calculated from the 10,000 bootstrap replicates. 3. Results and discussion 3.1. AUA assigns Met in the L. chlorophorum plastid In this work, we carefully inspected the codon/amino acid alignments, and found that in the L. chlorophorum plastid-encoded genes, AUA occurred at amino acid positions that are predominantly occupied by Met in the 17 green algal sequences (Fig. 1). The 11-proteinSTD alignment contains 65 amino acid positions where P15 out of the 17 green algal sequences possess Met (henceforth designated as ‘conserved Met positions’). Curiously, L. chlorophorum plastid-encoded genes utilized AUA, as well as the standard Met codon, AUG, at the codon positions corresponding to the conserved Met positions: AUA and AUG occurred in 31 and 30 out of the 65 cases, respectively (Table 1). This trend was observed in all the L. chlorophorum plastid-encoded genes examined here, except psbA (Table 1). In contrast, AUA was found in the L. chlorophorum sequence in only one out of 137 ‘conserved Ile positions,’ where P15 out of the 17 green algal sequences possess Ile (Table 1). These observations strongly suggest that AUA is read as Met, rather than Ile, in the L. chlorophorum plastid. Prior to this study, the standard code had been applied to in silico translation of L. chlorophorum plastid-encoded genes (Takishita et al., 2008; Matsumoto et al., 2011), and a statistical test had suggested that the amino acid composition in the L. chlorophorum plastid proteins was significantly biased (Matsumoto et al., 2011). However, AUA, which is one of the three Ile codons in the standard code, most likely assigns Met in the L. chlorophorum plastid (see above). Therefore, the difference between the amino acid frequency in the L. chlorophorum sequence and those in other green algal sequences may be artifactually magnified by misassignment
70
T. Matsumoto et al. / Molecular Phylogenetics and Evolution 60 (2011) 68–72
Fig. 1. Codon sequences at alignment positions predominantly occupied by methionine (Met). Codon sequences of three out of 64 ‘highly conserved Met’ positions, where P15 out of the 17 green algal sequences possess Met, are shown here. Codons were conceptually translated into amino acids (indicated by one letter code) according to the standard genetic code, except those of the Lepidodinium chlorophorum sequences. Left, the alignment position corresponding to the 371th amino acid residue in Chlamydomonas reinhardtii PsaA (GenBank accession no. NP_958375); Center, the position corresponding to the 549th amino acid residue in C. reinhardtii PsaB (CAA29287); Right, the 297th amino acid residue in C. reinhardtii RbcL (ACJ50136).
of AUA during in silico translation. We ran the v2 tests to examine the null hypothesis, which assumes that the overall amino acid frequency in the L. chlorophorum sequence is equal to the averaged amino acid frequency calculated from the 17 green algal
sequences. For the first test, assuming the standard code (i.e. AUA = Ile) for the L. chlorophorum plastid, the null hypothesis was rejected at the 0.1% level (p = 3.28 10 4). In sharp contrast, the second test, assuming the deviant AUA = Met code for the L. chlorophorum plastid, failed to reject the null hypothesis (p = 0.145). These results indirectly but strongly suggest that the misassignment of AUA (and the resultant incorrect numbers of Ile and Met) was responsible for the result from the first v2 test, as well as the test conducted in our previous study (Matsumoto et al., 2011), which assumed the standard code for the L. chlorophorum plastid. Next, we directly addressed the question whether the L. chlorophorum and green algal sequences contain significantly different frequencies of Met and Ile, by the bootstrap test. This test examined the null hypothesis, which assumes that the frequency of Ile/Met in the L. chlorophorum sequence equals that of a green algal sequence. We repeated the same test seven times, each exchanging a different green algal sequence in turn. When the standard code was applied to the L. chlorophorum plastid, the null hypothesis was rejected at the 1% level in each of the seven tests on the frequency of Ile (‘Standard’ column in Table 2). A similar, but much more extreme, trend was observed in the tests on the frequency of Met. When AUG was treated as a single Met codon, the null hypothesis was rejected with extremely small p values in the seven tests (p < 0.0001; Table 2). In sharp contrast, when the AUA = Met code (AUA and AUG for Met, and AUU and AUC for Ile) was applied to the L. chlorophorum plastid, the null hypothesis was never rejected (p = 0.66–0.98; ‘Deviant’ column in Table 2). These bootstrap tests clearly indicate that, when the deviant code (AUA = Met) was taken into account for the L. chlorophorum plastid, neither the frequency of Ile nor that of Met was significantly different between the L. chlorophorum and green algal sequences.
3.2. Putative evolution of the genetic code in the Lepidodinium plastids The host (dinoflagellate) phylogeny inferred from large subunit ribosomal RNA (LSU rRNA) sequences suggested an extremely close relationship between two species with green-colored plastids belonging to the genus Lepidodinium, L. chlorophorum and Lepidodinium viride (Note that the two rRNA sequences share >90% identity; Watanabe et al., 1987; Elbrächter and Schnepf, 1996; Hansen et al., 2007; Takishita et al., 2008). Based on the host LSU rRNA phylogeny, L. chlorophorum and L. viride are most likely separated from each other very recently (or are the same species), and
Table 1 Codon usages at the amino acid positions, predominantly occupied by methionine (Met) or isoleucine (Ile) in 17 green algal plastid proteins, in 11 Lepididinium chlorophorum plastid-encoded genes. Gene
psaA psaB psbA psbB psbC psbD atpA atpB rbcL tufA petD Total a b c
Codon/amino acid number
351 422 198 326 335 297 205 277 339 296 117 3163
Conserved Met positionsa
9 4 8 4 8 7 3 8 6 7 1 65
L. chlorophorum uses AUA
AUG
Others
6 3 0 3 2 3 2 5 3 3 1 31
3 1 8 1 5 4 1 3 3 1 0 30
0 0 0 0 1 0 0 0 0 3 0 4
Amino acid positions where P15 out of the 17 green algal sequences possess Met. Amino acid positions where P15 out of the 17 green algal sequences possess Ile. AUU or AUC.
Conserved Ile positionsb
16 20 10 12 16 11 12 16 9 10 5 137
L. chlorophorum uses AUA
AUYc
Others
0 0 0 0 0 0 0 1 0 0 0 1
15 18 10 9 13 10 9 15 9 10 5 123
1 2 0 3 3 1 3 0 0 0 0 13
T. Matsumoto et al. / Molecular Phylogenetics and Evolution 60 (2011) 68–72 Table 2 p values for bootstrap tests comparing the number of methionine/isoleucine residues in the Lepididinium chlorophorum sequence and that of green algal sequences. Green alga compared
Chlamydomonas reinhardtii Chlorella vulgaris Oltmannsiellopsis viridis Pseudendoclonium akinetum Picocystis salinarum Tetraselmis spp. Pedinomans minor
Isoleucine (Ile)
Methionine (Met)
Standarda
Deviantb
Standarda
Deviantb
0.0001 0.0002 0.0012 0.0002 0.0003 0.0045 <0.0001
0.6925 0.8518 0.9247 0.8732 0.8386 0.9813 0.6663
<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001
0.8716 0.9846 0.9683 0.9847 0.7969 0.8755 0.8390
a AUU, AUC, and AUA were considered as Ile codons, while AUG was the single Met codon. b AUU and AUC were considered as Ile codons, while AUA and AUG were considered as Met codons.
the green-colored plastid in ‘L. viride’ may also use the deviant AUA = Met code. The origin of the L. chlorophorum plastid has been assessed by phylogenetic analyses of plastid-encoded genes, albeit applying the standard genetic code (i.e. AUA = Ile) to the conceptual translation of L. chlorophorum plastid-encoded genes (Takishita et al., 2008; Matsumoto et al., 2011). We here re-analyzed the data set of 11 plastid proteins used in Matsumoto et al. (2011) by incorporating the AUA = Met code in the L. chlorophorum plastid. The revised 11-protein analysis recovered a robust clade comprising L. chlorophorum, members of core chlorophytes, and the chlorarachniophyte Bigelowiella natans, but failed to pinpoint the position of L. chlorophorum in this assemblage (Fig. S2). Importantly, no sign for the AUA = Met code was found in any genes encoded in the green algal plastid genomes considered in the 11-protein data set (data not shown). We additionally conducted an extensive survey of the AUA = Met code in green algal plastids by analyzing rbcL, one of the most broadly-sampled plastid-encoded genes in public databases. Again, we could not find any green algal plastids with the AUA = Met code (data not shown). These results suggest that the ancestral Lepidodinium cells captured an endosymbiotic green alga whose plastid used the standard code, and the reassignment of AUA codon then took place in the ancestral Lepidodinium plastid. Nevertheless, if the AUA = Met code were found in the plastids of as-yet-unknown core chlorophytes in the future, these algae are the most plausible candidate for the endosymbiont that gave rise to the current Lepidodinium plastids. According to the codon capture theory by Osawa and Jukes (1989), the plastids in Lepidodinium spp. may have undergone a sequence of selectively neutral events to reassign AUA from Ile to Met—disappearance of AUA from the plastid genome (step 1), disappearance of an Ile tRNA (tRNAIle) that read AUA from the translation system (step 2), emergence of a new Met tRNA (tRNAMet) that read AUA (step 3), and re-appearance of AUA as one of Met codons (step 4). In step 1, AUA may have disappeared from the genome by being substituted to synonymous codons AUY (Y = U or C). In organelles with the standard code (including the vast majority of plastids), three Ile codons, AUU, AUC, and AUA, are read by two Ile tRNA species: the tRNAIle with anticodon GAU (tRNAIle GAU ) and the tRNAIle with anticodon LAU (L, lysidylcytidine; tRNAIle LAU ) [see discussion in Turmel et al. (2009)]. The former tRNA can read AUC and AUU codons by the typical wobble-pairing between G at the first anticodon position and U/C at the third codon position (Crick, 1966). Although the codon-specificity of tRNAIle LAU in organelles remains unclear, tRNAIle LAU in Escherichia coli was experimentally shown to translate AUA as Ile (Muramatsu et al., 1988). Thus, we predict that tRNAIle LAU disappeared during the evolution of the genetic code in the Lepidodinium plastids, and AUA became an ‘unassigned’ codon, which is recognized by neither tRNAIle, tRNAMet, nor
71
any other tRNA species (step 2). The elimination of tRNAIle LAU was neutral in the ‘AUA-lacking’ genome that underwent step 1. Then, unassigned AUA may have been captured by a new tRNA that can read AUA as Met (step 3). The reassignment of AUA from Ile to Met has been documented in mitochondria of metazoans, fungi, and heterokont algae (Clark-Walker and Weiller, 1994; Ehara et al., 1997; Yokobori et al., 2001). Following the pioneering works on metazoan mitochondria bearing the AUA = Met code, we propose that the L. chlorophorum plastid possesses either tRNAMet with anticodon UAU (tRNAMet UAU ; U at the first anticodon position is likely modified), or tRNAMet with anticodon f5CAU (f5C, 5-formylcytidine; Met tRNAMet f 5 CAU ) for translation of AUA and AUG. tRNAUAU was found in the mitochondria of the blue mussel Mytilus edulis, and its anticodon likely wobble-pairs both AUA and AUG (Hoffmann et al., 1992). tRNAMet f 5 CAU has been identified in bovine mitochondria and those of the nematode Ascaris suum (Moriya et al., 1994; Watanabe et al., 1994). Wobble pairing between f5C at the first anticodon position and A/G at the third codon position is possible, since in vitro translation experiments confirmed that bovine tRNAMet f 5 CAU translated AUA, as well as AUG, in mRNA (Takemoto et al., 2009). Finally, after step 3, AUA could turn out as a Met codon in the genome as the substitution from AUG to AUA became synonymous (step 4). For our scenarios for the tRNA evolution in Lepidodinium plastids, the gene repertoire for both tRNAIle and tRNAMet species in the complete plastid genome is significant. If our conjecture on tRNAIle is correct, we would then find a single gene for tRNAIle GAU in the L. chlorophorum plastid genome. At the same time, the L. chlorophorum plastid genome likely bears a tRNAMet gene whose anticodon (DNA) sequences are CAT or TAT. Strictly speaking, though, the tRNA gene sequences are not sufficient for a full understanding of genetic code evolution in Lepidodinium plastids, since the modification of U or C at the first anticodon position of tRNAMet is critical for determining its codon recognition capacity. 4. Conclusions We here reported the deviant genetic code, AUA = Met, in the L. chlorophorum plastid. It is widely believed that Lepidodinium spp. replaced the typical dinoflagellate plastid (containing chl-a+c with a carotinoid peridinin) with a green algal plastid (containing chla+b without peridinin), but this unusual plastid history is unlikely to be related to the evolution of the plastid genetic code. Rather, serial mutations that altered the codon recognition capacities of plastid tRNAIle and tRNAMet, coupled with a certain type of codon usage bias, resulted in the AUA = Met code. Although only two types of deviant codes—UGA = Trp and AUA = Met—have been known for plastids so far (Lang-Unnasch and Aiello, 1999; Cai et al., 2003; Moore et al., 2008; this study), additional cases of plastid deviant codes will likely be identified in the future. Acknowledgments We thank R. Kamikawa (University of Tsukuba, Japan) for critical comments on this manuscript. We also thank Aaron Heiss (Dalhousie University, Canada) for proofreading of this manuscript. TM is a research fellow supported by the Japan Society for Promotion of Sciences for Young Scientists (No. 21508). YI and TH are supported by grants from the Japan Society for Promotion of Sciences (Nos. 21370031 and 20570219 were awarded to YI and TH, respectively). Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ympev.2011.04.010.
72
T. Matsumoto et al. / Molecular Phylogenetics and Evolution 60 (2011) 68–72
References Barrell, B.G., Bankier, A.T., Drouin, J., 1979. A different genetic code in human mitochondria. Nature 282, 189–194. Cai, X., Fuller, A.L., McDougald, L.R., Zhu, G., 2003. Apicoplast genome of the coccidian Eimeria tenella. Gene 321, 39–46. Capecchi, M.R., Klein, H.A., 1969. Characterization of three proteins involved in polypeptide chain termination. Cold Spring Harb. Symp. Quant. Biol. 34, 469– 477. Caskey, C.T., 1980. Peptide chain termination. Trends Biochem. Sci. 5, 234–237. Caskey, T., Scolnick, E., Tompkins, R., Goldstein, J., Milman, G., 1969. Peptide chain termination, codon, protein factor, and ribosomal requirements. Cold Spring Harb. Symp. Quant. Biol. 34, 479–488. Chapeville, F., Lipmann, F., Von Ehrenstein, G., Weisblum, B., Ray Jr., W.J., Benzer, S., 1962. On the role of soluble ribonucleic acid in coding for amino acids. Proc. Natl. Acad. Sci. USA 48, 1086–1092. Clark-Walker, G.D., Weiller, G.F., 1994. The structure of the small mitochondrial DNA of Kluyveromyces thermotolerans is likely to reflect the ancestral gene order in fungi. J. Mol. Evol. 38, 593–601. Crick, F.H., 1968. The origin of the genetic code. J. Mol. Biol. 38, 367–379. Crick, F.H., 1966. Codon-anticodon pairing: the wobble hypothesis. J. Mol. Biol. 19, 548–555. Ehara, M., Hayashi-Ishimaru, Y., Inagaki, Y., Ohama, T., 1997. Use of a deviant mitochondrial genetic code in yellow-green algae as a landmark for segregating members within the phylum. J. Mol. Evol. 45, 119–124. Elbrächter, M., Schnepf, E., 1996. Gymnodinium chlorophorum, a new, green, bloomforming dinoflagellate (Gymnodiniales, Dinophyceae) with a vestigial prasinophyte endosymbiont. Phycologia 35, 381–393. Hansen, G., Botes, L., de Salas, M., 2007. Ultrastructure and large subunit rDNA sequences of Lepidodinium viride reveal a close relationship to Lepidodinium chlorophorum comb. nov. (=Gymnodinium chlorophorum). Phycol. Res. 55, 25– 41. Hoffmann, R.J., Boore, J.L., Brown, W.M., 1992. A novel mitochondrial genome organization for the blue mussel, Mytilus edulis. Genetics 131, 397–412. Lang-Unnasch, N., Aiello, D.P., 1999. Sequence evidence for an altered genetic code in the Neospora caninum plastid. Int. J. Parasitol. 29, 1557–1562. Matsumoto, T., Shinozaki, F., Chikuni, T., Yabuki, A., Takishita, K., Kawachi, M., Nakayama, T., Inouye, I., Hashimoto, T., Inagaki, Y., 2011. Green-colored plastids in the dinoflagellate genus Lepidodinium are of core chlorophyte origin. Protist 162, 268–276. Minge, M.A., Shalchian-Tabrizi, K., Torresen, O.K., Takishita, K., Probert, I., Inagaki, Y., Klaveness, D., Jakobsen, K.S., 2010. A phylogenetic mosaic plastid proteome and
unusual plastid-targeting signals in the green-colored dinoflagellate Lepidodinium chlorophorum. BMC Evol. Biol. 10, 191–201. Moore, R.B., Oborník, M., Janouškovec, J., Chrudimsky´, T., Vancová, M., Green, D.H., Wright, S.W., Davies, N.W., Bolch, C.J., Heimann, K., Slapeta, J., Hoegh-Guldberg, O., Logsdon, J.M., Carter, D.A., 2008. A photosynthetic alveolate closely related to apicomplexan parasites. Nature 451, 959–963. Moriya, J., Yokogawa, T., Wakita, K., Ueda, T., Nishikawa, K., Crain, P.F., Hashizume, T., Pomerantz, S.C., McCloskey, J.A., Kawai, G., Hayashi, N., Yokoyama, S., Watanabe, K., 1994. A novel modified nucleoside found at the first position of the anticodon of methionine tRNA from bovine liver mitochondria. Biochemistry 33, 2234–2239. Muramatsu, T., Nishikawa, K., Nemoto, F., Kuchino, Y., Nishimura, S., Miyazawa, T., Yokoyama, S., 1988. Codon and amino-acid specificities of a transfer RNA are both converted by a single post-transcriptional modification. Nature 336, 179– 181. Osawa, S., Jukes, T.H., 1989. Codon reassignment (codon capture) in evolution. J. Mol. Evol. 28, 271–278. Osawa, S., Jukes, T.H., Watanabe, K., Muto, A., 1992. Recent evidence for evolution of the genetic code. Microbiol. Rev. 56, 229–264. Takemoto, C., Spremulli, L.L., Benkowski, L.A., Ueda, T., Yokogawa, T., Watanabe, K., 2009. Unconventional decoding of the AUA codon as methionine by mitochondrial tRNAMet with the anticodon f5CAU as revealed with a mitochondrial in vitro translation system. Nucl. Acids Res. 37, 1616–1627. Takishita, K., Kawachi, M., Noel, M.H., Matsumoto, T., Kakizoe, N., Watanabe, M.M., Inouye, I., Ishida, K., Hashimoto, T., Inagaki, Y., 2008. Origins of plastids and glyceraldehyde-3-phosphate dehydrogenase genes in the green-colored dinoflagellate Lepidodinium chlorophorum. Gene 410, 26–36. Turmel, M., Gagnon, M.C., O’Kelly, C.J., Otis, C., Lemieux, C., 2009. The chloroplast genomes of the green algae Pyramimonas, Monomastix, and Pycnococcus shed new light on the evolutionary history of prasinophytes and the origin of the secondary chloroplasts of euglenids. Mol. Biol. Evol. 26, 631–648. Watanabe, M.M., Takeda, Y., Sasa, T., Inouye, I., Suda, S., Sawaguchi, T., Chihara, M., 1987. A green dinoflagellate with chlorophylls a and b: morphology, fine structure of the chloroplast and chlorophyll composition. J. Phycol. 23, 382– 389. Watanabe, Y., Tsurui, H., Ueda, T., Furushima, R., Takamiya, S., Kita, K., Nishikawa, K., Watanabe, K., 1994. Primary and higher order structures of nematode (Ascaris suum) mitochondrial tRNAs lacking either the T or D stem. J. Biol. Chem. 269, 22902–22906. Yokobori, S., Suzuki, T., Watanabe, K., 2001. Genetic code variations in mitochondria: tRNA as a major determinant of genetic code plasticity. J. Mol. Evol. 53, 314–326.
Contents lists available at ScienceDirect
Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev
A deviant genetic code in the green alga-derived plastid in the dinoflagellate Lepidodinium chlorophorum Takuya Matsumoto a, Sohta A. Ishikawa a, Tetsuo Hashimoto b, Yuji Inagaki b,⇑ a b
Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba, Ibaraki 305-8572, Japan Center for Computational Sciences and Institute for Biological Sciences, University of Tsukuba, Tsukuba, Ibaraki 305-8577, Japan
a r t i c l e
i n f o
Article history: Received 10 October 2010 Revised 6 April 2011 Accepted 6 April 2011 Available online 22 April 2011 Keywords: Genetic code Codon reassignment tRNA Plastid Secondary endosymbiosis Dinoflagellates
a b s t r a c t We here report a deviant genetic code, in which AUA is read as methionine (Met) instead of isoleucine (Ile), in the green alga-derived plastid in the dinoflagellate Lepidodinium chlorophorum. Although L. chlorophorum cDNA sequences of 11 plastid-encoded genes were deposited in the GenBank database, the non-canonical usage of AUA in this dinoflagellate plastid has been overlooked prior to this study. We compared 11 plastid-encoded genes of L. chlorophorum with the corresponding genes of 17 green algal plastids. Intriguingly, AUA often occurred in the L. chlorophorum sequences at codon positions that are predominantly occupied by Met amongst the green algal sequences. Coincidentally, the L. chlorophorum sequences utilized few AUA codons at the positions predominantly occupied by Ile amongst the green algal sequences. These observations clearly indicated that both AUA and AUG encode Met, while AUU and AUC encode Ile, in the L. chlorophorum plastid. Despite the rapidly-evolving nature of L. chlorophorum plastid-encoded genes, our statistical tests incorporating the deviant code suggest no significant difference in amino acid composition among the L. chlorophorum plastid and the green algal plastids considered in this study. Finally, the possible evolutionary events required for the reassignment of AUA from Ile to Met in Lepitodinium plastids were discussed. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction The genetic code is one of the most fundamental rules for the flow of genetic information. Twenty amino acids and a signal for translational termination are specified by the code, comprising 64 types of three consecutive nucleotides (codons), and each codon specifies a single amino acid or termination signal in mRNA. Each tRNA carries three consecutive nucleotides (anticodon) complementary to a codon or codons, and charges a specific amino acid. In the ribosome, mRNA and tRNA interact with each other via codon-anticodon pairing, and tRNAs work as adaptor molecules, which convert a codon sequence in mRNA to the corresponding amino acid sequence (Chapeville et al., 1962; Osawa et al., 1992). The vast majority of living organisms share the same genetic code, and thus this code was most likely established in the last universal common ancestor (henceforth we designate this code as ‘standard’) (Crick, 1968). Nevertheless, codons in the standard code are re-assignable to some extent, as minor alternations in the code (i.e. deviant genetic codes) have been reported in several bacterial and eukaryotic lineages. For instance, AUA, which is read as isoleu⇑ Corresponding author. Address: Center for Computational Sciences, University of Tsukuba, 1-1-1 Tennoudai, Tsukuba, Ibaraki 305-8577, Japan. Fax: +81 29 853 6406. E-mail address: [email protected] (Y. Inagaki). 1055-7903/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2011.04.010
cine (Ile) in the standard code, was re-assigned to methionine (Met) in mammal mitochondria (Barrell et al., 1979). Transition from the standard to this deviant code, the ‘AUA = Met’ code, should have been associated with (i) secondary loss of the tRNA that used to translate AUA as Ile, and (ii) emergence of a new tRNA that recognizes AUA as Met (see review for Osawa et al. (1992)). Mitochondrial genomes have been extensively investigated to date (2228 complete genomes are listed in GenBank Entrez Genome as of August 2010). Mitochondrial sequence data have largely contributed to understanding the evolution of the genetic code, since various types of deviant codes have been identified in mitochondria, particularly those in metazoans (e.g., Yokobori et al., 2001). On the other hand, sequencing of plastid genomes is less advanced than that of mitochondrial genomes (185 complete plastid genomes are listed in GenBank Entrez Genome as of August 2010). Within plastid sequence data, only a single deviant code has been found to date. In the standard code, three codons, UAA, UAG, and UGA, are recognized by a protein factor (release factor) in ribosomes, and promote translation termination (Capecchi and Klein, 1969; Caskey et al., 1969; Caskey, 1980). Meanwhile, the plastids of apicomplexan parasites and their close relatives utilize UGA as tryptophan (Trp) (Lang-Unnasch and Aiello, 1999; Cai et al., 2003; Moore et al., 2008). However, it is unclear whether this ‘UGA = Trp’ code is the sole deviant code that has emerged in the whole of plastid evolution.
T. Matsumoto et al. / Molecular Phylogenetics and Evolution 60 (2011) 68–72
Plastid evolution in the dinoflagellate genus Lepidodinium is unique. Ancestral dinoflagellates most likely possessed red alga-derived plastids containing chlorophylls a and c (Chl-a + c) and the carotenoid peridinin. In sharp contrast, members of the genus Lepidodinium possess green-colored plastids containing chlorophylls a and b (Chl-a + b). The pigment composition (Watanabe et al., 1987), phylogenetic analyses of plastid-encoded genes (Takishita et al., 2008; Matsumoto et al., 2011), and a survey of nuclear genes encoding plastid-targeted proteins in Lepidodinium chlorophorum (Minge et al., 2010) clearly indicated that the ancestral Lepidodinium cells replaced the original plastids containing Chl-a + c with the Chl-a + b-containing plastid of an endosymbiotic green alga. We here report an AUA = Met code in the L. chlorophorum plastid. Our close assessment revealed that L. chlorophorum plastid-encoded genes utilize AUA, which assigns Ile in the standard code, at the codon positions predominantly occupied by Met amongst the 17 green algal plastid sequences considered in this study. As anticipated from the AUA = Met code, the L. chlorophorum plastid rarely uses AUA for the positions predominantly occupied by Ile amongst the green algal sequences. We reassessed the amino acid composition in the L. chlorophorum plastid in light of the AUA = Met code, and noticed that the biased amino acid frequency in the L. chlorophorum sequences was attributed to the misassignment of AUA. Finally, the putative evolution of the genetic code in the L. chlorophorum plastid was discussed by invoking the data from metazoan mitochondria with the same deviant code (Hoffmann et al., 1992; Moriya et al., 1994; Watanabe et al., 1994).
2. Materials and methods 2.1. Codon and amino acid alignments For L. chlorophorum, both cDNA and genomic sequences of 11 plastid-encoded genes (GenBank accession Nos. AB367934AB367941, AB561045, AB589330-AB589340, AB591240 and AB591241): (i) psaA (encoding photosystem I P700 apoprotein A1), (ii) psaB (encoding photosystem I P700 apoprotein A2), (iii) psbA (encoding photosystem II D1 apoprotein), (iv) psbB (encoding photosystem II CP47 apoprotein), (v) psbC (encoding photosystem II CP43 apoprotein), (vi) psbD (encoding photosystem II D2 apoprotein), (vii) atpA (encoding ATP synthase CF1 alpha subunit), (viii) atpB (encoding ATP synthase CF1 beta subunit), (ix) petD (encoding cytochrome b6/f complex subunit IV), (x) rbcL (encoding ribulose1,5-bisphosphate carboxylase/oxygenase large subunit), and (xi) tufA (encoding translation elongation factor Tu). Since the genomic sequence and the corresponding cDNA sequence were identical, we concluded that no RNA editing occurred in any of the L. chlolophorum plastid-encoded genes. We generated 11 plastid-encoded gene alignments including L. chlorophorum and 17 green algae closely related to the endosymbiont that gave rise to the Lepidodinium plastids (Matsumoto et al., 2011). The green algae considered in this study are three ulvophytes (Pseudendoclonium akinetum, Ulva arasakii, and Bryopsis hypnoides), four chlorophytes (Chlamydomonas reinhardtii, Scenedesmus obliquus, Oedogonium cardiacum, and Stigeoclonium helveticum), four trebouxiophytes (Chlorella vulgaris, Parachlorella kessleri, Oocystis solitaria, and Leptosira terrestris), two pedinophytes (strain YPF 701 and Pedinomans minor), three prasinophytes (Tetraselmis spp., Picocystis salinarum, and strain CCMP 1205), and Oltmannsiellopsis viridis (undesignated class). We manually assembled codon alignments of the 11 plastid-encoded genes. After exclusion of all ambiguously aligned codon positions and those including any gaps, the final psaA, psaB, psbA, psbB, psbC, psbD, atpA, atpB, rbcL, tufA, and petD alignments contained 351, 422, 198, 326, 335, 297, 205,
69
277, 339, 296, and 117 codon positions, respectively. Subsequently the codon sequences were converted into amino acid sequences based on the standard genetic code in which AUA assigns Ile (Fig. S1). The 11 amino acid alignments were concatenated into a single alignment (11-proteinSTD alignment). We also made a second 11-protein alignment by applying a deviant genetic code, in which AUA assigns methionine (Met), to the L. chlorophorum plastid-encoded genes (11-proteinDEV alignment). Note that the standard code was applied to the 17 green algal plastid-encoded genes in the second alignment. 2.2. v2 and bootstrap tests for amino acid frequencies The 11-proteinSTD and 11-proteinDEV alignments were subjected to the v2 test assessing whether the overall amino acid frequency in the L. chlorophorum sequence significantly departs from the average amino acid frequencies in the 17 green algal sequences. We also subjected the two alignments to bootstrap tests comparing the frequencies of Met and Ile in the L. chlorophorum sequence to those in the green algal sequences. By excluding phylogenetically redundant green algal taxa in the 11-proteinSTD alignment, we selected seven representative taxa, specifically C. reinhardtii (Chlorophyceae), C. vulgaris (Trebouxiophyceae), P. minor (Pedinophyceae), P. akinetum (Ulvophyceae), O. viridis (undesignated class), Tetraselmis (Prasinophyceae), and P. salinarum (Prasinophyceae). Although not shown here, no significant bias in amino acid frequency was detected in the seven green algal sequences. For each green algal sequence, we generated 10,000 bootstrap replicates and counted the frequencies of Ile and Met in each replicate. Finally, the observed frequency of Ile (or Met) in the L. chlorophorum sequence was compared to the distribution of the frequency of Ile (or Met) calculated from the 10,000 bootstrap replicates. 3. Results and discussion 3.1. AUA assigns Met in the L. chlorophorum plastid In this work, we carefully inspected the codon/amino acid alignments, and found that in the L. chlorophorum plastid-encoded genes, AUA occurred at amino acid positions that are predominantly occupied by Met in the 17 green algal sequences (Fig. 1). The 11-proteinSTD alignment contains 65 amino acid positions where P15 out of the 17 green algal sequences possess Met (henceforth designated as ‘conserved Met positions’). Curiously, L. chlorophorum plastid-encoded genes utilized AUA, as well as the standard Met codon, AUG, at the codon positions corresponding to the conserved Met positions: AUA and AUG occurred in 31 and 30 out of the 65 cases, respectively (Table 1). This trend was observed in all the L. chlorophorum plastid-encoded genes examined here, except psbA (Table 1). In contrast, AUA was found in the L. chlorophorum sequence in only one out of 137 ‘conserved Ile positions,’ where P15 out of the 17 green algal sequences possess Ile (Table 1). These observations strongly suggest that AUA is read as Met, rather than Ile, in the L. chlorophorum plastid. Prior to this study, the standard code had been applied to in silico translation of L. chlorophorum plastid-encoded genes (Takishita et al., 2008; Matsumoto et al., 2011), and a statistical test had suggested that the amino acid composition in the L. chlorophorum plastid proteins was significantly biased (Matsumoto et al., 2011). However, AUA, which is one of the three Ile codons in the standard code, most likely assigns Met in the L. chlorophorum plastid (see above). Therefore, the difference between the amino acid frequency in the L. chlorophorum sequence and those in other green algal sequences may be artifactually magnified by misassignment
70
T. Matsumoto et al. / Molecular Phylogenetics and Evolution 60 (2011) 68–72
Fig. 1. Codon sequences at alignment positions predominantly occupied by methionine (Met). Codon sequences of three out of 64 ‘highly conserved Met’ positions, where P15 out of the 17 green algal sequences possess Met, are shown here. Codons were conceptually translated into amino acids (indicated by one letter code) according to the standard genetic code, except those of the Lepidodinium chlorophorum sequences. Left, the alignment position corresponding to the 371th amino acid residue in Chlamydomonas reinhardtii PsaA (GenBank accession no. NP_958375); Center, the position corresponding to the 549th amino acid residue in C. reinhardtii PsaB (CAA29287); Right, the 297th amino acid residue in C. reinhardtii RbcL (ACJ50136).
of AUA during in silico translation. We ran the v2 tests to examine the null hypothesis, which assumes that the overall amino acid frequency in the L. chlorophorum sequence is equal to the averaged amino acid frequency calculated from the 17 green algal
sequences. For the first test, assuming the standard code (i.e. AUA = Ile) for the L. chlorophorum plastid, the null hypothesis was rejected at the 0.1% level (p = 3.28 10 4). In sharp contrast, the second test, assuming the deviant AUA = Met code for the L. chlorophorum plastid, failed to reject the null hypothesis (p = 0.145). These results indirectly but strongly suggest that the misassignment of AUA (and the resultant incorrect numbers of Ile and Met) was responsible for the result from the first v2 test, as well as the test conducted in our previous study (Matsumoto et al., 2011), which assumed the standard code for the L. chlorophorum plastid. Next, we directly addressed the question whether the L. chlorophorum and green algal sequences contain significantly different frequencies of Met and Ile, by the bootstrap test. This test examined the null hypothesis, which assumes that the frequency of Ile/Met in the L. chlorophorum sequence equals that of a green algal sequence. We repeated the same test seven times, each exchanging a different green algal sequence in turn. When the standard code was applied to the L. chlorophorum plastid, the null hypothesis was rejected at the 1% level in each of the seven tests on the frequency of Ile (‘Standard’ column in Table 2). A similar, but much more extreme, trend was observed in the tests on the frequency of Met. When AUG was treated as a single Met codon, the null hypothesis was rejected with extremely small p values in the seven tests (p < 0.0001; Table 2). In sharp contrast, when the AUA = Met code (AUA and AUG for Met, and AUU and AUC for Ile) was applied to the L. chlorophorum plastid, the null hypothesis was never rejected (p = 0.66–0.98; ‘Deviant’ column in Table 2). These bootstrap tests clearly indicate that, when the deviant code (AUA = Met) was taken into account for the L. chlorophorum plastid, neither the frequency of Ile nor that of Met was significantly different between the L. chlorophorum and green algal sequences.
3.2. Putative evolution of the genetic code in the Lepidodinium plastids The host (dinoflagellate) phylogeny inferred from large subunit ribosomal RNA (LSU rRNA) sequences suggested an extremely close relationship between two species with green-colored plastids belonging to the genus Lepidodinium, L. chlorophorum and Lepidodinium viride (Note that the two rRNA sequences share >90% identity; Watanabe et al., 1987; Elbrächter and Schnepf, 1996; Hansen et al., 2007; Takishita et al., 2008). Based on the host LSU rRNA phylogeny, L. chlorophorum and L. viride are most likely separated from each other very recently (or are the same species), and
Table 1 Codon usages at the amino acid positions, predominantly occupied by methionine (Met) or isoleucine (Ile) in 17 green algal plastid proteins, in 11 Lepididinium chlorophorum plastid-encoded genes. Gene
psaA psaB psbA psbB psbC psbD atpA atpB rbcL tufA petD Total a b c
Codon/amino acid number
351 422 198 326 335 297 205 277 339 296 117 3163
Conserved Met positionsa
9 4 8 4 8 7 3 8 6 7 1 65
L. chlorophorum uses AUA
AUG
Others
6 3 0 3 2 3 2 5 3 3 1 31
3 1 8 1 5 4 1 3 3 1 0 30
0 0 0 0 1 0 0 0 0 3 0 4
Amino acid positions where P15 out of the 17 green algal sequences possess Met. Amino acid positions where P15 out of the 17 green algal sequences possess Ile. AUU or AUC.
Conserved Ile positionsb
16 20 10 12 16 11 12 16 9 10 5 137
L. chlorophorum uses AUA
AUYc
Others
0 0 0 0 0 0 0 1 0 0 0 1
15 18 10 9 13 10 9 15 9 10 5 123
1 2 0 3 3 1 3 0 0 0 0 13
T. Matsumoto et al. / Molecular Phylogenetics and Evolution 60 (2011) 68–72 Table 2 p values for bootstrap tests comparing the number of methionine/isoleucine residues in the Lepididinium chlorophorum sequence and that of green algal sequences. Green alga compared
Chlamydomonas reinhardtii Chlorella vulgaris Oltmannsiellopsis viridis Pseudendoclonium akinetum Picocystis salinarum Tetraselmis spp. Pedinomans minor
Isoleucine (Ile)
Methionine (Met)
Standarda
Deviantb
Standarda
Deviantb
0.0001 0.0002 0.0012 0.0002 0.0003 0.0045 <0.0001
0.6925 0.8518 0.9247 0.8732 0.8386 0.9813 0.6663
<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001
0.8716 0.9846 0.9683 0.9847 0.7969 0.8755 0.8390
a AUU, AUC, and AUA were considered as Ile codons, while AUG was the single Met codon. b AUU and AUC were considered as Ile codons, while AUA and AUG were considered as Met codons.
the green-colored plastid in ‘L. viride’ may also use the deviant AUA = Met code. The origin of the L. chlorophorum plastid has been assessed by phylogenetic analyses of plastid-encoded genes, albeit applying the standard genetic code (i.e. AUA = Ile) to the conceptual translation of L. chlorophorum plastid-encoded genes (Takishita et al., 2008; Matsumoto et al., 2011). We here re-analyzed the data set of 11 plastid proteins used in Matsumoto et al. (2011) by incorporating the AUA = Met code in the L. chlorophorum plastid. The revised 11-protein analysis recovered a robust clade comprising L. chlorophorum, members of core chlorophytes, and the chlorarachniophyte Bigelowiella natans, but failed to pinpoint the position of L. chlorophorum in this assemblage (Fig. S2). Importantly, no sign for the AUA = Met code was found in any genes encoded in the green algal plastid genomes considered in the 11-protein data set (data not shown). We additionally conducted an extensive survey of the AUA = Met code in green algal plastids by analyzing rbcL, one of the most broadly-sampled plastid-encoded genes in public databases. Again, we could not find any green algal plastids with the AUA = Met code (data not shown). These results suggest that the ancestral Lepidodinium cells captured an endosymbiotic green alga whose plastid used the standard code, and the reassignment of AUA codon then took place in the ancestral Lepidodinium plastid. Nevertheless, if the AUA = Met code were found in the plastids of as-yet-unknown core chlorophytes in the future, these algae are the most plausible candidate for the endosymbiont that gave rise to the current Lepidodinium plastids. According to the codon capture theory by Osawa and Jukes (1989), the plastids in Lepidodinium spp. may have undergone a sequence of selectively neutral events to reassign AUA from Ile to Met—disappearance of AUA from the plastid genome (step 1), disappearance of an Ile tRNA (tRNAIle) that read AUA from the translation system (step 2), emergence of a new Met tRNA (tRNAMet) that read AUA (step 3), and re-appearance of AUA as one of Met codons (step 4). In step 1, AUA may have disappeared from the genome by being substituted to synonymous codons AUY (Y = U or C). In organelles with the standard code (including the vast majority of plastids), three Ile codons, AUU, AUC, and AUA, are read by two Ile tRNA species: the tRNAIle with anticodon GAU (tRNAIle GAU ) and the tRNAIle with anticodon LAU (L, lysidylcytidine; tRNAIle LAU ) [see discussion in Turmel et al. (2009)]. The former tRNA can read AUC and AUU codons by the typical wobble-pairing between G at the first anticodon position and U/C at the third codon position (Crick, 1966). Although the codon-specificity of tRNAIle LAU in organelles remains unclear, tRNAIle LAU in Escherichia coli was experimentally shown to translate AUA as Ile (Muramatsu et al., 1988). Thus, we predict that tRNAIle LAU disappeared during the evolution of the genetic code in the Lepidodinium plastids, and AUA became an ‘unassigned’ codon, which is recognized by neither tRNAIle, tRNAMet, nor
71
any other tRNA species (step 2). The elimination of tRNAIle LAU was neutral in the ‘AUA-lacking’ genome that underwent step 1. Then, unassigned AUA may have been captured by a new tRNA that can read AUA as Met (step 3). The reassignment of AUA from Ile to Met has been documented in mitochondria of metazoans, fungi, and heterokont algae (Clark-Walker and Weiller, 1994; Ehara et al., 1997; Yokobori et al., 2001). Following the pioneering works on metazoan mitochondria bearing the AUA = Met code, we propose that the L. chlorophorum plastid possesses either tRNAMet with anticodon UAU (tRNAMet UAU ; U at the first anticodon position is likely modified), or tRNAMet with anticodon f5CAU (f5C, 5-formylcytidine; Met tRNAMet f 5 CAU ) for translation of AUA and AUG. tRNAUAU was found in the mitochondria of the blue mussel Mytilus edulis, and its anticodon likely wobble-pairs both AUA and AUG (Hoffmann et al., 1992). tRNAMet f 5 CAU has been identified in bovine mitochondria and those of the nematode Ascaris suum (Moriya et al., 1994; Watanabe et al., 1994). Wobble pairing between f5C at the first anticodon position and A/G at the third codon position is possible, since in vitro translation experiments confirmed that bovine tRNAMet f 5 CAU translated AUA, as well as AUG, in mRNA (Takemoto et al., 2009). Finally, after step 3, AUA could turn out as a Met codon in the genome as the substitution from AUG to AUA became synonymous (step 4). For our scenarios for the tRNA evolution in Lepidodinium plastids, the gene repertoire for both tRNAIle and tRNAMet species in the complete plastid genome is significant. If our conjecture on tRNAIle is correct, we would then find a single gene for tRNAIle GAU in the L. chlorophorum plastid genome. At the same time, the L. chlorophorum plastid genome likely bears a tRNAMet gene whose anticodon (DNA) sequences are CAT or TAT. Strictly speaking, though, the tRNA gene sequences are not sufficient for a full understanding of genetic code evolution in Lepidodinium plastids, since the modification of U or C at the first anticodon position of tRNAMet is critical for determining its codon recognition capacity. 4. Conclusions We here reported the deviant genetic code, AUA = Met, in the L. chlorophorum plastid. It is widely believed that Lepidodinium spp. replaced the typical dinoflagellate plastid (containing chl-a+c with a carotinoid peridinin) with a green algal plastid (containing chla+b without peridinin), but this unusual plastid history is unlikely to be related to the evolution of the plastid genetic code. Rather, serial mutations that altered the codon recognition capacities of plastid tRNAIle and tRNAMet, coupled with a certain type of codon usage bias, resulted in the AUA = Met code. Although only two types of deviant codes—UGA = Trp and AUA = Met—have been known for plastids so far (Lang-Unnasch and Aiello, 1999; Cai et al., 2003; Moore et al., 2008; this study), additional cases of plastid deviant codes will likely be identified in the future. Acknowledgments We thank R. Kamikawa (University of Tsukuba, Japan) for critical comments on this manuscript. We also thank Aaron Heiss (Dalhousie University, Canada) for proofreading of this manuscript. TM is a research fellow supported by the Japan Society for Promotion of Sciences for Young Scientists (No. 21508). YI and TH are supported by grants from the Japan Society for Promotion of Sciences (Nos. 21370031 and 20570219 were awarded to YI and TH, respectively). Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ympev.2011.04.010.
72
T. Matsumoto et al. / Molecular Phylogenetics and Evolution 60 (2011) 68–72
References Barrell, B.G., Bankier, A.T., Drouin, J., 1979. A different genetic code in human mitochondria. Nature 282, 189–194. Cai, X., Fuller, A.L., McDougald, L.R., Zhu, G., 2003. Apicoplast genome of the coccidian Eimeria tenella. Gene 321, 39–46. Capecchi, M.R., Klein, H.A., 1969. Characterization of three proteins involved in polypeptide chain termination. Cold Spring Harb. Symp. Quant. Biol. 34, 469– 477. Caskey, C.T., 1980. Peptide chain termination. Trends Biochem. Sci. 5, 234–237. Caskey, T., Scolnick, E., Tompkins, R., Goldstein, J., Milman, G., 1969. Peptide chain termination, codon, protein factor, and ribosomal requirements. Cold Spring Harb. Symp. Quant. Biol. 34, 479–488. Chapeville, F., Lipmann, F., Von Ehrenstein, G., Weisblum, B., Ray Jr., W.J., Benzer, S., 1962. On the role of soluble ribonucleic acid in coding for amino acids. Proc. Natl. Acad. Sci. USA 48, 1086–1092. Clark-Walker, G.D., Weiller, G.F., 1994. The structure of the small mitochondrial DNA of Kluyveromyces thermotolerans is likely to reflect the ancestral gene order in fungi. J. Mol. Evol. 38, 593–601. Crick, F.H., 1968. The origin of the genetic code. J. Mol. Biol. 38, 367–379. Crick, F.H., 1966. Codon-anticodon pairing: the wobble hypothesis. J. Mol. Biol. 19, 548–555. Ehara, M., Hayashi-Ishimaru, Y., Inagaki, Y., Ohama, T., 1997. Use of a deviant mitochondrial genetic code in yellow-green algae as a landmark for segregating members within the phylum. J. Mol. Evol. 45, 119–124. Elbrächter, M., Schnepf, E., 1996. Gymnodinium chlorophorum, a new, green, bloomforming dinoflagellate (Gymnodiniales, Dinophyceae) with a vestigial prasinophyte endosymbiont. Phycologia 35, 381–393. Hansen, G., Botes, L., de Salas, M., 2007. Ultrastructure and large subunit rDNA sequences of Lepidodinium viride reveal a close relationship to Lepidodinium chlorophorum comb. nov. (=Gymnodinium chlorophorum). Phycol. Res. 55, 25– 41. Hoffmann, R.J., Boore, J.L., Brown, W.M., 1992. A novel mitochondrial genome organization for the blue mussel, Mytilus edulis. Genetics 131, 397–412. Lang-Unnasch, N., Aiello, D.P., 1999. Sequence evidence for an altered genetic code in the Neospora caninum plastid. Int. J. Parasitol. 29, 1557–1562. Matsumoto, T., Shinozaki, F., Chikuni, T., Yabuki, A., Takishita, K., Kawachi, M., Nakayama, T., Inouye, I., Hashimoto, T., Inagaki, Y., 2011. Green-colored plastids in the dinoflagellate genus Lepidodinium are of core chlorophyte origin. Protist 162, 268–276. Minge, M.A., Shalchian-Tabrizi, K., Torresen, O.K., Takishita, K., Probert, I., Inagaki, Y., Klaveness, D., Jakobsen, K.S., 2010. A phylogenetic mosaic plastid proteome and
unusual plastid-targeting signals in the green-colored dinoflagellate Lepidodinium chlorophorum. BMC Evol. Biol. 10, 191–201. Moore, R.B., Oborník, M., Janouškovec, J., Chrudimsky´, T., Vancová, M., Green, D.H., Wright, S.W., Davies, N.W., Bolch, C.J., Heimann, K., Slapeta, J., Hoegh-Guldberg, O., Logsdon, J.M., Carter, D.A., 2008. A photosynthetic alveolate closely related to apicomplexan parasites. Nature 451, 959–963. Moriya, J., Yokogawa, T., Wakita, K., Ueda, T., Nishikawa, K., Crain, P.F., Hashizume, T., Pomerantz, S.C., McCloskey, J.A., Kawai, G., Hayashi, N., Yokoyama, S., Watanabe, K., 1994. A novel modified nucleoside found at the first position of the anticodon of methionine tRNA from bovine liver mitochondria. Biochemistry 33, 2234–2239. Muramatsu, T., Nishikawa, K., Nemoto, F., Kuchino, Y., Nishimura, S., Miyazawa, T., Yokoyama, S., 1988. Codon and amino-acid specificities of a transfer RNA are both converted by a single post-transcriptional modification. Nature 336, 179– 181. Osawa, S., Jukes, T.H., 1989. Codon reassignment (codon capture) in evolution. J. Mol. Evol. 28, 271–278. Osawa, S., Jukes, T.H., Watanabe, K., Muto, A., 1992. Recent evidence for evolution of the genetic code. Microbiol. Rev. 56, 229–264. Takemoto, C., Spremulli, L.L., Benkowski, L.A., Ueda, T., Yokogawa, T., Watanabe, K., 2009. Unconventional decoding of the AUA codon as methionine by mitochondrial tRNAMet with the anticodon f5CAU as revealed with a mitochondrial in vitro translation system. Nucl. Acids Res. 37, 1616–1627. Takishita, K., Kawachi, M., Noel, M.H., Matsumoto, T., Kakizoe, N., Watanabe, M.M., Inouye, I., Ishida, K., Hashimoto, T., Inagaki, Y., 2008. Origins of plastids and glyceraldehyde-3-phosphate dehydrogenase genes in the green-colored dinoflagellate Lepidodinium chlorophorum. Gene 410, 26–36. Turmel, M., Gagnon, M.C., O’Kelly, C.J., Otis, C., Lemieux, C., 2009. The chloroplast genomes of the green algae Pyramimonas, Monomastix, and Pycnococcus shed new light on the evolutionary history of prasinophytes and the origin of the secondary chloroplasts of euglenids. Mol. Biol. Evol. 26, 631–648. Watanabe, M.M., Takeda, Y., Sasa, T., Inouye, I., Suda, S., Sawaguchi, T., Chihara, M., 1987. A green dinoflagellate with chlorophylls a and b: morphology, fine structure of the chloroplast and chlorophyll composition. J. Phycol. 23, 382– 389. Watanabe, Y., Tsurui, H., Ueda, T., Furushima, R., Takamiya, S., Kita, K., Nishikawa, K., Watanabe, K., 1994. Primary and higher order structures of nematode (Ascaris suum) mitochondrial tRNAs lacking either the T or D stem. J. Biol. Chem. 269, 22902–22906. Yokobori, S., Suzuki, T., Watanabe, K., 2001. Genetic code variations in mitochondria: tRNA as a major determinant of genetic code plasticity. J. Mol. Evol. 53, 314–326.