MBE Advance Access originally published online on November 23, 2005
Molecular Biology and Evolution 2006 23(5):856-865; doi:10.1093/molbev/msj066
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Proceedings of the SMBE Tri-National Young Investigators' Workshop 2005 |
Insight into the Diversity and Evolution of the Cryptomonad Nucleomorph Genome
Genome Atlantic and the Canadian Institute for Advanced Research, Program in Evolutionary Biology, Department of Biochemistry and Molecular Biology, Dalhousie University, Halifax, Nova Scotia, Canada
E-mail: c.lane{at}dal.ca.
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
The cryptomonads are an enigmatic group of marine and freshwater unicellular algae that acquired their plastids through the engulfment and retention of a eukaryotic ("secondary") endosymbiont. Together with the chlorarachniophyte algae, the cryptomonads are unusual in that they have retained the nucleus of their endosymbiont in a miniaturized form called a nucleomorph. The nucleomorph genome of the cryptomonad Guillardia theta has been completely sequenced and with only three chromosomes and a total size of 551 kb, is a model of nuclear genome compaction. Using this genome as a reference, we have investigated the structure and content of nucleomorph genomes in a wide range of cryptomonad algae. In this study, we have sequenced nine new cryptomonad nucleomorph 18S ribosomal DNA (rDNA) genes and four heat shock protein 90 (hsp90) gene fragments, and using pulsed-field gel electrophoresis and Southern hybridizations, have obtained nucleomorph genome size estimates for nine different species. We also used long-range polymerase chain reaction to obtain nucleomorph genomic fragments from Hanusia phi CCMP325 and Proteomonas sulcata CCMP704 that are syntenic with the subtelomeric region of nucleomorph chromosome I in G. theta. Our results indicate that (1) the presence of three chromosomes is a common feature of the nucleomorph genomes of these organisms, (2) nucleomorph genome size varies dramatically in the cryptomonads examined, (3) unidentified cryptomonad species CCMP1178 has the largest nucleomorph genome identified to date at
845 kb, (4) nucleomorph genome size reductions appear to have occurred multiple times independently during cryptomonad evolution, (5) the relative positions of the 18S rDNA, ubc4, and hsp90 genes are conserved in three different cryptomonad genera, and (6) interchromosomal recombination appears to be rapidly changing the size and sequence of a repetitive subtelomeric region of the nucleomorph genome between the 18S rDNA and ubc4 loci. These results provide a glimpse into the genetic diversity of nucleomorph genomes in cryptomonads and set the stage for more comprehensive sequence-based studies in closely and distantly related taxa.
Key Words: nucleomorph • genome • cryptomonads • ribosomal RNA genes • evolution
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Endosymbiosis has played a significant role in the genetic, biochemical, and ultrastructural diversification of the eukaryotic cell. In the case of photosynthetic eukaryotes, the primary endosymbiosis that gave rise to plastids (chloroplasts) probably occurred more than one billion years ago (Yoon et al. 2004
) and had a huge impact on the composition of the nuclear genome of plants and algae (Martin et al. 2002; Archibald 2005). Three extant eukaryotic lineages, the glaucocystophytes (or glaucophytes), red algae, and green algae (and their land plant relatives) harbor plastids that trace their ancestry directly back to the original cyanobacterial endosymbiont. The plastids of all other photosynthetic eukaryotes were acquired more recently through endosymbiotic mergers between two eukaryotic cells in what is referred to as secondary endosymbiosis. This process has occurred multiple times during eukaryotic evolution and has involved both green and red algal secondary endosymbionts (see Delwiche 1999; Keeling 2004; Archibald and Keeling 2005 for review). The molecular and morphological diversity of secondary plastid-containing algae is immense and includes a significant fraction of the primary producers in the world's oceans such as the haptophyte and dinoflagellate algae, heterokonts (e.g., diatoms and their multicellular relatives the brown algae), cryptomonads, and the chlorarachniophytes (Delwiche 1999; Keeling 2004). Remarkably, several instances of tertiary endosymbiosis have also been documented in the dinoflagellates, where specific genera have replaced their "ancestral" secondary plastid with that of another secondary plastid-containing alga (see Bhattacharya, Yoon, and Hackett 2003; Hackett et al. 2004 for comprehensive review). Unlike primary plastids, which are surrounded by two membranes, secondary plastids are characterized by the presence of three or four membranes surrounding the organelle (Delwiche 1999), a feature that results in a more complex plastid protein targeting pathway in secondary plastid-containing algae (reviewed by McFadden 1999; van Dooren et al. 2001).
The cryptomonad and chlorarachniophyte algae are of particular interest with respect to the origin and evolution of secondary plastids. This is because unlike all other secondary plastid-containing algae, they still retain the nucleus of their eukaryotic endosymbionts in a highly reduced form called a "nucleomorph." Nucleomorphs were first described by Greenwood (1974; Greenwood, Griffiths, and Santore 1977) in cryptomonads as tiny double-membrane–bound bodies nested between the inner and outer pairs of plastid membranes, with nuclear pore-like structures and an electron-dense region resembling a nucleolus (Morrall and Greenwood 1982). Histological studies revealed the presence of DNA within the nucleomorph (Hansmann, Falk, and Sitte 1985; Ludwig and Gibbs 1985), and gene sequencing and immunolocalization studies confirmed that cryptomonad and chlorarachniophyte nucleomorph genomes possess their own small subunit ribosomal (SSU) RNA (18S rRNA) genes distinct from those encoded in the host nucleus (Hansmann 1988; Douglas et al. 1991; Eschbach et al. 1991; Maier et al. 1991; McFadden et al. 1994), protein genes typical of eukaryotic nuclear genomes (Hofmann et al. 1994; Rensing et al. 1994; Gilson and McFadden 1996; Zauner et al. 2000), as well as telomere-like repeats on the ends of their chromosomes (Gilson and McFadden 1995; Gilson and McFadden 1996; Zauner et al. 2000). Maier et al. (1991) and Eschbach et al. (1991) pioneered the use of pulsed-field gel electrophoresis (PFGE) as a tool for investigating the karyotypic structure of nucleomorph genomes. Using this technique, these authors showed that the nucleomorph of the cryptomonad Rhodomonas salina (formerly Pyrenomonas salina) harbors just three chromosomes, 195, 225, and 240 kb in size (Eschbach et al. 1991). More recently, Rensing et al. (1994) showed that the presence of three chromosomes, each with rDNA cistrons, is a widespread feature of cryptomonad nucleomorphs, as is the presence of a nucleomorph-encoded heat shock protein 70 (hsp70) gene.
Genome-scale analyses have confirmed that the chlorarachniophyte and cryptomonad nucleomorphs harbor the smallest and most compact eukaryotic genomes known. The nucleomorph genomes of the cryptomonad Guillardia theta and the chlorarachniophyte Bigelowiella natans have been completely sequenced (Douglas et al. 2001; Gilson and McFadden 2002; G. I. McFadden, personal communication) and are similar in size and structure. Both genomes are comprised of three chromosomes: in G. theta, chromosomes I, II, and III are 196, 181, and 174 kb, respectively, with a total genome size of only 551 kb (Douglas et al. 2001). The B. natans genome is even smaller at 380 kb, with chromosomes of 145, 140, and 98 kb (Gilson and McFadden 2002). Both genomes are extraordinarily gene rich (approximately one gene per kilobase) and A + T biased (75%; Gilson, Maier, and McFadden 1997; Gilson 2001; Gilson and McFadden 2002). Overall, these similarities represent a remarkable instance of convergent evolution because molecular phylogenetic analyses have revealed that the G. theta and B. natans nucleomorphs are of independent origin: while the cryptomonad secondary endosymbiont is derived from a red alga (e.g., Douglas et al. 1991; Van der Auwera et al. 1998; Archibald et al. 2001), the endosymbiont of chlorarachniophytes was most likely a green alga (McFadden, Gilson, and Waller 1995; Ishida et al. 1997; Ishida, Green, and Cavalier-Smith 1999; Archibald et al. 2003). The nucleomorph genomes of both groups appear to have shrunk to approximately the same size and structure from unrelated free-living algal ancestors with presumably dozens of nuclear chromosomes.
While preliminary analyses of nucleomorph genomes in cryptomonad and chlorarachniophyte algae have provided a wealth of information on their basic structure and composition (see Gilson and McFadden 2002 for comprehensive review), very little is known about the diversity of nucleomorphs within the two lineages. In order to study the processes of nucleomorph-to-host-nucleus gene transfer and nucleomorph genome reduction, and to begin to understand why nucleomorphs have been retained in cryptomonads and chlorarachniophytes but have been lost in all other secondary plastid-containing algae, it is necessary to explore the diversity of nucleomorph genomes among different members of both groups. To that end, we are investigating the structure and composition of nucleomorph genomes in diverse cryptomonads using the complete nucleomorph genome of G. theta as a reference (Douglas et al. 2001). Here we present PFGE data indicating that the presence of three nucleomorph chromosomes is a common feature of all cryptomonads investigated thus far and that some species have nucleomorph genomes over 800 kb in size, much larger than that of G. theta. A comparison of nucleomorph genome fragments obtained by long-range polymerase chain reaction (PCR) from Hanusia phi CCMP325 and Proteomonas sulcata CCMP704 with the G. theta genome provides insight into the concerted evolution of cryptomonad nucleomorph chromosome ends.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
DNA Extraction, PCR, Cloning, and Sequencing
Cryptomonad cultures were obtained from public culture collections (table 1) and grown in the laboratory under designated conditions. DNA was extracted from cell pellets using a Tris-HCl digestion buffer (200 mM Tris-HCl (pH 7.5), 250 mM NaCl, 25 mM ethyleneglycol-bis(aminoethylether)-tetraacetic acid, 0.5% sodium dodecyl sulfate) at 50°C for 10 min. After incubation, samples were centrifuged for 5 min at 15,000xg, and the aqueous phase was subjected to two rounds of protein extraction with phenol and chloroform. Nucleic acids were precipitated from the aqueous layer and following centrifugation, pellets were washed and rehydrated in 50 µl of water and stored at –20°C.
|
PCR was used to amplify nucleomorph 18S rDNA, hsp90, ubc4-hsp90, 18S rDNA-ubc4, and 18S rDNA-hsp90 gene fragments. For 18S rDNA amplifications, primers were designed to preferentially amplify the nucleomorph copy of the gene (CrNMSSU.F1—CAGTAGTCATATGCTTGTCTTAAG and CrNMSSU.R1—TGTACAAAGGGCAGGGACGTATTCAGC). The thermal profile for PCR amplification of 18S rDNA included an initial denaturation cycle of 94°C for 5 min followed by 45 cycles of 94°C for 30 s, 50°C for 1 min, and 72°C for 4 min. A final extension step was performed at 72°C for 10 min followed by storage at 10°C until the samples were processed. For hsp90, a degenerate set of primers (hsp90F3—GCACCATTYGAYCTNTTYGANCC and hsp90R3—CATATTAGCACTCCANCCRTAYTCNCC) was designed based on an alignment of diverse eukaryotic hsp90 amino acid sequences. Reactions were performed as above with annealing temperatures of 48–55°C and an extension time of 2 min. Long-range PCR was also used to amplify nucleomorph genome fragments between the 18S rDNA and ubc4 loci in H. phi (hpSSUrRNA.R1—GAGCCATTCGCAGTTTCACGGTAC and hpubc4.R1—GGAGCTGATTGAGAGTAATACC) and between the 18S rDNA and hsp90 genes in P. sulcata (18S.NMrRNA.R2—GAGACATGCATGGCTTAATCC and prsuNMhsp90.R1—GGAAGATCTTCAGAGTCAATTACTCC). Reactions were performed as above with Platinum Taq DNA polymerase High Fidelity (Invitrogen, Carlsbad, Calif.) and an extension time of 4 min.
PCR products were purified using the MinElute Gel Extraction Kit (Catalog number 28604, Qiagen Sciences, Valencia, Calif.) and sequenced directly or cloned using the Topo TA Cloning Kit (Invitrogen, Carlsbad, Calif.) according to the manufacturer's protocol. Three to eight independent clones were chosen per transformation for sequencing using PCR and sequencing primers (SSUni1—CAAGGAAGGCAGCAGGCGCG, SSUni3—TTGATCAAGAACGAAAGT, and SSUni2—CTGGGGGGAGTATGGTCGC). Sequencing reactions were performed using the CEQ Dye Terminator Cycle Sequencing kit (Beckman Coulter, Inc., Fullerton, Calif.) and a Beckman Coulter CEQ8000. DNA sequences determined in this study have been deposited in GenBank under the following accession numbers: DQ228116–DQ228124.
PFGE
Agarose plugs for PFGE were prepared using the protocol of Eschbach et al. (1991) using 2 liters of log-phase culture for a final density of 1.5–2.0 x 108 cells. Plugs were run in a CHEF-DR III Pulsed Field Electrophoresis System (Bio-Rad Laboratories, Hercules, Calif.), on 1% agarose boric ethylenediamine tetraacetic acid (TBE) (1x TBE) gels in 0.5% TBE buffer, at 14.0°C. PFGE was performed using a 60-h run time, a voltage of 4.1 V/cm and a 30–10 s switch time.
Southern Blotting and Hybridization
Gels were blotted onto positively charged nylon membranes (Roche Diagnostics Corp., Indianapolis, Ill.) using the method described in Current Protocols in Molecular Biology (Wiley Interscience, http://www3.interscience.wiley.com/cgi-bin/mrhome/104554809/HOME). An 319-bp fragment of nucleomorph 18S rDNA was PCR amplified (SSUprb1—ATCATTCAAATTTCTGCC and SSUprb2—GCAGTTAAAAAGCTCGTAGTCG) and labeled using the PCR digoxygenin (DIG) Synthesis Kit (Roche Diagnostics Corp.) and used as a probe in Southern hybridizations. Hybridization reactions were performed overnight at 57°C, and membranes were processed using the DIG Luminescent Detection Kit and CDP-Star substrate (Roche Diagnostics Corp.). Probes for nucleomorph ubc4 and hsp90 genes were generated similarly and used in hybridizations at 45–50°C.
Data Analyses
Overlapping DNA fragments were aligned and edited in Sequencher 4.2 (Gene Codes Corporation, Ann Arbor, Mich.). Artemis (The Sanger Institute) was used to search for open reading frames (ORFs) in nucleomorph DNA and to construct G + C content profiles, and codon usage patterns were determined using codonw (http://bioweb.pasteur.fr/seqanal/interfaces/codonw.html). DNA fragments were searched for the presence of tRNA genes using tRNAScan (http://www.genetics.wustl.edu/eddy/tRNAscan-SE/).
Multiple sequence alignments of DNA and protein sequences were constructed and edited with MacClade 4.06 (W. Maddison and D. Maddison 2003). Ambiguously aligned data were removed from alignments prior to phylogenetic analysis. The nucleomorph 18S rDNA sequence alignment (1428 nucleotide positions and 43 taxa) was subjected to maximum likelihood (ML), neighbor joining (NJ), and bootstrap analyses in PAUP* v 4.0b10 (Swofford 2002), whereas Bayesian analyses were performed with MrBayes v 3.0 (Huelsenbeck and Ronquist 2003). Model parameters used in ML and NJ analyses were estimated in Modeltest (Posada and Crandall 1998). Ten random addition replicates under the heuristic search method, using Tree Bisection-Reconnection branch swapping, were completed under ML analysis. Bootstrap analyses (1,000 replicates) were calculated under NJ. Three independent runs of one million generations and one of four million generations were run in MrBayes using default values and the general time reversible model (GTR + I + 
) with parameters estimated during the analyses. Trees were sampled every 100 generations and "burn-in" values around 160,000 generations were determined by visual inspection of likelihood values. The first 400,000 generations were discarded to ensure stabilization, and the remaining trees were used to construct the consensus.
Amino acid sequences inferred from partial cryptomonad nucleomorph hsp90 genes were added to a preexisting alignment of eukaryotic hsp90 proteins. Phylogenetic trees were inferred from -corrected distance matrices (calculated using Tree-Puzzle 5.0 [http://www.tree-puzzle.de/] with eight rate categories and an invariable rate category) using Fitch-Margoliash in PHYLIP 3.6 (http://evolution.genetics.washington.edu/phylip.html) and weighted neighbor joining (Bruno, Socci, and Halpern 2000).
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Nucleomorph Genome Size Diversity
We used a combination of PFGE and Southern hybridizations to explore the diversity of nucleomorph genomes in a wide range of cryptomonad species. Preliminary experiments were performed on 15 different strains, nine of which proved amenable to chromosome preparation and PFGE analysis. The visibility of nucleomorph chromosomes under ethidium bromide staining varied from strain to strain, but Southern blots with a heterologous G. theta nucleomorph 18S rDNA probe produced clear hybridization signals for three chromosomes in each of the nine organisms (significant cross-hybridization with higher molecular weight host nuclear chromosomes also occurred; fig. 1). The known sizes of the yeast and G. theta nucleomorph chromosomes were used to construct a standard logarithmic curve against which the migration distances of these chromosomes were compared (data not shown), and sizes were interpolated and rounded to the nearest 5 kb. The relative nucleomorph chromosome sizes were consistent across multiple independent PFGE runs, although should be considered rough estimates because variations in the total amount of DNA per lane can influence the chromosome migration rate.
|
Overall, cryptomonad nucleomorph genome size was found to vary significantly in the different strains tested, between
570 and 845 kb (fig. 1). The nucleomorph genome of P. sulcata CCMP704 is most similar in size to that of G. theta (551 kb) and at 570 kb is significantly smaller than any of the other strains examined. Within strains classified as members of the genus Rhodomonas, nucleomorph genome size is quite variable, between 650 and 755 kb. Despite this variation, a common feature in these organisms is the presence of two large chromosomes that are close in size and a significantly smaller one. At 845 kb, the nucleomorph genome of the unidentified cryptomonad species 1178 is the largest identified to date. Our study reveals that members of the distantly related genus Chroomonas also possess large nucleomorph genomes, with three similarly sized chromosomes totaling over 800 kb (fig. 1).
Cryptomonad 18S rDNA Phylogeny
With the goal of placing the nucleomorph karyotype data in a phylogenetic context, we sequenced the nucleomorph 18S rDNA gene from each of the cryptomonad strains examined above, with the exception of R. salina CCMP1420 and P. sulcata CCMP704 (GenBank accession number AJ420692). The 18S rDNA genes were also sequenced from Rhodomonas baltica and unidentified cryptomonad CCMP2045, two species for which we were unable to obtain nucleomorph karyotype data. These genes were added to a comprehensive alignment of publicly available 18S rDNA sequences, and a variety of rooted and unrooted phylogenetic analyses were performed. In rooted analyses, a set of eight bangiophyte sequences were used as an outgroup to the cryptomonad nucleomorph genes, and the position of the root was found to lie between the Rhodomonas clade and all other cryptomonads (data not shown). While this topology is well supported and consistent with the nuclear 18S phylogenies of Deane et al. (2002), it is at odds with other studies, which place either the genus Cryptomonas or the Chroomonas/Hemiselmis clade at the base of the cryptomonad tree (Cavalier-Smith et al. 1996; Marin, Klingberg, and Melkonian 1998; Hoef-Emden, Marin, and Melkonian 2002). In the interest of obtaining better resolution among the ingroup taxa, the outgroup was removed and the analyses were repeated using an alignment containing more sites. A Bayesian phylogeny (identical topology to ML tree) is shown in figure 2, arbitrarily rooted between the Rhodomonas clade and the remaining taxa. The topology is largely congruent with the phylogenies presented in two recent comprehensive studies using nuclear 18S (Deane et al. 2002; Hoef-Emden, Marin, and Melkonian 2002) and nucleomorph 18S data (Hoef-Emden, Marin, and Melkonian 2002). This includes the resolution of a monophyletic Rhodomonas clade (which also includes Storeatula major and Rhinomonas pauca) and strongly-supported Cryptomonas and Chroomonas/Hemiselmis clades.
|
The 18S phylogeny allowed us to place formerly unidentified taxa in a generic context (unidentified cryptomonad species CCMP1178 and CCMP2045 both branched within the Rhodomonas clade) as well as interpret the PFGE data from an evolutionary perspective. For comparison, the nucleomorph genome size estimates shown in figure 1 and those taken from previous studies (Rensing et al. 1994
; Deane et al. 1998) are mapped onto the 18S rDNA phylogeny shown in figure 2. The size variations within the well-supported Rhodomonas, Cryptomonas, and Chroomonas/Hemiselmis clades, combined with the fact that nucleomorph genome sizes of over 800 kb are present within Rhodomonas and Chroomonas, suggest that nucleomorph genome size reductions have occurred multiple times independently during cryptomonad evolution, including a 23% size reduction within Rhodomonas, if CCMP1178 is considered a member of this genus.
Nucleomorph Genome Synteny
The complete nucleomorph genome of G. theta consists of three chromosomes (196, 181, and 174 kb in size), each with subtelomeric rDNA cistrons (comprised of 5S, 28S, 5.8S, and 18S genes) and a stretch of repetitive DNA encoding a set of five small hypothetical ORFs (Douglas et al. 2001). A gene for the ubiquitin conjugating enzyme (ubc4) follows this repetitive region on five of the six chromosome ends (missing on one end of chromosome I), and a single-copy gene encoding hsp90 is internal to ubc4 (Zauner et al. 2000; Douglas et al. 2001; fig. 1A). As a launch point for sequence-based nucleomorph genome diversity studies, we used degenerate PCR to isolate a 742-bp hsp90 gene fragment from H. phi CCMP325, P. sulcata CCMP704, R. salina CCMP1319, and Chroomonas mesostigmatica CCMP1168. Phylogenetic analysis of these sequences in the context of a range of eukaryotic cytosolic hsp90 proteins revealed that, as expected, the new genes were most similar to the G. theta nucleomorph hsp90 sequence (data not shown) and exhibited similar base composition biases (28%–30% G + C). These data are consistent with the hypothesis that the new hsp90 sequences are encoded in the nucleomorph genome in these species. PCR experiments using the same primers on Storeatula sp. CCMP1868 and Rhodomonas lens CCMP739 gDNAs failed to generate products of the expected size.
In an attempt to determine the genomic context of the hsp90 genes described above, we employed the "walking PCR" approach of Katz et al. (2000). We were successful in determining the complete 3' end of the hsp90 gene from C. mesostigmatica, but the maximum extension product went only 37 bp beyond the stop codon and did not reveal the identity of the adjacent gene. Similar attempts to extend the hsp90 gene fragment in P. sulcata failed, and we therefore focused on a more directed approach, designing gene-specific primers base on the arrangement of genes in the G. theta genome. To determine if hsp90 is next to ubc4 in other cryptomonad nucleomorph genomes, primers were designed to the ubc4 and hsp90 genes and used in PCRs on gDNAs from C. mesostigmatica, R. salina, P. sulcata, and H. phi, a cryptomonad closely related to G. theta (fig. 2; Deane et al. 1998). Bands of the expected size (1,600 bp) were obtained from H. phi and P. sulcata, although only the H. phi product proved to be a bona fide ubc4-hsp90 product. We then used long-range PCR to extend this genomic fragment across to the 5' end of the 18S rDNA gene in H. phi and PCR amplified and sequenced a fragment of nucleomorph DNA between the 18S and hsp90 loci in P. sulcata. Attempts to extend these fragments to the gamma tubulin gene (tubG), which is located three genes downstream of hsp90 on chromosome I in G. theta, were unsuccessful. We also tried and failed to determine the genomic context of the hsp70 gene in H. phi. Hsp70 has been shown to be present in a wide range of cryptomonad nucleomorph genomes (Rensing et al. 1994) and in G. theta is located near the center of chromosome I, flanked by genes for proteasome subunit B6 (prs B6) and U3 small nucleolar ribonucleoprotein IMP4 (Douglas et al. 2001). Degenerate primers designed to these three genes failed to give PCR products of the expected sizes, possibly because of genomic rearrangements or loss of one or more of these genes in the H. phi nucleomorph genome (see Discussion).
The results of the successful long-range PCR experiments are summarized in figure 3. A comparison of the H. phi and P. sulcata genomic fragments with the subtelomeric region of chromosome I in G. theta (fig. 3A) revealed that while the relative positions of the 18S rDNA, ubc4, and hsp90 genes are conserved, the intergenic spacers vary considerably and are comprised of repetitive and low-complexity sequence. The distance between the 5' end of the 18S rDNA and ubc4 genes in G. theta is 4,050 bp but is only 3,412 and 830 bp in H. phi and P. sulcata, respectively. As in G. theta, several short hypothetical ORFs were identified between the 18S rDNA and ubc4 genes in H. phi (fig. 3B), although these showed no similarity to those in the G. theta genome (Douglas et al. 2001). No tRNA genes were found. In P. sulcata, a short hypothetical ORF was present in the 400-bp region between ubc4 and hsp90 and another was adjacent to the 18S rDNA gene (fig. 3C). We were unable to sequence 300 bp of this fragment due to the presence of G + C-rich tracts that prematurely terminated extension reactions. Attempts to sequence through this area using smaller PCR and cloned products as template and with a variety of alternative primers and sequencing protocols were unsuccessful.
|
To explore the genomic structure of the P. sulcata nucleomorph further, we used a P. sulcata ubc4 probe in a Southern blot against PFGE-separated chromosomes and obtained positive hybridization signals for all three nucleomorph chromosomes in this organism and in G. theta (data not shown). This indicates that the ubc4 gene is present in at least one copy on all three chromosomes in P. sulcata, as is the case in G. theta (Douglas et al. 2001
) (and likely in H. phi). Apart from three instances of clonal length heterogeneity in poly-A tracts near the 3' end of ORF87 in H. phi (1-, 2-, or 3-bp differences), and another in P. sulcata, very little sequence variability was observed in the long-range PCR products described in this study. For example, three independent clones were completely sequenced for the P. sulcata fragment, and only eight ambiguities were apparent in 3.3 kb of sequence (some of these could also be PCR-induced errors even though a high-fidelity polymerase was used for amplifications). This suggests that within a given nucleomorph genome, the repeated sequences on the chromosome ends are very similar to one another. Attempts to determine the copy number and genomic location of the hsp90 gene in P. sulcata using Southern hybridizations were unsuccessful, perhaps due to the extremely low G + C content of the hsp90 probe. A sliding-window analysis of the G + C content of the subtelomeric region in G. theta, H. phi, and P. sulcata revealed considerable intrachromosomal variation, with a moderate G + C content between (and including) the 18S rDNA and ubc4 genes, and a sharp drop in G + C content at the beginning of hsp90 (fig. 3A–C). A codon usage analysis of ubc4 and hsp90 showed significant differences between the two genes: while hsp90 exhibits a codon usage pattern typical of nucleomorph housekeeping genes, that is, with a bias toward A + T-rich codons, ubc4 shows the reverse pattern. This is most pronounced in P. sulcata, where the ubc4 gene is 63% G + C overall and exhibits a strong bias toward G + C in silent codon positions.
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
We have investigated the karyotypic diversity of nine different cryptomonad strains and obtained the first nucleomorph comparative genomic data from within this poorly studied lineage. Our three-species comparison of a region of subtelomeric nucleomorph DNA reveals that while the relative positions of the rDNA cistron, ubc4, and hsp90 genes are conserved, the intergenic spacers bear no resemblance to one another in length or sequence. Indeed, the hypothetical ORFs assigned to the
4-kb region between the 18S rDNA and ubc4 loci in G. theta (Douglas et al. 2001) do not exist in H. phi or P. sulcata, and are thus not likely to be real genes. Our results are consistent with the hypothesis of Zauner et al. (2000), namely, that the process of gene conversion is actively homogenizing the terminal repeats of the nucleomorph chromosomes in G. theta. We conclude that such recombination appears to have led to a situation in which the noncoding sequences at each of the chromosome ends within a genome are identical (or nearly so), but are hypervariable between genomes, even between those that are closely related. Given that ubc4 is present on at least one end of each of the three P. sulcata nucleomorph chromosomes, our data are also consistent with the possibility that the ubc4 gene in H. phi and P. sulcata marks the boundary between repetitive sequences and single-copy DNA, as is the case in G. theta. This terminal repeat region is characterized by a moderate to high G + C content, and while an elevated G + C content for the rDNA genes can be explained by demands placed on their ability to maintain secondary structure, it is not obvious why the G + C content is elevated throughout the repeat region—most of which appears to be noncoding DNA—and drops dramatically at the ubc4/hsp90 junction (fig. 3). Clearly, the mutational and selective pressures influencing the base composition of the repetitive regions of the G. theta, H. phi, and P. sulcata nucleomorph genomes are very different from those acting on single-copy DNA. Intriguingly, a similar pattern of rDNA repeats and elevated G + C contents are seen in the chromosome extremities of the genome of the microsporidian parasite Encephalitozoon cuniculi (Katinka et al. 2001; Peyret et al. 2001).
With respect to cryptomonad nucleomorph genome size, our PFGE results compliment and expand those of Rensing et al. (1994), and indicate that nucleomorph genomes vary from 450 kb in size in Cryptomonas paramecium (Rensing et al. 1994) to well over 800 kb in unidentified species CCMP1178 and members of the genus Chroomonas (fig. 1). The karyotypic diversity within the genus Rhodomonas is surprising when compared to the small genome size differences between Hanusia, Guillardia, and Proteomonas. The PFGE results of Deane et al. (1998) indicate that the nucleomorph genome karyotype of H. phi is indistinguishable from that of G. theta, and we estimate no more than a 20-kb difference in nucleomorph genome size between these two genera and Proteomonas (fig. 1). In stark contrast, strains designated as R. salina vary by as much as 95 kb (fig. 1; Eschbach et al. 1991), and within Rhodomonas we observe a size difference of 100–200 kb, depending on whether one includes unidentified species CCMP1178. This species branches with Rhodomonas abbreviata in our trees, and previous nuclear 18S phylogenies (e.g., Deane et al. 2002; Hoef-Emden, Marin, and Melkonian 2002) indicate that R. abbreviata does not group closely with the other members of the genus and may be an outlier. Our karyotype data support this conclusion as all of the Rhodomonas species investigated showed similar karyotypes (two large chromosomes and one small one), the only exception being CCMP1178 (one large chromosome and two small). Even without considering CCMP1178, the variation in nucleomorph genome sizes within Rhodomonas far exceeds that found between some other genera. Whether this is the result of increased rates of nucleomorph genome evolution within Rhodomonas or is a function of the classification system imposed upon these taxa is unclear, and detailed analyses of both nucleomorph and nuclear sequences will be required to resolve this issue.
Significant nucleomorph genome size variation is also apparent within the genus Cryptomonas (Rensing et al. 1994; fig. 2), although it should be noted that only two species have been investigated thus far, including one that is nonphotosynthetic (C. paramecium). The heterogeneous evolutionary rate of the nucleomorph rDNA in C. paramecium is well documented (Hoef-Emden and Melkonian 2003; Hoef-Emden 2005), and preliminary karyotype data from our laboratory suggests that the small size of the C. paramecium nucleomorph genome is likely the exception in this genus, rather than the rule (C. Lane and J. Archibald, unpublished data). In contrast to Cryptomonas, the evolutionary rate of the Rhodomonas rDNA appears relatively uniform (fig. 2; Hoef-Emden, Marin, and Melkonian 2002), making the nucleomorph genome size variation in this genus all the more surprising.
Given that nucleomorph genome sizes vary dramatically in the cryptomonads examined, the most obvious question is why? It is well known that nuclear genomes vary dramatically in size and structure across the evolutionary tree of eukaryotes. Yet, nucleomorphs represent a special case of nuclear genome evolution, similar in many ways to the evolution of organellar DNA. It is possible that at least some of the nucleomorph genome size heterogeneity observed in this study is due to differences in the size and abundance of introns in different nucleomorphs and in the length of intergenic spacers. Furthermore, larger nucleomorph genomes with more noncoding DNA could be inherently more variable in size, due to the fact that recombination or insertion/deletion events would be less likely to disrupt essential coding regions. However, the extraordinary compactness of the G. theta nucleomorph genome, combined with the small size of the few spliceosomal introns it does possess (Douglas et al. 2001), suggests that differences in amount of noncoding DNA are unlikely to contribute greatly to genome size variation, at least in cryptomonads with relatively small nucleomorph genomes.
An alternative explanation is that cryptomonads differ significantly in the coding capacity of their nucleomorph genomes, that is, in the amount of gene loss and nucleomorph-to-host-nucleus gene transfer that has occurred during cryptomonad evolution. Evidence for nucleomorph genome dynamics comes from an analysis of the hsp70 locus by Rensing et al. (1994). These authors showed that this gene is located on the largest chromosome in some cryptomonad species (including G. theta) but on the second largest in others. Either the hsp70 gene has moved between nonhomologous nucleomorph chromosomes in different species or homologous chromosomes have changed significantly in size in different cryptomonad lineages (Rensing et al. 1994). At present we are unable to distinguish between these two possibilities.
If differential gene content is the major factor contributing to cryptomonad nucleomorph genome size variation, sequence data from species with large genomes will be particularly useful. For example, genes present in the large genomes of Chroomonas and Rhodomonas species, but absent in the smaller G. theta nucleomorph are good candidates for recent transfers to the G. theta nucleus. In the context of a robust phylogenetic framework, not only will it be possible to recognize and study such transfers but it should also be possible to identify gene transfers that have independently taken place multiple times. The data presented here thus provide a start point for detailed investigations into the pattern and process of nucleomorph gene transfer. Such studies will hopefully address one of the fundamental issues of nucleomorph genome evolution, that is, whether the nucleomorphs of cryptomonads and chlorarachniophytes have been reduced to an endpoint beyond which no further reduction is possible or whether they will eventually be eliminated like the endosymbiont nuclei of all other secondary plastid-containing algae.
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
We thank Tom Cavalier-Smith and Uwe Maier for cryptomonad genomic DNAs, Susan Douglas for H. phi cell pellets, Kristen Hoef-Emden and James Deane for SSU rDNA alignments, Paul Gilson for helpful discussion, and Krystal van den Heuval for generating nucleomorph-specific DNA probes. Two anonymous reviewers are also thanked for their helpful comments. This research was supported by Genome Atlantic and a discovery grant from the Natural Sciences and Engineering Research Council of Canada. J.M.A. is a scholar of the Canadian for Advanced Research, Program in Evolutionary biology.
|
Footnotes |
|---|
1 These authors contributed equally to this work
2 Present address: DNA Technologies, Halifax, Nova Scotia, Canada
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Archibald, J. M. 2005. Jumping genes and shrinking genomes—probing the evolution of eukaryotic photosynthesis using genomics. IUBMB Life 57:539–547.[Web of Science][Medline]
Archibald, J. M., T. Cavalier-Smith, U. Maier, and S. Douglas. 2001. Molecular chaperones encoded by a reduced nucleus—the cryptomonad nucleomorph. J. Mol. Evol. 52:490–501.[Web of Science][Medline]
Archibald, J. M., and P. J. Keeling. 2005. On the origin and evolution of plastids. Pp. 238–260 in J. Saap, ed. Microbial phylogeny and evolution. Oxford University Press, New York.
Archibald, J. M., M. B. Rogers, M. Toop, K. Ishida, and P. J. Keeling. 2003. Lateral gene transfer and the evolution of plastid-targeted proteins in the secondary plastid-containing alga Bigelowiella natans. Proc. Natl. Acad. Sci. USA 100:7678–7683.
Bhattacharya, D., H. S. Yoon, and J. D. Hackett. 2003. Photosynthetic eukaryotes unite: endosymbiosis connects the dots. Bioessays 26:50–60.
Bruno, W. J., N. D. Socci, and A. L. Halpern. 2000. Weighted neighbor joining: a likelihood-based approach to distance-based phylogeny reconstruction. Mol. Biol. Evol. 17:189–197.
Cavalier-Smith, T., J. A. Couch, K. E. Thorsteinsen, P. Gilson, J. A. Deane, D. R. A. Hill, and G. I. McFadden. 1996. Cryptomonad nuclear and nucleomorph 18S rRNA phylogeny. Eur. J. Phycol. 31:315–328.
Deane, J. A., D. R. A. Hill, S. J. Brett, and G. I. McFadden. 1998. Hanusia phi gen. et sp. nov. (Cryptophyceae): characterization of ‘Cryptomonas sp. phi’. Eur. J. Phycol. 33:149–154.
Deane, J. A., I. M. Strachan, G. W. Saunders, D. R. A. Hill, and G. I. McFadden. 2002. Cryptomonad evolution: nuclear 18S rDNA phylogeny versus cell morphology and pigmentation. J. Phycol. 38:1236–1244.[CrossRef][Web of Science]
Delwiche, C. F. 1999. Tracing the thread of plastid diversity through the tapestry of life. Am. Nat. 154(Suppl.):S164–S177.[CrossRef][Medline]
Douglas, S. E., C. A. Murphy, D. F. Spencer, and M. W. Gray. 1991. Cryptomonad algae are evolutionary chimaeras of two phylogenetically distinct unicellular eukaryotes. Nature 350:148–151.[CrossRef][Medline]
Douglas, S. E., S. Zauner, M. Fraunholz, M. Beaton, S. Penny, L. Deng, X. Wu, M. Reith, T. Cavalier-Smith, and U.-G. Maier. 2001. The highly reduced genome of an enslaved algal nucleus. Nature 410:1091–1096.[CrossRef][Medline]
Eschbach, S., C. J. Hofmann, U. G. Maier, P. Sitte, and P. Hansmann. 1991. A eukaryotic genome of 660 kb: electrophoretic karyotype of nucleomorph and cell nucleus of the cryptomonad alga, Pyrenomonas salina. Nucleic Acids Res. 19:1779–1781.
Gilson, P. R. 2001. Nucleomorph genomes: much ado about practically nothing. Genome Biol. 2:R1022.1–R1022.5.
Gilson, P. R., U. G. Maier, and G. I. McFadden. 1997. Size isn't everything: lessons in genetic miniaturisation from nucleomorphs. Curr. Opin. Genet. Dev. 7:800–806.[CrossRef][Web of Science][Medline]
Gilson, P., and G. I. McFadden. 1995. The chlorarachniophyte: a cell with two different nuclei and two different telomeres. Chromosoma 103:635–641.[Web of Science][Medline]
Gilson, P. R., and G. I. McFadden. 1996. The miniaturized nuclear genome of eukaryotic endosymbiont contains genes that overlap, genes that are cotranscribed, and the smallest known spliceosomal introns. Proc. Natl. Acad. Sci. USA 93:7737–7742.
———. 2002. Jam packed genomes—a preliminary, comparative analysis of nucleomorphs. Genetica 115:13–28.[CrossRef][Web of Science][Medline]
Greenwood, A. D. 1974. The Cryptophyta in relation to phylogeny and photosynthesis. Pp. 566–567 in J. V. Sanders and D. J. Goodchild, eds. Electron microscopy 1974. Australian Academy of Sciences, Canberra, Australia.
Greenwood, A. D., H. B. Griffiths, and U. J. Santore. 1977. Chloroplasts and cell compartments in Cryptophyceae. Br. J. Phycol. 12:119.
Hackett, J. D., D. M. Anderson, D. L. Erdner, and D. Bhattacharya. 2004. Dinoflagellates: a remarkable evolutionary experiment. Am. J. Bot. 91:1523–1534.
Hansmann, P. 1988. Ultrastructural localization of RNA in cryptomonads. Protoplasma 146:81–88.[CrossRef]
Hansmann, P., H. Falk, and P. Sitte. 1985. DNA in the nucleomorph of Cryptomonas demonstrated by DAPI fluorescence. Z. Naturforsch. 40c:933–935.
Hoef-Emden, K. 2005. Multiple independent losses of photosynthesis and differing evolutionary rates in the genus Cryptomonas (Cryptophyceae): combined phylogenetic analyses of DNA sequences of the nuclear and the nucleomorph ribosomal operons. J. Mol. Evol. 60:183–195.[Medline]
Hoef-Emden, K., B. Marin, and M. Melkonian. 2002. Nuclear and nucleomorph SSU rDNA phylogeny in the Cryptophyta and the evolution of cryptophyte diversity. J. Mol. Evol. 55:161–179.[CrossRef][Web of Science][Medline]
Hoef-Emden, K., and M. Melkonian. 2003. Revision of the genus Cryptomonas (Cryptophyceae): a combination of molecular phylogeny and morphology provides insights into a long-hidden dimorphism. Protist 154:371–409.[Medline]
Hofmann, C. J., S. A. Rensing, M. M. Hauber, W. F. Martin, S. B. Muller, J. Couch, G. I. McFadden, G. L. Igloi, and U. G. Maier. 1994. The smallest known eukaryotic genomes encode a protein gene: towards an understanding of nucleomorph functions. Mol. Gen. Genet. 243:600–604.[CrossRef][Web of Science][Medline]
Huelsenbeck, J., and F. Ronquist. 2003. MrBayes. Evolutionary Biology Centre, Uppsala University, Uppsala, Sweden.
Ishida, K., Y. Cao, M. Hasegawa, N. Okada, and Y. Hara. 1997. The origin of chlorarachniophyte plastids, as inferred from phylogenetic comparisons of amino acid sequences of EF-Tu. J. Mol. Evol. 45:682–687.[CrossRef][Web of Science][Medline]
Ishida, K., B. R. Green, and T. Cavalier-Smith. 1999. Diversification of a chimaeric algal group, the chlorarachniophytes: phylogeny of nuclear and nucleomorph small-subunit rRNA genes. Mol. Biol. Evol. 16:321–331.[Web of Science]
Katinka, M. D., S. Duprat, E. Cornillot et al. (17 co-authors). 2001. Genome sequence and gene compaction of the eukaryote parasite Encephalitozoon cuniculi. Nature 414:450–453.[CrossRef][Medline]
Katz, L. A., E. A. Curtis, M. Pfunder, and L. F. Landweber. 2000. Characterization of novel sequences from distantly related taxa by walking PCR. Mol. Phylogenet. Evol. 14:318–321.[CrossRef][Web of Science][Medline]
Keeling, P. J. 2004. Diversity and evolutionary history of plastids and their hosts. Am. J. Bot. 91:1481–1493.
Ludwig, M., and S. P. Gibbs. 1985. DNA is present in the nucleomorph of cryptomonads: further evidence that the chloroplast evolved from a eukaryotic endosymbiont. Protoplasma 127:9–20.[CrossRef]
Maddison, W., and D. Maddison. 2003. MacClade. Sinauer Associates, Sunderland, Mass.
Maier, U. G., C. J. Hofmann, S. Eschbach, J. Wolters, and G. L. Igloi. 1991. Demonstration of nucleomorph-encoded eukaryotic small subunit ribosomal RNA in cryptomonads. Mol. Gen. Genet. 230:155–160.[CrossRef][Web of Science][Medline]
Marin, B., M. Klingberg, and M. Melkonian. 1998. Phylogenetic relationships among the Cryptophyta: analyses of nuclear-encoded SSU rRNA sequences support the monophyly of extant plastid-containing lineages. Protist 149:265–276.
Martin, W., T. Rujan, E. Richly, A. Hansen, S. Cornelsen, T. Lins, D. Leister, B. Stoebe, M. Hasegawa, and D. Penny. 2002. Evolutionary analysis of Arabidopsis, cyanobacterial, and chloroplast genomes reveals plastid phylogeny and thousands of cyanobacterial genes in the nucleus. Proc. Natl. Acad. Sci. USA 99:12246–12251.
McFadden, G. I. 1999. Plastids and protein targeting. J. Eukaryot. Microbiol. 46:339–346.[Web of Science][Medline]
McFadden, G. I., P. R. Gilson, C. J. Hofmann, G. J. Adcock, and U. G. Maier. 1994. Evidence that an amoeba acquired a chloroplast by retaining part of an engulfed eukaryotic alga. Proc. Natl. Acad. Sci. USA 91:3690–3694.
McFadden, G. I., P. R. Gilson, and R. F. Waller. 1995. Molecular phylogeny of chlorarachniophytes based on plastid rRNA and rbcL sequences. Arch. Protistenkd. 145:231–239.
Morrall, S., and A. D. Greenwood. 1982. Ultrastructure of nucleomorph division in species of Cryptophyceae and its evolutionary implication. J. Cell Sci. 54:311–328.
Peyret, P., M. D. Katinka, S. Duprat, F. Duffieux, V. Barbe, M. Barbazanges, J. Weissenbach, W. Saurin, and C. P. Vivares. 2001. Sequence and analysis of chromosome I of the amitochondriate intracellular parasite Encephalitozoon cuniculi (Microspora). Genome Res. 11:198–207.
Posada, D., and K. A. Crandall. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14:817–818.
Rensing, S. A., M. Goddemeier, C. J. Hofmann, and U. G. Maier. 1994. The presence of a nucleomorph hsp70 gene is a common feature of Cryptophyta and Chlorarachniophyta. Curr. Genet. 26:451–455.[Medline]
Swofford, D. L. 2002. Phylogenetic analysis using parsimony (*and other methods), version 4.0B10 PPC. Sinauer Associates, Sunderland, Mass.
Van der Auwera, G., C. J. B. Hofmann, P. De Rijk, and R. De Wachter. 1998. The origin of red algae and cryptomonad nucleomorphs: a comparative phylogeny based on small and large subunit rRNA sequences of Palmaria palmata, Gracilaria verrucosa, and the Guillardia theta nucleomorph. Mol. Phylogenet. Evol. 10:333–342.[CrossRef][Web of Science][Medline]
van Dooren, G. G., S. D. Schwartzbach, T. Osafune, and G. I. McFadden. 2001. Translocation of proteins across the multiple membranes of complex plastids. Biochim. Biophys. Acta 1541:34–53.[Medline]
Yoon, H. S., J. D. Hackett, C. Ciniglia, G. Pinto, and D. Bhattacharya. 2004. A molecular timeline for the origin of photosynthetic eukaryotes. Mol. Biol. Evol. 21:809–818.
Zauner, S., M. Fraunholz, J. Wastl, S. Penny, M. Beaton, T. Cavalier-Smith, U.-G. Maier, and S. E. Douglas. 2000. Chloroplast protein and centrosomal genes, a tRNA intron, and odd telomeres in an unusually compact eukaryotic genome, the cryptomonad nucleomorph. Proc. Natl. Acad. Sci. USA 97:200–205.
CiteULike 
Connotea 
Del.icio.us What's this?
Related articles in MBE:
- Insight into the Diversity and Evolution of the Cryptomonad Nucleomorph Genome
MBE 2006 23: 1817.[Extract] [FREE Full Text]
This article has been cited by other articles:
|
|
 |
|
  J. M. Archibald and C. E. Lane Going, Going, Not Quite Gone: Nucleomorphs as a Case Study in Nuclear Genome Reduction J. Hered., September 1, 2009; 100(5): 582 - 590. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Troedsson, R. F. Lee, V. Stokes, T. L. Walters, P. Simonelli, and M. E. Frischer Development of a Denaturing High-Performance Liquid Chromatography Method for Detection of Protist Parasites of Metazoans Appl. Envir. Microbiol., July 15, 2008; 74(14): 4336 - 4345. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Khan and J. M. Archibald Lateral transfer of introns in the cryptophyte plastid genome Nucleic Acids Res., May 1, 2008; 36(9): 3043 - 3053. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Fong and J. M. Archibald Evolutionary Dynamics of Light-Independent Protochlorophyllide Oxidoreductase Genes in the Secondary Plastids of Cryptophyte Algae Eukaryot. Cell, March 1, 2008; 7(3): 550 - 553. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. Lane, K. van den Heuvel, C. Kozera, B. A. Curtis, B. J. Parsons, S. Bowman, and J. M. Archibald Nucleomorph genome of Hemiselmis andersenii reveals complete intron loss and compaction as a driver of protein structure and function PNAS, December 11, 2007; 104(50): 19908 - 19913. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Khan, N. Parks, C. Kozera, B. A. Curtis, B. J. Parsons, S. Bowman, and J. M. Archibald Plastid Genome Sequence of the Cryptophyte Alga Rhodomonas salina CCMP1319: Lateral Transfer of Putative DNA Replication Machinery and a Test of Chromist Plastid Phylogeny Mol. Biol. Evol., August 1, 2007; 24(8): 1832 - 1842. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||