Plastid-Derived Single Gene Minicircles of the Dinoflagellate Ceratium horridum Are Localized in the Nucleus
http://www.100md.com
分子生物学进展 2004年第7期
* Institute of Botany, Heinrich-Heine-University Düsseldorf, Düsseldorf, Germany
Cell Biology, Philipps-University Marburg, Marburg, Germany
E-mail: zauner@staff.uni-marburg.de.
Abstract
Recent reports show that numerous chloroplast-specific proteins of peridinin-containing dinoflagellates are encoded on minicircles—small plasmidlike molecules containing one or two polypeptide genes each. The genes for these polypeptides are chloroplast specific because their homologs from other photosynthetic eukaryotes are exclusively encoded in the chloroplast genome. Here, we report the isolation, sequencing, and subcellular localization of minicircles from the peridinin-containing dinoflagellate Ceratium horridum. The C. horridum minicircles are organized in the same manner as in other peridinin-containing dinoflagellates and encode the same kinds of plastid-specific proteins, as previous studies reported. However, intact plastids isolated from C. horridum do not contain minicircles, nor do they contain DNA that hybridizes to minicircle-specific probes. Rather, C. horridum minicircles are localized in the nucleus as shown by cell fractionation, Southern hybridization, and in situ hybridization with minicircle-specific probes. A high-molecular-weight DNA was detected in purified C. horridum plastids, but it is apparently not minicircular in organization, as hybridization with a cloned probe from the plastid-localized DNA suggests. The distinction between C. horridum and other peridinin-containing dinoflagellates at the level of their minicircle localization is paralleled by C. horridum thylakoid organization, which also differs from that of other peridinin-containing dinoflagellates, indicating that a hitherto underestimated diversity of minicircle DNA localization and thylakoid organization exists across various dinoflagellate groups.
Key Words: Dinoflagellates ? plastids ? minicircles ? peridinin
Introduction
The organization of the genetic material in plastids of peridinin-containing dinoflagellates, a group of photosynthetic, unicellular alveolate eukaryotes, is thought to be unique in that a minor portion of chloroplast proteins and ribosomal RNAs appear to be individually encoded on plasmidlike "minicircles" of approximately 2 to 3 kb in size (Zhang, Green, and Cavalier-Smith 1999; Barbrook and Howe 2000; Zhang, Cavalier-Smith, and Green 2001; Barbrook et al. 2001; Hiller 2001), rather than collectively on evolutionarily reduced cyanobacterial chromosomes of approximately 100 to 200 kb in length, as in all other photosynthetic organelles (Martin et al. 1998; Douglas 1998; Stoebe and Kowallik 1999). Moreover, recent work shows that gene transfer from the plastid into the cell nucleus is unusual high, making peridinin-containing dinoflagellates the "champions" in this field (Hackett et al. 2004).
Sequence similarities indicate that minicircle-encoded genes are of cyanobacterial origin (Zhang, Green, and Cavalier-Smith 1999), and it is generally assumed that these unusual molecules reside in the plastid of the species studied so far (Zhang, Green, and Cavalier-Smith 1999; McFadden 1999; Barbrook et al. 2001). However, in none of those pioneering studies on dinoflagellate minicircles has any direct evidence been reported that the minicircle DNA does, in fact, reside in the plastid. A recent study has shown that minicircle mRNA transcripts (though not minicircle DNA) are indeed found in the plastid of Symbiodinium (Takishita et al. 2003), providing additional indirect evidence that the minicircle DNA itself might be plastid localized in the peridinin-containing dinoflagellate Symbiodinium. Here, we report the cloning and characterization of minicircles with sequences potentially encoding nine chloroplast proteins from the peridinin-containing dinoflagellate Ceratium horridum. We have investigated the localization of the minicircles to determine where, within the cell, the minicircles reside in this dinoflagellate species, whose plastids can be physically isolated intact by cell fractionation techniques. Reinspection of the plastid morphology showed that C. horridum substantially differs from other phototrophic dinoflagellates, despite its possession of peridinin and minicircles.
Methods
Strain
Ceratium horridum (Cleve) was isolated by H.A. v. Stosch 1965 in Helgoland, Germany. The strain is deposited at SAG G?ttingen (www.gwdg.de/epsag/phykologia/epsag.html).
Isolation of Minicircles
C. horridum was harvested from cultures by filtration through a 22-mm mesh gauze. The cells were resuspended in 250-μl buffer P1 of the QIAprep Spin Miniprep Kit (QIAGEN) in a 1.5-ml tube and homogenized using a teflon pestle. Minicircles were isolated by following the manufacturers' instructions for plasmid isolation. Isolated Minicircles were restricted using either HindIII, XbaI, or BglII and subsequently cloned into the correspondent pUC18 plasmid. Shotgun sequencing was performed on a Li-Cor 4200 sequencer and sequences thus obtained were assembled using Sequencher (GeneCodes). Circularity was verified using an inverse PCR approach.
Isolation of C. horridum Chloroplasts
C. horridum (100 to 200 mg) was resuspended in an ice-cold isolation buffer (0.4M mannitol, 1mM MgCl2, 10mM Na2EDTA, 0.25% polyvinylpyrrolidone 90, 50mM Tris-HCl pH 7.5), homogenized in a 5-ml potter and filtered through a 5-μm mesh gauze. The chloroplasts were pelleted from the filtrate at 3,400g, resuspended in a 150-μl isolation buffer, loaded on 30% (v/v) Percoll (Amersham) in isolation buffer and centrifuged at 4,000g for 10 min.
Isolation of C. horridum Nuclei
C. horridum cells were suspended and homogenized as for plastid preparation (mannitol concentration was increased to 0.7 M). The homogenate was filtered through a 22-μm mesh gauze and centrifuged at 2,500g to 3000g. The pellet was resuspended and loaded on 30% to 40% (v/v) Percoll (Amersham). After centrifugation for 10 min and 3,400g the gradient was fractionated, and fractions containing intact nuclei were identified microscopically.
Electron Microscopy
Ultrathin sections of trypsin-treated Prorocentrum micans were prepared as described (Kowallik 1971) and recorded in a Philipps EM 301 electron microscope as described (Herrmann et al. 1975).
In Situ Hybridization
In situ hybridizations were performed as described (McFadden 1991). The probes used for hybridizations were a 1,244-bp biotinylated probe spanning the coding region of the psbB minicircle, a 1,038-bp probe from the RuBisCO gene of C. horridum, and a 66-bp (5'-CCAGTGGTCGTTGGTTGGTGTGTGGTTGGTGGCGGTGGGGGTTGTGCCGGCGTGGTTGTTGTTGGG-3') fragment cloned from the high molecular plastid DNA from C. horridum. As a negative control, in situ hybridizations were done with the secondary antibody only. Hybrids were immunodetected for electron microscopy with gold-coupled antibiotin antibodies and with Cy2-coupled antibiotin antibodies for confocal laser scanning microscopy.
Results and Discussion
Minicircles from Ceratium horridum were enriched from total dinoflagellate DNA by a procedure that exploits their plasmidlike size and conformation (figure 1, lane B). The most prominent circular minicircle size populations in C. horridum migrate in the range of 3.5 to 4 kb in agarose gels (figure 1, lane B), similar to the size range of cloned minicircles previously found for Heterocapsa triquetra, Amphidinium operculatum, and Amphidinium carterae (Zhang, Green, and Cavalier-Smith; Barbrook and Howe 2000; Zhang et al. 2001; Barbrook et al. 2001; Hiller 2001).
FIG. 1. DNA profiles from cell fractions of Ceratium horridum. Lane A: Lambda DNA cut with EcoRI/HindIII. Lane B: Minicircle purification of C. horridum cells. Lane C: Total-cell DNA, including minicircles, most of which band around 4 kb. Lane D: DNA from intact plastids isolated from Percoll pellet and subsequently lysed with CTAB. Lane E: Linearized pUC18 added as control. Lane F: Same fraction as lane D after DNase I treatment prior to CTAB lyses. Lane G: Lambda size marker. Lanes C* to G* Southern blot from lanes C to G and hybridization using the digoxiginin-labeled psbB coding region of Ceratium horridum minicircle as a probe. Lane H: DNA from isolated nuclei. Lane H*: Southern blot from Lane H, hybridized with a minicircle probe (psbB probe). Lane I: Lambda size marker. Lane K: DNA from isolated plastids. Lane L: DNA from the supernatant of the plastid isolation, harboring DNA from nuclei, mitochondria, and broken plastids. Lanes I*, K*, and L*: Southern blot of lanes I to K, hybridization with a digoxiginin-labeled probe isolated from the high-molecular-weight plastid DNA. Lanes I+, K+, and L+: Southern blot of lanes I-K, hybridized with a minicircle-specific psbB probe. MP indicates minicircle probe
Sequence comparisons showed that the C. horridum minicircles encode genes of cyanobacterial origin, homologs of which are plastome-localized in other algae. By cloning of isolated minicircles, we detected genes and gene fragments for plastid-related psaA, psaB, psbB, psbC, psbD, psbE, petB, ycf16, and ycf24. Whereas the genes psaA, psaB, psbB, psbC, psbD, psbE, and petB are contained in all plastid genomes sequenced to date, ycf16 and ycf24 are specific to plastid genomes from non–green algal lineages (Stoebe, Martin, and Kowallik. 1999). All minicircle-encoded proteins from peridinin-containing dinoflagellates isolated here and previously (Zhang, Green, and Cavalier-Smith 1999; Barbrook and Howe 2000; Zhang, Cavalier-Smith, and Green; Barbrook et al. 2001; Hiller 2001) share convincing amino acid similarity with their homologs from red algal chloroplast genomes (and plastids derived from red algae via secondary endosymbiosis). A significant portion of minicircles from C. horridum encode one protein per plasmid, except ycf16/ycf24, which are located on a single minicircle in tandem. As with other dinoflagellates (Barbrook et al. 2001; Hiller 2001), several C. horridum minicircles did not contain a recognizable coding region. We found no typical transit peptide coding regions contiguous with the minicircle open reading frames, consistent with findings from previous studies.
To physically localize the minicircles within the cell, intact plastids were purified from C. horridum by cell fractionation, and their nucleic acids were isolated. Purified plastids did not contain detectable amounts of minicircle DNA (figure 1, lanes D and F), whereas DNA from whole cells did (figure 1, lane C). This was confirmed by Southern hybridization with a minicircle-specific probe (figure 1, lanes C* to G*). Instead, the plastids release a high-molecular-weight DNA upon lysis (figure 1, lanes D and F). To further pinpoint the localization of these minicircles, we isolated nuclei from C. horridum. Minicircles can be observed in large numbers in the nuclear DNA (figure 1, lane H), which was confirmed by Southern hybridizations with a minicircle probe (figure 1, lane H*).
We tried to demonstrate the presence of minicircles in the dinoflagellate plastids by in situ hybridization and imaging with confocal laser scanning microscopy, using a 1,244 bp long stretch from the coding region of minicircle psbB from C. horridum (fig. 2f) as well as a fragment from high-molecular-weight plastid DNA (fig. 2b) as probes. An additional control was performed in which the secondary antibody was tested for unspecific reactions (not shown), showing the autofluorescence of the plastids. As shown in figure 2f, a minicircle probe labeled the nucleus in the space between the chromosomes. Furthermore, the high-molecular-weight plastid probe hybridized in the plastid (fig. 2b). That the psbB probe specifically labeled the interchromosomal regions of the nucleus was confirmed by conventional in situ hybridization techniques and visualization with electron microscopy (fig. 2g). As a further control, a probe spanning the C. horridum–specific RuBisCo labeled a chromosomal region, whereas a nonhomologous probe (phycoerythrin from cryptomonads) showed no significant hybridization (data not shown). Taken together, these findings indicate that the sequence of the psbB minicircle is primarily, and probably exclusively, localized extrachromosomal in the nucleus and not in the plastid of C. horridum.
FIG. 2. In situ hybridizations. (a) and (e) show phase contrast images of (b) and (f), respectively. In situ hybridizations are shown in (b) using a plastid genome–derived probe and a psbB probe in (e). (c) Visualizes a mechanically stressed C. horridum cell releasing the nucleus and plastids. These organelles (P = plastid; N = nucleus), before its separation, are shown in (d). (g) Hybridization using the minicircle-encoded psbB coding region of C. horridum as probe and visualizing the hybrids with the electron microscope. Numerous hybrids in the cell nucleus (N) and one in the plastid (P) are shown by arrows. The bar = 1 μm
In a recent study, minicircle-encoded mRNA for psbA was detected in the chloroplast of the genus Symbiodinium by in-situ hybridization (Takishita et al. 2003). These data, together with the results of Zhang, Green, and Cavalier-Smith (1999) on Prorocentrum, from which no 23S rRNA was detected, raises the possibility that in some dinoflagellate species, minicircles may be plastid located, whereas in others, the chloroplast genes have moved into the cell nucleus (Cavalier-Smith 2003). This hypothesis is complicated by the finding that in some dinoflagellate species, plastid DNAs has been identified with molecular sizes between 114 to 125 kb, consistent with plastid DNAs from other phototrophic organisms (Boczar, Liston, and Catollico 1991).
This conundrum of findings prompted us to start again from scratch to identify the genuine dinoflagellate plastid DNA from C. horridum. A plastid-localized DNA can be clearly seen inside the dinoflagellate stroma by traditional transmission electron microscopy of, for example, Prorocentrum micans (fig. 3). To find the plastid DNA in Ceratium, intact plastids were isolated from C. horridum. Isolated plastids that were lysed after treatment with DNase I released a high-molecular-weight DNA that is DNase sensitive (figure 1, lane K). C. horridum minicircles do not hybridize to the high-molecular-weight DNA isolated from such purified plastids (figure 1, lane K+), but do hybridize to the supernatant of this preparation, which contains broken nuclei (figure 1, lanes L and L+). Thus, the C. horridum plastid DNA is apparently distinct from minicircle DNA. However, in Southern experiments, a probe cloned from the plastid DNA fraction (see Methods) hybridized with the high-molecular-plastid DNA (figure 1, lane K*), as well as with the DNA isolated from the supernatant of the plastid preparation, which contains DNA from the other organelles and from broken plastids (figure 1, lane L*). Finally, in situ hybridization experiments with the probe for the high-molecular-weight DNA confirmed its plastid location (fig. 2b). Nevertheless, the chloroplast-located DNA of C. horridum remains mysterious. Sequences obtained from this genome do not show any coding region or significant similarities to other data base entries.
FIG. 3. Chloroplast nucleoid (black arrow) of Prorocentrum micans. The DNA was made visible after tryptic digestion of the stroma
Peridinin-containing dinoflagellates are clearly defined by their plastid morphology: a triple membrane envelope and thylakoid stacks in layers of three. Because our findings suggested that there may be lineage-specific differences of minicircle localization among peridinin-containing dinoflagellates, we reinspected the plastid morphotype of C. horridum. One of us (KVK) noticed that differences are obvious, because the C. horridum plastid shows stacks of thylakoids in rows of two in contrast to the stacks of three typical for other peridinin-containing dinoflagellates. Furthermore, the thylakoid lumen is opaque as in cryptophyte plastids. However, the plastids are devoid of phycobiliproteins but contain peridinin.
Our findings indicate that the organization of the dinoflagellate C. horridum minicircles is surprising because it is a nuclear extrachromosomal and not a plastid genome organization, although in evolutionary terms, it ultimately stems from the three-membrane-bounded plastid of these protists. DNA does exist in plastids from C. horridum, but it is distinct from the nuclear minicircles previously thought to exclusively represent the chloroplast DNA.
Where are plastid proteins expressed? If they are expressed in the cytoplasm of the host, how do they find their final target sites? If the minicircles of C. horridum (but possibly not those from other dinoflagellates [Takishita et al. 2003]) represent only vehicles for transferring plastid DNA into the cell nucleus, one would have to postulate that the plastid-located DNA does indeed encode a set of proteins known from other chloroplasts (Martin et al. 2002). Our preliminary analysis of the plastid genome shows yet no indications of such encoded genes. However, coding regions could be hidden and recognized only after an editing machinery, known from chloroplasts of higher plants (Bock 2000), has formed an mRNA that could be reasonably translated. If minicircles are the transportation molecules into the nucleus, one would suggest that dinoflagellates show an extended rate of plastid gene migration into the host. Hints for this comes from a recent publication showing that 15 genes that are limited to the plastid genome in all other eukaryotic phototrophs are nucleus encoded in a peridinin-containing dinoflagellate (Hackett et al. 2004).
As most of the minicircle-encoded proteins are known to be expressed in chloroplasts of other groups, one could argue that RNA import might provide the plastid with RNA messages. Such a roughly analogous situation is known, for example, in the case of tRNA import into mitochondria of trypanosomes (Schneider 2001). If the C. horridum nuclear minicircles specify chloroplast proteins, then these would be expected to encode a preprotein, harboring a signal sequence and a transit peptide to manage the transport across the plastid surrounding membranes. However, the coding region does not show any N-terminal extension that could serve as a signal sequence. Again, topogenic signals could be hidden in the genomic sequence and uncovered in the mRNA after RNA editing. Notably, editing can be observed in C. horridum minicircles (Zauner and Maier, unpublished data).
Dinoflagellates are unusual in many respects, as shown, for example, by their chromosome architecture and by the finding that peridinin-containing dinoflagellates express an unusual form II RuBisCo (Morse et al. 1995, Whitney, Shaw, and Yellowlees 1995, Rowan et al. 1996). C. horridum furthermore differs from other phototrophic dinoflagellates investigated thus far by its noncanonical plastid morphology, despite of the existence of peridinin, and by its minicircle localization, assuming that all other dinoflagellate minicircles described thus far are localized in the plastid. Thus, peridinin-containing dinoflagellates could be a heterogenous subgroup of alveolates, in which nature might have found a genetic playground.
Supplementary Material
Sequences are deposited in GenBank under accession numbers AF490356 to AF490364 and AF490368.
Acknowledgements
The authors would like to thank Marianne Johannsen (Marburg) and Dr. Franz Grolig (Marburg) for their help in electron and confocal laser scanning microscopy. Supported by the Deutsche Forschungsgemeinschaft.
Literature Cited
Barbrook, A. C., and C. J. Howe. 2000. Minicircular plastid DNA in the dinoflagellate Amphidinium operculatum. Mol. Gen. Genet. 263:152-158.
Barbrook, A. C., H. Symington, R. E. R. Nisbet, A. Larkum, and C. J. Howe. 2001. Organisation and expression of the plastid genome of the dinoflagellate Amphidinium operculatum. Mol. Genet. Genomics. 266:632-638.
Bock, R. 2000. Sense from nonsense: how the genetic information of chloroplasts is altered by RNA editing. Biochemie 82:549-557.
Boczar, B. A., J. Liston, and R. A. Catollico. 1991. Characterization of satellite DNA from three marine dinoflagellates (Dinophyceae): Glenodinium sp. and two members of the toxic genus Protogonyaulax. Plant Physiol. 97:613-618.
Cavalier-Smith, T. 2003. Genomic reduction and evolution of novel genetic membranes and protein-targeting machinery in eukaryote-eukaryote chimaeras (meta-algae). Philos. Trans. R. Soc. Lond. B Biol. Sci. 358:109-133.
Douglas, S. E. 1998. Plastid evolution: origins, diversity, trends. Curr. Opin. Genet. Dev. 8:655-661.
Hackett, J. D., H. S. Yoon, M. B. Soares, M. F. Bonaldo, T. L. Casavant, T. E. Scheetz, T. Nosenko, and D. Bhattacharya. 2004. Migration of the plastid genome to the nucleus in a peridinin dinoflagellate. Curr. Biol. 14:213-218.
Herrmann, R. G., H. J. Bohnert, K. V. Kowallik, and J. M. Schmitt. 1975. Size, conformation and purity of chloroplast DNA of some higher plants. Biochim. Biophys. Acta 378:305-317.
Hiller, R. G. 2001. ‘Empty’ minicircles and petB/atpA and psbD/psbE (cytb559 alpha) genes in tandem in Amphidinium carterae plastid DNA. FEBS Lett. 505:449-452.
Kowallik, K. V. 1971. The use of proteases for improved presentation of DNA in chromosomes and chloroplasts of Prorocentrum micans (Dinophyceae). Arch. Mikrobiol. 80:154-165.
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.
Martin, W., B. Stoebe, V. Goremykin, S. Hausmann, M. Hasegawa, and K. V. Kowallik. 1998. Gene transfer to the nucleus and the evolution of chloroplasts. Nature 393:162-165.
McFadden, G. I. 1991. In situ hybridization techniques: molecular cytology goes ultrastructural. Pp. 219–255 in J. L. Hall, C. R. Hawes, eds. Electron microscopy of plant cells. Academic Press, London.
McFadden, G. I. 1999. Chloroplasts: ever decreasing circles. Nature 400:119-120.
Morse, D., P. Salois, P. Markovic, and J. W. Hastings. 1995. A nuclear-encoded form II RuBisCO in dinoflagellates. Science 268:1622-1624.
Rowan, R., S. M. Whitney, A. Fowler, and D. Yellowlees. 1996. RuBisCO in marine symbiotic dinoflagellates: form II enzymes in eukaryotic oxygenic phototrophs encoded by a nuclear multigene family. Plant Cell 8:543-545.
Schneider, A. 2001. Unique aspects of mitochondrial biogenesis in trypanosomatids. Int. J. Parasitol. 31:1403-1415.
Stoebe, B., and K. V. Kowallik. 1999. Gene-cluster analysis in chloroplast genomics. Trends Genet. 15:344-347.
Stoebe, B., W. Martin, and K. V. Kowallik. 1998. Distribution and nomenclature of protein-coding genes in 12 sequenced chloroplast genomes. Plant Mol. Biol. Reptr. 16:243-255.
Takishita, K., M. Ishikura, K. Koike, and T. Maruyama. 2003. Comparison of phylogenies based on nuclear-encoded SSU rDNA and plastid-encoded psbA in the symbiotic dinoflagellate genus Symbiodinium. Phycologia 42:285-291.
Whitney, S. M., D. C. Shaw, and D. Yellowlees. 1995. Evidence that some dinoflagellates contain a ribulose-1,5-bisphosphate carboxylase/oxygenase related to that of the alpha-proteobacteria. Proc. R. Soc. Lond. B Biol. Sci. 259:271-275.
Zhang, Z., B. R. Green, and T. Cavalier-Smith. 1999. Single gene circles in dinoflagellates chloroplast genomes. Nature 400:155-159.
Zhang, Z., T. Cavalier-Smith, and B. R. Green. 2001. A family of selfish minicircular chromosomes with jumbled chloroplast gene fragments from a dinoflagellate. Mol. Biol. Evol. 18:1558-1565.(T. Laatsch*,1, S. Zauner,)
Cell Biology, Philipps-University Marburg, Marburg, Germany
E-mail: zauner@staff.uni-marburg.de.
Abstract
Recent reports show that numerous chloroplast-specific proteins of peridinin-containing dinoflagellates are encoded on minicircles—small plasmidlike molecules containing one or two polypeptide genes each. The genes for these polypeptides are chloroplast specific because their homologs from other photosynthetic eukaryotes are exclusively encoded in the chloroplast genome. Here, we report the isolation, sequencing, and subcellular localization of minicircles from the peridinin-containing dinoflagellate Ceratium horridum. The C. horridum minicircles are organized in the same manner as in other peridinin-containing dinoflagellates and encode the same kinds of plastid-specific proteins, as previous studies reported. However, intact plastids isolated from C. horridum do not contain minicircles, nor do they contain DNA that hybridizes to minicircle-specific probes. Rather, C. horridum minicircles are localized in the nucleus as shown by cell fractionation, Southern hybridization, and in situ hybridization with minicircle-specific probes. A high-molecular-weight DNA was detected in purified C. horridum plastids, but it is apparently not minicircular in organization, as hybridization with a cloned probe from the plastid-localized DNA suggests. The distinction between C. horridum and other peridinin-containing dinoflagellates at the level of their minicircle localization is paralleled by C. horridum thylakoid organization, which also differs from that of other peridinin-containing dinoflagellates, indicating that a hitherto underestimated diversity of minicircle DNA localization and thylakoid organization exists across various dinoflagellate groups.
Key Words: Dinoflagellates ? plastids ? minicircles ? peridinin
Introduction
The organization of the genetic material in plastids of peridinin-containing dinoflagellates, a group of photosynthetic, unicellular alveolate eukaryotes, is thought to be unique in that a minor portion of chloroplast proteins and ribosomal RNAs appear to be individually encoded on plasmidlike "minicircles" of approximately 2 to 3 kb in size (Zhang, Green, and Cavalier-Smith 1999; Barbrook and Howe 2000; Zhang, Cavalier-Smith, and Green 2001; Barbrook et al. 2001; Hiller 2001), rather than collectively on evolutionarily reduced cyanobacterial chromosomes of approximately 100 to 200 kb in length, as in all other photosynthetic organelles (Martin et al. 1998; Douglas 1998; Stoebe and Kowallik 1999). Moreover, recent work shows that gene transfer from the plastid into the cell nucleus is unusual high, making peridinin-containing dinoflagellates the "champions" in this field (Hackett et al. 2004).
Sequence similarities indicate that minicircle-encoded genes are of cyanobacterial origin (Zhang, Green, and Cavalier-Smith 1999), and it is generally assumed that these unusual molecules reside in the plastid of the species studied so far (Zhang, Green, and Cavalier-Smith 1999; McFadden 1999; Barbrook et al. 2001). However, in none of those pioneering studies on dinoflagellate minicircles has any direct evidence been reported that the minicircle DNA does, in fact, reside in the plastid. A recent study has shown that minicircle mRNA transcripts (though not minicircle DNA) are indeed found in the plastid of Symbiodinium (Takishita et al. 2003), providing additional indirect evidence that the minicircle DNA itself might be plastid localized in the peridinin-containing dinoflagellate Symbiodinium. Here, we report the cloning and characterization of minicircles with sequences potentially encoding nine chloroplast proteins from the peridinin-containing dinoflagellate Ceratium horridum. We have investigated the localization of the minicircles to determine where, within the cell, the minicircles reside in this dinoflagellate species, whose plastids can be physically isolated intact by cell fractionation techniques. Reinspection of the plastid morphology showed that C. horridum substantially differs from other phototrophic dinoflagellates, despite its possession of peridinin and minicircles.
Methods
Strain
Ceratium horridum (Cleve) was isolated by H.A. v. Stosch 1965 in Helgoland, Germany. The strain is deposited at SAG G?ttingen (www.gwdg.de/epsag/phykologia/epsag.html).
Isolation of Minicircles
C. horridum was harvested from cultures by filtration through a 22-mm mesh gauze. The cells were resuspended in 250-μl buffer P1 of the QIAprep Spin Miniprep Kit (QIAGEN) in a 1.5-ml tube and homogenized using a teflon pestle. Minicircles were isolated by following the manufacturers' instructions for plasmid isolation. Isolated Minicircles were restricted using either HindIII, XbaI, or BglII and subsequently cloned into the correspondent pUC18 plasmid. Shotgun sequencing was performed on a Li-Cor 4200 sequencer and sequences thus obtained were assembled using Sequencher (GeneCodes). Circularity was verified using an inverse PCR approach.
Isolation of C. horridum Chloroplasts
C. horridum (100 to 200 mg) was resuspended in an ice-cold isolation buffer (0.4M mannitol, 1mM MgCl2, 10mM Na2EDTA, 0.25% polyvinylpyrrolidone 90, 50mM Tris-HCl pH 7.5), homogenized in a 5-ml potter and filtered through a 5-μm mesh gauze. The chloroplasts were pelleted from the filtrate at 3,400g, resuspended in a 150-μl isolation buffer, loaded on 30% (v/v) Percoll (Amersham) in isolation buffer and centrifuged at 4,000g for 10 min.
Isolation of C. horridum Nuclei
C. horridum cells were suspended and homogenized as for plastid preparation (mannitol concentration was increased to 0.7 M). The homogenate was filtered through a 22-μm mesh gauze and centrifuged at 2,500g to 3000g. The pellet was resuspended and loaded on 30% to 40% (v/v) Percoll (Amersham). After centrifugation for 10 min and 3,400g the gradient was fractionated, and fractions containing intact nuclei were identified microscopically.
Electron Microscopy
Ultrathin sections of trypsin-treated Prorocentrum micans were prepared as described (Kowallik 1971) and recorded in a Philipps EM 301 electron microscope as described (Herrmann et al. 1975).
In Situ Hybridization
In situ hybridizations were performed as described (McFadden 1991). The probes used for hybridizations were a 1,244-bp biotinylated probe spanning the coding region of the psbB minicircle, a 1,038-bp probe from the RuBisCO gene of C. horridum, and a 66-bp (5'-CCAGTGGTCGTTGGTTGGTGTGTGGTTGGTGGCGGTGGGGGTTGTGCCGGCGTGGTTGTTGTTGGG-3') fragment cloned from the high molecular plastid DNA from C. horridum. As a negative control, in situ hybridizations were done with the secondary antibody only. Hybrids were immunodetected for electron microscopy with gold-coupled antibiotin antibodies and with Cy2-coupled antibiotin antibodies for confocal laser scanning microscopy.
Results and Discussion
Minicircles from Ceratium horridum were enriched from total dinoflagellate DNA by a procedure that exploits their plasmidlike size and conformation (figure 1, lane B). The most prominent circular minicircle size populations in C. horridum migrate in the range of 3.5 to 4 kb in agarose gels (figure 1, lane B), similar to the size range of cloned minicircles previously found for Heterocapsa triquetra, Amphidinium operculatum, and Amphidinium carterae (Zhang, Green, and Cavalier-Smith; Barbrook and Howe 2000; Zhang et al. 2001; Barbrook et al. 2001; Hiller 2001).
FIG. 1. DNA profiles from cell fractions of Ceratium horridum. Lane A: Lambda DNA cut with EcoRI/HindIII. Lane B: Minicircle purification of C. horridum cells. Lane C: Total-cell DNA, including minicircles, most of which band around 4 kb. Lane D: DNA from intact plastids isolated from Percoll pellet and subsequently lysed with CTAB. Lane E: Linearized pUC18 added as control. Lane F: Same fraction as lane D after DNase I treatment prior to CTAB lyses. Lane G: Lambda size marker. Lanes C* to G* Southern blot from lanes C to G and hybridization using the digoxiginin-labeled psbB coding region of Ceratium horridum minicircle as a probe. Lane H: DNA from isolated nuclei. Lane H*: Southern blot from Lane H, hybridized with a minicircle probe (psbB probe). Lane I: Lambda size marker. Lane K: DNA from isolated plastids. Lane L: DNA from the supernatant of the plastid isolation, harboring DNA from nuclei, mitochondria, and broken plastids. Lanes I*, K*, and L*: Southern blot of lanes I to K, hybridization with a digoxiginin-labeled probe isolated from the high-molecular-weight plastid DNA. Lanes I+, K+, and L+: Southern blot of lanes I-K, hybridized with a minicircle-specific psbB probe. MP indicates minicircle probe
Sequence comparisons showed that the C. horridum minicircles encode genes of cyanobacterial origin, homologs of which are plastome-localized in other algae. By cloning of isolated minicircles, we detected genes and gene fragments for plastid-related psaA, psaB, psbB, psbC, psbD, psbE, petB, ycf16, and ycf24. Whereas the genes psaA, psaB, psbB, psbC, psbD, psbE, and petB are contained in all plastid genomes sequenced to date, ycf16 and ycf24 are specific to plastid genomes from non–green algal lineages (Stoebe, Martin, and Kowallik. 1999). All minicircle-encoded proteins from peridinin-containing dinoflagellates isolated here and previously (Zhang, Green, and Cavalier-Smith 1999; Barbrook and Howe 2000; Zhang, Cavalier-Smith, and Green; Barbrook et al. 2001; Hiller 2001) share convincing amino acid similarity with their homologs from red algal chloroplast genomes (and plastids derived from red algae via secondary endosymbiosis). A significant portion of minicircles from C. horridum encode one protein per plasmid, except ycf16/ycf24, which are located on a single minicircle in tandem. As with other dinoflagellates (Barbrook et al. 2001; Hiller 2001), several C. horridum minicircles did not contain a recognizable coding region. We found no typical transit peptide coding regions contiguous with the minicircle open reading frames, consistent with findings from previous studies.
To physically localize the minicircles within the cell, intact plastids were purified from C. horridum by cell fractionation, and their nucleic acids were isolated. Purified plastids did not contain detectable amounts of minicircle DNA (figure 1, lanes D and F), whereas DNA from whole cells did (figure 1, lane C). This was confirmed by Southern hybridization with a minicircle-specific probe (figure 1, lanes C* to G*). Instead, the plastids release a high-molecular-weight DNA upon lysis (figure 1, lanes D and F). To further pinpoint the localization of these minicircles, we isolated nuclei from C. horridum. Minicircles can be observed in large numbers in the nuclear DNA (figure 1, lane H), which was confirmed by Southern hybridizations with a minicircle probe (figure 1, lane H*).
We tried to demonstrate the presence of minicircles in the dinoflagellate plastids by in situ hybridization and imaging with confocal laser scanning microscopy, using a 1,244 bp long stretch from the coding region of minicircle psbB from C. horridum (fig. 2f) as well as a fragment from high-molecular-weight plastid DNA (fig. 2b) as probes. An additional control was performed in which the secondary antibody was tested for unspecific reactions (not shown), showing the autofluorescence of the plastids. As shown in figure 2f, a minicircle probe labeled the nucleus in the space between the chromosomes. Furthermore, the high-molecular-weight plastid probe hybridized in the plastid (fig. 2b). That the psbB probe specifically labeled the interchromosomal regions of the nucleus was confirmed by conventional in situ hybridization techniques and visualization with electron microscopy (fig. 2g). As a further control, a probe spanning the C. horridum–specific RuBisCo labeled a chromosomal region, whereas a nonhomologous probe (phycoerythrin from cryptomonads) showed no significant hybridization (data not shown). Taken together, these findings indicate that the sequence of the psbB minicircle is primarily, and probably exclusively, localized extrachromosomal in the nucleus and not in the plastid of C. horridum.
FIG. 2. In situ hybridizations. (a) and (e) show phase contrast images of (b) and (f), respectively. In situ hybridizations are shown in (b) using a plastid genome–derived probe and a psbB probe in (e). (c) Visualizes a mechanically stressed C. horridum cell releasing the nucleus and plastids. These organelles (P = plastid; N = nucleus), before its separation, are shown in (d). (g) Hybridization using the minicircle-encoded psbB coding region of C. horridum as probe and visualizing the hybrids with the electron microscope. Numerous hybrids in the cell nucleus (N) and one in the plastid (P) are shown by arrows. The bar = 1 μm
In a recent study, minicircle-encoded mRNA for psbA was detected in the chloroplast of the genus Symbiodinium by in-situ hybridization (Takishita et al. 2003). These data, together with the results of Zhang, Green, and Cavalier-Smith (1999) on Prorocentrum, from which no 23S rRNA was detected, raises the possibility that in some dinoflagellate species, minicircles may be plastid located, whereas in others, the chloroplast genes have moved into the cell nucleus (Cavalier-Smith 2003). This hypothesis is complicated by the finding that in some dinoflagellate species, plastid DNAs has been identified with molecular sizes between 114 to 125 kb, consistent with plastid DNAs from other phototrophic organisms (Boczar, Liston, and Catollico 1991).
This conundrum of findings prompted us to start again from scratch to identify the genuine dinoflagellate plastid DNA from C. horridum. A plastid-localized DNA can be clearly seen inside the dinoflagellate stroma by traditional transmission electron microscopy of, for example, Prorocentrum micans (fig. 3). To find the plastid DNA in Ceratium, intact plastids were isolated from C. horridum. Isolated plastids that were lysed after treatment with DNase I released a high-molecular-weight DNA that is DNase sensitive (figure 1, lane K). C. horridum minicircles do not hybridize to the high-molecular-weight DNA isolated from such purified plastids (figure 1, lane K+), but do hybridize to the supernatant of this preparation, which contains broken nuclei (figure 1, lanes L and L+). Thus, the C. horridum plastid DNA is apparently distinct from minicircle DNA. However, in Southern experiments, a probe cloned from the plastid DNA fraction (see Methods) hybridized with the high-molecular-plastid DNA (figure 1, lane K*), as well as with the DNA isolated from the supernatant of the plastid preparation, which contains DNA from the other organelles and from broken plastids (figure 1, lane L*). Finally, in situ hybridization experiments with the probe for the high-molecular-weight DNA confirmed its plastid location (fig. 2b). Nevertheless, the chloroplast-located DNA of C. horridum remains mysterious. Sequences obtained from this genome do not show any coding region or significant similarities to other data base entries.
FIG. 3. Chloroplast nucleoid (black arrow) of Prorocentrum micans. The DNA was made visible after tryptic digestion of the stroma
Peridinin-containing dinoflagellates are clearly defined by their plastid morphology: a triple membrane envelope and thylakoid stacks in layers of three. Because our findings suggested that there may be lineage-specific differences of minicircle localization among peridinin-containing dinoflagellates, we reinspected the plastid morphotype of C. horridum. One of us (KVK) noticed that differences are obvious, because the C. horridum plastid shows stacks of thylakoids in rows of two in contrast to the stacks of three typical for other peridinin-containing dinoflagellates. Furthermore, the thylakoid lumen is opaque as in cryptophyte plastids. However, the plastids are devoid of phycobiliproteins but contain peridinin.
Our findings indicate that the organization of the dinoflagellate C. horridum minicircles is surprising because it is a nuclear extrachromosomal and not a plastid genome organization, although in evolutionary terms, it ultimately stems from the three-membrane-bounded plastid of these protists. DNA does exist in plastids from C. horridum, but it is distinct from the nuclear minicircles previously thought to exclusively represent the chloroplast DNA.
Where are plastid proteins expressed? If they are expressed in the cytoplasm of the host, how do they find their final target sites? If the minicircles of C. horridum (but possibly not those from other dinoflagellates [Takishita et al. 2003]) represent only vehicles for transferring plastid DNA into the cell nucleus, one would have to postulate that the plastid-located DNA does indeed encode a set of proteins known from other chloroplasts (Martin et al. 2002). Our preliminary analysis of the plastid genome shows yet no indications of such encoded genes. However, coding regions could be hidden and recognized only after an editing machinery, known from chloroplasts of higher plants (Bock 2000), has formed an mRNA that could be reasonably translated. If minicircles are the transportation molecules into the nucleus, one would suggest that dinoflagellates show an extended rate of plastid gene migration into the host. Hints for this comes from a recent publication showing that 15 genes that are limited to the plastid genome in all other eukaryotic phototrophs are nucleus encoded in a peridinin-containing dinoflagellate (Hackett et al. 2004).
As most of the minicircle-encoded proteins are known to be expressed in chloroplasts of other groups, one could argue that RNA import might provide the plastid with RNA messages. Such a roughly analogous situation is known, for example, in the case of tRNA import into mitochondria of trypanosomes (Schneider 2001). If the C. horridum nuclear minicircles specify chloroplast proteins, then these would be expected to encode a preprotein, harboring a signal sequence and a transit peptide to manage the transport across the plastid surrounding membranes. However, the coding region does not show any N-terminal extension that could serve as a signal sequence. Again, topogenic signals could be hidden in the genomic sequence and uncovered in the mRNA after RNA editing. Notably, editing can be observed in C. horridum minicircles (Zauner and Maier, unpublished data).
Dinoflagellates are unusual in many respects, as shown, for example, by their chromosome architecture and by the finding that peridinin-containing dinoflagellates express an unusual form II RuBisCo (Morse et al. 1995, Whitney, Shaw, and Yellowlees 1995, Rowan et al. 1996). C. horridum furthermore differs from other phototrophic dinoflagellates investigated thus far by its noncanonical plastid morphology, despite of the existence of peridinin, and by its minicircle localization, assuming that all other dinoflagellate minicircles described thus far are localized in the plastid. Thus, peridinin-containing dinoflagellates could be a heterogenous subgroup of alveolates, in which nature might have found a genetic playground.
Supplementary Material
Sequences are deposited in GenBank under accession numbers AF490356 to AF490364 and AF490368.
Acknowledgements
The authors would like to thank Marianne Johannsen (Marburg) and Dr. Franz Grolig (Marburg) for their help in electron and confocal laser scanning microscopy. Supported by the Deutsche Forschungsgemeinschaft.
Literature Cited
Barbrook, A. C., and C. J. Howe. 2000. Minicircular plastid DNA in the dinoflagellate Amphidinium operculatum. Mol. Gen. Genet. 263:152-158.
Barbrook, A. C., H. Symington, R. E. R. Nisbet, A. Larkum, and C. J. Howe. 2001. Organisation and expression of the plastid genome of the dinoflagellate Amphidinium operculatum. Mol. Genet. Genomics. 266:632-638.
Bock, R. 2000. Sense from nonsense: how the genetic information of chloroplasts is altered by RNA editing. Biochemie 82:549-557.
Boczar, B. A., J. Liston, and R. A. Catollico. 1991. Characterization of satellite DNA from three marine dinoflagellates (Dinophyceae): Glenodinium sp. and two members of the toxic genus Protogonyaulax. Plant Physiol. 97:613-618.
Cavalier-Smith, T. 2003. Genomic reduction and evolution of novel genetic membranes and protein-targeting machinery in eukaryote-eukaryote chimaeras (meta-algae). Philos. Trans. R. Soc. Lond. B Biol. Sci. 358:109-133.
Douglas, S. E. 1998. Plastid evolution: origins, diversity, trends. Curr. Opin. Genet. Dev. 8:655-661.
Hackett, J. D., H. S. Yoon, M. B. Soares, M. F. Bonaldo, T. L. Casavant, T. E. Scheetz, T. Nosenko, and D. Bhattacharya. 2004. Migration of the plastid genome to the nucleus in a peridinin dinoflagellate. Curr. Biol. 14:213-218.
Herrmann, R. G., H. J. Bohnert, K. V. Kowallik, and J. M. Schmitt. 1975. Size, conformation and purity of chloroplast DNA of some higher plants. Biochim. Biophys. Acta 378:305-317.
Hiller, R. G. 2001. ‘Empty’ minicircles and petB/atpA and psbD/psbE (cytb559 alpha) genes in tandem in Amphidinium carterae plastid DNA. FEBS Lett. 505:449-452.
Kowallik, K. V. 1971. The use of proteases for improved presentation of DNA in chromosomes and chloroplasts of Prorocentrum micans (Dinophyceae). Arch. Mikrobiol. 80:154-165.
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.
Martin, W., B. Stoebe, V. Goremykin, S. Hausmann, M. Hasegawa, and K. V. Kowallik. 1998. Gene transfer to the nucleus and the evolution of chloroplasts. Nature 393:162-165.
McFadden, G. I. 1991. In situ hybridization techniques: molecular cytology goes ultrastructural. Pp. 219–255 in J. L. Hall, C. R. Hawes, eds. Electron microscopy of plant cells. Academic Press, London.
McFadden, G. I. 1999. Chloroplasts: ever decreasing circles. Nature 400:119-120.
Morse, D., P. Salois, P. Markovic, and J. W. Hastings. 1995. A nuclear-encoded form II RuBisCO in dinoflagellates. Science 268:1622-1624.
Rowan, R., S. M. Whitney, A. Fowler, and D. Yellowlees. 1996. RuBisCO in marine symbiotic dinoflagellates: form II enzymes in eukaryotic oxygenic phototrophs encoded by a nuclear multigene family. Plant Cell 8:543-545.
Schneider, A. 2001. Unique aspects of mitochondrial biogenesis in trypanosomatids. Int. J. Parasitol. 31:1403-1415.
Stoebe, B., and K. V. Kowallik. 1999. Gene-cluster analysis in chloroplast genomics. Trends Genet. 15:344-347.
Stoebe, B., W. Martin, and K. V. Kowallik. 1998. Distribution and nomenclature of protein-coding genes in 12 sequenced chloroplast genomes. Plant Mol. Biol. Reptr. 16:243-255.
Takishita, K., M. Ishikura, K. Koike, and T. Maruyama. 2003. Comparison of phylogenies based on nuclear-encoded SSU rDNA and plastid-encoded psbA in the symbiotic dinoflagellate genus Symbiodinium. Phycologia 42:285-291.
Whitney, S. M., D. C. Shaw, and D. Yellowlees. 1995. Evidence that some dinoflagellates contain a ribulose-1,5-bisphosphate carboxylase/oxygenase related to that of the alpha-proteobacteria. Proc. R. Soc. Lond. B Biol. Sci. 259:271-275.
Zhang, Z., B. R. Green, and T. Cavalier-Smith. 1999. Single gene circles in dinoflagellates chloroplast genomes. Nature 400:155-159.
Zhang, Z., T. Cavalier-Smith, and B. R. Green. 2001. A family of selfish minicircular chromosomes with jumbled chloroplast gene fragments from a dinoflagellate. Mol. Biol. Evol. 18:1558-1565.(T. Laatsch*,1, S. Zauner,)