subcellular localization of minicircle dna in the dinoflagellate amphidinium massartii

Subcellular localization of minicircle DNA in thedinoflagellate Amphidinium massartiiSatomi Owari,1 Aiko Hayashi3 and Ken-ichiro Ishida2,3*1Graduate School of Life and Environmental Sciences, 2Faculty of Life and Environmental Sciences, University ofTsukuba, Tsukuba, and 3Graduate School of Natural Science and Technology, Kanazawa University, Kanazawa, Japan

SUMMARYPeridinin-containing dinoflagellates have small circularDNA molecules called minicircle DNAs, each of whichencodes one, or occasionally a few, plastid proteins orribosomal RNA. Dinoflagellate minicircle DNA is com-posed of two parts: a gene-coding sequence and anon-coding sequence that consists of several variableand core regions. The core regions are identical amongthe minicircle DNAs with different genes within aspecies or strain. Because such structure is very differ-ent from those of well known plastid DNAs, many func-tional and evolutionary questions have been raised forthe minicircle DNAs, and several studies that focus onanswering those questions are underway. However, thelocalization of minicircle DNA is still controversial:several lines of indirect evidence have implied plastidlocalization, whereas the nuclear localization ofminicircle DNA has also been suggested in a species. Inorder to understand the evolution and function ofminicircle DNA, it is important to know its preciselocalization. In this study, we sequenced two typicalminicircle DNAs, one encodes psbA and the otherencodes 23S rRNA genes, from an Amphidiniummassartii strain (TM16). To determine the subcellularlocalization of these minicircle DNAs, we performedDNA-targeted whole cell fluorescence in situ hybridiza-tion with A. massartii minicircle DNA-specific probesand demonstrated that minicircle DNAs were present inplastids. This study provides the first direct evidencefor the plastid localization of dinoflagellate minicircleDNAs.

Key words: Amphidinium massartii, dinoflagellates, insitu hybridization, minicircle DNA, plastid DNA.

INTRODUCTION

Plastids are known to have originated from a singleendosymbiotic cyanobacterium that was incorporatedinto a eukaryotic host cell (Reyes-Prieto et al. 2007;Gould et al. 2008). The genome of the endosymbioticcyanobacterium has been drastically reduced and

evolved into plastid DNA, which is usually circular, andencodes approximately several hundred genes forplastid proteins and ribosomal RNAs (rRNAs) (Palmer1990; Reith & Munholland 1993; Simpson & Stern2002; Ohta et al. 2003). In most peridinin-containingdinoflagellates, however, plastid proteins and rRNAs areencoded in uncommon DNA molecules. In 1999,Zhang et al. first reported a small circular DNA mol-ecule, termed minicircle DNA, from a peridinin-containing dinoflagellate Heterocapsa triquetra. Sincethen, minicircle DNAs have been reported fromseveral peridinin-containing dinoflagellate genera,including Heterocapsa (Zhang et al. 1999, 2002),Amphidinium (Barbrook & Howe 2000; Hiller 2001;Zhang et al. 2002), Proroceratium (Zhang et al. 2002),Symbiodinium (Moore et al. 2003), Ceratium (Laatschet al. 2004), Adenoides (Nelson & Green 2005), andAlexandrium (Iida et al. 2009, 2010), though aperidinin-containing species, Lingulodinium polyedrum(= Gonyaulax polyedra), seems to have approximately50–150 kbp of circular DNA molecules that encodeplastid genes (Wang & Morse 2006).

Dinoflagellate minicircle DNAs are 2–6 kbp circularDNA molecules, each of which is composed of uniquenon-coding region(s) and coding region(s) encoding asingle gene (or occasionally two or three genes) thatis usually found in plastid DNA (Zhang et al. 1999;Barbrook & Howe 2000). The sequence in a non-codingregion contains one or more core sequence(s), which isalmost identical among the minicircle DNAs of differentgenes within a species or a cultured strain, but thereare no sequence similarities between closely relatedspecies or strains (Zhang et al. 1999, 2002; Barbrook& Howe 2000; Barbrook et al. 2001; Hiller 2001;Nelson & Green 2005). The core region has beenspeculated to function as the noncanonical promoter(Zhang et al. 2002; Moore et al. 2003; Howe et al.2008; Nisbet et al. 2008) and replication origin, and to

*To whom correspondence should be addressed.Email: [email protected] Editor: I. Mine.Received 28 February 2013; accepted 28 June 2013.doi: 10.1111/pre.12037

Phycological Research 2014; 62: 1–8

bs_bs_banner

© 2013 Japanese Society of Phycology

anchor minicircle DNAs to the membrane (Howe et al.2008).

Although minicircle DNAs have been sequencedfrom many peridinin-containing dinoflagellates, thesubcellular localization of those DNA molecules is stillcontroversial; no direct evidence for plastid localizationof minicircle DNA has been demonstrated. Severallines of indirect evidence have implied, however, thatminicircle DNAs are present in plastids; sequenceanalyses have not revealed any targeting sequence fordelivering the gene products to the plastids (Howe et al.2008); mRNAs of psbA were detected in plastids of adinoflagellate Symbiodinium sp. by in situ hybridization(Takishita et al. 2003); and in Amphidinium carteraeand L. polyedrum, PsbA protein synthesis was inhibitedby chloramphenicol, which is known to block proteinsynthesis in the plastids of other organisms (Wang et al.2005). On the other hand, on the basis of cell frac-tionation, Southern blotting, and in situ hybridization,Laatsch et al. (2004) suggested that minicircle DNAsare present outside the plastids in Ceratium horridum.

To understand the evolution and function ofminicircle DNAs, it is important to know the truelocalization of these DNA molecules. In terms ofminicircle DNAs, Amphidinium carterae is one of themost studied peridinin-containing dinoflagellates, andmany minicircle DNA sequences have been reported inthis species (Howe et al. 2008). However, it has ahighly reticulate plastid (Murray et al. 2004) thatmakes determining the localization of minicircle DNAsdifficult under a fluorescent microscope. For thepresent study, we selected Amphidinium massartii, theclosest relative to A. carterae having a large centralplastid with radially arranged thick lobes (Murray et al.2004), which would aid in microscopic observation.

Here, we sequenced full minicircle DNAs of psbAand the 23S rRNA gene from a cultured strain ofA. massartii and determined the structure of minicircleDNA, including the core region. In order to determinethe subcellular localization of minicircle DNAs in theA. massartii cell, we performed a DNA-targeted wholecell fluorescence in situ hybridization (FISH) withnewly made specific probes, and we found that theA. massartii minicircle DNAs are likely to be present inthe plastids.

MATERIALS AND METHODS

Algal culture

A culture of the dinoflagellate Amphidinium massartii(TM16), which had been established from a samplecollected from Shirahama Beach (Wakayama prefec-ture, Japan) by Dr M. Tamura (currently at Akita Uni-versity, Japan), was kindly provided by Dr T. Horiguchi(Hokkaido University, Japan). The phylogenetic position

of this strain was confirmed by our preliminaryphylogenetic analyses by using the 28S rRNA genesequences. The alga was cultured in Daigo IMKmedium (Wako Pure Chemical, Osaka, Japan) under50 μmol photon m−2 s−1 with a 14-h light/10-h darkcycle.

DNA extraction

A well-grown culture near the stationary phase wasused for DNA extraction because the copy number ofminicircle DNA is expected to be high (Koumandou &Howe 2007). DNA extraction was performed accordingto the standard phenol extraction method of Koumandouand Howe (2007), with slight modifications. Cellswere collected by a centrifugation at 2500 g for 5 min at4°C and resuspended in TEN buffer (20 mM Tris-HCl,0.5 mM ethylenediaminetetraacetic acid, 0.1 M NaCl,pH 8) with 1% (w/v) sodium dodecyl sulfate and500 μg mL−1 Proteinase K. The mixture was incubatedfor 1 h at 50°C, and DNA was extracted by the phenolextraction method. The DNA was precipitated by ethanoland finally dissolved in sterilized distilled water.

Sequence analysis

The nuclear-encoded 28S rRNA gene sequence wasamplified from the genomic DNA by a polymerase chainreaction (PCR). The primers used in this study are listedin Table 1. The PCR was performed using the Ex taqpolymerase (Takara Bio, Shiga, Japan) with the follow-ing conditions: 94°C for 2 min; then 35 cycles of 94°Cfor 30 s, 50°C for 30 s, and 72°C for 30 s; followed bya single extension at 72°C for 7 min. An amplifiedproduct was cloned into the pGEM T-easy Vector(Promega, Madison, WI, USA) and sequenced with aGenetic Analyzer 3100 (Applied Biosystems, FosterCity, CA, USA) by using Big Dye Terminator v3.1 CycleSequencing Kit (Applied Biosystems).

To obtain sequences of minicircle DNA genes, psbAand the 23S rRNA gene, we used degenerate primersmade on the basis of dinoflagellate sequences(Table 1). PCR analyses were carried out using Ex taqpolymerase (Takara Bio) with the following conditions:94°C for 2 min; followed by 35 cycles of 94°C for 30 s,47°C–55°C for 30 s, and 72°C for 3 min; and a singleextension at 72°C for 7 min. The amplified DNA frag-ments were sequenced as described above. To obtainfull-length minicircle DNA sequences, outward primerswere then constructed on the basis of the obtainedpsbA and 23S rRNA gene sequences (Table 1). PCRand sequencing were performed as described above. Allsequences acquired in this study were deposited inGenBank with accession numbers: AB818490 (psbA


2 S. Owari et al.

minicircle DNA), AB818491 (23S rRNA geneminicircle DNA), and AB818952 (28S rRNA genefragment).

Similarity search of the non-coding region wasperformed using YASS (Noé & Kucherov 2005) withstandard parameters. Secondary structure of the coresequence was predicted by using RNAfold with stand-ard parameters (Hofacker et al. 1994).

Probe synthesis

Digoxigenin-labeled DNA probes for the 28S rRNAgene, minicircle DNAs psbA, 23S rRNA gene, and thecore sequence were prepared from the plasmid cloneswith specific primers (Table 1) by using PCR DIGProbe Synthesis Kit (Roche Diagnostics, Mannheim,Germany) according to the manufacturer’s instructions.For a negative probe, human tissue type plasminogenactivator (tPA) probe was amplified using the controlplasmid and primers from the kit. The probes weredenatured just before use by incubation at 90°C for4 min.

Southern blotting

Genomic DNA of A. massartii was electrophoresed on a1% agarose gel in TAE buffer and neutrally transferredto a Hybond-N + nylon membrane (GE HealthcareBioscience, Piscataway, NJ, USA) by capillary blottingaccording to the manufacturer’s instructions. Each ofthe probes was hybridized and detected using DIGNucleic Acid Detection Kit (Roche Diagnostics) accord-ing to the manufacturer’s instructions.

DNA-targeted whole-cell FISH

Cells were collected by low-speed centrifugation(500 g, 20°C for 5 min). The cells were fixed with 3%paraformaldehyde in PHEM-NaCl buffer (60 mMPIPES, 25 mM HEPES, 10 mM EGTA, 2 mM MgCl2,pH 6.9, 3% NaCl) for 15 min on ice. The cells weredehydrated and autofluorescence originating from chlo-rophylls was diminished in moderation by using anethanol series of 35%, 70%, 85%, and 85% for 5 minon ice, after which the cells were hydrated using anethanol series of 70% and 35% for 5 min on ice andtwice in 2 × standard saline citrate (SSC) (300 mMNaCl, 30 mM trisodium citrate, pH 7.0) for 5 min atroom temperature. After 100 μg mL−1 RNase A (RocheDiagnostics) was added, the cells were settled ontocoverslips coated with 0.1% (w/v) poly-L-lysine solution(Wako Pure Chemical), and incubated for 2 h at 37°C ina moist chamber to digest RNA. The cells were perme-ated three times with PBS-T (0.2% (v/v) Tween-20 inPBS (13 mM NaCl, 7 mM Na2HPO4, 3 mM NaH2PO4,pH 7.2)) for 5 min at room temperature and overnightat 4°C. The cells were washed three times in 2 × SSC atroom temperature.

The cells were incubated in a hybridization buffer(50% deionized formamide, 2 × SSC, 50 mM sodiumphosphate; pH 7) for 1.5 h at 37°C in a moist chamber.After removing the buffer, 100 μL of fresh hybridizationbuffer containing 10% dextran sulfate was mixed with1 µL of the denatured DIG-labeled DNA probe (pre-pared as described above) and applied to eachcoverslip. The coverslips on a glass slides were coveredwith another coverslip and incubated for 7 min at 85°C

Table 1. Primers for sequence acquisition and probe synthesis

Primer name Primer sequence (5′–3′) Application

Primers for sequence acquisitionDIR† ACCCGCTGAATTTAAGCATA LSU rRNA gene forward28-1483R† GCTACTACCACCAAGATCTGC LSU rRNA gene reverseDinopsbAF1i YTWTAYATHGGWTGGTTYGG Degenerate psbA inward forwardDinopsbAR1i TCRTGCATWACYTCNATNCC Degenerate psbA inward reverseDinopabAF2o TGGCCWGTWATHGGWATHTGG Degenerate psbA outward forwardDinopabAR2o CKRATWCCATCDATATCDACTG Degenerate psbA outward reverseDino23SF1i TMRTRTTYATCAASCGACTGT Degenerate 23S rRNA gene inward forwardDino23SR1i AAGCCGACATCGAGGTGCMAA Degenerate 23S rRNA gene inward reverseDino23SF2o TACYCWAGGGATAACAGGCTTA Degenerate 23S rRNA gene outward forwardDino23SR2o CGTTCWATSTTTCATGCAGGTC Degenerate 23S rRNA gene outward reverse

Primers for probe synthesisAMlsuF1 GCACCAGCAACCAACTGATC nuclear LSU rRNA gene forward28-1483R† GCTACTACCACCAAGATCTGC nuclear LSU rRNA gene reverseAMpsbAR1 GCACGGTTGAGGATATCAGC minicircle psbA coding region forwardAMpsbAF1 TGGGGTTCTTTCGTTCAAAC minicircle psbA coding region reverseAMcoreF1 GGAAATAACCCCTAGACTTTAACGG minicircle non-coding core region forwardAMcoreR1 TAGTTTAACGTCATTGCGGACCAGA minicircle non-coding core region reverse

†From Murray et al. (2004).


3Localization of minicircle DNA

to keep the probes denatured and to denature thegenomic DNAs, then placed overnight at 37°C forhybridization. Hybridized cells were washed three timesin 2 × SSC containing 50% formamide for 5 min at45°C and then three times in 2 × SSC for 5 min atroom temperature to remove unhybridized DNA probe.

For detection, the cells were treated with blockingbuffer (Roche Diagnostics) for 1 h at room temperatureand then incubated with an anti-digoxigenin antibodyproduced in a mouse (11333062910, 1:125 dilutionin blocking buffer; Roche) for 1 h at 37°C in a moistchamber. The cells were then washed with PBS-T,treated with an anti-mouse immunoglobulin G antibodyconjugated with fluorescein isothiocyanate (F0257,1:50 dilution in blocking buffer; SIGMA Aldrich, St.Louis, MO, USA) and washed with PBS-T. Theimmunostained cells were mounted with Slow-Fadeanti-fade reagent (Molecular Probes, Eugene, OR, USA)containing 0.5 μg mL−1 4′,6-diamidino-2-phenylindoledihydrochloride (DAPI). The samples were observedunder a DMRD epifluorescence microscope (LeicaMicrosystem, Wetzlar, Germany) equipped with aDP71 CCD camera (OLYMPUS, Tokyo, Japan). Imagesfor the samples treated with the psbA, core sequence,and tPA probes were taken with the same exposureconditions (ISO 800), whereas those treated with the28S rRNA gene probe were captured with a less sensi-tive exposure condition (ISO 400). The samples werealso observed under a LSM 710 confocal laser-scanningmicroscope (Carl Zeiss, Jena, Germany) using the Zensoftware.

RESULTS

Minicircle DNA sequences ofAmphidinium massartii

We newly sequenced complete minicircle DNAsequences encoding the psbA and 23S rRNA genesfrom a cultured strain of A. massartii (TM16) in order toconstruct probes to determine the localization ofminicircle DNAs in the cell. The size of the psbAminicircle DNA was 3569 bp (A + T content = 59.9%).The coding sequence was 1023 bp long and had anATG start codon and a TAG stop codon (Fig. 1a). Thesize of the 23S rRNA gene minicircle DNA was3922 bp (A + T content = 61.8%); however, the exactregion of the gene could not be determined like 23SrRNA gene minicircle DNAs of other dinoflagellates(Fig. 1a). In comparison of non-coding regions of thosetwo minicircle DNA sequences, we identified a 198-bpidentical region between the two minicircle DNAs,which was defined as the core region (Fig. 1b). The coreregion had many tandem A (2–6 bp) and T (2–5 bp)nucleotides, and its A + T content was 69.7%. The coreregion was predicted to form at least two stem-loops by

RNAfold centroid structure: a 59-nt imperfect stem-loop structure composed of three short (6–9 bp long)stems (solid arrows in Fig. 1b) and a 38-nt imperfectstem-loop structure with a 4-bp and a 7-bp stem (openarrows in Fig. 1b). As expected, the core sequenceshowed no similarity to any core sequences reportedfrom other dinoflagellates. Other regions of the non-coding sequences had no similarity between those twominicircle DNAs; however, dispersal distribution ofshort repeat sequences and short inverted repeatsequences was observed.

Localization of minicircle DNAs by FISH

To determine the localization of minicircle DNAs in thecell, we chose 932 bp of psbA and 169 bp of the coreregion sequence of A. massartii minicircle DNA as spe-cific probes. We also constructed a nuclear-encoded28S rRNA gene-specific probe (700 bp) as a positivecontrol and a tPA human gene specific probe (169 bp)as a negative control. The specificity of each probe wasconfirmed by total genome Southern blotting (Fig. 2).The nuclear 28S rRNA gene probe gave a strong signalfor the broad mass of nuclear genomic DNA at highmolecular weight (Fig. 2, lane 2). The psbA probeshowed a weak band at about 3.5 kb and a strong band

Fig. 1. Map of psbA and 23S rRNA gene minicircle DNAs of

Amphidinium massartii. (a) Gray areas represent coding regions;

black areas represent cores of non-coding regions; white areas

represent nonconserved regions of non-coding regions. N and C

are the N-terminus and C-terminus, respectively, of PsbA protein.

5′ and 3′ are the 5′ end and 3′ end of the 23S rRNA gene. (b) The

198-bp core consensus sequence in the non-coding region of

psbA and 23S rRNA gene minicircles. Arrows show putative

stem-loop structures by RNAfold (solid arrows: a 59-nt imperfect

stem-loop structure, open arrows: a 38-nt imperfect stem-loop

structure).


4 S. Owari et al.

at about 5 kb (Fig. 2, lane 3), which probably corre-spond to linear monomers and relaxed monomericcircles, respectively. Another minicircle specific probe,the core region probe, gave a weak band at about 3 kband several strong bands at about 4 kb, 5 kb, and 6 kb(Fig. 2, lane 4). The probes for the nuclear encoded28S rRNA gene, minicircle DNA psbA, and theminicircle DNA core region were sufficient to identifyspecific bands, which show that each probe can detectspecific loci in A. massartii. In contrast, no signal wasdetected from the tPA probe, indicating that this probesis suitable for a negative control (lane 1 in Fig. 2).

Using these probes, we used FISH analysis onA. massartii cells. The tPA probe (negative control) gaveno signal at all (Fig. 3a), whereas the nuclear 28S rRNAgene probe (positive control) gave a signal as a brightspot near the edge of the nucleus in 93% of observedcells (Fig. 3b).

Epifluorescence microscopy revealed that the probefor psbA (Fig. 3c) gave spotted signals scattered withinthe chlorophyll autofluorescence, in 97% of observedcells. The probe for the core region (Fig. 3d) gavethe similar spotted signals throughout the plastidautofluorescence in 96% of observed cells, though theseveral signals sometimes gathered and made a largecluster in the plastid autofluorescence (Fig. 3d). Noneof the signals merged with DAPI staining, which labelsthe nucleus.

Fluorescence in situ hybridization observations usinga confocal laser-scanning microscope enabled us toidentify the localization of minicircle DNA in detail. Thecore probe resulted in spotted signals that were actuallylocated along the outer edge of the autofluorescence,

between thylakoids and the cytoplasm (Fig. 4), indicat-ing that minicircle DNA is present at the periphery ofthe plastid. Similar results were obtained with the psbAprobe (data not shown).

DISCUSSION

Minicircle DNA sequences for psbA and23S rRNA genes

We have sequenced two minicircle DNAs from the dino-flagellate A. massartii, one encodes psbA and the otherencodes 23S rRNA gene. These sequences demon-strated the typical ‘one gene-one circle’ organizationof dinoflagellate minicircle DNA that contains coreregion(s) in the non-coding sequence. The codingsequences of both minicircle DNAs were homologouswith those of other dinoflagellates, and the non-codingsequences (including the core region) had no similaritywith those of any other dinoflagellates. Other features ofA. massartii minicircle DNAs, such as the molecularsize, AT content, and start and stop codons, weresimilar to those of other dinoflagellate minicircle DNAs(Zhang et al. 1999, 2002; Barbrook & Howe 2000;Hiller 2001; Nisbet et al. 2004; Nelson & Green 2005;Barbrook et al. 2006; Iida et al. 2009).

Core regions are known to show considerable vari-ation among species and strains, and they containseveral direct and inverted repeats in many cases(Zhang et al. 1999, 2002). We found a core region inthe non-coding sequence of A. massartii minicircleDNA that was predicted to contain at least two stem-loop structures. Although the significance of thoserepeats and the function of core regions are stillunknown, the core region of A. massartii minicircleDNA has normal characteristics. We therefore think thatthe A. massartii minicircle DNA can be used to studysubcellular localization as a representative minicircleDNA in dinoflagellates.

Localization of minicircle DNAs inA. massartii cells

Our FISH technique using the minicircle DNA-specificprobes gave clear spotted signals within theA. massartii plastid. These signals are not likely to benonspecific signals or artifacts because these probeswere highly specific as confirmed by total genomeSouthern blotting. In addition, our in situ hybridizationprocedure worked well for A. massartii cells, as shownby the positive and negative control probes (Fig. 3a,b).The single dot signal at the periphery of the nucleusgiven by the positive control (the nuclear 28S rRNAgene probe) is the expected positive result, because itis the same profile as the 28S rRNA gene localizationshown by FISH in another dinoflagellate Prorocentrum

Fig. 2. Total genome Southern blotting for A. massartii. Lane 1:

tPA-specific probe (negative control). Lane 2: nuclear encoded

28S rRNA gene probe which hybridized with nuclear DNA (indi-

cated by a bracket). Lane 3: minicircle DNA encoded psbA probe

showing a weak band at about 3.5 kb and a strong band at about

5 kb, which are probably linear monomers and relaxed monomeric

circles, respectively (arrowheads). Lane 4: minicircle encoded

core region probe that gave a weak band at about 3 kb and several

strong bands at about 4 kb, 5 kb, and 6 kb (arrowheads) which

indicates the presence of several minicircle DNAs with different

genes.



micans (Geraud et al. 1991). Therefore, we concludedthat our FISH results from the minicircle DNA-specificprobes indeed show the localization of minicircle DNAsin A. massartii cells.

Localization of minicircle DNAs in dinoflagellatecells has been debated since the discovery of dinoflag-ellate minicircle DNA. Several circumstantial lines ofevidence have indicated plastid localization of the

Fig. 3. Fluorescence in situ hybridization (FISH) images with specific probe for controls and minicircle DNAs in Amphidinium massartii.

(a) Unhybridized probe specific for tPA could not detect any signal as a control. (b) Localization of a nuclear gene detected by a probe

specific for the 28S rRNA gene in the nucleus as a control. (c) Localization of the coding region of minicircle DNAs detected by a specific

probe for psbA in the plastid. (d) Localization of a core region detected by a specific probe for the core sequence in the plastid. (1)

Fluorescein isothiocyanate (FITC) images (green); (2) 4′6′-diamidino-2-phenylindole dihydrochloride (DAPI) stained images (blue); (3)

Plastid chlorophyll autofluorescence images (red); (4) Merged images of FITC, DAPI, and chlorophyll pictures. Scale bar, 5 μm.

Fig. 4. Localization of minicircle DNAs by FISH images by using confocal laser-scanning microscopy. Consecutive serial sections of

minicircle DNA FISH images by using a core region probe for minicircle DNAs. The cell images were acquired in 1-μm intervals from the

center to the ventral side. Fluorescein isothiocyanate images (green), DAPI stained images (blue), and plastid chlorophyll autofluorescence

images (red) are merged. Scale bar, 5 μm.


6 S. Owari et al.

minicircle DNA. However, the localization of minicircleDNA in dinoflagellate cells has remained unclear. Thepresent study provides the first direct evidence forplastid localization of minicircle DNAs in a peridinin-containing dinoflagellate. The fragmentation of plastidDNA into minicircle DNAs would have occurred withinthe plastid in the peridinin-containing dinoflagellatelineage. However, we still cannot rule out the possibilitythat minicircle DNA is present in the nucleus ofC. horridum (Laatsch et al. 2004) because we did notstudy C. horridum in the present study.

Our FISH observations with epifluorescence and con-focal laser-scanning microscopy indicated a spotty dis-tribution of minicircle DNAs in the plastids. Confocallaser-scanning microscopy further revealed that thespots of minicircle DNAs are actually present at theperiphery of the plastids. This distribution pattern ofminicircle DNA indicates that the copies of minicircleDNA are not evenly distributed, but rather formnucleoid-like aggregations in the plastids and may beanchored to the plastid envelope. It is not clear how thecopies of minicircle DNA are packed into the aggrega-tion or how it is anchored to the envelope membrane. Inhigher plant plastids, a DNA-binding protein, PENDprotein, is known to be involved in anchoring plastidDNA to the plastid envelope (Sato et al. 1998; Sato &Ohta 2001). It would be interesting to investigatewhether there is any DNA-binding protein like PEND inthe peridinin-containing dinoflagellate plastid, and ifthe core region or the stem-loop structure in the non-coding sequence is involved in it.

ACKNOWLEDGMENTS

We thank Dr M. Tamura and Dr T. Horiguchi (HokkaidoUniversity, Japan) for providing the culture strainA. massartii (TM 16). We thank Dr Y. Hirakawa(Tsukuba University, Japan) for helpful suggestions.

REFERENCES

Barbrook, A. C. and Howe, C. J. 2000. Minicircular plastidDNA in the dinoflagellate Amphidinium operculatum. Mol.Gen. Genet. 263: 152–8.

Barbrook, A. C., Symington, H., Nisbet, R. E. R., Larkum, A.and Howe, C. J. 2001. Organisation and expression ofthe plastid genome of the dinoflagellate Amphidiniumoperculatum. Mol. Gen. Genet. 266: 632–8.

Barbrook, A. C., Santucci, N., Plenderleith, L. J., Hiller, R. G.and Howe, C. J. 2006. Comparative analysis of dinoflag-ellate chloroplast genomes reveals rRNA and tRNA genes.BMC Genomics 7: 297.

Geraud, M. L., Herzog, M. and Soyergobillard, M. O. 1991.Nucleolar localization of rRNA coding sequences in

Prorocentrum micans Ehr. (dinomastigote, kingdomProtoctist) by in situ hybridization. Biosystems 26: 61–74.

Gould, S. B., Waller, R. R. and McFadden, G. I. 2008. Plastidevolution. Annu. Rev. Plant Biol. 59: 491–517.

Hiller, R. G. 2001. ‘Empty’ minicircles and petB/atpA andpsbD/psbE (cytb 559 α) genes in tandem in Amphidiniumcarterae plastid DNA. FEBS Lett. 505: 449–52.

Hofacker, I. L., Fontana, W., Stadler, P. F., Bonhoeffer, L. S.,Tacker, M. and Schuster, P. 1994. Fast folding and com-parison of RNA secondary structures. Monatsh. Chem.125: 167–88.

Howe, C. J., Nisbet, R. E. R. and Barbrook, A. C. 2008. Theremarkable chloroplast genome of dinoflagellates. J. Exp.Bot. 59: 1035–45.

Iida, S., Kobiyama, A., Ogata, T. and Murakami, A. 2009.Identification of transcribed and persistent variants of thepsbA gene carried by plastid minicircles in a dinoflagellate.Curr. Genet. 55: 583–91.

Iida, S., Kobiyama, A., Ogata, T. and Murakami, A. 2010.Differential DNA rearrangements of plastid genes, psbAand psbD, in two species of the dinoflagellateAlexandrium. Plant Cell Physiol. 51: 1869–77.

Koumandou, V. L. and Howe, C. J. 2007. The copy numberof chloroplast gene minicircles changes dramaticallywith growth phase in the dinoflagellate Amphidiniumoperculatum. Protist 158: 89–103.

Laatsch, T., Zauner, S., Stoebe-Maier, B., Kowallik, K. V. andMaier, U. G. 2004. Plastid-derived single gene minicirclesof the dinoflagellate Ceratium horridum are localized in thenucleus. Mol. Biol. Evol. 21: 1318–22.

Moore, R. B., Ferguson, K. M., Loh, W. K. W.,Hoegh-Guldberg, C. and Carter, D. A. 2003. Highly organ-ized structure in the non-coding region of the psbAminicircle from clade C Symbiodinium. Int. J. Syst. Evol.Microbiol. 53: 1725–34.

Murray, S., Jorgensen, M. F., Daugbjerg, N. and Rhodes, L.2004. Amphidinium revisited. II. Resolving speciesboundaries in the Amphidinium operculatum speciescomplex (Dinophyceae), including the descriptions ofAmphidinium trulla sp. nov. and Amphidinium gibbosum.comb. nov. J. Phycol. 40: 366–82.

Nelson, M. J. and Green, B. R. 2005. Double hairpinelements and tandem repeats in the non-coding region ofAdenoides eludens chloroplast gene minicircles. Gene358: 102–10.

Nisbet, R. E. R., Koumandou, V. L., Barbrook, A. C. and Howe,C. J. 2004. Novel plastid gene minicircles in the dinoflag-ellate Amphidinium operculatum. Gene 331: 141–7.

Nisbet, R. E. R., Hiller, R. G., Barry, E. R., Skene, P.,Barbrook, A. C. and Howe, C. J. 2008. Transcript analysisof dinoflagellate plastid gene Minicircles. Protist 159:31–9.

Noé, L. and Kucherov, G. 2005. YASS: enhancing the sensi-tivity of DNA similarity search. Nucleic Acids Res. 33:540–3.



Ohta, N., Matsuzaki, M., Misumi, O. et al. 2003. Completesequence and analysis of the plastid genome of the uni-cellular red alga Cyanidioschyzon merolae. DNA Res. 10:67–77.

Palmer, J. D. 1990. Contrasting modes and tempos of genomeevolution in land plant organelles. Trends Genet. 6: 115–20.

Reith, M. and Munholland, J. 1993. A high-resolution genemap of the chloroplast genome of the red alga Porphyrapurpurea. Plant Cell 5: 465–75.

Reyes-Prieto, A., Weber, A. P. M. and Bhattacharya, D. 2007.The origin and establishment of the plastid in algae andplants. Annu. Rev. Genet. 41: 147–68.

Sato, N. and Ohta, N. 2001. DNA-binding specificity anddimerization of the DNA-binding domain of the PENDprotein in the chloroplast envelope membrane. NucleicAcids Res. 29: 2244–50.

Sato, N., Ohshima, K., Watanabe, A. et al. 1998. Molecularcharacterization of the PEND protein, a novel bZIP proteinpresent in the envelope membrane that is the site ofnucleoid replication in developing plastids. Plant Cell 10:859–72.

Simpson, C. L. and Stern, D. B. 2002. The treasure trove ofalgal chloroplast genomes. Surprises in architecture andgene content, and their functional implications. PlantPhysiol. 129: 957–66.

Takishita, K., Ishikura, M., Koike, K. and Maruyama, T. 2003.Comparison of phylogenies based on nuclear-encoded SSUrDNA and plastid-encoded psbA in the symbiotic dinoflag-ellate genus Symbiodinium. Phycologia 42: 285–91.

Wang, Y. L. and Morse, D. 2006. The plastid-encoded psbAgene in the dinoflagellate Gonyaulax is not encoded on aminicircle. Gene 371: 206–10.

Wang, Y. L., Jensen, L., Hojrup, P. and Morse, D. 2005.Synthesis and degradation of dinoflagellate plastid-encoded psbA proteins are light-regulated, not circadian-regulated. Proc. Natl. Acad. Sci. U.S.A. 102: 2844–9.

Zhang, Z. D., Green, B. R. and Cavalier-Smith, T. 1999.Single gene circles in dinoflagellate chloroplast genomes.Nature 400: 155–9.

Zhang, Z. D., Cavalier-Smith, T. and Green, B. R. 2002.Evolution of dinoflagellate unigenic minicircles and thepartially concerted divergence of their putative repliconorigins. Mol. Biol. Evol. 19: 489–500.


8 S. Owari et al.

subcellular localization of minicircle dna in the dinoflagellate amphidinium massartii

Documents