Kimihiro Terasawa

Genome sequence of the ultrasmall unicellular red alga Cyanidioschyzon merolae 10D

Small, compact genomes of ultrasmall unicellular algae provide information on the basic and essential genes that support the lives of photosynthetic eukaryotes, including higher plants. Here we report the 16,520,305-base-pair sequence of the 20 chromosomes of the unicellular red alga Cyanidioschyzon merolae 10D as the first complete algal genome. We identified 5,331 genes in total, of which at least 86.3% were expressed. Unique characteristics of this genomic structure include: a lack of introns in all but 26 genes; only three copies of ribosomal DNA units that maintain the nucleolus; and two dynamin genes that are involved only in the division of mitochondria and plastids. The conserved mosaic origin of Calvin cycle enzymes in this red alga and in green plants supports the hypothesis of the existence of single primary plastid endosymbiosis. The lack of a myosin gene, in addition to the unexpressed actin gene, suggests a simpler system of cytokinesis. These results indicate that the C. merolae genome provides a model system with a simple gene composition for studying the origin, evolution and fundamental mechanisms of eukaryotic cells.

A 100%-complete sequence reveals unusually simple genomic features in the hot-spring red alga Cyanidioschyzon merolae

BackgroundAll previously reported eukaryotic nuclear genome sequences have been incomplete, especially in highly repeated units and chromosomal ends. Because repetitive DNA is important for many aspects of biology, complete chromosomal structures are fundamental for understanding eukaryotic cells. Our earlier, nearly complete genome sequence of the hot-spring red alga Cyanidioschyzon merolae revealed several unique features, including just three ribosomal DNA copies, very few introns, and a small total number of genes. However, because the exact structures of certain functionally important repeated elements remained ambiguous, that sequence was not complete. Obviously, those ambiguities needed to be resolved before the unique features of the C. merolae genome could be summarized, and the ambiguities could only be resolved by completing the sequence. Therefore, we aimed to complete all previous gaps and sequence all remaining chromosomal ends, and now report the first nuclear-genome sequence for any eukaryote that is 100% complete.ResultsOur present complete sequence consists of 16546747 nucleotides covering 100% of the 20 linear chromosomes from telomere to telomere, representing the simple and unique chromosomal structures of the eukaryotic cell. We have unambiguously established that the C. merolae genome contains the smallest known histone-gene cluster, a unique telomeric repeat for all chromosomal ends, and an extremely low number of transposons.ConclusionBy virtue of these attributes and others that we had discovered previously, C. merolae appears to have the simplest nuclear genome of the non-symbiotic eukaryotes. These unusually simple genomic features in the 100% complete genome sequence of C. merolae are extremely useful for further studies of eukaryotic cells.

Organization, developmental dynamics, and evolution of plastid nucleoids

The plastid is a semiautonomous organelle essential in photosynthesis and other metabolic activities of plants and algae. Plastid DNA is organized into the nucleoid with various proteins and RNA, and the nucleoid is subject to dynamic changes during the development of plant cells. Characterization of the major DNA-binding proteins of nucleoids revealed essential differences in the two lineages of photosynthetic eukaryotes, namely nucleoids of green plants contain sulfite reductase as a major DNA-binding protein that represses the genomic activity, whereas the prokaryotic DNA-binding protein HU is abundant in plastid nucleoids of the rhodophyte lineage. In addition, current knowledge on DNA-binding proteins, as well as the replication and transcription systems of plastids, is reviewed from comparative and evolutionary points of view. A revised hypothesis on the discontinuous evolution of plastid genomic machinery is presented: despite the cyanobacterial origin of plastids, the genomic machinery of the plastid genome is fundamentally different from its counterpart in cyanobacteria.

Purification and characterization of organellar DNA polymerases in the red alga Cyanidioschyzon merolae

DNA polymerase γ, a mitochondrial replication enzyme of yeasts and animals, is not present in photosynthetic eukaryotes. Recently, DNA polymerases with distant homology to bacterial DNA polymerase I were reported in rice, Arabidopsis, and tobacco, and they were localized to both plastids and mitochondria. We call them plant organellar DNA polymerases (POPs). However, POPs have never been purified in the native form from plant tissues. The unicellular thermotrophic red alga Cyanidioschyzon merolae contains two genes encoding proteins related to Escherichia coli DNA polymerase I (PolA and PolB). Phylogenetic analysis revealed that PolB is an ortholog of POPs. Nonphotosynthetic eukaryotes also have POPs, which suggested that POPs have an ancient origin before eukaryotic photosynthesis. PolA is a homolog of bacterial DNA polymerase I and is distinct from POPs. PolB was purified from the C. merolae cells by a series of column chromatography steps. Recombinant protein of PolA was also purified. Sensitivity to inhibitors of DNA synthesis was different in PolA, PolB, and E. coli DNA polymerase I. Immunoblot analysis and targeting studies with green fluorescent protein fusion proteins demonstrated that PolA was localized in the plastids, whereas PolB was present in both plastids and mitochondria. The expression of PolB was regulated by the cell cycle. The available results suggest that PolB is involved in the replication of plastids and mitochondria.

Characterization of cell-cycle-driven and light- driven gene expression in a synchronous culture system in the unicellular rhodophyte Cyanidioschyzon merolae

The unicellular rhodophyte Cyanidioschyzon merolae, having a single plastid and a single mitochondrion, is suitable for the analysis of the cell cycle involving the division of organelles. In conventional methods of synchronous culture of algae, light/dark cycles have been used as signals for synchronization, and the gene expression promoted by light is not separated from the gene expression related to cell cycle progression. We previously devised a novel synchronous culture system with controlled photosynthesis, which is triggered by 6 h-light/18 h-dark cycles combined with different levels of CO(2). The cells do not enter S-phase and consequently do not divide after the minimum light period without CO(2) supplementation, but do divide after a light period with 1 % CO(2). In this way, we can compare a dividing cycle and a non-dividing cycle. We examined changes in the expression of 74 genes throughout the cell cycle by quantitative RT-PCR. The expression of genes for two cyclins (cyclin C and H) and two CDKs (CDKA and CDKD) as well as metabolic enzymes was promoted by light, whereas the expression of genes for G1/S or G2/M cyclins and CDKs as well as DNA replication enzymes and proteins related to organellar division was promoted only in the dividing cycles. These results suggested that C. merolae has a checkpoint for G1/S progression, which is regulated by nutrients within the 6 h light period.

Occurrence and characterization of PEND proteins in angiosperms

The PEND protein is a DNA-binding protein in the inner envelope membrane of the developing chloroplast. It consists of a short pre-sequence, an N-terminal DNA-binding domain (cbZIP), a central repeat domain, and a C-terminal transmembrane domain. PEND homologs have been detected in various angiosperms, including Arabidopsis thaliana, Brassica napus, Medicago truncatula, cucumber and cherry. Monocot homologs have also been detected in barley and rice, but sequence conservation was low in monocots. PEND-related sequences have not been detected in non-flowering plants and algae. Green fluorescent protein fusions consisting of the N-terminal as well as full-length PEND homologs in A. thaliana and B. napus were targeted to chloroplasts, and localized to nucleoids and chloroplast periphery, respectively. Immunoblot analysis suggested that crucifer homologs were present in chloroplasts probably as a dimer, as in the case of pea. These results suggest that PEND protein is present in angiosperms, and the homologs in crucifers are functionally analogous to the PEND protein in pea.

Plastid localization of the PEND protein is mediated by a noncanonical transit peptide.

Plastid envelope DNA‐binding protein (PEND) is a DNA‐binding protein with a chloroplast basic region‐zipper domain at its N‐terminus and a transmembrane domain at its C‐terminus. The localization of PEND to the inner envelope membrane was demonstrated in a targeting experiment using isolated membranes and green fluorescent protein‐tagged fusion proteins. An N‐terminal sequence analysis showed that the presequence is 15 amino acids long; however, based on neural network‐based prediction tools, this short peptide is not predicted to be a chloroplast‐targeting sequence. In the present study we confirmed, by the digestion of intact chloroplasts, that PEND is located in the envelope membrane. We then demonstrated that the N‐terminal 88‐amino acid sequence is sufficient for plastid import in vitro. The transient expression of green fluorescent protein‐tagged fusion proteins revealed that neither the N‐terminal 29‐amino acid sequence nor the 16‐amino acid sequence directed green fluorescent protein to chloroplasts, but that the N‐terminal 66‐amino acid sequence was sufficient for correct targeting. These results suggest that targeting of PEND to the chloroplast requires both the presequence and the basic region, whereas postimport processing cleaves only the presequence. Interestingly, deletion of the presequence in the green fluorescent protein‐tagged 88‐amino acid construct resulted in targeting to the nucleus. This raises the possibility of plastid‐to‐nuclear signal transduction by the relocalization of PEND.