Identification of microRNAs in the Toxigenic Dinoflagellate Alexandrium catenella by High-Throughput Illumina Sequencing and Bioinformatic Analysis

Abstract

Micro-ribonucleic acids (miRNAs) are a large group of endogenous, tiny, non-coding RNAs consisting of 19-25 nucleotides that regulate gene expression at either the transcriptional or post-transcriptional level by mediating gene silencing in eukaryotes. They are considered to be important regulators that affect growth, development, and response to various stresses in plants. Alexandrium catenella is an important marine toxic phytoplankton species that can cause harmful algal blooms (HABs). To date, identification and function analysis of miRNAs in A. catenella remain largely unexamined. In this study, high-throughput sequencing was performed on A. catenella to identify and quantitatively profile the repertoire of small RNAs from two different growth phases. A total of 38,092,056 and 32,969,156 raw reads were obtained from the two small RNA libraries, respectively. In total, 88 mature miRNAs belonging to 32 miRNA families were identified. Significant differences were found in the member number, expression level of various families, and expression abundance of each member within a family. A total of 15 potentially novel miRNAs were identified. Comparative profiling showed that 12 known miRNAs exhibited differential expression between the lag phase and the logarithmic phase. Real-time quantitative RT-PCR (qPCR) was performed to confirm the expression of two differentially expressed miRNAs that were one up-regulated novel miRNA (aca-miR-3p-456915), and one down-regulated conserved miRNA (tae-miR159a). The expression trend of the qPCR assay was generally consistent with the deep sequencing result. Target predictions of the 12 differentially expressed miRNAs resulted in 1813 target genes. Gene ontology (GO) analysis and the Kyoto Encyclopedia of Genes and Genomes pathway database (KEGG) annotations revealed that some miRNAs were associated with growth and developmental processes of the alga. These results provide insights into the roles that miRNAs play in the growth of A. catenella, and they provide the basis for further studies of the molecular mechanisms that underlie bloom growth in red tides species.

Fig 1. Length distribution and abundance of sequenced small RNA sequences in A. catenella libraries.

Among the small RNAs in the range of 18–25 nt, the most abundant size was 24 nt, followed by 22 nt and 25 nt.

Fig 2. Venn diagram of identified known miRNAs in the two libraries.

The diagram not only shows the numbers of miRNA that were expressed in the lag phase and the logarithmic phase preferentially but also the co-expressed miRNAs in both phases.

Fig 3. Expression abundance of miRNA families in two growth phases in A. catenella.

X-axis represents the known miRNA families, Y-axis represents the normalized expression level of each miRNA family.

Fig 4. miRNA families with more than two members.

Fig 5. The occurrence of A. catenella miRNAs that appeared also in other selected species.

Gma: Glycine max, osa: Oryza sativa, ptc: Populus trichocarpa, mdm: Malus domestica, zma: Zea mays, mes: Manihot esculenta, vvi: Vitis vinifera, lus: Linum usitatissimum, aly: Arabidopsis lyrata, nta: Nicotiana tabacum, ath: Arabidopsis thaliana, cme: Cucumis melo, bdi: Brachypodium distachyon, sbi: Sorghum bicolor, mtr: Medicago truncatula, stu: Solanum tuberosum, tcc: Theobroma cacao, rco: Ricinus communis, bna: Brassica napus, ppt: Physcomitrella patens, bra: Brachypodium distachyon, sly: Solanum lycopersicum, cpa: Carica papaya, ghr: Gossypium hirsutum, aqc: Aquilegia caerulea, hvu: Hordeum vulgare, Gma: Glycine max, osa: Oryza sativa, ptc: Populus trichocarpa, mdm: Malus domestica, zma: Zea mays, mes: Manihot esculenta, vvi: Vitis vinifera, lus: Linum usitatissimum, aly: Arabidopsis lyrata, nta: Nicotiana tabacum, ath: Arabidopsis thaliana, cme: Cucumis melo, bdi: Brachypodium distachyon, sbi: Sorghum bicolor, mtr: Medicago truncatula, stu: Solanum tuberosum, tcc: Theobroma cacao, rco: Ricinus communis, bna: Brassica napus, ppt: Physcomitrella patens, bra: Brachypodium distachyon, sly: Solanum lycopersicum, cpa: Carica papaya, ghr: Gossypium hirsutum, aqc: Aquilegia caerulea, hvu: Hordeum vulgare.

Fig 6. First nucleotide bias of known miRNAs in A. catenella.

The figure indicates the first nucleotide of 18–25 nt miRNAs. U had the greatest frequency among miRNAs of 20–23 nt.

Fig 7. qPCR validation of the differentially expressedtae-miR159a.

Fig 8. qPCR validation of the differentially expressedaca-miR-3p-456915.

Normalized expression level (2^–ΔΔCT) of tae-miR159a and aca-miR-3p-456915 under different growth phases and conditions.

Fig 9. Partial gene ontology (GO) classification annotated for predicted target genes of 12 differentially expressed miRNAs.

The figure shows partial GO enrichment of the 1813 predicted target genes in three GO ontologies: biological processes, cellular component and molecular function.