Identification And Expression Analysis Of MicroRNAs In Ginkgo Biloba Var. Epiphylla Mak

Ginkgo biloba L., as one of the world’s oldest tree species, belongs to Ginkgoales. Given the slow rate of evolution of the genus and the special evolution status, the ginkgo occupies an outstanding research value in aspects of plant classification, phyletic evolution, morphological anatomy and biology research. Ginkgo biloba had occurred obvious morphological variation on leaves and seeds because of the natural hybridization and artificial selection. Ginkgo biloba var. epiphylla Mak.(G. biloba var. epiphylla) is a special germplasm, and the peculiar pattern of these abnormal ovuliferous structure was the partly epiphyllous ovules, that developed on the leaf blades. The grafted G. biloba var. epiphylla shared similar growth characteristics with the parent tree, indicating the genetic stability of this mutant plant. But little is known about the mechanism of G. biloba var. epiphylla abnormal ovuliferous. It has important theoretical significance to study the molecular mechanism and genetic regulation mechanism in the abnormal ovuliferous development.MicroRNA(miRNA), as one class of non-coding small RNAs, regulate gene expression by directly cleavage or translational inhibition. It is well known that miRNAs play crucial roles in plant development, including organ development, morphogenesis, hormone signal transduction, stress responses etc.. Researches on the miRNA level in G. biloba var. epiphylla will illuminate the molecular mechanism mediated by miRNA and provide the theory basis for the molecular mechanism of abnormal ovuliferous germination and development.In this study, two small RNA libraries constructed from epiphyllous ovule leaves and normal leaves of G. biloba var. epiphylla were sequenced on an Illumina/Solexa platform. Combined with bioinformatics methods and qRT-PCR technology, differentially expressed miRNA and the target genes were identified. The main results were as follows:1. Overall analysis of small RNA sequencingTwo small RNA libraries were constructed from epiphyllous ovule leaves and normal leaves and sequenced using the Solexa/Illumina sequencing platform. The deep sequencing yielded 22,401,208 and 19,042,060 raw reads from the epiphyllous ovule leaves and normal leaves, respectively. After removing adapters, contaminants, inserts, and low quality sequences, 21,717,220 and 18,380,325 clean reads, with lengths ranging from 18 to 30 nt, were obtained. Among the clean reads, 2,585,993 and 2,640,731 unique sequences were obtained. In total, 88.49% of the clean sRNA reads, representing 15.91% of the unique sRNA sequences, were found in both libraries. The length distributions of the sRNAs in both libraries revealed a major peak at 21 nt and a minor peak at 24 nt. The total number of the 21 nt redundant sequences accounted for 65.96% and 56.77% respectively in the two libraries.2. Conserved miRNAs analysisAfter further removal of other non-coding sRNAs(rRNA, tRNA, snRNA, snoRNA) by BLAST searching against the GenBank and Rfam database(9.1), the candidate sRNAs were BLASTN searched against the currently known mature plant miRNAs in miRBase 20.0 to identify conserved miRNAs in G. biloba var. epiphylla. Overall, the 82 miRNA sequences from 23 miRNA families were identified. Most of the identified miRNA families were conserved in a variety of plant species in addition to miR3522, miR1507, miR1509 and miR1314. The expression abundances among the different families and family members varied a lot. The conserved miRNA families with more members(miR168, miR172, miR156, miR390, miR396) had higher abundance of expression, while the miRNA families with only one member showed lower expression levels. Among the 82 conserved miRNAs, 25 miRNAs from 16 families were found to be differentially expressed(fold changes >1.5, p-values ≤0.01) between the two datasets. Sixteen of the 25 differentially expressed miRNAs were more highly expressed in epiphyllous ovule leaves than that in normal leaves. To validate the conserved miRNAs obtained from the Solexa sequencing, qRT-PCR was performed on 15 selected miRNAs. In general, all the miRNA expression profiles were similar to those detected by the Solexa deep sequencing.3. Novel mi RNAs analysisFifty-three putative novel mi RNAs, including 25 miRNA*, were detected using the MIREAP algorithm. Thirty-four of the novel candidate miRNAs were expressed in both databases, 11 were specifically expressed in epiphyllous ovule leaves and 8 were detected only in normal leaves. The size distribution of the novel candidate miRNAs ranged from 20 to 23 nt, with 21 nt being the most abundant(66.0%). The first nucleotide bias of these candidate miRNAs was the common 5’ terminal uridine(U) nucleotide. The pre-miRNAs ranged from 67 to 268 bp in length, with an average of 116 bp. The average minimum free energy(MFE) value obtained for these precursors was-46.0kcal/mol. The read counts for these novel mi RNAs in the two datasets were low. Compared with the miRNA expression levels in normal leaves, 10 were significantly up-regulated and 11 were significantly down-regulated in epiphyllous ovule leaves. The differentially expressed up-/down- regulated miRNAs had fold changes that ranged from 2 to 14 fold. The expression profiles of novel miRNAs verified by qRT-PCR were also similar to those detected by the Solexa deep sequencing.4. The target genes regulated by differentially expressed miRNAsTo determine the functions of the identified miRNAs, putative target genes were predicted using psRNA Target. More than half of the target genes were annotated as unknown or hypothetical proteins. A small part of these annotated target genes were conserved transcription factors related to the growth development and environmental adaptation. To evaluate the potential functions of targets of the differentially expressed miRNAs, gene ontology(GO) annotations were assigned to the targets. A total of 111 and 88 unigenes were identified as potential targets of 38 up-regulated and 34 down-regulated miRNAs, respectively. The functions of the genes targeted by the up-regulated miRNAs in epiphyllous ovule leaves were more abundant in biological regulation, development process, reproduction, signalling process and catalytic activity. The results suggested the G. biloba var. epiphylla miRNAs targeted genes participated in a broad range of physiological functions. Our data suggested that differentially expressed miRNAs may play important roles during the ovule formation and development though regulating plant hormone synthesis and the ovule development related genes. The expression profiles of 10 target genes were also analyzed by qRT-PCR. Except for gbi-miR396 b and its target gene, the expression profiles of the other putative target genes were opposite to those of their corresponding miRNAs.We constructed two sRNA libraries of G. biloba var. epiphylla during partly epiphyllous ovule initiation from epiphyllous ovule leaves and from normal leaves. Significant differences in the regulation patterns of both the conserved and novel miRNAs in the two databases were identified. Target prediction showed that some of the differentially expressed miRNAs may regulate epiphyllous ovule formation. Besides, qRT-PCR technology was used to verify expression levels of some miRNAs and their targets. Our results suggested that diverse and complex miRNAs were involved in epiphyllous ovule formation and development in post-transcription regulation.