Takara 6110

[Tarja Rabito]

Thisis an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

New miRNAs are evolutionarily important but their functional evolution remains unclear. Here we report that the evolution of a microRNA cluster, mir-972C rewires its downstream regulatory networks in Drosophila. Genomic analysis reveals that mir-972C originated in the common ancestor of Drosophila where it comprises six old miRNAs. It has subsequently recruited six new members in the melanogaster subgroup after evolving for at least 50 million years. Both the young and the old mir-972C members evolved rapidly in seed and non-seed regions. Combining target prediction and cell transfection experiments, we found that the seed and non-seed changes in individual mir-972C members cause extensive target divergence among D. melanogaster, D. simulans, and D. virilis, consistent with the functional evolution of mir-972C reported recently. Intriguingly, the target pool of the cluster as a whole remains relatively conserved. Our results suggest that clustering of young and old miRNAs broadens the target repertoires by acquiring new targets without losing many old ones. This may facilitate the establishment of new miRNAs in existing regulatory networks.

Using next-generation sequencing techniques, previous studies have identified a large cohort of new miRNAs across taxa (Berezikov, 2011). In Drosophila, we have reported that the birth and death of miRNAs is extremely rapid (Lyu et al., 2014). It has been shown that over 40% of the miRNAs are only observed in the specific lineages (Lyu et al., 2014). Among these evolutionarily young miRNAs, 95% of them likely arose from scratch, as their seeds and precursors are different from that of the existing miRNAs (Lyu et al., 2014). It appears that these newly-evolved miRNAs have introduced a wide array of novel miRNA-mRNA interactions. Similar to new protein-coding genes, young miRNAs are inclined to express in testes, and they exhibit strong signatures of positive selection (Lyu et al., 2014; Mohammed et al., 2014). Understanding the mechanisms through which new miRNAs and targets evolve will provide key insights into the evolutionary processes of new genes. For example, how novel components originated and integrated into biological networks.

Our previous study has identified a Drosophila-specific miRNA cluster that we refer as mir-972C (Lyu et al., 2014). In Drosophila melanogaster, it consists of at least 12 miRNA members including mir-972, making it the largest new miRNA cluster in this species by far (Marco et al., 2013; Lyu et al., 2014; Mohammed et al., 2018). We speculated that the mir-972C is evolutionarily important, as it is highly expressed in testes and its DNA sequences exhibit strong signal of positive selection (Lyu et al., 2014). To understand how newly-evolved miRNAs influence gene regulatory networks, we investigated the evolution of miR-972C sequences and the regulatory networks within Drosophila. We found that not only the members of mir-972C vary across species, their sequences also undergo rapid changes, which cause the evolution of the target repertoire. In the end, we discussed the selective forces that may drive the evolution of new miRNA clusters in the long-term.

We searched for mir-972C sequences in the Drosophila genomes using BLAT (Kent, 2002) with default parameters and an E-value threshold of 0.001. Homologs of the mir-972C members in each species were identified using BLAST (Altschul et al., 1990) with queries of the known precursor sequences (miRBase Release 22.1) and an E-value threshold of 0.001. miRNA homologs from different species were aligned using MUSCLE (Edgar, 2004) with default parameters. To validate these miRNAs in D. melanogaster, D. simulans, D. pseudoobscura, and D. virilis, we used miRDeep2 (Friedlnder et al., 2012) to map the small RNA sequencing reads (see Supplementary Table S1 for the information of the libraries) back to the genomic sequences of the entire cluster with default parameters. We used five standards that derived from a publication (Fromm et al., 2015) to validate miRNAs: 1) at least one miR* read; 2) at least 20 reads mapping to miR and miR* in total; 3) a hairpin structure with at least 13 paired nucleotides in miR:miR* duplex; 4) The top 3 iso-miR reads account for 85% of the miR arm reads and 5) the miR:miR* duplex to background reads ratio is >1. A maximum parsimony analysis was used to infer the origination of the mir-972C members by assuming that a miRNA emerged in the most recent common ancestor of all species bearing verified homologs.

Total RNAs were extracted from the samples using TRIzol (Thermo Fisher Scientific, catalog no. 15596026) for qRT-PCR and RNA-seq analyses. To quantify miRNA expression, total RNAs were reverse-transcribed into cDNAs using stem-loop reverse transcription and analyzed using the TaqMan qRT-PCR method following the miRNA UPL (Roche Diagnostics) probe assay protocol (He et al., 2016). The 2S RNA was used as the endogenous control (see Supplementary Table S4 for the qRT-PCR primers). cDNA libraries for each RNA sample were sequenced using the Illumina HiSeq 2000 at the Beijing Genomics Institute (Shenzhen). Reads were mapped to the genomes using TopHat (v.1.3.1) with a parameter -r 20 (Trapnell et al., 2009). Gene expression was estimated by FPKM (Fragments Per Kilobase per Million) using Cufflinks (v.2.1.1) with default parameters (Trapnell et al., 2010). Differentially expressed genes were determined using Cuffdiff (v.2.1.1) with default parameters (Trapnell et al., 2010). Non-expressed genes (FPKM = 0) were removed from the further analyses.

To confirm the trans effects of miRNA evolution on target repression, we expressed dme-mir-975 and dsi-mir-975 in S2 cells, respectively, and measured the expression of nine predicted targets of dme-mir-975. gDNA was removed from total RNAs by using TURBO DNA-free kit (Thermo Fisher Scientific, catalog no. AM 1907). Total RNAs were transcribed into cDNAs with the PrimeScript first strand cDNA synthesis kit (TAKARA Bio, catalog no. 6110A) and followed by a qPCR analysis using the SYBR Premix Ex Taq II kit (TAKARA Bio, catalog no. RR82WR). rp49 was used as an internal control. Primers used are listed in Supplementary Table S5.

Evolutionary history of the mir-972 cluster (mir-972C) in Drosophila. This cluster includes twelves miRNAs with distinct seeds. New (yellow boxes) and old (blue boxes) mir-972C members validated through deep sequencing (see Methods) are shown. miRNA homologs are colored in black. Deletions and insertions observed in sequence alignments are represented by dashed lines and inverted triangles. The genomic region is not drawn to scale. The phylogenetic tree was previously reconstructed based on whole-genome sequences (Clark et al., 2007).

To investigate the origin and evolution of mir-972C, we searched for the orthologs of these miRNA genes in other Drosophila species, and in Aedes (mosquito) and Apis (honey bee) which have diverged from Drosophila 250 and 300 million years ago, respectively (Honeybee Genome Sequencing Consortium 2006; Yeates and Wiegmann, 1999). We found homologous sequences in all the seven Drosophila genomes surveyed (D. simulans, D. yakuba, D. erecta, D. ananassae, D. pseudoobscura, D. mojavensis, and D. virilis), but failed to detect any homologs in the mosquito or the honey bee genomes. This result indicates that mir-972C most likely emerged in the common ancestor of Drosophila between 60 and 250 million years ago. After origination, individual members of mir-972C have undergone rapid birth and death. In the D. pseudoobscura genome, the homologous sequences of mir-2499 and mir-979 were identified. Using the conspecific testes library, we were unable to detect the expression of these sequences, suggesting the loss of the entire cluster in this species (Figure 1). The distribution of individual miRNAs also varies across the remaining species. For example, mir-973/974/975/976/977/978 sequences are represented in all the species except D. pseudoobscura, while other miRNAs have been lost in various lineages (Figure 1).

After showing the evolution of the mir-972C members, we continued to investigate the sequence changes in individual miRNAs among D. melanogaster, D. simulans, and D. virilis. We are particularly interested in the alterations in the seed sequences as they are responsible for target recognition. The precursor alignments (Supplementary Figure S1) reveal two types of seed changes: 1) seed shifting, in which the dominant mature miRNA is shifted due to the changes in Drosha and Dicer processing (e.g., mir-976), and 2) arm switching, in which the mature miRNA switches to the other arms of the precursor (e.g., mir-975). We further inferred the time that the these two types of events occurred on the phylogenetic tree (Figure 2A). As shown in Figure 2A, six of the nine events occurred after the split of D. melanogaster and D. simulans in the recent 4 million years. Both the new and the old mir-972C members are involved in this seed innovation. The arm switching of mir-975 occurred after the split of D. virilis and D. melanogaster/D. simulans but it is unclear on which branch (Figure 2A). mir-978 is the only member that experienced both seed shifting and arm switching, and its seed is different among all three species (Supplementary Figure S1).

Evolution of miR-972C targets. (A) miR-972C seed innovations. Arm switching (blue circles) and seed shifts (yellow circles) were inferred and denoted along ancestral (grey) and recent (light blue) branches. (B) Venn diagrams depict the number of shared targets of individual miR-972C member targets or pooled cluster targets. (C) Functional evolution of targets. GO categories of shared and lineage-specific targets are indicated on the corresponding evolutionary branches.

3a8082e126