Disruption of miRNA sequences by TALENs and CRISPR/Cas9 induces varied lengths of miRNA production

Abstract

MicroRNAs (miRNAs) are 20-24 nucleotides (nt) small RNAs functioning in eukaryotes. The length and sequence of miRNAs are not only related to the biogenesis of miRNAs but are also important for downstream physiological processes like ta-siRNA production. To investigate these roles, it is informative to create small mutations within mature miRNA sequences. We used both TALENs (transcription activator-like effector nucleases) and clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9) to introduce heritable base pair mutations in mature miRNA sequences. For rice, TALEN constructs were built targeting five different mature miRNA sequences and yielding heritable mutations. Among the resulting mutants, mir390 mutant showed a severe defect in the shoot apical meristem (SAM), a shootless phenotype, which could be rescued by the wild-type MIR390. Small RNA sequencing showed the two base pair deletion in mir390 substantially interfered with miR390 biogenesis. In Arabidopsis, CRISPR/Cas9-mediated editing of the miR160* strand confirmed that the asymmetric structure of miRNA is not a necessary determinant for secondary siRNA production. CRISPR/Cas9 with double-guide RNAs successfully generated mir160a null mutants with fragment deletions, at a higher efficiency than a single-guide RNA. The difference between the phenotypic severity of miR160a mutants in Col-0 versus Ler backgrounds highlights a diverged role for miR160a in different ecotypes. Overall, we demonstrated that TALENs and CRISPR/Cas9 are both effective in modifying miRNA precursor structure, disrupting miRNA processing and generating miRNA null mutant plants.

Keywords: Arabidopsis; CRISPR/Cas9; TALENs; gene editing; miRNA; rice.

Figure 1

Five genomic loci targeted by TALENs and representative genotypes of mutant lines for five MIR genes. Underlined letters indicate sequences selected for design of paired TALENs, TALEN‐L and TALEN‐R for each MIR gene, while letters in red correspond to the sequences of mature miRNAs. Solid black bars denote the transcribed regions of MIR genes, and numbers flanking the bars are coordinates of genomic regions of MIR genes. All expected mutation sites were designed to be within the sequences for mature miRNAs. Individual rice lines with mutated genomic sequences aligned with their wild‐type (WT) sequences are shown under the genome loci. All mutations in T0 plants were monoallelic. Dashed lines denote nucleotides deleted and bold lower letters are for nucleotides inserted. Changes of nucleotides (insertion, + and deletion, −) are presented at the far‐right sides of sequences.

Figure 2

miRNA expression in MIR mutants. (a) Expected mature miRNA sequences in four mutant lines. Bold upper letters in strikethrough are the TALEN‐induced nucleotides, and bold lower letters represent TALEN‐introduced insertions. (b) Expression level of mature miRNA affected by TALEN‐induced mutations. The expression change in mutated alleles varies from abolished expression to increased expression. (c) 21‐nt miRNA abundance in wild type and three mutant lines. Each mutant is depicted in three columns: blue columns for wild‐type miRNA abundance in Kitaake (kit); orange columns for wild‐type miRNA abundance in mutant; grey columns for mutated miRNA abundance in mutant. The numbers above each column indicate the reads of the miRNAs.

Figure 3

miR390 pre‐miRNA cleavage sites detected by mapped 5´ ends of small RNAs. (a) Cleavage pattern of wild‐type miR390. (b) Cleavage pattern of mutated miR390. The sequences are arranged for the predicted secondary structure of pre‐miRNA. Red letters are mature miRNA sequence, and blue letters are sequence of miRNA*. The arrows indicate the positions of the precursor cuts located below the number of reads corresponding small RNA sequences identified. The sequences besides the secondary structures show the most abundant small RNAs detected in the library. The numbers after the sequence show the length of the small RNAs, and the Italic numbers after the length indicate the abundance of small RNAs.

Figure 4

Mutation in MIR390 causes rice shootless phenotype. (a) Shootless phenotype of mir390. Genotypes of germinating seeds are indicated above the picture. The mutant homozygous for miR390−/− exhibits obvious shootless phenotype while the heterozygous and wild‐type seeds show normal germination and growth. (b) Segregation ratio of miR390‐/+ progeny in germination. (c) Schematic construct of MIR390 genomic region. Red bar represents transcribed region of MIR390, numbers below are lengths of corresponding region of promoter, transcribed region and terminator of MIR390. (d) Restoration of shoot phenotype by genomic fragment of MIR390 from callus cells derived from the embryos of shootless mutant.

Figure 5

Arabidopsis MIR160a gene editing using CRIPSR/Cas9 with single‐guide RNA. (a) The design of CRISPR/Cas9 for the miR160a* strand. miR160 and miR160a* strands are in bold and underlined. PAM and double‐strand break (DSB) sites are labelled. The pairing sequence between the guide RNA (gRNA) and the genomic DNA is in blue. (b) The single ‘T’ insertion on the miR160a* strand makes the miRNA precursor into an asymmetric structure that produces asymmetric miRNA/miRNA* duplex. (c) The two‐week‐old wild‐type and mir160a*+1 plants. Scale bars represent 1 cm. (d) Secondary structure of miR160a precursor from wild‐type and mir160a*‐Δ5 plants. The secondary structures are redrawn based on the mfold result.

Figure 6

Arabidopsis MIR160a gene editing using CRIPSR/Cas9 with double‐guide RNA. (a) The design of CRISPR/Cas9 with double‐guide RNAs. miR160 and miR160a* strands are in bold and underlined. PAM and double‐strand break (DSB) sites are labelled. The pairing sequences between the guide RNA (gRNA) and the genomic DNA are labelled in blue and orange. The resultant 47‐ or 48‐bp fragment deletions are represented below. (b) The two‐week‐old wild‐type and mir160a‐Δ47 plants; scale bars represent 1 cm. (c) Flower phenotypes of the mir160a‐Δ47 mutants. (d) Siliques at the same developmental stage from wild‐type and mir160a‐Δ47 mutants. (e) Developing seeds of the wild‐type and mir160a‐Δ47 mutants. Blue arrows indicate delayed or aborted developing seeds; white arrows indicate unfertilized ovules.