Undesired small RNAs originate from an artificial microRNA precursor in transgenic petunia (Petunia hybrida)

Abstract

Although artificial microRNA (amiRNA) technology has been used frequently in gene silencing in plants, little research has been devoted to investigating the accuracy of amiRNA precursor processing. In this work, amiRNAchs1 (amiRchs1), based on the Arabidopsis miR319a precursor, was expressed in order to suppress the expression of CHS genes in petunia. The transgenic plants showed the CHS gene-silencing phenotype. A modified 5' RACE technique was used to map small-RNA-directed cleavage sites and to detect processing intermediates of the amiRchs1 precursor. The results showed that the target CHS mRNAs were cut at the expected sites and that the amiRchs1 precursor was processed from loop to base. The accumulation of small RNAs in amiRchs1 transgenic petunia petals was analyzed using the deep-sequencing technique. The results showed that, alongside the accumulation of the desired artificial microRNAs, additional small RNAs that originated from other regions of the amiRNA precursor were also accumulated at high frequency. Some of these had previously been found to be accumulated at low frequency in the products of ath-miR319a precursor processing and some of them were accompanied by 3'-tailing variant. Potential targets of the undesired small RNAs were discovered in petunia and other Solanaceae plants. The findings draw attention to the potential occurrence of undesired target silencing induced by such additional small RNAs when amiRNA technology is used. No appreciable production of secondary small RNAs occurred, despite the fact that amiRchs1 was designed to have perfect complementarity to its CHS-J target. This confirmed that perfect pairing between an amiRNA and its targets is not the trigger for secondary small RNA production. In conjunction with the observation that amiRNAs with perfect complementarity to their target genes show high efficiency and specificity in gene silencing, this finding has an important bearing on future applications of amiRNAs in gene silencing in plants.

Figure 1. Phenotype of amiRchs1 transgenic flowers.

(A) Opening flower bud of amiRchs1 transgenic plants (total RNA was extracted at this stage). (B) V26 (wild-type) flower. (C) amiRchs1 transgenic flower. (D–E) qRT-PCR detection of mRNA levels of the CHS-A (D) and CHS-J (E) gene in V26 and amiRchs1 transgenic petals. Data were normalized against petunia ubiquitin gene and were means of three biological pools (each with three technical replicates); the error bars indicate SD.

Figure 2. Mapping of target cleavage products by 5′ RLM-RACE.

The cleavage sites and the number of sequenced clones corresponding to each site are indicated by arrows. For most of the sequenced clones, the 5′ end was at the expected position, opposite to nucleotides 10–11 of amiRchs1. (A) Cleavage of CHS-A mRNA. (B) Cleavage of CHS-J mRNA. (C) Agarose gel showing products after 5′ RLM-RACE PCR amplification. M, Marker; W, water.

Figure 3. Processing of amiRchs1 and miR319 precursors.

(A) Small RNA sequences from the miRBase database (v20 http://miRbase.org/index.shtml) were incorporated into the predicted stem-loop structure of the Arabidopsis miR319a precursor. Small RNAs cloned fewer than 5 times are also indicated. (B) Small RNA sequences from amiRchs1 transgenic petals incorporated into the scheme for the amiRchs1 precursor. Only small RNAs cloned more than 5 times are indicated. (C) Processing intermediates detected using 5′ RLM-RACE PCR amplification. The positions of cleavage sites, as revealed by 5′ RACE, and the number of sequenced clones corresponding to each site, are indicated by black arrows. The four sites marking the origins of B1 and B2 small RNAs, and corresponding to the four marked fragments in the polyacrylamide gel, are indicated by blue arrows (Sites 1, 2, 3 and 4). Site II corresponds to the cleavage site that marks the origin of the most frequent B1 small RNAs, but the processing intermediates had not been sampled by random sequencing of RACE PCR products. Left inset: Polyacrylamide gel showing fragments after 5′ RACE. (D) Small RNA sequences from amiRchs1 transgenic petals incorporated into the scheme for the petunia miR319a precursor. Only small RNAs cloned more than 5 times are indicated. B1, B1^*, B2, and B2^* correspond to the regions producing the sequences previously designated as miR319a.1, miR319a.1^*, miR319a.2 and miR319a.2^* , , respectively. B3 corresponds to the region between B1 and B2, B3^* corresponds to the region between B1^* to B2^*, and B3⁺ corresponds to a region longer than B3, in which the 3′ ends of B3 sequences stretched into the middle of B1.

Figure 4. Size distribution of small RNAs mapped to the amiRchs1 precursor.

Figure 5. Relative proportions of small RNAs arising from different regions along the miR319 precursors.

The regions are the same as indicated in Figure 3.

Figure 6. Alignments of selected small RNAs and some of their potential mRNA targets.

Included are small RNA length, hybridization energy, percentage of free energy compared to a perfectly complementary target and small RNA reads identified in this study.