Combining artificial miRNA-based technology with virus-induced gene silencing in functional analysis assays

Virus-induced gene silencing (VIGS) is a well-established reverse genetics technology for assessment of gene functions in plants. VIGS is a transient loss-of-function assay that involves three steps: engineering viral genomes to include fragments of host genes that are targeted to be silenced, infecting the plant hosts and suppressing the target genes expression by post-transcriptional gene silencing (PTGS), the defense mechanism deployed by plants against virus infections. Suppression of specific mRNA accumulation allows correlation between gene silencing and the deriving phenotype, providing clues on gene functions. However, the efficiency of this technology may be affected by various factors, including virus vector properties and susceptibility of plant host species. In several cases, weak and/or non-homogeneous distribution in the plant (or in the single leaf) of VIGS may generate results not fully coherent, particularly in terms of correlation between phenotype and accumulation levels of the specifically suppressed RNA. This often limits the extensive application of the technique to more permissive plant species such as Nicotiana benthamiana. Aiming at increasing VIGS efficiency in functional studies, particularly in key crop species, we produced and tested new constructs using a Tobacco rattle virus (TRV)-based vector in tomato (Solanum lycopersicum) and other solanaceous crops. This innovative approach consisted in cloning into the TRV vector a short fragment of a host gene containing at its termini mutations designed for the expression of small interfering RNAs (siRNAs) that mimicked a microRNA (miRNA) structure. The recently developed artificial microRNAs (amiRNAs) technology modifies an endogenous gene silencing mechanism that processes natural miRNA precursors to small silencing RNAs targeting transcripts for degradation. Based on natural miRNA structures, amiRNAs are commonly designed to contain mismatches at specific nucleotides with respect to their target sites. Unlike the conventional amiRNA strategy, where target-specific 21nt small silencing RNAs derive from longer double-stranded RNA (dsRNA) precursors that are processed in the nucleus by DCL1, we designed a vector where amiRNA-like small RNAs are generated when viral intermediate dsRNA forms are targeted by the host PTGS machinery in the cytoplasm. In the viral vector, we inserted mutant sequences designed to contain at both their 5' and 3' termini one or two mismatches at selected positions. Mismatched sequences were computed by the WMD3 web tool (wmd3.weigelworld.org), an algorithm that generates all possible amiRNAs using full-length target gene sequences as input. Short (110nt) amiRNA-like containing sequences were compared for their VIGS efficiency with wild-type sequences, shorter- or longer-sized inserts and inverted-repeat constructs. Upon inoculation of our constructs , VIGS established earlier and more extensively than its wild-type counterpart in tomato, N. benthamiana and N. tabacum. For instance, suppression of the tomato reporter gene magnesium chelatase (ChlI or SU) with the VIGS-amiRNA-like construct produced the typical yellow phenotype earlier and more extensively than its wild-type counterpart, and seven days in advance as compared to the standard TRV-PDS (phytoene desaturase) VIGS vector. Quantitative RT-PCR confirmed the efficiency of our VIGS-amiRNA-like constructs in terms of post-transcriptional suppression of host target mRNAs. Our results are discussed in the light of their beneficial contribution to the functional analysis of genes putatively involved in plant-virus interactions