Virus-based microRNA expression for gene functional analysis in plants

Abstract

Traditional virus-induced gene silencing (VIGS) is a powerful virus-based short interfering RNA-mediated RNA silencing technique for plant functional genomics. Besides short interfering RNAs, microRNAs (miRNAs) have also been shown to regulate gene expression by RNA silencing in various organisms. However, plant virus-based miRNA silencing has not been reported. In addition, a number of plant miRNAs have been identified or predicted, while their functions are largely unknown. Thus, there is an urgent need for the development of new technologies to study miRNA function. Here, we report that a modified cabbage leaf-curl geminivirus vector can be used to express artificial and endogenous miRNAs in plants. Using this viral miRNA expression system, we demonstrate that VIGS using artificial miRNAs, dubbed as "MIR VIGS," was effective to silence the expression of endogenous genes, including PDS, Su, CLA1, and SGT1, in Nicotiana benthamiana. Silencing of SGT1 led to the loss of N-mediated resistance to Tobacco mosaic virus. Furthermore, using this viral miRNA expression system, we found that viral ectopic expression of endogenous miR156 and miR165 but not their mutants in N. benthamiana resulted in earlier abnormal developmental phenotypes, and expression of miR165 induced abnormal chlorotic spots on leaves. These results demonstrate that the cabbage leaf-curl geminivirus-based miRNA expression system can be utilized not only to specifically silence genes involved in general metabolism and defense but also to investigate the function of endogenous miRNAs in plants.

Figure 1.

Agroinfection of plants with the modified CaLCuV vector. A, Construction of CaLCuV T-DNA vector. CR, Common region; LB, left border; RB, right border. B, Agroinfection of N. benthamiana plants with CaLCuV pCVA and pCVB (right), but not the control vector pCAMBIA1301 (left), causes the leaf-curl symptom. Photographs were taken of N. benthamiana plants at 3 wpi. C, PCR showing the existence of CaLCuV infection in plants infiltrated with Agrobacterium containing pCVA and pCVB (right lane) but not pCAMBIA1301 (middle lane). The positions of primers OTY77/OTY78 for PCR detection are indicated. M, Marker.

Figure 2.

MIR VIGS of marker genes. A, The designed amiRNA sequences and their related target mRNA sequences. B, Silencing phenotypes of marker genes in N. benthamiana plants coinfiltrated with CaLCuV, CaLCuV-amiR-PDS, CaLCuV-amiR-Su, or CaLCuV-amiR-CLA1. Photographs were taken at 35 dpi. C, End-point RT-PCR (30 cycles) analyses showing the existence of mature amiRNAs (amiR-PDS, amiR-Su, and amiR-CLA1) in inoculated plants. D, Semiquantitative RT-PCR analysis of the effects of MIR VIGS on the levels of the PDS,Su, and CLA1 mRNAs. Lanes 1 to 5 represent PCRs with 18, 21, 24, 27, and 30 cycles, respectively. PCR products to the left of the marker (M) are from the silenced plants, and those to the right are the nonsilenced controls. Actin mRNA was used as an internal control, and a no-RT control (C) was also included.

Figure 3.

TMV response to MIR VIGS of NbSGT1. A, The amiR-SGT1 sequence and its cognate target sequence. B, N. benthamiana plant coinfiltrated with TMV-GFP and CaLCuV (top row) or CaLCuV-amiR-SGT1 (bottom row). Photographs were taken at 20 dpi. C, RT-PCR detection of TMV in plants agroinfected by CaLCuV-amiR-SGT1. D, End-point stem-loop RT-PCR (30 cycles) analysis showing the existence of mature amiRNAs (amiR-SGT1) in inoculated plants. E, Semiquantitative RT-PCR analysis of the effects of MIR VIGS on NbSGT1 mRNA. Lanes 1 to 5 represent PCRs with 18, 21, 24, 27, and 30 cycles, respectively. PCR products to the left of the marker (M) are from the silenced plants, and those to the right are the nonsilenced controls. Actin mRNA was used as an internal control, and a no-RT control (C) was also included.

Figure 4.

Comparison of MIR VIGS and SIR VIGS with different insertions. Phenotypes of N. benthamiana plants coinfiltrated with Agrobacterium carrying pCVB and pCVA or its derivatives are shown. Photographs were taken at 4 wpi.

Figure 5.

CaLCuV-based expression of Arabidopsis miR156b and miR165a in N. benthamiana plants. A, Plant phenotypes associated with CaLCuV, CaLCuV-miR156, CaLCuV-miR156-m, CaLCuV-miR165, and CaLCuV-miR165-m at 10 dpi (top row) or 16 dpi (bottom row). B, End-point stem-loop RT-PCR (22 cycles) analysis showing the levels of mature miR156 and miR165 in inoculated plants. C, Semiquantitative RT-PCR analysis of the effects of CaLCuV-based expression of Arabidopsis miR156b and miR165a on their cognate target genes. Lanes 1 to 5 represent PCRs with 18, 21, 24, 27, and 30 cycles, respectively. PCR products to the left of the marker (M) are from the silenced plants, and those to the right are the nonsilenced controls. Actin mRNA was used as an internal control, and a no-RT control (C) was also included.