Extensive 3' modification of plant small RNAs is modulated by helper component-proteinase expression

Abstract

RNA silencing is an evolutionarily conserved process in eukaryotes that represses gene expression by using 21- to 24-nt guide RNAs to mediate mRNA cleavage or translational inhibition. Plants have two distinct groups of silencing-associated small RNAs (smRNAs): the micro RNAs (miRNAs) and the small interfering RNAs (siRNAs). A recent report by Yu et al. [Yu, B., Yang, Z., Li, J., Minakhina, S., Yang, M., Padgett, R. W., Steward, R. & Chen, X. (2005) Science 307, 932-935] has shown that plant miRNAs are modified at their 3' termini with a methyl group. Here, we show that a large fraction of all silencing-associated smRNAs in tobacco are modified; this modification occurs on the 2' hydroxyl of the terminal ribose and significantly reduces the cloning efficiency of these modified smRNAs. Expression of the strong silencing suppressor P1/helper-component proteinase results in a marked decrease in the 3'-terminal modification of viral siRNAs but does not significantly affect the modification of endogenous miRNAs and 24-nt siRNAs. The differential modification mediated by helper-component proteinase expression implies that exogenous and endogenous smRNAs are processed through independent pathways that are isolated by subcellular compartmentalization and/or the association with distinct Dicer complexes. The degree of terminal modification may play an important role in regulating the extent to which primary smRNA signals can be amplified by RNA-dependent RNA polymerases.

Fig. 1.

General resistance of smRNAs to periodate treatment. (A) smRNAs were radiolabeled with PNK (lanes 3, 5, 7, and 9) and then subjected to periodate treatment (lanes 4, 6, 8, and 10). Only RNA from virally infected HC-Pro⁺ plants showed significant sensitivity to periodate (lane 8). A synthetic 7-nt internal control RNA was added to all RNA fractions before chemical treatment to confirm the efficient periodate-mediated elimination of the terminal nucleoside from unmodified RNA. A 24-nt RNA marker reacts to completion when exposed to periodate, shifting downward by ≈2 nt in mobility (lanes 1 and 2). Note that the ≈15-nt double bands in the two uninfected samples are also fully sensitive to periodate (lanes 3-6). (B) The 24-nt-long RNAs from both the infected and uninfected plant samples were gel-purified and subjected to periodate treatment separately. Quantification of cleavage products indicates that cleavage in uninfected plants is weakly dependent on HC-Pro expression (7% and 5% in lanes 3 and 5, respectively, after background subtraction using the - periodate lanes). In infected plants, HC-Pro effects a 2-fold increase in periodate cleavage (12% and 6% in lanes 7 and 9, respectively).

Fig. 2.

RNA-DNA ligation with T4 RNA ligase. (A) The 21- to 24-nt smRNAs from infected HC-Pro⁺ or HC-Pro^- plants (lanes 5-12) or the 24-nt smRNAs from both infected and uninfected HC-Pro⁺ and HC-Pro^- plants (lanes 13-28) were gel-purified and ligated to a 16-nt adenylated DNA oligonucleotide. A synthetic 18-nt RNA control ligation is also shown (lanes 1-4). The ligation was analyzed at four time points for all of the samples: 1 min, 1 h, 2 h, and overnight. The slower-migrating species correspond to ligated RNA-DNA ligation products; the faster-migrating bands that emerge during ligation correspond to circularized RNA (only these bands remain labeled upon treatment with CIP, indicating the circular nature of the RNA). The RNA-DNA ligation product from infected HC-Pro⁺ plants that corresponds to the 21- to 22-nt RNAs is significantly enriched (lanes 5-8) relative to the 24-nt RNAs. This effect does not occur in the absence of HC-Pro expression (lanes 9-12). The 24-nt RNAs from either infected or uninfected plants (HC-Pro⁺ or HC-Pro^-) preferentially form self-ligated circular species, with a 3:1 ratio between this circular species and the DNA-RNA ligation products. In contrast, the ligation of the 18-nt control RNA (having the terminal sequence of... AAG-3′) is much faster and only gives a negligible amount of circular product (lanes 1-4). (B) Histograms showing the smRNA distribution before ligation (blue) or in the RNA-DNA ligated population (red) from viral-infected HC-Pro⁺ plants (Left) or HC-Pro^- plants (Right).

Fig. 3.

Source and size distribution of cloned smRNA populations. Data are shown for uninfected (Upper) and infected (Lower) plants lacking (Left) or expressing (Right) HC-Pro. The green bars indicate smRNA clones derived either from tRNA or rRNA (light green bars) or from other endogenous sequences (dark green bars). The red and orange bars indicate Y-Sat- and CMV-derived smRNA clones, respectively. All smRNAs that were successfully cloned and sequenced are included in these histograms. smRNA clones of 21-22 nt are dominant in virally infected HC-Pro-expressing plants (Lower Right), consistent with the marked periodate sensitivity (Fig. 1 A, lane 8) and ligation efficiency (Fig. 2B Left) exhibited by these RNAs. Note also the large excess of 21- to 22-nt smRNA seen in uninfected HC-Pro⁺ plants (Upper Right) relative to that observed in Fig. 1.