siRNA-dependent and -independent post-transcriptional cosuppression of the LTR-retrotransposon MAGGY in the phytopathogenic fungus Magnaporthe oryzae

Abstract

The LTR-retrotransposon MAGGY was introduced into naive genomes of Magnaporthe oryzae with different genetic backgrounds (wild-type, and MoDcl1 [mdl1] and MoDcl2 [mdl2] dicer mutants). The MoDcl2 mutants deficient in MAGGY siRNA biogenesis generally showed greater MAGGY mRNA accumulation and more rapid increase in MAGGY copy number than did the wild-type and MoDcl1 mutants exhibiting normal MAGGY siRNA accumulation, indicating that RNA silencing functioned as an effective defense against the invading element. Interestingly, however, regardless of genetic background, the rate of MAGGY transposition drastically decreased as its copy number in the genome increased. Notably, in the MoDcl2 mutant, copy-number-dependent MAGGY suppression occurred without a reduction in its mRNA accumulation, and therefore by a silencing mechanism distinct from both transcriptional gene silencing and siRNA-mediated RNA silencing. This might imply that some mechanism possibly similar to post-transcriptional cosuppression of Ty1 retrotransposition in Saccharomyces cerevisiae, which operates regardless of the abundance of target transcript and independent of RNA silencing, would also function in M. oryzae that possesses the RNA silencing machinery.

Figure 1.

MoDcl2 but not MoDcl1 is responsible for MAGGY siRNA biogenesis in M. oryzae. Northern blot analysis of MAGGY siRNAs in M. oryzae transformants with different genetic backgrounds (wild-type, MoDcl1-KO, MoDcl2-KO). Small RNA fractions were prepared and separated on a 17.5% polyacrylamide gel. Equal loading of total RNA was estimated by ethidium bromide staining of predominant RNAs.

Figure 2.

Southern blot analysis of the LTR-retrotransposon MAGGY in spheroplast regenerants. Copy number dynamics of MAGGY was assessed in a series of spheroplast regenerants with different genetic backgrounds (wild-type, MoDcl1-KO and MoDcl2-KO). Southern hybridization was performed as described previously (13). SG, Spheroplast generations.

Figure 3.

Copy number dynamics of MAGGY in Dicer mutants of M. oryzae. The copy number of MAGGY (A) and the rate of MAGGY transposition (B) in each regenerant series are shown. The rate of MAGGY transposition was calculated by dividing the copy number increment in the next generation by copy number.

Figure 4.

Copy-number-dependent MAGGY suppression in MoDcl2-KO mutants is not associated with transcriptional repression or mRNA degradation. Accumulation of MAGGY mRNA (A) and siRNAs (B) in the R0, R3 and R7 generations was examined by northern blots. RNA samples from two regenerant series each of wild-type, MoDcl1-KO and MoDcl2-KO were subjected to the analysis. Equal loading was estimated by ethidium bromide staining of rRNA or predominant RNAs.

Figure 5.

M.oryzae endogenous transposons are not generally targeted by the RNA silencing machinery. (A) Transcript levels of the endogenous transposons, MG-SINE, Pot2 and MGR583 were examined by semi-quantitative RT-PCR. Two independent M. oryzae strains each of wild-type, MoDcl1-KO and MoDcl2-KO were used. Actin and elongation factor 1a (EF1) genes were used as internal controls. (B) Northern blot analysis of MAGGY siRNAs in three natural MAGGY carriers of M. oryzae isolates from rice (lanes 1 and 2) and foxtail millet (lane 3), and one of non-MAGGY carrier (wheat isolate; lane 4). A Br48 transformant with MAGGY was also subjected to the analysis (lane 5). Small RNA fractions were prepared and separated on a 17.5% polyacrylamide gel as described previously [Kadotani et al. (14)]. Equal loading of total RNA was estimated by ethidium bromide staining of predominant RNAs.