Reply to Serra et al.: Nucleotide substitutions in plant viroid genomes that multiply in phytopathogenic fungi

This is our response to the letter by Serra et al. (1), which questioned our recent paper (2) describing plant viroid infections in phytopathogenic fungi. In this study, full-length monomeric cDNA clones of seven plant viroid RNA genomes were produced using oligonucleotide synthesis (2). We opted for the monomeric version of viroid cDNA clone, considering that it requires shorter sequences for oligonucleotide synthesis and because of noting an article showing the infectivity of the monomeric cDNA clone of a viroid (3). We constructed plasmid constructs for in vitro transcriptions by placing the viroid cDNA sequence immediately after a T7 promoter sequence and incorporating a restriction site (HindIII or SpeI) into the 3′ terminus (2). Due to the large number of samples, the infectivity of monomeric viroid cDNA transcripts was assayed by RT-PCR, and indeed, all of the RT-PCR products derived from the detection of viroid RNA accumulation in plants, yeast, and fungi inoculated with viroid RNA transcripts were verified by DNA sequencing. First, to test the infectivity of plus strand in vitro transcripts of viroid cDNA clones, each transcript was inoculated to Nicotiana benthamiana plants by mechanical rub-inoculation. At 7 d after inoculation, viroid RNAs were detected by RT-PCR in the uninoculated upper leaves of all plants inoculated with the seven viroid transcripts. Each inoculation experiment included mock-inoculated plants as a negative control, and similar results were obtained from three independent experiments (2). It is unclear whether the discrepancies between the inoculation results for peach latent mosaic viroid and avocado sunblotch viroid (ASBVd) reported by Serra et al. (1) and our paper (2) are due to the different nature of RNA inoculums used in these two studies.

After screening for 21 viroid–fungal host combinations, we obtained six combinations that showed stable viroid accumulation in the fungi (2). For each inoculation experiment, a viroid-free fungus was also cultured in parallel and subjected to RT-PCR analysis. In all RT-PCR detections on fungi (more than 48 sets), viroids were not detected in any negative control samples. Most of the viroid infections were asymptomatic in fungal hosts, but viroids stably accumulated through at least eight subcultures. Moreover, we observed that the viroids could be transmitted horizontally through hyphal anastomosis and vertically through conidia (2). The accumulation of hop stunt viroid (HSVd) and ASBVd was elevated in RNA silencing-deficient fungal strains (dcl mutants) (2), showing that plant viroids are targeted by the fungal antiviral RNA silencing.

Here, we present the partial genome sequences of HSVd that accumulated in the Fusarium graminearum Δdcl2 mutant (4), and ASBVd that accumulated in the Cryphonectria parasitica Δdcl2 mutant (5) after successive subcultures. Our analysis revealed the presence of nucleotide sequence substitutions in the genomic RNA of HSVd and ASBVd that accumulated in fungi after eight subcultures (Fig. 1). In addition, the sequence junction of the circularized (+) genomic RNAs of HSVd was determined by inverse RT-PCR using previously reported primers (2) (Fig. 1). These results provide additional supporting evidence for the occurrence of plant viroid multiplication in fungi and suggest that genome evolution or adaptation may occur during viroid multiplication in fungi.

Fig. 1.

Nucleotide sequence substitutions in the genomic RNAs of HSVd and ASBVd that were maintained in fungal hosts (the Δdcl2 mutants of Fusarium graminearum and Cryphonectria parasitica, respectively) after eight subcultures. RT-PCR was carried out using PrimeSTAR HS DNA Polymerase (Takara) with forward primer, 5′-TTCTCAGAATCCAGCGAGAGG-3′, and reverse primer, 5′-GACAAAAAGCAGGTTGGGACGAAC-3′, for HSVd, and forward primer, 5′-AGAACAAGAAGTGAGGATATGATTAAAC-3′, and reverse primer, 5′-GAAGATAGAGGAGTAAACCTTGCG-3′, for ASBVd. The PCR products were ligated to a plasmid vector pGEM-T Easy (Promega) after A-tailing, and multiple plasmid clones were subjected to sequencing. Schematic representation of proposed genomic structures of HSVd (6) and ASBVd (7, 8) is shown. The position of the nucleotide substitutions and the number of mutated clones per total clones are presented (red letters). The black arrows and the green line mark the sequenced genomic region. Most of the nucleotide substitutions in each clone are single-site substitutions (except for ASBVd M3/M7) with interchanges occuring between the same purine (adenine/guanine) or pyrimidine (cytosine/uracil) nucleic acids. The nucleotides at the junction of circularized (+) genomic RNAs of HSVd accumulated in the fungal host after the third subculture are highlighted red (the 5′-start) and blue (the 3′-end). Sequence substitutions were observed at the junction site (and the M4 site of the sequenced region), but no additional sequence (nonviroid sequence) was present. The black-filled and open boxes indicate (+) and (−) sequence of hammerhead structures, respectively. CCR, central conserved region; TCR, terminal conserved region.