Expression of the tobacco MDP92 gene tended to be suppressed in association with N gene activation
As a starting point for this study, we analyzed the secondary RNA structure of intron 4 of the N gene to estimate its possible role in N-mediated resistance, showing that its RNA sequence could contain stem-loop structures with a 21-bp perfectly complementary stem (Fig. S2a). It was possible that this predicted stem could produce small RNAs to regulate defense responses. As a result of database searches for the potential targets of possible intron 4-derived small RNAs, we found that a tobacco MYB-related mRNA (mRNA_127264) contained the 21-nt sequence in the 3’-UTR, which was complementary to the intron 4 sequence (Fig. S2b). The mRNA encodes a protein annotated as MDP92, which contains two helix-loop-helix (HLH) domains conserved in the R2R3-type MYB transcription factor (Fig. S3a and b). We also found other potential targets within non-functional genomic regions, such as an extremely long 5'-UTR, within an intron, or in a pseudogene (data not shown), which we decided not to analyze.
On the assumption that the possible intron 4-derived small RNAs could affect MDP92 expression, RT-qPCR with a temperature shift assay was performed to investigate whether MDP92 expression changes during N activation upon TMV infection. HR was synchronously induced by shifting to 20 ˚C after 40 h of incubation of TMV-inoculated NN tobacco at 30 ˚C, as the N gene is inactivated at 30 ˚C and activated at 20 ˚C (Wright et al. 2000). In TMV-inoculated plants, transcript levels of N increased sharply from 2 h after temperature shift (hat), peaked at 4 hat, and gradually decreased thereafter, while they were relatively constant in mock-inoculated plants (Fig. 1a and b). On the other hand, in both TMV- and mock-inoculated plants, transcript levels of MDP92 decreased from 0 to 6 hat (Fig. 1c). Although we did not find a statistically significant difference, the MDP92 transcript level in TMV-inoculated plants was higher at 0 (1.1-fold) and 2 (1.2-fold) hat, but lower at 4 (0.73-fold), 6 (0.70-fold), and 8 (0.87-fold) hat than that in mock-inoculated plants (Fig. 1d). These results led us to speculate that MDP92 expression might be downregulated in response to temperature change and further suppressed in association with N activation.
To eliminate the effect of temperature change on MDP92 expression, we then examined NN tobacco plants maintained at the constant temperature of 20 ˚C after inoculation with TMV. N transcript levels in TMV-inoculated plants continued to increase from 0 to 3 days after inoculation (dai) as virus infection progressed (Fig. 1e, f, and i). MDP92 transcript levels were similar between TMV- and mock-inoculated plants at 0, 2, and 3 dai (Fig. 1g and h). However, there was a different expression pattern at 1 dai, where MDP92 transcript levels increased in mock-inoculated plants but remained unchanged in TMV-inoculated plants (Fig. 1g and h). Based on these results, we hypothesized that MDP92 expression without TMV infection might temporarily be upregulated in response to wounding by inoculation but suppressed during the early stage of N activation induced by TMV infection.
We conducted RNA-Seq analysis of small RNAs in TMV-infected NN tobacco plants and detected known N-related miRNAs and small RNAs (Li et al. 2012), but not the intron 4-derived small RNAs (Table S3). Therefore, we did not show that small RNAs were produced from intron 4 of the N gene under the current experimental conditions.
Transient overexpression of MDP92 promoted ToMV intercellular movement in N. benthamiana leaves that induced N resistance
Our finding that MDP92 expression tended to be suppressed during N activation raised the possibility that MDP92 might have an impact on N-mediated virus resistance. To investigate this possibility, we performed a transient expression assay by agroinfiltration in N. benthamiana. In this assay, an MDP92 coding sequence was transiently expressed with an infectious cDNA of a ToMV variant encoding ER-targeting GFP (ToMV-erGFP) together with the N coding sequence including its introns (gN-Int1234). A GUS cDNA was used as a negative control. N. benthamiana was used to evaluate the effect of N-mediated virus resistance without inducing visible cell death (Bhattacharjee et al. 2009; Mestre and Baulcombe 2006; Sasaki et al. 2013; 2021).
At 72 h after infiltration (hai), the size of individual fluorescent infection sites and virus accumulation per leaf were significantly higher in MDP92-expressing leaves than GUS-expressing leaves (Fig. 2a, b, and d). Meanwhile, the average number of fluorescent infection sites was similar between the two treatments (Fig. 2c). These results suggested that transient overexpression of MDP92 enhanced virus intercellular movement during N-mediated resistance.
Different expression patterns of the endogenous MDP92 gene between TMV-infected and uninfected nn tobacco plants
Since nn tobacco also has the MDP92 gene, it was possible that MDP92 might be involved in N-independent basal resistance. To test this possibility, we compared MDP92 expression patterns between nn tobacco plants with and without TMV infection. MDP92 transcript levels in mock-inoculated plants were slightly higher at 1, 2, and 3 dai than at 0 dai, possibly due to the response to wounding (Fig. 3a). Compared to mock-inoculated plants, MDP92 transcript levels in TMV-inoculated plants were significantly lower at 1 dai, but similar at 2 and 3 dai (Fig. 3a). The transcript levels of MDP92 in TMV-inoculated plants at 1, 2, and 3 days after inoculation were 0.6, 1.4, and 1.3 times those in the control plants, respectively (Fig. 3b). Virus RNA levels increased sharply between 2 and 3 dai (Fig. 3c). These results suggest that, similar to the results for NN tobacco, the expression of MDP92 in nn tobacco, which would be upregulated by wounding in the absence of TMV infection, may need to be suppressed during the early induction of basal resistance to the virus, but later regulated to levels comparable to those in control plants.
Transient expression of MDP92 increased ToMV infection sites in N. benthamiana
To further elucidate the role of MDP92 in virus resistance in the absence of the N gene, we examined the effect of transient expression of MDP92 or GUS on ToMV-erGFP infection in N. benthamiana. The number of fluorescent infection sites in MDP92-expressing leaves was significantly higher than that in GUS-expressing leaves at 48 hai, but not at 72 hai (Fig. 4a and c). Consistently, the virus accumulation level in MDP92-expressing leaves was significantly higher than that in GUS-expressing leaves at 48 hai (Fig. 4d). However, the fluorescence size of individual infection sites was similar between the two treatments (Fig. 4b). These results suggested that transient overexpression of MDP92 enhanced virus amplification in initially infected cells. We then performed an inoculation assay using protoplasts transiently overexpressing MDP92 or GUS. For inoculation, we used in vitro transcripts of TocJ-GFP, which is a chimeric ToMV encoding GFP and the CP of tobacco mild green mosaic virus (Hori and Watanabe 2003). RT-qPCR analysis showed that viral RNA levels were twice as high in MDP92-expressing protoplasts as in GUS-expressing protoplasts (Fig. 4e). These collective data suggested that overexpressed MDP92 may attenuate basal resistance that limits virus amplification at the single cell level.
Different effects of transient overexpression of MDP92 in the presence or absence of N resistance on the expression of defense-related genes
Based on the finding that transient overexpression of MDP92 can negatively affect N resistance and basal resistance in N. benthamiana, we investigated its effect on the expression of selected defense-related genes (i.e. NbPR1a, NbPR4, NbHin1, and NbHsr203j). NbPR1a and NbPR4 were selected as marker genes for the salicylic acid (SA) and jasmonic acid (JA) signaling pathways, respectively (Diao et al. 2019; Pascual et al. 2015). NbHin1 and NbHsr203j were used as resistance-related markers (Peng et al. 2019; Takagi et al. 2022). RT-qPCR analysis was performed on ToMV-erGFP-infected leaves expressing MDP92 or GUS with or without N resistance induction. Regardless of N resistance, transient overexpression of MDP92 did not affect the transcript level of NbPR1a at 48 or 72 hai (Fig. 5a and e). For NbPR4 (Fig. 5b and f), NbHin1 (Fig. 5c and g), and NbHsr203j (Fig. 5d and h), there were some cases where the transcript levels were significantly different between MDP92 and GUS-expressing leaves. In the presence of N resistance, the transcript level of NbHsr203j was significantly lower in MDP92-expressing leaves than in GUS-expressing leaves at 48 hai (Fig. 5d). In the absence of N resistance, the transcript levels of both NbPR4 and NbHin1 were significantly higher at 48 hai in MDP92-expressing leaves than in GUS-expressing leaves (Fig. 5f and g). These results suggested that MDP92 is differently involved in the regulation of these defense-related genes during N resistance and basal resistance.
To confirm that MDP92 functions as a nuclear transcription factor, we examined the subcellular localization of YFP-fused MDP92 (MDP92-YFP). At 24 hai, MDP92-YFP was localized only in the DAPI-stained nucleus (Fig. S4) in N. benthamiana. This suggested that MDP92 may be a transcription factor that regulates gene expression in the nucleus.
Transient overexpression of MDP92 enhanced PVX infection sites in N. benthamiana
To further investigate whether MDP92 affects basal resistance to viruses other than ToMV, PXV-erGFP was inoculated to N. benthamiana leaves transiently expressing MDP92 or GUS. The average number of fluorescent sites and the level of virus accumulation in MDP92-expressing leaves were significantly higher than in GUS-expressing leaves at 48 hai (Fig. 6a, c, and d). The average size of individual fluorescent sites was comparable between MDP92- and GUS-expressing leaves at 48 and 72 hai (Fig. 6b). In addition, gene expression analysis of defense-related genes during PVX-erGFP infection showed that the significant difference in transcript levels was observed for NbPR4 and NbHin1 at 48 hai (Fig. 6f and g), but not for NbPR1a and NbHsr203j (Fig. 6e and h). These results suggested that MDP92 plays a role in attenuating basal resistance to the early infection with PVX as well as ToMV, which is associated with NbPR4 and NbHin1 expression.
In addition, we performed RT-qPCR to examine the expression levels of an N. benthamiana orthologous gene to MDP92, NbMDP92 (Fig. S3c, Table. S4), using total RNA from the samples with or without the transient expression of MDP92 and/or infection with ToMV-erGFP or PVX-erGFP. However, little or no expression of endogenous NbMDP92 was detected in any of the samples tested (data not shown). These results suggested that NbMDP92 expression is always close to the detection limit and does not affect our results from the MDP92 transient expression experiments described above.