miR171c and miR172a confer resistance to tomato mottle mosaic virus in tobacco by targeting SCL and AP2 transcription factors

Abstract

Background

Tomato mottle mosaic virus (ToMMV) is an emerging tobamovirus that posing a significant threat to the yield and quality of solanaceous crops globally. To elucidate the regulatory role of small RNAs in the host response to ToMMV, we used Nicotiana tabacum var. Xanthi nc as a model system and performed small RNA sequencing across three distinct disease stages: mosaic symptom onset, linear leaf symptom development, and the recovery phase. By integrating qRT-PCR validation with functional assays, including miRNA overexpression, virus-induced gene silencing, and biological inoculation, several miRNAs specifically associated with ToMMV resistance were identified.

Results

Functional characterization revealed that overexpression of miR171c, miR172a, and a novel miRNA (novel-miR363) significantly suppressed ToMMV accumulation and thereby enhancing plant resistance, whereas silencing of miR171c resulted in an approximately threefold increase in viral accumulation. In contrast, overexpression of another novel miRNA, novel-miR1022, promoted viral accumulation and facilitated viral infection. Further analysis through target prediction and experimental validation confirmed that miR171c and miR172a target transcripts encoding Scarecrow-like (SCL) and APETALA2 (AP2) class transcription factors, respectively. Specifically, miR171c appears to confer resistance by repressing NtSCL6 and NtSCL22, whereas miR172a likely mediates this effect through the downregulation of NtRAP2-7.

Conclusion

These findings elucidate key miRNA-mediated regulatory modules governing ToMMV resistance, advancing the understanding of host defense mechanisms against tobamoviruses.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12870-026-09581-9.

Background

MicroRNAs (miRNAs) are a class of endogenous non-coding RNAs, typically 21–24 nucleotides in length, that regulate gene expression through target cleavage, translational inhibition, and RNA-dependent DNA methylation. Notably, the first two mechanisms function post-transcriptionally by guiding the RNA-induced silencing complex (RISC) to complementary mRNAs sequences [1–3]. Increasing evidence highlights the pivotal roles of miRNAs in plant immunity against pathogens [4]. For instance, during antiviral defense in rice, miR319 targets TCP21 to suppress jasmonic acid‑mediated immunity, thereby facilitating rice ragged stunt virus (RRSV) infection [5]. Furthermore, RRSV manipulates host development by dynamically regulating miR164, miR319 and miR156, along with their target genes (OsCUC1, OsTCP1 and OsSPL14/OsSPL17), to alter leaf morphogenesis [6]. Similarly, in tobacco (Nicotiana tabacum), miR172 has been shown to negatively regulate resistance to tobacco mosaic virus (TMV) by targeting the AP2 transcription factor TOE3, conversely, silencing miR172 enhances resistance [7]. Concurrently, tobacco miR171c modulates developmental plasticity by negatively regulating SCL genes to maintain apical dominance [8]. Despite the established roles of the miR171 and miR172 families in developmental and stress adaptation across plant species [7–10], their regulatory functions in response to tomato mottle mosaic virus (ToMMV) infection in tobacco remain uncharacterized.

ToMMV is an emerging tobamovirus initially identified in Mexican tomatoes in 2013, and has since spread rapidly worldwide, causing severe yield and quality losses in solanaceous crops [11–13]. The virus is transmitted mechanically and via contaminated seeds, soil and water, and infects a broad range of plants including tomato, pepper and tobacco [14, 15]. Infected tobacco plants exhibit severe systemic symptoms, including mosaic, blistering, deformation, linear leaf distortion, necrosis, and shortened growth cycles [16–18]. Ultrastructural analyses revealed that ToMMV compromises organelle integrity, leading to the depletion of the mitochondrial matrix and cristae, as well as severe disruption of chloroplast membranes [19].

To elucidate the role of small RNAs in the host response to ToMMV, Nicotiana tabacum var. Xanthi nc was used as a model system, with integrated small RNA and transcriptome sequencing conducted at three distinct symptomatic stages (mosaic, linear leaf, and recovery) after ToMMV inoculation. By combining qRT‑PCR validation with functional assays, including transient and systemic miRNA overexpression, virus‑induced gene silencing, and biological inoculation, several miRNAs associated with ToMMV resistance were identified. Notably, we demonstrated that miR171c and miR172a confer enhanced resistance by targeting SCL and AP2 transcription factors, respectively. Collectively, these findings uncover previously uncharacterized miRNA‑mediated regulatory modules involved in antiviral defense and provide insights into host resistance mechanisms against ToMMV.

Materials and methods

Plant materials and processing method

N. tabacum var. Xanthi nc plants were inoculated with the infectious clone pCB301-ToMMV, and leaf samples were collected at three distinct disease symptom stages: (1) 15 days post-inoculation (dpi), from initially symptomatic leaves showing mosaic symptoms (designated Nt1; Fig. 1A); (2) 30 dpi, from upper leaves exhibiting linear leaf symptoms (designated Nt2; Fig. 1B); (3) 45 dpi, from newly emerged leaves displaying mottling, mosaic, or blistering (rather than linear symptoms; designated Nt3; Fig. 1C). Control groups (CK1, CK2, CK3; Fig. 1D-F) were collected at the corresponding time points from N. tabacum var. Xanthi nc plants inoculated with the empty vector pCB301. For each treatment, leaf tissue was pooled from three individual plants to minimize biological variability, and three independent biological replicates were prepared per group. Immediately after collection, all samples were flash-frozen in liquid nitrogen for 30 min and stored at -80 °C. Upon completion of sampling across all three symptomatic stages, the samples were submitted to Shanghai Majorbio Bio-pharm Biotechnology Co., Ltd for combined transcriptome and small RNA sequencing.

Fig. 1.

Symptomatic progression, tissue sampling locations and protocols in Nicotiana tabacum var. Xanthi nc post-ToMMV inoculation at three intervals. A At 15 days post ToMMV inoculation (dpi), initial symptomatic leaves exhibited mosaic, marking the first sampling time point. B At 30 dpi, leaves exhibiting linear leaf symptoms in the upper foliage, marking the second sampling time point. C At 45 dpi, new leaves displaying mottling, mosaic, or blistering instead of linear leaves, marking the third sampling time point. D-F N. tabacum var. Xanthi nc inoculated with the pCB-301 empty vector at 15, 30, and 45 dpi, respectively, as controls

Small RNA sequencing and bioinformatics analysis

Total RNA was extracted from 18 samples using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). RNA concentration and purity were assessed using a Nanodrop 2000 spectrophotometer, while RNA integrity was evaluated by agarose gel electrophoresis. The RNA integrity number (RIN) was determined using an Agilent 2100 Bioanalyzer system. Small RNA libraries were constructed by Shanghai Majorbio Bio-pharm Technology Co., Ltd., and sequencing was performed on the Illumina HiSeq 2500 platform. Raw sequencing data were subjected to stringent quality control, including the removal of low-quality reads, sequences containing ambiguous bases, adapter-free reads, and reads outside the 18–30 nucleotide (nt) length range, yielding high-quality clean reads. Clean reads were aligned against the Repbase and Rfam databases using Bowtie to filter out non-coding RNAs and repetitive sequences, retaining unannotated sequences potentially harboring miRNAs. For miRNA identification, clean reads (18–30 nt) were mapped to precursor miRNAs (pre-miRNAs) and mature miRNAs in the miRBase database using miRDeep2 to annotate known miRNAs. Unannotated small RNA sequences were aligned to the N. tabacum reference genome, followed by secondary structure prediction, Dicer cleavage site identification, and minimum free energy calculation to characterize novel miRNAs. Statistical analyses were performed on the sequence features, length distribution, and precursor information of the newly predicted miRNAs.

To identify differentially expressed miRNAs (DEMs) across three stages of ToMMV infection in N. tabacum var. Xanthi nc compared to controls, transcriptomic data of known and novel miRNAs were analyzed using DESeq2, which employs a negative binomial distribution model for significance testing. Predicted target genes of both known and novel miRNAs were identified using computational tools (TargetFinder, psRobot, and RNAhybrid), followed by functional annotation via EggNOG, Gene Ontology (GO), and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases.

Validation of differentially expressed miRNAs by qRT-PCR

To elucidate the regulatory mechanisms of miRNAs in N. tabacum var. Xanthi nc during ToMMV infection, particularly in the linear-leaf formation stage, eight significantly differentially expressed miRNAs (miR171c, miR172a, miR408, novel-miR516, novel-miR1022, novel-miR432, novel-miR363, and novel-miR667) were selected for validation. Stem-loop reverse transcription (RT) primers (Table S1) were designed using the Vazyme miRNA Primer Design Tool (https://www.vazyme.com/gongju.html), followed by cDNA synthesis from total RNA using miRNA 1st Strand cDNA Synthesis Kit (by stem-loop) (Nanjing Vazyme Biotech Co., Ltd.). Quantitative PCR (qPCR) was performed in triplicate (three biological replicates with three technical replicates each) using a Roche LightCycler^® 480 system with ChamQ Universal SYBR qPCR Master Mix (Vazyme Biotech Co., Ltd.), using the actin gene of N. tabacum as an endogenous control. Relative expression levels were calculated using the 2^−ΔΔCt method, with statistical analysis conducted using SPSS 19.0 and GraphPad Prism 8.

Cloning and sequencing analysis of miRNA precursors

Based on the precursor sequences of miR171c, miR172a, miR408, novel-miR516, novel-miR1022, novel-miR432, novel-miR363, and novel-miR667, gene-specific amplification primers were designed using Primer Premier 6.0 (Table S2). The full-length miRNA precursors were amplified via PCR using cDNA templates derived from N. benthamiana and N. tabacum var. Xanthi nc leaves at the linear-leaf formation stage following ToMMV infection. PCR amplification was performed using Takara PrimeSTAR^® GXL DNA Polymerase according to the manufacturer’s protocol. The PCR products were purified by agarose gel electrophoresis (AGE), cloned into the pMD19-T Vector (Takara Bio Inc.), and transformed into Escherichia coli DH5α competent cells. Positive clones were screened by colony PCR, and validated plasmids were sent for Sanger sequencing (Beijing Tsingke Biotech Co., Ltd.). For evolutionary analysis, mature miRNA sequences of the miR171, miR172, and miR408 families were retrieved from miRBase (Release 22.1, https://www.mirbase.org/), including homologs from soybean (Glycine max), maize (Zea mays), peanut (Arachis hypogaea), A. thaliana, rice (Oryza sativa), S. lycopersicum, and grape (Vitis vinifera). Multiple sequence alignment was conducted using MEGA7, and a maximum-likelihood (ML) phylogenetic tree was constructed with 1000 bootstrap replicates. Additionally, the secondary stem-loop structures of miR171c, miR172a, miR408, and novel miRNA precursors were predicted using the RNAfold WebServer (http://rna.tbi.univie.ac.at/cgi-bin/RNAWebSuite/RNAfold.cgi), based on transcriptome-derived precursor sequences.

Construction of overexpression and silencing vectors

For the construction of miRNA overexpression vectors (exemplified by miR171c), we performed overlap extension PCR using plasmid pRS300 containing the A. thaliana miR319a precursor as template. Through sequential PCR amplifications with primer combinations I (amiR-miR171s), II (amiR-miR171a), III (amiR-miR171*s), IV (amiR-miR171*a), A and B (Table S3), we precisely replaced the 20-bp endogenous miR319a mature sequence in the precursor structure with a designed 21-bp artificial miR171c (amiR171c) sequence. The initial PCR generated three distinct fragments: (a) the 5’ arm containing the modified miRNA sequence, (b) the central loop region, and (c) the 3’ arm, which were subsequently gel-purified, quantified, and mixed in equimolar ratios as templates for a second overlap PCR using primers A and B to generate the final insert (fragment d). This product was then cloned into the pRS300 backbone, transformed into E. coli DH5α competent cells, and sequence-verified before being subcloned into PCV and PVX vectors through homologous recombination using specific primers (Table S4), yielding constructs for transient (PCV-based) and systemic (PVX-based) miRNA overexpression, respectively.

For construction of the silencing vector using miR171c as an example, we designed a short tandem target mimic (STTM-nta-miR171c) structure, which was synthesized by BGI Genomics Co., Ltd. through whole-gene synthesis. The transformed bacterial stab culture was streaked onto LB agar plates, and 800 µL of ampicillin-containing LB medium (100 µg/mL) was aliquoted into 1.5 mL microcentrifuge tubes for single colony inoculation, followed by incubation at 37 °C with shaking at 220 rpm. Positive clones containing the silencing fragment were identified by colony PCR using specific primers (Table S4) and subsequently expanded in liquid culture under the same conditions. Plasmid DNA was extracted from the positive E. coli cultures using the SanPrep Column Plasmid Mini-Preps Kit (Sangon Biotech Co., Ltd.), with concentration determined by Nanodrop 2000 spectrophotometer and integrity verified by 0.8% AGE. The recipient vectors PMV-STTM171c and PVX were digested with Cla I/Sal I, while PCV was linearized with Apa I. The purified miRNA overexpression fragment and STTM silencing fragment were independently ligated into their respective digested vectors. The ligation products were transformed into E. coli DH5α, and verified positive clones were subsequently transformed into Agrobacterium tumefaciens strain GV3101 for plant transformation.

Agrobacterium-mediated inoculation

The A. tumefaciens culture was streaked onto LB solid medium supplemented with 50 µg/mL kanamycin (Kan) and 50 µg/mL rifampicin (Rif), followed by incubation at 28 °C for 36–48 h. A single colony was selected and inoculated into LB liquid medium containing 50 µg/mL Kan and 50 µg/mL Rif, and cultured until the OD₆₀₀ reached 0.6–0.8. Following verification of the target gene, bacterial cells from 2 mL of culture were harvested by centrifugation at 3,500 × g for 5 min. The supernatant was discarded, and the bacterial pellet was resuspended in sterile ddH₂O containing 10 mmol/L MgCl₂, 10 mmol/L MES (pH 5.6), and 200 µmol/L acetosyringone (Ace). The suspension was adjusted to an OD₆₀₀ of 0.6-1.0 and incubated in the dark at room temperature for 3 h. Leaves of plants to be inoculated were punctured using a 1 mL syringe needle, and the Agrobacterium suspension was infiltrated into the abaxial side of the punctured area using the needle-free syringe. Each plant received approximately 0.3 mL of the suspension, with a minimum of 10 plants per treatment. Post-inoculation, plants were transferred to a greenhouse maintained at 25 °C under a 16 h light/8 h dark cycle for subsequent growth and analysis.

Prediction, sequence analysis, and expression analysis of miR171c and miR172a target genes

Based on the high complementarity between plant miRNAs and their target genes, potential targets of miR171c and miR172a were predicted using the online software psRNATarget (https://www.zhaolab.org/psRNATarget/) against the N. benthamiana database [20]. Subsequently, the candidate target genes were amplified for sequence analysis and quantified by qRT-PCR using the primers listed in Table S5. Bioinformatics analyses were performed using multiple platforms (EggNOG, GO, KEGG) for functional annotation of predicted targets.

Results

Overview of small RNA sequencing data analysis

Small RNA sequencing generated a total of 240.52 M high-quality raw reads, with each sample yielding at least 10.01 M reads and Q30 scores exceeding 87.29% (Table S6). Alignment of clean reads to the N. tabacum miRNA reference genome revealed robust mapping rates of 54.91% (Nt1) vs. 44.95% (CK1), 56.38% (Nt2) vs. 54.86% (CK2), and 54.05% (Nt3) vs. 48.85% (CK3), thereby ensuring data reliability for downstream analysis. Correlation analysis confirmed excellent reproducibility among biological triplicates, with Pearson correlation coefficients > 0.9 within both the ToMMV-infected (Nt1-Nt3) and control (CK1-CK3) groups (Fig. 2A). Principal component analysis (PCA) revealed that PC1 and PC2 accounted for 71.96% and 14.88% of the total transcriptional variance, cumulatively explaining 86.84%. Biological replicates within the CK and Nt groups clustered compactly, and the two cohorts were distinctly segregated along the PC1 axis, indicating strong transcriptomic changes induced by ToMMV infection with distinct stage-specific patterns (Fig. 2B). A total of 460 mature miRNAs were identified in Nt1, comprising 153 known and 307 novel miRNAs, compared to 419 miRNAs (151 known, 268 novel) in the corresponding control. Similarly, Nt2 yielded 459 miRNAs (152 known, 307 novel) vs. 420 miRNAs (154 known, 266 novel) in the control, while Nt3 contained 460 miRNAs (153 known, 307 novel) vs. 416 miRNAs (148 known, 268 novel) in the control. This conserved distribution pattern across experimental groups confirms the robustness of miRNA profiling.

Fig. 2.

The small RNA sequencing data analysis of Nicotiana tabacum var. Xanthi nc inoculated with ToMMV. A Correlation analysis of small RNA sequencing data between ToMMV-inoculated treatment groups (Nt1, Nt2, Nt3) and mock-inoculated controls (CK1, CK2, CK3). Sample/group names are labeled on the top and diagonal; colored squares indicate the correlation between two samples; higher values denote stronger correlation and greater similarity, reflecting good biological reproducibility. B Principal component analysis of transcriptome profiles. Sample point distances reflect inter-sample distances; shorter distances indicate higher similarity and better biological replicates. The x-axis shows the contribution of PC1 to sample separation, and the y-axis shows the contribution of PC2. C Bar chart of differentially expressed miRNAs in ToMMV-inoculated treatment groups versus mock-inoculated controls. The x-axis represents the different comparison groups, and the y-axis represents the number of differentially expressed miRNAs; cutoff: p<0.05, |FC|>2

Comparative transcriptomic analysis revealed 310 DEMs in Nt1, with 207 significantly upregulated and 103 downregulated. Similarly, 185 DEMs were identified in Nt2, including 126 upregulated and 59 downregulated, while Nt3 exhibited 281 DEMs with 192 upregulated and 89 downregulated (Fig. 2C).

Bioinformatic prediction of miRNA targets and pathway enrichment across distinct symptomatic stages of ToMMV-infected N. tabacum var. Xanthi nc

Given the established role of miRNAs in the post-transcriptional regulation of multiple target genes via complex regulatory networks, functional characterization of the identified DEMs was pursued through systematic target prediction and annotation. Target prediction analysis yielded 2,893 putative targets for Nt1 DEMs (1,289 annotated), 2,006 for Nt2 DEMs (908 annotated), and 2,987 for Nt3 DEMs (1,308 annotated). GO analysis suggested broader functional diversity in Nt1 and Nt3, with significantly more annotated targets across various GO categories compared to Nt2 (Fig. 3A), implying more extensive transcriptional reprogramming during the mosaic and late symptomatic phases relative to the linear leaf formation stage.

Fig. 3.

Functional enrichment of predicted miRNA targets and expression profiles of representative miRNAs during ToMMV infection in Nicotiana tabacum var. Xanthi nc. A Functional annotation of differentially expressed miRNA target gene GO in different stages. B-D KEGG pathway enrichment of differentially expressed miRNA target genes in three stages (p<0.05). E Venn plot analysis of common and unique sequences between N. tabacum var. Xanthi nc samples and control samples in three stages. F Relative expression of eight differentially expressed miRNAs in ToMMV-infected N. tabacum var. Xanthi nc at the linear leaf symptom stage. Data were presented as mean ± SD (n=3). ns, not significant; *p<0.05; **p<0.01 were based on Student's t-test

KEGG pathway enrichment analysis provided evidence for stage-specific metabolic perturbations during ToMMV infection: Nt1 showed significant enrichment in glycerolipid metabolism, sphingolipid metabolism, fatty acid biosynthesis, and mitophagy-yeast (Fig. 3B); Nt2 displayed distinct alterations in meiosis-yeast, ascorbate and aldarate metabolism, fatty acid biosynthesis, ubiquinone and other terpenoid-quinone biosynthesis, amino sugar and nucleotide sugar metabolism, along with cell cycle-yeast pathways (Fig. 3C); and Nt3 exhibited reactivation of Nt1-associated pathways (glycerolipid metabolism, sphingolipid metabolism, fatty acid biosynthesis, and mitophagy-yeast), accompanied by persistent changes in the mRNA surveillance pathway and meiosis-yeast (Fig. 3D). These results suggest a dynamic yet partially conserved metabolic reprogramming across different infection stages.

Notably, fatty acid biosynthesis emerged as a consistently dysregulated pathway throughout all three stages, whereas glycerolipid metabolism and mitophagy‑yeast showed recurrent patterns in both early (Nt1) and late (Nt3) infection phases. These computational findings provide a rationale for hypothesizing the potential roles of these pathways in viral pathogenesis and host defense.

Screening and identification of miRNAs in ToMMV-infected N. tabacum var. Xanthi nc at the linear leaf formation stage

Comparative analysis of DEMs across the three infection stages identified 134 common miRNAs (37.22% of total DEMs), encompassing both known and novel miRNAs. Stage-specific DEMs included 39 (10.83%) unique to Nt1, 22 (6.11%) to Nt2, and 17 (4.72%) to Nt3 (Fig. 3E), suggesting their potential roles in mediating distinct plant-virus interactions and symptom development during each phase. Notably, eight DEMs were exclusively associated with the linear leaf formation stage, including three conserved miRNAs (nta-miR171c, nta-miR172a, and nta-miR408) and five novel miRNAs (novel-miR363, novel-miR432, novel-miR516, novel-miR667, and novel-miR1022), implicating their potential regulatory involvement in this specific symptomatic response.

To validate the impact of ToMMV on these eight miRNAs in N. tabacum var. Xanthi nc, samples were collected at approximately 30 dpi during the linear leaf formation, including both ToMMV-infected leaves and empty vector-inoculated controls (four biological replicates for each group). qRT-PCR analysis revealed significant transcriptional changes: nta-miR171c, nta-miR172a, nta-miR408, novel-miR363, and novel-miR516 were markedly upregulated, with expression levels approximately 1.5-, 3.7-, 10-, 40-, and 6-fold higher than controls, respectively. Conversely, novel-miR432, novel-miR667, and novel-miR1022 were downregulated, exhibiting about 50%, 50%, and 70% of control levels, respectively (Fig. 3F). These results indicate that ToMMV infection significantly alters the expression profiles of these eight miRNAs during linear leaf formation in N. tabacum var. Xanthi nc, implying their potential contribution to either viral pathogenesis or host resistance.

Validation of miRNA overexpression and silencing effects on ToMMV infection in N. benthamiana

To elucidate the functional roles of these eight miRNAs during ToMMV infection, we successfully generated both transient and systemic overexpression vectors, with the exception of novel-miR516 (which failed in transient overexpression vector construction), the seven successful PCV-amiRNA constructs were agroinfiltrated into N. benthamiana leaves along with the empty vector control (PCV). At 2 dpi, qRT-PCR confirmed robust overexpression, ranging from approximately 1.5-fold (miR408, novel-miR1022), 2.5-fold (miR172a and novel-miR667), and 2.7-3.0-fold (miR171c, novel-miR363, novel-miR432) relative to empty vector controls (Fig. 4A). Co-infiltration of these PCV-amiRNA constructs with ToMMV infectious clones revealed distinct effects on viral accumulation at 2 dpi: overexpression of miR171c, miR172a, and novel-miR363 reduced ToMMV RNA levels by 40–50%, while novel-miR432 and novel-miR1022 enhanced accumulation by 2.7- and 4-fold, respectively. No significant changes were observed for miR408 and novel-miR667 (Fig. 4B).

Fig. 4.

Validation of the effects of miRNA overexpression and silencing on ToMMV infection in Nicotiana benthamiana by qRT-PCR and phenotypic analysis. A,C qRT-PCR analysis of the relative expression levels of miRNAs in N. benthamiana plants following transient (A) or systemic (C) overexpression versus controls. B,D qRT-PCR quantification of ToMMV accumulation in N. benthamiana following transient (B) or systemic (D) overexpression of individual miRNAs and subsequent viral inoculation. E Phenotypic symptoms of systemic miRNA overexpression in N. benthamiana following ToMMV infection at 10 dpi versus controls. F Analysis of miR171c silencing efficiency during the time course of 7 to 10 days post-transformation in N. benthamiana. G Relative ToMMV expression in N. benthamiana with PVX-mediated nta-miR171c silencing versus controls at 10 dpi. H Phenotype in PVX-STTM171c-silenced N. benthamiana at 10 dpi with ToMMV. Note: All data were presented as mean±SD (n=4). ns, not significant; *p<0.05; **p<0.01; ***p<0.001 were based on Student's t-test

To assess systemic overexpression, PVX-amiRNA constructs were infiltrated, with PVX-GUS as control. Systemic expression analysis showed minimal miRNA overexpression at 7 dpi (0.8-1.2-fold), but significant induction at 10 dpi (1.5-2.8-fold) (Fig. 4C), establishing this later timepoint as optimal for subsequent ToMMV challenge assays. At this stage, overexpression miR171c, miR172a, miR408, novel-miR363, and novel-miR432 suppressed viral accumulation to 50.9%, 39.4%, 63%, 41%, and 79.5% of control levels, respectively. Conversely, overexpression of novel-miR516, novel-miR667, and novel-miR1022 enhanced viral accumulation by 1.4, 6.2, and 3-fold (Fig. 4D). Phenotypically, these accumulation patterns closely mirrored symptom severity. N. benthamiana plants overexpressing miR171c or miR172a exhibited only mild chlorosis, while those overexpressing miR408, novel-miR363, or novel-miR432 displayed more pronounced chlorosis. In contrast, plants overexpressing novel-miR516, novel-miR667 or novel-miR1022, developed systemic chlorosis and mosaic symptoms in newly emerging leaves (Fig. 4E). Integrated analyses from both transient and systemic platforms consistently demonstrated that ectopic expression of miR171c, miR172a, and novel-miR363 suppresses ToMMV accumulation, implying their potential roles in enhancing host antiviral resistance. Conversely, novel-miR1022 exhibited a pronounced proviral effect, facilitating viral infection. These opposing regulatory phenotypes were robustly reproduced across four independent biological replicates, confirming the reliability of these functional roles in modulating plant-virus interactions.

To further validate the antiviral capacity of miR171c, we employed a silencing approach using PVX-STTM171c. qRT-PCR analysis confirmed progressive knockdown, with endogenous miR171c expression levels decreasing to 70% and 40% of PVX-GUS controls at 7 and 10 dpi, respectively, with optimal silencing achieved by 10 dpi (Fig. 4F). Challenging the miR171c-silenced N. benthamiana plants with ToMMV at 10 dpi resulted in a significant 3-fold increase in viral accumulation compared to controls (p < 0.01; Fig. 4G), accompanied by characteristic chlorosis in systemic leaves (Fig. 4H), which demonstrated that downregulation of miR171c in N. benthamiana enhanced viral susceptibility. These findings complemented and extended our overexpression results, collectively establishing additional evidence that miR171c acts to restrict ToMMV infection and may contribute fundamentally to antiviral defense mechanisms in N. benthamiana.

Sequence analysis and phylogenetic comparison of miR171c and miR172a

Comprehensive sequence analysis revealed contrasting characteristics between the miR171 and miR172 families in N. tabacum var. Xanthi nc. The miR171 family comprises three members (nta-miR171a/b/c), while the miR172 family encompasses ten isoforms (nta-miR172a-k). WeBLOGO analysis of the mature sequences demonstrated substantial nucleotide variation among nta-miR171 members (Fig. 5A), whereas nta-miR172 sequences exhibited high conservation with minimal polymorphisms (Fig. 5B).

Fig. 5.

Sequence analysis and phylogenetic comparison of miR171c and miR172a. A,B Base conservation analysis of mature sequence of miR171(A) and miR172 (B) family members in N. tabacum var. Xanthi nc. C,D Comparative sequence alignment of miR171c (C) and miR172a (D) from various species

Phylogenetic reconstruction based on 165 miR171 sequences from 27 plant genera and 250 miR172 sequences from 44 genera further elucidated their evolutionary relationships. nta-miR171c clustered most closely with orthologs from osi-miR171c of Oryza sativa, bdi-miR171c of Brachypodium distachyon, ppe-miR171c of Prunus persica, and vvi-miR171c of Vitis vinifera (Fig. 5C). Meanwhile, nta-miR172a grouped primarily with orthologs from Solanaceae members stu-miR172a of Solanum tuberosum and sly-miR172a of S. lycopersicum, as well as monocots osa-miR172a of O. sativa, zma-miR172a of Zea mays, sbi-miR172a of Sorghum bicolor, and ppe-miR172a of P. persica (Fig. 5D). These conserved phylogenetic patterns strongly imply functional preservation of these miRNA families across angiosperms.

Prediction and sequence analysis of miR171c and miR172a target genes

Bioinformatic analysis using target prediction algorithms identified 12 and 45 putative target genes for miR171c and miR172a, respectively. GO and KEGG pathway analyses revealed distinct functional categories for these targets: miR171c predominantly regulated Scarecrow-like (SCL) proteins and ABC transporters, while miR172a targets include AP2-like ethylene-responsive transcription factor TOE3, ubiquitin receptor RAD23b-like, floral homeotic protein APETALA 2-like, 1,4-dihydroxy-2-naphthoyl-CoA thioesterase 1-like, and protein PHR1-LIKE 1-like (Table 1). The top-ranked predicted target genes, namely NtSCL6 (TC134811) and NtSCL22 (TC127385) for miR171c, and RAP2-7 (EH368241) for miR172a, were selected for further validation via PCR amplification. The resulting amplified products for these three target genes were 3126 bp, 2897 bp, and 4570 bp in length, respectively.

Table 1.

Target genes predict for differentially express miR171c and miR172a

miRNA	target gene name	target gene description
nta-miR171c	LOC107763769	ABC transporter C family member 10-like
	LOC107810886	ABC transporter C family member 10-like
	LOC107788613	DEAD-box ATP-dependent RNA helicase 32-like
	LOC107759671	scarecrow-like protein 15
	LOC107769694	scarecrow-like protein 15
	LOC107802311	scarecrow-like protein 22
	LOC107807449	scarecrow-like protein 22
	LOC107810717	scarecrow-like protein 22
	LOC107763536	scarecrow-like protein 6
	LOC107763537	scarecrow-like protein 6
	LOC107807450	scarecrow-like protein 6
nta-miR172a	LOC107812324	AP2-like ethylene-responsive transcription factor TOE3
	LOC107759353	AP2-like ethylene-responsive transcription factor TOE3
	LOC107784366	AP2-like ethylene-responsive transcription factor TOE3
	LOC107783385	AP2-like ethylene-responsive transcription factor TOE3
	LOC107819773	AP2-like ethylene-responsive transcription factor TOE3
	LOC107810888	AP2-like ethylene-responsive transcription factor TOE
	LOC107778909	ethylene-responsive transcription factor RAP2-7-like
	LOC107796949	ethylene-responsive transcription factor RAP2-7-like
	LOC107818199	ethylene-responsive transcription factor RAP2-7-like
	LOC107776994	floral homeotic protein APETALA 2-like
	LOC107804251	floral homeotic protein APETALA 2-like
	LOC107803429	floral homeotic protein APETALA 2-like
	LOC107820989	floral homeotic protein APETALA 2-like
	LOC107791419	protein NRT1/ PTR FAMILY 5.10-like
	LOC107780079	protein NRT1/ PTR FAMILY 5.10-like
	LOC107817759	protein PHR1-LIKE 1-like

Phylogenetic analysis was conducted to characterize the evolutionary relationships of the predicted miR171c targets, NtSCL6 (TC134811) and NtSCL22 (TC127385). This analysis incorporated 16 sequences spanning N. tabacum (two sequences), N. tomentosiformis, N. sylvestris, S. lycopersicum, S. tuberosum, S. pennellii, S. verrucosum, C. annuum, Lycium barbarum, L. ferocissimum, Ipomoea nil, I. triloba, Salvia miltiorrhiza, Apium graveolens, and Arabidopsis lyrata subsp. lyrata. The resulting phylogenetic tree demonstrated that target gene TC127385 shares the closest evolutionary relationship with N. tabacum SCL22 and N. tomentosiformis SCL6. Conversely, TC134811 exhibited the highest sequence similarity to the N. tabacum SCL6 transcription factor (Fig. 6A). Similarly, phylogenetic reconstruction was performed for the miR172a target, EH368241, using 12 sequences from N. tabacum (two sequences), N. attenuata, N. tomentosiformis, N. sylvestris, C. annuum, S. dulcamara, S. lycopersicum, S. stenotomum, S. tuberosum, Chenopodium quinoa and Amaranthus tricolor. This analysis revealed EH368241 clustering within the clade of AP2-like homologous proteins from Solanaceae plants (Fig. 6B). These findings highlight the strong evolutionary conservation of these regulatory targets among Solanaceae species.

Fig. 6.

Sequence, binding site, and functional analysis of miR171c and miR172a target genes. A,B Systematic evolutionary analysis of the target genes TC134811 and TC127385 of miR171c (A), and EH368241 of miR172a (B). Note: The black filled triangle, square and circle represent TC134811, TC127385 and EH368241 from this study. Using MEGA7 software and neighborhood method, establish a tree with 1000 repeated samples. C Binding sites of miR171c and miR172a on their target genes. D Relative expression of miR171c/miR172a target genes in ToMMV-infected N. tabacum var. Xanthi nc at the linear leaf symptom stage (30 dpi). E qRT-PCR analysis of relative target gene expression in N. benthamiana in response to transient overexpression of miR171c or miR172a. Note: Data were presented as mean±SD (n=3). ***p<0.001 were based on Student's t-test in Fig.6D and 6E

Functional analysis of miR171c and miR172a target gene binding sites and their regulatory roles

Sequence alignment revealed perfect complementarity between miR171c and its putative target genes, NtSCL6 and NtSCL22, as well as between miR172a and NtRAP2-7. Notably, complete base pairing was observed from the first nucleotide, with no mismatches at the critical 10–11 nucleotide positions, which are essential for miRNA-mediated cleavage (Fig. 6C). GO annotation confirmed that all three target genes encode proteins involved in DNA binding and transcription factor activity, consistent with their predicted roles as transcription regulators (Table 2). To assess the impact of ToMMV infection on these miRNA-regulated target genes, N. tabacum var. Xanthi nc plants were agroinfiltrated with ToMMV. At 30 dpi, qRT-PCR analysis of RNA extracted from both infected and mock-inoculated plants revealed that ToMMV infection significantly downregulated the expression of NtSCL6, NtSCL22 and NtRAP2-7, reducing their transcript levels to approximately 30% of the control groups (Fig. 6D), suggesting that ToMMV may suppress the expression of these miR171c- and miR172a-regulated genes. To further validate the negative regulatory relationships between these miRNAs and their targets, miR171c and miR172a were transiently overexpressed in N. benthamiana leaves. qRT-PCR at 2 dpi revealed significant downregulation of their respective target genes NtSCL6, NtSCL22 and NtRAP2-7 (Fig. 6E). Collectively, these findings validate the predicted target specificity and confirm the negative post-transcriptional regulation of these genes by miR171c and miR172a.

Table 2.

Target Gene GO Function

Gene	Biological Process	Molecular Function	Cellular Component
TC127385	regulation of DNA-templated transcription (GO:0006355)	sequence-specific DNA binding (GO:0043565);	nucleus (GO:0005634)
		DNA-binding transcription factor activity(GO:0003700).
TC134811	regulation of DNA-templated transcription (GO:0006355)	sequence-specific DNA binding (GO:0043565);	nucleus (GO:0005634)
		DNA-binding transcription factor activity(G0:0003700).
EH368241	regulation of DNA-templated transcription (G0:0006355)	DNA binding(GO:0003677); DNA-binding transcription factor activity(GO:0003700).	-

Discussion

Small RNA sequencing analysis of N. tabacum var. Xanthi nc across three symptomatic stages of ToMMV infection revealed substantial reprogramming of the host miRNA landscape. Comparative profiling identified numerous DEMs at all infection stages, with upregulated miRNAs predominating (207/310 in Nt1, 126/185 in Nt2, and 192/281 in Nt3). Notably, the majority of these DEMs corresponded to novel miRNAs rather than previously characterized sequences. These observations are largely consistent with previous reports on miRNA expression alterations in tobacco following potato virus Y (PVY) infection [21]. The prevalence of novel miRNAs suggests that ToMMV infection induces extensive and stage‑specific remodeling of the endogenous miRNAome. However, the functional roles of these novel miRNAs remain largely unknown and warrant further investigation.

GO and KEGG enrichment analyses were performed on the computationally predicted targets of DEMs. The results indicated that the formation and transition of mosaic and linear leaf symptoms may be associated with dysregulated pathways, including mitophagy, glycerolipid metabolism, and fatty acid biosynthesis. Mitophagy can induce ATP deficiency and excessive accumulation of reactive oxygen species (ROS), and thereby disrupting of photorespiration and carbon metabolism. Such disruptions potentially reduce cell division rates, inhibiting leaf expansion, and impair photosynthetic efficiency [22]. Furthermore, as precursors of glycerolipids, fatty acids are critical structural components of thylakoid membranes and directly influence photosynthetic performance [23].

Plants employ miRNAs as critical regulators of antiviral defense by post-transcriptionally repressing target genes. In this study, our findings demonstrate that overexpression of miR171c or miR172a significantly suppressed ToMMV accumulation, whereas silencing of miR171c increased viral titer by approximately threefold. Experimental validation suggested that miR171c directly represses NtSCL6 and NtSCL22, and miR172a represses NtRAP2‑7 (Fig. 6E). These results are consistent with previous reports on miRNA‑mediated antiviral immunity. For example, in wheat, infection by Puccinia striiformis f. sp. tritici induces the upregulation of miR408, which activates lignin biosynthesis and hypersensitive response through its target TaCLP1, a positive regulator [24]. In tobacco, miR172 negatively regulates resistance to TMV by targeting TOE3, accordingly, silencing of miR172 enhances disease resistance [7]. Similarly, in rice, overexpression of osa‑miR171b reduces susceptibility to rice stripe virus (RSV), accompanied by reciprocal regulation of its targets OsSCL6-IIa/b/c [10]. Our study reveals a novel role for the miR171c‑NtSCL6/NtSCL22 and miR172a‑NtRAP2‑7 modules in conferring resistance to ToMMV. Notably, the same miR171c‑SCL regulatory axis has previously been implicated in governing developmental phenotypes, such as the loss of apical dominance in tobacco [8], suggesting that this specific signaling module can be co‑opted by the host to mediate both developmental processes and antiviral defense responses.

Previous studies have demonstrated that viral infections can manipulate host miRNA networks to alter leaf morphogenesis. For instance, RRSV infection in rice dynamically regulates the expression of miR164, miR319, miR156, along with their targets (OsCUC1, OsTCP1, OsSPL14/OsSPL17), thereby disrupting normal leaf development [6]. Similarly, in tomato, infection with tomato leaf curl new delhi virus (ToLCNDV) significantly upregulates miR159/319 and miR172, suggesting their involvement in the development of leaf curl symptoms [9]. In our previous ultrastructural study, we also observed that the formation of various leaf malformations in ToMMV‑infected tobacco is primarily associated with severe chloroplast damage [19]. In the present study, although miR171c and miR172a were shown to modulate ToMMV accumulation, it remains unclear whether they directly contribute to the development of these leaf symptoms via chloroplast‑related pathways. This potential mechanistic link warrants further investigation.

Conclusions

By profiling small RNAs during ToMMV infection in N. tabacum cv. Xanthi nc, we identified eight miRNAs specifically responsive at the linear leaf stage. Functional characterization revealed that miR171c, miR172a, and novel-miR363 enhance host resistance, whereas novel-miR1022 facilitated viral infection. Mechanistically, miR171c and miR172a mediate defense by repressing NtSCL6, NtSCL22 and NtRAP2-7, respectively. Collectively, these findings uncover the miR171c-NtSCL6, miR171c-NtSCL22 and miR172a-NtRAP2-7 axes as critical regulatory modules in the ToMMV-tobacco interaction, offering new insights and potential targets for engineering virus-resistant crops.

Authors’ contributions

FL, XJC and GLT contributed to the concept and design of the study. YYL, YLC, SZ and HYX jointly completed the experimental work. YYL, YLC and GLT completed data analyses. YYL wrote the first draft of the manuscript. FL, XJC, PXL and GLT revised the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by the Open Research Project of Yunnan Provincial Key Laboratory of Plant Pathology (szbkf2024004); the Key Science and Technology Program of Henan Province (No. 232102111082); the National Natural Science Foundation of China (31660039).

Data availability

All data supporting the findings of this study are available within the paper and its Supplementary Information.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.