HTS analysis of resistance induction against PPV by four hairpin constructs in Nicotiana benthamiana Domin

Abstract

Plum pox virus (PPV) is the most devastating viral disease of the stone fruits worldwide. Inefficiency of the traditional control measures against PPV along with its globally widespread distribution and the economic importance of stone fruits, signify the necessity and importance of PPV resistance programs. In the present study, Agrobacterium-mediated transformation of Nicotiana benthamiana Domin was performed using four inverted repeat constructs derived from UTR/P1, HCPro, HCPro/P3, and CP regions of PPV-T isolate KyEsAp301. The efficacy of the constructs for inducing virus resistance in transgenic plants was evaluated by inoculation with PPV-D, -M, and -T strains. The potential of hairpin structures in the production of siRNAs and miRNAs in both wild-type and transgenic plants was compared by small RNA high-throughput sequencing. Although the four PPV genomic regions were used for transgenic resistance in previous experiments, small RNA high-throughput sequencing was first time used in this study to demonstrate the efficacy of the PPV constructs and to determine expression profiles of siRNAs and miRNAs. The results revealed that the potentials of hairpin constructs in producing siRNAs and their accumulation in target regions were significantly different. Expression profiles of several known and novel miRNAs were dramatically changed in response to PPV infection in both wild-type and transgenic plants, demonstrating plausible involvement of these miRNAs in plant-virus interactions. Based on the abundance of siRNAs and lack of PPV virus accumulation in transgenic plants harboring UTR/P1 and CP hairpin construct, we have concluded that UTR/P1 and CP are likely the best viral regions for induction of resistance against PPV.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-97546-7.

Introduction

Stone fruits are of great economic importance for their dietary and health benefits. The productivity of stone fruits is threatened by a severe viral disease known as Sharka, caused by plum pox virus (PPV; Potyvirus plumpoxi; Potyviridae), since its first report in Bulgaria on Prunus domestica cv. Kjustendil¹, PPV has been found in all continents except Australia². The virus is transmitted by aphids and infected propagative materials³ and can cause up to 100% fruit drop in susceptible hosts⁴.

The serious threat of incurable Sharka disease and the failures of quarantine efforts in many countries brought attention to alternative approaches such as genetic engineering techniques, for instance, the application of RNA silencing mechanism for induction of resistance against viruses. RNA silencing or RNA interference (RNAi) process, besides its significant role in gene regulation in eukaryotic organisms, is also involved in defense mechanisms such as antiviral defense in plants and invertebrates^5,6. The essence of this evolutionally conserved mechanism is to regulate gene expression either by restricting transcription, which is known as the transcriptional gene silencing (TGS) process, or by sequence-specific degradation of cellular messenger RNA or viral RNA known as posttranscriptional gene silencing (PTGS)⁷. PTGS is a mechanism that endogenous or exogenous double-stranded RNA (dsRNA) precursors are cleaved by Dicer-like (DCL) ribonuclease (RNase) ІІІ enzymes to produce 21–24 nucleotide (nt) small interfering RNAs (siRNAs) duplexes containing 2-nt 3́ overhangs at both ends. Most plants, including Nicotiana benthamiana, encode four distinct DCL proteins^8,9. DCL2, and DCL3, are responsible for producing 22, and 24, respectively, while DCL1 and DCL4 are responsible for 21 nt siRNAs^8,9. After unwinding siRNA duplexes, one strand is loaded into an RNA-induced silencing complex (RISC), which cleavages the complementary target RNA molecules¹⁰.

Before the RNA silencing mechanism was discovered, the primary evidence was observed in engineered plants expressing virus-encoded sequences, especially the virus coat protein, which was resistant to the same virus or closely related virus strains. Lindbo et al.¹¹ ‘s study revealed the virus resistance’s underlying mechanism and its association with cytoplasm-localized degradation of the transgene mRNA. After the discovery of the PTGS mechanism and the role of dsRNAs in triggering RNAi, integration of intron-containing hairpin-RNA (ihpRNA) constructs derived from virus genes into plant host genome for induction of resistance became widespread, and considering its high resistance efficiency in transgenic (TG) plants, this strategy has been widely used in different crops^12,13.

The PPV genome consists of a positive-sense single-stranded RNA of about 9,770 nucleotides (nt) encoding a long open reading frame (ORF), which is translated into a polyprotein of about 355 kDa. This polyprotein is processed by three virus-encoded proteinases to produce 1 mature protein product: P1-pro, HC-Pro, P3, 6K1, CI, 6K2, VPg, NIa-pro, NIb, and CP, as reported for other potyviruses^14–16. Another protein, P3 N-PIPO, is produced RNA molecules by the transcriptional slippage event¹⁶. The proteins are multifunctional and are involved in several processes. The coat protein (CP) is associated with virion assembly, RNA binding, virus translation, virus cell-to-cell and long-distance movement, and aphid transmission^16,17. The helper-component protease (HC-Pro) plays roles in viral uncoating, movement, translation initiation, aphid transmission, and RNA silencing suppression^18,19. The P1 protein is involved in virus replication and host adaptation of potyviruses^18,20. The P1 also enhances the silencing suppression ability of the HC-Pro²¹.

Due to the critical roles of P1, HC-Pro, and CP genes (mainly CP) in pathogenesis and their highly conserved nt sequences, these genes have been used to convey resistance against numerous plant viruses²². To confer resistance to PPV, P1, HC-Pro, and P3 genes were used in Nicotiana benthamiana Domin^23,24, and CP gene was used to transform N. benthamiana and Prunus domestica L. (plum)²⁵. Nicola-Negri et al.²³ showed that the 5’UTR/P1 gene construct conferred long-lasting resistance to nine PPV isolates. In another study, an ihpRNA construct derived from the 5´ part of the P1 gene and the 3´ portion of CP was utilized to transform N. benthamiana and plum²⁴. Based on the high success rate of producing resistant plants transformed with UTR/P1^23,26, or 5′ portion of the P1 gene²⁴, the UTR/P1 region was suggested to be used for establishing PPV-resistant stone fruits¹⁵.

Several studies have addressed successful resistance induction against PPV by ihpRNA constructs derived from different genomic regions of the virus. However, the efficiency and potential of the constructs in induction of resistance varied considerably. Selecting the most appropriate gene for constructing the hairpin structure and the length of inverted repeat seems to be critical for the induction of persistent and successful resistance in TG plants. In the previous studies mentioned above, high throughput sequencing (HTS) technology was not used to detect expression profiles of siRNAs and miRNAs.

HTS technology offers a powerful tool for assessing the abundance and accumulation patterns of siRNAs targeting the viral genome and changes that occur in the host’s microRNA (miRNA) expression profile in response to virus infection. The expression profiles of siRNAs and miRNAs and investigation of probable differences in their accumulation in TG plants harboring different hairpin constructs may help to choose the best hairpin construct candidate for stone fruits’ transformation to induce resistance against PPV.

In the case of PPV, both PPV-D and PPV-M strains, which are prevalent in Europe, were selected as the source for ihpRNA constructs¹⁵. However, different virus isolates prevail in various countries, and PPV-T is the most common strain in Türkiye^27–29. In the present study, N. benthamiana plants were transformed with four different hairpin structures derived from an isolate of the PPV-T strain. The TG plants were challenged by virus inoculation with three different PPV strains, and one from each of the PPV-infected TG plants, PPV-infected wild-type (WT) plants, and uninfected (mock) WT plants were subjected to HTS analysis. Subsequently, for the first time, expression profiles of miRNAs and siRNAs were analyzed to compare the efficacy of different hairpin constructs in the induction of transgenic resistance against PPV.

Materials and methods

Designing and cloning of hairpin constructs

PPV-T isolate KyEsAp301^28,29 was used as the virus source to amplify four different gene fragments and construct intron hairpin RNA vectors. As potential silencing targets, four primer pairs targeting UTR/P1 (752 pb), HC-Pro (649 bp), HC-Pro/P3 (594 bp), and CP (990 bp) (Table 1) were used to amplify the fragments. The attB1 and attB2 sites, which serve as the binding sites for recombination proteins, were added to the 5’ end of each primer. The recombination sites are required for Gateway Technology, which is based on the bacteriophage lambda site-specific recombination system. The four PPV regions were amplified by RT-PCR from the above-mentioned PPV-T isolate (Fig. 1f). Then, each of the four purified attB flanked PCR products was cloned into the entry vector pDONR221, using a BP clonase enzyme. Subsequently, the fragments were cloned into a pHELLSGATE12 binary vector using a recombination-based Gateway cloning technique by the LR clonase enzyme (Invitrogen). The pHELLSGATE12 contains two attR sites in reverse orientation to create the intron-containing hairpin structure. The recombinant clones were then transformed into One Shot^® TOP10 Electrocomp™ Escherichia coli cells by electroporation. The presence of the pHELLSGATE12 binary vectors and the genes of interest in this vector were examined by colony PCR using the nptII forward and reverse and gene-specific forward and reverse primers. In addition, CaMV 35 S promoter forward and the gene-specific reverse primer (each fragment); OCS terminator forward and gene-specific reverse primer (each fragment) were used for amplification and sequencing to ensure the correct orientation of inserts in the pHELLSGATE12 binary vector. Constructed binary vectors pHELLSGATE12 containing the ihpRNA constructs were transformed into Agrobacterium tumefaciens strain EHA105 by electroporation. After observation of single colonies on YEP agar medium containing selective antibiotics (rifampicin 50 µg/ml and spectinomycin 100 µg/ml), individual colonies were sub-cultured and subjected to colony PCR using the gene-specific primers. The formal identification of the specimens was performed by Prof. Dr. Kahraman Gürcan.

Table 1.

Primer sequences for amplification of desired fragments.

Name	Sequences	Product size (bp)	Reference
UTR/P1-F	5´´ GGGGACAAGTTTGTACAAAAAAGCAGGC AATATAAAAACTCAACACAACATTCA 3´	752	Nicola-Negri et al.23
UTR/P1-R	5´ GGGGACCACTTTGTACAAGAAAGCTGGGT CGCTTACGCCCTATGATCTC 3´
HC-Pro-F	5´ GGGGACAAGTTTGTACAAAAAAGCAGGC TGATGAGCGCACAAGACTACA 3´	649	Nicola-Negri et al.23
HC-Pro-R	5´ GGGGACCACTTTGTACAAGAAAGCTGGGT AACCTGCCTTTGCTATGAACA 3´
HC-Pro/P3-F	5´ GGGGACAAGTTTGTACAAAAAAGCAGGC TTCACAAAGACAGTGCGTGA 3´	594	Nicola-Negri et al.23
HCPro/P3-R	5´ GGGGACCACTTTGTACAAGAAAGCTGGGT AACTGCTGCATGTTCGTCAA 3´
CP-F	5´ GGGGACAAGTTTGTACAAAAAAGCAGGC GCTGATGAAAAGGAGGACGA 3´	990	This study
CP-R	5´ GGGGACCACTTTGTACAAGAAAGCTGGGT CTCCCCTCACACCGAGGA 3´

Note: The underlined sequences are attB1 and attB2 sites.

Fig. 1.

Different stages of regeneration of transgenic plants. (a-c) Generation of transgenic shoots from explants. (d) Transgenic plants growing in turf. (e) inoculum source for PPV-D, -M, and -T strains. (f) Amplification of genes of interest for hairpin construction. Lanes 1: UTR/P1 752pb, 2: HC-Pro 649 bp, 3: HC-Pro/P3 594 bp and 4: CP 990 bp. M: 100 bp DNA ladder. (g) Agarose gel electrophoresis of RT-PCR products of 5 selected plants harboring four hairpin constructs showing resistance. Lanes 1 to 5: Plants with UTR/P1 hairpin construct; Lanes 6 to 10: Plants with HC-Pro/P3 hairpin construct; Lanes 11 to 15: Plants with HCPro hairpin construct; Lanes 16 to 20: Plants with CP hairpin construct. M: 100 bp DNA ladder. The images (f and g) depict agarose gel electrophoresis results from RT-PCR experiments. The presented bands accurately represent the PCR products without any manipulation. While minor adjustments may have been made to align the panels within the figure layout, no significant cropping affecting the integrity of the data was performed. The images retain the original banding patterns, ensuring that the necessary regions are visible in a single, consistent line for proper interpretation.

Transformation and regeneration of transgenic plants

N. benthamiana seeds were sterilized with 75% (v/v) absolute ethanol for 1 min and 0.75% NaOCl for 20 min and eventually washed three times with sterile distilled water. The sterilized seeds were sowed on an MS agar medium (pH = 5.8). The aseptic seeds were germinated, and the seedlings were cultured in a growth chamber at 26 ± 1 °C under a photoperiod of 16 h-light/8 h-dark. Leaf explants of 0.5 cm² cut from leaves one to two centimeters in diameter were placed on the pre-culture medium (MS medium supplemented with 5µM 6-Benzylaminopurine: BAP) with adaxial side up. The explants on the pre-culture medium were cultured at 25 °C under a 12-hour light regime for 2 days. Subsequently, each group of the pre-cultured explants was soaked in A. tumefaciens (EHA105) cell suspensions harboring each of the ihpRNA constructs separately for 30 min. The soaked explants were blotted on a sterile filter paper and then placed on co-cultivation medium (MS medium supplemented with 5µM BAP, 3.7 g/L 2-morpholinoethanesulfonic acid (MES) buffer (pH = 5.4) and 38 mg/L of acetosyringone) at 28ºC for 2 days in the dark. After 2 days, the leaf discs were transferred to the selection medium (MS medium supplemented with 100 mg/L kanamycin and 500 mg/L cefotaxime). The explants were transferred to a fresh selection medium at two-week intervals. After about a month, the 5 mm or longer shoots from each explant were excised and transferred to the rooting medium supplemented with the appropriate selection agents. The rooting stage was completed under the same conditions applied for the regeneration step for about 2 weeks. The plantlets with well-established root systems were transferred to pots containing turf soil in a climate room. Once the plantlets were fully acclimated, leaf samples were collected from transformed N. benthamiana plants, and the presence of the transgenes was evaluated by PCR using gene-specific and the neomycin phosphotransferase II (nptII) primers. The seeds from self-pollinated transgenic plants were collected for further experiments.

Resistance assay

In this study, one isolate of three strains was used as a virus source while determining the resistance in transgenic NB plants. A PPV-T isolate KyEsAp301 hosted in an apricot tree in Kayseri, a PPV-M isolate is MrPl220 in a plum tree, and again an apricot tree infected by PPV-D isolate AkOrAp430^27,29–31. For this, first, the isolates were transferred from stone fruit trees to N.B. plants, and then contaminated NB plants were used as inoculum sources for resistance assay.

Mechanical inoculation of virus isolates from apricots to N. benthamiana

Symptomatic leaves of the trees infected with isolates given above were brought to the laboratory in the late spring and were homogenized with inoculation buffer [(phosphate-buffered saline) PH = 7.2 supplemented with 2% (w/v) polyvinylpyrrolidone (PVP), and 0.2% (w/v) sodium diethyl-dithiocarbamate (DIECA)]. The homogenate was rubbed onto celite-dusted leaves of over 200 seedlings of N. benthamiana at the 3–5 leave stage. Inoculated plants were kept in a controlled condition at 25 °C, 70% humidity, and inspected weekly for symptom expression for 6 weeks. In addition to observation of the viral symptoms in the plants, infections in N. benthamiana were evaluated by RT-PCR using PPV-specific primers. Several infected N. benthamiana plants for each strain were kept in big pots in climate rooms and used as a virus source as needed for the resistance assays of TG plants. The WT healthy seeds were also grown as control plants and were inoculated mechanically, as described above.

Resistance assay of transgenic N. benthamiana plants

The T0 progeny seeds of N. benthamiana containing each of the four hairpin constructs were sown on turf soil to raise mature T1 lines for the resistance assays. The T1 lines were evaluated by PCR using gene-specific primers and then were mechanically inoculated when they were four to five broad leaf stages, separately, using three virus strains kept in N. benthamian plants. Mechanical inoculation was performed as described above. The resistance assays were conducted in a randomized complete block design with nine replications. After visual observation of virus symptoms on control WT plants at 15 days post inoculation (dpi), DAS-ELISA was performed for the evaluation of virus infection in TG plants. Five individuals were selected for the ELISA test from each line of TG plants. Leaf samples were collected from each plant, and they were ground in extraction buffer (PBS pH = 7.2–7.4, % 2 PVP, % 0.2 bovine serum albumin (BSA)) at a 1:10 ratio (w/v), and ELISA test was carried out according to the manufacturer’s protocol (Loewe Biochemica GmbH, Sauerlach, Germany). Samples with OD405 values more significant than twice that of NCx (average OD405 value of the negative controls) were considered positive.

After getting ELISA results at 15 dpi, five plants with the least viral symptoms lines with four different hairpin constructs were subjected to RT-PCR using P1/P2 primers, which amplifies a 243 bp fragment of CP of the virus³². In TG plants harboring the CP hairpin construct, additional primer pairs amplifying other virus genes were used.

Library preparation and small RNA sequencing

Leaf pools from the most resistant lines harboring each of the hairpin constructs (UTR/P1, HC-Pro, HC-Pro/P3, and CP), infected (WT+) and uninfected (mock) wild-type (WT-) control plants were made by collecting three different leaves from three replicates of each plant. The total RNA of collected samples was extracted using a mirVana™ miRNA isolation kit (Life Technologies, Thermo Fisher Scientific, Waltham, MA, USA). The purity of the isolated RNA was evaluated using the BioSpec-nano spectrophotometer (Shimadzu, Kyoto, Japan) and subjected to library construction using TruSeq Small RNA Sample Prep Kit (Illumina, San Diego, USA). Briefly, adapters were ligated to the 3′ and 5′ ends of 1 µg of isolated RNA. Then cDNA constructs were created by reverse transcription followed by amplification using two primers that annealed to the adapter ends to enrich RNA fragments with adapter molecules on both ends. Small RNA libraries were selected by running the amplified cDNA on 6% Novex TBE PAGE Gels (Life Technologies CA, United States), then were cut and purified by ethanol precipitation. Quality control and validation of libraries were performed via fragment analysis using a Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) and sequenced by single-end sequencing 50 bp on an Illumina Hiseq 4000 at LC Sciences (Houston, Texas, USA).

miRNA and siRNA identification and target prediction

Raw reads were subjected to ACGT101-miR software (LC Sciences, Houston, Texas, USA) to remove adapter dimers, junk, low complexity, common RNA families (rRNA, tRNA, snRNA, and snoRNA), and repeats. The remaining clean sequences were mapped to the miRNA sequences available in the miRBase database (version 21.0). Those unique sequences mapped to the mature miRNAs were considered as known miRNAs. The unmapped sequences were blasted to GenBank, and the hairpin RNA structures containing sequences were predicated from the flank 120 nt sequences using RNAfold software. The criteria for secondary structure prediction were described by Soichot et al.³³. The sequences having all those criteria were considered as potential candidate miRNAs.

To detect the siRNAs produced in TG and inoculated wild-type plants, the clean reads of each library were mapped to PPV-M isolate isMrPl220 genome (acc. no. MF370984) using Bowtie v1.2.1.1 and Geneious Prime 2019.1.3 (https://www.geneious.com)³⁴. The read coverage of the reads was calculated in Samtools v.1.16.1³⁵. The normalized siRNAs were mapped to hairpin constructs (UTR/P1, HC-Pro, HC-Pro/P3, 6K1, CI, 6K2, NIa-VPg, NIa-Pro, Nib, and CP) using the shinyCircos tools³⁶. Targets of the identified miRNAs were predicted using the Web-based psRNATarget program (http://plantgrn.noble.org/psRNATarget/)³⁷ with default parameters. In addition, Blast2Go version 5.2 was used for gene ontology enrichment analysis of predicted target genes, and their functional groups were categorized into three parts: biological processes, cellular components, and molecular function. Data analysis and plots were conducted using the ggplot2 package³⁸.

miRNA expression profiling analysis

To compare the expression levels of the miRNAs among libraries, chi-square, and Fisher’s exact tests were performed on the normalized data. The log2 ratio was used as the threshold to compare differences in miRNA expression levels. The miRNAs were considered to be significantly upregulated or down-regulated when the p values of both the chi-square test and Fisher’s exact test were ≤ 0.05, the log2 ratio| was ≥ 1, and more than 100 reads were present in at least one library after normalization³⁹. Expression levels of miRNAs were categorized into three groups: high (the number of reads in miRNAs is higher than the average copy of the data set), medium (the number of reads in miRNAs is higher than 10 and less than average), and low (the number of reads in miRNAs is less than 10 and less than average).

Results

Evaluation of PPV resistance in transgenic plants

Transforming N. benthamiana plants with UTR/P1, HC-Pro, HC-Pro/P3, and CP ihpRNA constructs resulted in 10, 5, 6, and 11 TG lines, respectively (Fig. 1a-e). The TG lines and the control plants were inoculated separately with PPV-T, PPV-D, and PPV-M strains. The virus symptoms, including chlorotic mosaics with dark green islands and leaf puckering, were observed on wild-type control plants at 15 dpi. Several lines showed no virus symptoms in TG plants harboring UTR/P1 and CP ihpRNA constructs. The DAS-ELISA results confirmed that they were virus-free in all three viral inoculation trials. The DAS-ELISA results for TG plants containing HC-Pro ihpRNA construct showed that line 8−1 with one infected plant line in PPV-D strain inoculations was considered the most resistant. In TG plants containing HC-Pro/P3 hairpin construct TG plants, 5−4 and 5–6 lines, out of six lines, were utterly resistant against PPV (Table 2).

Table 2.

Resistance analysis of transgenic Nicotiana benthamiana plants harboring hairpin constructs challenged with three PPV virus strains.

Hairpin constructs		PPV-D		PPV-T		PPV-M
		Infected/transgenic	ELISA positive	Infected/transgenic	ELISA positive	Infected/transgenic	ELISA positive
UTR/P1 transgenic lines	10 − 4	0/6	0/5	0/5	0/5	0/8	0/5
	11 − 1	0/6	0/5	0/7	0/5	0/7	0/5
	12 − 3	0/8	0/5	0/9	0/5	0/9	0/5
	12 − 4	0/7	0/5	0/9	0/5	0/9	0/5
	12 − 5	0/9	0/5	0/9	0/5	0/8	0/5
	15	0/8	0/5	0/8	0/5	0/9	0/5
	18 − 2	0/6	0/5	1/7	1/5	0/5	0/5
	18 − 3	0/6	0/5	0/5	0/5	0/6	0/5
	18 − 5	0/5	0/5	0/8	0/8	0/6	0/5
	18 − 6	0/7	0/5	1/6	1/5	NA	NA
HC-Pro transgenic lines	2	2/5	2/5	NA	NA	3/9	2/5
	6	NA	NA	0/5	0/5	1/5	1/5
	7 − 1	1/5	1/5	0/5	0/5	1/7	1/5
	7 − 2	1/7	1/5	1/7	1/5	2/7	2/5
	8 − 1	1/9	1/5	0/8	0/5	0/8	0/5
HC-Pro/P3 transgenic lines	1–2	5/7	5/5	2/6	2/5	5/5	5/5
	5 − 3	5/8	5/5	2/7	2/5	5/5	5/5
	5 − 4	0/8	0/5	0/9	0/5	0/7	0/5
	5–6	0/5	0/5	0/8	0/5	0/8	0/5
	5–8	3/8	3/5	0/5	0/5	7/7	5/5
	5–9	-	-	2/7	2/5	6/6	5/5
CP transgenic lines	3 − 1	0/6	0/5	0/6	0/5	2/9	2/5
	4 − 1	0/9	0/5	0/6	0/5	2/9	2/5
	4 − 3	0/9	0/5	0/9	0/5	0/6	0/5
	11 − 1	1/8	1/5	0/9	0/5	0/9	0/5
	11 − 3	0/9	0/5	0/8	0/5	0/7	0/5
	12 − 2	0/9	0/5	0/5	0/5	1/6	1/5
	20 − 1	0/8	0/5	1/8	1/5	0/8	0/5
	20 − 3	1/8	1/5	0/7	0/5	1/8	1/5
	21 − 3	0/9	0/5	1/8	1/5	0/6	0/5
	21 − 4	1/9	1/5	0/8	0/5	-	-
	21 − 5	0/5	0/5	0/9	0/5	0/9	0/5
Wild type		9/9	5/5	9/9	5/5	9/9	5/5

NA is not available.

The presence of PPV in the tested plants was evaluated by RT-PCR using P1/P2 primer at 60 dpi. The results showed no amplification in the plants harboring UTR/P1 and HC-Pro/P3 constructs. Nonetheless, PPV was detected in all the TG plants harboring HC-Pro and CP hairpin construct (Fig. 1g). However, since the P1/P2 primer targets CP, repeated RT-PCR assays with PPV-specific primers spanning different viral genomic regions did not show amplicon.

Deep sequencing, sRNA, miRNA, and siRNA profiling changes

Deep sequencing of six samples generated raw reads ranging from 7 764 101 (WT-) to 16 091 996 (WT+) (Additional file 1: Table S1). After removing adapter sequences and junk reads, an abundant number of small non-coding RNAs (sRNA) were identified, of which almost half were unique (Additional file 1: Table S2). While among total reads, 24 nt sRNAs were the most abundant, among unique reads, the majority of sRNAs were 21 nt for all libraries except for the WT + sample, which depicted different sRNA length profiles (Fig. 2a-b). Pearson correlation analysis (Additional file 3: Fig S1) showed that R values are almost equal to 1, showing the perfect positive linear relationship among the TG lines. The R values among each pair of the TG lines were higher than that among the pairs of any TG line and PPV-infected WT, indicating differential expression of sRNAs.

Fig. 2.

Length distribution of counts of (a) total, (b) unique small non-coding RNAs (sRNA), (c) miRNAs, and (d) siRNAs.

In total, 770 miRNAs were detected by screening sRNAs (Additional file 1: Table S3). Summary features of pre-miRNAs and predicted miRNAs for each library were presented in Additional file 1, Table S4. Among the detected miRNAs, 460 were identified for the first time in this study, mainly for newly reported 5p or 3p sequence-derived miRNA candidates; 227 were identified as different; and 83 were identified as known confirming miRNA sequences in miRbase. Concerning the length size of miRNAs, 21-nt miRNAs were most abundant (41.95%), followed by 24 nt (14.03%) and 22 nt (13.90%) miRNAs (Fig. 2c, Additional file 1: Table S5).

Abundant numbers of siRNA were detected by aligning the clean small sequences to the PPV genome. The profile of screened siRNAs in all libraries revealed an accumulation of 21- to 25-nt siRNA species targeting the PPV genome. A total of 153,839 reads of 21–25-nt siRNA species were detected in wild-type plants (WT+), where PPV was successfully multiplied and established. (Fig. 3a). While lower numbers of siRNAs (5261 in UTR/P1, 2971 in HCPro, 2637 in HCPro/P3, and 5205 in CP) were identified for TG lines. Among resistant TG plants, the total number of 21–25-nt siRNA species in UTR/P1 and CP ihpRNA constructs was considerably higher than in HC-Pro and HC-Pro/P3 ihpRNA constructs. In this library, 22-nt siRNAs were the most abundant, followed by 21-nt, 23-nt, 24-nt, and 25-nt siRNAs (Figs. 2d and 3a; Additional file 2: Table S16-S21). Circos plot of the siRNAs to the PPV genome highlighted a positive correlation between siRNA abundance and those four regions used for hairpin construction (Fig. 3b). In infected wild-type plants, the siRNAs were distributed throughout the virus genome. In contrast, in resistant TG plants, siRNAs accumulation in those four selected regions was detected (Fig. 3b).

Fig. 3.

Size distribution and Circos plot of siRNA. (a) Size distribution of the 21–25 nucleotide siRNAs in transgenic plant harboring UTR/P1, HC-Pro, HC-Pro/P3 and CP hairpin constructs, and (b) Circos plot of the normalized (reads per million) from outside to inside circles: UTR/P1 (green), HC-Pro (orange), HC-Pro/P3(pink), CP (blue), Infected -wild type(brown), uninfected (teal).

Differentially expressed MiRNAs and prediction of potential target genes

For identifying miRNAs involved in TG lines, the expression of miRNAs in TG lines and WT- and WT + samples were compared. Among the 770 identified miRNAs, 96 miRNAs exhibited a high level of expression while 442 and 252 miRNAs were considered as medium and low-level expression, respectively (Additional file 1: Table S6), and among six lines, 366, 93, and 73 miRNAs exhibited significant differential expression with a p value ≤ 0.001, ≤ 0.01, and ≤ 0.05, respectively (Additional file 1: Table S6). Differentially expressed miRNAs for TG lines and WT- and WT+ were presented in Additional file 1: Table S7-S14. A venn diagram showing the number of differentially expressed miRNAs between TG lines and wild plants was depicted (Fig. 4). To study the changes in the expression profile of miRNAs and the possible connection between miRNA profiling and viral resistance, the significantly differentially expressed (either up-regulated or down-regulated) miRNAs were detected in TG lines and wild plants (Table 3) using thresholds set by Kuchenbauer et al. (2008). Sixty-five miRNAs were found to be the most significantly differentially expressed (≥ 1.5-fold change, ≥ 150 sequence counts, P < 0.001), and only two miRNAs (miR169h-5p_L+2R-1_1ss22GC and miR169h-5p_L+2R-1_1ss22GC) were upregulated in TG plants compared to WT- plant. However, 44 miRNAs were downregulated, and 18 miRNAs were upregulated in response to PPV infection in TG plants (Table 3).

Fig. 4.

Venn diagram showing differentially expressed miRNA comparisons. (a) Comparison of miRNAs between each transgenic line harboring UTR/P1, HC-Pro, HC-Pro/P3, CP hairpin constructs, and PPV positive (WT+) and PPV negative (WT-) samples. (b) Comparison of miRNAs among transgenic lines harboring UTR/P1, HC-Pro, HC-Pro/P3, CP hairpin constructs.

Table 3.

The most differentially expressed MiRNA species (counts > 150 and > 1.5-fold change) in TG lines in comparison with wild-type PPV negative (WT-) and wild-type PPV positive samples (WT+).

	miR_name*	Fold_change				Expression level	Sequence in miRbase
		UTR/P1	HCPro	HCPro/P3	CP
	transgenic/WT-
Up regulated	nta-miR156 g_L + 1	2.28	2.33	3.45	2.08	high	Diff
	ata-miR169 h- 5p_L + 2R- 1_1 ss22GC	3.73	8.42	5.00	4.60	high	Diff
	transgenic/WT+
Down regulated	miR11609-p3_2 ss10GA21 AG	-inf	-inf	-inf	-inf	high	New
	miR7490-p3_2 ss5 AT18GT	-inf	-inf	-inf	-inf	high	New
	PC- 3p- 24596_182	-inf	0.06	0.05	0.04	high	New
	miR397-p3_1 ss6GA	0.04	-inf	-inf	0.02	middle	New
	miR393a- 3p_2 ss12 TC20 TC	0.01	0.01	0.00	0.00	high	Diff
	miR398a	0.04	0.03	0.01	0.04	high	Yes
	miR398-p5_2 ss8 TC18 CT	0.04	0.06	0.01	0.01	high	New
	miR398-p3_1 ss4GA	0.12	0.05	0.01	0.10	high	New
	miR398a_L + 2R- 2	0.25	0.28	0.09	0.31	high	Diff
	miR6025a-p3	0.02	0.03	0.02	0.02	high	New
	miR6025b-p5	0.02	0.03	0.02	0.02	high	New
	miR391 - 3p_L + 1	0.16	0.03	0.10	0.01	high	Diff
	miR391_L + 1R- 1_1 ss5 CT	0.39	0.09	0.29	0.06	high	Diff
	miR162	0.20	0.22	0.20	0.25	high	Diff
	miR171f	0.07	0.03	0.05	0.05	high	Yes
	miR171a-p5_1 ss5 AG	0.10	0.03	0.05	0.04	high	New
	miR6155-p3	0.03	0.03	0.05	0.02	middle	New
	miR6147-p3	0.05	0.05	0.03	0.03	high	New
	miR408	0.17	0.09	0.05	0.18	high	Yes
	miR408-p5_1 ss2 CT	0.11	0.06	0.01	0.08	high	New
	miR408_R + 1	0.40	0.17	0.12	0.42	high	Diff
	miR160a-p3	0.05	0.05	0.04	0.03	high	New
	miR160a- 3p	0.05	0.05	0.04	0.03	high	Yes
	miR164a-p3	0.07	0.04	0.04	0.05	high	New
	miR166a-p5	0.49	0.36	0.41	0.07	high	New
	miR166c- 5p_1 ss10 TG	0.11	0.07	0.06	0.13	high	Diff
	miR166c-p5	0.05	0.07	0.06	-inf	high	New
	miR166b-p5	0.21	0.11	0.19	0.25	high	New
	miR168a_1 ss21 CA	0.08	0.09	0.07	0.11	high	Diff
	miR168b- 3p_R + 1	0.09	0.11	0.10	0.11	high	Diff
	miR168c-p3	0.10	0.12	0.12	0.14	high	New
	miR168a_R + 1	0.18	0.27	0.16	0.15	high	Diff
	miR403 - 5p_2 ss14 AT18 AG	0.09	0.13	0.07	0.09	high	Diff
	PC- 5p- 1760_1591	0.07	0.14	0.09	0.11	high	New
	PC- 3p- 4344_960	0.13	0.09	0.08	0.08	middle	New
	miR396a- 3p	0.24	0.15	0.24	0.08	high	Diff
	miR1919_1 ss2 AT	0.16	0.16	0.17	0.11	high	Diff
	miR390b- 3p_R- 1_1 ss20 CT	0.16	0.14	0.09	0.17	middle	Diff
	miR390a	0.26	0.28	0.29	0.24	high	New
	miR6020b	0.21	0.21	0.22	0.39	high	Yes
	miR172c-p5_1 ss5 CT	0.28	0.24	0.26	0.32	middle	New
	miR8036 - 3p_1 ss2 AT	0.34	0.29	0.29	0.44	high	Diff
	miR6026_2 ss3 CT18 AC	0.38	0.24	0.47	0.37	middle	Diff
	PC- 3p- 8146_611	0.60	0.47	0.27	0.37	middle	New
Up regulated	miR156 d- 3p_R- 1	1.79	1.95	2.68	2.19	middle	Diff
	miR159	2.41	2.05	3.27	2.20	middle	Yes
	miR171c	1.92	2.02	2.27	2.22	middle	Yes
	miR171e	2.66	2.41	2.72	2.48	middle	Yes
	miR171 h_1 ss3GA	2.66	2.41	4.50	2.48	middle	Diff
	miR482a	3.58	3.92	3.52	2.60	middle	New
	miR6025a	1.86	1.64	2.26	2.56	middle	Yes
	miR394	2.76	1.70	2.06	2.44	middle	New
	miR396a	1.92	1.70	2.13	2.26	middle	Diff
	miR396a- 5p	1.92	1.70	2.13	2.26	middle	Yes
	miR6300	3.13	3.56	4.26	4.70	middle	New
	miR2916-p3_1 ss15 TC	5.98	18.47	6.14	10.45	middle	New
	miR2916-p5_1 ss3 AG	3.20	7.24	2.86	5.21	middle	New
	miR6149a	2.91	3.55	2.22	2.90	middle	Diff
	PC- 5p- 4228_975	3.21	2.59	3.19	6.09	middle	New
	miR6155	2.42	2.08	1.80	2.41	middle	Yes
	miR7122 - 5p_1 ss18 TG	1.81	1.64	1.77	1.94	high	Diff

* The meaning of the abbreviations used for naming miRNAs was explained in Additional file 1: Table S3.

Responses of the miRNAs target genes to PPV infection in TG plants were studied by gene ontology (GO) analysis (Fig. 5). GO analysis of biological processes demonstrated that genes involved in the carbohydrate metabolic process were most affected by miRNAs (Fig. 5). Molecular function analysis showed that responsive miRNAs mostly affected target genes encoding transferase activity, kinase activity, and lyase activity. Another group of genes affected by miRNAs during PPV infection (Fig. 5) were the genes of the cellular component (CC) category: plastid, endocytic vesicle, and phagocytic vesicle (Additional file 1: Table S15).

Fig. 5.

Gene ontology (GO) enrichment analysis of miRNAs’ target genes, showing the most significantly enriched GO terms of the miRNA target genes in the cellular components, molecular function, and biological processes.

Discussion

Transformation of plants with pathogens-derived hairpin RNAs is an effective and widely used strategy for producing resistant plants against several plant viruses, including PPV¹⁵. In the case of PPV, previous studies employed the hairpin constructs derived from conserved genomic regions: UTR/P1, HC-Pro, HC-Pro/P3, and CP from isolates belonging to PPV-D or PPV-M strains. In the present study, the same regions were chosen from a PPV-T isolate, which successfully inducted resistance against PPV-D, -M, and -T strains. Similar results have been reported using the same genomic regions driven from PPV-M²³and PPV-D²⁵, which suggests these genomic regions have the necessary criteria for effective and successful siRNA production. Therefore, regardless of their origin, they can be considered as highly potent silencing constructs.

The abundance, complexity, and diversity of processed miRNA and siRNAs in WT (-/+) and TG plants harboring different ihpRNA constructs were evaluated by small RNA sequencing. The expressions of miRNAs are often changed after plant virus infections and several miRNAs have been involved in self-defensive processes in response to the viruses⁴⁰. Regarding the expression profile of the known miRNAs, as shown in Table 3, the expression of several miRNA species was dramatically varied following PPV infection in TG plants. To illustrate the important functions of the known miRNAs detected in this study, three miRNAs (miR391, miR390b, and miR169 h) were exemplified here. miR391 (also known as Sly-miR4376) having a 5′ terminal uridine (5′-U) was found to be mainly associated with AGO1 protein in PTGS of transgenes⁴¹. Recently, overexpression of miR391 in TG Arabidopsis lines demonstrated post-transcriptional silencing of the ACA- 10 gene by this miRNA⁴². miR390, already known to be functionally important for many years⁴³, has recently been implicated in resistance to apple anthracnose caused by Colletotrichum gloeosporioides⁴⁴. Likewise, miR169’s role as a negative regulator in rice immunity against the blast fungus by repressing the expression of nuclear factor Y-A genes was confirmed recently⁴⁵. Taken together, higher expression of these miRNAs in this study confirms their involvement in the plant defense system. However, novel and known miRNAs identified in this study should be further examined to determine their possible contribution to PPV resistance.

The comparison of the distribution of enriched GO terms in PPV-infected and mock libraries revealed that biological processing, transcription regulation, transcription, protein phosphorylation, and defense response-related GO terms decreased upon PPV infection. In general, viruses can reprogram and alter plant gene expression, leading to different cellular stress responses and developmental defects in plants. Incompatible plant-virus interactions, where the host plant does not recognize the viral particle, the virus can interfere with the accumulation and function of host vital elements such as proteins and nucleic acids as well as essential reactions such as host defense responses⁴⁶. A phenomenon known as host gene shut-off, which is the down-regulation of the expression of host genes, has been reported in virus-infected plants^47,48. In PPV, the HC-Pro and P1 genes have been demonstrated to be involved in suppressing RNA silencing⁴⁹. As the viral RNA silencing suppressors usually promote infection by interfering with miRNAs and trans-acting small interfering RNA expression⁸, biological and developmental defects by perturbing or interfering with signaling pathways are expected. GO analysis in this study indicated that “plastid” was the organelle most affected. Whereas one of the typical symptoms induced by PPV infection was the appearance of a yellow line pattern on leaves; implying that the plastid is one of the most critical components in cells affected by virus infection⁵⁰.

On the other hand, the sequence distribution profile of TG siRNAs on the PPV genome revealed that the siRNAs targeted different regions throughout the PPV genome. However, an outstanding accumulation of TG siRNAs on hotspot regions of the genome was observed. These hotspots in each TG plant were the regions utilized for hairpin construction. Interestingly, the amount of siRNA accumulation in the corresponding regions differed among these four resistant TG plants. The higher accumulation of siRNAs on targeted regions in resistant TG plants harboring UTR/P1 and CP hairpin constructs was interpreted by the higher capacity of these genomic regions in producing 21 - 25nt siRNAs compared to two other hairpin constructs. To obtain optimum resistance in TG plants using virus-derived ihpRNA constructs, selecting the best target sites from the virus genome is crucially important. The level of introduced resistance in hosts to the viruses varies widely depending on the genomic regions used for hairpin construction. As several studies illustrated, a positive correlation existed between siRNA population and resistance^51–54. In the present study, the number of 21–25 nt siRNAs in UTR/P1 and CP ihpRNA-containing TG plants were significantly higher than TG plants harboring HC-Pro and HC-Pro/P3 hairpin constructs, suggesting that the UTR/P1 and CP regions can be considered for better choices for hairpin construction and induction of RNA silencing and resistance. Our results were in line with the study of Di Nicola-Negri et al.²⁶, using UTR/P1 ihpRNA construct from the PPV-M strain, in which they reported the high capacity of the UTR/P1 region of the PPV genome in inducing resistance against seven isolates of D, M and REC strains and PPV-C and EA which are phylogenetically distantly related to D, M and T strains.

Mapping screened siRNAs from infected wild-type library to the PPV genome revealed that the 21–25-nt siRNAs from both polarities were uniformly distributed throughout the whole genome, indicating involvement and activation of the host RNAi mechanism and antiviral defense in response to viral infection. As this result shows, despite the production of a large set of siRNAs that targeted the virus genome, virus multiplication successfully occurred, and symptoms appeared in infected wild-type plants, indicating that PPV overcame the plant defense mechanism. Antiviral defense response in plants mainly consists of accumulation of 21-nt and 22-nt species, produced by DCL4 and DCL2, respectively⁵⁵. These two classes of siRNAs were also highly accumulated in infected wild-type plants. 22-nt siRNAs, with a slight difference compared to 21-nt siRNAs, were the most abundant class of siRNAs. Although 21-nt siRNAs are generally the most frequent siRNA species in antiviral defense response, in some cases, higher or the same abundance of 22-nt siRNAs have been reported^53,56, and this can be considered as evidence of various host reactions to different viruses. In the TG plants harboring the UTR/P1, HC-Pro, HC-Pro/P3, and CP hairpin constructs, 21-nt siRNA species were the most abundant. The second most abundant class was the 22-nt species. The dominant frequency of 21-nt and 22-nt among siRNA classes of TG and virus-infected plants in this study agreed with previous findings, which demonstrated that in siRNA-directed RNA degradation pathway, endogenous or exogenous long dsRNA or hrpRNA are processed mostly by DCL4 and DCL2 into 21- and 22-nt siRNAs, respectively^57–59.

Conclusion

The introduction of persistent and durable resistance against PPV is the main goal in its management and considering the laborious process of stone fruit plant transformation, choosing the best hairpin construct is crucially important. Based on our results, UTR/P1 and CP hairpin constructs had higher potential than HC-Pro and HC-Pro/P3 hairpin constructs in producing siRNAs. Hence, they are the best candidates for inducing resistance by RNAi mechanism. The presented study is an essential step for the induction of persistent and efficient resistance to PPV in stone fruit plants by the RNAi mechanism.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

MGS experimented and collected data. KG, ST, and MA advised study and wrote manuscript. YT, MY, AK, AN, HK, BÖ, and VR edited the manuscript. All authors read and approved the final manuscript.

Funding

This research was conducted at the Plant Biotechnology Laboratory of Erciyes University Genome and Stem Cell Center and was supported by the Research Fund of Erciyes University (FCD- 2020 - 9555) and the Scientific and Technological Research Council of Türkiye (project number 112O022).

Data availability

The datasets supporting the conclusions of this article are included within the article and its additional files. The miRNAseq datasets supporting the conclusions of this article are available in the NCBI repository, [BioProject ac. no. PRJNA1179906, https://www.ncbi.nlm.nih.gov/bioproject/1179906].

Declarations

Competing interests

The authors declare no competing interests.

Statement specifying permissions

For this study, we acquired permission to study Nicotiana benthamiana issued by the Agricultural and Forestry Ministry of Türkiye.

Statement on experimental research and field studies on plants

The either cultivated or wild-growing plants sampled comply with relevant institutional, national, and international guidelines and domestic legislation of Türkiye.