Fig 1.
Preparation of saturated mutant pools for selected stem regions, analysis of structural alterations caused by base substitutions in the stem region of the PSTVd secondary structure, and mutant pool inoculation strategy.
(A) The secondary structure of PSTVd and the design of saturated mutant pools on selected stem regions. The secondary structure of PSTVd consists of 26 stems (S) and 27 loops. The 26 stems are numbered. Three stems, S3, S15, and S26 were chosen for mutant pool preparation and are highlighted in outlined rectangles. The red Ns represent the degenerate regions, with N representing A, U, G, or C. The numbers of all possible sequences in the degenerate regions were calculated. Mutant pools were prepared through PCR using forward primers that contained these degenerate regions. (B) Principle of structural alterations caused by base substitutions in the stem region of PSTVd secondary structure. A base substitution in a canonical base pair within stem regions can lead to three possible mutated base pairs, as shown in this example. These include one purine/pyrimidine base pair (group 1) and two other types of base pairs (purine/purine or pyrimidine/pyrimidine, group 2). Consequently, the ratio of group 2 base pairs to group 1 base pairs is 2.0. See text for the comparisons of 3D structure of different base pairs. (C) Inoculation strategy of stem mutant pools. Mutant pools of S3, S15, and S26 described in Fig 1A were delivered to the first two fully expanded true leaves of 3-week-old N. benthamiana plants through agroinoculation. Three biological replicates were included, with one plant per replicate. Using the S3 mutant pool as an example, samples were taken from the inoculated region (IR), the margin of the inoculated leaves (LM), and the upper systemic leaves (Sys) at various time points. (D) Analysis of the earliest infection time of S3 mutant pool at sites of IR, LM, and Sys. Samples collected from the IR, LM, and Sys sites at various time points were subjected to RNA blot analysis to determine the earliest time of infection. Circular form PSTVd (PSTVd-C), which is the functional form, was detected. Ethidium bromide staining of ribosomal RNA serves as a loading control.
Fig 2.
Analysis of sequence diversity and types of base pairs in selected stem regions of PSTVd quasispecies derived from stem mutant pools.
(A) Analysis of the stem region sequence diversity of PSTVd quasispecies from mutant pool, IR, LM, and Sys samples. A total of 30 libraries were sequenced using the Illumina NovaSeq 6000 platform, and normalized Shannon entropy of the pool and the three selected loop sequences was calculated as described in the text. Values range from 0 to 1, with higher values indicating greater sequence diversity. (B) Read number of the top 20 most frequently detected variants from deep-sequenced libraries. The number of reads for the top 20 most frequently detected variants from the deep-sequenced libraries were normalized to a per-thousand basis and plotted on a log2 scale. (C) The number of unique sequences in deep sequencing libraries. The number of unique sequences of all deep sequencing libraries were normalized to a per-thousand basis and presented. Data are expressed as mean ± standard deviations (normalized Shannon entropy and number of unique sequences per thousand reads) or average values (reads number of the top 20 most frequently detected variants per thousand reads) for IR, LM, and Sys samples and a single value for the inoculum pools. Student’s t-test was applied for comparisons: *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, ns, not statistically significant. (D) Sequences of both strands of stems S3, S15, and S26 were isolated from the deep sequencing dataset, and the base pairs formed by two strands of the same stem were analyzed. By comparing them to the WT sequence as a reference, the mutated base pairs at each position were analyzed. The weighted number of group 1 and group 2 mutated base pairs was calculated and compared between different samples (see text for details). Data are expressed as the mean ± standard deviations for IR, LM, and Sys samples, and as a single value for the inoculum pools. Student’s t-test was applied for data comparisons: *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, ns, not statistically significant.
Fig 3.
Structural models of selected PSTVd loops and preparation of saturated mutant pools for those loops.
The PSTVd secondary structure consists of 27 loops (L), denoted by numbers in red (L1-L27). To analyze the constraints on sequence diversity within loop regions, saturated mutant pools were prepared for four selected loops: L1, L6, L15, and L27. Structural models of these four loops are outlined in red rectangles, along with the corresponding loop IDs from the BGSU RNA site (http://rna.bgsu.edu/rna3dhub/) and the names of the RNAs from which the models were derived. The structure models of L1, L6, and L15 originated from the large subunit ribosomal RNA (LSU rRNA). The structural model for loop 27 corresponds to the loop region found within the histone 3’ untranslated region (UTR) stem-loop structure in animal cells. Bases in the structural models are labeled accordingly. In cases where model sequences differ from PSTVd sequences, the PSTVd sequences (in black) are provided in brackets. Saturated mutant pools were prepared for each of these loops, as outlined in black rectangles. However, due to technical limitations, saturating mutagenesis was only performed on the lower strand of L15. Ns represent nucleotides A, U, G, or C. The number of mutants in each pool was calculated and presented.
Fig 4.
Analysis of loop region sequence diversity in PSTVd quasispecies derived from loop mutant pools and structural modeling of variants.
Using the same methods described in Fig 2, deep sequencing data from a total of 40 libraries derived from mutant pools of L1, L6, L15, and L27 were processed to calculate (A) normalized Shannon entropy, (B) read numbers per thousand reads of the top 20 most frequently detected unique sequences, and (C) number of unique sequences per thousand reads. (D) Using JAR3D the unique sequences of loop mutants in each library were aligned to their corresponding structural models. The output cutoff scores represent the compatibility of the unique sequences with the structural models. Higher cutoff scores indicate a higher level of compatibility. Weighted average cutoff scores of unique sequences in each library were calculated as described in the text. Data are expressed as the mean ± standard deviations for IR, LM, and Sys samples, and as a single value for the pools. Student’s t-test was applied for data comparisons: *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, ns, not statistically significant.
Fig 5.
WT PSTVd can carry a mutant with a nearly correct structure in systemic trafficking.
(A) RNA blot detection of PSTVd in N. benthamiana plants inoculated with WT and mutants. Two variants with positive cutoff scores (U180C and C323A), and two double mutants with negative scores (C38U/C323U and G36U/A325G) were used to test whether they can be carried by WT PSTVd in systemic infection. In vitro transcripts (300 ng each) of the four mutants and WT PSTVd were prepared and rub-inoculated onto the first two true leaves of two-week-old N. benthamiana plants, with three plants allocated for each treatment. Additionally, mixed co-inoculation (mix) was carried out by combining 37.5 ng of in vitro transcripts from each of the four mutants with 150 ng of WT in vitro transcripts (total 300 ng per plant). At 20 dpi, total RNA samples were prepared from systemic leaves and PSTVd was detected by RNA blot analysis, with ethidium bromide staining of rRNA serving as a loading control. Circular PSTVd (PSTVd-C), the functional form, is shown. A sample positive for PSTVd (P) and a mock (M) control were included. (B) RT-PCR, a more sensitive method, was also used to detect PSTVd with actin as an endogenous control. (C) Progeny from three plants of the WT and mixed co-inoculation groups were analyzed by Sanger sequencing. The number of sequenced clones is indicated, with novel mutations highlighted in bold.