Compensatory sequence variation between trans-species small RNAs and their target sites

Essential revisions:

Some experimental data are required to show that Cuscuta HS-sRNAs indeed use sequence variations to target homologous genes with corresponding sequence variations in different plant species or members within the same gene family in the same host plant. The following experiments are suggested.

Overall we've addressed this by adding new data and analyses that demonstrate targeting of homologous mRNAs, with synonymous sequence variants, from Nicotiana benthamiana. The newly added data are in new Figure 6, and Supplementary files 9 and 10.

1) Generate complementation lines of seor1, or apk2b, or other 1-2 of the 19 examples that contain synonymous mutations. The lines contain the mutated sequences in the synonymous sites but not cleaved by the Hi-sRNAs, and compare the infection results with the mutant and normal complementation lines in Arabidopsis. The reviewers realized that generating stable transgenic lines takes a long time and thus suggested the authors to provide results if they already have the lines.

We agree that these experiments would be informative. We have started producing the plant materials for these experiments, but they are not ready in time for this manuscript.

2) Perform 5' RACE experiments in non-Arabidopsis plants infected by different Cuscuta species to confirm that the predicted targets are cleaved by HI-sRNAs.

Confirmation of targeting in Nicotiana benthamiana was shown for Cuscuta campestris miRNAs by 5’ RACE and secondary siRNA identification in our prior publication (Shahid et al., 2018). To address this question in the context of this project, we re-analyzed the prior sRNA-seq data with our genome-free pipeline, confirming targeting for many of these transcripts through the presence of secondary siRNAs (now shown in Supplementary files 9 and 10). We show a more in-depth example of a sRNA superfamily targeting multiple TIR1 homologs in N. benthamiana (Figure 6), which illustrates a clear case where synonymous variants of a target (comparing N. benthamiana and A. thaliana) are compensated for by miRNA superfamily variation.

3) Examine the cleavage of target sequences from different hosts by a few representative Cuscuta HI-sRNAs using reporter assays in tobacco leaves.

Thank you for the suggestion. While potentially informative, we feel this experiment is outside of the scope of our present study. We do worry that artificial reporter assays might not reflect the true situation during parasite infestation, primarily because of the different modes of microRNA delivery.

4) Perform qRT-PCR or RNA-seq analysis in Arabidopsis and other hosts infected by Cuscuta species to show that the target genes are indeed down-regulated by Cuscuta HI-sRNAs.

Done. In the modified Figure 3 we show an analysis of confirmed targets based on mRNA-seq in Arabidopsis, confirming a general trend of target down-regulation during parasite infestation. We have also now performed qRT-PCR experiments from C. campestris – N. benthamiana samples for several confirmed targets. In general these N. benthamiana targets also trend toward down regulation, with the TIR1 example significantly down-regulated (newly added Figure 6).