Systemic RNAi in planarians depends on spread of RNPs from active stem cells

  1. Sudheesh Allikka Parambil
  2. Karina Ascunce Gonzalez
  3. Axel Poulet
  4. Levi Cruz
  5. Hae-Lim Lee
  6. Mariana Witmer
  7. Josien C. van Wolfswinkel

Reviewed by

Read the full article

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

RNAi is a powerful cellular defense mechanism against genomic invaders that rely on dsRNA intermediates, such as viruses and mobile elements. Due to its specificity and ease of use, RNAi is also widely used as an experimental and therapeutic strategy to reduce levels of specific RNAs. For many systems however, delivery of the silencing agents into each individual cell is a major challenge. A mechanistic understanding of the processes involved in intercellular spread in systems with effective systemic RNAi thus is key to improve applications, and to increase our understanding of this defense mechanism. Remarkably, outside of C. elegans and plants, our molecular understanding of systemic RNAi remains very limited.

We here investigated the highly-effective systemic RNAi of the planarian S. mediterranea , which can be triggered by introduction of dsRNA via food or injection and rapidly spreads through the entire body. Notwithstanding its efficiency, we find no evidence of an RNA amplification mechanism or of transgenerational effects as are found in C. elegans , and rather find that planarian RNAi effects are limited in time. We identify the biogenesis factors involved in the planarian RNAi mechanism, and find that these are independent of the miRNA pathway, enabling the separation of the effects from these small RNA pathways. Surprisingly, we find that planarian systemic RNAi relies on active stem cells. Further, we identify Argonaute-siRNA complexes as the mobile agent that effectuates systemic spread of RNAi throughout the tissue.

These findings provide new insights into the mechanisms by which small RNAs spread between cells, and by which organisms can extend protection to all their cells upon encounter of a novel invading element. Additionally, our findings may have important implications for the design of effective applications in other systems.

Article activity feed

  1. Review Commons

    Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

    Learn more at Review Commons

    Reply to the reviewers

    The authors do not wish to provide a response at this time.

  2. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #4

    Evidence, reproducibility and clarity

    RNAi is remarkably efficient in planaria, yet no mechanism for the amplification of the RNAi signal has thus far been observed. In this manuscript, the authors analyse the mechanisms of RNAi spread in planaria. Starting from some basic observations on the identity of the Dicer and Argonaute proteins required for RNAi, the authors performed a set of elegant experiments to conclude that cycling stem cells likely take up dsRNA and excrete Ago-siRNA complexes, which are then taken up by other cells to mediate RNAi. In addition, the authors provide compelling evidence that RNAi is indeed independent of an amplification mechanism.

    Overall, I found the experiments and results compelling and the manuscript a pleasure to read. I have only a few suggestions for consideration, none of which are essential to support the main conclusions:

    • What does the arrest of stem cell proliferation do to the expression of RNAi genes (with and without dsRNA stimulation)?
    • Page 9: top panel. Is there a control that the dsRNA generated by RNaseIII is functional? I.e. that the defect is indeed due to an uptake effect and not the quality of the siRNA preparation itself? (In our hands silencing of siRNA prepared with bacterial RNaseIII has not been efficient at all). As a side note: no method is provided for the RNaseIII treatment.
    • Have the authors analyzed which of the Argonautes are present in the preparations generated with Q-sepharose?

    Data presentation:

    • For all figure legends: please make sure to state animals, number of repeats, define boxplots and what the individual data points represent. Please provide statistics where quantitative statements are made.

    Minor points:

    • First paragraph results: The statement that Ago1 and 3 were "closer to the nematode-specific WAGOs" does not seem correct, (horizontal distance to the miRNA-AGOs is still lower than the the WAGOs). I suggest removing the statement.
    • Use of checkmarks: please define when a checkmark vs cross was indicated? E.g., does a checkmark indicate that 100% of the animals showed efficient RNAi, or a majority of animals?
    • Many of the legends contain conclusions. While this may be a matter of taste/style, I would suggest to introduce conclusions only sparingly, if at all, in the legends
    • Some of the font sizes are rather small on print size (e.g. Fig 1A, S4i). In Fig 1A the black font on dark blue background is hard to distinguish.

    Textual suggestion:

    • Abstract "that rely on dsRNA intermediates, such as viruses" > ".. such as those from viruses..."
    • Materials and Methods: The lowerscript numbers for the ion show as squares in my pdf.

    Significance

    Strength/weaknesses:

    I found the experimental support robust and well supported and I did not find weaknesses that jeopardize the conclusions.

    Significance:

    One of the most intriguing features of RNAi is the systemic spread of a silencing signal across an organism's body. This has received significant attention in C. elegans and plants, but for other organisms, this is much less well explored. Planaria have a very efficient RNAi response, which the authors propose is due to uptake of an initial dsRNA by stem cell and excretion of an Argonaute-siRNA complex, which is then taken up by distal cells in an endocytic mechanism. I find this an intriguing mechanism that to differ from mechanisms for RNAi spread observed in other organisms.

    The work will be of interest to those interested in small RNA pathways (and RNA biology in general) and has practical implications for scientists working on planaria. The fact that small RNAs spread in an Argonaute-siRNA complex in an organism should also be of interest for cell biologists.

    My field of expertise: Small RNA pathways and antiviral defense in insects. No experience working with planaria.

  3. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    In RNA interference (RNAi), double-stranded RNA (dsRNA) is processed into small interfering RNAs (siRNAs), which can function locally or act as mobile RNA species that spread between cells. In organisms such as nematodes and plants, the underlying mechanisms and key factors involved in this process, including transporters such as SID-1, have been well characterized. While systemic RNAi has also been reported in other animals, the underlying mechanisms remain largely unclear. In this context, the authors focus on planarians as one such model to investigate these processes. In the planarian S. mediterranea, gene knockdown by dsRNA injection is commonly employed, and the RNAi effect is known to spread rapidly throughout the organism. However, given the absence of RNA-dependent RNA polymerase (RdRP), the mechanism by which RNAi signals are efficiently propagated remains unclear. In this study, the authors provide several important insights into this question.

    First, the authors carefully evaluated the duration of the RNAi effect. In addition, they systematically examined the involvement of known RNAi-related factors and demonstrated that this process depends on ago1 and ago3. Second, interestingly, the authors find that initiation of systemic RNAi depends on neoblasts. Third, Argonaute-siRNA complexes play a crucial role in systemic RNAi. This differs markedly from the nematode system, in which dsRNA itself is transported, highlighting an intriguing mechanistic distinction. Finally, the authors suggest that distinct Argonaute proteins may function at different stages of RNAi propagation. Ago1 + Ago3 play essential roles in the initial phase of systemic RNAi in neoblast, Ago3 but not Ago1 silences the target in the differentiated cells. While the phenomenon described here is highly interesting, the underlying mechanism remains to be fully elucidated. In particular, how different Argonaute proteins functionally coordinate with each other, especially with respect to the transfer of siRNAs between Argonaute complexes, is still unclear and represents an important direction for future studies.

    The study is supported by well-designed control experiments, and the results are consistent with and support the authors' conclusions.

    I have no major concerns about this manuscript. The study is well conducted, and I only have minor comments that could further improve the manuscript.

    (Minor) While the authors have examined the effects of irradiation on the donor, it would be interesting to test the reciprocal experiment in which the recipient is irradiated. In particular, assessing whether the addition of donor lysate to irradiated recipients can recapitulate the observed RNAi propagation would further strengthen the proposed model.

    (Minor) The purity of the AGO complexes obtained via the TraPR anion-exchange procedure is not entirely clear. The authors may consider providing additional evidence of purity (e.g., visualization of small RNAs with T4-PNK), which would strengthen the conclusions.

    (Minor) Figure 4H is not referred to in the main text. The authors may consider incorporating a description of this panel into the Results section for clarity.

    Significance

    Overall, given the substantial amount of data and the overall high quality of the present study, further mechanistic dissection would likely be beyond the scope of the current manuscript. I look forward to future work from the authors addressing these mechanistic questions in more detail. RNAi has been widely used in stem cell research in planarians. In light of the findings presented in this study, however, previous studies that combine UV irradiation and RNAi may warrant careful re-evaluation. In this regard, the present work has important implications and is likely to have a broad impact on the field.

  4. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    In this paper, the authors set out to understand how dsRNA elicits a system-wide RNAi effect using planarians as a model system. This is an important question, because it gets at evolution of these processes in different animal models and because knowing more about how RNAi works can allow scientists to tweak their approach for a better knock down efficiency. Importantly, though the system-wide mechanism of RNAi is fairly well understood in C. elegans and in some plants, it isn't clear how conserved these mechanisms are. Some aspects of this paper are quite convincing, including identification of the responsible Argonaute and Dicer proteins. Further, the identification of potential Sid-1 homologs that may allow for import of dsRNA is new. However, the role for Ago-3 was recently reported in Sasidharan, et al (Science Advances, 2026), which is not cited in this manuscript. Perhaps more importantly, several key aspects of the argument set up in this paper are not adequately supported and there are key gaps in the mechanism proposed that prevent its publication in this form. Major and minor suggestions follow:

    Major issues:

    1. The argument that siRNAs must be generated in stem cells that are cycling is not well supported.

    a. The authors only use one approach to reduce stem cell numbers, lethal irradiation. In addition to causing loss of stem cells, lethal irradiation causes wide-spread DNA damage and organismal/cellular stress responses. By 6 days after lethal irradiation, other progenitor cells are lost as well. Epidermal progenitors are known to be very abundant and to play signaling and/or metabolic roles in planarian physiology, so their loss may also be impactful. The authors should consider other orthogonal approaches to eliminate stem cells and to rule out other potential mechanisms.

    b. The authors use camptothecin in planarians and claim that it reduces cell divisions of stem cells. To my knowledge, this drug has not been shown to work in planarians before. The concentration used is also higher than in published studies. The authors should show whether stem cells are lost after this drug treatment (through levels of stem cell markers or stem cell counting) and should clarify the timing of the treatment relative to the RNAi, which is not clear from the figure legend or methods section. The authors should also discuss possible alternative interpretations of this piece of data (e.g. potential off-target effects). Without more information, it is hard to interpret the data relative to the irradiation results. The authors also do not provide any insight into how or why dividing/cycling stem cells would be important for the systemic RNAi mechanism they propose.

    c. ago-1, ago-3, and dcr-2 were shown to be enriched in stem cells (Fig. 3C), but these genes are also expressed in differentiated cells in single-cell sequencing data. Therefore, it isn't intuitive that non-stem cells would lack the capacity to generate siRNAs.

    d. Is there a way to directly test the hypothesis that stem cells are only generated in stem cells, potentially by blocking transport in some way and then visualizing new siRNAs with a miRNA/siRNA version of ISH OR FACS and sequencing? If the Ago-1/3-siRNA complexes are indeed transported by EVs as per Sasidharan, et al then the ESCRT(RNAi) approach might be useful in blocking movement of siRNAs? Or, could the authors show that dsRNAs are preferentially taken up by stem cells using the type of experiment shown in Fig. S5H?

    1. It isn't clear from the manuscript how the authors believe that Ago-1/3-siRNA complexes exit and enter cells. The diagram in Fig. 5F describes the complex as moving between cells either through vesicles or extracellularly. How do the authors propose that Ago-siRNA complexes pass through the plasma membrane given that they are not known to go through the secretory pathway? Or once endocytosed, how do they exit the vesicle? Uncertainty on these points makes the molecular mechanism proposed here seem poorly supported by the data provided in the paper.
    2. One key result in the paper is the transplant of "AGO complexes" that are purified from lysate. The authors writing about this experiment implies that they are transplanting a fairly pure material representing these RNPs and no others. However, the approach described is unlikely to result in purification of highly specific protein complexes. At a minimum, gels that illustrate protein/complex purity should be provided. Preferably, though, mass spectrometry and sequencing would be provided to detail siRNAs and proteins in this sample.
    3. The Sasidharan, et al (Science Advances, 2026) paper should be cited and also the findings of this paper should be put in the context of that work.

    Minor changes:

    1. In several experiments, quantitative assessment of impacts (e.g. eye size or ovo/opsin transcript levels) rather than subjective eye scoring would be preferable for rigor and for statistical analysis of changes rather than check/X (e.g. 4F-H).
    2. F1 and F2 terminology for regenerates is probably not accurate since F in those terms stands for "filial" and is used to denote offspring.
    3. The images in Figure 2 (A, C) are quite hard to see on the printed page. Using white for fluorescence might improve contrast and visibility.
    4. The element of time seems to be very important for transmission of the RNAi effect in sexual offspring. Instead of the claim that hatchlings from RNAi crosses have no effect (Fig. 2H), the detail provided in the results section seems to indicate that there is a time-limited effect. These findings should be clarified with progeny sorted by time of egg laying and with a better sense of time between RNAi injection and hatch. Further, even in animals that do regenerate eyes, it would be nice to see a quantification of transcript as a clearer readout of whether some knockdown persists.
    5. This is more of a curiosity question, but it would be interesting to know how the differences in Ago1/2/3 protein structure might relate to their function, particularly in terms of the PAZ and MID domains.

    Significance

    This paper provides some new insight into mechanisms underlying systemic RNAi in planarians. Some of the results are quite preliminary and the overall interpretation of data is not yet well founded. However, there are some highlights, including the potential identification of dsRNA transporters that will be interesting to those in the planarian field.

  5. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    Summary

    In this manuscript, the authors characterized the molecular mechanism of systemic RNAi in planarian Schmidtea mediterranea (Sme) through loss-of-function genetic perturbations. They genetically identified key protein factors involved in the siRNA pathway and assessed the systemic RNAi efficacy at the molecular level. Notably, they find that the proliferating stem cells (neoblasts) are specifically required for systemic RNAi in Sme. They further propose that the requirement of neoblasts in systemic RNAi is mediated by spreading of the RISC RNP to differentiated cells.

    Major Comments

    1. The authors show that in Sme systemic RNAi strictly relies on the presence of neoblasts, which is one of the most interesting finding. It is important to understand the mechanism, specifically whether neoblasts are generally required for dsRNA processing or for conferring mRNA slicing activity. Although the authors claimed in Figure 5 that neoblasts are required for siRNA biogenesis, the results provided do not directly support this claim. An alternative scenario is that dsRNA can still be processed into siRNA in the absence of neoblasts, but the resulting siRNA subsequently fails to function without the neoblast AGOs or other signals. To directly confirm that neoblasts are required for dsRNA processing, one additional experiment should be performed in which irradiated worms are injected with dsRNA, followed by small RNA cloning and sequencing to detect whether processed siRNAs are present.
    2. An alternative mechanism to interpret the role of neoblasts is that, instead of processing dsRNA and/or spreading RISC RNP, the neoblasts may function as regulatory cells that provide signals licensing the dsRNA processing and target slicing in differentiated cells. Under this scenario, the requirement for sid-1 and vha-16 could instead be interpreted as necessitate the dsRNA transfer from the initial uptake tissues (parenchyma for be injection or intestine for feeding) to the target tissues. To rule out this possibility, isolated neoblasts from naive donors could be transplanted into irradiated recipient worms who have been injected with dsRNA, and whether such transplantation can restore the systemic RNAi in the recipient animals could then be tested phenotypically. One caveat is that any positive result may be due to the proliferation of the donor neoblasts in the recipient. This can be addressed by performing the same transplantation but using neoblasts isolated from camptothecin-treated worms, which would limit the proliferative contribution.
    3. In Figure 5E, the authors show that recipient ago-3 is required for systemic RNAi, and they suggest in the Discussion a plausible model in which recipient AGO-3 is required for nuclear RNAi for transcriptional target repression. However, this result appears inconsistent with the results in Figure 1C-D, where ago-3 KD did not abolish systemic RNAi. This contradiction should be acknowledged in text and further investigated. One possible interpretation is that the presence of the neoblast ago-3 from the donor lysate may have an antimorphic effect and interferes with the recipient AGO(s) (presumably ago-1 in this case) during target silencing , implying that homogeneity of AGO(s), or at least homogeneity of ago-1, is required for such systemic RNAi. Although the underlying mechanism remains difficult to interpret, such hypothesis could be tested by injecting lysate from ago-3 KD donor into ago-3 KD recipient. If AGO homogeneity is indeed required, such transfer treatment should no longer abolish systemic RNAi in the recipient in Figure 5D. Additionally, the target genes used for the systemic RNAi in Figure 1C/D and Figure 5E are different. To exclude the possibility that this discrepancy is target-specific, either six1/2 should be tested in the whole worm RNAi assay in Figure 1 or opsin should be used in the transfer assay in Figure 5.
    4. The authors claim in Figure 4 that the systemic RNAi is mediated by secreted RISC. This claim is not unexpected, because naked siRNA generally suffers poor half-life in vivo and therefore must be stabilized by bounding to AGO to evade the endogenous ribonucleases. Nevertheless, the alternative hypothesis that the transferred RNAi is mediated by the spread of naked RNA, though unlikely, should be experimentally excluded. Specifically, the isolated RNA and lysate with protease in Figure 4F (which failed to induce RNAi in the host worm) should be tested to confirm whether they contain siRNAs. This can be done by cloning and sequencing the sRNA in the lysate.
    5. The authors assigned ago-1 and ago-3 to the siRNA pathway and ago-2 to the miRNA pathway. This is an important conclusion for subsequent sRNA studies in planarians. However, the evidence provided in the current manuscript is insufficient to exclude ago-2 from the siRNA pathway, especially given that DDH catalytic triad is present in AGO-2. The observed redundancy between ago-1 and ago-3 to maintain functional RNAi can only support the involvement of these two AGO genes in the siRNA pathway but does not exclude AGO-2. To more rigorously test whether ago-2 should be excluded from the siRNA pathway, double RNAi of ago-2 and ago-1, as well as of ago-2 and ago-3, should be performed, and ago-2 should only be excluded from the siRNA-pathway if such double KD do not further reduce the RNAi efficacy compared with individual KD.
    6. The results shown in Figure 1F, where exposure to exogenous dsRNA can enhance the endogenous transcription of ago-1 and ago-3 in Sme, are particularly interesting. The authors should discuss whether this phenomenon is related to nuclear RNAi. In addition, it has been reported that exposure to exogenous dsRNA can increase the AGO/DICER protein levels without increasing the mRNA level (PMID32194567), and this should be compared with the present findings. Importantly, the result also suggests a potential strategy to improve the Sme RNAi efficiency. Accordingly, it would be valuable to test whether the increased ago-1/3 transcript levels caused by introducing exogenous dsRNA can lead to higher RNAi efficacy, both in terms of target silencing depth and the duration of RNAi effectiveness.
    7. Figure 2I-J provides remarkable evidence that the systemic RNAi in Sme is independent of RdRP. This result should be highlighted in the final paragraphs of the Introduction and mentioned in the Abstract.

    Minor Comments:

    1. The authors show that ago-1 + ago-3 KD only slightly perturbed the miRNA levels. However, this observation can be interpreted in at least two ways: (a) these two AGO genes are not involved in the miRNA pathway; or (b) these two genes are expressed at low abundance (which was mentioned later in the paper), such that their KD only mildly perturb their associated miRNAs, especially if these miRNAs are also associated with AGO-2. Scenario (a) seems less likely to be true because ago-2 is enriched in neoblasts (Figure 3C), whereas many conserved miRNAs have been reported to be enriched in Xins in Sme (Sasidharan et al 2013). This issue should be therefore discussed. In addition, the gene expression levels of the three ago genes from previously published bulk RNAseq datasets should be included in the figure.
    2. The illustration in Figure 1A is not fully accurate. In the miRNA pathway, target repression also includes mRNA degradation (which is conventionally referred to as mRNA decay or mRNA destabilization), which is in fact the dominant mode of miRNA-mediated repression. Therefore, "mRNA decay" should be added in addition to "translational inhibition". In the siRNA pathway, mRNA degradation is not directly mediated by RISC itself, but by the downstream exonucleases (i.e., XRN-1); therefore, the term "mRNA slicing" should be used instead for the siRNA part. Additionally, it has been shown that C. elegans RDE-1 is also associated with miRNAs (PMID 36790166), so the functional assignment in the model should be adjusted accordingly.
    3. In Figure S1C, the authors claimed that ago-1 and ago-3 exhibit more divergent PAZ and MID domains according to the AF modeling. However, this divergence may simply reflect the lower sequence conservation of AGO-1 and AGO-3 relative to AGO-2, which is shown in the phylogeny in Figure S1A. To address this caveat, Robetta modeling should be performed for both the full-length proteins in the comparative modeling mode due to the length of AGO proteins, or de novo modeling of the PAZ and MID domain. Structural the alignment in reference to solved AGO structures such as 4W5Q or 6N4O should be shown. If the MID/PAZ domains divergency remains evident, it should be quantified using backbone RMSD relative to known AGO structures.
    4. In Figure S1C, a second structural view should also be included to better illustrate the AGO architecture. The duplex channel within the PIWI-MID lobe should be clearly visible in one of the views. The L2 domain, or at least helix-7, should be labeled. If possible, the relative position of helix-7 to the guide RNA should also be shown. All the predicted structural models should be included in the supplemental files.
    5. The authors suggest that the spread of functional RISC from neoblasts depends on EVs. The evidence involving vha-16 is convincing, but to directly validate the presence of EVs that cargo RISC, CsCl ultracentrifugation would be informative. Although this experiment is beyond the scope of the current manuscript, the need for direct EV validation should be discussed.
    6. In Figure 2G, the authors show that although zfp-1 restores the homeostatic mRNA level at week 5, its downstream target prog-1 and agat-3 fail to recover. It remains unclear whether this is due to the delay of newly translated zfp-1 to activate the downstream targets, or due to translational suppression of zfp-1. Therefore, the mRNA levels of prog-1 and agat-3 should be further monitored beyond week 5.
    7. In Figure 3, the authors use co-expression by in situ hybridization to demonstrate the expression of ago in neoblasts. To provide the whole-animal context, co-expression of smedwi-1 and ago genes should also be confirmed using the current Sme scRNAseq datasets.
    8. The authors proposed in the Discussion that AGO-1 may sponge unwound RNA duplex and this facilitates the dsRNA transfer. This interpretation seems unlikely, because the ago-1 single KD, which would abolish such dsRNA transfer, did not show phenotypes in terms of systemic RNAi defect. Also, such scenario suggests that loss-of-function of ago-1 may be antimorphic since the sponged dsRNA were released, and thus co-KD of ago-1 and ago-3 should result in more efficient RNAi. These concern should be discussed.
    9. The Discussion states that AGO-1 is required in the donor, but this is inconsistent with the results in Figure 5D, where ago-1 KD in the donor did not abolish RNAi in the recipients. This inconsistency between results and text should be corrected.
    10. In Figure S5I, the authors show that short dsRNA generated by RNase III digestion failed to induce systemic RNAi in sid-1 loss-of-function condition. However, the alternative explanation is that RNase III digestion produced short dsRNAs that may result in siRNA with suboptimal length for AGO loading or functioning. This caveat should be mentioned, and the length profile of the RNase III digestion products should be shown by high density urea gel electrophoresis or HPLC.
    11. In all the transfer assays, one concern is the lysate may contain viable neoblasts, so that any observed results could be attributed to the proliferation of the donor-derived neoblasts rather than the transfer of RNAi materials. Therefore, a cell viability test using Calcein AM or other equivalent assay should be conducted to confirm the absence of live cells in the lysate preparation protocol.
    12. In the second paragraph of the introduction, when comparing the siRNA and miRNA pathways, the difference in base-pairing configuration with the target site should be introduced with appropriate reference.
    13. In the last paragraph of the introduction, the claim that the results may have implications for the design of effective RNAi-based therapies is too vague. Given that the current therapeutic siRNA delivery methods are already robust in clinical applications, the authors should more specifically explain how their findings in Sme might inform therapeutic development.
    14. In the last sentence of the second last paragraph of the introduction and Figure S5I, "RNAse" should be corrected to "RNase".
    15. In the first Results subsection, the second last paragraph, first sentence, one left parenthesis is missing.
    16. Throughout the Discussion, the term "AGO-RNA". If the authors intend to express a distinction from RISC, how this terminology differs from RISC should be justified. Otherwise, RISC would be more appropriate.
    17. Statistical significance should be shown in Figure 2E, 2G, 3A, 3C, S2A, S2D, S2G, S3D, S5D, S5G, S5J.
    18. Molecular weight should be labeled in Figure S5L.
    19. In Figure 2J, where the y-axis indicates % prevalence, the down-facing bars (antisense reads) should also be labeled as positive values on the y-axis. Displaying them as negative percentages (-20%) is incorrect.
    20. The small RNA cloning procedure should be described in the Methods. Basic information of sRNA sequencing, including read numbers, biotype distribution, proportion mapping to the triggering dsRNA, should be included too.
    21. The methods used to measure RNA and protein concentrations should be included in the Methods section.
    22. The irradiation protocol, including dosage, should be included in the Methods.
    23. In the Methods section, subscripts of chemical formulas are rendered as squares throughout the text. This formatting issue should be corrected.
    24. In the Results section, first subsection, second paragraph and first sentence. The cited data should be Suppl Figure S2D, not the current S2A-C.
    25. The manuscript uses inconsistent formatting for supplemental figures (for example, "Suppl Figure S1B,C" versus "Suppl Figure 2A-C"). The formatting should be standardized.

    Significance

    Planarians have long been appreciated as a robust model organisms for studying gene function in animal regeneration, and one major advantage of this system is its highly efficient systemic RNAi. However, the molecular basis of the RNAi machinery has not been thoroughly investigated, and detailed RNAi efficacy hasn't been evaluated. This study therefore provides important value by characterizing the molecular components underlying systemic RNAi in Sme, which contributes to both fundamental understanding and to potential optimization of RNAi-based experiments in Sme.

    In addition, the manuscript reports that stem cells are required for systemic RNAi in differentiated cells in Sme, a finding that has not been described in other organisms. Although the underlying mechanism remains unresolved, this observation offers potentially important implications for both RNA biology and stem cell biology.

  6. Version published to 10.64898/2026.03.03.709314 on bioRxiv