Mutational profile of the regenerative process and de novo genome assembly of the planarian Schmidtea polychroa

  1. Ádám Póti
  2. Dávid Szüts
  3. Jelena Vermezovic

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Planarians are organisms with a unique capacity to regenerate any part of their body. New tissues are generated in a process that requires many swift cell divisions. How costly is this process to an animal in terms of mutational load remains unknown. Using whole genome sequencing, we defined the mutational profile of the process of regeneration in the planarian species Schmidtea polychroa . We assembled de novo the genome of S. polychroa and analyzed mutations in animals that have undergone regeneration. We observed a threefold increase in the number of mutations and an altered mutational spectrum. High allele frequencies of subclonal mutations in regenerated animals suggested that relatively few stem cells with high expansion potential regenerated the animal. We provide, for the first time, the draft genome assembly of S. polychroa , an estimation of the germline mutation rate for a planarian species and the mutational spectrum of the regeneration process of a living organism.

Article activity feed

  1. Review Commons

    Note: This rebuttal was posted by the corresponding author to Review Commons. 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 #3

    Evidence, reproducibility and clarity

    The manuscript under evaluation investigates the mutation rate between generations and during regeneration in the planarian species S. polychroa. Abundant adult pluripotent stem cells, the potential somatic stem cell contributions to the germ line, and the regeneration of entire animals from tiny tissue fragments provide interesting conceptual background to these questions. Specifically, the authors assemble a draft genome of a purportedly triploid S. polychroa strain with an accompanying de novo transcriptome. Via shallow whole genome shot gut sequencing, the authors subsequently attempt to estimate the germline mutation rate and the effect of regeneration on the number and the spectrum of detectable mutations. As stated in the abstract, the study "...provides, for the first time, the draft genome assembly of S. polychroa, the germline mutation rate for a planarian species and the mutational spectrum of the regeneration process of a living organism". As detailed under "major points" below, the draft genome assembly is not great and significant concerns remain regarding both regeneration rate statements.

    Major points

    Genome and transcriptome assembly:

    1. As the authors state, ploidy is known to be highly variable across S. polychroa strains and ploidy is an important variable in the estimation of mutation rates. The authors should therefore provide additional experimental evidence that their strain is indeed triploid (e.g., through karyology or FACS genome size estimation).
    2. Assembly quality assessments: As stated, the authors intend to use a de novo assembled short-read transcriptome as an independent assessment of assembly quality. However, no information on the quality of the transcriptome reference is provided. Salient quality control metrics on the size, completeness, and assembly contiguity of their transcriptome need to be provided. The published and publically available S. polychroa transcriptome on PlanMine [cited by the authors] might provide a further useful reference.
    3. The assembly is highly fragmented ( 8700 contigs) and the authors detect a significant number of scaffolding errors within contigs (Fig. 3d) . Moreover, the large number of multi mappers might indicate that the draft assembly has been insufficiently purged of haplotigs. This is especially a concern in a potentially triploid species and should be assessed via a kmer-based approach such as Merqury. Ideally, an independent Illumina dataset from the mutation screening and the reads used for polishing should be used. The analysis of synteny and structural variation relative to S. mediterranea in Figure three is therefore of questionable relevance and should be dropped or significantly condensed.

    Mutation analysis:

    1. The SNP analysis is based on a pipeline that the authors published previously. Beyond this, the manuscript provides insufficient information in terms of technical details (e.g., the salient IsoMut settings, how duplicate reads were treated, and how the SNP calling approach guards against the ever-present possibility of sequencing artifacts). In addition, the authors should discuss whether the pipeline adequately accounts for the i) triploid genome and ii) the fragmented and potentially insufficiently purged assembly. In absence of such information, the validity of the results remains difficult to ascertain.
    2. The statistical support for the germline mutation rate is weak (Fig. 5a). The sample size of the experiment is rather low for such studies, with only four Filia in each group. Furthermore, in both the control and regenerate group, two of the filia are siblings. This creates a data dependence that is not adequately discussed or taken into account for the analysis. For example, the authors conduct a t-test to determine if the number of detected mutations differs between the groups. A critical assumption of the t-test is that the samples are independent, an assumption that is violated when a relatedness structure is present. Similarly, this dependence could further influence the analysis of the mutational profile. Hence the authors either have to increase the sample size or temper the interpretation of their results, including the claimed estimation of germline mutation rate.
    3. The possibility of somatic mosaicism that the authors discuss extensively in the context of regeneration also complicates the interpretation of the clonal mutations between parents and filia. First, somatic mosaicism has been already demonstrated in a different planarian species and discussed in multiple reviews (e.g., PMID: 31221097, PMID: 35862435). This literature needs to be cited adequately.
      Second, the plausible contribution of individual somatic stem cells to the germ line leaves open the possibility that the observed parent/offspring differences in the control group also reflect rare pre-existing allele heterogeneities within the parental population of pluripotent stem cells. Therefore, clonal differences between parents and offspring cannot simply be attributed to germline mutations. Third, low, but measurable rates of sex are known to occur even in predominantly parthenogenetic S. polychroa populations (e.g., PMID: 16721392; PMID: 15293852]. These studies need to be cited and the possibility of parental genome contributions needs to be explicitly examined, as it would violate the requirement for an isogenic background stated on page 4. Overall, this means that the author's claim of providing the first quantification of the germ cell mutation rate in planarians is therefore insufficiently justified.
      The possibility of somatic mosaicism also impacts the interpretation of the apparent increase in genetic variation during regeneration. Given the limited depth of the sequencing assays, it remains difficult to refute the null hypothesis that the apparent increase in the mutation load of regenerates represents a subsampling of the standing genetic variation in the parent animals (and without the single-cell populational bottleneck of parthenogenetic reproduction). Also, the claims regarding the mutagenic nature of the regeneration process should therefore need to be dropped or significantly toned down.

    Minor comments:

    Fig.2a: The S. polychroa genome size estimates from genomesize.com Table S1 and Figure 2 a: The entries from the animal genome Size Database need to be removed from the figure, as this is published background information and not an analysis result.

    Page 6: The text description of the transcriptome backmapping results (Fig. 3A) is confusing: "...17.4 % were not mapped by GMAP,... the remaining transcripts were mapped as duplicates, at multiple positions or in two chimeric fragments... The authors need to insert the fraction of single mappers, as otherwise, they imply that they only obtain multi-mappers.

    Page 7: PlanMine needs to be cited as the source of the orthology information between S.med and S. pol.

    Page 10: What is the COSMIC database? Please explain/reference.

    Fig. 4 and 5: The experimental set-up cartoon in Fig. 4a is confusing and should more clearly illustrate which of the experimental groups involved regeneration, how many individuals were sequenced, and the meaning of the A/B terminology in subsequent graphs. Moreover, the authors need to ensure consistent symbol use, e.g., Fr1, 2, 3, 4 instead of the current F1, F2, ... in Fig. 5a.

    The authors discuss the potential contribution of methylation to the observed mutation spectrum and conclude that it might not be present in S. mediterranea and S. polychroa. Indeed, the lack of measurable mC in the planarian genome has already been demonstrated (PMID: 24063805). Please cite and shorten the respective text section.

    Fig. 6e: The authors estimate the number of stem cells in tail segments using H3P staining. However, H3P marks only a short segment of the cell cycle and therefore underestimates the number of resident stem cells. This caveat needs to be discussed.

    Typo in Figure S2: The splice site is marked wirth a black

    Significance

    Planarian flatworms harbor abundant pluripotent adult stem cells. These cells are the only division-competent cells in the animal and of pivotal importance to planarian biology. For example, they enable the regeneration of entire animals from tiny tissue pieces or the re-formation of the germ line in sexually reproducing strains. How planarians maintain their genetic identity in the face of abundant pluripotent adult stem cells and a strict soma/germline divide raises many intriguing questions.

    The manuscript provides a preliminary and highly fragmented draft genome assembly of the planarian species S. polychroa, which adds to the available planarian genome information. Based on the genome assembly, the manuscript attempts to measure generational mutation rates during parthenogenetic reproduction and regeneration. The quality of the SNP detection is somewhat difficult to evaluate in the current manuscript and the possibility of somatic heterogeneity in the parents raises concerns regarding the interpretation of the supposed germline mutation rate. The data provide further evidence for somatic mosaicism, which has already been demonstrated in a different planarian species. The extent by which regeneration is mutagenic per se or uncovers standing genetic variation due to the inherent population sub-sampling also remains unclear. Overall, the manuscript stands out as one of the first intra-organismal population genomics studies in the field. But I think not all its claims are sufficiently supported by data.

    I am a planarian biologist with experience in planarian genomics. I am not an explicit population genetics/genomics expert.

  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 #2

    Evidence, reproducibility and clarity

    Summary:

    This paper presents the genome of the planarian Schmidtea polychroa, a sister species to the widely used regenerative model S. mediterranea. The preliminary analysis of the genome is sound and the authors establish that they have produced a good draft assembly. The authors then leverage this assembly to ask novel questions about mutation rates in regenerative planarians, a question that has not been well addressed previously. They use a clear and logical approach to measure mutation rates in relation to both development and regeneration and establish that different clones of stem cells exist in planarians and contribute to regeneration and production of germ cells.

    Overall the work is of a high quality, and logically explained.

    Major comment.

    I could find no statement about raw data availability or metadata availability, or availability of intermediate data analysis files. This makes proper review and consideration of the authors analysis essentially impossible, so all of this must be taken on trust and easy to do but helpful further analyses in the context of the existing data structure can't really be suggested. This is a great shame. Furthermore, in this reviewers' opinion this goes against the principle of open science. A revision must address this issue unless the reviewers' are planning to publish this paper in the 1990s print format only (forgive the sarcasm, but I hope the authors can concede this isn't a good way forward). Without data access the full impact of their very exciting work cannot land. If I have missed reference to access to the data or a GitHub link etc in the paper then I apologise, but I have looked extensively. Sequence Data submission can take time so they should do this in advance and perhaps share the link to the unreleased link too reviewers, and intermediate files and metadata can be shared on GitHub or the like.

    Minor comments.

    I enjoyed reading the paper immensely and I think it touches on many important and interesting theoretical ideas in the field.
    With regards to their comments on methylation they should note that previous work on S. mediterranea has rigorously shown that methylation is absent or very low, probably any residual comes from base scavenging from the calf liver food source. Additionally, DMNT 1 and 3 are absent from S. mediterranea, so canonical enzymes for methyl-cytosine addition and maintenance are not available. Citing this would be useful (Jaber-Hijazi et al, 2013, Developmental Biology https://doi.org/10.1016/j.ydbio.2013.09.020). Other work suggests there might be methylation in this group using retriction enzyme based approaches.

    I have some questions that relate to evolution of a parthenogen.
    Did the authors ever detect homozygous changes between "parental" generations and offspring or changes from a heterozygous state to a homozygous state?
    Given the parthogenic and triploid nature of Polychroa did the authors detect high levels of/accumulation of heterozygous alleles generally in the genome?

    Significance

    This paper will eventually be very significant to the regenerative biology community as it will give us comparative genomic capabilities for the well-established model S. mediterranea. Although not commented on much in the manuscript this is very important. Furthermore, the paper begins the important work of characterising mutation rates in this group of animals that avoid the ageing process entirely, this work will be another important foundation stone in understanding the phenomena that allow for this in this group of animals.

    I have expertise in genome assembly, analysis and annotation, as well in assessing variation in genomes. I also have expertise in the model system used here and its general biology, including regeneration.

  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 #1

    Evidence, reproducibility and clarity

    In this paper, using the triploid biotype of planarian Schmidea polychroa, the first half of the paper presents the results of the analysis of genome structure and the second half shows that (de novo) mutations in individuals that undergo regeneration are passed on by the next generation.

    While I think this paper contains interesting biological findings, I am skeptical about its novelty. I was convinced by the results and discussion of the analysis of genome structure, but the results and that of the analysis of (de novo) mutation were very confusing. This may be due to my lack of knowledge in this field. But even so, the author needs to improve this manuscript so that the general reader will better understand it.

    Major comments:

    1. The author mentions that it is important to note that this study was conducted using a parthenogenetic triploid biotype. However, I think that the parthenogenesis undergoing by a triploid biotype of S. polychroa is very unusual. It is not typical apomictic parthenogenesis. Triploid oocytes arise by meiosis from hexaploid oocytes derived from triploid adult somatic stem cells called neoblasts. On the other hand, haploid sperm arise by meiosis from diploid spermatogonia derived from neoblasts. Embryogenesis of triploid eggs then occurs by pseudogamy. Occasional sex is also known to occur even if the offspring's chromosome number remains triploid. I think this background is important information to give the reader. Also, don't the authors need to treat the results in this paper with this complex phenomenon also taken into account?
    2. Fig.4B-C: Analysis by lineage-specific mutations of parental controls.
      The authors do not specifically mention or discuss this result. What about the accumulation of mutations within such populations in typical parthenogenesis (daphnia and aphids)? In other words, are the results in Fig. 4B-C due to the special mechanism for parthenogenesis in the triploid S.polychroa as described above?
    3. Throughout this paper, the authors show that regeneration increases de novo mutations in the progeny. The authors conclude that many of the mutations occurred in neoblasts during regeneration. However, I would like you to explain the biological significance of this results in S. polychroa, which naturally does not reproduce by fission and regeneration. There are already reports of mutations accumulating in neoblasts in Dugesia japonica, which reproduce aexually by fission. For these reasons, I do not think this paper presents extremely novel results.
    4. p15, Discussion:
      "Tissue regeneration is best seen in the liver of mammals, and the regrowth of relapsed tumours following surgery can also be considered an example of a regenerative process. Mutagenesis accompanying these processes is relevant to subsequent tumorigenesis or the development of resistance, and the planarian system can provide a useful model for the mutagenic effect of tissue regeneration."

    Isn't it an overstatement to associate the regenerative system of planaria with the liver regeneration of mammals?

    1. p10, Results:
      "We compared the two de novo spectra to the spectrum of germline heterozygous SNPs, present in all animals, and found that the pattern of germline substitutions resembled more closely the de novo spectrum of the control group (Fig 5D, Fig S3), implying that regeneration has a minor contribution to germline mutations in S. polychroa populations."
      p14, Discussion:
      "The high similarity of the spectrum of heterozygous SNPs and de novo mutations of control animals suggests that the species primarily reproduces in a non-regenerative manner. The increased mutation rate and the altered mutation spectrum upon regeneration confirmed our hypothesis that regeneration is a mutagenic process."

    I was very confused by these sentences and it took me some time to understand them. Triploid S. polychroa naturally does not reproduce by fission and regeneration, namely a non-regenerative manner. I do not understand why the author insists on this. Please explain the results for the regenerated case in Fig. 5D (0.88) in a way that is also easy to understand. Also, what is the biological significance of asserting here that de novo mutation by regeneration increases in a species that does not increase by regeneration and division in the first place?

    Minor comments:

    1. The author should add a schematic diagram showing the distribution of reproductive organs in Fig.1 to help the reader understand that the ovaries are not included in the regenerative fragment.
    2. P12, line12: Fig 6D-E, it's F, not E, right?
    3. P9, line 8:
      "these mutations were missing from the original egg but were present in the egg laid by the parent and thus represent the total mutation load of a generation."

    The author mentions that the de novo mutation found in offspring derived from parents that do not undergo regeneration was already present in the eggs, but I can find no evidence of this. Can you rule out the possibility that these mutations occurred between hatching and adulthood?

    1. p10, Results:
      "Interestingly, the majority of mutations were shared in the siblings F4A and F4B. This suggests that the germ cells of these animals were descendants of the same stem cell, which underwent a high number of cell divisions early during the regeneration process prior to oocyte differentiation. The same finding also confirms that the detected clonal filial mutations were present in the respective oocyte and were not generated by embryonic cell divisions."

    The shared de novo mutations detected in the siblings (F4A and F4B) derived from the parent that underwent regeneration in Fig. 5A suggest that the germ cells of these siblings are descended from the same stem cell. The authors say that these mutations occurred in a large number of cell divisions early in the regenerative process prior to oocyte differentiation.

    So why is there no shared de novo mutation in the siblings (Fc4A and Fc4B) derived from the non-regenerating parent in Fig. 5A? As mentioned in Minor comment 3, the author states that the de novo mutations were already present in the parent-laid eggs, but when did these mutations, which are not shared, arise?

    1. p11, Results:
      "Interestingly, in the case of FR4A-FR4B sibling pair, shared de novo mutations present in both were subclonal in R4 in a proportion comparable to the other samples (7/15 by WGS, 46.7%), while the three unique mutations could not be detected in R4 by the PCR approach, indicating again that the unique mutations, which amounted to approximately 10% of total clonal filial mutations in these two animals, arose late during germ cell regeneration."

    "during germ cell regeneration." the expression is too vague to know which stage you are referring to. In relation to minor comment 4, why not create a new chart to clearly show when the expected mutations occurred?

    1. p12, Results:
      "Altogether 7/30 regenerant mutations were detected in PR animals, and these included those with the highest AF in the regenerants (Fig. 6C). This suggests that parental animals, even before regeneration, contained a diverse set of stem cells, and some of the detected de novo mutations in the filial generation resulted from the expansion of mutation-containing stem cell clones contributing also to germ cells in the regenerant animals."

    If the mutation in the offspring is derived from the parent (PR) prior to the time of tail amputation, wouldn't it be wouldn't it be strange to assume that it is a de novo mutation?

    1. p12, Results:
      "The remaining 23/30 R- subclonal mutations may have arisen during regeneration. On average, ~250 dividing neoblasts were detected in cut tails of animals from the same population as the sequenced individuals, as determined by immunofluorescence of phosphorylated H3 histone (Fig 6D-E). However, the high proportions of body cells carrying regenerant-specific mutations suggest that certain stem cells contribute to disproportionately large parts of the regenerated body, including the germline."

    I did not quite understand the relevance of this discussion to the photos shown here of the M period (Fig. 6e).

    Significance

    General assessment: This paper contains important biological information. The finding that mutations in planarian stem cells cause diversity in the next generation of parthenogenesis is very interesting. However, I think that the author needs to carefully explain and change his argument, for example, that the mutations were caused by regeneration, which does not naturally occur in the species used.

    Advance: The finding that accumulation of mutations is occurring in planarian stem cells has already been reported in Dugesia japonica. Please cite the papers and clarify what is the key finding in this paper.

    Audience: Basic Research_Evolutionary Ecology, Developmental Biology (Stem Cells), Reproductive Biology

    Please define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate.

    My field of study is reproductive biology. I am familiar with the transcriptome but unfamiliar with genome analysis.

  5. Version published to 10.1101/2023.07.20.549885v1 on bioRxiv