The Transmission of Intergenerational Epigenetic Information by Sperm microRNAs

Round 1

Reviewer 1 Report

Title: The transmission of intergenerational epigenetic information by sperm microRNAs

Reviewer:

Grace S. Lee1 & Colin C. Conine summarize the contributions to the field with emphasis on miRNAs. It is a very valuable choice with such a complex system and variation of a large number of noncoding RNAs. miRNAs have the advantage at this stage of being short (stable and mobile), each of them having specificity for a given mRNAs and almost all mRNAs transcribed in a cell are targeted by a known miRNAs, with few exceptions. In other words targets and functions could be evaluated.

Epigenetic heredity is an important and highly debated topic whose most often considered tools are DNA methylation, histones modifications and RNA alterations.

The current perspective on microRNAs has the clear advantage that the epigenetic change can be experimentally induced directly by RNAs. It covers almost all possibilities of inducing RNA variations.

It is a faithful review of each contribution to the field and a description of the articles without taking a personal position. 

Hereafter are just a few suggestions  and comments:

Authors: lines 230- 237

Reviewer:
Regarding a role of epididysomes: the phenotypes transferred via the epididysomes are not studied further than F0, which should be specified. In case if it is /or not transferred further? 

Authors: 85, 86

Histone modifications are considered as demonstrated to transmit epigenetic information mitotically to maintain cell-fate through cell cycles in species ranging from yeast to mammals.

Reviewer:
I am not sure we can conclude from the results presented in these articles, and even with the modified proteins that are fixed forever it`s an artefact, it is stuck that's all, like a glue.

Authors: 226

major remodelers of the sperm small RNA payload: 1) The small RNAs present in tran     

Reviewer:

I found the word major too strong, a remodeler perhaps but not major.

Authors: 80-81

In all instances, the single sense strand of the miRNA rather than duplex miRNA was injected, suggesting that the functions these injected miRNAs are initiating are possibly AGO-independent.

Reviewer:

I am not sure about the independent AGO conclusion, because single-stranded miRNA won`t stay single-stranded for long as such it would have to hybridize to a complementary sequence.

Authors: 588

Finally, we must return to the big question – how can a sperm miRNA generate an altered phenotype in a fully developed offspring?

Reviewer:

We certainly agree that this is definitely the most important direction for future investigators:

“That is a question”.

Author Response

We would like to thank reviewer 1 for their gracious and useful comments on our manuscript. Below we address the comments accordingly.

Reviewer 1

Grace S. Lee1 & Colin C. Conine summarize the contributions to the field with emphasis on miRNAs. It is a very valuable choice with such a complex system and variation of a large number of noncoding RNAs. miRNAs have the advantage at this stage of being short (stable and mobile), each of them having specificity for a given mRNAs and almost all mRNAs transcribed in a cell are targeted by a known miRNAs, with few exceptions. In other words targets and functions could be evaluated.

Epigenetic heredity is an important and highly debated topic whose most often considered tools are DNA methylation, histones modifications and RNA alterations.

The current perspective on microRNAs has the clear advantage that the epigenetic change can be experimentally induced directly by RNAs. It covers almost all possibilities of inducing RNA variations.

It is a faithful review of each contribution to the field and a description of the articles without taking a personal position. 

Hereafter are just a few suggestions and comments:

Authors: lines 230- 237

Reviewer:
Regarding a role of epididysomes: the phenotypes transferred via the epididysomes are not studied further than F0, which should be specified. In case if it is /or not transferred further? 

            We thank the reviewer for this comment. In attempt to clarify this concern we have added a sentence to the end of the “Toxins” section that reads. “However, the ability of epigenetic information in epididymosomes to produce transgenerational epigenetically inherited phenotypes in succeeding generations of progeny (F₂ and beyond) has not yet been determined.”         

Authors: 85, 86

Histone modifications are considered as demonstrated to transmit epigenetic information mitotically to maintain cell-fate through cell cycles in species ranging from yeast to mammals.

Reviewer:
I am not sure we can conclude from the results presented in these articles, and even with the modified proteins that are fixed forever it`s an artefact, it is stuck that's all, like a glue.

While histone modifications as carriers of epigenetic information across cell-divisions is a controversial concept. It is clear that silencing marks such as H3K9me3 can transmit inherited information important for cell-state. Please see the following citations below which we have included in the text.

Reinberg, D. & Vales, L. D. Chromatin domains rich in inheritance. Science (New York, N.Y 361, 33-34, doi:10.1126/science.aat7871 (2018).

Audergon, P. N. et al. Epigenetics. Restricted epigenetic inheritance of H3K9 methylation. Science (New York, N.Y 348, 132-135, doi:10.1126/science.1260638 (2015).

Authors: 226

major remodelers of the sperm small RNA payload: 1) The small RNAs present in tran     

Reviewer:

I found the word major too strong, a remodeler perhaps but not major.

We took out the word “major”.

Authors: 80-81

In all instances, the single sense strand of the miRNA rather than duplex miRNA was injected, suggesting that the functions these injected miRNAs are initiating are possibly AGO-independent.

Reviewer:

I am not sure about the independent AGO conclusion, because single-stranded miRNA won`t stay single-stranded for long as such it would have to hybridize to a complementary sequence.

We thank the review for this comment as “AGO-independent” is misleading. What we meant here is that injected single stranded miRNA is functioning through AGO by non-canonical loading and/or targeting (perhaps to chromatin or another unexpected function). We have changed the sentence to read “In all instances, the single sense strand of the miRNA rather than duplex miRNA was injected, suggesting that the functions these injected miRNAs are initiating are possibly through a non-canonical miRNA pathway (see below).”

Authors: 588

Finally, we must return to the big question – how can a sperm miRNA generate an altered phenotype in a fully developed offspring?

Reviewer:

We certainly agree that this is definitely the most important direction for future investigators:

“That is a question”.

We changed this to say “the prominent lingering question”.

Reviewer 2 Report

The authors wrote an article focusing on the male transgenerational effects driven by sperm miRNAs. In general, is a well written review manuscript describing the latest data published for miRNAs on transgenerational studies, starting with a brief and clear explanation for miRNA biogenesis, followed with their function on sperm and embryo development. Then, the text brings those together to describe alterations on the offspring phenotypes due to different stressors, centering the attention on the lack of causality for miRNAs, with a couple of exceptions properly described on the literature. The Conclusions section agrees with the body of the text including the stoichiometry issue, and the “non-canonical” miRNA functions that could explain the regulation of the expression programs at embryonic stages.

There are some specific comments that I hope the authors find useful for the manuscript described below:

• Page 2, lines 58-68: The sentences that explain why they focus on the paternal transmission are weak because they lack bibliographic support. A reformulation of them sentences with a strong bibliographic support or key reported examples will certainly aid in explaining the paternal transmission focus of the article.

• Page 2, lines 72-76: the authors wrote “Heritable DNA methylation is critical to development as it is required to properly transmit imprinting from one generation to the next5; however, de novo establishment and maintenance of DNA methylation patterns outside of genomically imprinted regions for the purpose of transferring the experience of parents to progeny is not well established”. Is the experience transferred? or the molecular information generated due to the parent’s experience? Please clarify these sentences.

• Page 2, lines 95-97: the authors wrote “For example, genetically eliminating specific histone PTMs using mutants for chromatin modifying proteins in the male germline has been shown to illicit phenotypes in succeeding generations”. Please, can the authors explain the meaning of “illicit phenotypes”? I understand the meaning of the sentence, but probably the authors can reformulate it with a more precise term.

• Page 3, lines 102-124: This paragraph is totally correct, but lacks the bibliographic support for all the effects described here.

• Page 5, lines 205-207: the authors wrote “Hundreds of miRNAs are expressed during sperm development, some constitutively, such as let-7 miRNAs, and others during specific stages of the process” this sentence should be cited.

• Page 5, lines 225-244: In this paragraph the authors neatly explain the input of the sperm small RNA payload. Nevertheless, the remodeling of the payload also includes the “output” (or degradation) of the small RNAs. For this reason, and to be compliant with the remodeling concept, I suggest the authors to include information about the degradation of these ncRNAs during the epididymal transit.

• Page 6, lines 289-292: the authors wrote “In a follow up study, gene expression phenotypes associated with caput embryos were successfully rescued by injecting synthetic RNAs mimicking some of the most abundant sperm miRNAs acquired in the epididymis which originate from genomic clusters (see above).” The first part of this sentence is fine, but the last part is vague, genomic clusters of…? genomic clusters defined by…? Please reformulate to clarify the sentence.

• Page 7, lines 298-299: Why the authors don’t refer to Figure 2?

• Page 7, lines 315-320: Is there any bibliography supporting this hypothesis? Needle shearing is less potent than sonication. This can be observed when preparing chromatin. With the needle is not possible to shear chromatin for a proper ChIP. My question arises following the detailed description of Kuretake at al. Biol Reprod 1996. Cite 51 supports miRNA embebed in the sperm nucleus, but not the transient vesicle-sperm interaction. Please, provide information supporting the proposed hypothesis.

• Page 7, lines 322-324: the authors wrote “Treating mature mouse sperm treated with lysolecithin, pronase, and RNases A and H removes 90% of the RNA associated with sperm.” Please, reformulate the sentence to avoid the redundancy.

• Page 13, lines 27-29: the authors wrote “Further, microinjection of two miRNAs, miR-221 and miR-222, which both? Naturally target Kit, into naïve WT zygotes also produced mice with white tail tips.” Please, reformulate.

• Page 14, lines 86-87: the authors wrote “they are none-the-less interesting in the context of this review as they are initiated by the microinjection of RNAs…” Why none-the-less and not nonetheless? in any case, for non-measurable things, nevertheless is more appropriate.

• Page 15, lines 112-114: the authors wrote “Similarly in mice, male mice fed a HFD to induce obesity also produced F1 offspring of both genders which displayed increased bodyweight and insulin resistance compared to control progeny” Please, reformulate to avoid redundancy.

• On the written sections Diet & Metabolism, Stress, and Toxins, the authors explain what is described on the Table 1. Why there is no reference to Table 1? Additionally, I suggest the reformulation of the paragraphs to avoid the excess of redundancy between the text and the table. For example, the list of miRNAs that is already described on the table do not have to be re-described on the text, except for key examples. This will make the text easier to read.

• Page 17, line 243: should say affects instead of effects?

• Page 18, line 314: Delete the “(“, because there is no parenthesis closure on the paragraph

• Page 18, lines 316-317: should say “the fact that…” instead of “that fact that…”

• Page 21, line 459: Homogenize with numbers please.

• Table 1. Although the table is truly informative, in my opinion, the term “paradigm” does not fit here. I see different kinds of environmental alterations applied to the subjects of study. My suggestion would be to reformulate the head of the first column. Also, I would suggest separating the examples in mouse, rat, and human to make it more explicit for the readers.

Author Response

We thank the reviewer for their comments and recommendations. We think that they will certainly improve the manuscript.

Reviewer 2

The authors wrote an article focusing on the male transgenerational effects driven by sperm miRNAs. In general, is a well written review manuscript describing the latest data published for miRNAs on transgenerational studies, starting with a brief and clear explanation for miRNA biogenesis, followed with their function on sperm and embryo development. Then, the text brings those together to describe alterations on the offspring phenotypes due to different stressors, centering the attention on the lack of causality for miRNAs, with a couple of exceptions properly described on the literature. The Conclusions section agrees with the body of the text including the stoichiometry issue, and the “non-canonical” miRNA functions that could explain the regulation of the expression programs at embryonic stages.

There are some specific comments that I hope the authors find useful for the manuscript described below:

• Page 2, lines 58-68: The sentences that explain why they focus on the paternal transmission are weak because they lack bibliographic support. A reformulation of them sentences with a strong bibliographic support or key reported examples will certainly aid in explaining the paternal transmission focus of the article.

We thank the reviewer for this comment and have rewritten this section with bibliographic support. “The study of epigenetic inheritance via the maternal gamete is complicated by the direct impacts of the maternal environment on the fetus (F₁) during gestation⁵. Further, because F₁ generation female animals complete gametogenesis during gestation, the gametes that will make the F₂ generation are also directly exposed to the altered environmental. Therefore, in female mammals, the F₂ and F₃ generation of exposed females must be studied to demonstrate non-genetic intergenerational and transgenerational effects, respectively. Additionally, to eliminate the confounding effects of maternal care, F₂ and F₃ animals must be raised by unexposed animals⁶. An alternative method to study intergenerational inheritance through the female germline is to isolate eggs exposed to the altered environmental condition, use IVF to fertilize the egg with naive sperm, and implant the embryo into a surrogate naive female.”

• Page 2, lines 72-76: the authors wrote “Heritable DNA methylation is critical to development as it is required to properly transmit imprinting from one generation to the next5; however, de novo establishment and maintenance of DNA methylation patterns outside of genomically imprinted regions for the purpose of transferring the experience of parents to progeny is not well established”. Is the experience transferred? or the molecular information generated due to the parent’s experience? Please clarify these sentences.

We thank the reviewer for this comment; however, it is unclear to us what the reviewer is asking.  DNA methylation on imprinted regions of the genome is clearly transmitted from one generation to the next to establish imprinting in successive generations of progeny. However, DNA methylation outside of imprinted regions has not been demonstrated to be inherited during fertilization and then protected against erasure during the global demethylation process that occurs after fertilization.

• Page 2, lines 95-97: the authors wrote “For example, genetically eliminating specific histone PTMs using mutants for chromatin modifying proteins in the male germline has been shown to illicit phenotypes in succeeding generations”. Please, can the authors explain the meaning of “illicit phenotypes”? I understand the meaning of the sentence, but probably the authors can reformulate it with a more precise term.

 Thank you for this correction. We changed the phrase to “induce phenotypes”.

• Page 3, lines 102-124: This paragraph is totally correct, but lacks the bibliographic support for all the effects described here.

     We added additional appropriate citations in this paragraph by citing a review by Sharma for the introduction sentence and later with regard to sperm/egg ratio Lessells et al. The sentences in between are set up for the topic of the review and are all cited through the text.

• Page 5, lines 205-207: the authors wrote “Hundreds of miRNAs are expressed during sperm development, some constitutively, such as let-7 miRNAs, and others during specific stages of the process” this sentence should be cited.

            Thank you. This sentence certainly needed a proper citation. We have cited the following review:

Reza, A. et al. Roles of microRNAs in mammalian reproduction: from the commitment of germ cells to peri-implantation embryos. Biol Rev Camb Philos Soc 94, 415-438, doi:10.1111/brv.12459 (2019).

• Page 5, lines 225-244: In this paragraph the authors neatly explain the input of the sperm small RNA payload. Nevertheless, the remodeling of the payload also includes the “output” (or degradation) of the small RNAs. For this reason, and to be compliant with the remodeling concept, I suggest the authors to include information about the degradation of these ncRNAs during the epididymal transit.

      This is a valid point. However, very little is known about the degradation of small RNAs during epididymal maturation. We have added the following to the end of this paragraph: “Contrarily, degradation or loss of RNAs in sperm during epididymal transit is poorly understood. During maturation in the epididymis, sperm discharge much of their RNAs through the loss of the cytoplasmic droplet⁵⁰. Taken together, the RNAs present in sperm are the result of an equilibrium of RNAs acquired by fusion with epididymosomes, the loss of RNAs through the cytoplasmic droplet, and RNAs retained during development in the testis.”

• Page 6, lines 289-292: the authors wrote “In a follow up study, gene expression phenotypes associated with caput embryos were successfully rescued by injecting synthetic RNAs mimicking some of the most abundant sperm miRNAs acquired in the epididymis which originate from genomic clusters (see above).” The first part of this sentence is fine, but the last part is vague, genomic clusters of…? genomic clusters defined by…? Please reformulate to clarify the sentence.

    We modified this sentence to improve clarity. “In a follow up study, gene expression phenotypes associated with caput embryos were successfully rescued by injecting synthetic RNAs mimicking some of the most abundant sperm miRNAs acquired in the epididymis (see above)”.

• Page 7, lines 298-299: Why the authors don’t refer to Figure 2?

             We added a reference to Figure 2 as suggested.

• Page 7, lines 315-320: Is there any bibliography supporting this hypothesis? Needle shearing is less potent than sonication. This can be observed when preparing chromatin. With the needle is not possible to shear chromatin for a proper ChIP. My question arises following the detailed description of Kuretake at al. Biol Reprod 1996. Cite 51 supports miRNA embebed in the sperm nucleus, but not the transient vesicle-sperm interaction. Please, provide information supporting the proposed hypothesis.

     We thank the review for the comment. As state in the manuscript, this is a hypothesis to explain the data that we hinted at in the below citation which is now cited. Additionally, we’ve added the following sentence to make it clear that there is currently not actual data supporting this hypothesis: “However, the dynamic localization of miRNAs transmitted by epididymosomes to sperm during epididymal transit has not been characterized.”

Conine, C. C., Sun, F., Song, L., Rivera-Perez, J. A. & Rando, O. J. Sperm Head Preparation and Genetic Background Affect Caput Sperm ICSI Embryo Viability: Cauda-Enriched miRNAs Only Essential in Specific Conditions. Developmental cell 55, 677-678, doi:10.1016/j.devcel.2020.11.021 (2020).

• Page 7, lines 322-324: the authors wrote “Treating mature mouse sperm treated with lysolecithin, pronase, and RNases A and H removes 90% of the RNA associated with sperm.” Please, reformulate the sentence to avoid the redundancy.

                        Edited as suggested.

• Page 13, lines 27-29: the authors wrote “Further, microinjection of two miRNAs, miR-221 and miR-222, which both? Naturally target Kit, into naïve WT zygotes also produced mice with white tail tips.” Please, reformulate.

                        Edited as suggested.

• Page 14, lines 86-87: the authors wrote “they are none-the-less interesting in the context of this review as they are initiated by the microinjection of RNAs…” Why none-the-less and not nonetheless? in any case, for non-measurable things, nevertheless is more appropriate.

                  Edited as suggested.

• Page 15, lines 112-114: the authors wrote “Similarly in mice, male mice fed a HFD to induce obesity also produced F1 offspring of both genders which displayed increased bodyweight and insulin resistance compared to control progeny” Please, reformulate to avoid redundancy.

                        Edited as suggested.

• On the written sections Diet & Metabolism, Stress, and Toxins, the authors explain what is described on the Table 1. Why there is no reference to Table 1? Additionally, I suggest the reformulation of the paragraphs to avoid the excess of redundancy between the text and the table. For example, the list of miRNAs that is already described on the table do not have to be re-described on the text, except for key examples. This will make the text easier to read.

We added references to Table 1 throughout the manuscript and eliminated lists of miRNAs in the text as suggested.

• Page 17, line 243: should say affects instead of effects?

                        Thank you for catching this. This typo was corrected.

• Page 18, line 314: Delete the “(“, because there is no parenthesis closure on the paragraph

                        Edited as suggested.

• Page 18, lines 316-317: should say “the fact that…” instead of “that fact that…”

                        Edited as suggested

• Page 21, line 459: Homogenize with numbers please.

                        Edited as suggested

• Table 1. Although the table is truly informative, in my opinion, the term “paradigm” does not fit here. I see different kinds of environmental alterations applied to the subjects of study. My suggestion would be to reformulate the head of the first column. Also, I would suggest separating the examples in mouse, rat, and human to make it more explicit for the readers.

We changed the first column title to “Experimental Design” instead of “Paradigm”.  Regarding the second suggestion, we would prefer to keep the content in this order with the examples of different model organisms because many of these are placed in this order because they are from the same publication and because they are placed in the same order as discussed in the text.

Reviewer 3 Report

This review by Conine and coauthor covers an exciting field of sperm small RNA mediated epigenetic inheritance. The article summarizes the molecular functions of miRNAs in the male germline and sperm, and discusses the paternal environmental exposure (focuses on the examples including paramutation, diet & metabolism, stress, toxins) and their effects on sperm miRNAs and offspring phenotypes. At the end, the authors also provides important prospectives and gaps. This article is expected to be very useful to researchers in the epigenetics. It is generally balanced and well covered with literatures except for missing one important concept as discussed below.

Although the authors clearly stated the focus of the review on the microRNAs, other well documented sperm small RNAs (e.g., tsRNAs) in transmitting parental phenotypes are worth to mention. Also, the sperm small RNA modifications are reported to be able to transmit parental phenotypes to the offspring via shaping early embryo development. This line of research is an emerging field; thus it would be nice to include in this review.

Other minor suggestions:

Table 1: The references is suggested to put in a seperate column or detail citations (e.g., author, year) can be provided to help readers. Also suggest to concise texts in the table, bullet the major points…

Conclusions: The first two paragraph can be at least partially moved to introduction, summarize the idea that the ancestral environment can influence the phenotype would suffice.

Author Response

We thank the reviewer for their supportive comments and suggestions for improving the manuscript.

Reviewer 3

This review by Conine and coauthor covers an exciting field of sperm small RNA mediated epigenetic inheritance. The article summarizes the molecular functions of miRNAs in the male germline and sperm, and discusses the paternal environmental exposure (focuses on the examples including paramutation, diet & metabolism, stress, toxins) and their effects on sperm miRNAs and offspring phenotypes. At the end, the authors also provides important prospectives and gaps. This article is expected to be very useful to researchers in the epigenetics. It is generally balanced and well covered with literatures except for missing one important concept as discussed below.

Although the authors clearly stated the focus of the review on the microRNAs, other well documented sperm small RNAs (e.g., tsRNAs) in transmitting parental phenotypes are worth to mention. Also, the sperm small RNA modifications are reported to be able to transmit parental phenotypes to the offspring via shaping early embryo development. This line of research is an emerging field; thus it would be nice to include in this review.

            We added a section in the Introduction about tsRNAs. “Two classes of small non-coding RNAs have been implicated as carriers of epigenetic information that transfer information through sperm, microRNAs (miRNAs or miRs) and transfer RNA fragments (tRFs or tsRNAs). The composition of sperm small RNAs is unique in that rather than being predominately comprised of microRNAs like most somatic cells, ~80% of their composition are fragments of transfer RNAs and ribosomal RNAs (rRFs or rsRNAs)^21,22. While originally disregarded as remnants of degradation after spermatogenesis, they have become increasingly recognized as a class of small regulatory RNAs with the ability to contribute non-genetically inherited information intergenerationally. For example, mice fed altered diets exhibit regulated levels of specific tRFs in their sperm (for example tRF-Gly-GCC) and produce progeny with metabolic phenotypes different from their control counterparts^23,24. Interestingly, this inheritance can be recapitulated in animals developed from naive zygotes microinjected with tRFs purified from altered diet sperm²⁴. These findings demonstrate that tRFs can function as carriers of epigenetically inherited information from sperm in mammals. However, much is still to be learned about how tRFs regulate gene expression and whether different species of tRFs from different tRNA isoacceptors have gene regulatory functions.”

Other minor suggestions:

Table 1: The references is suggested to put in a seperate column or detail citations (e.g., author, year) can be provided to help readers. Also suggest to concise texts in the table, bullet the major points…

            We thank the reviewer for the suggestion have edited the text as suggested.

Conclusions: The first two paragraph can be at least partially moved to introduction, summarize the idea that the ancestral environment can influence the phenotype would suffice.

            We thank the reviewer, however we believe this is a stylistic preference and would prefer to keep this text in the Conclusions.