Paternal obesity alters the sperm epigenome and is associated with changes in the placental transcriptome and cellular composition

Article activity feed

1. 

   eLife

   Feb 9, 2024

   Author Response

   Reviewer #1 (Public Review):

   Using a HFD mouse model, the authors examined the H3K4me3 mark in sperm and placental tissues followed by correlation to the transcriptomic changes in the placental tissues of the male and female offspring. The hypothesis that the authors tried to test was that sperm histone epimutations affect placental function, thereby leading to metabolic disorders in offspring. The strength of this work includes the interesting idea and the initial data generated. However, the entire study remains purely correlative without any validation experiment to support the correlation. The conclusion needs to be further supported by bigger sample size and more functional analyses demonstrating the causal relationship among the histone epimutations detected, the dysregulated mRNA expression in the placenta, and …

   Author Response

   Reviewer #1 (Public Review):

   Using a HFD mouse model, the authors examined the H3K4me3 mark in sperm and placental tissues followed by correlation to the transcriptomic changes in the placental tissues of the male and female offspring. The hypothesis that the authors tried to test was that sperm histone epimutations affect placental function, thereby leading to metabolic disorders in offspring. The strength of this work includes the interesting idea and the initial data generated. However, the entire study remains purely correlative without any validation experiment to support the correlation. The conclusion needs to be further supported by bigger sample size and more functional analyses demonstrating the causal relationship among the histone epimutations detected, the dysregulated mRNA expression in the placenta, and the phenotypes in offspring.

   Functional data: We appreciate that we should have emphasized and written more clearly that we had indeed phenotyped the placentas and offspring metabolic health from the same model we derived the placenta tissue from as we reported in (Jazwiec et al., 2022)(PMID: 35377412). This was referenced in our submitted manuscript (Lines 105-107; 131-133; 135-139; 147-150; 232-235; 270-273; 297-300; 384-386; 433-435; 441-448; 507-514). We have made this more apparent in the manuscript by expanding our description of the offspring phenotypes in the introduction and clarified that it was from this model that the placenta’s used in this study were derived from (Jazwiec et al., 2022) (PMID: 35377412).

   Regarding effect and sample size: It appears that on review the animal numbers used for the ChIP-seq were confused with the number of replicates by the reviewers. These details were in Supplementary file 1a. There were 3 replicates per experimental group and each replicate contained sperm from pooled samples that was equalized in cell number and comprised of sperm from n=7 control males, or n=16 HFD males. For the RNA-seq n=4 placentas were used from each experimental group from both males and females for a total N of 16. Although the sample size is moderate, we followed the Canadian Council of Animal Care guideline which calls for the use of the lowest animal number that elicits significant effects (CCAC guidelines p6 “Consideration must also be given to reduction, to determine the fewest number of animals appropriate to provide valid information and statistical power, while still minimizing the welfare impact for each animal”).

   Validation: We used a high standard of computational validation and visualization strategies, to ensure confidence in genomic data. This also allowed for a comprehensive understanding of the biological and physiological impacts of paternal obesity on the sperm epigenome and placenta transcriptome. In our experimental design we also included biological and technical replicates. Together these methods provide robustness checks of the experimental data and support our conclusions. These are the validation strategies we used:

   Technical and experimental validation

   • We evaluated the quality of sequencing data using metrics of read quality, alignment and coverage. These are summarized in Supplementary file 1a.

   • Visualized and performed statistical analysis of data to check for anomalies and discrepancies, Pearson correlation analysis shown on heatmap to look for variance and patterns in samples- all here highly correlated (Figure 2 – Figure supplement 1 B and Figure 4 – Figure supplement 1 A). We checked for batch effects and normalized the data (Figure 4 – Figure supplement 1 B) we used PCA plot analysis as a second check for sample behaving oddly (Figure 2 – Figure supplement 1 C and Figure 4 – Figure supplement 1 C).

   • We used a deconvolution approach to improve the biological meaning of our bulk RNA-seq data (Figure 6, Figure 5 – Figure supplement 1 and 2).

   • Performed functional enrichment analysis to gain insight into biological functions, pathways, and genome ontology and visualized individual regions identified to be altered as a confirmation (Figure 2 D and 2 E; Figure 4 E and F; Figure 6, Figure 2 – Figure supplement 1 E; Figure 3 – Figure supplement 1). Comparison to external data sets:

   • We compared our data with external data sets using the same tissues and cell and to our prior studies: a) We compared ChIP-seq data from this obesity model with our former obesity ChIP-seq data (Figure 2 – Figure supplement 1); b) re-analyzed and compared placenta RNA-seq data from an in utero exposure hypoxia model that shared similar offspring and placenta phenotypes as we observed in the obesity model (Figure 6 and Figure 6 – Figure supplement 1).

   • We used a deconvolution approach to improve the biological meaning of our bulk RNA-seq data (Figure 6, Figure 5 – Figure supplement 1 and 2). Statistical Significance and False Discovery Rate (FDR):

   • We applied statistical tests and multiple testing corrections to reduce the likelihood of false positives (See also response 1 for additional testing added to the revised manuscript)

   Causation versus correlation: We agree that the relationship between the sperm epigenome and placenta transcriptome is correlative, however this is the current state of the field for studies of paternal epigenetic transmission of environmental information. To take this study to the point where causation can be implied would require the generation of a sperm epigenome edited mouse model where we target genes implicated in placental function. Indeed, this targeting approach is well underway in our research program.

   Reviewer #2 (Public Review):

   This study follows up on previous work from this group, and others, relating paternal diet to changes in sperm epigenetics, and offspring phenotypes. The authors focus on paternal diet (high-fat diet versus a control chow), sperm chromatin, and molecular changes in the placenta associated with offspring development.

   The text is well written and the figures are generally well presented and clear. The sperm epigenetic analyses and analysis of the placenta epigenetics and gene expression are generally well performed. The study provides new insight into how paternally mediated intergenerational epigenetic inheritance could involve placenta-embryo signaling.

   A major weakness is that the high-fat diet used was from a different manufacturer than the control (lower fat) diet. Therefore, it is difficult to judge whether the effects are due to a change in fat levels, or the many other molecules that are likely to differ in chow between different manufacturers. Other weaknesses include lack of methodological detail in parts, low n values for some experiments, and the need for more mechanistic data.

   Diets: It is worth reminding that we are studying the effects of obesity and not diet. Indeed, HFD induces metabolic dysfunction while the control does not. Although it is fair to point out that the composition of the control diet should be kept in mind, considering the desired outcomes within the scope of the study, the diets elicited the desired phenotypic effects serving as a model for obesity. We see this experimental design as a strength, as in this study we compared this model to our previous published obesity model (Pepin, Lafleur, Lambrot, Dumeaux, & Kimmins, 2022) (PMID: 35183795), and there was significant overlap in the regions of differential enrichment detected between both models even though they were conducted in different research settings, with different mouse substrain and different diet combinations. In our opinion this demonstrates that we are measuring robust effects of paternal obesity that can be replicated under different conditions. This comparative study design has been lacking in the field of epigenetic inheritance.

   Animal numbers and replicates: It appears that on review the animal numbers used for the ChIP-seq were confused with the number of replicates by the reviewers. These details were in Supplementary file 1a. There were 3 replicates per experimental group and each replicate contained sperm from pooled samples that was equalized in cell number and comprised of sperm from n=7 control males, or n=16 HFD males. For the RNA-seq n=4 placentas were used from each experimental group from both males and females for a total N of 16. Although the sample size is moderate, we followed the Canadian Council of Animal Care guideline which calls for the use of the lowest animal number that elicits significant effects (CCAC guidelines p6 “Consideration must also be given to reduction, to determine the fewest number of animals appropriate to provide valid information and statistical power, while still minimizing the welfare impact for each animal”).

   Whilst the authors may have achieved their aims, more data is needed to inform a potential mechanism.

   It is difficult in studies on paternal epigenetic inheritance to attribute a mechanism and we agree that the relationship between the obesity altered sperm epigenome and the placenta abnormalities are correlative. However, the novelty in our study is that we postulate a new mechanism for paternal transmission of metabolic disease that implicates the placenta and demonstrate this via an altered placenta transcriptome and placenta developmental abnormalities described here and in our previous paper on this model ((Jazwiec et al., 2022); PMID: 35377412). The next steps for the field to address causation/mechanism requires generation of a sperm epigenome edited mouse model where we induce and track histone methylation changes at specific genes to the tissues in the next generation. Indeed, this targeting approach is underway in our research program.

   Reviewer #3 (Public Review):

   This study represents a useful addition to the authors' previous study examining the effects of paternal high-fat diet on offspring metabolism and gene expression in offspring (PMID: 35183795). It differs from the previous study in some of the details of the experimental model (age of sire when exposed to the diet manipulation, mouse substrain, and the nature of the control diet) and the results are largely in line with previous findings. The major finding is that many genes at which sperm H3K4me3 signal is altered also have altered expression in the placenta; some of these genes are paternally imprinted, providing a paternal-specific epigenetic signature. Strengths of the study include establishment of an important dataset correlating the sperm epigenome with gene expression in placental tissue, leading to an interesting and provocative conclusion. Weaknesses include a relatively superficial analysis of the dataset, revealing broad patterns but few specific conclusions, reliance on correlative analysis to draw conclusions, and absence of validation studies. Deconvolution analysis of bulk RNA-seq data helps to account for differences in cell composition between placental datasets, but does not add additional insight toward the central question of how sperm epigenetic state contributes to offspring gene expression. Overall the advance over previous work is relatively small.

   Specific points:

   1. The analysis as it stands is limited. To compare sperm H3K4me3 and placental expression, numbers of overlapping genes are provided, but no statistical analysis is done to indicate the significance of the overlap.

   Fisher’s exact test to overlap paternal obesity-associated differentially enriched regions of H3K4me3 deH3K4me3) with female and male placenta differentially enriched genes (Figure 4 – Figure supplement 1 Di and ii).

   1. There is little direct connection to biological systems or validation of differential enrichment/expression analysis. Gene ontology enrichments for genes differentially enriched for H3K4me3 in sperm or differentially expressed in placenta (broken up by sex) are performed, but the biological significance of these categories is not clear.

   We used a high standard of computational validation and visualization strategies, to ensure confidence in genomic data. This also allowed for a comprehensive understanding of the biological and physiological impacts of paternal obesity on the sperm epigenome and placenta transcriptome. In our experimental design we also included biological and technical replicates. Together these methods provide robustness checks of the experimental data and support our conclusions. The validation strategies we used are detailed in response 17.

   We revised the text to expand discussion on the observed enriched gene ontology terms, as well as the biological significance and functions of the genes we refer to in this section:

   Lines 222-227: “The placenta is a rich source of hormone production, is highly vascularized, and secretes neurotransmitters (Hemberger, Hanna, & Dean, 2020; Rosenfeld, 2021). Disruption in these functions is suggested in the significantly enriched pathways that included genes involved in the transport of cholesterol, angiogenesis, and neurogenesis (Figure 4 C-D, Supplementary file 1e-f). Other significantly enriched processes included genes implicated in nutrient and vitamin transport (Figure 4 C-D).”

   Lines 441-463:“Many of the DEGs in the paternal obese-sired placentas were involved in the regulation of the heart and brain. This is in line with paternal obesity associated to the developmental origins of neurological, cardiovascular, and metabolic disease in offspring (Andescavage & Limperopoulos, 2021; Binder, Beard, et al., 2015; Binder et al., 2012; Chambers et al., 2016; Cropley et al., 2016; de Castro Barbosa et al., 2016b; T. Fullston et al., 2012; Tod Fullston et al., 2013; Grandjean et al., 2015; Huypens et al., 2016; Jazwiec et al., 2022; Mitchell, Bakos, & Lane, 2011; Ng et al., 2010; Pepin et al., 2022; Perez-Garcia et al., 2018; Terashima et al., 2015; Thornburg et al., 2016; Thornburg & Marshall, 2015; Ueda et al., 2022; Wei et al., 2014). The brain-placenta and heart-placenta axes refer to their developmental linkage to the trophoblast which produces various hormones, neurotransmitters, and growth factors that are central to brain and heart development (Parrettini, Caroli, & Torlone, 2020; Rosenfeld, 2021). This is further illustrated in studies where placental pathology is linked to cardiovascular and heart abnormalities (Andescavage & Limperopoulos, 2021; Thornburg et al., 2016; Thornburg & Marshall, 2015). For example, in a study of the relationship between placental pathology and neurodevelopment of infants, possible hypoxic conditions were a significant predictor of lower Mullen Scales of Early Learning (Ueda et al., 2022). A connecting factor between the neural and cardiovascular phenotypes is the neural crest cells which make a critical contribution to the developing heart and brain (Hemberger et al., 2020; Perez-Garcia et al., 2018). Notably, neural crest cells are of ectodermal origin which arises from the TE (Prasad, Charney, & García-Castro, 2019), which is in turn governed by paternally-driven gene expression. It is worth considering the routes by which TE dysfunction may be implicated in the paternal origins of metabolic and cardiovascular disease. First, altered placenta gene expression beginning in the TE could influence the specification of neural crest cells which are a developmental adjacent cell lineage in the early embryo. TE signaling to neural crest cells could alter their downstream function. Second, altered trophoblast endocrine function will influence cardiac and neurodevelopment (Hemberger et al., 2020).”

   1. The overall effect size is small. In most cases the magnitude of differences is minor, and it is not clear which of these changes are significant over noise. For example, the y-axis for the metagene plots in Figure 2B does not start at zero, so the total range of the difference in H3K4me3 is small. In Figure 6C, DEGs detected in hypoxic placenta after deconvolution analysis do not look very different compared to control.

   Thank-you for pointing out that the scales were different in Figure 2 Bi and ii. They have been revised to show the same Y axis scale beginning at zero for comparison of regions that gained and lost H3K4me3 making the differences in H3K4me3 more readily visible. The heatmap shown in Figure 6 C visualizes the DEGs in hypoxic vs control placenta where 1477 DEGS were identified in our re-analysis using a convolution approach applied to the bulk-seq data set from Chu et al., 2019. We do not share the view that they are not well visualized in the heat map.

   1. Deconvolution analysis was done on bulk RNA-seq data from placenta, and the numbers of DEGs identified with this analysis compared to the original analysis are shown, but is not clear how the deconvolution analysis changes the specific biological conclusions. In addition, the reference dataset for deconvolution is a published dataset generated in another lab, and it is unclear how comparable the reference sample is to the samples analyzed in this study, or how robust this analysis is when using a dataset generated under different conditions.

   The deconvolution analysis allows to infer cellular composition within a tissue and suggests that there are changes in cell-type proportion that could change placenta function and improves the detection of differentially expressed genes (Aliee & Theis, 2021; Campbell et al., 2023; Kuhn, Thu, Waldvogel, Faull, & Luthi-Carter, 2011) (PMID: 34293324; 36914823; 21983921).

   As per the published dataset used as a reference sample for the deconvolution analysis, it was ideal -we specifically chose this dataset for this analysis as the tissue of origin matched for the same mouse strain and developmental type points as our samples and those used in the Chu et al., 2019 analysis. We used the Chu et al., 2019 data set for comparative validation, and to further explore whether the biological effects of paternal obesity were like those of a hypoxic placenta. We have revised the text to more clearly show the biological relevance and interpretation of this analysis (see author response 12)

   We revised the text to clarify the biological implications of this analysis:

   Lines 282-290: “This reduction in the number of detected DEGs before versus after accounting for cellular composition suggests that changes in cell-type proportions at least partly drive tissue-level differential expression. This is consistent with the recent finding that preeclampsia-associated cellular heterogeneity in human placentas mediates previously detected bulk gene expression differences (Campbell et al., 2023). There were similarities between the bulk RNA-seq and deconvoluted analysis in that there was overlap of DEGs detected before and after adjusting for cell-type proportions (Figure 5 – Figure supplement 3 G and H, Fisher’s exact test P=1.8e-105 and P=0e+00, respectively). This differential gene expression analysis accounting for cellular composition provides insight into how paternal obesity may impact placental development and function and underscores the contribution of cellular heterogeneity in this process.”

   Reviewer #4 (Public Review):

   The members of the Kimmins lab perform a dietary study in mice to investigate the impact of obesity of fathers on the development of their offspring. To do so, they expose male mice to a high fat diet and determine the distribution and occupancy levels of the histone H3 lysine 4 trimethylation (H3K4me3) mark in spermatozoa and perform gene expression studies on placenta tissue obtained from mouse embryos during mid-gestation development. The authors report changes in H3K4me3 occupancy in sperm as well as in transcriptomes of placentas of male and female embryonic offspring. While the authors perform extensive computational analysis of the transcriptomic and chromatin immunoprecipitation data, the authors do not go much beyond making correlative statements at mainly the genome wide level between changes for H3K4me3 in sperm and transcriptional changes in placenta, the latter of which are in part related to changes in cellular composition (as deduced from transcriptional data). Given that both parental mice had the same genetic background, it was not possible to deduce parental specific contributions to transcriptional changes as observed in placentas of offspring. In all, the study falls short in increasing mechanistic insights into this important biological phenomenon.

   It is difficult in studies on paternal epigenetic inheritance to attribute a mechanism and we agree that the relationship between the obesity altered sperm epigenome and the placenta abnormalities are correlative. However, the novelty in our study is that we postulate a new mechanism for paternal transmission of metabolic disease that implicates the placenta and demonstrate this via an altered placenta transcriptome and placenta developmental abnormalities described here and in our previous paper on this model ((Jazwiec et al., 2022); PMID: 35377412). The next steps for the field to address causation/mechanism requires generation of a sperm epigenome edited mouse model where we induce and track histone methylation changes at specific genes to the tissues in the next generation. Indeed, this targeting approach is underway in our research program.

   Read the original source
2. 

   eLife

   Feb 9, 2024

   eLife assessment

   This important study presents data suggesting that HFD-induced histone epimutations in sperm may impact the transcriptome of the placenta, thereby contributing to the paternal transmission of paternal metabolic disorders to offspring. Although the hypothesis is interesting and the evidence presented is compelling, more careful statistical analyses and functional validation experiments are needed to further strengthen the conclusion.

   Read the original source
3. 

   eLife

   Feb 9, 2024

   Reviewer #1 (Public Review):

   Using a HFD mouse model, the authors examined the H3K4me3 mark in sperm and placental tissues followed by correlation to the transcriptomic changes in the placental tissues of the male and female offspring. The hypothesis that the authors tried to test was that sperm histone epimutations affect placental function, thereby leading to metabolic disorders in offspring. The strength of this work includes the interesting idea and the initial data generated. However, the entire study remains purely correlative without any validation experiment to support the correlation. The conclusion needs to be further supported by bigger sample size and more functional analyses demonstrating the causal relationship among the histone epimutations detected, the dysregulated mRNA expression in the placenta, and the phenotypes in …

   Reviewer #1 (Public Review):

   Using a HFD mouse model, the authors examined the H3K4me3 mark in sperm and placental tissues followed by correlation to the transcriptomic changes in the placental tissues of the male and female offspring. The hypothesis that the authors tried to test was that sperm histone epimutations affect placental function, thereby leading to metabolic disorders in offspring. The strength of this work includes the interesting idea and the initial data generated. However, the entire study remains purely correlative without any validation experiment to support the correlation. The conclusion needs to be further supported by bigger sample size and more functional analyses demonstrating the causal relationship among the histone epimutations detected, the dysregulated mRNA expression in the placenta, and the phenotypes in offspring.

   Read the original source
4. 

   eLife

   Feb 9, 2024

   Reviewer #2 (Public Review):

   This study follows up on previous work from this group, and others, relating paternal diet to changes in sperm epigenetics, and offspring phenotypes. The authors focus on paternal diet (high-fat diet versus a control chow), sperm chromatin, and molecular changes in the placenta associated with offspring development.

   The text is well written and the figures are generally well presented and clear. The sperm epigenetic analyses and analysis of the placenta epigenetics and gene expression are generally well performed. The study provides new insight into how paternally mediated intergenerational epigenetic inheritance could involve placenta-embryo signaling.

   A major weakness is that the high-fat diet used was from a different manufacturer than the control (lower fat) diet. Therefore, it is difficult to judge …

   Reviewer #2 (Public Review):

   This study follows up on previous work from this group, and others, relating paternal diet to changes in sperm epigenetics, and offspring phenotypes. The authors focus on paternal diet (high-fat diet versus a control chow), sperm chromatin, and molecular changes in the placenta associated with offspring development.

   The text is well written and the figures are generally well presented and clear. The sperm epigenetic analyses and analysis of the placenta epigenetics and gene expression are generally well performed. The study provides new insight into how paternally mediated intergenerational epigenetic inheritance could involve placenta-embryo signaling.

   A major weakness is that the high-fat diet used was from a different manufacturer than the control (lower fat) diet. Therefore, it is difficult to judge whether the effects are due to a change in fat levels, or the many other molecules that are likely to differ in chow between different manufacturers. Other weaknesses include lack of methodological detail in parts, low n values for some experiments, and the need for more mechanistic data.

   Whilst the authors may have achieved their aims, more data is needed to inform a potential mechanism.

   This study adds to our understanding of how changes in paternal diet may alter sperm epigenetics and offspring development. The novelty is in the link to gene expression in the placenta associated with offspring development in utero.

   Read the original source
5. 

   eLife

   Feb 9, 2024

   Reviewer #3 (Public Review):

   This study represents a useful addition to the authors' previous study examining the effects of paternal high-fat diet on offspring metabolism and gene expression in offspring (PMID: 35183795). It differs from the previous study in some of the details of the experimental model (age of sire when exposed to the diet manipulation, mouse substrain, and the nature of the control diet) and the results are largely in line with previous findings. The major finding is that many genes at which sperm H3K4me3 signal is altered also have altered expression in the placenta; some of these genes are paternally imprinted, providing a paternal-specific epigenetic signature. Strengths of the study include establishment of an important dataset correlating the sperm epigenome with gene expression in placental tissue, leading to …

   Reviewer #3 (Public Review):

   This study represents a useful addition to the authors' previous study examining the effects of paternal high-fat diet on offspring metabolism and gene expression in offspring (PMID: 35183795). It differs from the previous study in some of the details of the experimental model (age of sire when exposed to the diet manipulation, mouse substrain, and the nature of the control diet) and the results are largely in line with previous findings. The major finding is that many genes at which sperm H3K4me3 signal is altered also have altered expression in the placenta; some of these genes are paternally imprinted, providing a paternal-specific epigenetic signature. Strengths of the study include establishment of an important dataset correlating the sperm epigenome with gene expression in placental tissue, leading to an interesting and provocative conclusion. Weaknesses include a relatively superficial analysis of the dataset, revealing broad patterns but few specific conclusions, reliance on correlative analysis to draw conclusions, and absence of validation studies. Deconvolution analysis of bulk RNA-seq data helps to account for differences in cell composition between placental datasets, but does not add additional insight toward the central question of how sperm epigenetic state contributes to offspring gene expression. Overall the advance over previous work is relatively small.

   Specific points:

   1. The analysis as it stands is limited. To compare sperm H3K4me3 and placental expression, numbers of overlapping genes are provided, but no statistical analysis is done to indicate the significance of the overlap.

   2. There is little direct connection to biological systems or validation of differential enrichment/expression analysis. Gene ontology enrichments for genes differentially enriched for H3K4me3 in sperm or differentially expressed in placenta (broken up by sex) are performed, but the biological significance of these categories is not clear.

   3. The overall effect size is small. In most cases the magnitude of differences is minor, and it is not clear which of these changes are significant over noise. For example, the y-axis for the metagene plots in Figure 2B does not start at zero, so the total range of the difference in H3K4me3 is small. In Figure 6C, DEGs detected in hypoxic placenta after deconvolution analysis do not look very different compared to control.

   4. Deconvolution analysis was done on bulk RNA-seq data from placenta, and the numbers of DEGs identified with this analysis compared to the original analysis are shown, but is not clear how the deconvolution analysis changes the specific biological conclusions. In addition, the reference dataset for deconvolution is a published dataset generated in another lab, and it is unclear how comparable the reference sample is to the samples analyzed in this study, or how robust this analysis is when using a dataset generated under different conditions.

   Read the original source
6. 

   eLife

   Feb 9, 2024

   Reviewer #4 (Public Review):

   The members of the Kimmins lab perform a dietary study in mice to investigate the impact of obesity of fathers on the development of their offspring. To do so, they expose male mice to a high fat diet and determine the distribution and occupancy levels of the histone H3 lysine 4 trimethylation (H3K4me3) mark in spermatozoa and perform gene expression studies on placenta tissue obtained from mouse embryos during mid-gestation development. The authors report changes in H3K4me3 occupancy in sperm as well as in transcriptomes of placentas of male and female embryonic offspring. While the authors perform extensive computational analysis of the transcriptomic and chromatin immunoprecipitation data, the authors do not go much beyond making correlative statements at mainly the genome wide level between changes for …

   Reviewer #4 (Public Review):

   The members of the Kimmins lab perform a dietary study in mice to investigate the impact of obesity of fathers on the development of their offspring. To do so, they expose male mice to a high fat diet and determine the distribution and occupancy levels of the histone H3 lysine 4 trimethylation (H3K4me3) mark in spermatozoa and perform gene expression studies on placenta tissue obtained from mouse embryos during mid-gestation development. The authors report changes in H3K4me3 occupancy in sperm as well as in transcriptomes of placentas of male and female embryonic offspring. While the authors perform extensive computational analysis of the transcriptomic and chromatin immunoprecipitation data, the authors do not go much beyond making correlative statements at mainly the genome wide level between changes for H3K4me3 in sperm and transcriptional changes in placenta, the latter of which are in part related to changes in cellular composition (as deduced from transcriptional data). Given that both parental mice had the same genetic background, it was not possible to deduce parental specific contributions to transcriptional changes as observed in placentas of offspring. In all, the study falls short in increasing mechanistic insights into this important biological phenomenon.

   Read the original source
7. 

   Version published to 10.1101/2022.08.30.503982v1 on bioRxiv

   Aug 31, 2022