Cataloguing the mouse small intestinal transcriptome in duodenum and ileum during the first postnatal month

  1. Luiz Fernando Silva Oliveira
  2. Radhika S Khetani
  3. Nicholas Makogonov
  4. Elizabeth S Partan
  5. Yu-Syuan Wu
  6. Venkata Siva Dasuri
  7. Amanda W Harrington
  8. Oluwabunmi Olaloye
  9. Jeffrey Goldsmith
  10. David T Breault
  11. Liza Konnikova
  12. Shannan J Ho Sui
  13. Amy E O’Connell

  • Curated by eLife

    eLife logo

    eLife Assessment

    This study presents a useful inventory of genes that are up- and down-regulated in the mouse small intestine (duodenum and ileum) during the first postnatal month, along with a comparison to the human intestine and organoids; the data were collected and analyzed using solid and validated methodology and can be used as a starting point for additional validation of specific markers and for follow-up functional studies.

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

In the first postnatal month, the developing mouse intestine shifts from an immature to a mature intestine that will sustain the organism throughout the lifespan. Here, we surveyed the mouse intestine in C57Bl/6 mice by bulk RNA-Seq to evaluate the changes in gene expression over time from the day of birth through 1 month of age in both the duodenum and ileum. Using trends identified in the RNA-Seq analyses, we further evaluated expression of epithelial and mesenchymal cell markers, epithelial regulators, and immune cell markers. We confirmed key changes with qRT-PCR and immunofluorescence. In addition, we compared some findings to humans using human intestinal biopsies and organoids. This dataset can serve as a reference for other groups considering the role of single molecules or molecular families in early intestinal and postnatal development, expanding the limited literature on postnatal gene expression in the developing intestine.

Article activity feed

  1. eLife

    eLife Assessment

    This study presents a useful inventory of genes that are up- and down-regulated in the mouse small intestine (duodenum and ileum) during the first postnatal month, along with a comparison to the human intestine and organoids; the data were collected and analyzed using solid and validated methodology and can be used as a starting point for additional validation of specific markers and for follow-up functional studies.

  2. eLife

    Reviewer #1 (Public review):

    Summary:

    In this manuscript, the authors aimed to clarify the transcriptional changes across murine postnatal small intestinal development (0 days to 1 month) in both the duodenum and ileum, a period that shows morphological similarity to 20-30-week-old fetal humans. This is an especially critical stage in human intestinal development, as necrotizing enterocolitis (NEC) usually manifests during these stages.

    Strengths:

    The authors assessed numerous timepoints between 0 days and 1 month in the postnatal mouse duodenum and ileum using bulk RNA transcriptomics of bulk-isolated tissues. Cellular deconvolution, based on relative marker expression, was used to clarify epithelial, mesenchymal, and immune cell proportions in the bulk RNA sequencing data. They confirmed some transcriptional targets found in vivo primarily in mouse via qrtPCR and immunohistochemistry, but also in human fetal tissues and isolated organoids, and are of decent quality.

    Weaknesses:

    The overall weakness of this study, as mentioned by the authors themselves, is that the bulk transcriptomic data generated for the study was isolated from non-fractionated bulk intestinal tissue. This makes it difficult to interpret much of this data in regards to cellular fractions found across developmental time. It is difficult to rationale the approach here- as even isolation protocols of epithelial-only or mesenchyme-only tissues for bulk RNA sequencing are well established. The authors address some of these concerns using cellular deconvolution, however, further verification of changes in cell populations is very limited. The rationale of using postnatal mouse as a comparison to NEC is also a little unclear- perhaps some of the developmental processes are similar, however, the environments are completely different (microbiome present in mouse postnatal development, for example). In its current form, this work falls short in providing any new insights into either NEC pathogenesis or murine Sl postnatal development.

  3. eLife

    Reviewer #2 (Public review):

    Summary:

    This work presents a valuable resource by generating a comprehensive bulk RNA sequencing catalogue of gene expression in the mouse duodenum and ileum during the first postnatal month. The central findings of this work are based on analysis of this dataset. Specifically, the authors characterized molecular shifts that occur as the intestine matures from an immature to an adult-like state, investigating both temporal changes and regional differences between the proximal and distal small intestine. They also sought to validate key findings through complementary methods and to provide comparative context with human intestinal samples derived from premature infants. This study will provide a solid reference dataset for the community of researchers studying postnatal gastrointestinal development and diseases that arise during these stages. However, the study lacks in functional validation of the interpretations.

    Strengths:

    (1) The inclusion of numerous time points (day 0 through 4 weeks) and comparative analyses throughout the first postnatal month.

    (2) Validation of key interpretations of RNA-seq data by other methods

    (3) Linking mouse postnatal development to human premature infant development, enhancing its clinical relevance. The inclusion of human intestinal biopsy and organoid data for comparison further strengthens this link.

    (4) The investigation covers a wide array of developmental gene categories with known significance, including epithelial differentiation markers (e.g., Vil1, Muc2, Lyz1), intestinal stem cell markers (e.g., Lgr5, Olfm4, Ascl2), mesenchymal markers (e.g., Pdgfra, Vim), Wnt signaling components (e.g., Wnt3, Wnt5a, Ctnnb1), and various immune genes (e.g., defensins, T cell, B cell, ILC, macrophage markers).

    Weaknesses:

    (1) The primary limitation is that there is no functional validation. The study primarily focuses on interpretation of RNA expression. This is a common limitation of transcriptomic "atlas" studies, but the functional and mechanistic relevance of these interpretations remains to be determined.

    (2) The data are derived from bulk RNA-Seq of full-thickness intestinal tissue. While this approach helps capture rare cell types and both epithelial and mesenchymal components simultaneously, it does not provide cell-type specific gene expression profiles, which might obscure important nuances. Future investigations using single-cell sequencing would be a logical follow-up.

    (3) The day 4 samples were omitted due to quality issues, which might have led to missing some dynamic changes, especially given that some ISC genes show dynamic changes around day 6.

  4. eLife

    Reviewer #3 (Public review):

    Summary:

    This study uses bulk mRNA sequencing to profile transcriptional changes in intestinal cells during the early postnatal period in mice-a developmental window that has received relatively little attention despite its importance. This developmental stage is particularly significant because it parallels late gestation in humans, a time when premature infants are highly vulnerable to necrotizing enterocolitis (NEC). By sampling closely spaced timepoints from birth through postnatal week four, the authors generate a resource that helps define transcriptional trajectories during this phase. Although the primary focus is on murine tissue, the authors also present limited data from human fetal intestinal biopsy samples and organoids.

    Strengths:

    The close temporal sampling in mice offers a detailed view of dynamic transcriptional changes across the first four weeks after birth. The authors leverage these close timepoints to perform hierarchical clustering to define relationships between developmental stages. This is a useful approach, as it highlights when transcriptional states shift most dramatically and allows for functional predictions about classes of genes that vary over time. This high-level analysis provides an effective entry point into the dataset and will be useful for future investigations. The inclusion of human fetal intestinal samples, although limited, is especially notable given the scarcity of data from late fetal timepoints. The authors are generally careful in their presentation of results, acknowledging the limitations of their approach and avoiding over-interpretation. As they note, this dataset is intended as a foundation for their lab and others, with secondary approaches required to more fully explore the biological questions raised. The revised manuscript also addresses many of the issues of clarity and presentation raised in the original review.

    Weaknesses:

    A limitation of the study is the use of bulk mRNA sequencing on unsorted cell populations to draw conclusions about individual cell types. This is particularly challenging for intestinal epithelial cells, where few genes are exclusively expressed in single cell types. For instance, markers such as Lgr5 and Olfm4 are enriched in intestinal stem cells (ISCs), but are also expressed by other absorptive and secretory lineages. Although the authors incorporate marker-based deconvolution analyses, the epithelial signatures used here are not among the cell populations originally validated for mMCPcounter. Conclusions that specific epithelial populations increase or decrease across timepoints would benefit from more robust systematic, complementary validation.

    Validation of the sequencing data in this study remains limited. The authors have added imaging/IF-based validation of mesenchymal populations; however, validation otherwise relies primarily on qPCR, which measures expression using a similar modality rather than providing orthogonal support. Many of the epithelial cell type-specific dynamics inferred from the bulk sequencing and deconvolution analyses are therefore not independently validated. Moreover, the regional differences reported in Lgr5, Olfm4, and Ascl2-appearing much higher in proximal than distal samples-were not recapitulated by qPCR validation of Olfm4. Additional orthogonal validation of these cell type-specific dynamics would be valuable in future studies.

    Finally, the human fetal biopsy samples represent a potentially valuable component of the study given the scarcity of data from these developmental timepoints, but the conclusions that can currently be drawn from these samples are limited. The central figure presenting these data shows immunofluorescence for LGR5, a single stem cell marker, and the staining at day 35 is difficult to interpret. The data support localization of LGR5-positive cells to crypts as early as 26 weeks, but more extensive analysis would be needed to draw broader conclusions about human intestinal development. Thus, while inclusion of these samples adds value to the study, this component remains relatively limited in scope.

  5. eLife

    Author response:

    The following is the authors’ response to the original reviews.

    Public Reviews:

    Reviewer #1 (Public review):

    Summary:

    In this manuscript, the authors aimed to clarify the transcriptional changes across murine postnatal small intestinal development (0 days to 1 month) in both the duodenum and ileum, a period that shows morphological similarity to 20-30 week old fetal humans. This is an especially critical stage in human intestinal development, as necrotizing enterocolitis (NEC) usually manifests during these stages.

    Strengths:

    The authors assessed numerous timepoints between 0 days and 1 month in the postnatal mouse duodenum and ileum using bulk RNA transcriptomics of bulk-isolated tissues. Cellular deconvolution, based on relative marker expression, was used to clarify immune cell proportions in the bulk RNA sequencing data. They confirmed some transcriptional targets found in vivo primarily in mouse via qrtPCR and immunohistochemistry, but also in human fetal tissues and isolated organoids, and are of decent quality.

    Weaknesses:

    The overall weakness of this study, as mentioned by the authors themselves, is that the bulk transcriptomic data generated for the study were isolated from non-fractionated bulk intestinal tissue. This makes it difficult to interpret much of this data regarding cellular fractions found across developmental time. It is difficult to rationalize the approach here, as even isolation protocols of epithelial-only or mesenchyme-only tissues for bulk RNA sequencing are well established. The authors address some of these concerns using cellular deconvolution for immune cell populations, which I think might be helpful if they expanded this analysis to other cell types (mesenchyme, endothelium, glia). However, I would assume that bulk isolations across developmental time are going to be influenced primarily by the bulk of tissue-type found at each time point - primarily epithelium. But this is also confirmed by the immune transcripts becoming more apparent later in their time series, as this system becomes more established during weaning. This study might also be strengthened by comparison with data that is publicly available for early fetal stage development in humans. Comparisons between the duodenum and ileum could be strengthened by what we already know from adult data, from both epithelial- and mesenchyme-isolated fractions. The rationale of using the postnatal mouse as a comparison to NEC is also a little unclear- perhaps some of the developmental processes are similar, however, the environments are completely different. For example, even in early postnatal mouse development, you would find microbial activity and milk.

    Reviewer #2 (Public review):

    Summary:

    This work presents a valuable resource by generating a comprehensive bulk RNA sequencing catalogue of gene expression in the mouse duodenum and ileum during the first postnatal month. The central findings of this work are based on an analysis of this dataset. Specifically, the authors characterized molecular shifts that occur as the intestine matures from an immature to an adult-like state, investigating both temporal changes and regional differences between the proximal and distal small intestine. A key objective was to identify gene expression patterns relevant to understanding the region-specific susceptibility and resistance to necrotizing enterocolitis (NEC) observed in humans during the postnatal period. They also sought to validate key findings through complementary methods and to provide comparative context with human intestinal samples. This study will provide a solid reference dataset for the community of researchers studying postnatal gastrointestinal development and diseases that arise during these stages. However, the study lacks functional validation of the interpretations.

    Strengths:

    (1) The inclusion of numerous time points (day 0 through 4 weeks) and comparative analyses throughout the first postnatal month.

    (2) Validation of key interpretations of RNA-seq data by other methods.

    (3) Linking mouse postnatal development to human premature infant development, enhancing its clinical relevance, particularly for NEC research. The inclusion of human intestinal biopsy and organoid data for comparison further strengthens this link.

    (4) The investigation covers a wide array of developmental gene categories with known significance, including epithelial differentiation markers (e.g., Vil1, Muc2, Lyz1), intestinal stem cell markers (e.g., Lgr5, Olfm4, Ascl2), mesenchymal markers (e.g., Pdgfra, Vim), Wnt signaling components (e.g., Wnt3, Wnt5a, Ctnnb1), and various immune genes (e.g., defensins, T cell, B cell, ILC, macrophage markers).

    Weaknesses:

    (1) The primary limitation is that there is no functional validation. The study primarily focuses on the interpretation of RNA expression. This is a common limitation of transcriptomic "atlas" studies, but the functional and mechanistic relevance of these interpretations remains to be determined.

    (2) The data are derived from bulk RNA-Seq of full-thickness intestinal tissue. While this approach helps capture rare cell types and both epithelial and mesenchymal components simultaneously, it does not provide cell-type-specific gene expression profiles, which might obscure important nuances. Future investigations using single-cell sequencing would be a logical follow-up.

    (3) The day 4 samples were omitted due to quality issues, which might have led to missing some dynamic changes, especially given that some ISC genes show dynamic changes around day 6.

    Reviewer #3 (Public review):

    Summary:

    This study uses bulk mRNA sequencing to profile transcriptional changes in intestinal cells during the early postnatal period in mice - a developmental window that has received relatively little attention despite its importance. This developmental stage is particularly significant because it parallels late gestation in humans, a time when premature infants are highly vulnerable to necrotizing enterocolitis (NEC). By sampling closely spaced timepoints from birth through postnatal week four, the authors generate a resource that helps define transcriptional trajectories during this phase. Although the primary focus is on murine tissue, the authors also present limited data from human fetal intestinal biopsy samples and organoids. In addition, they discuss potential links between observed gene expression changes and factors that may contribute to NEC.

    Strengths:

    The close temporal sampling in mice offers a detailed view of dynamic transcriptional changes across the first four weeks after birth. The authors leverage these close timepoints to perform hierarchical clustering to define relationships between developmental stages. This is a useful approach, as it highlights when transcriptional states shift most dramatically and allows for functional predictions about classes of genes that vary over time. This high-level analysis provides an effective entry point into the dataset and will be useful for future investigations. The inclusion of human fetal intestinal samples, although limited, is especially notable given the scarcity of data from late fetal timepoints. The authors are generally careful in their presentation of results, acknowledging the limitations of their approach and avoiding over-interpretation. As they note, this dataset is intended as a foundation for their lab and others, with secondary approaches required to more fully explore the biological questions raised.

    Weaknesses:

    One limitation of the study is the use of bulk mRNA sequencing to draw conclusions about individual cell types. It has been documented that a few genes are exclusively expressed in single cell types. For instance, markers such as Lgr5 and Olfm4 are enriched in intestinal stem cells (ISCs), but they are also expressed at lower levels in other lineages and in differentiating cells. Using these markers as proxies for specific cell populations lowers confidence in the conclusions, particularly without complementary validation to confirm cell type-specific dynamics.

    Validation of the sequencing data was itself limited, relying primarily on qPCR, which measures expression at the same modality rather than providing orthogonal support. It is unclear how the authors selected the subset of genes for validation; many key genes highlighted in the sequencing data were not assessed. Moreover, the regional differences reported in Lgr5, Olfm4, and Ascl2, appearing much higher in proximal samples than in distal ones, were not recapitulated by qPCR validation of Olfm4, and this discrepancy was not addressed. Resolving such inconsistencies will be important for interpreting the dataset.

    The basis for linking particular gene sets to NEC susceptibility rests largely on their spatial restriction to the distal intestine and their temporal regulation between early (day 0-14) and later (weeks 3-4) developmental stages. While this is a reasonable approach for generating hypotheses, the correlations have limited interpretive power without experimental validation, which is not provided here. Many factors beyond NEC may drive regional and temporal differences in intestinal development.

    Finally, the contribution of human fetal biopsy samples is minimal. The central figure presenting these data (Figure 4A) shows immunofluorescence for LGR5, a single stem cell marker. The staining at day 35 is not convincing, and the conclusions that can be drawn are limited to confirming the localization of LGR5-positive cells to crypts as early as 26 weeks.

    We have added additional figures and significantly edited the manuscript based on the insightful suggestions of the reviewers. We added epithelial and mesenchymal deconvolutions as well as additional protein-based analyses using IF. We are very pleased with the improvements since then initial submission, and hope the reviewers will feel similarly.

    Recommendations for the authors:

    Reviewing Editor Comments:

    Two of the reviewers mentioned that the correlations with NEC may be speculative and/or hypothesis-generating. For example, as noted by Reviewer 3, "While this is a reasonable approach for generating hypotheses, the correlations have limited interpretive power without experimental validation, which is not provided here." Based on this, I would recommend that the authors remove the correlations between NEC susceptibility and gene expression changes from the results section and include this only in the discussion section. All reviewers felt that some additional validation of major findings from bulk RNA-seq is warranted. Reviewer 1 suggested additional comparisons with single-cell data or isolated populations; however, as noted by other reviewers, some additional validation using protein markers (i.e., immunofluorescence) would suffice to bolster findings and claims. Reviewers 2 and 3 provide several specific, detailed suggestions below that will improve the clarity of the manuscript through text revisions.

    Thank you to the editor and reviewers for critically evaluating the manuscript and suggesting these changes to strengthen the work.

    We agree that linking the study to NEC was overly theoretical, and have revised the focus of the paper to focus on early postnatal development, removing the speculative links to NEC susceptibility.

    We have added additional antibody-based immunofluorescence data to consider cognate protein expression in Figure 10. We expanded the evaluation of the mesenchymal cell populations, which is particularly poorly characterized in this period.

    Reviewer #1 (Recommendations for the authors):

    My primary concern is for the authors to adequately deconvolute their bulk RNA datasets and to perform additional comparison analyses with either bulk or single-cell isolated epithelium and/or mesenchyme from either adult mice (for regionality) or human fetal data.

    We thank the reviewer for this suggestion. We performed an epithelial cell deconvolution and a mesenchymal cell deconvolution using published datasets and add this information as figure 4. We were unable to perform the mesenchymal deconvolution at the time of submission, because this compartment is understudied, however a recent preprint published mouse single-cell data that we were able to use for this analysis.

    Reviewer #2 (Recommendations for the authors):

    I think that this study will serve as a solid reference for those interested in postnatal intestinal development, and currently, there is a key lack of studies during these stages. I'd strongly suggest making some of the figures more "readable". For example, the font size along many axes and figure legends is very challenging to read when printed out. There are also some minor errors that I caught, but the authors should probably look through more thoroughly:

    We appreciate the reviewer’s enthusiasm for the study. We have attempted to make the figure text more legible.

    (1) In the Results section, the text states "stained tissue sections for CD19 (Fig. 5, column 4)". This should be Fig. 8.

    Corrected, thank you.

    (2) In the abbreviations section, MHCII is misspelled.

    Corrected, thank you.

    (3) The abstract has an extra period after ileum.

    Corrected, thank you.

    Reviewer #3 (Recommendations for the authors):

    (1) Some elements lack precise language that could lead to misconceptions. For example, I disagree with the use of the term duodenum to designate samples taken from the entire proximal half of the small intestine. The duodenum is well established to represent only a small portion of the proximal-most intestine.

    I think this may be a misunderstanding? The RNA samples were taken from the first 2cm of small bowel and the ileal samples form the last 2cm, so for any RNA work these labels are appropriate. For the histology, we prepped the samples initially by bisecting the SI, however the images presented are only from the duodenum or ileum, which we could determine by standardizing the Swiss roll orientation during the prep as we describe in the methods. We have added text to clarify this to the reader in the methods: “We analyzed only the segments that corresponded to the duodenum and ileum for the histology, in order to maintain consistency with the RNA data.”

    (2) At the start of the Results section, it would be helpful to emphasize that the initial data presented are from murine tissue rather than human (this is stated in the introduction, but it is easy to lose track since both mouse and human tissue are used throughout). A graphic overview of all samples obtained would also aid clarity. The species should be clearly noted in each figure legend to avoid confusion.

    Thank you for these great suggestions. We added language in the results section, added a logo designating mouse or human to each figure, and also added a graphical overview of the samples (Fig. --).

    (3) The statement "PCA of all samples showed that duodenal samples and ileal samples intermixed and did not segregate by tissue type in early development but segregated more at later time points (Supp. Fig. 1C)" is not clearly supported by the data as presented; this panel requires timepoint annotations.

    Thank you for pointing this out. We have added this data as a new panel in this figure, 1D.

    (4) The heatmaps in Figure 1C-D could be moved to the supplement, with main-figure heatmaps showing the subset of samples actually used for downstream analyses (e.g., with days 4 and 6 removed).

    We have removed Fig. 1 C-D to the supplement, thank you for the suggestion. Fig. 1 E-F now become C-D and contain the heatmap of the samples actually used.

    (5) Two batches are labeled in Figures 1C-D but are not mentioned in the text. Batches should be described in the Methods and noted in figures. Please clarify how batch effects in the sequencing data were assessed.

    Thank you for asking us to clarify this for readers. We have added a section to the methods to explain the need for the batches (too many samples for one run on the sequencer we were using at the time) and how they were handled in the data analysis.

    (6) While the rationale for focusing on specific timepoints and regions potentially relevant to NEC susceptibility is appreciated, the interpretation that these transcriptional changes directly relate to NEC pathology is highly speculative. In my view, these points could be minimized in the Results and moved to the Discussion.

    We agree that linking the study to NEC was overly theoretical, and have revised the focus of the paper to focus on early postnatal development, removing the speculative links to NEC susceptibility.

    (7) Figure 3A: Could EpCAM be included as a canonical epithelial marker?

    We have included EpCAM in the epithelial cell markers (Fig. 3A).

    (8) The use of the word 'expansion' in the phrase "suggesting that epithelial expansion is represented by cluster 2" is unclear and should be clarified or better supported.

    Thank you, this word was imprecise. We have revised the text to: “suggesting that epithelial villus growth, potentially in the form of increased length or increased density, is represented by cluster 2.”

    (9) Please consider incorporating the newly identified intestinal stem cell marker Fgfbp1 (Capdevila et al. and Malagola et al., Cell 2024) in the ISC marker panel.

    We have now included Fgfbp1 (removed Ascl2 to make space given it was redundant with Olfm4 and Lgr5), however we note that the paper referenced by the reviewer suggests that Fgfbp1 marks a distinct population of stem cells, rather than redundantly marking the Lgr5+ISC we previously sought to highlight in this panel.

    (10) The conclusions that can be drawn from the human organoid data in Figure 4B are unclear. In the Methods, please note when cells were collected (i.e., days post-passage) and whether this was consistent across samples.

    Thank you. The samples were consistently grown to confluence, collected 5 days after the most recent passage to allow the organoids to recover. We have added this to the details on this experiment’s methodology.

    (11) In discussing regional Wnt regulation, consider citing and relating findings to Maimets et al., Nat Commun 2022. Other relevant literature on zonation could also be included in the discussion.

    Thank you for this suggestion. We pulled some of the genes indicated in this Maimets study to associate with zonation and discussed them in the context of the Wnt figure (Fig. 5D) and added some discussion on this.

    (12) More precise language and references would strengthen descriptions of analysis approaches. For example: "We sought to confirm some of these findings using the newer RNAseq immune cell deconvolution approach that combines key genes..." - is this a published method that can be explicitly cited?

    Thank you for noting this. We have changed the language introducing this section and added references for deconvolution.

    (13) The figure callout in "the proportion of CD19+ increased over time as a percentage of the total PP lesions (Supplementary Figure 3C), indicating enlargement of the germinal centers of time" appears incorrect. The conclusion that this reflects germinal center enlargement is not, in my view, supported by the data.

    We acknowledge the reviewer’s point that relative proportionality of the PP may not be strictly a function of size, and have removed this conclusion statement accordingly.

    (14) Phrases such as "the data was very messy and hard to interpret" could be replaced with more precise language appropriate to a scientific report (though I sympathize!).

    Thank you for pointing out this overly informal language, we have revised.

    (15) Ensure consistent lettering conventions across figure legends.

    Corrected for consistency, thank you.

    (15) In Figure 2, larger axis text and an indication of the developmental timeline would improve clarity. The meaning of the x-axis label "GeneRatio" should also be clarified.

    Given the amount of text in this figure, it is impossible to enlarge the text more, unfortunately. We wanted to show more the family names and cluster patterns. Persons uses an e-version (majority of readers today) can easily zoom in to see more detail if desired. We added text to the figure legend “Gene ratio on the y-axis indicates the proportion of your significantly altered genes that are involved in a specific biological pathway or function.”

    (16) In Figure 3 (and other similar mRNAseq/validation panels), maintaining a consistent color code between mRNAseq and qPCR validation would aid interpretation. In panel F, the key's colors should align with the graph.

    We adjusted the qPCR figures to correspond with the RNASeq.

    (17) In Figure 4, please specify which intestinal region the samples are from, as this is especially relevant given the regional differences observed in the murine data.

    Thank you, these are from ileum, we have designated this more clearly in the figure.

  6. Version published to 10.7554/elife.108645.2 on eLife
  7. Version published to 10.7554/elife.108645 on eLife
  8. Version published to 10.1101/2024.09.25.612672v3 on bioRxiv
  9. eLife

    eLife Assessment

    This study presents a useful inventory of genes that are up- and down-regulated in the mouse small intestine (duodenum and ileum) during the first postnatal month; the data were collected and analyzed using solid and validated methodology and can be used as a starting point for additional validation of specific markers and for follow-up functional studies. Some aspects of the study were incomplete, with claims being only partially supported by the data, and it is suggested that additional validation be performed. The authors attempted to correlate gene expression changes with periods of high and low NEC susceptibility, but these correlations are speculative and not supported by functional follow-up studies. Discussion of gene expression changes with NEC susceptibility would be more appropriate to include in the Discussion section and to be tempered in the results section.

  10. eLife

    Reviewer #1 (Public review):

    Summary:

    In this manuscript, the authors aimed to clarify the transcriptional changes across murine postnatal small intestinal development (0 days to 1 month) in both the duodenum and ileum, a period that shows morphological similarity to 20-30 week old fetal humans. This is an especially critical stage in human intestinal development, as necrotizing enterocolitis (NEC) usually manifests during these stages.

    Strengths:

    The authors assessed numerous timepoints between 0 days and 1 month in the postnatal mouse duodenum and ileum using bulk RNA transcriptomics of bulk-isolated tissues. Cellular deconvolution, based on relative marker expression, was used to clarify immune cell proportions in the bulk RNA sequencing data. They confirmed some transcriptional targets found in vivo primarily in mouse via qrtPCR and immunohistochemistry, but also in human fetal tissues and isolated organoids, and are of decent quality.

    Weaknesses:

    The overall weakness of this study, as mentioned by the authors themselves, is that the bulk transcriptomic data generated for the study were isolated from non-fractionated bulk intestinal tissue. This makes it difficult to interpret much of this data regarding cellular fractions found across developmental time. It is difficult to rationalize the approach here, as even isolation protocols of epithelial-only or mesenchyme-only tissues for bulk RNA sequencing are well established. The authors address some of these concerns using cellular deconvolution for immune cell populations, which I think might be helpful if they expanded this analysis to other cell types (mesenchyme, endothelium, glia). However, I would assume that bulk isolations across developmental time are going to be influenced primarily by the bulk of tissue-type found at each time point - primarily epithelium. But this is also confirmed by the immune transcripts becoming more apparent later in their time series, as this system becomes more established during weaning. This study might also be strengthened by comparison with data that is publicly available for early fetal stage development in humans. Comparisons between the duodenum and ileum could be strengthened by what we already know from adult data, from both epithelial- and mesenchyme-isolated fractions. The rationale of using the postnatal mouse as a comparison to NEC is also a little unclear- perhaps some of the developmental processes are similar, however, the environments are completely different. For example, even in early postnatal mouse development, you would find microbial activity and milk.

  11. eLife

    Reviewer #2 (Public review):

    Summary:

    This work presents a valuable resource by generating a comprehensive bulk RNA sequencing catalogue of gene expression in the mouse duodenum and ileum during the first postnatal month. The central findings of this work are based on an analysis of this dataset. Specifically, the authors characterized molecular shifts that occur as the intestine matures from an immature to an adult-like state, investigating both temporal changes and regional differences between the proximal and distal small intestine. A key objective was to identify gene expression patterns relevant to understanding the region-specific susceptibility and resistance to necrotizing enterocolitis (NEC) observed in humans during the postnatal period. They also sought to validate key findings through complementary methods and to provide comparative context with human intestinal samples. This study will provide a solid reference dataset for the community of researchers studying postnatal gastrointestinal development and diseases that arise during these stages. However, the study lacks functional validation of the interpretations.

    Strengths:

    (1) The inclusion of numerous time points (day 0 through 4 weeks) and comparative analyses throughout the first postnatal month.

    (2) Validation of key interpretations of RNA-seq data by other methods.

    (3) Linking mouse postnatal development to human premature infant development, enhancing its clinical relevance, particularly for NEC research. The inclusion of human intestinal biopsy and organoid data for comparison further strengthens this link.

    (4) The investigation covers a wide array of developmental gene categories with known significance, including epithelial differentiation markers (e.g., Vil1, Muc2, Lyz1), intestinal stem cell markers (e.g., Lgr5, Olfm4, Ascl2), mesenchymal markers (e.g., Pdgfra, Vim), Wnt signaling components (e.g., Wnt3, Wnt5a, Ctnnb1), and various immune genes (e.g., defensins, T cell, B cell, ILC, macrophage markers).

    Weaknesses:

    (1) The primary limitation is that there is no functional validation. The study primarily focuses on the interpretation of RNA expression. This is a common limitation of transcriptomic "atlas" studies, but the functional and mechanistic relevance of these interpretations remains to be determined.

    (2) The data are derived from bulk RNA-Seq of full-thickness intestinal tissue. While this approach helps capture rare cell types and both epithelial and mesenchymal components simultaneously, it does not provide cell-type-specific gene expression profiles, which might obscure important nuances. Future investigations using single-cell sequencing would be a logical follow-up.

    (3) The day 4 samples were omitted due to quality issues, which might have led to missing some dynamic changes, especially given that some ISC genes show dynamic changes around day 6.

  12. eLife

    Reviewer #3 (Public review):

    Summary:

    This study uses bulk mRNA sequencing to profile transcriptional changes in intestinal cells during the early postnatal period in mice - a developmental window that has received relatively little attention despite its importance. This developmental stage is particularly significant because it parallels late gestation in humans, a time when premature infants are highly vulnerable to necrotizing enterocolitis (NEC). By sampling closely spaced timepoints from birth through postnatal week four, the authors generate a resource that helps define transcriptional trajectories during this phase. Although the primary focus is on murine tissue, the authors also present limited data from human fetal intestinal biopsy samples and organoids. In addition, they discuss potential links between observed gene expression changes and factors that may contribute to NEC.

    Strengths:

    The close temporal sampling in mice offers a detailed view of dynamic transcriptional changes across the first four weeks after birth. The authors leverage these close timepoints to perform hierarchical clustering to define relationships between developmental stages. This is a useful approach, as it highlights when transcriptional states shift most dramatically and allows for functional predictions about classes of genes that vary over time. This high-level analysis provides an effective entry point into the dataset and will be useful for future investigations. The inclusion of human fetal intestinal samples, although limited, is especially notable given the scarcity of data from late fetal timepoints. The authors are generally careful in their presentation of results, acknowledging the limitations of their approach and avoiding over-interpretation. As they note, this dataset is intended as a foundation for their lab and others, with secondary approaches required to more fully explore the biological questions raised.

    Weaknesses:

    One limitation of the study is the use of bulk mRNA sequencing to draw conclusions about individual cell types. It has been documented that a few genes are exclusively expressed in single cell types. For instance, markers such as Lgr5 and Olfm4 are enriched in intestinal stem cells (ISCs), but they are also expressed at lower levels in other lineages and in differentiating cells. Using these markers as proxies for specific cell populations lowers confidence in the conclusions, particularly without complementary validation to confirm cell type-specific dynamics.

    Validation of the sequencing data was itself limited, relying primarily on qPCR, which measures expression at the same modality rather than providing orthogonal support. It is unclear how the authors selected the subset of genes for validation; many key genes highlighted in the sequencing data were not assessed. Moreover, the regional differences reported in Lgr5, Olfm4, and Ascl2, appearing much higher in proximal samples than in distal ones, were not recapitulated by qPCR validation of Olfm4, and this discrepancy was not addressed. Resolving such inconsistencies will be important for interpreting the dataset.

    The basis for linking particular gene sets to NEC susceptibility rests largely on their spatial restriction to the distal intestine and their temporal regulation between early (day 0-14) and later (weeks 3-4) developmental stages. While this is a reasonable approach for generating hypotheses, the correlations have limited interpretive power without experimental validation, which is not provided here. Many factors beyond NEC may drive regional and temporal differences in intestinal development.

    Finally, the contribution of human fetal biopsy samples is minimal. The central figure presenting these data (Figure 4A) shows immunofluorescence for LGR5, a single stem cell marker. The staining at day 35 is not convincing, and the conclusions that can be drawn are limited to confirming the localization of LGR5-positive cells to crypts as early as 26 weeks.

  13. Version published to 10.7554/elife.108645.1 on eLife
  14. Mentioned by @aeo-mdphd.bsky.social on Bluesky

    Also posting our last preprint, which we recently updated with searchable data tables. If you're interested in what your favorite gene is doing in the small intestine from P0-P28, now you can find out! #intestine #immunity #SciSky #BlueSci #RNASeq #RNAFind! www.biorxiv.org/content/10.1...

  15. Version published to 10.1101/2024.09.25.612672v2 on bioRxiv
  16. Mentioned by @biorxiv-devbio.bsky.social on Bluesky

    Cataloguing the postnatal small intestinal transcriptome during the period of susceptibility to necrotizing enterocolitis https://www.biorxiv.org/content/10.1101/2024.09.25.612672v1

  17. Mentioned by @biorxivpreprint.bsky.social on Bluesky

    Cataloguing the postnatal small intestinal transcriptome during the period of susceptibility to necrotizing enterocolitis https://www.biorxiv.org/content/10.1101/2024.09.25.612672v1

  18. Version published to 10.1101/2024.09.25.612672v1 on bioRxiv