Histone methyltransferase DOT1L differentially affects the development of dendritic cell subsets

  1. Rianne G. Bouma
  2. Willem-Jan de Leeuw
  3. Aru Z. Wang
  4. Muddassir Malik
  5. Joeke G.C. Stolwijk
  6. Veronique A.L. Konijn
  7. Anne Mensink
  8. Natalie Proost
  9. Maarten K. Nijen Twilhaar
  10. Tibor van Welsem
  11. Negisa Seyed Toutounchi
  12. Alsya J. Affandi
  13. Jip T. van Dinter
  14. Fred van Leeuwen
  15. Joke M.M. den Haan

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

Dendritic cells (DCs) are important orchestrators of immune responses. Their development in the bone marrow is controlled by transcription factors, but epigenetic mechanisms remain poorly understood. DOT1L is emerging as a key epigenetic regulator in immune cells. By mapping DOT1L-mediated histone H3K79 methylation in canonical DC subsets, we observed that DOT1L modified common as well as DC subset-specific genes. In vitro- or in vivo- induced deletion of Dot1l followed by in vitro cell culture resulted in a decrease in myeloid progenitors and plasmacytoid DCs (pDCs) and an increase in cDC2s, while cDC1s remained unchanged. In vitro generated Dot1l- KO DCs were unable to produce IFNα upon stimulation. Moreover, transcriptomes of Dot1l -KO DC subsets exhibited enrichment of antigen presentation pathways and MHC class II surface levels were upregulated in pDCs. Mechanistically, inhibition of DOT1L linked the observed effects to its methyltransferase activity. Together, our data indicate that in DCs DOT1L differentially affects the development of canonical subsets and suppresses antigen presentation pathways.

Article activity feed

  1. Review Commons

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

    Learn more at Review Commons

    Reply to the reviewers

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

  2. Review Commons

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

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    This manuscript investigates the role of DOT1L and its H3K79 methyltransferase activity in dendritic cell (DC) differentiation. The authors employ a combination of in vitro FLT3L/SCF bone marrow culture systems, in vivo inducible knockout models, and genome-wide H3K79me2 ChIP-seq and RNA-seq analyses to demonstrate that DOT1L influences the balance between pDC and cDC2 differentiation, while leaving cDC1 development largely unaffected. The study further identifies transcriptional and epigenetic programs associated with these changes, linking DOT1L deficiency to altered antigen presentation pathways and loss of pDC-associated transcription factors. The paper provides valuable insights into DC biology. However, some of the key conclusions rely heavily on in vitro systems and short-term tamoxifen deletion models, which limit the interpretation of the in vivo data. Strengthening or clearly defining these limitations would substantially improve the paper's impact and clarity.

    Major Comments

    1. To strengthen the paper, the authors could follow one of two alternative strategies:

    (1) Validate their in vitro observations through in vivo experiments, or

    (2) Focus on deepening and refining their in vitro findings, moving the limited in vivo data to the supplementary material and explicitly acknowledging the limitations of the tamoxifen-inducible system.

    Strategy 1 - Strengthen in vivo validation

    -   The experiments presented in Figures 3 and 5 could be repeated in a competitive bone marrow chimera setting (e.g. CD45.1/CD45.2 irradiated hosts reconstituted with a 1:1 mix of WT CD45.1⁺ and Dot1l-KO CD45.2⁺ cells).
-   This design would allow dissection of direct (cell-intrinsic) versus indirect effects of DOT1L deficiency and could mitigate confounding effects of incomplete or asynchronous deletion.
-   After reconstitution, mice could be maintained on tamoxifen-supplemented chow for a longer period to ensure efficient recombination and adequate time for observing phenotypic consequences.
-   Flow cytometric analysis of spleen and bone marrow should use more refined panels to explore DC precursor and subset deficiencies. Suggested reference panels: Rodrigues et al., Immunity 2024; Minutti et al., Nat. Immunol. 2024; Zhu et al., Nat. Immunol. 2015.

    Strategy 2 - Refine in vitro system and reposition in vivo data - The authors could replicate their differentiation assays under conditions that emulate the chimera approach by co-culturing WT (CD45.1⁺) and Dot1l-KO (CD45.2⁺) bone marrow cells. - This would reveal potential competition or cross-talk between WT and mutant cells and provide clearer mechanistic insight into cell-intrinsic versus extrinsic effects. - The authors should examine how tamoxifen itself affects differentiation and measure the kinetics of deletion and H3K79me loss to better contextualize the dynamic response. - It would also be valuable to assess which cDC2 subtypes (A vs. B) are preferentially affected by Dot1l deficiency, again using more sophisticated flow cytometry panels (see references above). If this in vitro-focused strategy is adopted, the in vivo data could be moved to the supplementary material, with explicit acknowledgment that the inducible deletion model and the gradual nature of H3K79me dilution limit the interpretation of the in vivo findings.

    1. In Figures 2 and 3, the efficiency of H3K79me2 depletion following Dot1l excision should be assessed directly. Although DOT1L is the sole H3K79 methyltransferase, the dilution kinetics of H3K79me2 can vary depending on the proliferation rate. Quantifying the H3K79me2 signal in bone marrow-derived cell culture samples would clarify whether the deletion window allowed complete loss of the methylation mark.
    2. Several observations are not discussed in sufficient depth:
      • The finding that Dot1l deletion increases antigen-presentation signatures might reflect stress or activation rather than lineage fate change.
      • The authors could also acknowledge that DOT1L's effect might be indirect, acting through cytokine feedback loops or altered progenitor proliferation, especially given the co-expression of Kit, Flt3, and Irf8 in early DC progenitors.
      • Moreover, because H3K79 methylation is primarily associated with transcriptional elongation rather than initiation, the observed transcriptional changes could result from broader alterations in chromatin accessibility or polymerase processivity, rather than direct promoter regulation. Discussing this mechanistic aspect would help clarify whether DOT1L's role in DC differentiation reflects a direct control of lineage-defining gene expression or a secondary consequence of disrupted transcriptional elongation dynamics.

    Minor Comments

    1. Terminology: The manuscript repeatedly refers to "mature" DCs-please clarify whether this means activated or fully differentiated cells.
    2. Ontogeny statements:
      The assertion that DCs of lymphoid origin are well established should be softened; the lymphoid contribution to some DC lineages remains under discussion.
    3. Transitional DCs (tDCs):
      The equivalence between tDCs and pre-cDC2As remains controversial. This should be acknowledged.
    4. Cytokine supplementation:
      The inclusion of SCF in the FLT3L-based differentiation assays should be justified, it is not a standard procedure.
    5. Macrophage contamination:
      The presence of C1qa, C1qb, and C1qc transcripts in some datasets suggests possible macrophage contamination. Please discuss how this was controlled for or how it might affect interpretation.

    Significance

    This study provides important insights into the epigenetic regulation of DC differentiation by DOT1L. The conclusions would be more compelling if supported by in vivo validation or, alternatively, if the limitations of the current in vivo data were transparently acknowledged and the focus shifted toward mechanistic in vitro depth.

    With these revisions, the manuscript would represent a valuable contribution to understanding how chromatin modification integrates with transcriptional control in shaping dendritic cell fate.

  3. Review Commons

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

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    Bouma et al. present a comprehensive analysis of DOT1L-mediated histone H3K79 methylation across canonical DC subsets. By mapping the methylation landscape, the authors demonstrate that DOT1L regulates both shared and subset-specific gene programs. They show that in vitro or in vivo deletion of Dot1l, followed by in vitro differentiation, results in reduced myeloid progenitors and pDCs alongside an increase in cDC2s, while cDC1 numbers remain largely unaffected. Functionally, Dot1l-deficient DCs fail to produce IFNα upon stimulation. Transcriptomic profiling reveals enrichment of antigen presentation pathways in Dot1l-KO subsets, with upregulated MHC class II surface expression in pDCs. Mechanistically, pharmacological inhibition of DOT1L links these effects to its methyltransferase activity. Collectively, the data suggest that DOT1L differentially regulates canonical DC subset development and represses antigen presentation pathways.

    The manuscript is well-written and technically sound. However, several conclusions would benefit from deeper discussion or additional experimental validation.

    Major Comments

    1. Interpretation of DC balance changes and cell-cycle effects

    The authors propose that DOT1L loss skews DC differentiation toward a pDC-like phenotype. However, DOT1L deletion or inhibition, and the consequent global loss of H3K79 methylation, is well known to downregulate key cell-cycle genes (e.g., Cyclin D1, Cyclin E, CDK4/6, MCM family) while upregulating cell-cycle inhibitors (e.g., Cdkn1a and b). These transcriptional changes are associated with slower proliferation, G1 arrest or delayed S-phase entry, and reduced DNA replication fork progression. Importantly, blocking DNA synthesis (e.g., with aphidicolin or mitomycin C) during early culture inhibits DC emergence, underscoring that proliferation is essential for differentiation. The authors should discuss how their findings align with this established literature. Could the observed DC subset shifts result from impaired cell-cycle progression rather than lineage-specific transcriptional reprogramming? A more detailed consideration of this point is needed.

    1. Discrepancy between in vitro and in vivo pDC phenotypes

    The in vitro data show a marked reduction in pDCs, yet in vivo pDC numbers appear unchanged. Although the discussion briefly mentions proliferation differences, this discrepancy deserves a clearer explanation or experimental follow-up.

    Minor Comments

    • Clarify statistical methods, specify biological replicate numbers, and indicate whether corrections for multiple comparisons were applied to transcriptomic analyses.
    • The introduction is somewhat lengthy and repetitive; condensing it would improve focus.
    • In the discussion sometimes it is not clear the distinction between findings and speculation.
    • Ensure consistent gene name formatting throughout (e.g., Dot1l, Dot1L).

    Significance

    The current manuscript fills a gap in knowledge, and this is its major strength. Other strengths are clarity and technical appropriateness.

    The major weakness is that the work is mainly descriptive. Mechanistic insights into DOT1L-dependent transcriptional regulation are still weak. The proposed mechanism -that DOT1L maintains pDC identity through H3K79 methylation at key transcription factors (Tcf4, SpiB, Irf8)- is intriguing but currently lacks functional evidence. The authors should consider validating this model experimentally, by modulating the expression of these genes without affecting DOT1L activity. Also the model suggesting that DOT1L indirectly represses antigen presentation via the Fbxo11-Ciita pathway is interesting but remains speculative. Additional mechanistic data would help support this claim.

  4. Review Commons

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

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    Summary:

    In this study, Bouma et al. investigate the epigenetic mechanisms involved in dendritic cell (DC) development, focusing on the role of the lysine methyltransferase DOT1L, which mediates histone H3 lysine 79 (H3K79) methylation. The authors first show that Dot1l is expressed across most DC subsets and their progenitors. Consistently, DOT1L activity was detected in these subsets, as ChIP-seq analysis revealed an enrichment of H3K79 methylation marks around the transcription start sites of numerous genes that regulate DC fate. These marks were associated with active transcription, as confirmed by RNA sequencing. To assess the functional role of Dot1l in DC development, the authors used Rosa26Cre-ERT2 × Dot1l^flox/flox mice. Bone marrow (BM) cells from these mice were treated in vitro with tamoxifen and cultured with FLT3L and SCF to induce DC differentiation. Dot1l deletion impaired the development of plasmacytoid DCs (pDCs) and enhanced the generation of conventional DC2 (cDC2), while leaving cDC1 development unaffected. Similarly, in vivo tamoxifen treatment of Rosa26Cre-ERT2 × Dot1l^flox/flox mice for three days led to a comparable impairment of DC development upon in vitro culture of BM cells. Beyond mature DCs, Dot1l deletion also disrupted the ability of BM cells to generate common myeloid progenitors (CMPs), monocyte-dendritic cell progenitors (MDPs), and common DC progenitors (CDPs). These effects were attributed to the methyltransferase activity of DOT1L, as pharmacological inhibition of DOT1L produced similar outcomes. Interestingly, while in vivo tamoxifen treatment altered the frequencies of progenitor populations (MDP, CDP, CMP) in the BM, it did not significantly change the frequency of pDCs in the BM or spleen. Moreover, an increase in the cDC2 population was observed only in the BM, with no effect detected in the spleen. With these findings the authors claim that epigenetic regulation of gene expression by DOT1L is important for proper dendritic cell development.

    Major comments.

    While this study demonstrates that DOT1L regulates DC development in vitro, its inducible deletion in vivo using tamoxifen does not appear to significantly affect the overall distribution or function of DCs. Therefore, further investigation is needed to clarify the role of DOT1L in regulating DC fate under physiological conditions. The authors analyzed DC populations at only two time points (3 and 12 days) following tamoxifen-induced Dot1l deletion. As noted in the discussion, these time points are relatively early considering the lifespan of DCs, which often extends beyond this period. It would thus be important to assess the effects of Dot1l deletion over a longer duration (e.g., at least one month) to fully evaluate its impact on DC development. In addition to the BM, an extensive analysis of DCs population should be carried in the spleen as well as lymph nodes. Given the broad activity of the Rosa26-Cre system, prolonged deletion may affect overall mouse health and/or the function of other cell types that contribute to DC development; therefore, using a DC-specific Cre driver (e.g., CD11c-Cre) would provide a more targeted approach. Alternatively, competitive BM chimera experiments could be performed by reconstituting irradiated control mice with a 1:1 mixture of BM cells from Rosa26Cre-ERT2 × Dot1l^flox/flox and Rosa26Cre-ERT2 × Dot1l^wt/flox mice, both pre-treated with tamoxifen in vitro. Such experiments would offer more definitive evidence for the role of DOT1L in DC development in vivo. Aside from this point, the data and methods are clearly presented, and the figures are largely self-explanatory. All experiments were adequately replicated three times. Statistical analyses were primarily performed using t-tests, and ANOVA with multiple comparisons when appropriate. Since these are parametric tests that assume a normal distribution, it would be important to confirm whether the analyzed samples meet this assumption. If not, non-parametric tests should be used instead.

    Minor comments.

    It would be informative to show how specific Dot1l expression is in DCs and their progenitors compared with other immune lineages (e.g., lymphocytes) and their precursors. The data suggest that DOT1L regulates H3K79 methylation of both shared and subset-specific genes among DC populations. The authors could elaborate on how this regulation achieves cell-type specificity-perhaps through differential Dot1l expression levels across DC subsets.

    Interestingly, Dot1l deletion both in vitro and in vivo markedly reduces the frequency of common DC progenitors (CDPs), which give rise to cDC1 and cDC2. The authors should discuss how such a substantial loss of progenitors does not proportionally affect downstream cDC populations. Although in vivo tamoxifen-induced deletion of Dot1l in Rosa26Cre-ERT2 × Dot1l^flox/flox mice does not significantly alter the overall distribution of DC subsets (pDCs and cDCs), it appears to modify their phenotype. It would therefore be valuable to examine how Dot1l loss impacts the functional properties of individual DC subsets. While pDC responsiveness to CpG stimulation seems preserved in the absence of Dot1l, assessing how cDCs respond to TLR3 and TLR4 stimulation and their capacity to activate T cells would provide important additional insights.

    Significance

    General assessment: Bouma et al. present compelling evidence that DOT1L is an important regulator of DC differentiation in vitro from bone marrow-derived cells. They further demonstrate that DOT1L regulates DC development through its lysine methyltransferase activity, mediating histone H3K79 methylation. While these in vitro findings are robust and well supported, the physiological relevance of DOT1L function in vivo remains less clearly established. Additional experiments would help to strengthen the conclusions regarding its role under physiological conditions.

    Advance: While numerous transcription factors have been described as key regulators of DC subset development and fate, the role of epigenetic regulation in this process remains relatively understudied and poorly understood. This study addresses this important gap in the literature and provides novel insights into the role of H3K79 methylation mediated by DOT1L in controlling DC development.

    Audience: This paper will be of interest for a specialized audience in the field of the regulation of dendritic cell ontogeny. This work could influence additional research to investigate the epigenitc regulation of DCs development.

  5. Review Commons

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

    Learn more at Review Commons

    Reply to the reviewers

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

  6. Review Commons

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

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    This manuscript investigates the role of DOT1L and its H3K79 methyltransferase activity in dendritic cell (DC) differentiation. The authors employ a combination of in vitro FLT3L/SCF bone marrow culture systems, in vivo inducible knockout models, and genome-wide H3K79me2 ChIP-seq and RNA-seq analyses to demonstrate that DOT1L influences the balance between pDC and cDC2 differentiation, while leaving cDC1 development largely unaffected. The study further identifies transcriptional and epigenetic programs associated with these changes, linking DOT1L deficiency to altered antigen presentation pathways and loss of pDC-associated transcription factors. The paper provides valuable insights into DC biology. However, some of the key conclusions rely heavily on in vitro systems and short-term tamoxifen deletion models, which limit the interpretation of the in vivo data. Strengthening or clearly defining these limitations would substantially improve the paper's impact and clarity.

    Major Comments

    1. To strengthen the paper, the authors could follow one of two alternative strategies:

    (1) Validate their in vitro observations through in vivo experiments, or

    (2) Focus on deepening and refining their in vitro findings, moving the limited in vivo data to the supplementary material and explicitly acknowledging the limitations of the tamoxifen-inducible system.

    Strategy 1 - Strengthen in vivo validation

    -   The experiments presented in Figures 3 and 5 could be repeated in a competitive bone marrow chimera setting (e.g. CD45.1/CD45.2 irradiated hosts reconstituted with a 1:1 mix of WT CD45.1⁺ and Dot1l-KO CD45.2⁺ cells).
-   This design would allow dissection of direct (cell-intrinsic) versus indirect effects of DOT1L deficiency and could mitigate confounding effects of incomplete or asynchronous deletion.
-   After reconstitution, mice could be maintained on tamoxifen-supplemented chow for a longer period to ensure efficient recombination and adequate time for observing phenotypic consequences.
-   Flow cytometric analysis of spleen and bone marrow should use more refined panels to explore DC precursor and subset deficiencies. Suggested reference panels: Rodrigues et al., Immunity 2024; Minutti et al., Nat. Immunol. 2024; Zhu et al., Nat. Immunol. 2015.

    Strategy 2 - Refine in vitro system and reposition in vivo data - The authors could replicate their differentiation assays under conditions that emulate the chimera approach by co-culturing WT (CD45.1⁺) and Dot1l-KO (CD45.2⁺) bone marrow cells. - This would reveal potential competition or cross-talk between WT and mutant cells and provide clearer mechanistic insight into cell-intrinsic versus extrinsic effects. - The authors should examine how tamoxifen itself affects differentiation and measure the kinetics of deletion and H3K79me loss to better contextualize the dynamic response. - It would also be valuable to assess which cDC2 subtypes (A vs. B) are preferentially affected by Dot1l deficiency, again using more sophisticated flow cytometry panels (see references above). If this in vitro-focused strategy is adopted, the in vivo data could be moved to the supplementary material, with explicit acknowledgment that the inducible deletion model and the gradual nature of H3K79me dilution limit the interpretation of the in vivo findings.

    1. In Figures 2 and 3, the efficiency of H3K79me2 depletion following Dot1l excision should be assessed directly. Although DOT1L is the sole H3K79 methyltransferase, the dilution kinetics of H3K79me2 can vary depending on the proliferation rate. Quantifying the H3K79me2 signal in bone marrow-derived cell culture samples would clarify whether the deletion window allowed complete loss of the methylation mark.
    2. Several observations are not discussed in sufficient depth:
      • The finding that Dot1l deletion increases antigen-presentation signatures might reflect stress or activation rather than lineage fate change.
      • The authors could also acknowledge that DOT1L's effect might be indirect, acting through cytokine feedback loops or altered progenitor proliferation, especially given the co-expression of Kit, Flt3, and Irf8 in early DC progenitors.
      • Moreover, because H3K79 methylation is primarily associated with transcriptional elongation rather than initiation, the observed transcriptional changes could result from broader alterations in chromatin accessibility or polymerase processivity, rather than direct promoter regulation. Discussing this mechanistic aspect would help clarify whether DOT1L's role in DC differentiation reflects a direct control of lineage-defining gene expression or a secondary consequence of disrupted transcriptional elongation dynamics.

    Minor Comments

    1. Terminology: The manuscript repeatedly refers to "mature" DCs-please clarify whether this means activated or fully differentiated cells.
    2. Ontogeny statements:
      The assertion that DCs of lymphoid origin are well established should be softened; the lymphoid contribution to some DC lineages remains under discussion.
    3. Transitional DCs (tDCs):
      The equivalence between tDCs and pre-cDC2As remains controversial. This should be acknowledged.
    4. Cytokine supplementation:
      The inclusion of SCF in the FLT3L-based differentiation assays should be justified, it is not a standard procedure.
    5. Macrophage contamination:
      The presence of C1qa, C1qb, and C1qc transcripts in some datasets suggests possible macrophage contamination. Please discuss how this was controlled for or how it might affect interpretation.

    Significance

    This study provides important insights into the epigenetic regulation of DC differentiation by DOT1L. The conclusions would be more compelling if supported by in vivo validation or, alternatively, if the limitations of the current in vivo data were transparently acknowledged and the focus shifted toward mechanistic in vitro depth.

    With these revisions, the manuscript would represent a valuable contribution to understanding how chromatin modification integrates with transcriptional control in shaping dendritic cell fate.

  7. Review Commons

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

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    Bouma et al. present a comprehensive analysis of DOT1L-mediated histone H3K79 methylation across canonical DC subsets. By mapping the methylation landscape, the authors demonstrate that DOT1L regulates both shared and subset-specific gene programs. They show that in vitro or in vivo deletion of Dot1l, followed by in vitro differentiation, results in reduced myeloid progenitors and pDCs alongside an increase in cDC2s, while cDC1 numbers remain largely unaffected. Functionally, Dot1l-deficient DCs fail to produce IFNα upon stimulation. Transcriptomic profiling reveals enrichment of antigen presentation pathways in Dot1l-KO subsets, with upregulated MHC class II surface expression in pDCs. Mechanistically, pharmacological inhibition of DOT1L links these effects to its methyltransferase activity. Collectively, the data suggest that DOT1L differentially regulates canonical DC subset development and represses antigen presentation pathways.

    The manuscript is well-written and technically sound. However, several conclusions would benefit from deeper discussion or additional experimental validation.

    Major Comments

    1. Interpretation of DC balance changes and cell-cycle effects

    The authors propose that DOT1L loss skews DC differentiation toward a pDC-like phenotype. However, DOT1L deletion or inhibition, and the consequent global loss of H3K79 methylation, is well known to downregulate key cell-cycle genes (e.g., Cyclin D1, Cyclin E, CDK4/6, MCM family) while upregulating cell-cycle inhibitors (e.g., Cdkn1a and b). These transcriptional changes are associated with slower proliferation, G1 arrest or delayed S-phase entry, and reduced DNA replication fork progression. Importantly, blocking DNA synthesis (e.g., with aphidicolin or mitomycin C) during early culture inhibits DC emergence, underscoring that proliferation is essential for differentiation. The authors should discuss how their findings align with this established literature. Could the observed DC subset shifts result from impaired cell-cycle progression rather than lineage-specific transcriptional reprogramming? A more detailed consideration of this point is needed.

    1. Discrepancy between in vitro and in vivo pDC phenotypes

    The in vitro data show a marked reduction in pDCs, yet in vivo pDC numbers appear unchanged. Although the discussion briefly mentions proliferation differences, this discrepancy deserves a clearer explanation or experimental follow-up.

    Minor Comments

    • Clarify statistical methods, specify biological replicate numbers, and indicate whether corrections for multiple comparisons were applied to transcriptomic analyses.
    • The introduction is somewhat lengthy and repetitive; condensing it would improve focus.
    • In the discussion sometimes it is not clear the distinction between findings and speculation.
    • Ensure consistent gene name formatting throughout (e.g., Dot1l, Dot1L).

    Significance

    The current manuscript fills a gap in knowledge, and this is its major strength. Other strengths are clarity and technical appropriateness.

    The major weakness is that the work is mainly descriptive. Mechanistic insights into DOT1L-dependent transcriptional regulation are still weak. The proposed mechanism -that DOT1L maintains pDC identity through H3K79 methylation at key transcription factors (Tcf4, SpiB, Irf8)- is intriguing but currently lacks functional evidence. The authors should consider validating this model experimentally, by modulating the expression of these genes without affecting DOT1L activity. Also the model suggesting that DOT1L indirectly represses antigen presentation via the Fbxo11-Ciita pathway is interesting but remains speculative. Additional mechanistic data would help support this claim.

  8. Review Commons

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

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    Summary:

    In this study, Bouma et al. investigate the epigenetic mechanisms involved in dendritic cell (DC) development, focusing on the role of the lysine methyltransferase DOT1L, which mediates histone H3 lysine 79 (H3K79) methylation. The authors first show that Dot1l is expressed across most DC subsets and their progenitors. Consistently, DOT1L activity was detected in these subsets, as ChIP-seq analysis revealed an enrichment of H3K79 methylation marks around the transcription start sites of numerous genes that regulate DC fate. These marks were associated with active transcription, as confirmed by RNA sequencing. To assess the functional role of Dot1l in DC development, the authors used Rosa26Cre-ERT2 × Dot1l^flox/flox mice. Bone marrow (BM) cells from these mice were treated in vitro with tamoxifen and cultured with FLT3L and SCF to induce DC differentiation. Dot1l deletion impaired the development of plasmacytoid DCs (pDCs) and enhanced the generation of conventional DC2 (cDC2), while leaving cDC1 development unaffected. Similarly, in vivo tamoxifen treatment of Rosa26Cre-ERT2 × Dot1l^flox/flox mice for three days led to a comparable impairment of DC development upon in vitro culture of BM cells. Beyond mature DCs, Dot1l deletion also disrupted the ability of BM cells to generate common myeloid progenitors (CMPs), monocyte-dendritic cell progenitors (MDPs), and common DC progenitors (CDPs). These effects were attributed to the methyltransferase activity of DOT1L, as pharmacological inhibition of DOT1L produced similar outcomes. Interestingly, while in vivo tamoxifen treatment altered the frequencies of progenitor populations (MDP, CDP, CMP) in the BM, it did not significantly change the frequency of pDCs in the BM or spleen. Moreover, an increase in the cDC2 population was observed only in the BM, with no effect detected in the spleen. With these findings the authors claim that epigenetic regulation of gene expression by DOT1L is important for proper dendritic cell development.

    Major comments.

    While this study demonstrates that DOT1L regulates DC development in vitro, its inducible deletion in vivo using tamoxifen does not appear to significantly affect the overall distribution or function of DCs. Therefore, further investigation is needed to clarify the role of DOT1L in regulating DC fate under physiological conditions. The authors analyzed DC populations at only two time points (3 and 12 days) following tamoxifen-induced Dot1l deletion. As noted in the discussion, these time points are relatively early considering the lifespan of DCs, which often extends beyond this period. It would thus be important to assess the effects of Dot1l deletion over a longer duration (e.g., at least one month) to fully evaluate its impact on DC development. In addition to the BM, an extensive analysis of DCs population should be carried in the spleen as well as lymph nodes. Given the broad activity of the Rosa26-Cre system, prolonged deletion may affect overall mouse health and/or the function of other cell types that contribute to DC development; therefore, using a DC-specific Cre driver (e.g., CD11c-Cre) would provide a more targeted approach. Alternatively, competitive BM chimera experiments could be performed by reconstituting irradiated control mice with a 1:1 mixture of BM cells from Rosa26Cre-ERT2 × Dot1l^flox/flox and Rosa26Cre-ERT2 × Dot1l^wt/flox mice, both pre-treated with tamoxifen in vitro. Such experiments would offer more definitive evidence for the role of DOT1L in DC development in vivo. Aside from this point, the data and methods are clearly presented, and the figures are largely self-explanatory. All experiments were adequately replicated three times. Statistical analyses were primarily performed using t-tests, and ANOVA with multiple comparisons when appropriate. Since these are parametric tests that assume a normal distribution, it would be important to confirm whether the analyzed samples meet this assumption. If not, non-parametric tests should be used instead.

    Minor comments.

    It would be informative to show how specific Dot1l expression is in DCs and their progenitors compared with other immune lineages (e.g., lymphocytes) and their precursors. The data suggest that DOT1L regulates H3K79 methylation of both shared and subset-specific genes among DC populations. The authors could elaborate on how this regulation achieves cell-type specificity-perhaps through differential Dot1l expression levels across DC subsets.

    Interestingly, Dot1l deletion both in vitro and in vivo markedly reduces the frequency of common DC progenitors (CDPs), which give rise to cDC1 and cDC2. The authors should discuss how such a substantial loss of progenitors does not proportionally affect downstream cDC populations. Although in vivo tamoxifen-induced deletion of Dot1l in Rosa26Cre-ERT2 × Dot1l^flox/flox mice does not significantly alter the overall distribution of DC subsets (pDCs and cDCs), it appears to modify their phenotype. It would therefore be valuable to examine how Dot1l loss impacts the functional properties of individual DC subsets. While pDC responsiveness to CpG stimulation seems preserved in the absence of Dot1l, assessing how cDCs respond to TLR3 and TLR4 stimulation and their capacity to activate T cells would provide important additional insights.

    Significance

    General assessment: Bouma et al. present compelling evidence that DOT1L is an important regulator of DC differentiation in vitro from bone marrow-derived cells. They further demonstrate that DOT1L regulates DC development through its lysine methyltransferase activity, mediating histone H3K79 methylation. While these in vitro findings are robust and well supported, the physiological relevance of DOT1L function in vivo remains less clearly established. Additional experiments would help to strengthen the conclusions regarding its role under physiological conditions.

    Advance: While numerous transcription factors have been described as key regulators of DC subset development and fate, the role of epigenetic regulation in this process remains relatively understudied and poorly understood. This study addresses this important gap in the literature and provides novel insights into the role of H3K79 methylation mediated by DOT1L in controlling DC development.

    Audience: This paper will be of interest for a specialized audience in the field of the regulation of dendritic cell ontogeny. This work could influence additional research to investigate the epigenitc regulation of DCs development.

  9. Version published to 10.1101/2025.09.05.674429 on bioRxiv