Écrire un commentaire

Title: Mammalian condensin I controls higher-order chromosome organization and homologous recombination in meiotic prophase I

Authors: Laurine Dal Toe, Boubou Diagouraga, Alexandre Nore, Audrey Bost, Estelle Grosjean, Julien Cau, Thomas Robert

Overall assessment: In this preprint, Dal Toe et al. investigate the role of condensin I in mammalian meiotic prophase I, focusing on the NCAPD2 subunit. Using a germline conditional knockout mouse model together with immuno-FISH, quantitative cytology, and super-resolution microscopy, the authors characterize changes in meiotic chromosome organization, synaptonemal complex architecture, and recombination-associated protein dynamics following NCAPD2 depletion.

The study addresses an important and relatively underexplored question in mammalian meiosis and provides evidence that NCAPD2 contributes to several aspects of chromosome organization, including lateral chromatin compaction, synaptonemal complex (SC) architecture, and the dynamics of recombination-associated proteins. Interpretation of several of these phenotypes, however, is complicated by incomplete penetrance of the conditional knockout, and many of the reported effects are quantitatively modest. In addition, some of the mechanistic interpretations extend beyond what is directly established by the current cytological measurements.

Overall, the data support contributions of NCAPD2 to meiotic chromosome organization and recombination progression, while some of the more specific causal relationships proposed throughout the manuscript remain less directly supported by the current data. More clearly distinguishing direct observations from inferred mechanisms, accounting for the mosaic nature of NCAPD2 depletion and the biological-replicate structure of the experiments, and calibrating the wording of the conclusions to the magnitude and nature of the measured effects would strengthen the manuscript.

Major points:

• Incomplete penetrance of the conditional knockout complicates interpretation of both positive and negative phenotypes. The Ncapd2 cKO population contains a substantial fraction of cells with WT-like NCAPD2 levels. The authors report that approximately 20–30% of leptotene-to-pachytene nuclei retain WT-like NCAPD2 signal, with this fraction increasing markedly at diplonema. This heterogeneity is likely to dilute positive phenotypes and may therefore underestimate the consequences of NCAPD2 depletion. More importantly, it also complicates interpretation of negative results, because cells retaining near-WT NCAPD2 levels may contribute disproportionately to measurements in which little or no phenotype is detected. The manuscript itself recognizes that later-stage cells may be enriched for cells in which depletion was incomplete or delayed. Where technically feasible, the authors could consider classifying cells according to NCAPD2 depletion status in the relevant analyses. It would also be helpful to distinguish more clearly between the reported 100% transmission of the deleted allele and the efficiency or timing of NCAPD2 protein depletion in the meiotic cells being analyzed. This distinction may be particularly important when interpreting fertility, progression through prophase I, and phenotypes described as NCAPD2-independent.

• The biological-replicate structure and magnitude of the effects could be more clearly incorporated into the statistical analysis. Many cytological comparisons appear to use individual nuclei, or in some cases multiple measurements within nuclei, as independent observations, although these observations are nested within a much smaller number of biological replicates. Most figure legends report nuclei as n and use Mann–Whitney tests at the cell level; for example, Figure 2 includes approximately 55–80 nuclei per condition for several FISH measurements. This becomes particularly important because many of the reported genotype effects are quantitatively modest. Large numbers of nuclei can increase statistical power and yield small P values, but this does not necessarily indicate that the effect is similarly consistent across independent animals. The authors could consider reporting the number of mice contributing to each experiment and showing the distribution of measurements across individual animals where feasible. Analysis at the biological-replicate level, or alternatively a hierarchical/mixed-effects framework that accounts for the nesting of cells within animals, may better reflect the experimental structure. Reporting effect sizes and exact P values in addition to significance symbols would also make the strength and reproducibility of the evidence easier to evaluate.

• The interpretation of the whole-chromosome FISH normalization would benefit from additional clarification. For the whole-chromosome FISH experiments (e.g. Figure 2), the authors normalize FISH-signal area to total DAPI nuclear area and interpret an increase in this ratio as evidence that the chromosome occupies a larger space and is therefore less compact. Because this measurement is a ratio, however, a difference could in principle arise from changes in the chromosome FISH area, nuclear area, or both. It would therefore be helpful to present the absolute FISH areas and corresponding nuclear areas alongside the normalized values, if these measurements are available. This would make it easier to determine the physical basis of the observed 1.3- to 1.6-fold differences and to evaluate how strongly the normalized measurements support a change in global chromosome compaction.

• Additional support would be helpful to establish an altered recombination outcome. The manuscript interprets changes in MLH1, HEI10, and HEIP1 foci as evidence that NCAPD2 affects recombination progression and outcome. These cytological markers are informative for the dynamics and designation of recombination-associated sites, but they do not by themselves establish changes in final crossover number, crossover resolution, or other recombination products. The late-pachytene changes reported for these markers are also relatively modest, on the order of ~10%. The authors could consider distinguishing more consistently between altered recombination-marker dynamics/progression and altered recombination products or outcomes, particularly in the Abstract, Results headings, Discussion, and Figure 7 model. If the authors wish to retain stronger conclusions regarding recombination outcome, additional readouts of final crossover products or chiasmata would provide more direct support.

• Additional structural evidence would help distinguish the proposed nested-loop model from other explanations for the increased chromatin extension. The fosmid-FISH experiments show that the assayed chromatin regions extend farther from the chromosome axis in Ncapd2 cKO spermatocytes. These measurements provide evidence for altered chromatin compaction or radial extension. However, the assay measures the spatial extent of a labeled chromatin region relative to the axis and does not directly reveal how that chromatin is folded within the measured region. Thus, the data do not by themselves distinguish a nested-loop configuration from other possible changes in chromatin organization. The authors could consider presenting the nested-loop configuration more consistently as a working model rather than as an established structural intermediate. If the authors wish to strengthen the specific nested-loop interpretation, additional approaches that more directly interrogate three-dimensional chromatin architecture could help distinguish this model from alternative explanations for the increased chromatin extension.

• More precise causal language would help align the manuscript’s conclusions with the strength of the current evidence. Terms such as “control,” “drive,” and “regulate” are used throughout the manuscript to describe the effects of NCAPD2 on chromatin organization, SC components, and recombination-associated proteins. This is particularly prominent in the Abstract, where condensin I is described as controlling higher-order chromosome organization, structural-protein localization, recombination-associated protein turnover, SC organization, and recombination outcome. Given that many of the observed changes are quantitatively modest and that NCAPD2 depletion is incomplete, the current data appear more directly consistent with partial contributions of NCAPD2 to these processes, which may be direct or indirect. In several cases, the wording also implies a specific regulatory relationship even though the relevant causal step has not been directly identified. For example, “NCAPD2 loss alters DMC1 focus dynamics” more directly describes the measured phenotype than “NCAPD2 controls DMC1 loading.” The authors could consider revisiting this terminology throughout the manuscript and using more descriptive language where the mechanism remains unresolved. This would help distinguish the observed phenotypes from the proposed regulatory relationships and improve the precision of the manuscript’s central conclusions.

Minor points:

• (Results, Figure S3) Because REC8 and RAD21L mark the underlying cohesin-defined chromosome axes and their recruitment patterns and overall signal intensities appear unchanged in the Ncapd2 cKO, measurements using these markers could provide an orthogonal assessment of the conclusion that axis length is unaffected. If suitable data are available, super-resolution analysis might also help distinguish whether the increased SC width reflects altered spacing of the cohesin-defined axes or changes primarily involving SYCP3 and/or central-region components.

• (Results, p. 6, line 119; Figure 1C and Figure S1A) The description that NCAPD2 foci “colocalize with DAPI-stained chromatin” is somewhat difficult to interpret because DAPI broadly labels chromatin throughout the nucleus. The authors could consider using wording such as “distributed throughout chromatin” or “chromatin-associated,” which may more precisely describe the observed localization pattern.

• (Results, pp. 12–14; Figures 3F–G and 4) The manuscript appears in several places to use “SC central region” and “central element” interchangeably. Because these terms refer to related but non-identical structural features of the SC, using them more consistently may help avoid ambiguity, particularly when describing the localization or recruitment of specific proteins.

• (Results, p. 7, lines 143–150; Figure 1 and Figure S1) A few quantitative descriptions could benefit from clarification. For example, the statement that NCAPD2 signal is lost in “80%–70% of foci” is difficult to reconcile with the following sentence, which describes 20%–30% of nuclei as retaining WT-like NCAPD2 signal. Clarifying the denominator and whether these percentages refer to foci or nuclei would make the extent of NCAPD2 depletion easier to interpret. Reported sample sizes could also be checked for consistency between the main text and figure legends.

I reviewed this preprint as part of the Transparent and Accountable Peer Review training program. This report was written with the support of my research mentor.

Competing interests

The authors declare that they have no competing interests.

Use of Artificial Intelligence (AI)

The authors declare that they did not use generative AI to come up with new ideas for their review.