Organization of DNA replication origin firing in Xenopus egg extracts : the role of intra-S checkpoint

Article activity feed

1. 

   eLife

   Feb 1, 2021

   ###Author Response:

   ###Summary:

   This paper uses numerical simulations to model DNA replication dynamics in an in vitro Xenopus DNA replication system, both in unperturbed conditions and upon intra-S-checkpoint inhibition. The current work extends previous studies by the authors that recapitulated some but not all features of the replication program. The new model is superior as it can model both the frequency and the distribution of observed initiation events. Although the reviewers found the work in principle interesting and well executed, they have identified limitations of the study, both with respect to model validation and the extent to which the findings represent new biological insights into origin regulation and replication dynamics.

   We would like to thank the referees and the editor to have read and commented our work. The

   ###Author Response:

   ###Summary:

   This paper uses numerical simulations to model DNA replication dynamics in an in vitro Xenopus DNA replication system, both in unperturbed conditions and upon intra-S-checkpoint inhibition. The current work extends previous studies by the authors that recapitulated some but not all features of the replication program. The new model is superior as it can model both the frequency and the distribution of observed initiation events. Although the reviewers found the work in principle interesting and well executed, they have identified limitations of the study, both with respect to model validation and the extent to which the findings represent new biological insights into origin regulation and replication dynamics.

   We would like to thank the referees and the editor to have read and commented our work. The main message that we grasp from the three referees comments is that this work lacks “ new biological insights into origin regulation and replication dynamics”.

   To our knowledge, this work is the first one to clearly show that:

   • The origin clustering is not regulated by intra-S checkpoint in Xenopus egg extract as was proposed previously [1].

   • The variability of the rate of DNA synthesis close to replication forks is a necessary ingredient to describe the dynamic of replication origin firing.

   • Heterogenous firing probabilities in the embryonic Xenopus system

   We believe that the common referees conclusion arises because these important conclusions were not clearly and explicitly stated in our manuscript. Hence, we modified our manuscript to explicitly state these new insights. Please find below our detailed answers to the referee’s comments, criticisms and suggestions.

   ###Reviewer #1:

   The current work by Goldar and colleagues uses numerical simulations to model the spatiotemporal DNA replication program in an in vitro Xenopus DNA replication system. By comparing modeled data and experimental DNA combing data generated during unperturbed S-phase replication and upon intra-S checkpoint inhibition (which the authors published previously), the authors find that DNA replication in Xenopus extracts can be modeled by segmenting the genome in regions of high and low probability of origin activation, with the intra-S-phase checkpoint regulating origins with low but not high firing probability. Recapitulating the kinetics of global and local S-phase replication under different conditions through mathematical simulations represents an important contribution to the field. However, one concern I have pertains to the generality of the model, as the authors did not explore whether the model can accurately simulate replication under other conditions (e.g., checkpoint activation).

   In this work we showed that the same combination of processes can recapitulate several observations on the spatio-temporal pattern of DNA replication (as measured by DNA combing) in unchallenged and checkpoint inhibited conditions. Following the referee’s suggestion to “explore whether the model can accurately simulate replication under other conditions”, we also applied our methodology to a condition where Chk1 is over-expressed. We were able to reproduce the pattern of DNA replication as measured by DNA combing and found, as expected, that the over-expression of Chk1 reduces the rate of origin firing, but only by reducing the number of available limiting factors and not the capacity of potential origins to fire. This analysis was added to our manuscript and discussed.

   Major comments:

   1. In figure 1a and 1c, the authors show data that were previously published by the authors. Yet, the displayed values in 1a and 1c differ from those displayed in Figure 10 of Platel et al, 2015. This discrepancy should be explained.

   The discrepancy results from the thresholding of the optical signal and the smoothing of the experimental data in Platel et al, 2015. In the work presented here, we decided to model raw profiles after the thresholding. While the absolute values of the extracted data are different from those in Platel et al 2015, the trends of I(f) and fork density profiles are similar. We stated this point clearly in the caption of figure 2.

   1. The authors test whether their model can simulate replication when S-phase is perturbed by Chk1 inhibition, but not under opposite conditions of Chk1 activation. This important analysis should be included.

   The experimental mean chosen for activating or inhibiting (manipulating) the checkpoint in Platel et al 2015 was respectively to overexpress Chk1 protein, or to inhibit its activity using the specific inhibitor UCN-01. We further analysed Chk1 overexpression combed fibres and add this new analysis to our manuscript (See above).

   1. Although the MM4 model developed by the authors is in agreement with previously published experimental DNA combing data measured in the Xenopus system, it is unclear whether it can also accurately predict the replication program in other systems. Comparing simulated data with experimental data from another metazoan system would serve as an important additional validation of the authors' model.

   We agree with the referee that the generality of this model has to be tested by comparing it with experimental data from other metazoan. Unfortunately, to our knowledge, there is no available DNA combing data in other metazoan where the effect of inhibition ( and now “activation”) of intra S checkpoint have been measured concomitantly with cells under unchallenged growth 3 conditions. If the referee is aware of such an available data we will be happy to analyse them. It is possible to compare a simulation of our model with replication timing profiles measured by NGS techniques, by introducing in the model a distribution of length for regions where the probability of origin firing is high. This will result in a timing profile where we can define TTR and CTR as it has been done in human cell lines [2]. However, this requires the addition of a supplementary parameter: the length of domains with high probability of origin firing. This would complexify the model and cannot be justified on a statistical ground based on combing data (see annex 1 of our new manuscript, this model corresponds to MM6)

   ###Reviewer #2:

   Here the authors expand on their prior modeling of origin activity (Platel 2015) in xenopus extracts. Their prior work, while successful in some estimates, failed to reproduce the tight distribution of interorigin ("eye to eye") distances. Here the authors generate a series of nested models (MM1-MM4) of increasing complexity to describe the distribution and frequency of observed initiation events in an unperturbed S-phase. Not surprisingly, the fit improves with the increasing complexity of each model.

   The improvement of the concordance between the model and the data was assessed by 2 statistical methods (F test and AIC) in order to avoid overfitting of data. Both tests showed that the increasing complexity of the model were necessary to explain the variability of measured data. In fact, one could still increase the complexity of the model (for example one could use our fictitious model to fit the data ). In this case, the F test and AIC score show that the better representation of the data by the model is due to the increase in the complexity and not the necessity of considered processes. We included this discussion in annex 1 of our new manuscript.

   The authors then built an even more complex model based on prior published work to generate in silico data for which they tested their MM4 model. I admit to being a little lost at this point as to why the authors were using simulated data to assess their model and identify key parameters.

   The in silico data helps us to verify the quantitative ability of our model and validate the analysis process that we propose.

   Finally, the authors compare prior published experimental data from an unperturbed S-phase and one with an abrogated intra s-phase checkpoint (chk1 inhibition) and three parameters stood out J (rate limiting factor), 𝜃 (fraction of the genome with high origin initiation activity), and Pout (probability of remaining origins to fire) which suggests that Chk1 limits the probability of origin activation outside of the regions of the genome with high origin activation efficiency and modulates the activity of the rate limiting factor (J). These conclusions are consistent with prior observations in other systems. In summary, the authors apply elegant modeling approaches to describe xenopus in vitro replication dynamics and the effects of Chk1 inhibition, but the work fails to reveal new principles of eukaryotic origin regulation and replication dynamics.

   See above

   The most powerful modeling approaches are those that reveal a new or unexpected mode of regulation (or parameter) that can then be experimentally tested.

   We agree with the referee, and thank him for his comment. We re-wrote part of our manuscript to explicitly indicate “the new principles of eukaryotic origin regulation and replication dynamics” that our analysis implies.

   Additional points:

   This was a very specialized manuscript and would be difficult to read for general biologists. The terms/parameters were only defined in a table and many of the figures would not be parsable by a broad audience.

   We re-wrote part of the manuscript to make it more readable, and transfer technical details in annexes. We added a new subfigure Fig 1a to better explain combing parameters

   Figure 1. Sets off the challenge at hand -- that the previous model couldn't account for the distribution of "eye to eye" distances; but this is never assessed in similar format with the newer model. I assume this is captured in the appendix 1 figures, but was uncles if this was eye length or gap length.

   The referee is correct, this is represented in figures in annexes sections, where we showed that our modelling approach can reproduce in a satisfactory manner replication fraction of measured fibres, I(f), fork density, eye length distribution, gap length distribution and eye-to-eye distribution in all considered conditions. Following the referee’s suggestion we added in our new manuscript a figure comparable to figure 2 for our new model in the main text.

   ###Reviewer #3:

   General assessment:

   The authors arrive at a plausible model of DNA replication kinetics that reasonably fits six types of plots from fiber-combing data on Xenopus cell-free extracts, for normal and challenged cases. However, although the mechanisms postulated and the parameters inferred all seem reasonable, they rely on untested hypotheses and a single type of data (combing).

   All hypothesis used in this model have already been proposed and tested in existing literature, as stated in the discussion (lines 309-315 in the new manuscript) where all used hypothesis are explained and referenced.

   We use DNA combing data, and compare our conclusions to observations in the literature obtained by other techniques. Indeed, DNA combing (and in general DNA fibre stretching technique combined with optical detection) has the unique ability to allow working directly on distribution of parameters like eye-to-eye distances, eye-length…. Hence the data are not biased by any type of population averaging (as it is the case in the NGS our other classical biochemical techniques ).

   To truly convince, the authors need further experiments to test specific hypothesized mechanisms.

   This is not the purpose of this work and we do not propose any molecular mechanism. We look for essential ingredients necessary to reproduce spatio-temporal dynamic of DNA replication.

   Techniques such as Repli-Seq or perhaps FORK-seq (recently developed by one of the authors here) might give direct information on the variation of initiation efficiency across the genome.

   We analyse data from Xenopus invitro system that has been extensively used to investigate spatio-temporal pattern of DNA replication. Unfortunately, the referenced genome of this organism is not assembled accurately enough to allow techniques such as Repli-Seq or FORKseq that require mapping procedure on a reference genome. Furthermore, these techniques require a cell population containing more than 107 individuals [3], here we are working with 200000 to 500000 nuclei. Hence without changing model system these techniques could not be applied.

   Substantive Concerns:

   1. The authors refer to each case (MM1-5) as a unique model, but each has more complexity and defines a class of models.

   MM1-5 belong all to the unique class of nucleation and growth process defined as KJMA model. All models are variants of this model. We do not understand the point of referee, if the referee means that each case can represent the data not in a unique manner, we agree with him/here and this is the reason we used a genetic algorithm and not a gradient descent algorithm to minimise the difference between the data and the considered model.

   For example, in fitting MM1, the simplest of all the cases (and with, by far, the worst fit), the fork velocity was fixed at 0.5 kb/min. And yet the real fork velocity is described as having v ~ 0.5 kb/min. Shouldn't this also be a parameter in the fit?

   We chose to keep the velocity as a constant and close to the observed experimental value, as in Xenopus egg extract it is assumed that the fork velocity is constant [4]. But indeed, one could consider fork velocity as a fitting parameter (see the answer to the next point), but this is not in accordance with experimental observations.

   1. Under replication stress, forks can stall, giving an effectively two populations of forks, as proposed by the authors in an earlier work (Ciardo et al., Genes 2019; cf. Fig. 1). Strangely, that paper is not referred to or discussed in this manuscript. Why not?

   Indeed, instead of self-citation of a review article we preferred to refer to original experimental works. Furthermore, in order to change the mean of eye to eye distribution by only changing the speed of replication forks, one should consider that the speed of replication forks should have a value higher than 10kb/min‼! which has not been reported in any organism. To be conservative, we ran a model where the speed of replication forks could take several values ranging between 0 to 3kb/min. The model failed to fit the experimental data. (see the new manuscript and annex 1). Hence, we consider that the best model is the one with constant speed.

   1. Continuous vs. discrete potential origins: The density was fixed to be random at 1 potential origin per 2.3 kb (or 1 kb in part of the paper). How robust are findings to these assumed densities?

   If we consider the density as a free parameter, the model converges with a density of 1 origin every 2.3 kb.

   In general, there does not seem to be a huge difference between the two cases, for the type of data explored. Perhaps it is not worth looking at the discrete case here?

   The difference is that in the “discrete” case the distribution of origins is not continuous and hence there naturally exists a distance between two fired origin where the origin firing is inhibited. The existence of such an origin firing exclusion zone was shown to be necessary to model replication dynamic as measured by DNA combing [5,6].

   1. The definition of goodness of the fit (GoF) should be made more explicitly. How is the norm calculated? There is an implicit sum - the elements should be defined explicitly. Also, the ensemble average < yexp > is not defined. More broadly, it is not clear why we need a custom GoF statistic when it would seem that standard ones (chi square, or - ln likelihood) could serve equally well.

   The defined GoF is a classical normalised chi squared as defined in annex 1. We modified the text to include explicitly the summation over the data points. By definition is the average value of an experimental data series. GoF is not a custom defined criterion but the classical normalised chi square [7].

   Note that those statistics (when proper normalization is used) can also work for global fits where each local fit is to a quantity with different units.

   References:

   1. Ge XQ, Blow JJ. Chk1 inhibits replication factory activation but allows dormant origin firing in existing factories. The Journal of Cell Biology. 2010;191: 1285–1297. doi:10.1083/jcb.201007074

   2. Pope BD, Ryba T, Dileep V, Yue F, Wu W, Denas O, et al. Topologically associating domains are stable units of replication-timing regulation. Nature. 2014;515: 402–405. doi:10.1038/nature13986

   3. Petryk N, Kahli M, d’Aubenton-Carafa Y, Jaszczyszyn Y, Shen Y, Silvain M, et al. Replication landscape of the human genome. Nat Commun. 2016;7: 10208. doi:10.1038/ncomms10208

   4. Marheineke K, Hyrien O. Control of Replication Origin Density and Firing Time in Xenopus Egg Extracts ROLE OF A CAFFEINE-SENSITIVE, ATR-DEPENDENT CHECKPOINT. J Biol Chem. 2004;279: 28071–28081. doi:10.1074/jbc.M401574200

   5. Löb D, Lengert N, Chagin VO, Reinhart M, Casas-Delucchi CS, Cardoso MC, et al. 3D replicon distributions arise from stochastic initiation and domino-like DNA replication progression. Nature Communications. 2016;7: 11207. doi:10.1038/ncomms11207

   6. Jun S, Herrick J, Bensimon A, Bechhoefer J. Persistence length of chromatin determines origin spacing in Xenopus early-embryo DNA replication: quantitative comparisons between theory and experiment. Cell Cycle. 2004;3: 223–229.

   7. Bevington P, Robinson DK. Data Reduction and Error Analysis for the Physical Sciences. McGraw-Hill Education; 2003.

   Read the original source
2. 

   eLife

   Aug 21, 2020

   ###Reviewer #3:

   General assessment:

   The authors arrive at a plausible model of DNA replication kinetics that reasonably fits six types of plots from fiber-combing data on Xenopus cell-free extracts, for normal and challenged cases. However, although the mechanisms postulated and the parameters inferred all seem reasonable, they rely on untested hypotheses and a single type of data (combing). To truly convince, the authors need further experiments to test specific hypothesized mechanisms. Techniques such as Repli-Seq or perhaps FORK-seq (recently developed by one of the authors here) might give direct information on the variation of initiation efficiency across the genome.

   Substantive Concerns:

   1. The authors refer to each case (MM1-5) as a unique model, but each has more complexity and defines a class of models. For example, in fitting …

   ###Reviewer #3:

   General assessment:

   The authors arrive at a plausible model of DNA replication kinetics that reasonably fits six types of plots from fiber-combing data on Xenopus cell-free extracts, for normal and challenged cases. However, although the mechanisms postulated and the parameters inferred all seem reasonable, they rely on untested hypotheses and a single type of data (combing). To truly convince, the authors need further experiments to test specific hypothesized mechanisms. Techniques such as Repli-Seq or perhaps FORK-seq (recently developed by one of the authors here) might give direct information on the variation of initiation efficiency across the genome.

   Substantive Concerns:

   1. The authors refer to each case (MM1-5) as a unique model, but each has more complexity and defines a class of models. For example, in fitting MM1, the simplest of all the cases (and with, by far, the worst fit), the fork velocity was fixed at 0.5 kb/min. And yet the real fork velocity is described as having v ~ 0.5 kb/min. Shouldn't this also be a parameter in the fit?

   2. Under replication stress, forks can stall, giving an effectively two populations of forks, as proposed by the authors in an earlier work (Ciardo et al., Genes 2019; cf. Fig. 1). Strangely, that paper is not referred to or discussed in this manuscript. Why not?

   3. Continuous vs. discrete potential origins: The density was fixed to be random at 1 potential origin per 2.3 kb (or 1 kb in part of the paper). How robust are findings to these assumed densities? In general, there does not seem to be a huge difference between the two cases, for the type of data explored. Perhaps it is not worth looking at the discrete case here?

   4. The definition of goodness of the fit (GoF) should be made more explicitly. How is the norm calculated? There is an implicit sum - the elements should be defined explicitly. Also, the ensemble average < yexp > is not defined. More broadly, it is not clear why we need a custom GoF statistic when it would seem that standard ones (chi square, or - ln likelihood) could serve equally well. Note that those statistics (when proper normalization is used) can also work for global fits where each local fit is to a quantity with different units.

   Read the original source
3. 

   eLife

   Aug 21, 2020

   ###Reviewer #2:

   Here the authors expand on their prior modeling of origin activity (Platel 2015) in xenopus extracts. Their prior work, while successful in some estimates, failed to reproduce the tight distribution of interorigin ("eye to eye") distances. Here the authors generate a series of nested models (MM1-MM4) of increasing complexity to describe the distribution and frequency of observed initiation events in an unperturbed S-phase. Not surprisingly, the fit improves with the increasing complexity of each model. The authors then built an even more complex model based on prior published work to generate in silico data for which they tested their MM4 model. I admit to being a little lost at this point as to why the authors were using simulated data to assess their model and identify key parameters. Finally, the authors compare prior …

   ###Reviewer #2:

   Here the authors expand on their prior modeling of origin activity (Platel 2015) in xenopus extracts. Their prior work, while successful in some estimates, failed to reproduce the tight distribution of interorigin ("eye to eye") distances. Here the authors generate a series of nested models (MM1-MM4) of increasing complexity to describe the distribution and frequency of observed initiation events in an unperturbed S-phase. Not surprisingly, the fit improves with the increasing complexity of each model. The authors then built an even more complex model based on prior published work to generate in silico data for which they tested their MM4 model. I admit to being a little lost at this point as to why the authors were using simulated data to assess their model and identify key parameters. Finally, the authors compare prior published experimental data from an unperturbed S-phase and one with an abrogated intra s-phase checkpoint (chk1 inhibition) and three parameters stood out J (rate limiting factor), 𝜃 (fraction of the genome with high origin initiation activity), and Pout (probability of remaining origins to fire) which suggests that Chk1 limits the probability of origin activation outside of the regions of the genome with high origin activation efficiency and modulates the activity of the rate limiting factor (J). These conclusions are consistent with prior observations in other systems. In summary, the authors apply elegant modeling approaches to describe xenopus in vitro replication dynamics and the effects of Chk1 inhibition, but the work fails to reveal new principles of eukaryotic origin regulation and replication dynamics. The most powerful modeling approaches are those that reveal a new or unexpected mode of regulation (or parameter) that can then be experimentally tested.

   Additional points:

   This was a very specialized manuscript and would be difficult to read for general biologists. The terms/parameters were only defined in a table and many of the figures would not be parsable by a broad audience.

   Figure 1. Sets off the challenge at hand -- that the previous model couldn't account for the distribution of "eye to eye" distances; but this is never assessed in similar format with the newer model. I assume this is captured in the appendix 1 figures, but was uncles if this was eye length or gap length.

   Read the original source
4. 

   eLife

   Aug 21, 2020

   ###Reviewer #1:

   The current work by Goldar and colleagues uses numerical simulations to model the spatiotemporal DNA replication program in an in vitro Xenopus DNA replication system. By comparing modeled data and experimental DNA combing data generated during unperturbed S-phase replication and upon intra-S checkpoint inhibition (which the authors published previously), the authors find that DNA replication in Xenopus extracts can be modeled by segmenting the genome in regions of high and low probability of origin activation, with the intra-S-phase checkpoint regulating origins with low but not high firing probability. Recapitulating the kinetics of global and local S-phase replication under different conditions through mathematical simulations represents an important contribution to the field. However, one concern I have pertains to …

   ###Reviewer #1:

   The current work by Goldar and colleagues uses numerical simulations to model the spatiotemporal DNA replication program in an in vitro Xenopus DNA replication system. By comparing modeled data and experimental DNA combing data generated during unperturbed S-phase replication and upon intra-S checkpoint inhibition (which the authors published previously), the authors find that DNA replication in Xenopus extracts can be modeled by segmenting the genome in regions of high and low probability of origin activation, with the intra-S-phase checkpoint regulating origins with low but not high firing probability. Recapitulating the kinetics of global and local S-phase replication under different conditions through mathematical simulations represents an important contribution to the field. However, one concern I have pertains to the generality of the model, as the authors did not explore whether the model can accurately simulate replication under other conditions (e.g., checkpoint activation).

   Major comments:

   1. In figure 1a and 1c, the authors show data that were previously published by the authors. Yet, the displayed values in 1a and 1c differ from those displayed in Figure 10 of Platel et al, 2015. This discrepancy should be explained.

   2. The authors test whether their model can simulate replication when S-phase is perturbed by Chk1 inhibition, but not under opposite conditions of Chk1 activation. This important analysis should be included.

   3. Although the MM4 model developed by the authors is in agreement with previously published experimental DNA combing data measured in the Xenopus system, it is unclear whether it can also accurately predict the replication program in other systems. Comparing simulated data with experimental data from another metazoan system would serve as an important additional validation of the authors' model.

   Read the original source
5. 

   eLife

   Aug 21, 2020

   ##Preprint Review

   This preprint was reviewed using eLife’s Preprint Review service, which provides public peer reviews of manuscripts posted on bioRxiv for the benefit of the authors, readers, potential readers, and others interested in our assessment of the work. This review applies only to version 1 of the manuscript.

   ###Summary:

   This paper uses numerical simulations to model DNA replication dynamics in an in vitro Xenopus DNA replication system, both in unperturbed conditions and upon intra-S-checkpoint inhibition. The current work extends previous studies by the authors that recapitulated some but not all features of the replication program. The new model is superior as it can model both the frequency and the distribution of observed initiation events. Although the reviewers found the work in principle interesting and well executed, …

   ##Preprint Review

   This preprint was reviewed using eLife’s Preprint Review service, which provides public peer reviews of manuscripts posted on bioRxiv for the benefit of the authors, readers, potential readers, and others interested in our assessment of the work. This review applies only to version 1 of the manuscript.

   ###Summary:

   This paper uses numerical simulations to model DNA replication dynamics in an in vitro Xenopus DNA replication system, both in unperturbed conditions and upon intra-S-checkpoint inhibition. The current work extends previous studies by the authors that recapitulated some but not all features of the replication program. The new model is superior as it can model both the frequency and the distribution of observed initiation events. Although the reviewers found the work in principle interesting and well executed, they have identified limitations of the study, both with respect to model validation and the extent to which the findings represent new biological insights into origin regulation and replication dynamics.

   Read the original source
6. 

   Version published to 10.1101/2020.06.22.164673 on bioRxiv

   Jun 23, 2020