Review Reports - Functional Aspects of Fibrin Structure Alterations by Tranexamic Acid in the Inhibition of Fibrinolysis

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study addresses an important and clinically relevant question regarding the context-dependent effects of tranexamic acid (TXA) on fibrinolysis. The integration of multiple complementary techniques (confocal microscopy, viscoelastic assays, turbidimetry, plasminogen activation assays, and SEM) is a major strength and provides a coherent experimental framework. The observation that pre-incorporated TXA can accelerate fibrinolysis under physiological plasminogen concentrations is particularly novel and may help explain inconsistencies observed in prophylactic clinical settings.

However, several aspects require clarification and improvement:

First, the physiological relevance of the in vitro models is limited. The experiments are performed in purified systems lacking key modulators such as platelets, FXIII crosslinking, TAFI, α2-antiplasmin dynamics, and flow conditions. The authors should more explicitly discuss how these missing components might influence their findings and temper the translational interpretation. Second, the mechanistic conclusion that structural alterations of fibrin are the primary driver of the paradoxical effect is not fully established. While enhanced plasminogen activation is excluded, alternative mechanisms such as altered plasmin diffusion, fibrin binding kinetics, or tPA distribution within the clot are not directly investigated. The conclusions should therefore be moderated. Third, the definition of “prophylactic” versus “therapeutic” conditions in the experimental setup is somewhat simplified and may not fully reflect in vivo dynamics. This distinction should be clarified and discussed more carefully. Fourth, the choice and range of TXA concentrations should be better justified in relation to clinically observed plasma and tissue levels. A more detailed dose–response analysis would strengthen the study. Finally, while the manuscript is generally well written, parts of the Discussion are overly dense and could be streamlined to improve clarity and focus on the novel contributions of the study. Overall, this is a valuable and thought-provoking study, but addressing the points above would significantly strengthen its mechanistic rigor and translational relevance.

Author Response

We would like to thank the reviewers for their careful evaluation of our manuscript and for their valuable and constructive comments. We greatly appreciate the time and effort they devoted to reviewing our work. We are pleased that the assessment of our work was generally positive and that our findings are rated as novel and of significant interest. At the same time the Reviewers had some suggestions to improve the interpretation of our data and for additional discussion of in vivo relevance. We addressed all of these suggestions as detailed in our responses below. Responding to the Section Managing Editor’s comment we have reduced the percentage of self- citations from 27 to 19.29%.

However, several aspects require clarification and improvement:

Response: The Discussion section has been expanded with potential interference of other lysine-dependent interactions (alpha2-antiplasmin, TAFI, FXIII-crosslinking).

Second, the mechanistic conclusion that structural alterations of fibrin are the primary driver of the paradoxical effect is not fully established. While enhanced plasminogen activation is excluded, alternative mechanisms such as altered plasmin diffusion, fibrin binding kinetics, or tPA distribution within the clot are not directly investigated.

Response: A discussion of these limitations related to the experimental setting has now been added to the manuscript on p. 14 with 2 relevant references.

The conclusions should therefore be moderated. Third, the definition of “prophylactic” versus “therapeutic” conditions in the experimental setup is somewhat simplified and may not fully reflect in vivo dynamics. This distinction should be clarified and discussed more carefully.

Response: This distinction has now been clarified in the Introduction: “the aim of the present study was to investigate the molecular and structural basis of the TXA effects in prophylactic and therapeutic clinical context applying two in vitro experimental models: 1) a therapeutic scenario mimicked by lysis of fibrin clots with entrapped tPA at the stage of fibrin formation; 2) a prophylactic scenario mimicked by lysis of fibrin clots formed in the absence of tPA and subsequently attacked by extrinsic tPA.”

Fourth, the choice and range of TXA concentrations should be better justified in relation to clinically observed plasma and tissue levels. A more detailed dose–response analysis would strengthen the study.

Response: The relevant section has been added to the manuscript, and an additional citation has been included. In the Methods section “TXA applied at varying concentrations in the range 0–128 μM (0-20 mg/L, with the upper limit of the concentrations used in our experiments corresponding to a therapeutically relevant TXA concentration in human plasma)[19]” In the Discussion section “Previous studies indicate that substantial inhibition of fibrinolysis occurs at plasma TXA concentrations of approximately 60-95 µM, while concentrations in the range of 30-60 µM may already exert partial antifibrinolytic effects[19]. In our experimental settings we used TXA concentrations ranging from 0 to 128 µM, which are within the therapeutically relevant plasma range in human and remain below the peak concentrations reported in plasma after intravenous TXA administration[19].”

Finally, while the manuscript is generally well written, parts of the Discussion are overly dense and could be streamlined to improve clarity and focus on the novel contributions of the study. Overall, this is a valuable and thought-provoking study, but addressing the points above would significantly strengthen its mechanistic rigor and translational relevance.

Response: The Discussion section of the manuscript has been expanded and revised to improve clarity and readability. In the revised Discussion section we emphasized the limitations related to the in vitro nature of the study.

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for giving me the opportunity to review “Functional aspect of fibrin structure alterations by tranexamic acid in the inhibition of fibrinolysis.” The manuscript addresses a very interesting research topic concerning the action of the one of antifibrinolytic agent which could also have completely opposite effects. However, I have few comments and some issues require clarification before recommending the article for publications.

The introduction lacks a clear statement of the study’s objectives and intended contributions. Furthermore, the rationale for the chosen concentrations of TXA remains unclear, as the manuscript does not relate them to pharmacologically relevant levels. Therefore it is difficult to interpret whether the relatively high concentrations mentioned by the authors in the introduction, showing different effects, were included in the study. Does the elucidation of the molecular mechanisms underlying TXA's action apply to both low concentrations and much higher concentrations?
Regarding the choice of methods and conditions of experiments: It is curious that the authors did not perform experiments in more physiological environmental, in the presence of red blood cells. They even showed in the previous article (J Thromb Haemost. 2024; 22:794-804) that antifibrinolytic potency of TXA was increased in the presence of red blood cells. Did I understand correctly that the all experiments were conducted exclusively in a fully artificial medium? If so, could the authors provide a detailed description of the experimental conditions, including the composition of the system (for example: ionic strength or the presence of albumin) as well as key parameters such as temperature and pH?”.
Why only relatively high concentrations of thrombin (mainly 80 nM) were used? It is well documented that thrombin concentration significantly determines the structure, stability and kinetics of fibrin clot formation [Blood Rev. 2007;21,131 - doi: 10.1016/j.blre.2006.11.001; Transfus Apher Sci. 2008;38,15 - doi: 10.1016/j.transci.2007.12.005.; Biomolecules2024, 14, 230 - https://doi.org/10.3390/biom14020230]. High thrombin concentrations favour the formation of dense, rigid networks of thin fibres, characterised by increased resistance to fibrinolysis, whereas low concentrations lead to the formation of more porous structures composed of thick fibres, which are significantly more susceptible to degradation. In fact, extremely low concentration of thrombin (<1 nM or <0.1 U/ml) are sufficient to cleave fibrinopeptides and catalyse fibrin polymerization.

This study presented the effect of TXA depending on the concentration of plasminogen, but it is not known whether a similar relationship may occur in the case of thrombin, which is responsible for the formation of fibrin. Fibrin is a key cofactor in the conversion of plasminogen to plasmin, accelerating this process by acting as a binding surface for both plasminogen and tissue plasminogen activator (tPA). Even if only high concentrations of thrombin were used, it would be worthwhile to discuss the effect of the amount of added thrombin on the obtained results. I appreciate that the effect of TXA on clot structure was taken into account in the discussion, but I am also surprised that the issue of the amount of thrombin used in the research models was omitted.

Do the asterisks shown in Figure 1C really indicate statistical differences between the lysis in the absence and presence of TXA? Considering the number of cases (n) and the differences between the mean (median) values presented in the figure, I rather doubt it and assume that it is an error in the graphics. The quality of panels B-D in Figure 1 would need improvement.
The authors showed that the A_max values in Figure 3 are less consistent, and while they try to explain the increase in maximum absorbance with high plasminogen concentrations, they ignore the lower A_max values for the lowest plasminogen concentration (or the increase for TXA128). What is the reason? What was the maximum value of A_max recorded in the absence of plasminogen? Additionally, Table 1 and Figure 3 present largely overlapping data. To improve clarity and avoid redundancy, the authors are encouraged to either retain only one of these representations or combine them into a single, more comprehensive presentation.

Overall, the manuscript presents an interesting study and has the potential to make a valuable contribution; however, it requires minor revisions, as mentioned above. Certainly, some figures require improvement to enhance their clarity and quality. The Discussion section, although generally well-written, contains several overly elaborate passages that appear more appropriate for a review article than for a focused discussion of the presented results. The authors should consider shortening some sections to improve coherence and relevance. Furthermore, the manuscript would benefit from a more explicitly acknowledgment the limitations of the study, which are currently not insufficiently addressed in the concluding section.

Minor comments: It is worth calculating TXA concentrations more precisely in discussion section (page 15) - taking into account the molar weight of this compound (157.21 g/mol), the concentration of 10-20 mg/mL corresponds more to value 64-128 mM (not 75-150), and could also explain the choice of concentrations used in this study.

Please specify which summary statistics are presented (e.g., median, mean). It would also be helpful to include a brief summary of the statistical analysis in the Methods section.

Author Response

The introduction lacks a clear statement of the study’s objectives and intended contributions. Furthermore, the rationale for the chosen concentrations of TXA remains unclear, as the manuscript does not relate them to pharmacologically relevant levels. Therefore it is difficult to interpret whether the relatively high concentrations mentioned by the authors in the introduction, showing different effects, were included in the study. Does the elucidation of the molecular mechanisms underlying TXA's action apply to both low concentrations and much higher concentrations?

Response: A justification for the therapeutically relevant TXA concentrations used in our experiments has been added to the relevant sections of the manuscript: In the Methods section “TXA applied at varying concentrations in the range 0–128 μM (0-20 mg/L, with the upper limit of the concentrations used in our experiments corresponding to a therapeutically relevant TXA concentration in human plasma)[19]” In the Discussion section “Previous studies indicate that substantial inhibition of fibrinolysis occurs at plasma TXA concentrations of approximately 60-95 µM, while concentrations in the range of 30-60 µM may already exert partial antifibrinolytic effects[19]. In our experimental settings we used TXA concentrations ranging from 0 to 128 µM, which are within the therapeutically relevant plasma range in human and remain below the peak concentrations reported in plasma after intravenous TXA administration[19].”.

The objectives of the study have been described in greater detail in the Introduction section: “the aim of the present study was to investigate the molecular and structural basis of the TXA effects in prophylactic and therapeutic clinical context applying two in vitro experimental models: 1) a therapeutic scenario mimicked by lysis of fibrin clots with entrapped tPA at the stage of fibrin formation; 2) a prophylactic scenario mimicked by lysis of fibrin clots formed in the absence of tPA and subsequently attacked by extrinsic tPA.”.

Regarding the choice of methods and conditions of experiments: It is curious that the authors did not perform experiments in more physiological environmental, in the presence of red blood cells. They even showed in the previous article (J Thromb Haemost. 2024; 22:794-804) that antifibrinolytic potency of TXA was increased in the presence of red blood cells. Did I understand correctly that the all experiments were conducted exclusively in a fully artificial medium? If so, could the authors provide a detailed description of the experimental conditions, including the composition of the system (for example: ionic strength or the presence of albumin) as well as key parameters such as temperature and pH?”.

Response: We agree with the reviewer that the presence of cellular components, including erythrocytes, leukocytes, and platelets, would likely affect our experimental system and could influence the observed effects. As now explicitly formulated in the Introduction, our aim was to examine the apparently contradictory effects of TXA in a purified system, thereby allowing these effects to be assessed under controlled conditions and with minimal confounding from additional blood components. Experiments were conducted using purified proteins under physiologically relevant conditions (pH 7.4), in the absence of additional plasma proteins such as albumin as described in the Materials and Methods section. However, our findings also indicate that further studies in more complex models, closer to physiological conditions, will be necessary to determine how these cellular factors modify the effects observed here. This will be an important subject for future investigation. In response to the Reviewer’s comment, we have added a section to the Discussion addressing the potential contribution of the above-mentioned cellular factors.

Why only relatively high concentrations of thrombin (mainly 80 nM) were used? It is well documented that thrombin concentration significantly determines the structure, stability and kinetics of fibrin clot formation [Blood Rev. 2007;21,131 - doi: 10.1016/j.blre.2006.11.001; Transfus Apher Sci. 2008;38,15 - doi: 10.1016/j.transci.2007.12.005.; Biomolecules2024, 14, 230 - https://doi.org/10.3390/biom14020230]. High thrombin concentrations favour the formation of dense, rigid networks of thin fibres, characterised by increased resistance to fibrinolysis, whereas low concentrations lead to the formation of more porous structures composed of thick fibres, which are significantly more susceptible to degradation. In fact, extremely low concentration of thrombin (<1 nM or <0.1 U/ml) are sufficient to cleave fibrinopeptides and catalyse fibrin polymerization.

Response: We agree with the reviewer that thrombin concentration is an important determinant of fibrin structure and its susceptibility to lysis which we have also investigated in several studies (for example, Blood 2011;117(2):661-8). However, the main goal of our current study (i.e. to compare lysis with intrinsic and extrinsic tPA application) limited our options to vary the thrombin concentration. The relatively high thrombin concentration was necessary to prevent any extensive loss of fibrinogen in the clotting stage of the intrinsic assay, where tPA and plasminogen are already intermixed with fibrinogen when clotting is triggered by thrombin.

Do the asterisks shown in Figure 1C really indicate statistical differences between the lysis in the absence and presence of TXA? Considering the number of cases (n) and the differences between the mean (median) values presented in the figure, I rather doubt it and assume that it is an error in the graphics. The quality of panels B-D in Figure 1 would need improvement.

Response: Figure 1 was improved, and an erroneous asterisk was removed from panel 1C. Paired set of measurements, namely distance run (µm; n = 4), was analyzed using the Mann–Whitney U test and the indicated pairwise comparisons showed statistically significant differences.

The authors showed that the A_max values in Figure 3 are less consistent, and while they try to explain the increase in maximum absorbance with high plasminogen concentrations, they ignore the lower A_max values for the lowest plasminogen concentration (or the increase for TXA128). What is the reason? What was the maximum value of A_max recorded in the absence of plasminogen? Additionally, Table 1 and Figure 3 present largely overlapping data. To improve clarity and avoid redundancy, the authors are encouraged to either retain only one of these representations or combine them into a single, more comprehensive presentation.

Response: Although Fig. 3 and Table 1 present results from the same series of measurements, we think that both of them are needed, because they illustrate different aspects of the assay. The figure shows the original traces of the assay and such primary data are essential to appreciate the nature of the assay and the source of the secondary data. The table allows for presentation of the numeric values extracted from the primary data and their statistical evaluation. Importantly, in the Table now we have added the analysis of the Amax in the absence of TXA that shows an independent, statistically significant thickening effect of plasminogen on fibrin fibers (according to Kruskal-Wallis test). At the same time, an independent method (Fig. 5) shows an isolated thickening effect of TXA too. Because TXA and plasminogen compete for the same binding sites of fibrin, and in addition TXA binds plasminogen, at low plasminogen, the independent thickening effect of TXA is seen, whereas high concentration, plasminogen blocks TXA and thus, their combined action reflects the relative strength of their effects.

Response: The Discussion section has been revised, and the limitations of the study are now more clearly explained. Because Reviewer 1 requested a more detailed discussion of physiological factors not represented in our system, we were unable to substantially shorten the Discussion. However, we reorganized and refined it to improve clarity and focus.

Response: The corresponding molar concentrations have been specified more precisely. As noted in the manuscript, the concentrations used in the cited study remain markedly supraphysiologic, being approximately 170–350-fold higher than reported peak plasma levels after intravenous administration and roughly 1000-fold higher than the commonly cited antifibrinolytic target range of 10–15 mg/L. Indeed, this justifies the concentration range applied in our current study.

Please specify which summary statistics are presented (e.g., median, mean). It would also be helpful to include a brief summary of the statistical analysis in the Methods section.

Response: Now, all statistical methods have been summarized in a separate section of the Methods, and corrected in the Figure Legends.