Cell softness regulates tumorigenicity and stemness of cancer cells

Abstract Identifying and sorting highly tumorigenic and metastatic tumor cells from a heterogeneous cell population is a daunting challenge. Here, we show that microfluidic devices can be used to sort marker‐based heterogeneous cancer stem cells (CSC) into mechanically stiff and soft subpopulations. The isolated soft tumor cells (< 400 Pa) but not the stiff ones (> 700 Pa) can form a tumor in immunocompetent mice with 100 cells per inoculation. Notably, only the soft, but not the stiff cells, isolated from CD133+, ALDH+, or side population CSCs, are able to form a tumor with only 100 cells in NOD‐SCID or immunocompetent mice. The Wnt signaling protein BCL9L is upregulated in soft tumor cells and regulates their stemness and tumorigenicity. Clinically, BCL9L expression is correlated with a worse prognosis. Our findings suggest that the intrinsic softness is a unique marker of highly tumorigenic and metastatic tumor cells.

1.It would be helpful to have a bit more background on how the various microweir design paramet ers would theoret ically be expect ed to affect st iffness-sort ing.Were these dimensions arrived at in an ad hoc fashion, or did the aut hors do comput at ional design?If so, how well do the experiment s mat ch the predict ions? 2. In Fig EV1 there seems to be disconnect between the text and fig for the Jas result.The text indicates that Jas stiffens cells and produces fewer cells at the outlet.However, the figure is missing any indication that Jas-induced differences are statistically significant.Are symbols missing from the figure, or is the text incorrect?
3. The choice of ANOVA and t-tests for many of the statistical comparisons is puzzling given that the distributions seem so non-Gaussian (e.g.EV1A, EV3E).This is more than an academic detail given the enormous spread in some of these data sets.Could the authors justify this decision, and do the statistical differences hold when tests for non-normal distributions are used (e.g.Mann-Whitney or Kruskal-Wallis)? 4. Do the soft cells remain uniformly soft after culture in the fibrin gels, or are softness distributions re-established?The former would suggest that the softness cues are functionally important for colony formation (and later invasion), whereas the latter would suggest that softness is more of a selection marker that happens to capture the most tumorigenic cells.
5. There appears to be some conceptual overlap from earlier studies involving some of these authors, e.g.Tan et al Nat Comm 5: 4619 (2014).The authors should incorporate this paper more explicitly in their introduction and explain more clearly the new advances here.
6. Based on publicly available tumor sequence databases (e.g.TCGA), is BCL9L expression associated with malignancy and/or poor prognosis?Referee #2: In this study, the investigators explore the potential of cellular softness as a marker of tumorinitiating and metastatic cancer cells.They use a specific microfluidic chip to isolate stiff and soft cancer cells from different malignancies, and this is followed by characterization of the isolated cells in vitro and in vivo.The results indicate that soft cancer cells have tumor initiating capacity and are more metastatic than stiff cells.Interestingly, this is independent of previously identified cancer stem cell (CSC) markers.The authors use transcriptomic analysis to identify BCL9L, a co-factor of beta-catenin and regulator of Wnt signaling, as a functional mediator of aggressivity in soft cancer cells.Finally, the investigators analyze BCL9L expression in clinical cancer samples and identify association between BCL9L and poor patient outcome.This is a very interesting and overall thorough study.However, I do have some critical points that need to be addressed.Details below.
1.The investigators characterize soft cancer cells and demonstrate that this property is an excellent marker of tumor-initiating ability and cancer aggressiveness across different cancer types.Growing evidence indicates that tumor-initiating properties can be surprisingly plastic and context dependent, suggesting that cancer cells may gain or lose these properties depending on the microenvironment.The authors state that physical traits may be more stable than expression of CSC marker genes.With this in mind, I think it would be important to show that different levels of stiffness are indeed stable properties.What happens when selected soft cells are grown for some time in culture?Do they maintain the soft phenotype?What about growth of selected cells in stiff matrix rather than soft matrix?Does it influence the cellular stiffness.
2. Following up on the previous point, is it fair to use soft matrix when colony formation is measured to compare soft and stiff cells?How would the comparison turn out in stiff matrix?Would this now benefit stiff cells?3.In Figure 6B and 6E, the aut hors address correlat ion bet ween tumor grade and the percent age of soft cells or the levels of BCL9L in pat ient samples.However, the graphs present ed are not appropriat e to address this.A simple correlat ion plot wit h the paramet ers on x and y axis would be more appropriat e.For st at ist ical analysis, the invest igat ors use Pearson's correlat ion test and thus the correlat ion coefficient R should be included.

4.
It is int erest ing to see how much bet ter of a marker the cellular soft ness can be compared to previously recognized CSC markers (Figure 3 and EV3).Not ably, a number of the CSC markers (ALDH, Side Populat ion and CD133) were significant ly associat ed wit h increased soft ness.Moreover, the combinat ion of ALDH and soft ness, as markers to isolat e cancer cells, modest ly improved tumor init iat ion in mice compared to soft ness alone (Fig 3B).Not all soft cells express CSC markers and not all CSC marker posit ive cells are soft .Therefore, it would be insight ful to explore the combinat ion of soft ness and CSC markers a bit furt her.Does the combinat ion of soft ness and ot her CSC markers select for more enhanced tumor-init iat ion or met ast at ic abilit y compared to soft ness alone.Furt her analysis of the double posit ive cells (soft + CSC markers) is warrant ed.

5.
The result s present ed in this manuscript should be discussed furt her in the cont ext of the large amount of st udies showing posit ive associat ion bet ween increased tissue st iffness and aggressive cancer behavior.Here the aut hors show that soft cells are more mobile and invasive.However, subst ant ial evidence suggest s that cross-linked collagen mediat ed by LOX family members and leading to increased st iffness is associat ed wit h increased invasion and aggressive cancer behavior.
6.It is not clear to this reviewer what is being compared in Figure 6D.
7. For BCL9L knockout , CRISPR.EFS.GFP posit ive cells were select ed by flow cyt omet ry based on GFP expression.However, if guide RNA against GFP was used as a cont rol, would this not affect select ion of CRISPR.EFS.GFP posit ive cells?8.It is not clear how the met ast asis index is derived.What does it mean specifically?9.There is a discordance bet ween text and Figure 2E legend.The figure legend st at es "(E) the same as (B)", essent ially primary tumor analysis.However, the main text st at es that met ast at ic colonizat ion of the lungs was analyzed.
10. Tot al bet a-cat enin expression is not a good marker of act ive Wnt signaling.Nuclear bet acat enin or a report er are more reliable.

RESPONSES TO REVIEWERS
We would like to express our sincere thanks to the reviewers for their critical and constructive comments.We respond point-by-point to each of their comments and criticisms.We feel that their comments have helped us on substantially improving and strengthening the manuscript and clarifying some issues.

RESPONSE TO REVIEWER #1:
This study presents a microfluidic device for separating cells according to mechanical compliance ("softness") and goes on to show that softer cells are more tumorigenic, with both softness and tumorigenicity being regulated by BCL-9.This is a novel and well-designed study that nicely integrates microdevices, mouse models, analysis of patient samples, and mechanistic culture studies.It goes beyond many past studies in its rigorous molecular characterization of soft vs stiff cells and use of multiple experimental paradigms.My main criticism is that it's not clear that softness is directly responsible for driving differences in malignancy, versus being a more passive biomarker.But either possibility would be significant, and the work is likely to be of great interest in any case.I just have a few minor comments for the authors to address, after which I would recommend publication.I don't view any of my suggested experiments as truly essential to publication.
1.It would be helpful to have a bit more background on how the various microweir design parameters would theoretically be expected to affect stiffness-sorting.Were these dimensions arrived at in an ad hoc fashion, or did the authors do computational design?If so, how well do the experiments match the predictions?Response: We thank the reviewer's pertinent comment.Tumor cells usually have the size around 20-25 μm (Anal Chem.2010;82:6629-35; Int J Cancer.2016;138:2894-904).In our preliminary experiments, we actually tested different size of gaps, including 10 μm, 15 μm and 20 μm.When bulk tumor cells passed through the channel, we found that the 10 μm-microfluidic chips yielded very few soft cells, while 20 μm gap allowed many stiff cells to pass through the channel (unpublished data).Therefore, we designed a 15 μm gap to isolate soft tumor cells.On the other hand, we designed the 18 μm gap to allow soft passing, and isolate stiff tumor cells through inversely washing the channels.
In addition, we selected the 40 μm height of microweir based on previous studies (Proc Natl Acad Sci.2012;109: 18707-18712; J Chromatogr A. 2009;1216: 8289-8295), thus allowing tumor cells to pass through the channel in a single cell form.We also referred to the 11th Sep 2020 1st Authors' Response to Reviewers previous studies and tested different widths of microweir structure, spaces between microweir structures, and total numbers of microweir structures.
According to the reviewer's suggestion, we added the tumor cell size background and the references in the revised manuscript, page 5 line 15.We thank the reviewer's indicating this statistical issue for Fig. EV1.In the revised manuscript, we added the P value between the PBS control and the Jas group.

The choice of ANOVA and t-tests for many of the statistical comparisons is puzzling given that the distributions seem so non-Gaussian (e.g. EV1A, EV3E). This is more than an academic detail given the enormous spread in some of these data sets. Could the authors justify this decision, and do the statistical differences hold when tests for non-normal distributions are used (e.g. Mann-Whitney or Kruskal-Wallis)?
Response: We thank the reviewer for indicating the statistical issue.Accordingly, we did the statistical analysis by using the Mann-Whitney or Kruskal-Wallis method and similar results were obtained.
According the reviewer's suggestion, we revised the corresponding figure legends and material methods, in the revised manuscript, page 34 line 22.

Do the soft cells remain uniformly soft after culture in the fibrin gels, or are softness distributions re-established? The former would suggest that the softness cues are functionally important for colony formation (and later invasion), whereas the latter would suggest that softness is more of a selection marker that happens to capture the most tumorigenic cells. Response:
We thank the reviewer's comments.Stiffness is an inherent feature of a cell, which is mainly provided by F-actin filaments (Science 1993;260:1124-7).Different cell types display varying levels of stiffness, which matches the stiffness of the local extracellular matrix, allowing the cells to properly sense and respond to the surrounding mechanical microenvironments (Science 2009;324:1673-7; Nature Methods 2018;15:491-8).
In the revised manuscript, we isolated soft MCF-7 cells by using the microfluidic chip and then seeded these soft cells in soft 3D fibrin gel for 4, 6, 12, 24, 48, 72 or 96 hr.The cellular stiffness of these cells was detected by AFM.The result showed that during the observation period, these soft MCF-7 cells always kept their mechanical character of being softness, suggesting that stiffness is an intrinsic character of a cell.
According to the reviewer's suggestion, we added this result in the revised manuscript, page 7 line 20 and Fig EV2B . 5.There appears to be some conceptual overlap from earlier studies involving some of these authors, e.g.Tan et al Nat Comm 5: 4619 (2014).The authors should incorporate this paper more explicitly in their introduction and explain more clearly the new advances here.Response: According to the reviewer's suggestion, we added this information in the section of introduction in the revised manuscript, page 3 line 20.
"This suggests that soft cells should survive in a soft stroma and stiff cells function optimally in a stiff niche.As a support, we have demonstrated that 3D soft fibrin matrices promote H3K9 demethylation and increase Sox2 expression and self-renewal of melanoma stem cells, whereas stiff ones exert opposite effects (Nat Commun.2014;5:4619).In line with this notion, cells with various degrees of stiffness can co-exist within the same tumor tissue, due to the heterogeneity of the tumor mechanical microenvironments (Nat Nanotechnol.2012;7:757-65; Nat Mater.2014;13:631-7).Despite such understanding, direct evidence that the cellular softness functions as a basic feature for tumorigenic cells remains elusive.In this study, we develop a method to separate soft cells from stiff ones and provide evidence that these soft cells are highly tumorigenic and possess the ability to metastasize".

Based on publicly available tumor sequence databases (e.g. TCGA), is BCL9L expression associated with malignancy and/or poor prognosis?
Response: We thank the reviewer's comment.In our original manuscript, we indeed investigated whether BCL9L expression is associated with patients' prognosis, and found that the higher expression of BCL9L in tumor tissues was correlated with a poorer prognosis of patients with melanoma, breast cancer and other cancer types such as pancreatic cancer, lung cancer and liver cancer (please see original Fig 6F and EV5J).

RESPONSE TO REVIEWER #2:
Referee #2: In this study, the investigators explore the potential of cellular softness as a marker of tumor-initiating and metastatic cancer cells.They use a specific microfluidic chip to isolate stiff and soft cancer cells from different malignancies, and this is followed by characterization of the isolated cells in vitro and in vivo.The results indicate that soft cancer cells have tumor initiating capacity and are more metastatic than stiff cells.Interestingly, this is independent of previously identified cancer stem cell (CSC) markers.The authors use transcriptomic analysis to identify BCL9L, a co-factor of beta-catenin and regulator of Wnt signaling, as a functional mediator of aggressivity in soft cancer cells.Finally, the investigators analyze BCL9L expression in clinical cancer samples and identify association between BCL9L and poor patient outcome.This is a very interesting and overall thorough study.However, I do have some critical points that need to be addressed.Details below.

The investigators characterize soft cancer cells and demonstrate that this property is an excellent marker of tumor-initiating ability and cancer aggressiveness across different cancer types. Growing evidence indicates that tumor-initiating properties can be surprisingly plastic and context dependent, suggesting that cancer cells may gain or lose these
properties depending on the microenvironment.The authors state that physical traits may be more stable than expression of CSC marker genes.With this in mind, I think it would be important to show that different levels of stiffness are indeed stable properties.What happens when selected soft cells are grown for some time in culture?Do they maintain the soft phenotype?What about growth of selected cells in stiff matrix rather than soft matrix?Does it influence the cellular stiffness.

Response:
We thank the reviewer's pertinent comments on cancer cells gaining or losing tumor-initiating properties depending on the microenvironment.Stiffness is an inherent feature of a cell, which is mainly provided by F-actin filaments (Science 1993;260:1124-7).Different cell types display varying levels of stiffness, which matches the stiffness of the local extracellular matrix, allowing the cells to properly sense and respond to the surrounding mechanical microenvironments (Science 2009;324:1673-7; Nature Methods 2018;15:491-8).
In the revised manuscript, we isolated soft MCF-7 cells by using the microfluidic chip and cultured the cells in soft 3D fibrin gel or stiff flask for 4, 6, 12, 24, 48, 72 and 96 hr.Stiffness of the cells was determined by atomic force microscopy (AFM).We found that the softness of the isolated soft MCF-7 cells could be kept in soft fibrin gels, which, however, started to become stiff following 4-hour culture in stiff flask, and reached the peak 12 hours later.
According to the reviewer's suggestion, we added this result in the revised manuscript, page 7 line 20 and Fig EV2B and C.

Following up on the previous point, is it fair to use soft matrix when colony formation is measured to compare soft and stiff cells? How would the comparison turn out in stiff matrix? Would this now benefit stiff cells?
Response: Seeding the soft MCF-7 cells isolated by the microfluidic chip in 90 Pa soft 3D fibrin gels could keep the softness of the cells, however, seeding soft tumor cells into 1050 Pa stiff 3D fibrin gels induced cells into dormancy (Cancer Res.2018;78 (14):3926-3937).In contrast to soft MCF-7 cells, if seeding the stiff MCF-7 cells into 90 Pa soft 3D fibrin gels, the formed colony number decreased dramatically and displayed a differentiated morphology (please see original Figure 2A); however, such stiff cells grew well in rigid plate.Thus, only the soft cells can grow colonies in soft fibrin gels but stiff ones cannot.This means that we can use colony formation in soft 3D fibrin gels to further verify the softness of the isolated soft tumor cells through microfluic chip.

Response:
We thank the reviewer's comment.Accordingly, we presented the graphs for Figure 6B and 6E with the "Soft cells (%)" or "Relative BCL9L level" on x axis and "Grade of cancer" on y axis, and also included the correlation coefficient R in the revised manuscript.

It is interesting to see how much better of a marker the cellular softness can be compared to previously recognized CSC markers (Figure 3 and EV3). Notably, a number of the CSC markers (ALDH, Side Population and CD133) were significantly associated with increased softness. Moreover, the combination of ALDH and softness, as markers to isolate cancer cells, modestly improved tumor initiation in mice compared to softness alone (Fig 3B). Not all soft cells express CSC markers and not all CSC marker positive cells are soft. Therefore, it would be insightful to explore the combination of softness and CSC markers a bit further. Does the combination of softness and other CSC markers select for more enhanced tumor-initiation or metastatic ability compared to softness alone. Further analysis of the double positive cells (soft + CSC markers) is warranted. Response:
We appreciate the reviewer's comment.In our unpublished data, we had sorted the soft cells from bulk 4T1 and ALDH + 4T1 cells, respectively, by using microfluidic chip.Then, we injected these different soft cells into the mammary fat pads of NSG mice (100 cells/mouse).Twelve weeks later, the tumor formation was recorded.The result showed that the soft 4T1 cells and the soft ALDH + 4T1 cells had the similar tumorigenic capability (5/10 versus 6/10).Thus, the intrinsic softness is a suitable characteristic to mark tumorigenic cells.
According to the reviewer's concern, we added this result in the revised manuscript, page 9 line 17 and Fig EV3B.

The results presented in this manuscript should be discussed further in the context of the large amount of studies showing positive association between increased tissue stiffness and aggressive cancer behavior.
Here the authors show that soft cells are more mobile and invasive.However, substantial evidence suggests that cross-linked collagen mediated by LOX family members and leading to increased stiffness is associated with increased invasion and aggressive cancer behavior.

Response:
We thank the reviewer's constructive comment.Accordingly, we discussed the positive association between increased tissue stiffness and aggressive cancer behavior as below: "Published reports have shown a positive association between increased tissue stiffness and aggressive cancer behavior, prompting to propose a model of cancer progression that depends on static or dynamic tumor tissue stiffening (Proc Natl Acad Sci. 2019;116:3502-3507;Cancer cell. 2005;8:241-254;Nat Cell Biol. 2015;17:678-688).However, there has been no evidence to support the notion that the cells coming out of the stiffened tumor stroma are tumorigenic and metastatic cancer cells.It has also been unclear whether the tumorigenic cells are stiff cells or soft cells.Despite the overall stiffness, local microenvironments for tumor stiffness are highly heterogeneous (Nat Nanotechnol.2012;7:757-65).Increased tissue stiffness may be attributed to more extracellular matrices, which are likely to limit blood vessel distribution and lead to tumor hypoxia.a common phenomenon in tumor microenvironments.It is known that hypoxia in primary tumors is associated with an increased metastasis and a worse prognosis in cancer patients (Nature. 2006;440:1222-6;Science. 2016;352:175-80;Nat Rev Cancer. 2014;14:430-9).Recently, we demonstrated that hypoxia promote human breast tumor-repopulating cell development (Oncogene.2019;38:6970-6984).Of note, hypoxic areas may be very soft due to local tissue necrosis and matrix degradation.Thus, increased tissue stiffness may result in more soft tumor cells at the hypoxic sites, favoring an aggressive cancer behavior.Consistently, studies by Superfine et al showed that cancer cells with the highest migratory and invasive potential are five times less stiff than cells with the lowest migration and invasion potential (Cancer Res.2011;71:5075-80)".
According to the reviewer's suggestion, we added this information in the revised manuscript, page 15 line 8. 6D.

Response:
We thank the reviewer's indicating this detail.In original Figure 6D, we stained the tissue sections from 6 high-grade melanoma of patients with anti-BCL9L antibody.The result showed that BCL9L was highly expressed in melanoma tissues.As a comparison, 3 patients' low-grade melanoma were stained and the result showed a decreased expression of BCL9L.
According to the reviewer's concern, we added the result from the 3 low-grade melanoma samples in the revised manuscript, revised Fig 6D .7. For BCL9L knockout, CRISPR.EFS.GFP positive cells were selected by flow cytometry based on GFP expression.However, if guide RNA against GFP was used as a control, would this not affect selection of CRISPR.EFS.GFP positive cells?Response: We thank the reviewer's indicating this error.In the revise manuscript, we corrected the original sentence with "These SGRNAs were cloned into the pL-CRISPR.EFS.RFP vector plasmid (addgene, #57819)" and replaced "GFP-positive cells were sorted by flow cytometry" with "RFP-positive cells were sorted by flow cytometry".

It is not clear how the metastasis index is derived. What does it mean specifically?
Response: We thank the reviewer's indicating this detail.The metastasis index was defined as the percentage of metastasis tumor areas to total lung areas based on calculation from 10 slides, according to the protocol we published before (Cancer Immunol Res.2018;6:1046-1056).
According to the reviewer's concern, we added this information in the revised manuscript, page 37 line 2. 9.There is a discordance between text and Figure 2E legend.The figure legend states "(E) the same as (B)", essentially primary tumor analysis.However, the main text states that metastatic colonization of the lungs was analyzed.

Response:
We thank the reviewer's indicating this error.In the revised manuscript, we revised the figure legend for Figure 2E as "Stiff or soft B16 or MP-1 cells (100 or 10 cells) were injected into the NSG mice by tail veil".

Response:
We thank the reviewer's constructive comments.Accordingly, we isolated the total nuclear proteins from stiff or soft 4T1, MCF-7, B16 or MP-1 cells and determined the expression of β-catenin by western blot.The result showed that the expression of nuclear β-catenin was increased in the soft tumor cells, which was consistent with the result from immunostaining of β-catenin (please see original Figure EV5C).
According to the reviewer's concern, we added this result in the revised manuscript, page 1st Oct 2020 1st Revision -Editorial Decision Thank you for submit ting a revised version of your manuscript .Your st udy has now been seen by bot h original referees, who find that their main concerns have been addressed and support publicat ion of the revised manuscript .There now remain only a few mainly edit orial issues that have to be addressed before I can ext end formal accept ance of the manuscript : 1. Please address the remaining minor point from reviewer #1.
Please let me know if you have any further questions regarding any of these points.You can use the link below to upload the revised files.
Thank you again for giving us the chance to consider your manuscript for The EMBO Journal.I look forward to receiving the final version.
- -----------------------------------------------Referee #1: In general, the aut hors have responded well to my comment s.My only remaining comment has to do wit h their response to my comment #4 and reviewer 2's comment #1 regarding the st abilit y of cell soft ness values.The dat a in EV2B-C nicely address the concern, but the associat ed text is unclear as writ ten and should be revised ("Meanwhile, we found that the soft ness of the isolat ed soft MCF-7 cells could be kept in soft fibrin gels, which, however, st art ed to become st iff following 4hour cult ure in convent ional two-dimensional (2D) rigid dishes, and reached the peak 12 hours lat er (Fig EV2B 3 and C)").That minor concern aside, the work seems ready for publicat ion.
Referee #2: The invest igat ors have made significant revisions to the manuscript and my crit ical point s have been thoroughly addressed.I do not have furt her comment s on this int erest ing st udy.

RESPONSES TO THE EDITOR AND REVIEWERS
We would like to express our sincere thanks to you and both reviewers for their critical and constructive comments.We respond point-by-point to their comments.We feel that their comments have helped us on improving and strengthening the manuscript.

RESPONSE TO THE EDITOR
1. Please address the remaining minor point from reviewer #1.

Response:
We appreciate the reviewer's comment.Accordingly, in the revised manuscript, we revised the sentence as below: Meanwhile, we found that the isolated soft MCF-7 cells could maintain their softness in the soft fibrin gels, while, the soft cells, if seeded in rigid culture plates for 4 hours, started to become stiff and reached the stiff peak 12 We set the sample size according to our previously published articles and guide of the Animal Center of Chinese Academy of Medical Sciences.
We dose not set up any inclusion/exclusion criteria, and have not deleted any samples and data.
Yes. Age-and sex-matched mice were selected and randomly assigne to different groups.
Yes.For identification of a normal distribution，we use the D'Agostino-Pearson omnibus normality test to detect the normal distribution of the data.If the P-Value of the D'Agostino-Pearson omnibus normality test is larger than 0.1, we assume a normal distribution.
Mice experiments were age and sex-matched and randomized.
The histology analysis of mice lung sections, the wound-healing assay, cell migration assay and cell invasion assays were performed blindly.
The mice lung tissue H&E staining and the pathological evaluation were performed blindly.a statement of how many times the experiment shown was independently replicated in the laboratory.

Data
Any descriptions too long for the figure legend should be included in the methods section and/or with the source data.
In the pink boxes below, please ensure that the answers to the following questions are reported in the manuscript itself.Every question should be answered.If the question is not relevant to your research, please write NA (non applicable).We encourage you to include a specific subsection in the methods section for statistics, reagents, animal models and human subjects.
definitions of statistical methods and measures: a description of the sample collection allowing the reader to understand whether the samples represent technical or biological replicates (including how many animals, litters, cultures, etc.).

A-Figures
Reporting Checklist For Life Sciences Articles (Rev.June 2017) This checklist is used to ensure good reporting standards and to improve the reproducibility of published results.These guidelines are consistent with the Principles and Guidelines for Reporting Preclinical Research issued by the NIH in 2014.Please follow the journal's authorship guidelines in preparing your manuscript.

PLEASE NOTE THAT THIS CHECKLIST WILL BE PUBLISHED ALONGSIDE YOUR PAPER
Journal Submitted to: The EMBO Journal Corresponding Author Name: Yuying Liu and Bo Huang

YOU MUST COMPLETE ALL CELLS WITH A PINK BACKGROUND ê
Is there an estimate of variation within each group of data?
Is the variance similar between the groups that are being statistically compared?6.To show that antibodies were profiled for use in the system under study (assay and species), provide a citation, catalog number and/or clone number, supplementary information or reference to an antibody validation profile.e.g., Antibodypedia (see link list at top right), 1DegreeBio (see link list at top right).
7. Identify the source of cell lines and report if they were recently authenticated (e.g., by STR profiling) and tested for mycoplasma contamination.
* for all hyperlinks, please see the table at the top right of the document 8. Report species, strain, gender, age of animals and genetic modification status where applicable.Please detail housing and husbandry conditions and the source of animals.9.For experiments involving live vertebrates, include a statement of compliance with ethical regulations and identify the committee(s) approving the experiments.
Fig EV1 there seems to be disconnect between the text and fig for the Jas result.The text indicates that Jas stiffens cells and produces fewer cells at the outlet.However, the figure is missing any indication that Jas-induced differences are statistically significant.Are symbols missing from the figure, or is the text incorrect?Response: -hour later (FigEV2B and C).
Do the data meet the assumptions of the tests (e.g., normal distribution)?Describe any methods used to assess it.

data shown in figures should satisfy the following conditions:
the data were obtained and processed according to the field's best practice and are presented to reflect the results of the experiments in an accurate and unbiased manner.figurepanelsinclude only data points, measurements or observations that can be compared to each other in a scientifically meaningful way.Source Data should be included to report the data underlying graphs.Please follow the guidelines set out in the author ship guidelines on Data Presentation.Please fill out these boxes ê (Do not worry if you cannot see all your text once you press return) a specification of the experimental system investigated (eg cell line, species name).Sample sizes were determined on the basis of estimates from preliminary experiments to conducted reasonable statistical analyses.graphs include clearly labeled error bars for independent experiments and sample sizes.Unless justified, error bars should not be shown for technical replicates.if n< 5, the individual data points from each experiment should be plotted and any statistical test employed should be justified the exact sample size (n) for each experimental group/condition, given as a number, not a range;

Each figure caption should contain the following information, for each panel where they are relevant: 2. Captions B-Statistics and general methods the
assay(s) and method(s) used to carry out the reported observations and measurements an explicit mention of the biological and chemical entity(ies) that are being measured.an explicit mention of the biological and chemical entity(ies) that are altered/varied/perturbed in a controlled manner.