Subpopulations of Organoid-Forming Cells Have Different Motility

Round 1

Reviewer 1 Report

In the manuscript by Jimenez et al. the authors analyzed organoid-formation of OPSCC cells using a specialized microscope.

Overall, the quality of the figures and significance of the content is quite low. The pictures from the analyses software of the microscope are very pixelated and the provided scale-bar is quite counter-intuitive since a progressive gradient would make much more sense, but instead it appears to be a heatmap. Fig. 1 shows a more complex sphere formation assay of (apparently) a single experiment performed in 1 replicate. Using Fig. 2 the authors aim to show some organoid behave different than others. This is no surprising finding and given that only 5 organoids were analyzed (and the experiment was performed once?) any conclusions drawn from this experiments are highly underpowered and should be thoroughly verified. In addition, it is not depicted which organoids on the left side correspond to the graphs on the right side. Lastly, the authors performed a FACS-sorting and generated 3 population which they analyzed for what they claim is motility. Judging from the presented data the authors did instead analyze sphere formation again. Hence, the claim that different motilities are mediated by different sub-population cannot be drawn.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript was prepared very well. The introduction section justifies the purpose of the study. I congratulate the authors for the preparation of the manuscript

However, I have the following comments:

Introduction

Line 42: You must provide some data from existing OPSCC "in vivo" or "in vitro" models. Also, you could explain some of the limitations that might justify the creation of your model.

Materials and Methods

The methodology is correct and very well described. However, why do they choose CD44 and NGRR in those three phenotypes? add some information that allows the reader to understand it

Results

The results are excellently described and meet the objectives set. The figures are representative of the main findings and contain critical information. However, the figure captions are somewhat confusing.

Figure 1a: indicate in figure 1a if it has been used using digital holography microscopy

Figure 1c: Indicate which are the two different densities

Figure 2a: explain the abbreviation (DHM). Indicate the different time points.

Figure 2b: Do organoids 1-5 make any difference?

Figure 3: explain all abbreviations.

Figure 3b: Explains 5 single cells

Discussion

Line 150: provide an adequate reference, after of tumor cells.

Line 150: provides what the quantitative methods are and why they are scarce. Besides, indicate any advantages and disadvantages of these.

You must indicate that you share your model with others, please incorporate some reference

Could your model be applied to generate with other cell lines models in different types of cancer? please explain briefly

Could it be applied immediately in drug tests, or does it need to be fine-tuned according to the drugs used? Explain this situation, because if so, it would be a strong point in your discussion

Please include a list of limitations of the study

Comments for author File: Comments.docx

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript entitled: „Sub-populations of organoid-forming cells have different motility“ presents a method for evaluation of the cancer-based organoids obtained from single-cells suspension by measuring optical properties, by using Digital Holographic microscopy. Here is introduced a new approach to study the evaluation of the organoid initiation and formation in real-time, with a HoloMonitor M4 instrument.

Firstly, the authors reported on the workflow to choose the best pathway to obtain organoids, the conditions on the use of holographic phase-contrast microscopy, the flow cytometry, and phenotypic study. Then, there are presented the results obtained: the successfully organoid formation from Oropharyngeal Squamous Cell Carcinoma (OPSCC) cells from three different phenotypes: CD44+/NGFR+, CD44+/NGFR- and CD44-/NGFR- in hydrogel and their tracking using HoloMonitor M4 equipment for 24 hours.

Discussion part presented the main results and I suggest to authors to present more discussion about:

- the comparison of the present approach to the other techniques used by other groups to monitorize 3D cell cancer cultures, such as: Moving into a New Dimension: Tracking Migrating Cells with Digital Holographic Cytometry in 3D, Anette Gjörloff Wingren, Cytometry Part A _ 95A: 144–146, 2019.

• The use of Holography microscopy in diverse applications, e.g.: Label-free fingerprinting of tumor cells in bulk flow using inline digital holographic microscopy, DHANANJAY KUMAR SINGH et al, 8, No. 2 | 1 Feb 2017 | BIOMEDICAL OPTICS EXPRESS 536.

Please correct at lines 105-106, the following phrase: “HoloMonitor M4 monitored the formation of organoids up to a volume of approximately 10⁵ um³” replace by “HoloMonitor M4 monitored the formation of organoids up to a volume of approximately 10⁵  μm³”.

In general, the data are strong and convincingly shows the necessity to develop and to present these kinds of studies to determine the early stage formation of a tumor. This manuscript is well written and concise.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Answer to Reviewer 1

We appreciate the comments on our manuscript describing a method of imaging early events of tumor organoid formation. We have updated the manuscript according to the feedback, and provide a point-by-point rely below:  

1. In the manuscript by Jimenez et al. the authors analyzed organoid-formation of OPSCC cells using a specialized microscope.
2. Overall, the quality of the figures and significance of the content is quite low.
3. The pictures from the analyses software of the microscope are very pixelated

About the quality of the images, it is not possible to increase it using the holomonitor M4, and the pixelation-effect comes after applying a light layer onto the image, which shows the roughness of the object’s surface.

4. the provided scale-bar is quite counter-intuitive since a progressive gradient would make much more sense, but instead it appears to be a heatmap.

We have changed the scaled color bar to a progressive gradient in the figures.

5. 1 shows a more complex sphere formation assay of (apparently) a single experiment performed in 1 replicate.

We have specified the number of replicates included in the different figures. In the case of figure 1, it includes 3 replicates per each of the two conditions.

6. Using Fig. 2 the authors aim to show some organoid behave different than others. This is no surprising finding and given that only 5 organoids were analyzed (and the experiment was performed once?) any conclusions drawn from this experiments are highly underpowered and should be thoroughly verified.
7. In regards of figure 2 we monitored the formation of 5 organoids from 5 individual cells, to show that it is possible to capture the heterogeneity of the culture at the selected seeding density range. It is not depicted which organoids on the left side correspond to the graphs on the right side.

We are including a modified version of figure 2a, which displays the 5 organoids tracked over time in the 3 time points shown. Also, we include an image depecting the movement of the 5 organoids in the XY plane, as the cells aggregate over time.

Anwer2: Judging from the data it rather seems like O5 was moving into a larger clump of cells that was already present from the beginning and therefore the resulting organoid is larger because there were more cells to beginn with rather than different growth kinetics

Unpublished figure 1 (corresponds to Figure 2 in the manuscript). Organoid formation events monitored over 15 hours using Digital Holographic microscopy (DHM). (a) DHM images captured at different time points (0, 7.5 and 15 hours) in a hydrogel embedded culture system. (b) Increase in area and optical volume patterns of 5 organoids tracked over 15 hours at 30 different time points. Organoids 1-5 in figure 2b are circled and numbered as O1-O5 in figure 2a.

Unpublished Figure 2. 2D moment plot of the 5 organoids, tracked over 15 hours, depicted in figure 2. Each coloured line represents the movement of a single organoid as they aggregate, and move in the XY plane from their original position (0,0) at time 0 hours.

8. Lastly, the authors performed a FACS-sorting and generated 3 population which they analyzed for what they claim is motility. Judging from the presented data the authors did instead analyze sphere formation again. Hence, the claim that different motilities are mediated by different sub-population cannot be drawn.

In the case of figure 3, we aim to display that tumor organoid forming cells move different through the extracellular matrix, and therefore we first identify which of the three phenotypes (CD44⁺/NGFR⁺, CD44⁺/NGFR^- and CD44^-/NGFR^-) are able to grow into organoids, and then we compare their motility pattern. Accumulated motility is defined as the movement of cells embedded in the hydrogel over time, since cell migration can be considered as a cellular response to a chemoattractant or repellent, and we do not show that data (lines 80-82).

Anwer2: It is true that double-negative cells largely fail to develop into larger structure, on the other NGF3 seems to be largely dispensable. Nonetheless, is this experiment again largely underpowered with only 2 cells per condition analysed and only one cell line used. Given that both markers have been already described as stemness related I do not see the novelty of this experiment.