Summary

Integrative Toolkit to Analyze Cellular Signals: Forces, Motion, Morphology, and Fluorescence

Published: March 05, 2022
doi:

Summary

The Integrative Toolkit to Analyze Cellular Signals (iTACS) platform automates the process of simultaneously measuring a wide variety of chemical and mechanical signals in adherent cells. iTACS is designed to facilitate community-driven development and enable researchers to use all platform features regardless of their educational background.

Abstract

Quantitative assessment of cellular forces and motion advanced considerably over the last four decades. These advancements provided the framework to examine insightful mechanical signaling processes in cell culture systems. However, the field currently faces three problems: lack of quality standardization of the acquired data, technical errors in data analysis and visualization, and perhaps most importantly, the technology remains largely out of reach for common cell biology laboratories. To overcome these limitations, we developed a new experimental platform – Integrative Toolkit to Analyze Cellular Signals (iTACS). iTACS consists of two components: Acquisition and Training Module (AcTrM) and Analysis and Visualization Module (AnViM). AcTrM is based on µManager – an NIH-ImageJ-based microscope control software – and facilitates user self-training and automation of common image acquisition protocols. AnViM is based on NIH-ImageJ and facilitates user-friendly automation of data analysis and insightful visualization of results. These experiments involve culturing adherent cells on hydrogels, imaging fiducial markers embedded in the hydrogel, and finally extracting from these images a comprehensive mechanical characterization of the cells. Currently, iTACS enables the user to analyze and track a wide array of properties, including morphology, motion, cytoskeletal forces, and fluorescence of individual cells and their neighboring region. The quality standardization issue was addressed in AcTrM with, a reference image-guided refocusing technique. The technical issues in data analysis were addressed in AnViM with a multi-pronged image segmentation procedure, a user-friendly approach to identify boundary conditions, and a novel cellular property-based data visualization. AcTrM is designed to facilitate the straightforward transformation of basic fluorescence microscopes into experimental cell mechanics rigs, and AnViM is equipped to enable users to measure cellular mechanical signals without requiring an engineering background. iTACS will be available to the research community as an open-source suite with community-driven development capabilities.

Introduction

Commonly used optical imaging and data analysis tools employ hardware and software technologies that are nearly antiquated. The lag in translation and implementation of advances in electronic devices, computational approaches, and mathematical analysis into common experimental cell biology tools is a major constraint on the pace of growth in our knowledge of cellular physiology. Currently, cell biology researchers find molecular biology tools within the reach, but tools based on engineering principles to be out of reach. One such engineering-principles-based tool is Monolayer Stress Microscopy (MSM)1,2. While MSM has been adapted and studied in various laboratories worldwide, its use is primarily confined to labs with engineering expertise3,4,5,6,7,8,9.

NIH-ImageJ is one of the most popular open-source tools among cell biology researchers10. User community contribution-driven advancements have been central to its popularity11,12. ImageJ has features that allow users to develop applications with a mix of an advanced programming language and simplified scripting approaches. These features facilitate users with basic programming knowledge to implement, adapt, and advance any new contribution to the software. Building on these qualities of NIH-ImageJ, we have developed the Integrative Toolkit to Analyze Cellular Signals (iTACS), which enables a low-cost integration of desired hardware and software tools to automate the measurement of a wide variety of chemical and mechanical signals across adherent cells11,12.

iTACS comprises two components: Acquisition and Training Module (AcTrM) and Analysis and Visualization Module (AnViM). AcTrM is built on µManager – an NIH-ImageJ-based image acquisition application – to enable users to set up time-lapse measurements of traditional optical properties and a variety of physical properties of adherent cells in multiple samples12. AcTrM facilitates user training through concise directions included in the graphical interface. In addition, it has a novel feature of reference image-based autofocusing that is designed to facilitate real-time measurements of physical forces and enable quality standardization of the acquired data.

AnViM is built on ImageJ plugins, expedited software, and file handling scripts that enable users to quantitatively assess more than 50 properties, including cellular shape, size, orientation, speed and direction of motion, tractions exerted on the extracellular matrix (ECM), and on neighboring cells, contractile and shear moments of both individual adherent cells and their neighboring region. AnViM facilitates users to quantify the cellular physical properties without mastering the underlying technical background11. Furthermore, it enables data analysis in an interactive or batch processing mode. It generates heat maps revealing spatial variation and graphs showing the temporal variation of the properties of individual cells.

In a typical experiment, the user cultures cells on an elastic hydrogel with appropriate extracellular matrix proteins on the top surface and two types of embedded fluorescent markers. Essentially, images of these fluorescent markers before and after culturing the cells are sufficient to quantify the forces within and around individual cells2,13. AnViM maps these results onto individual cells of the adherent cluster and generates insightful images and graphs.

Protocol

NOTE: Samples examined using the iTACS platform are cells adhered to a soft substrate. The protocol for assessing mechanical and chemical signals is divided into two sequential parts: Acquisition and Training Module (AcTrM) and Analysis and Visualization Module (AnViM).

1. Acquisition and Training Module (AcTrM)

NOTE: AcTrM automates the process of data acquisition and user self-training. Before any data acquisition, prepare a soft substrate capable of providing information necessary to quantify forces that cells exert on it.

  1. Hydrogel preparation
    NOTE: The goal here is to prepare a 1,250 Pa shear modulus, approximately 100 µm thickness, and 22 mm diameter polyacrylamide (PAA) hydrogel.
    1. Prepare polyacrylamide solution by following the method of Yeung et al. and cast the hydrogels following steps described by Trepat et al.14,15. One exception to the procedure is that the beads embedded immediately underneath the cells are 0.5 µm in diameter and emit yellow fluorescence.
    2. Skip the step of attaching 2 µm beads to the coverglass if the viewing area is expected to have a large cell-free region.
      NOTE: Hydrogel preparation protocol is now fairly established in the field16. Throughout the subsequent description, the 0.5 µm beads are referred to as `top beads', and the 2 µm beads are referred to as `bottom beads'. However, the bottom beads are optional where the imaged region contains a large cell-free area. The top bead pattern from such a cell-free region will serve the purpose of the bottom bead pattern.
    3. Mount the hydrogel with embedded fluorescent beads on the microscope stage.
    4. Allow 15-20 min for the temperature of the plate to attain a steady-state.
      NOTE: The plate's top surface is functionalized with extracellular matrix proteins, but the cells are not yet seeded on the hydrogel.
  2. Reference image acquisition
    1. Part 1: Creating a position list
      NOTE: Manual input in this step includes following the AcTrM prompts to choose locations with the best bead distribution. No previously created files are used in this step. Key new folders produced in this step include `t0imgs', `tnimgs', and `textfiles' folders, and the file produced includes the position list file. See Figure 2 for the visual description of the following steps that guide the creation of a position list using AcTrM. The demonstrated example acquires data at 20x magnification.
      1. Start µManager 2.0 – Beta.
      2. Run AcTrM by clicking on the AcTrM.bsh button in the IMBL Resources window.
      3. In the window titled Define Acquisition Properties, select the appropriate objective, tolerance for the accuracy of lateral position (i.e., XY) recovery, step size of the rough and refined refocusing (i.e., z) operation, and value of acceptable image correlation coefficient for focus recovery.
        NOTE: A finer tolerance for the lateral position recovery will slow position recovery, but a substantial tolerance will need position correction during data analysis and may cause difficulty in focus recovery. A smaller step size will slow the refocusing operation, but a very high step size may cause rapid movement with opportunities to miss the focus.
      4. In the AcTrM – Step 0 window, select Fresh Acquisition to create a position list.
        NOTE: The window titled AcTrM – Step 1 lists all the key next steps. These steps involve moving the stage, adjusting the focus, and clicking on buttons in the µManager. These steps guide in constructing the list of positions suitable for data acquisition.
      5. Click on Live de Micromanager to visualize the samples for manual adjustments listed in the AcTrM – Step 1 window.
      6. Follow the steps listed on this window.
      7. Now acquire the image of the beads. Select the appropriate channel for the top beads.
      8. Ensure to look at the bottom beads as well. To do so, select the channels for the bottom beads. A blurry image of bottom beads is seen.
        NOTE: Channels can be switched from the preset dropdown menu on the main µManager window. If the bottom beads are used in the experiment, ensure that they are present in the selected position. These beads will appear blurry.
      9. Upon selecting the appropriate position, click on Mark in the Stage Position List window to save the position.
        NOTE: The best position is defined by a dense and uniform distribution of the top beads embedded immediately underneath the top surface (i.e., top beads) and at least two beads attached to the cover glass (i.e., bottom beads) (Supplementary Figure S1). Focus recovery is faster if the bottom beads appear as large, out-of-focus rings. However, if experiments are done in one position with no need for focus or lateral position recovery, ignore instructions related to the bottom beads image.
      10. Follow steps 6-9 described in Figure 2 to include additional positions into the list.
        NOTE: Choose a few additional positions so that a round of elimination can allow for the best positions chosen on the plate.
    2. Part 2: Acquisition of reference images
      NOTE: This step does not involve any manual input. Previously created key files used in this step include `textfiles/*.pos'. Key new files and folders produced in this step include those within `t0imgs' and `tnimgs' folders. See Figure 3 and Supplementary Figure S2 for the visual description of the following steps that guide the acquisition of reference images using AcTrM. The AcTrM – Step 2 window also lists all the key steps.
      1. Following the steps listed in AcTrM – Step 2 window, make the choices in the Multi-Dimensional Acquisition window. For example, to perform a long-time lapse, indicate a number of images to be taken, select the first channel as phase channel, and then select the next for the top beads and the one after that is for bottom beads. Click on Close in the Multi-Dimensional Acquisition window, and then click on OK in the ActrM – Step 2 window.
      2. In the AcTrM – Step 3 window, choose to engage the reference-image-based XYZ position recovery. The choice is generally yes. However, if images are acquired at one position with no room for focus or stage drift, the choice in AcTrM – Step 3 window will be No.
      3. In the AcTrM – XYZ Recovery Options window, set the channel for rough XYZ recovery, whether to perform refined XYZ recovery, region, and channel for refined XYZ recovery and the direction to begin refocusing (Supplementary Figure S4). When completed, AcTrM will acquire reference images.
        NOTE: Typical choices for the channel will be the bottom beads channel. XYZ recovery works best when carried out with the full image in the current implementation. Reference images will typically include a transmitted light image of the hydrogel, a fluorescent image of top beads, and a fluorescent image of the bottom beads. Each image set's content may vary based on the choices made in the Multi-Dimensional Acquisition window. Still, the software will acquire the reference image set at each position determined in Figure 2. At the end of this step, three folders are created in the chosen directory: `t0imgs', `tnimgs', and `textfiles'. The `t0imgs' folder contains reference images in the format recognized by µManager and used in subsequent steps; `tnimgs' folder contains separate TIFF images with filenames `c0_p0_t0.tif', `c1_p0_t0.tif', etc. Here, `c' stands for the channel, `p' stands for the position, and `t' stands for the time. Following these letters are the channel number, position number, and frame number, respectively. The folder `textfiles' contains the position list in XML format and user choices for image acquisition and position recovery.
  3. Cell seeding and growth
    NOTE: The iTACS platform has the flexibility to accommodate sample preparation protocols used in common in vitro assessment of mechanical behavior of adherent cells, including sparse cells, fully confluent monolayer, network formation assay, and monolayers with holes or significant gaps.
    1. Culture rat pulmonary microvascular endothelial cells to confluence in a flask17,18.
    2. Detach the cells using trypsin. Resuspend the cells in a culture medium containing 10% fetal bovine serum to a concentration 1 x 106 cells/mL.
    3. Place a 5 µL droplet of the resuspended cells on a partially dry hydrogel surface and place it in the cell culture incubator.
      NOTE: After 2 days in the culture medium containing 10% fetal bovine serum, the cells in that droplet form a crowded island of cells1.
  4. Automated image acquisition for the remaining experiment
    NOTE: Manual input for this step includes following the AcTrM prompts to resume the image acquisition. Key previously created files used by this step include `textfiles/*.pos' and bottom bead image files in the folder `t0imgs'. Key new files produced in this step include updated position list files and images `tnimgs/*_t*.tif'.
    1. Ensure that the microscope environmental control system reaches stable tissue culture conditions.
    2. Gently mount the plate containing cultured cells onto the microscope stage.
    3. Allow 15-20 min for the temperature and humidity to stabilize.
      NOTE: See Figure 4 for the visual description of the following steps that guide the acquisition of images for the remaining experiment using AcTrM.
    4. Start µManager 2.0 – Beta.
    5. Run AcTrM by clicking on the AcTrM.bsh button in the IMBL Resources window.
      NOTE: Ignore the choices made in the Define Acquisition Properties window but use the choices made in the previous step.
    6. In the AcTrM – Step 0 window, select Continue Paused Acquisition.
    7. Select the magnification and directory identified in the Multi-Dimensional Acquisition window in the earlier step where data folders ('t0imags', `tnimgs', and `textfiles') were saved.
      NOTE: The title of the window guides the user to choose the appropriate directory.
    8. In the Repositioning the Plate window, select choices for repositioning the plate (three noncollinear positions) along with the channel used for repositioning.
      NOTE: Saved images will be seen in the camera. If overlapping images display as a red,green, and black color image, perform manual adjustment.
    9. Click on Accept to proceed with the acquisition.
      NOTE: This step overcomes the slight shifts from the irreproducibility of plate alignment when the plate is remounted on the microscope stage. Follow AcTrM prompts to expedite alignment at each of the three positions by showing a composite image of the reference image shown in red (typically of the bottom beads) and the image currently observed through the objective shown in green. Getting the green sufficiently close to red leaves less work for the software. When the overlap is perfect, the composite image appears yellow and black. Sufficiently close is typically when the same bottom beads are visible in both red and green images, and the corresponding red and green out-of-focus rings are touching each other. After the plate repositioning is complete, AcTrM takes the stage to each selected position and recovers XYZ position by matching the reference image with what is currently seen through the camera. The first lateral (XY) position is matched, and then the focus (Z) is matched. Rough recovery is followed by refined recovery of the position and focus. Updates of the current status of the experiment are displayed on the screen through a four-part window shown in Supplementary Figure S3. The acquired images are saved in the `tnimgs' folder following `*_t1.tif', `*_t2.tif', indicating the time count for the image set. If XYZ recovery identifies an updated position, the new position list is generated and saved in the `textfiles' folder.

2. Analysis and Visualization Module (AnViM)

  1. Setting up automated data analysis
    NOTE: Manual input in this step includes identifying the location of the `tnimgs' folder. Key previously created files used in this step include `tnimgs/*.tif'. Key new files and folders produced in this step include the `analysis' folder in the same parent folder as `tnimgs' and folders for each position `analysis/p*'. Within each `analysis/p*', this step creates a new `*.tif' image files within `phs' and `tny ' folders. These images are cropped and de-drifted images of cells and top beads. The other files created include `analysis/analysisChoices.txt' which lists analysis choices, `analysis/p*/skipAnalysis.txt', which lists instances where analysis is skipped due to significant preexisting drift, and `analysis/p*/part_shift_values_pixel_degree.dat' which lists estimated drift values. AnViM does not alter the raw data in the `tnimgs' folder. See Figure 5 for the visual description of the following steps that guide the setting up of automated data analysis using AnViM.
    1. Start the Fiji software and select the first option in the MSM drop-down menu labeled as MSM – pre-processing.
    2. Using the file browser dialog, select the folder containing the `tnimgs' folder, which contains the analyzed data.
    3. Define the channel for the images. Following the guidelines from Supplementary Figure S5, identify the channel numbers of the transmitted light image of the cells (phase contrast image of the cells), bottom beads image and top beads image. The following three checkboxes (`Move files to the position folder', `Do additional correction for rigid motion', and `Crop data and save in position folder') are typically checked. Finally, define tolerance for rejecting heavily drifted data, pixel size, and sides of the image where the cells cross.
    4. Respond to the prompt for enhancing the brightness and contrast of the bottom bead images so the beads appear prominently. Adjust this using the slider bar in the menu and click on OK.
    5. Now perform position correction to get rid of the shifts, if any. Once done, an analysis folder is created.
      NOTE: Contrast enhancement is typically not needed, but this provision is available for experiments where exposure to lasers needs to be minimized. After that choice, AnViM copies the files from the `tnimgs' folder to the `analysis' folder. The files corresponding to each position are saved in folders `analysis/p0, analysis/p1', etc. Within each of these position folders, AnViM makes `cels', `defs', and `refs' folders containing the original transmitted light image, top beads images, and bottom beads images, respectively (Supplementary Figure S6). AnViM then analyzes rigid motion in the images of the bottom beads and creates a `phs' folder containing a corrected transmitted light image of the cells and a `tny' folder containing updated top beads images. Finally, user choices of the channels, operations, tolerance, pixel size, and boundary-crossing are saved in the `analysisChoices.txt' file (Supplementary Figure S6).
  2. Quantification of deformation of hydrogel and monolayer
    NOTE: It is important to note that quantification of movement from a sequence of images is a rapidly evolving field19. The technology is constantly optimized for features, including speed, accuracy, specific features within the raw images, and specific deformation patterns. Hence, it is likely that some users may use a different deformation quantification approach than the one presented here.
    1. Part 1: Quantification of hydrogel deformation
      NOTE: Manual inputs in this step include identifying data directory and selection of grid resolution. Key previously created files used in this step include `p*/tny/*.tif'. Key new files produced in this include `p*/displacement/*_disp.dat' that lists displacement vectors of the top surface of the hydrogel. See Figure 6 for the visual description of the following steps to engage, via AnViM, the Particle Image Velocimetry analysis on the top beads images20.
      1. Start the Fiji software, and from the MSM drop-down menu, select MSM – Gel Deformation.
      2. From here, select the option that is suitable for the experiment.
      3. Using the file browser dialog, select the parent directory of the `analysis' folder containing the analyzed data.
      4. In the Parameters for Calculating Gel Deformation window, select appropriate cross-correlation window size, noise level, and threshold (Supplementary Figure S7)20.
        NOTE: The results are saved in a newly created `displacement' directory within each position folder `analysis/p0, analysis/p1', etc. This is where all the output files are stored.
    2. Part 2: Quantification of monolayer deformation
      NOTE: Manual inputs in this step include identifying the data directory and selection of grid resolution. Key previously created files used in this step include `p*/tny/*.tif'. Key new files produced in this include `p*/velocity/*_vel.dat' that lists motion vectors of the cells. See Figure 7 for the visual description of the steps to engage, via AnViM, the Particle Image Velocimetry analysis on the top beads images20. The procedure is similar to the one followed for quantifying hydrogel deformation and selects MSM – Cellular Motion from the MSM drop-down menu. The results are saved in a newly created `velocity' directory within each position folder `analysis/p0', `analysis/p1', etc.
  3. Quantification of cell-ECM and cell-cell forces
    NOTE: Manual input in this step includes identifying the data directory, indicating the hydrogel stiffness, and responding to cellular cluster segmentation prompts to detect cells from the no-cell region. Key previously created files used in this step include `p*/displacement/*_disp.dat'. Key new files produced in this step include: `p*/traction/*_trac.dat', which lists forces exerted by the cells on the hydrogel; `p*/traction/*_domain.dat', which lists the location of grid points containing cells; and input files `p*/traction.in, p*/clusterInput.txt' and `p*/prestress.in' that record user choices for this step. Following the first quantitative assessment of cell-ECM and cell-cell forces, several variations to the technique have been developed1,15. The variations focus on particular cases of substrates, cells, experimental conditions, or numerical tools7,8,21,22. See Figure 8 for the visual description to engage via AnViM, the Fourier Transform Traction Microscopy, and Monolayer Stress Microscopy analysis on the hydrogel deformation data1,2,15.
    1. Start the Fiji software, and from the MSM drop-down menu, select MSM – Cell-ECM and Cell-Cell Forces (third option in the dropdown menu).
    2. Using the file browser dialog, select the parent directory of the `analysis' folder containing the folders `tnimgs' and `analysis' containing analyzed data. Click on Select.
    3. In the Parameters for Calculating Cellular Forces window, enter the shear modulus, thickness of the hydrogel and the anticipated noise level. Select OK. This lets it execute traction via the MATLAB function.
      NOTE: Following these inputs, AnViM computes cell-ECM forces.
    4. After that, indicate whether the monolayer is confluent. If the entire cell image is covered with cells, then the answer is Yes. In that case, the cell-cell forces will be computed across the whole frame. On the other hand, if a portion of the image does not have cells, then the answer is No. In that case, follow AnViM prompts to facilitate segmentation of the image region that contains cells.
    5. If the answer is no, manually draw a polygon around the non-cell object when the software prompts to do so, and then select an appropriate method(s) for segmentation. The software will ask for the color (black or white) of the cells.
    6. Check the Fill Spots Automatically option in the software and click on OK.
      NOTE: The steps for segmenting a non-confluent monolayer will be covered in the first part of `Section 2.4. Cell-ECM forces are stored in `*_trac.dat ' files in a newly created directory `traction' within each position folder, and the input to the cell-ECM force calculation software is saved in the file `traction.in' (Supplementary Figure S8). Cell-cell forces are stored in `*_prestress.dat' files in a newly created directory `prestress' within each position folder, and the input to the cell-cell force calculation software is saved in the file `prestress.in' (Supplementary Figure S9).
  4. Mapping grid point values on individual cells
    NOTE: One of the current focuses of iTACS is to introduce a straightforward approach to interpret the measured mechanical signals in the field. This approach is beneficial for examining mechanical interactions among neighboring cells of a cluster18. Overall, the properties quantified until now are on regularly spaced grid points across the cellular cluster. This data identifies median, mean, and standard deviation of selected properties within the morphological boundary of individual cells and assigns those as cellular physical properties/signals. From these, the standard deviation is used to indicate variability, the difference between mean and median is used to indicate the nature of distribution, and the median value is used to indicate the overall state of the cells for the selected properties.
    1. Part 1: Segmentation of the image region that contains cells
      NOTE: Manual input to this step includes identifying data directory and responding to the segmentation prompts to detect cells from the no-cell region. Key previously created files used in this step include `p*/phs/*.tif'. Key new files produced in this step include `p*/phs/bwImages/*.tif' which are binary images separating cell regions from no-cell regions, `p*/phs/textResults/frameNum_percentHealed.txt', which lists percent image area covered by cells in each instance, and `p*/clusterInput.txt', which records user choices entered in the AcTrM interface. See Figure 9 for the visual description of the following steps to segment the image region that contains cells. This part must be completed before computing the cell-cell forces. In that case, these steps need not be repeated. Also, if these steps are conducted before the cell-cell force calculations, they are not requested at the beginning of the force calculations.
      1. Start the Fiji software, and from the MSM drop-down menu, select Segmentation – Cellular Cluster.
      2. Using the file browser dialog, select the position directory `analysis/p0', `analysis/p1', etc., that contains properties quantified on regularly spaced grid points.
      3. Indicate whether the monolayer is confluent.
        NOTE: The steps below are invoked only when the monolayer is not confluent. Follow AnViM prompts to apply the novel multi-pronged approach to identify the image regions that contain cells. The process involves four methods – each approaching segmentation in a different manner. One or multiple of these methods used in combination cover a wide variety of images of the cells. Therefore, try different approaches to find which combination works best for their data.
      4. Adjust the `Domain connection factor' and `Smear factor' to obtain optimum segmentation.
        NOTE: The `Smear factor' makes the regions that contain cells larger.
      5. Indicate whether the cells appear black or white in the binary image.
      6. In the Directions to Clean the Boundary Image window, choose whether to clean the image automatically or manually.
        NOTE: Unwanted features of the images are black spots on the pixels occupied by cells and white spots on the pixels disconnected from the cell cluster to be analyzed. Analysis can only be done on the regions that are connected. So multiple disconnected regions must be analyzed separately.
    2. Part 2: Segmentation of the individual cells in the images
      NOTE: Manual inputs to this step include identifying the data directory and responding to the segmentation prompts to detect individual cells within the image. Key previously created files used in this step include `p*/phs/*.tif' and `p*/phs/bwImages/*.tif'. Key new files produced include `p*/phs/textResults/*.csv', which contain cellular morphological properties. See Figure 10 for the visual description of the following steps to segment individual cells of the monolayer.
      1. Start the Fiji software, and from the MSM drop-down menu, select Segmentation – Individual Cells.
      2. Using the file browser dialog, select the position directory `analysis/p0', `analysis/p1', etc., that contains properties quantified on regularly spaced grid points.
      3. In the Choose the Input Parameters window, indicate the maximum aspect ratio, prominence of how much the cell center stands out compared to the cell boundary, and two blurring parameters to guide the detection of the cells (Supplementary Figure S10).
      4. On the image stack for each position, draw a polygon on the smallest normal cell. This is used to calculate the area. Anything smaller than this will not be considered as a cell by the software.
      5. Next draw a polygon around the largest normal cell and indicate whether the cell center or the cell-cell interface is brighter.
        NOTE: AnViM then creates files `phs/textResults/0001.csv', `phs/textResults/0002.csv' , etc., for each frame containing information on cells within that frame. At this stage, this file includes morphological information about the cells, including area, centroid, perimeter, orientation, circularity, aspect ratio, roundedness, solidity, distance from the no-cell region. The length unit in these properties is pixels, and the angle is in degrees. Finally, this file is updated to contain cellular pixel intensity, motion, and forces in the subsequent steps.
    3. Part 3: Mapping of the pixel intensities on the cells
      NOTE: Manual inputs in this step include identifying the data directory and selecting mapping properties and parameters. Key previously created files used in this step include `p*/phs/*.tif' (or `p*/fluo/*.tif ') and `p*/phs/bwImages/*.tif'. Key new files produced in this step include `p*/phs/textResults/*.csv', which lists cellular morphological properties. See Figure 11 for the visual description of the following steps to assess the pixel intensities in the region encompassed by individual cells and their neighboring region and define them as properties of the cells.
      1. Start the Fiji software, and from the MSM drop-down menu, select Map on Cells – Intensities.
      2. Using the file browser dialog, select the position directory `analysis/p0', `analysis/p1', etc., that contains properties quantified on regularly spaced grid points.
      3. Select Detect Cell Division or Quantify Cellular Fluorescence when prompted by the software. Define the size of the neighboring region. Check both Collect Properties of the Neighboring Region and Collect Properties of the Cells boxes. Click on OK.
      4. In the Purpose of Recording the Cellular Intensity window, indicate what type of image to be used for intensity mapping, the size of the neighboring region, and whether data is collected for individual cells or both cells and their neighboring regions (Supplementary Figure S11).
        NOTE: Mapping the pixel intensities of a phase-contrast image allows detection of cell division events. Mapping fluorescent image intensities will enable detection of fluctuations in the fluorescently tagged cytoplasmic molecules. The output of the above steps is the mean, median, standard deviation, minimum, and maximum pixel intensity values for individual cells. These numbers are entered as new columns in the files `phs/textResults/0001.csv', `phs/textResults/0002.csv' , etc.
    4. Part 4: Mapping of the forces and motion on the cells
      NOTE: Manual inputs to this step include identifying the data directory and selecting mapping properties and parameters. Key previously created files used in this step include `p*/phs/textResults/*.csv', `p*/velocity/*_vel.dat', `p*/displacement/*_disp.dat', `p*/traction/*_trac.dat', `p*/prestress/*_prestress.dat', and `p*/phs/bwImages/*.tif '. No new files are created during this step. Instead, the new information (cellular force and motion properties) are added to the files `p*/phs/textResults/*.csv '. See Figure 12 for the visual description of the steps to assess the forces and motion in the region encompassed by individual cells and their neighboring region and define them as properties of the cells.
      1. Choose from the MSM drop-down menu Map on Cells – Motion/Force Data (Supplementary Figure S12).
      2. Then, follow the steps listed in step 2.4.3.
        NOTE: New columns are added to the files `phs/textResults/0001.csv', `phs/textResults/0002.csv', etc. and include the mean, median, and standard deviation of speed, the orientation of velocity, average cytoskeletal tension, tension anisotropy, strain energy in the hydrogel, the orientation of highest tension, and magnitude of cell-ECM traction (Supplementary Figure S13).
  5. Visualization of results
    1. Part 1: Tracking of the cellular identities across the experiment
      NOTE: Manual inputs to this step include identifying the data directory and selecting mapped properties to visualize. Key previously created files used in this step include `p*/phs/textResults/*.csv'. Key new files produced during this step include `p*/phs/textResults/cellTimeTrace_*.csv', which contains a time trace of the cellular data. See Figure 13 for the visual description of the following steps to track the properties of individual cells over the entire duration of the experiment.
      1. Start the Fiji software, and from the MSM drop-down menu, select Results – Track Data.
      2. Using the file browser dialog, select the position directory `analysis/p0', `analysis/p1', etc., containing cellular properties to be tracked. When done click on Select.
      3. In the Choose Starting Point for Tracking window, choose the starting frame for monitoring, and the number of frames simultaneously tracked (Supplementary Figure S14). Always start from frame number 2 as velocity cannot be determined for frame number 1. When done, click on OK.
      4. In the Check the Variables to Make Tracks window, choose the variables either by typing the variable names or selecting the terms from the checkboxes (Supplementary Figure S15). Then, click on OK.
        NOTE: Tracking generates `phs/textResults/cellTimeTrace_*.csv' files with each column containing a unique cell number and each successive row representing the consecutive time instance.
    2. Part 2: Generation of time tracks of the assessed cellular properties
      NOTE: Manual inputs to this step include identifying the data directory and selecting mapped properties to visualize. Key previously created files used in this step include `p*/phs/textResults/cellTimeTrace_*.csv'. Key new files produced during this step include `p*/phs/trackedDataPlots/*.tif', which contain time track plots and `p*/phs/trackedDataPlots/*.csv', which contain plot data. See Figure 14 for the visual description of the following steps to generate time tracks of the assessed cellular properties.
      1. Start the Fiji software, and from the MSM drop-down menu, select Results – plot.
      2. Using the file browser dialog, select the position directory `analysis/p0', `analysis/p1', etc., containing tracked cellular properties to be plotted.
      3. Choose whether to limit the plotting to cells with unbroken tracks by clicking on the Yes or No button.
        NOTE: This option ignores the cells, which could not be detected in one or more time instances.
      4. Choose the number of variables to be plotted simultaneously and click on OK.
        NOTE: Currently, AnViM allows a maximum of three variables to be displayed in the same plot.
      5. Choose individual variables for plotting using a drop-down menu.
        NOTE: These steps create a tagged image file format (TIFF) file and a comma-separated data file within the `phs/trackedDataPlots/' folder. The filename is composed of variable names separated by an underscore and ends with a cell number.
    3. Part 3: Generation of heat maps of the assessed cellular properties
      NOTE: Manual inputs in this step include identifying the data directory and selecting mapped properties to visualize. Key previously created files used in this step include `p*/phs/textResults/cellTimeTrace_*.csv'. Key new files generated in this step include `p*/phs/segmentedImages/cellPropMaps/*.tif', which are the heatmaps of cellular properties. See Figure 15 for the visual description of the steps to generate heat maps of the assessed cellular properties.
      1. Choose from the MSM drop-down menu Results – Picture.
      2. Follow the steps described in step 2.5.2.
        NOTE: The output is stored as TIFF file files in the `phs/segmentedImages/cellPropMaps/' folder. The filenames are the time instance number and variable name separated by an underscore.

Representative Results

We present here two of the key outputs for the demonstrated example. The first output is the time trace of cellular speed and cytoskeletal tension for cell number 1 (Figure 16). The properties are shown on a shared vertical axis to facilitate the visual association between the properties, and the horizontal axis indicates time instance number. In this experiment, successive frames were acquired at a 15 min interval. The second output is an array of heat maps 1 h into the experiment (Figure 17). The properties shown here include spread area, orientation, circularity, speed, direction of motion, maximum tension orientation, cytoskeletal tension, substrate tractions, and tension anisotropy of individual cells.

Figure 1
Figure 1: Structure of Integrative Toolkit to Analyze Cellular Signals (iTACS). Two key components of iTACS are Acquisition and Training Module (AcTrM) and Analysis and Visualization Module (AnViM). AcTrM can use various hydrogel preparation techniques that currently exist for preparing hydrogels that can be held firmly on a microscope stage, any cell seeding, and a growth protocol that retains cells in one focal plane. AnViM can use various techniques to quantify the hydrogel and monolayer deformation, cell-ECM forces, and cell-cell forces. All these user-preferred components of the force measurements protocol can be accommodated in iTACS, and they have been identified with dashed boxes. The components identified with solid boxes are novel contributions to cellular force measurement technology. Visualization in the AnViM focuses on the median value and variability of the properties across individual cells. Please click here to view a larger version of this figure.

Figure 2
Figure 2: Reference image acquisition – part 1. Steps for creating a position list using AcTrM. Please click here to view a larger version of this figure.

Figure 3
Figure 3: Reference image acquisition – part 2. Steps for acquiring reference images using AcTrM. Detailed views of steps 2, 4, and 6 are presented in Supplementary Figures S2, S3, and S4, respectively. Please click here to view a larger version of this figure.

Figure 4
Figure 4: Automated image acquisition for the remaining experiment. Steps for resuming image acquisition to assess cellular behavior using AcTrM. Please click here to view a larger version of this figure.

Figure 5
Figure 5: Setting up automated data analysis. Steps to begin automated image analysis using AnViM. The software recognizes the image format used by AcTrM. A detailed view of the panels in steps 3 and 5 is presented in Supplementary Figure S5 and Supplementary Figure S6, respectively. Please click here to view a larger version of this figure.

Figure 6
Figure 6: Quantification of deformation of hydrogel and monolayer – part 1. Steps to engage, via AnViM, the Particle Image Velocimetry implementation of Tseng, Q. et al., PNAS (2012)20 to quantify deformation of the top surface of the hydrogel. Users can also implement within AnViM other approaches to quantify hydrogel deformation. A detailed view of step 3 is presented in Supplementary Figure S7. Please click here to view a larger version of this figure.

Figure 7
Figure 7: Quantification of deformation of hydrogel and monolayer – part 2. Steps to engage, via AnViM, the Particle Image Velocimetry implementation of Tseng, Q. et al., PNAS (2012)20 to quantify the local motion of individual cells. Users can also implement within AnViM other approaches to quantify cellular motion. Please click here to view a larger version of this figure.

Figure 8
Figure 8: Quantification of cell-ECM and cell-cell forces. Steps to perform image analysis to engage, via AnViM, the Fourier Transform Traction Microscopy implementation of Trepat et al., Nature Physics (2009)15 to quantify forces exerted by the cells on the hydrogel, and the Monolayer Stress Microscopy implementation of Tambe et al., Nature Materials (2011)1 to quantify forces within individual cells and between neighboring cells. Users can also implement within AnViM other approaches to quantify cell-ECM and cell-cell forces. A detailed view of step 6 is presented in Supplementary Figure S8 and Supplementary Figure S9. Please click here to view a larger version of this figure.

Figure 9
Figure 9: Mapping grid point values on individual cells – part 1. Steps to segment the image regions containing cells using a novel multi-pronged approach. This approach can be used to segment phase contrast, brightfield, or fluorescence images of the cells. Please click here to view a larger version of this figure.

Figure 10
Figure 10: Mapping grid point values on individual cells – part 2. Steps to segment individual cells of a monolayer using a novel multi-pronged approach developed in AnViM. This approach can be used to segment phase contrast, brightfield, or fluorescence images of the cells. A detailed view of step 2 is presented in Supplementary Figure S10. Please click here to view a larger version of this figure.

Figure 11
Figure 11: Mapping grid point values on individual cells – part 3. Steps to assess pixel intensities in the region within individual cells and within the neighboring region of individual cells using AnViM. The intensities assessed include transmitted light intensity and fluorescence intensity. This part maps the median value and standard deviation of the pixel intensities within individual cells and within a neighboring region of individual cells. A detailed view of step 2 is presented in Supplementary Figure S11. Please click here to view a larger version of this figure.

Figure 12
Figure 12: Mapping grid point values on individual cells – part 4. Steps to assess forces and motion properties of the grid points within individual cells and within the neighboring region of individual cells using AnViM. This part maps the median value and standard deviation of the properties within individual cells and within a neighboring region of individual cells. A detailed view of steps 2 and 3 are presented in Supplementary Figure S12 and Supplementary Figure S13. Please click here to view a larger version of this figure.

Figure 13
Figure 13: Visualization of results – part 1. Steps to track properties of individual cells over the entire duration of the experiment using AnViM. A detailed view of steps 2 and 3 are presented in Supplementary Figure S14 and Supplementary Figure S15. Please click here to view a larger version of this figure.

Figure 14
Figure 14: Visualization of results – part 2. Steps to generate time traces of the assessed properties using AnViM. The user has the option of plotting up to three properties in one graph. Time traces are generated for either all cells or only those cells for which tracking was successful across the entire experiment. A detailed view of step 5 is presented in Supplementary Figure S16 and Supplementary Figure S17. Please click here to view a larger version of this figure.

Figure 15
Figure 15: Visualization of results – part 3. Steps to generate heat maps of the assessed properties using AnViM. Heat maps are generated for all frames following the starting frame and all the selected properties. A detailed view of step 3 is presented in supplemental Figure S18. Please click here to view a larger version of this figure.

Figure 16
Figure 16: Time traces for cell ID 1. Two properties displayed are the cellular cytoskeletal tension ("avgtenMedian") and cellular speed ("speedMedian"). Both cellular cytoskeletal tension and cellular speed are quantified as the median value across the grid points within the cells. The two properties are plotted on the same axis with arbitrary units to visualize relationships between the assessed properties. Additional variable names are listed in Supplementary Table S1. Please click here to view a larger version of this figure.

Figure 17
Figure 17: Heat maps of the properties of individual cells across the analyzed monolayer. Each cell is colored with the median value of the property indicated in the panel. Thus, deep red indicates the maximum cellular value in the color spectrum, and deep blue indicates the minimum cellular value across the analyzed monolayer. As described in Tambe et al., PLoS One (2013)2, the cells located closer to the boundary have mechanical forces affected by unknown properties of the cells outside the image. Hence the heat map is generated for cells far from the boundary. Please click here to view a larger version of this figure.

Figure S1: Sample images of the top and bottom fluorescent bead. Please click here to download this File.

Figure S2: A detailed view of step 2 from Figure 3. Please click here to download this File.

Figure S3: A detailed view of step 4 from Figure 3. Please click here to download this File.

Figure S4 A: detailed view of step 6 from Figure 3. Please click here to download this File.

Figure S5: A detailed view of step 3 from Figure 5. Please click here to download this File.

Figure S6: A detailed view of step 5 from Figure 5. Please click here to download this File.

Figure S7: A detailed view of step 3 from Figure 6. Please click here to download this File.

Figure S8: A detailed view of cell-ECM force output of step 6 from Figure 8. Please click here to download this File.

Figure S9: A detailed view of cell-cell force output of step 6 from Figure 8. Please click here to download this File.

Figure S10: A detailed view of step 2 from Figure 10. Please click here to download this File.

Figure S11: A detailed view of step 2 from Figure 11. Please click here to download this File.

Figure S12: A detailed view of step 2 from Figure 12. Please click here to download this File.

Figure S13: A detailed view of the output of step 3 from Figure 12. Please click here to download this File.

Figure S14: A detailed view of step 2 in Figure 13. Please click here to download this File.

Figure S15: A detailed view of step 3 from Figure 13. Please click here to download this File.

Figure S16: A detailed view of data files generated in step 5 from Figure 14. Please click here to download this File.

Figure S17: A detailed view of a plot generated in step 5 from Figure 14. Please click here to download this File.

Figure S18: A detailed view of a heat map and the file containing the range of the color spectrum generated in step 4 from Figure 15. Please click here to download this File.

Table S1: A list of selected properties quantified by iTACS. Please click here to download this Table.

Discussion

Adherent cells use both mechanical and chemical signals to survive, grow, and function. A wide variety of microscopy software optimizes the user experience in assessing the chemical signals through fluorescence-based imaging. However, assessment of the mechanical signals involves capabilities that are not available in the standard microscopy software. Moreover, the assessment of mechanical signals is most efficient when data acquisition is integrated with data analysis. The lack of a unified platform that meets the unique needs of mechanical signal assessment has been a major technological gap in experimental cell biology. The Integrative Toolkit to Analyze Cellular Signals (iTACS) is designed to meet this gap. The two components of iTACS, AnViM and AcTrM, equip users with the necessary capabilities to quantify cellular properties of four broad categories: forces, motion, morphology, and fluorescence/brightness. Across these categories, iTACS is currently capable of revealing more than 50 unique aspects of individual adherent cells. These aspects comprise specific properties of each broad category, including their representative value and variability across the cell (Supplementary Table S1). For example, within the forces, there are tensile forces across the cytoskeleton, anisotropy of this tension, the orientation of maximum tension, and shear stress across the cell-ECM interface that has a profound influence on the behavior of adherent cells1,3,6.

A novel approach to examine the mechanical behavior of individual cells of a monolayer
Individual cells of a monolayer are engaged in an exchange of signals of chemical and mechanical nature3. These two types of signals are transmitted across the cellular monolayer in a different manner23. However, the mechanical signal transmission knowledge lags behind that of chemical signal transmission. This knowledge gap coincides with a sustained lack of simple and intuitive approaches to assess cellular mechanical signals. The novel data mapping approach described here is equipped to fill this gap. Such mapping reveals that fluctuation of intrinsic cytoskeletal tension in the neighboring region of a cell serves as relaxation, fluidization, and anchoring signals that regulate changes in cellular shape, size, and speed of the cell18. Maps of the properties of neighboring regions exhibit "multicellular subdivision" patterns where cells within the subdivision are exposed to a relatively uniform microenvironment and the cells at the boundary of the subdivision are exposed to a remarkably nonuniform microenvironment18.

Accessibility of the force measurement technology
A variety of protocols exists to make PAA hydrogels, analyze hydrogel deformation and cellular motion, and quantify cell-ECM and cell-cell forces1,2,7,8,9,13,14,15,18,20,21,24,25,26,27,28,29,30,31,32. However, these developments remain out of the reach of common cell biology laboratories and confined to laboratories with engineering expertise. By automating the technical aspects of these approaches and integrating them under a unified and user-friendly platform, the goal of iTACS is to make the assessment of mechanical signals a routine activity in experimental cell biology research and education.

ImageJ allows users to develop applications using approaches that would require little or no training11. iTACS is largely built using simple scripting approaches to facilitate community-driven continued development. A bulk of AcTrM is programmed using BeanShell scripts, and the bulk of AnViM is programmed using ImageJ Macros. These scripts and guidance for implementing these capabilities on the user's microscope are available through GitHub (https://github.com/IntegrativeMechanobiologyLaboratory/iTACS).

Quality standardization of the acquired images
Although the elastic-substrate-based techniques to quantify physical forces in adherent cells have been developed and implemented in various labs, the protocol still lacks standardization. One area that needs standardization most is the quality of the acquired top beads images (Supplementary Figure S1). Significant issues arise from the drift in focus throughout the experiment. Our novel reference-image-based refocusing approach makes such a focusing an objective process. The parameters defined in the very first step of AcTrM impose necessary objective quality limits. Other standardization measures can be programmed in future versions of AcTrM.

The broad applicability of iTACS
In addition to quantifying numerous aspects of adherent cells, the iTACS structure facilitates its use for various experimental protocols and needs. AcTrM allows software-guided user self-training. High-speed imaging required by, for example, simultaneous assessment of cytoplasmic calcium fluctuations is currently limited by the speed of repositioning and refocusing hardware and is best done at one location at a time. However, the current implementation is well equipped for long-term imaging, interrupted imaging, where the sample cannot be retained on the microscope stage for the entire duration of the experiment. Since the reference images are acquired at the beginning of the experiment, iTACS enables real-time imaging of mechanical signals, opening new avenues in drug-screening applications. AnViM allows users to provide highly technical information in layman's terms. The ability to quantify a broad spectrum of cellular properties and track them throughout the experiment constitutes critical capabilities needed to discover new intercellular communication mechanisms.

For the future development of iTACS, we have identified four focus areas: (1) enhancement of data acquisition and data analysis speed, (2) implementation of approaches to assess new cellular signals13, (3) development of workshops and education modules on iTACS-based cellular signal assessment, (4) development of low-cost automation solutions.

Declarações

The authors have nothing to disclose.

Acknowledgements

D.T.T. thank staff affiliated with the Center for Lung Biology at the University of South Alabama for stimulating discussions on the experimental cell biology research needs. These discussions were crucial in initiating the development of iTACS.

This work was supported in part by grants from the National Institute of Health/National Heart Lung Blood Institute, P01 HL66299 and R37 HL60024 (Stevens), R01-HL118334 (Alvarez), F32-HL144040-01 (Xu), and from the University of South Alabama through Abraham Mitchell Cancer Research Fund (Singh, Palanki, Tambe), Research and Scholarly Development Grant (Tambe), Honors College, and Summer Undergraduate Research Fellow (Nguyen).

Materials

Reagents and components used in prepare glass surface for hydrogel coating
(3-Aminopropyl)trimethoxysilane, 97% Aldrich chemistry 13822565
2% Bis Solution Bio-rad 1610142
3-(Trimethoxysilyl)propyl methacrylate,98% Acros organics 2530850
40% Acrylamide Solution Bio-rad 1610140
Glass bottom 35 mm dish/ 6 or 12 or 24 well plates MatTek or CellVis
Glutaraldehyde, EM Grade, 25% Polysciences 1909100
Sodium Hydroxide Sigma-aldrich 1002074706
Reagents and components used in preparing suitable hydrogel
2% Bis Solution Bio-rad 1610142
40% Acrylamide Solution Bio-rad 1610140
Ammonium Persulfate Bio-rad 1610700
Cover Slips Electron Microscopy Sciences 7222301
Dulbecco's Phosphate Buffered Saline (1M) Gibco 14190136
FluoSpheres carboxylate 0.2 um, yellow-green(505/515) Invitrogen F8811
FluoSpheres carboxylate 0.5 um, red(580/605) Invitrogen F8812
FluoSpheres carboxylate 2.0 um, red(580/605) Invitrogen F8826
Rain-X
TEMED Bio-rad 1610801
Reagents used in coating extracellular matrix on the hydrogel
Collagen Type I Rat Tail Corning 354236
HEPES(1M) Gibco 15630080
Phosphate Buffered Saline (1M) Gibco 10010023
Sulfo-SANPAH CovaChem 102568434
Microscope hardware used in the current study
Camera Hamamatsu Flash 4.0 LT sCMOS Camera C11440-42U
H117 ProScanTM Stages Prior Scientific
Light source- Lambda DG4 and Lambda DG5 Sutter instrument company
Microscope Nikon eclipse TE2000-S 550372
ProScan III Universal Microscope Automation Controller Prior Scientific
Stagetop incubator ibidi 11922
Stepper Motor Focus Drive Prior Scientific

Referências

  1. Tambe, D. T., et al. Collective cell guidance by cooperative intercellular forces. Nature Materials. 10 (6), 469-475 (2011).
  2. Tambe, D. T., et al. Monolayer stress microscopy: limitations, artifacts, and accuracy of recovered intercellular stresses. PloS One. 8 (2), 55172 (2013).
  3. Das, T., et al. A molecular mechanotransduction pathway regulates collective migration of epithelial cells. Nature Cell Biology. 17 (3), 276-287 (2015).
  4. Vedula, S. R., et al. Mechanics of epithelial closure over non-adherent environments. Nature Communications. 6, 6111 (2015).
  5. Lamason, R. L., et al. Rickettsia Sca4 reduces vinculin-mediated intercellular tension to promote spread. Cell. 167 (3), 670-683 (2016).
  6. Sunyer, R., et al. Collective cell durotaxis emerges from long-range intercellular force transmission. Science. 353 (6304), 1157-1161 (2016).
  7. Dong, L., Oberai, A. A. Recovery of cellular traction in three-dimensional nonlinear hyperelastic matrices. Computer Methods in Applied Mechanics and Engineering. 314, 296-313 (2017).
  8. Nier, V., et al. Kalman inversion stress microscopy. Biophysical Journal. 115 (9), 1808-1816 (2018).
  9. Serrano, R., et al. Three-dimensional Monolayer Stress Microscopy. Biophysical Journal. 117 (1), 111-128 (2019).
  10. Schneider, C. A., Rasband, W. S., Eliceiri, K. W. NIH Image to ImageJ: 25 years of Image Analysis. Nature Methods. 9 (7), 671-675 (2012).
  11. Schindelin, J., et al. Fiji – an Open Source platform for biological image analysis. Nature Methods. 9 (7), 676-682 (2012).
  12. Edelstein, A., Amodaj, N., Hoover, K., Vale, R., Stuurman, N. Computer control of microscopes using microManager. Current Protocols in Molecular Biology. , 20 (2010).
  13. Patel, N. G., et al. Unleashing shear: Role of intercellular traction and cellular moments in collective cell migration. Biochemical and Biophysical Research Communications. 522 (2), 279-285 (2020).
  14. Yeung, T., et al. Effects of substrate stiffness on cell morphology, cytoskeletal structure, and adhesion. Cell Motility and the Cytoskeleton. 60 (1), 24-34 (2005).
  15. Trepat, X., et al. Physical forces during collective cell migration. Nature Physics. 5, 426-430 (2009).
  16. Aratyn-Schaus, Y., Oakes, P. W., Stricker, J., Winter, S. P., Gardel, M. L. Preparation of complaint matrices for quantifying cellular contraction. Journal of Visualized Experiments: JoVE. (46), (2010).
  17. King, J., et al. Structural and functional characteristics of lung macro- and microvascular endothelial cell phenotypes. Microvascular Research. 67 (2), 139-151 (2004).
  18. Patel, G., et al. Mechanical signaling in a pulmonary microvascular endothelial cell monolayer. Biochemical and Biophysical Research Communications. 519 (2), 337-343 (2019).
  19. Kähler, C. J., et al. Main results of the 4th International PIV Challenge. Experiments in Fluids. 57 (6), 97 (2016).
  20. Tseng, Q., et al. Spatial organization of the extracellular matrix regulates cell-cell junction positioning. Proceedings of the National Academy of Sciences of the United States of America. 109 (5), 1506-1511 (2012).
  21. Alvarez-Gonzalez, B., et al. Two-layer elastographic 3-D traction force microscopy. Scientific Reports. 7, 39315 (2017).
  22. Makarchuk, S., Beyer, N., Gaiddon, C., Grange, W., Hebraud, P. Holographic traction force microscopy. Scientific Reports. 8 (1), 3038 (2018).
  23. Janmey, P. A., Miller, R. T. Mechanisms of mechanical signaling in development and disease. Journal of Cell Science. 124 (1), 9-18 (2011).
  24. Dembo, M., Oliver, T., Ishihara, A., Jacobson, K. Imaging the traction stresses exerted by locomoting cells with the elastic substratum method. Biophysical Journal. 70 (4), 2008-2022 (1996).
  25. Dembo, M., Wang, Y. L. Stresses at the cell-to-substrate interface during locomotion of fibroblasts. Biophysical Journal. 76 (4), 2307-2316 (1999).
  26. Butler, J. P., Tolic-Norrelykke, I. M., Fabry, B., Fredberg, J. J. Traction fields, moments, and strain energy that cells exert on their surroundings. American Journal of Physiology: Cell Physiology. 282 (3), 595-605 (2002).
  27. Saez, A., et al. Traction forces exerted by epithelial cell sheets. Journal of Physics. Condensed Matter. 22 (19), 194119 (2010).
  28. Deforet, M., et al. Automated velocity mapping of migrating cell populations (AVeMap). Nature Methods. 9 (11), 1081-1083 (2012).
  29. Polio, S. R., Smith, M. L. Patterned hydrogels for simplified measurement of cell traction forces. Methods in Cell Biology. 121, 17-31 (2014).
  30. Toyjanova, J., et al. 3D Viscoelastic traction force microscopy. Soft Matter. 10 (40), 8095-8106 (2014).
  31. Serra-Picamal, X., Conte, V., Sunyer, R., Munoz, J. J., Trepat, X. Mapping forces and kinematics during collective cell migration. Methods in Cell Biology. 125, 309-330 (2015).
  32. Charrier, E. E., et al. A novel method to make viscoelastic polyacrylamide gels for cell culture and traction force microscopy. APL Bioengineering. 4 (3), 036104 (2020).

Play Video

Citar este artigo
Nguyen, A., Battle, K., Paudel, S. S., Xu, N., Bell, J., Ayers, L., Chapman, C., Singh, A. P., Palanki, S., Rich, T., Alvarez, D. F., Stevens, T., Tambe, D. T. Integrative Toolkit to Analyze Cellular Signals: Forces, Motion, Morphology, and Fluorescence. J. Vis. Exp. (181), e63095, doi:10.3791/63095 (2022).

View Video