Analysis of Cell Migration within a Three-dimensional Collagen Matrix

Summary

Cell migration is a biological phenomenon that is involved in a plethora of physiological, such as wound healing and immune responses, and pathophysiological processes, like cancer. The 3D-collagen matrix migration assay is a versatile tool to analyze the migratory properties of different cell types within in a 3D physiological-like environment.

Abstract

The ability to migrate is a hallmark of various cell types and plays a crucial role in several physiological processes, including embryonic development, wound healing, and immune responses. However, cell migration is also a key mechanism in cancer enabling these cancer cells to detach from the primary tumor to start metastatic spreading. Within the past years various cell migration assays have been developed to analyze the migratory behavior of different cell types. Because the locomotory behavior of cells markedly differs between a two-dimensional (2D) and three-dimensional (3D) environment it can be assumed that the analysis of the migration of cells that are embedded within a 3D environment would yield in more significant cell migration data. The advantage of the described 3D collagen matrix migration assay is that cells are embedded within a physiological 3D network of collagen fibers representing the major component of the extracellular matrix. Due to time-lapse video microscopy real cell migration is measured allowing the determination of several migration parameters as well as their alterations in response to pro-migratory factors or inhibitors. Various cell types could be analyzed using this technique, including lymphocytes/leukocytes, stem cells, and tumor cells. Likewise, also cell clusters or spheroids could be embedded within the collagen matrix concomitant with analysis of the emigration of single cells from the cell cluster/ spheroid into the collagen lattice. We conclude that the 3D collagen matrix migration assay is a versatile method to analyze the migration of cells within a physiological-like 3D environment.

Introduction

Like cell fusion (for an overview see ^1,2) cell migration is another biological phenomenon that is involved in a plethora of physiological processes including embryonic development, wound healing and immune responses (for review see ³). However, the ability to migrate is also a prerequisite for tumor cells to metastasize (for review see ^3,4).

Cell migration is a complex, not yet fully understood process that is directed by the interplay of several signal transduction pathways initiated by various ligand (e.g., cytokines, chemokines, growth factors, hormones, extracellular matrix components) receptor (e.g., receptor tyrosine kinases, chemokine receptors, integrins) interactions⁵ ultimately causing the reorganization of the actin cytoskeleton concomitant with de- and reassembly of focal adhesion complexes and integrin-mediated signaling⁶.

To analyze cell migration several in vitro and in vivo cell migration assays have been developed in the past decades, including the Boyden chamber/transwell assay⁷, scratch assay/wound healing assay^8-10, three-dimensional (3D) collagen matrix migration assay¹¹ as well as intravital imaging/microscopy (for review see ¹²). Each of these cell migration assays has pros and cons, e.g., concerning costs and need of equipment, handling or reliability of obtained data.

Both the Boyden chamber/transwell assay and the scratch assay/wound healing assay are easy, low-cost and well-developed assays to measure cell migration in vitro^7-10. In the Boyden chamber/transwell assay cells are seeded on top of an insert containing pores (about 8 µm in diameter) - the so-called upper compartment⁷. Optional, the insert could be coated with extracellular matrix components, e.g., fibronectin, collagen, etc., to mimic a more physiological environment. Likewise, endothelial cells could be grown on top of the insert, thereby mimicking an endothelial cell barrier¹³. Those cells that have passed through the pores during a defined time interval into the lower compartment harboring media and supplements, such as growth factors and chemokines, are used as a read-out to quantify cell migration (or extravasation).

In the scratch assay/wound healing assay cells are seeded in plates and are grown to confluency¹⁰. In dependence of the experimental setting plates could be pre-coated with extracellular matrix components, such as fibronectin. After creating a scratch/wound by scraping the cell monolayer single cells from each side of the scratch/wound can migrate into the gap, thereby filling/healing it¹⁰. The distance between the two sides of the scratch/wounds is determined in dependence of time and is used as a read-out for the migratory activity of the cells¹⁰. However, to discriminate between cell proliferation (which could also result in filling/healing of the scratch/wound) and cell migration it is recommended to combine the assay with time-lapse video microscopy and single cell tracking¹⁰.

However, both the Boyden chamber/transwell assay and scratch assay/wound healing assay, are rather imperfect concerning a physiological-like cellular environment. In the Boyden chamber/transwell assay cells have to migrate through a plastic pore, whereas in the scratch assay/wound healing assay cells are seeded on a two-dimensional pre-coated plastic plate. Likewise, it is well recognized that the migratory behavior differs markedly between a two-dimensional and 3D environment³. For instance, three-dimensional-matrix adhesions of fibroblasts differ from focal and fibrillar adhesions characterized on two-dimensional substrates in their content of α5β1 and αvβ3 integrins, paxillin, other cytoskeletal components, and tyrosine phosphorylation of focal adhesion kinase¹⁴. Likewise, cells embedded within a 3D environment also displayed an altered migratory behavior¹⁵. Thus to analyze cell migration more accurately a migration assay is recommended allowing to measure the migration of single cells within a 3D physiological or physiological-like environment.

Intravital imaging/microscopy is the gold-standard for measuring cell migration within a 3D physiological context. This does not only belong to extracellular matrix-cell interactions, but also to the interactions among different cell types, such as tumor cells and endothelial cells during extravasation¹⁶ or lymphocyte trafficking within the lymph node¹⁷, which, to date, is possible due to improved fluorescence microscopy techniques, such as 2-photon confocal laser scanning microscopy, the use of vital fluorescent dyes and transgenic mouse strains expressing fluorescent proteins derivatives^12,16,17. Additionally, intravital imaging/microscopy could be combined with manual and automated cell tracking¹⁸. However, because of the need of a 2-photon confocal laser scanning microscopy as well as animals (and appropriate transgenic animal models) intravital imaging/microscopy is a rather cost-intensive technique.

To overcome the limitations of the Boyden chamber/transwell assay and the scratch assay/wound healing assay and to analyze the migration of different cell types within a 3D environment the 3D collagen matrix migration assay was developed^11,19. Thereby, migrating cells are embedded within a 3D collagen fiber network, which more resembles to the in vivo situation. Conjointly, due to time-lapse video microscopy real cell migration is measured allowing the determination of several migration parameters as well as their alterations in response to pro-migratory factors or inhibitors. Various cell types could be analyzed using this technique, including lymphocytes and leukocytes^11,20, hematopoietic stem/progenitor cells^21-24, and tumor cells^5,25-29. In addition to single cells also cell clusters or spheroids could be embedded within the collagen matrix concomitant with analysis of the emigration of single cells from the cell cluster/ spheroid into the collagen lattice^30,31.

This protocol presents an overview about a simple, but powerful technique to analyze the migratory behavior of different cell types within a 3D environment – an in vitro method yielding in results that are close to the in vivo situation.

Protocol

1. Preparation of Migration Chambers

1. Prepare a paraffin wax/petroleum jelly (1:1) mix and heat until the mixture has melted. Using a paint-brush and draw 2-3 layers of the paraffin wax/petroleum jelly (1:1) mix in the middle of the glass slide in accordance to Figures 1B-1D.
   NOTE: We are using common glass slides (76 x 26 x 1.0-1.5 mm (W/D/H))
2. Apply the melted paraffin wax/petroleum jelly mix rapidly on the glass slide to avoid solidification while drawing. Ensure the paraffin wax/petroleum jelly layer is about 2-2.5 cm in length and 0.3-0.5 cm in width. The thickness of the paraffin wax/petroleum jelly layer should be about 0.1-0.15 cm.
3. Add a coverslip to the solidified paraffin wax/petroleum jelly mix and seal it with two to three layers paraffin wax/petroleum jelly mix (Figures 1E,1F). Use 4-8 migration chambers for a common cell migration experiment. Place migration chambers in an upright position in a rack.

2. Preparation of the Collagen Suspension Cell Mix

1. Harvest (e.g., with 0.25% Trypsin/EDTA) and count cells of interest. Resuspend the cells in complete media containing fetal calf serum (FCS), antibiotics and recommended supplements. Prepare 4-8 1.5 ml reaction tubes and 20 µl cell suspension containing 4-6 x 10⁴ cells (the total number of cells depends on the cell type to be analyzed) and fill up to 50 µl with complete media.
   NOTE: Additional compounds (e.g., growth factors, chemokines, inhibitors, etc.) are added to the cell suspension. Use appropriate stock solutions such that the final volume of cell suspension does not exceed 50 µl.
2. Prepare a total amount of 452 µl final collagen suspension for four cell migration experiments. Add 50 µl 10x MEM (pH 5.1-5.5) and 27 µl of 7.5% sodium bicarbonate solution (pH 9.0-9.5) to a 1.5 ml reaction tube and mix thoroughly. Observe the solution color turn from yellow-orange to intense purple (pH 9.0-9.5). Add 375 µl liquid collagen suspension to the 10x MEM/sodium bicarbonate solution and mix thoroughly. The pH of the final collagen suspension should be about 7.5 (indicated by a light purple color).
   NOTE: Check the pH of the 7.5% sodium bicarbonate solution. If the pH is about 7.5 place the sodium bicarbonate solution with an open lid in an incubator (37 °C, 5% CO₂) to be saturated with CO₂ and increase the pH (about 9.0 - 9.5). The pH of collagen suspension cell mix is crucial for the stability of the collagen network. If it is too low the collagen lattice will collapse during the cell migration experiment.
   NOTE: For this protocol, use liquid collagen (pH 1.9-2.2) from bovine hind (2.9-3.3 mg/ml collagen; 95% collagen type I, 5% collagen type IV). Examples of further collagen lattice preparation protocols using collagen type I from other sources, such as porcine or rat, are given in ^32,33. Do not change the volumes of the used solutions as this will alter the ultimate collagen concentration of 1.67 mg/ml of the lattice. A higher or lower collagen concentration has an impact on the density of collagen fibers and the average distance between them concomitant with the cells migratory behavior³⁴.
3. Add 100 µl of the final collagen suspension to 50 µl cell suspension and mix thoroughly.The final collagen suspension (1.67 mg/ml collagen) is slightly viscous. To transfer the correct amount (100 µl), pipette the final collagen solution once up and down before transferring it to the cell suspension.
   NOTE: Once the collagen suspension cell mix is combined with the 10x MEM/sodium bicarbonate solution and the pH value has changed to 7.5 the collagen fibers immediately start to polymerize.
4. Transfer the final collagen suspension cell mix from the reaction tubes to the migration chambers. Gently tap the migration chamber to equally distribute the collagen suspension cell mix on the bottom of the migration chamber (Figure 1H). Place the migration chambers in an upright position in a rack and incubate for about 30 min (37 °C, 5% CO₂) to allow polymerization of the collagen fibers.
5. Notice that the polymerized collagen lattice is slightly turbid but still being light purple color. Fill up the migration chambers with complete medium or complete medium with supplements of interest and seal the migration chamber with paraffin wax/petroleum jelly mix (Figures 1I,1J).

3. Recording and Analysis of Cell Migration

This section describes the recording of cell migration by time-lapse video microscopy and analysis of cell migration by manual cell tracking.

1. Switch on the microscope and the microscope stage heater. Adjust the temperature to 37 °C. Use a 10X objective. Place migration chamber under a microscope and focus approximately 50 cells or more in the field of view.
2. Record cell migration in time-lapse mode using a multi-camera video surveillance software application. Save cell migration movies in an appropriate format, e.g., “*.avi” or “*.mov”. Link the cell migration movie files to a database containing supporting information, such as cell type and experimental conditions.
   NOTE: We are recording lymphocytes and HSPCs for up to 2 hr using a time-lapse factor of 1:80, which means that 0.75 sec in time-lapse is equal to 1 min in real-time. Tumor cells are slow moving cells and are thus recorded for at least 16 hr using a time-lapse factor of 1:1,800 (0.5 sec in time-lapse is equal to 15 min in real-time).
3. Analyze cell migration movies using an appropriate cell tracking software application, like ImageJ software plugins (an overview is given in ¹⁸). For an unbiased analysis it is of crucial importance that cells will be randomly selected without the knowledge whether cells are migratory active or not.
   NOTE: We are using a self-developed software application to manually track the paths of at least 30 cells per experiment by following the moving cells accurately with the mouse cursor (which is positioned to the nucleus). While tracking, the xy-coordinates of the tracked cells are automatically determined in accordance to the used time-lapse mode. For lymphocytes/HSPCs xy-coordinates are determined each 0.75 sec (equal to 1 min realtime), whereas for tumor cells the xy-coordinates are determined each 0.5 sec (equal to 15 min realtime).
4. Determine a total of 60 xy-coordinates per cell for a typical cell migration experiment. This is equal to 1 hr real-time for lymphocytes/HSPCs and 15 hr real-time for tumor cells. A “,” indicates that a cell has not moved between two time points, whereas a “numerical value” indicates the distance in “pixels” a cell has moved between 2 time points (Figure 2A).
   NOTE: Manual cell tracking requires experience. Particularly fast migrating cells are difficult to track and each cell type exhibit a distinct migratory behavior that might be triggered by supplemented factors. In case of cell division while tracking, randomly choose one daughter cell for continuing tracking. Cells that migrate out of the sight field or die while tracking are not neglected, but are considered for analysis.

4. Data Analysis

1. Analyze cell tracking data by using an appropriate software application, e.g., a spreadsheet program.
   1. Analyze cell tracking data by copying the cell tracking raw data (Figure 2A), into a self-designed spreadsheet template. Multiply sum of the “numerical values”, representing the total distance a cell has migrated with a correction factor to convert “pixel” into “µm”.
   2. Determine two cell migration parameters: locomotory activity (which is equivalent to migration rate) and time of active movement (Figures 2C,2D).
   3. Identify the cells that have migrated lesser than 25 µm (threshold level to reduce the number of false-positive cells) as non-moving cells. Replace “numerical values” of these cells with “,”.
      NOTE: The parameter “locomotory activity” (or migration rate) represents the percentage of cells of the tracked cell population that have moved between two time points (Figure 2D). A “numerical value” is defined as a moving cell, whereas “,” is defined as a non-moving cell. The parameter “time of active movement” (time active) represents the percentage of the total time a cell has migrated in relation to the time frame of the observation period (Figure 2C). A “numerical value” indicates that a cell has moved, whereas “,” indicates that a cell has not moved. This parameter is further used to determine the number of cells that has not moved.
2. Calculate statistical significance and display cell migration data.
   1. Calculate statistical significance of the mean locomotory activity of the cells (migration rate) using the Mann-Whitney U test. Consider p-value < 0.05 as significant.
   2. Display the mean locomotory activity of cells as xy-diagram, BoxPlot diagram, or as a bar chart diagram. Display single cell-based data, such as time of active movement or speed, as a bar chart diagram or as a histogram.
      NOTE: If the chosen spreadsheet software application does not contain a BoxPlot chart or histogram chart tool an online search should be performed for looking for suitable tutorials.

Results

The used 3D-collagen matrix migration assay combined with time-lapse video-microscopy and computer-assisted cell tracking allows for the determination of various cell migration parameters including both population-based parameter (e.g., mean locomotory activity) and single cell-based parameters (e.g., time of active movement, speed, distance migrated). An example of the obtained cell tracking data sets, data processing and data presentation are given in Figure 2. A cell tracking data fi...

Discussion

The ability to migrate is a hallmark of tumor cells⁴. Without the ability to detach from the primary tumor and to migrate through the surrounding connective tissue tumor cells won’t be able to seed secondary lesions, which are the main cause of death of nearly all cancer patients. Because of this relationship many studies are focusing on cancer cell migration. The aim of these studies is the identification of novel target molecules and target pathways that efficiently block tumor cell migration, thereby ...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Leica DM IL inverted microscope	Leica, Wetzlar, Germany
Microscope stage heater	Distelkamp Electronic, Kaiserslautern, Germany
JVC C1431 video camera	JVC, Bad Vilbel, Germany
Axis 241Q video server	Axis communication GmbH, Ismaning, Germany
Mac G5 Computer	Apple Macintosh
iMac	Apple Macintosh
FileMaker Pro	FileMaker GmbH, Unterschleißheim, Germany
Multi-camera video surveillance software(Security Spy)	Bensoftware, London, UK
Runtime Revolution Media 2.9.0	RunRev Ltd., Edinburgh, UK
Paraffin	Applichem GmbH, Darmstadt, Germany	A4264
Petroleum jelly	local drug store
Purecol (liquid collagen)	Nutacon BV, Leimuiden, The Netherlands		contains 2.9 - 3.3 mg/ml bovine collagen (95% collagen type I, 5% collagen type IV)
10x MEM	Sigma Aldrich, Taufkirchen, Germany	M0275
7.5% sodium bicarbonate solution	Sigma Aldrich, Taufkirchen, Germany	S8761
EGF	Sigma Aldrich, Taufkirchen, Germany	E9644
U73122	Merck Millipore, Darmstadt, Germany	662035	dissolve first in CHCl3; reconstitute in DMSO just prior to use