This protocol details a customizable method to measure cell migration in response to chemoattractants that may also be used to determine the diffusion rate of a drug out of a polymer matrix.
Cell migration is a vital part of immune responses, growth, and wound healing. Cell migration is a complex process that involves interactions between cells, the extracellular matrix, and soluble and non-soluble chemical factors (e.g., chemoattractants). Standard methods for measuring the migration of cells, such as the Boyden chamber assay, work by counting cells on either side of a divider. These techniques are easy to use; however, they offer little geometric modification for different applications. In contrast, microfluidic devices can be used to observe cell migration with customizable concentration gradients of soluble factors1,2. However, methods for making microfluidics based assays can be difficult to learn.
Here, we describe an easy method for creating cell culture chambers to measure cell migration in response to chemical concentration gradients. Our cell migration chamber method can create different linear concentration gradients in order to study cell migration for a variety of applications. This method is relatively easy to use and is typically performed by undergraduate students.
The microchannel chamber was created by placing an acrylic insert in the shape of the final microchannel chamber well into a Petri dish. After this, poly(dimethylsiloxane) (PDMS) was poured on top of the insert. The PDMS was allowed to harden and then the insert was removed. This allowed for the creation of wells in any desired shape or size. Cells may be subsequently added to the microchannel chamber, and soluble agents can be added to one of the wells by soaking an agarose block in the desired agent. The agarose block is added to one of the wells, and time-lapse images can be taken of the microchannel chamber in order to quantify cell migration. Variations to this method can be made for a given application, making this method highly customizable.
In order for vital processes such as wound healing, immune responses, and embryonic development to occur, cell migration must take place. Cell migration involves the interaction between cells and neighboring cells, the extracellular matrix, and soluble chemical cues (attractants or repellants). As an example, in the process of wound healing, fibroblasts play an integral role in fibrogenesis and wound contraction, where the cells are recruited to the site of injury to synthesize collagen in order to form the extracellular matrix3. Numerous mechanisms behind the migration of fibroblasts to a wound site have been studied, and they include different mechanical, physical, electrical, and chemotactic factors4. Fibroblasts respond especially well to different concentration gradients of growth factors. These different growth factors work together to optimize tissue regeneration5. While observing the chemotactic response of fibroblasts to growth factor concentrations, one can study the pattern of directional migration of fibroblasts and how they orient themselves around physical obstacles in order to reach their destination. Therefore, the goals of this study were to first develop a system in which fibroblast growth could be tracked under guidance by physical barriers and to secondly model the growth of fibroblasts as they navigate through the system.
Currently, the Boyden chamber assay is the most widely used system to measure the migration of cells6. The Boyden chamber consists of a two-chamber multi-well plate where each well may contain medium with or without chemoattractants7. A filter membrane provides a porous interface between the two chambers in each well; this creates a barrier so that cells cannot pass through unless it is by active migration. Typically, for the Boyden chamber, a chemoattractant is added to the lower chamber, and the system is allowed to equilibrate to form a gradient between the upper and lower wells4. One problem with the Boyden chamber assay is that steep gradients end up forming along a single axis perpendicular with the surface of the membrane. This causes the difference in the chemoattractant concentration between the upper and lower wells to be a lower than what was originally expected. Due to this constraint, the Boyden chamber assay makes it hard to correlate specific cell responses with particular gradient characteristics, such as the slope and the concentration difference. Without these measurements, it is hard to study multi-gradient signal integration.
To address some of the constraints of the traditional Boyden chamber assay, microfluidic assays have been developed to form customizable concentration gradients1. Standard methods for creating microfluidic systems require clean rooms for lithographic techniques. These techniques can be difficult to learn especially in a standard classroom setting. Thus, we have designed a chamber system for measuring cell migration that can be made without using a clean room. Using our system, the wells of the assay can be adjusted to a preferred size, and a linear concentration gradient of customized slope can be produced. This allows for accurate measurement of chemotaxis from random movement. The design is an inexpensive and easy-to-use system to model cell growth in response to different chemical stimuli.
1. Producing a Microchannel Chamber to Create a Concentration Gradient
Figure 1: Computer-Aided Design (CAD) Representation of Microchannel Chamber. This image depicts the CAD drawing of the acrylic insert needed to create the microchannel chamber. 1A) Top view of CAD drawing, 1B) Side view of CAD drawing with dimensions in inches, 1C) Top view of CAD drawing with dimensions in inches Please click here to view a larger version of this figure.
2. Plating the Cells in a Microchannel Chamber
3. Dextran Soaked Agarose Blocks
4. Time-lapse of the Dextran Diffusion to Assess the Soluble Factor Concentration Gradient
5. Time-lapse of Cell Growth
Figure 2 shows the movement of the cell front across the channel in response to a gradient of fetal bovine serum placed at the opposite end of the channel from where the cells are plated. The cell front is shown at 48 h (Figure 2A), 72 h (Figure 2B), 96 h (Figure 2C), and 120 h (Figure 2D) post plating. The movement of the cell front was tracked with these time-lapse images, and the migration distance and growth rate were calculated by the MATLAB polyfit function. From this, the average cell front velocity was found to be 13.4 µm/h (Figure 3). Figure 4 shows the fluorescent gradient established by placing a dextran soaked agarose block at one end of the channel and allowing dextran to diffuse through the channel for 12 h. A time-lapse image was taken at every hour, and the fluorescence across the distance of the channel was measured and plotted as shown in Figure 4 and compared to a 1D diffusion model.
Figure 2: Cell Front Movement. Time-lapse images of the cell front in the microchannel chamber moving through the channel. The cell front marked in orange lines, and the migration distance is included. This migration distance was measured from the marking on the plate to the front of the cell front as shown. A 1 mm scale bar is shown in white. A) 48 h post plating, B) 72 h post plating, C) 96 h post plating, D) 120 h post plating. Please click here to view a larger version of this figure.
Figure 3: Growth Rate Calculation. Cell front movement tracked with time-lapse pictures. Growth rate was calculated using the MATLAB polyfit function (roipoly) to yield average cell front velocity of 13.4 µm/h. Please click here to view a larger version of this figure.
Figure 4: Fluorescent Gradient with Dextran. Fluorescent intensity measured across the distance of the channel where each line represents the gradient at a particular hour. ImageJ was used to calculate the fluorescence intensity. This can be done using the Analyze | Measure tool. First, choose Set Measurements under Analyze and select intensity. Then, choose Measure under Analyze in order to get average pixel intensity. This can be plotted versus distance in microns as shown in this figure. Please click here to view a larger version of this figure.
Our microchannel chamber may be used for a multitude of purposes, including determining the cell migration rate in response to growth factors and chemoattractants and measuring the diffusion rate of a drug from a polymer matrix. It is possible to utilize our microchannel chamber to grow cells and place a chemoattractant at one end of the chamber. The cells grow in response to the chemoattractant, and the cell migration can be quantified by taking time-lapse images that may be analyzed in order to determine the cell migration rate. Representative results of this method are shown in Figure 3, where the growth rate was determined to be 13.4 µm/h using the MATLAB polyfit function.
Another usage of our microchannel chamber is to measure the diffusion rate of a drug out of a polymer matrix. A fluorescently tagged dextran of a similar molecular weight to the drug of interest may be used to model the diffusion of the drug of interest. It is important to note that the samples must be handled carefully as a large amount of convection will disrupt the gradient. The representative results of this method are shown in Figure 4. An advantage of this model in comparison to the Boyden chamber assay is that it is more applicable to adherent cell types because of the flat surface; whereas, the cells must fall through a small tube in the Boyden chamber before adhering. Another limitation of the Boyden chamber assay is that it is an endpoint assay; thus, chemotaxis cannot be visually observed. In contrast, with our method, chemotaxis is observed through the time-lapse images9. The use of micropipettes containing chemoattractants have also been used a method of observing cell migration. However, the main drawback of this method is low throughput10. The Dunn chemotaxis chamber is another assay that is used in conjunction with time-lapse imaging like the method detailed here11,12. The chamber consists of two concentric circles carved on the face of the glass slide in order to create an inner and outer well, and a bridge separates the two wells. The chemoattractant is placed in the outer well, and the cells are allowed to migrate from the inner well to the outer well. This migration is quantified using time-lapse images that are analyzed using image analysis software. Similar to the Boyden assay, one of the limitations of the Dunn chamber assay is that it cannot be easily modified to create custom soluble factor concentration gradients.
Another simple technique for assessing cell migration is an in vitro scratch13. This method is fast and inexpensive as it only requires creating a scratch in the cell monolayer and capturing of time-lapse images of cell migration close to the previously created scratch. However, the major disadvantage of this method is that it is not suitable to measure responses to a chemical factor concentration gradient. Our method offers the potential to create a chemical gradient and to quantify cell migration relative to the chemical gradient. Our method may be combined with the scratch method as a scratch may be created in the cell monolayer, while a chemotactic factor is present in one of the wells in the microchannel chamber. This allows for the quantification and modeling of wound healing. Another potential application of our device is to determine the response of cancer cells to agents that inhibit cell growth.
Other microfluidics based systems have also been used for cell migration studies14,15. These studies utilize lithographic techniques with clean rooms to create master molds in silicon and/or laser machining to create microfluidic channels in silicon wafers. Silicon wafer micro machining can be expensive in comparison to our method that utilizes cheaper materials (PDMS and acrylic) to manufacture a microchannel chamber. Our device overcomes the limitation of other PDMS based microfluidic devices in which it was not possible to create small 3D channels15. With our method, we have successfully created PDMS based device with small diameter channels. Other studies have used a PDMS-based microfluidic device with two layers of microchannels and a multi-channel gas mixer that is computer controlled in order to create arbitrary gradients of oxygen to study the migration of cells under different gradients of oxygen16. There is potential for our device to be used in a similar fashion with the addition of a gas mixing chamber at one end of the microchannel chamber.
Microfluidic devices are often made using soft lithographic techniques; these can be adapted to create cell migration chambers. These methods involve printing a mask with channel patterns with computer-aided design (CAD). These masks are then projected on a master mold coated with a photosensitive resist through optical lithography. After the master mold is made, PDMS is poured on the master and allowed to harden in order to create a PDMS mold that can be used for microfluidic studies17,18,19. Chemoattractant or chemorepulsive compounds are placed in the chamber in high concentration in one of the reservoirs to create gradients17. While these techniques are similar to our set up, they can be expensive due to the high cost of creating the master mold for the system and the use of clean rooms. In addition, these methods can be difficult for students to learn in a class based setting. The inserts used in our set-up to create the channels can be created with a variety of methods, including 3D printing and laser cutting. Thus, our system allows for the creation of a wide variety of molds for relatively low cost particularly as 3D printing becomes more affordable and widely available.
PDMS based microfluidic devices for measuring cell migration and growth have been used for a wide range of applications. For instance, in neuroscience research studies20, devices were fabricated using soft lithography techniques in which the PDMS component was placed on a tissue culture dish or glass in order to form two compartments that were separated by a physical barrier with micron-sized grooves to allow neurons to grow across the compartments. Time-lapse images were taken to show neurons crossing the barrier. The method of micropatterning used to produce this device is complex, while the production method for our device is simple and can be used by researchers at all levels. Another study also used an easy low-cost method for producing a similar PDMS microfluidic device using a master mold created from scotch tape in order to create an impression in the PDMS to produce microchannels. An advantage of our approach over the one described in this method is that the channels in our method are fabricated in one step, while this protocol requires adhesion between two PDMS layers which creates problems with medium flow to the cells 21.
In conclusion, our microchannel chamber method is customizable and may be used to obtain different results. The method allows for the creation of a concentration gradient in order to suit the user's needs. Furthermore, our device offers an advantage over existing devices due to its relatively low cost and ease of use for researchers at all levels.
The authors have nothing to disclose.
The authors acknowledge Clemson University’s Creative Inquiry program and NSF CBET1254609 for providing funding for this project.
Sylgard 184 Silicone Elastomer Kit | Sigma-Aldrich | 761036 | poly(dimethylsiloxane) 2-part kit including silicone elastomer base and silicone elastomer curing agent |
Acrylic Sheets | US Plastic | 44200 | |
Disposable Petri Dishes | Falcon | 25373-041 | |
Fluorescein isothiocyanate–dextran | Sigma-Aldrich | FD20s-100MG | |
Agarose, Type I, Low EEO | Sigma-Aldrich | A6013-100G | |
Dulbecco's Modified Eagle's Medium | Fisher Scientific | 11965092 | Cell media components |
Fetal Bovine Serium | Fisher Scientific | 16000036 | Cell media components |
Penicillin-streptomycin | Fisher Scientific | 15140148 | Cell media components |
Phosphate Buffered Saline (PBS) | Fisher Scientific | BP24384 | |
EVOS XL Cell Imaging System | Thermo Fisher Scientific | AME3300 | Instrument used for taking time-lapse images |
Versa LASER | Universal Laser Systems, Inc. | Model number VLS2.30 | Laser cutter used for cutting plastic |