Materials
SU-8 2050 and its developer were purchased from MicroChem Corp. (USA) and used for microfabrication of microfluidic master mold on a silicon wafer obtained from Nano-BAZAR (Iran). SYLGARD® 184 polydimethylsiloxane (PDMS) kit, and Tygon tubing were purchased from Dow Corning (USA). MCF7 and MDA-MD-231 breast cancer cell lines were purchased from Pasteur Institute (Tehran, Iran). Cell culture reagents including Dulbecco’s phosphate buffer saline (DPBS), Dulbecco’s modified eagle medium (DMEM), fetal bovine serum (FBS), trypsin-Ethylenediaminetetraacetic acid (EDTA), penicillin/streptomycin, and GelTrex were purchased from Gibco (Thermofisher Scientific, USA) while CellTracker were obtained from Invitrogen (Thermofisher Scientific, USA).
Design and fabrication of the gradient generator microfluidic device
The design of the gradient generator microfluidic chip is shown in Figure S1. The design consists of (i) main culture channel, (ii) four signal channels, (iii) hydrogel channels separating signal channels from main culture channels, and (iv) reagent reservoirs. The hydrogel channels are designed based on the capillary effect, so they offer easy and reproducible hydrogel filling while preventing the hydrogel solution from entering the signal or culture channels (Figure S1 B).
The microfluidic chips were fabricated using photo-lithography and soft-lithography approaches as described previously.17 Briefly, a master mold was fabricated by patterning SU-8 on a silicon wafer. Then PDMS base and curing agents were mixed together with a 10:1 volumetric ratio, poured on the master mold, degassed in a vacuum desiccator, and baked on a hot plate (RH digital, IKA, Germany) set at 80 °C. Different baking durations were used to optimize the bonding strength of the PDMS layer to polystyrene (PS, Falcon™ tissue culture dishes and plates, Corning, USA), while 30 min was selected as the optimized value for the rest of the experiments. After backing, PDMS was cut, peeled off from the master mold and punched using 1 mm and 4 mm disposable biopsy punches (KAI instruments, Japan) to make inlets for the hydrogel and reservoirs, respectively. Before experiments, the PDMS chips were cleaned using transparent tapes, rinsed with ethanol and sterile distilled water, and stored at room temperature.
Evaluation of PDMS/PS bonding strength
To evaluate PDMS/PS bonding strength, a PDMS-based microfluidic layer with a long square microchannel (100 μm × 100 μm) was fabricated and placed on a PS surface followed by applying gentle pressure for removing any air between the surfaces. Then, water containing a red dye (for better visualization) was injected into the microfluidic channel at a rate of 1 μL min−1 using an accurate syringe pump (AL-1000, World Precision Instruments, USA). Fluid flow was monitored under an inverted microscope (Leica DM IL LED) and the length of the microchannel filled with the fluid immediately before leakage was measured. The bonding strength was calculated using the following equation:18
(1)
where the first term stands for the flow resistance in the microfluidic channel and the second term indicates the capillary pressure. In this equation, η, L, q, h and γ are dynamic viscosity of water, the filled length of the microchannel, low rate, width or height of the microchannel and surface tension of the water. Also, θPDMS and θPS are water contact angles on PDMS and PS surfaces, respectively. The contact angles were measured to be ~110° and ~80° for PDMS and PS, respectively.
Characterization of mass transport in the microfluidic device
To characterize transport, diffusion rate and stability of a chemotactic factor, finite element simulations were performed in COMSOL Multiphysics 5.4 using “Free and Porous Media Flow” module coupled with “Transport of Diluted Species” module. A 2D model was developed corresponding to the actual microchannel dimensions and discretized with reasonably fine triangular meshes. “No-slip” boundary condition was considered for the fluid flow. The velocity field was first obtained by solving the model using a stationary solver. Subsequently, the mass transport of the chemotactic factor was assessed in the pre-solved velocity field using “No flux” boundary condition and a time-dependent solver. The results were then exported and evaluated using Microsoft Excel software.
Cell culture
Two breast cancer cell lines, including MCF7 and MDA-MD-231 were purchased from Pasteur Institute (Tehran, Iran) and cultured in DMEM culture medium, supplemented with 10% FBS and 1% penicillin-Streptomycin at 37 °C in a humidified atmosphere of 95% air and 5% CO2. At 70–80% confluency, cells were washed with DPBS, harvested with 0.025% trypsin–0.01% EDTA, followed by trypsin deactivation, centrifuged at 1500 rpm for 5 min, resuspended in the new medium, and subcultured or used in the experiments.
Preparation of the microfluidic device for cell studies
The PDMS microchips were directly mounted on the six-well cell culture plate by applying gentle pressure on the PDMS such that the channel side faces onto the surface of the cell culture plate to form the sealed microchannel network. Then, 2 µl GelTrex was gently injected into each hydrogel microchannel using a 10 µl pipette. As a result of the capillary effect, GelTrex precursor was easily confined between the micro-posts designed in the hydrogel microchannels (Figure S1). By incubating the microfluidic device in a humid chamber at 37 °C for 8 minutes, GelTrex was transformed to the solid-state and therefore the hydrogel between the micro-posts isolated the cell culture chamber from the signal channel. Culture medium (37 °C) was then pipetted into the reservoirs to fill the channels and prevent dehydration of the gel. The devices were kept inside the incubator before cell seeding.
For cell seeding, the medium from both culture channel reservoirs was removed followed by adding cell suspension (5×106 cells/ml) to one reservoir and left to equilibrate. Due to the pressure difference, cell suspension quickly flowed toward the outlet reservoir which resulted in a uniform cell seeding. The device was then incubated at 37°C for 2 hrs to allow cell attachment. The cell-containing solution in the reservoirs was then replaced with a fresh culture medium to remove excess cells from the reservoirs. The well plate containing microfluidic devices was finally placed in a cell culture incubator and cellular behavior was monitored each day.
Invasion assay
For evaluating the functionality of the microfluidic chips, a cell invasion assay was designed and performed within the microchips. Since the designed microchip contains four signal and one culture channel, two signal microchannels were filled with free-serum medium to serve as controls and the other two signal channels were filled with 20% FBS medium as test conditions. The cell culture channel was then filled with a medium having 5% FBS. The medium in each reservoir was replaced with a fresh corresponding medium every day to ensure a stable chemotactic factor gradient across the cell culture chamber during the cell invasion experiment. Following the completion of cell adhesion and establishment of the chemical gradients, breast cancer cell invasion was monitored and photographed using a digital camera (Canon EOS 1300D) mounted on a phase-contrast inverted microscope (Leica DM IL LED).
Statistical analysis
All tests were performed at least in triplicates and data were presented as means ± standard deviation. The invasion of the cells at each selected area was quantified by measuring the change in the ratio of hydrogel scaffold area occupied by cells to the total hydrogel area, using the ImageJ software. The comparison between the groups was performed using Student’s T-test and data were presented as *P<0.05, *P<0.005 and ***P<0.0005, where P stands for P-value.