3D printed inserts for reproducible high throughput screening of cell migration

Cell migration is a fundamental and complex phenomenon that occurs in normal physiology and in diseases like cancer. Hence, understanding cell migration is very important in the fields of developmental biology and biomedical sciences. Cell migration occurs in 3 dimensions (3D) and involves an interplay of migrating cell(s), neighboring cells, extracellular matrix, and signaling molecules. To understand this phenomenon, most of the currently available techniques still rely on 2-dimensional (2D) cell migration assay, also known as the scratch assay or the wound healing assay. These methods suffer from limited reproducibility in creating a cell-free region (a scratch or a wound). Mechanical/heat related stress to cells is another issue which hampers the applicability of these methods. To tackle these problems, we developed an alternative method based on 3D printed biocompatible cell inserts, for quantifying cell migration in 24-well plates. The inserts were successfully validated via a high throughput assay for following migration of lung cancer cell line (A549 cell line) in the presence of standard cell migration promoters and inhibitors. We also developed an accompanying image analysis pipeline which demonstrated that our method outperforms the state-of-the-art methodologies for assessing the cell migration in terms of reproducibility and simplicity.


Instructions to use:
1. Open ImageJ/FIJI program.2. Click on 'Process' menu → 'Batch' option → 'Macro' option.3. Enter 'Input' folder location and 'Output' folder location.Input folder is the one where RGB image files are stored; whereas output folder is the one where binary images will be stored.4. Copy-paste above given macro script in black space.
(Note: Make sure to change cyan-colored highlighted areas of macro script as per the necessity.For example, the image scale should be fixed as per the magnification of microscope used for imaging.) 5. Click on 'Process' and results will be summarized in a separate window.Save these results in csv format.

2
Method S2: Use of CellTracker software for determining average cell velocities.
We used CellTracker program (Piccinini et al., 2015) to analyze the cell movement direction as well as average cell velocities under all given treatments.For the same, first all the time lapse images were combined to prepare a stack with total of 9 frames (each frame denoting 12 h period) using FIJI software (Images→Stack→Images to stack).The stacks were converted to 8-bit grayscale format and then cropped to focus one corner of cell-free area along with some clear cellular structures.Then, a stack of images was opened in CellTracker program.All the procedure for 'Vignetting Correction' and 'Automatic Alignment' was performed as instructed by CellTracker developers (Piccinini et al., 2015).We selected 'Manual Tracking' option.We chose 'Linear Interpolation (Faster)' mode.We also fixed 'Maximum Cell Displacement' value to 600 and 'Cell Diameter' value to 20.After that, we selected 10 different cells at random and tracked their migration in each frame to generate X-Y coordinates.Finally, by clicking on 'Statistics' menu, we determined cell displacement, average cell velocities, and cell movement directions.As given in figure S1 (A-E In order to compare the effect of individual treatment with the respective control, we analyzed the cell migration results using two-way ANOVA test.The results obtained from treatment-wise comparison (Figure S2) suggests that during the initial 24 h no significant cell migration changes (in terms of filling up the cell-free area by cell proliferation and cell migration) were observed for any treatment groups.However, at 48 h, cell proliferation and cell migration were faster in EGF treated cells compared to DMEM (without serum), colchicine, and doxorubicin treated cells.At this time point, no significant difference was observed in the migration of cells treated with serum supplemented medium and EGF supplemented medium.After 48 h, a rapid decrease in % cell-free area denotes synergistic action of EGF with FBS, which is evident from figure S2 as two-way ANOVA test showed significant difference in % cell-free area for EGF treatment compared to all other groups past this time point.

4
Method S4: Use of CellTracker software for determining average cell velocities.
Finally, we also checked the cell morphology for given treatments.From bright field images as provided in figure S3, in case of DMEM, DMEM supplemented with 10% FBS, and DMEM supplemented with serum and EGF, the cells looked healthy with normal epithelial morphology.By contrast, the cells treated with colchicine and doxorubicin showed poor morphological features.Very few cells showed epithelial cell like morphology.For majority of cells, the shape was either oval or polygonal.After 24 h, the majority of treated cells showed oval morphology with destroyed cell membranes indicating cell death due to the action of colchicine and doxorubicin (Thorn et al., 2011, Wang et al., 2019, Oh et al., 2022).All these images showed change in morphology owing to the treatment with various supplements and now because of our 3D printed cell inserts with which cell monolayer was grown.
), the colored and numbered lines indicate X and Y coordinates of randomly selected cells across 9 frames (Z-axis).It is clear that the cell movement is very high (denoted by erratic wavy line) in EGF treatment (FigureS1-C), intermediate (denoted by lesser erratic wavy lines) in FBS treatment (FigureS1-B) and low (denoted by even lesser erratic wavy lines with some straight lines) in DMEM treatment (FigureS1-A).In case of Colchicine and Doxorubicin treatment, the cell movement was the lowest or none (represented by constant and straight colored lines) (FigureS1-D and E).This validates our cell migration kinetics data.The bar graph (FigureS1-F) shows average cell velocities (µm/h) for given treatments.EGF supplemented and serum supplemented DMEM showed statistically significant and higher cell velocities (One-way ANOVA, p-value < 0.05) as compared to other treatments.3MethodS3: Use of CellTracker software for determining average cell velocities.