Quantification of F-actin and β-tubulin contents from fluorescence imaging

We assumed that the intensity of the reflected color of the antibody directly correlates with the mass of the marked cellular protein. For example, the F-actin filaments were visualized using Phalloidin (i.e., iFluor 488), which appears green in the confocal images. We quantified the intensity of a specific color using the Z-stack, which is composed of 2D images of cross sections throughout the sample thickness [Figs. 2(a) and S5(a)]. In each experiment, we had up to 12 wells for the non-exposed control and after 60 and 120 min exposure to PWSSs. All the confocal images of a single experiment were acquired at a single setting of the microscope.

The wells of the present study have a net cultured surface of 50.27 mm² (i.e., a diameter of 8 mm), which allowed acquisition of multiple samples of 578 × 578 μm (i.e., 1024 × 1024 pixels) from each well [see Fig. 2(a)]. The intensity data of the stained colors of the Z-stack of cross-sectional images were exported by the Leica software LAS X. We converted the original uncompressed LIF file of a single color (i.e., a single channel) into a TIFF grayscale format (i.e., intensity values of green from 0 to 255 for each pixel) to be used by Matlab software. Then, we further selected smaller sub-samples of 200 × 200 μm (i.e., 352 × 352 pixels) from each sample, which are more common in cell mechanics [see Fig. 6(b)]. The data of each sub-sample were assembled in a 3D matrix A(X = 352, Y = 352, and Z = N) where N is the number of all cross sections in the Z-stack. The total amount of color intensity for each color for all the pixels of all the cross sections of the selected sub-sample will be

Note that M_color is a dimensionless number representing the sum of color intensity from all the cross sections throughout the co-culture thickness [shown in Fig. 2(c)].

Similarly, one can compute the color intensity of the Z-stack cross sections (i.e., slices) within each layer of the EECs and MSMCs of the uterine wall co-culture model. For this analysis, the relevant cross sections of the EECs and MSMCs layers were selected manually, and Eq. (1) was computed for the range k of the selected slices. The borders of each layer between the substrate and the apical end of the EEC were selected with a resolution between 1 and 2 cross sections out of 40–70. An example for the amount of F-actin filaments (i.e., green) vs Z-axis in a specific sub-sample is shown in Fig. S5(c) in the supplemental material for the complete co-culture and the EECs and MSMCs layers.

The relative alteration of the mass of a specific molecule (e.g., actin filaments) due to PWSSs (as shown in Figs. 7 and ​and8)8) was computed with respect to the non-exposed control wells in a specific experiment. The total intensity within the cross sections of either the EECs or the MSMCs (Mcolor/relativecell) of the sub-samples of exposed to PWSS was scaled by the averaged total intensity of the cross sections of either the EECs or MSMCs from all the control sub-samples (i.e., 10–15 per well). Hence,

where the superscript “cell” stands for either the EECs or the MSMCs. N_cell is the number of cross sections in the layer of either EECs or MSMCs, and Q is the total number of the control sub-samples per test. This analysis is useful for evaluation of the relative amount of polymerization or de-polymerization of the actin or tubulin filaments in response to a biochemical or biophysical provocation.