Data from: An in vitro model of stiffened colonic mucosa exhibits altered epithelial behavior
Data files
Apr 23, 2026 version files 985.50 KB
-
Figure1AB_Rheometry.csv
1.66 KB
-
Figure1C_FRAP.csv
3.04 KB
-
Figure2__WRN.csv
2.74 KB
-
Figure3_-WRN.csv
2.81 KB
-
Figure4_5_3DCryptIMARISdata_NEW.csv
522.66 KB
-
Figure4_5_AspectRatio_NEW.csv
742 B
-
Figure4_SupFigureS2_3DCrypts.prism
346.58 KB
-
Figure5_CocultureCrypts.prism
85.99 KB
-
Figure6_RNASeqData_NEW.csv
13.33 KB
-
README.md
5.94 KB
Abstract
Stiffening of the extracellular matrix underlying the epithelial cells of the large intestine is associated with aging as well as many diseases. Yet the impact of the stiffened matrix on epithelial physiology remains poorly understood. A 2D and 3D microphysiological model of the large intestine was developed using a collagen scaffold with a physiologic or excessive stiffness (Young’s moduli of 2.84 ± 0.85 kPa and 15.9 ± 0.73 kPa) by altering the collagen concentration within the substrate. Diffusion of a 10 and 40 kDa fluorescent dextran was significantly different between the physiologic and stiff scaffold (97.8 vs 79.8 µm2/s [10 kDa] and 68.2 vs 56.8 µm2/s [40 kDa], respectively). When primary human epithelial cells of the large intestine were grown as a 2D monolayer, cultures on the physiologic scaffold grew to a significantly higher density with more proliferative and fewer differentiated cells than cultures on the stiffened scaffold. Three-dimensional crypt arrays were also fabricated with the physiologic and stiff substrates, populated with cells, and a growth factor gradient applied. The cell density, proliferation, and height-to-width ratio was significantly greater for cells on the physiologic scaffold relative to that of cells on the stiffened scaffolds. Placement of a layer of intestinal fibroblasts below the epithelium on the crypt arrays did not mitigate the impact of the stiffened substrate. Bulk-RNA sequencing revealed 378 genes that were significantly upregulated and 385 genes significantly downregulated in the stiffened vs physiologic scaffolds. This work demonstrates that a molded collagen hydrogel can be used to mimic the biophysical characteristics of a stiffened intestinal stroma, recapitulating physiology observed in vivo. This in vitro model of polarized crypts with a tunable underlying substrate will enable an improved understanding of intestinal epithelial cell morphology, stem cell maintenance and lineage allocation within a stiffened environment.
Dataset DOI: 10.5061/dryad.fxpnvx15v
Description of the data and file structure
https://doi.org/10.5061/dryad.tqjq2bw7b
Data characterizing epithelial cell behavior while cultured within an in vitro model built on a normal stiffness or pathological stiffness collagen hydrogel scaffold with or without supportive primary fibroblasts. This work was conducted entirely at the University of Washington in Seattle, WA. The datasets enclosed herein represent the raw data collected for this original research.
For any questions, please contact Angelo Massaro (angelom@uw.edu). The corresponding article has been accepted for publication in Biofabrication:
Massaro Angelo , Villegas Novoa Cecilia, Allbritton Nancy L. An In Vitro Model of Stiffened Colonic Mucosa Exhibits Altered Epithelial Behavior. Biofabrication
Data files include:
Files and variables:
Figure1AB_Rheometry.csv
Rheological observations of two concentrations of collagen hydrogel (5.75 mg/mL versus 12.5 mg/mL) subject to parallel plate rheometry, three replicates each. Shear storage and loss moduli were measured and are reported in Pascals (Pa). Young's modulus was subsequently determined through conversion and is also reported in Pascals (Pa). The values reported represent a the mean measurement across a frequency sweep of varied rotational frequencies from 0.215 - 2.15 rad/s.
Figure1C_FRAP.csv
Characterizing diffusion through two concentrations of collagen hydrogel (5.75 mg/mL versus 12.5 mg/mL) that was soaked in fluorescent dextran. Measured intensity values and the change in time between frames of captured images for fluorescence recovery after photobleaching (FRAP) are reported. Using the radius of the bleached region (microns) and the radius of the region of interest (microns), the diffusion coefficient of four different sized fluorescent dextrans (4, 10, 40, 150 kDa) was calculated and reported in microns^2 per second.
Figure2__WRN.csv
The fluorescence intensity above a set empirical threshold is reported from images taken of epithelial monolayers that were grown atop collagen hydrogel with varied stiffness in the presence of exogenously provided growth factors Wnt, R-spondin, and Noggin. The area occupied by fluorescent markers for nuclei, proliferative nuclei (EdU), mucin 2 (muc2), and cytokeratin-20 (KRT20) is reported and ultimately all values are normalized to the area taken up by nuclei (Hoechst or H). Values are reported as a proportion of the total area imaged and thus have no units (i.e. proportions).
Figure3_-WRN.csv
The fluorescence intensity above a set empirical threshold is reported from images taken of epithelial monolayers that were grown atop collagen hydrogel with varied stiffness without exogenously provided growth factors Wnt, R-spondin, and Noggin. The area occupied by markers for nuclei, proliferative nuclei (EdU), mucin 2 (muc2), and cytokeratin-20 (KRT20) is reported and ultimately all values are normalized to the area taken up by nuclei (Hoechst or H). Mean values from three experimental replicates are reported as a proportion of the total area imaged and are thus have no units (i.e. proportions).
Figure4_5_3DCryptIMARISdata_NEW.csv
For 3D molded crypts on collagen scaffolds with varied stiffness that contain epithelial cells, fibroblasts, or both in coculture, the number and position of all nuclei and specifically proliferative nuclei is reported as measured in the analysis software IMARIS. The number of nuclei is reported as a count (count, number of cells) and the position is reported as a distance from the base of each crypt in microns (um). Each column displays measurements from a single crypt for either Hoechst positive cells (i.e. measuring all nuclei) or EdU positive cells (i.e. measuring all proliferating cells). The data are organized left to right in the following order: 5.75_Epithelium alone, 5.75_Coculture, 5.75_Fibroblasts alone, 12.5_Epithelium alone, 12.5_Coculture, 12.5_Fibroblasts alone.
Figure4_5_AspectRatio_NEW.csv
A sample of epithelial cells grown on varied stiffness collagen hydrogels is reported. The height and width was measured using Olympus' CellSens software and these values are reported in microns (um). The proportion of height to width, i.e. aspect ratio, is calculated by dividing the height by the width and is thus unitless.
Figure4_SupFigureS2_3DCrypts.prism
Files from GraphPad Prism are provided wherein the values for individual epithelium only crypts, like crypt height (microns), nuclei density (cells/micron), proliferative cell count and location (count, microns), is reported as data tables and plots. Additionally, the area that is above an empirically set threshold for EdU-positive cells (i.e., proliferative cells) across a low magnification picture of 3D crypt arrays is also reported.
Figure5_CocultureCrypts.prism
Files from GraphPad Prism are provided wherein the values for individual coculture (epithelium plus fibroblasts) crypts, like crypt height (microns), nuclei density (cells/micron), proliferative cell count and location (count, microns), is reported as data tables and plots. Additionally, the aspect ratio (height/width) for these cocultured epithelial cells is reported as a unitless proportion.
Figure6_RNASeqData_NEW.csv
Selected RNA sequencing data is shown as Fragments Per Kilobase of transcript per Million mapped reads (FPKM, normalized). For each subgroup (top to bottom) a set of genes is highlighted and for the final two subgroups the output of a Gene Ontology (GO) analysis is shown as determined from the significantly upregulated or downregulated genes when comparing epithelium grown on physiologic or stiffened scaffolds.