VEGFA dependent endothelial lumen formation in scaffold-free tissue engineering: a hybrid in vitro and in silico approach to improve vascularization of tissue: Spheroid simulation data
Data files
Dec 23, 2020 version files 464.54 MB
-
Analysis.zip
910.91 KB
-
README.docx
17.44 KB
-
Simulation_Code.zip
356.70 KB
-
Simulation_Lattices.zip
463.26 MB
Abstract
Vascularization is a tightly regulated process involving a complex interplay of multiple bio-signaling events. Among these events, lumen formation is a crucial rate-limiting step in successful anastomosis of tissue engineered constructs in a host. We have developed scaffold-free endothelial-fibroblast based constructs with preformed microvessel networks that are a promising approach to rapid host-implant anastomosis. However, rapid formation of a continuous endothelial lumen within the microvessels remains elusive. We propose that pretreatment of these constructs with vascular endothelial growth factor A, a potent proangiogenic molecule, can improve large caliber lumen formation within constructs during in vitro self-assembly.
Here, we present a hybrid in vitro and in silico approach to evaluating endothelial-fibroblast spheroid self-assembly and the effects of VEGF-A on lumen formation. Our VEGF-A dosing experiments ultimately demonstrated that our endothelial cells responded to VEGF-A by increased clustering tendency and hollow lumen formation. This behavior was coincident with an increase in polarization markers as well as a time and dose dependent increase in diameter of lumens. We used these observations to develop a simulation model using the Cellular Potts framework, a modeling framework that is well-adapted to tissue self-assembly. The model explores how changes in interfacial energy costs between cell types in our model as well as changes in basic cell parameters such as response to VEGF-A, growth and proliferation can alter dynamics of lumen formation within a construct. From our model, we ran a series of simulated experiments to generate tissues with an optimal distribution of large caliber vessels. Our analysis of the resulting regression model identified several statistically significant modifiable factors that could be mapped back to potential spheroid pretreatment strategies. These strategies could augment VEGF-mediated lumen formation and provide rationale for alterations in the design of our constructs.
Methods
Simulations were run using CompuCell 3D version 3.7.5. Software available from Glazier et. al in https://compucell3d.org/
CompuCell 3D is an open source software developed and described in:
Multi-Scale Modeling of Tissues Using CompuCell3D – M. Swat, Gilberto L. Thomas, Julio M. Belmonte, A. Shirinifard, D. Hmeljak, J. A. Glazier, Computational Methods in Cell Biology, Methods in Cell Biology 110: 325-366 (2012)
Main simulation files are available in <Simulation_Code> folder
Run the .cc3d files in each run folder (e.g. Polar1). The final outputs during each run are stored as .vtk files in <Simulation_Lattice> Folder.
A comprehensive ReadMe file is included in dataset.
Usage notes
README file included in data set:
GENERAL INFORMATION
Title of Dataset: Tissue Spheroid simulation data for manuscript titled: VEGFA dependent endothelial lumen formation in scaffold-free tissue engineering- a hybrid in vitro and in silico approach to improve vascularization of tissue
Author Information
A. Principal/Corresponding Investigator Contact Information
Name: Michael J. Yost, PhD
Institution: Medical University of South Carolina
Address: 173 Ashley Ave. Room 605, Charleston SC 29425
Email: yostm@musc.edu
Date of data collection: 5/1/2019- 6/10/2019
DATA & FILE OVERVIEW
Data uploaded to Dryad was used to generate Figure 8 and Figure 9 in our submitted manuscript (“VEGFA dependent endothelial lumen formation in scaffold-free tissue engineering- a hybrid in vitro and in silico approach to improve vascularization of tissue”).
1. Files and Instructions for running code and spheroid assembly simulations
Main Folders: “Simulation Code” and “Simulation Lattices”. Simulation code includes python and packaged “Compucell 3D” (.cc3d) files to run spheroid assembly simulations using Compucell3d version 3.7.5. As a heuristic algorithm, each simulation run will be different for users. Our outputs are stored as .vtk files in “Simulation Lattices” folder. The .vtk files should allow users to see the lattice arrangement of our simulation runs.
CompuCell 3D is a software available at <compucelll3d.org> developed by Dr. James A. Glazier and his team at the Biocomplexity Institute at Indiana University. The citation for the software is as follows:
Multi-Scale Modeling of Tissues Using CompuCell3D – M. Swat, Gilberto L. Thomas, Julio M. Belmonte, A. Shirinifard, D. Hmeljak, J. A. Glazier, Computational Methods in Cell Biology, Methods in Cell Biology 110: 325-366 (2012)
Simulations can be run from by opening the corresponding .cmd file for each run. Example: to run the first simulation run, access folder “Simulation Code>Polar1>Polar1.cc3d).” Code can be viewed through Compucell3D by selecting option “File>Start Twedit++”. This will provide access to the base python code, xml files, and the custom python scripts for each run. Variable are tabulated in the manuscript and described in comments.
To view our output for each run, open the corresponding lattice description summary file. This can be done through CompuCell3D by selecting “File>Open Lattice Description Summary File”. Example: to run output of first simulation, access folder “Simulation Lattices>Polar1>LatticeData>Polar1cc3dLDF.dml.” This will allow the software to cycle through the lattice outputs (e.g. Polar1_cc3d_00500.vtk looks at first simulation at 500 monte-carlo-steps).
2. Methods for processing the data:
Factorial design using D-optimal algorithm resulted in the 67 simulation runs used to evaluate factors important in large caliber vessel formation in our tissue models. Statgraphics centurion version 18.1.10 was used to create our optimal design. The file “QuadraticModel_scaledContinuous.dxpx” can be run on the software to view process of generating our simulation runs and the resulting quadratic model. The excel file “D-Optimal Design” Includes the output experimental run criteria in the tab “QuadraticModel_scaledContinuous.” The results from our simulations are stored in the “response” tab of the same excel file.
If you have questions about how to view the data or questions about the methods used for processing the data, please direct your questions to Michael J. Yost, PhD (yostm@musc.edu).