Tip cell overtaking occurs as a side effect of sprouting in computational models of angiogenesis - PubMed (original) (raw)
Tip cell overtaking occurs as a side effect of sprouting in computational models of angiogenesis
Sonja E M Boas et al. BMC Syst Biol. 2015.
Abstract
Background: During angiogenesis, the formation of new blood vessels from existing ones, endothelial cells differentiate into tip and stalk cells, after which one tip cell leads the sprout. More recently, this picture has changed. It has become clear that endothelial cells compete for the tip position during angiogenesis: a phenomenon named tip cell overtaking. The biological function of tip cell overtaking is not yet known. From experimental observations, it is unclear to what extent tip cell overtaking is a side effect of sprouting or to what extent it is regulated through a VEGF-Dll4-Notch signaling network and thus might have a biological function. To address this question, we studied tip cell overtaking in computational models of angiogenic sprouting in absence and in presence of VEGF-Dll4-Notch signaling.
Results: We looked for tip cell overtaking in two existing Cellular Potts models of angiogenesis. In these simulation models angiogenic sprouting-like behavior emerges from a small set of plausible cell behaviors. In the first model, cells aggregate through contact-inhibited chemotaxis. In the second model the endothelial cells assume an elongated shape and aggregate through (non-inhibited) chemotaxis. In both these sprouting models the endothelial cells spontaneously migrate forwards and backwards within sprouts, suggesting that tip cell overtaking might occur as a side effect of sprouting. In accordance with other experimental observations, in our simulations the cells' tendency to occupy the tip position can be regulated when two cell lines with different levels of Vegfr2 expression are contributing to sprouting (mosaic sprouting assay), where cell behavior is regulated by a simple VEGF-Dll4-Notch signaling network.
Conclusions: Our modeling results suggest that tip cell overtaking can occur spontaneously due to the stochastic motion of cells during sprouting. Thus, tip cell overtaking and sprouting dynamics may be interdependent and should be studied and interpreted in combination. VEGF-Dll4-Notch can regulate the ability of cells to occupy the tip cell position in our simulations. We propose that the function of VEGF-Dll4-Notch signaling might not be to regulate which cell ends up at the tip, but to assure that the cell that randomly ends up at the tip position acquires the tip cell phenotype.
Figures
Fig. 1
Overview of the workflow. We studied the biological relevance and the driving mechanisms of tip cell overtaking. a As a first step, we asked whether tip cell overtaking can be a side effect of sprouting. We studied tip cell overtaking in two computational models of angiogenic sprouting (the contact inhibition model and cell the elongation model), with different sprouting dynamics. We quantified tip cell overtaking and cell kinetics during simulations of these models and compared the results with similar in vitro experiments of Arima et al. [6]. b As a next step, we asked if tip cell overtaking can be regulated by VEGF-Dll4-Notch signaling. We added a VEGF-Dll4-Notch signaling network to each cell in the two models of angiogenic sprouting. Simulations are initialized with spheroids that contain a mix of wild type (WT) cells and Vegfr2 +/− cells. Due to signaling, cells can switch between four phenotypes during sprouting: WT tip cell, WT stalk cell, Vegfr2 +/− tip cell, and Vegfr2 +/− stalk cell. At the end of the simulations we quantified the percentage of sprout tips that were occupied by WT cells and compared the simulation results to experimental results of Jakobsson et al. [5]
Fig. 2
Leader identification and tip cell overtaking in the contact inhibition and cell elongation model. Sprouts formed from a spheroid in 30,000 MCS by (a) the contact inhibition model and by (b) the cell elongation model. Red cells at the sprout tips indicate the identified leader cells. Tip cell overtaking occurs in the (c) contact inhibition model as well as in (d) the cell elongation model. Two images of the same sprouts are shown for each model, with the lower sprout being at a later time point than the upper sprout. The center of mass is depicted with a colored dot for each cell and the displacement of the leader cells in time is visualized with the arrows. The mean tip cell overtake rate per sprout, calculated over 15 independent stochastic simulations, is 0.67 (±1.32) overtakes per 20,000 MCS for the contact inhibition model and 4.59 (±5.24) overtakes per 20,000 MCS for the cell elongation model
Fig. 3
Analysis of cell migration within sprouts. The position of each cell is orthogonally projected onto the sprout elongation axis and plotted against sprouting time in minutes for (a) a sprout in a murine aortic ring assay (Figure (a) is adapted from [6]), (b) in the contact inhibition model and (c) in the cell elongation model; arrows indicate tip cell overtake events. The standard deviation std(θ/π) is given for (d) anterograde moving cells (θ < π/2) and (**e**) retrograde moving cells (θ > π/2) for the experimental observations by Arima et al. [6] (exp), for the contact inhibition model (contact) and for the cell elongation model (long). f Directional motility represents the percentage of cells moving anterograde (blocked pattern), retrograde (diagonal striped pattern) or stopped (horizontally striped pattern). Mean square displacement (MSD) of cells, calculated by the projection of the center of mass on the sprout elongation axis, plotted against sprout time for (g) the contact inhibition model and for (h) the cell elongation model. The fitted curve following MSD = 2_Dt_ + (vt)2 is shown in blue, with D the dispersion coefficient and v the sprout elongation velocity
Fig. 4
Dll4 patterning by tip cell selection. a Checkerboard-like patterning of tip and stalk cells in a simulation of the contact inhibition model. The red color indicates high levels of Dll4 (tip cells) and blue indicates low levels of Dll4. b Checkerboard-like patterning of tip and stalk cells in a simulation of the cell elongation model. Figures (c–j) are images from a simulation of the contact inhibition model. c–e Enlarged view of a sprout in which branching occurs over time, at the location of the white circle in panel (c). f–h Enlarged view of two fusing sprouts (anastomosis) in time, indicated by the white circle in panel (f). i–k Enlarged view of a sprout in which tip cell overtaking occurs in time at the location of the white circle in panel (i). The cell annotated with a square overtakes the tip cell position from the cell annotated with a star
Fig. 5
Relative cell positions at sprout tips. Enlarged view of a sprout tip in a simulation of (a) the contact inhibition model and of (b) the cell elongation model. WT tip cells are colored red, Vegfr2 haploid tip cells dark purple and Vegfr2 stalk cells light purple. The leader cells of the sprouts are marked with yellow stars. The leader cell is the contact inhibition model has relatively little cell-cell contact compared to other cells in the sprout, while the leader cell in the cell elongation model is in contact with other cells for a large part of its membrane due to the multi-cellular composition of the sprout tip
Fig. 6
Leader cell identification. Schematic representation of a sprout to illustrate the identification of the leader cell. Line e is drawn through nodes B and E to find T, the furthest lattice site in the sprout on line e. Line a is perpendicular to line e and through T. The cell in which E is located and its neighbors that are on line a, are candidates to become the leader cell. The cell with the lattice site farthest from B (indicated with a star) and is connected to B through at least an equal amount of cells, will become the leader cell (indicated in red)
References
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Other Literature Sources