Research Papers

Mechanical Interaction of Angiogenic Microvessels With the Extracellular Matrix

[+] Author and Article Information
Lowell T. Edgar, Steve A. Maas

Department of Bioengineering and Scientific Computing and Imaging Institute,
University of Utah,
Salt Lake City, UT 84112

James B. Hoying

Division of Cardiovascular Therapeutics,
Cardiovascular Innovation Institute,
University of Louisville,
Louisville, KY 40202

Urs Utzinger

Department of Biomedical Engineering,
University of Arizona,
Tucson, AZ 85721;
Department of Electrical
and Computer Engineering,
University of Arizona,
Tucson, AZ 85721

Clayton J. Underwood

Department of Bioengineering
and Scientific Computing and Imaging Institute,
University of Utah,
Salt Lake City, UT 84112

Laxminarayanan Krishnan

Parker H. Petit Institute
for Bioengineering & Bioscience,
Georgia Institute of Technology,
Atlanta, GA 30332

Brenda K. Baggett

Department of Biomedical Engineering,
University of Arizona,
Tucson, AZ 85721

James E. Guilkey

Department of Mechanical Engineering,
University of Utah,
Salt Lake City, UT 84112

Jeffrey A. Weiss

Department of Bioengineering and Scientific Computing and Imaging Institute,
University of Utah,
72 South Central Campus Dr., Room 3750,
Salt Lake City, UT 84112
e-mail: jeff.weiss@utah.edu

1Corresponding author.

Contributed by the Bioengineering Division of ASME for publication in the Journal of Biomechanical Engineering. Manuscript received October 25, 2013; final manuscript received December 27, 2013; accepted manuscript posted February 5, 2014; published online February 5, 2014. Editor: Victor H. Barocas.

J Biomech Eng 136(2), 021001 (Feb 05, 2014) (15 pages) Paper No: BIO-13-1503; doi: 10.1115/1.4026471 History: Accepted February 05, 2013; Received October 25, 2013; Revised December 27, 2013

Angiogenesis is the process by which new blood vessels sprout from existing blood vessels, enabling new vascular elements to be added to an existing vasculature. This review discusses our investigations into the role of cell-matrix mechanics in the mechanical regulation of angiogenesis. The experimental aspects of the research are based on in vitro experiments using an organ culture model of sprouting angiogenesis with the goal of developing new treatments and techniques to either promote or inhibit angiogenic outgrowth, depending on the application. Computational simulations were performed to simulate angiogenic growth coupled to matrix deformation, and live two-photon microscopy was used to obtain insight into the dynamic mechanical interaction between angiogenic neovessels and the extracellular matrix. In these studies, we characterized how angiogenic neovessels remodel the extracellular matrix (ECM) and how properties of the matrix such as density and boundary conditions influence vascular growth and alignment. Angiogenic neovessels extensively deform and remodel the matrix through a combination of applied traction, proteolytic activity, and generation of new cell-matrix adhesions. The angiogenic phenotype within endothelial cells is promoted by ECM deformation and remodeling. Sensitivity analysis using our finite element model of angiogenesis suggests that cell-generated traction during growth is the most important parameter controlling the deformation of the matrix and, therefore, angiogenic growth and remodeling. Live two-photon imaging has also revealed numerous neovessel behaviors during angiogenesis that are poorly understood such as episodic growth/regression, neovessel colocation, and anastomosis. Our research demonstrates that the topology of a resulting vascular network can be manipulated directly by modifying the mechanical interaction between angiogenic neovessels and the matrix.

Copyright © 2014 by ASME
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Fig. 1

An initial microvessel fragment within a collagen matrix imaged using two-photon microscopy. Endothelial cells and pericytes were imaged using autofluorescence (green), collagen fibrils were imaged using second-harmonic generation (red). Scale bar 20 μm.

Grahic Jump Location
Fig. 2

Methods for preparing organ cultures of microvessel fragments suspended within a type-I collagen gel originally proposed by Hoying et al. [15]. (a) Rat microvessel fragments were isolated from epididymal fat pads of retired breeder Sprague–Dawley rats. (b) The pads were minced and subjected to limited digestion. (c) The solution was then sequentially filtered through 350 μm and then 30 μm sterile nylon membrane filters to remove unminced chunks of tissue and single cells from the microvessel population. (d) Microvessels were then resuspended in liquid type-I collagen. (e) The collagen solution was poured into custom rectangular Teflon molds and polymerized at 37 °C and 95% humidity for 30 min to create the three-dimensional vascularized constructs.

Grahic Jump Location
Fig. 3

Angiogenic growth within unconstrained (top) and long axis constrained (bottom) vascularized constructs, showing images of different cultures at different time points of growth after initial seeding. Unconstrained gels were free-floating and allowed to contract in all directions. For the long axis constrained constructs, contraction along the long-axis was constrained by inserting a stainless steel mesh into the gel prior to polymerization. For both conditions, neovessel sprouting typically began at Day 3 of culture and well-established vascular networks formed by Day 6. Microvessels in the unconstrained constructs grew into a randomly oriented network, while vessels in the long axis constrained construct were highly aligned along the constrained axis, indicated by the black double arrow. Scale bar 200 μm.

Grahic Jump Location
Fig. 4

Angiogenic neovessels remodel the extracellular matrix during growth [18,19]. (a) Normalized cross-sectional area of the rectangular gels at Day 1, Day 6, and Day 10 of growth. The cross-sectional area of the gel significantly decreased over time as microvessels applied traction and contracted the matrix inward. (b) Normalized levels of mRNA expression for proteolytic matrix metalloproteases at Day 0, Day 1, Day 6, and Day 10. Expression of these proteolytic enzymes significantly increased over culture time. (c) Harmonic oscillation viscoelastic testing was used to measure the dynamic stiffness of the vascularized constructs at Day 1, Day 6, and Day 10. The cultures were tested at equilibrium strain levels of 2%, 4%, and 10% at frequencies of 0.05, 0.1, 1.0, and 5.0 Hz. Increased MMP expression during growth reduced the dynamic stiffness of the cultures by Day 6. However, by Day 10 the dynamic stiffness was increased compared to the stiffness of initial cultures, even though proteolytic mRNA expression remained high. This was likely due to compaction of the matrix, reducing the fluid phase of the constructs.

Grahic Jump Location
Fig. 5

Angiogenic neovessels apply traction to the ECM and create new cell-matrix adhesions during growth [18,19]. (a) Two-photon image of a neovessel sprout, indicated by the yellow arrow. Vessels were imaged using autofluorescence (green) and the extracellular collagen matrix was imaged using second-harmonic generation (white). As neovessels applied traction and contracted the matrix, they drew collagen fibrils closer, condensing and reorientating ECM fibrils toward the neovessel tip. Scale bar 20 μm. (b) Normalized expression for matrix molecules fibronectin (FN), decorin (DCN), tenascin C (TNC), and hyaluronan (HAS) at Day 0, Day 1, Day 6, and Day 10 of growth. During angiogenesis, expression of matrix molecules that promote angiogenesis increased (FN, DCN, TNC) while expression for the angiogenesis-inhibiting molecule HAS decreased.

Grahic Jump Location
Fig. 6

Applied load and mechanical boundary conditions affect the topology of the vascular network and induce alignment [29]. (a) When vascularized constructs were free-floating and able to contract in all directions, microvessels grew into a randomly-oriented network (shape control). When a 6% static or cyclic stretch was applied along the long-axis, microvessels were highly aligned along the direction of applied stretch. However, simply constraining the long-axis of the construct (no stretch) resulted in the same amount of vascular alignment, and applying load did not produce additional alignment over the no stretch case (b) and (c). Microvessel orientation angles were measured between vessels and the long-axis (x-axis) φ and the vessels and the vertical direction (z-axis) δ. (b) Microvessels in shape control constructs exhibited no preference in φ, while vessels within the no stretch, static stretch, and cyclic stretch groups were significantly aligned along the long-axis (x-axis). (c) Microvessels in all the experimental conditions exhibited preferential alignment within the horizontal plane, as indicated by the large percentage of vessels oriented perpendicular to the vertical axis (z-axis).

Grahic Jump Location
Fig. 7

Implantation of prealigned vascularized constructs into immunocompromised mice demonstrates that preexisting vascular organization has no effect on subsequent vascular alignment [30]. (a) Confocal images of prealigned vascularized constructs at Day 7, prior to implantation (pre-implant, left). Constructs were prealigned by constraining the long-axis of the rectangular gels with a stainless steel mesh. Prior to implantation, constructs were either cut from the frame (unframed, middle) or left within the stainless steel frame (framed, right). Implants were removed and imaged at Day 30, revealing that microvessels within the unframed constructs had lost the initial prealignment and had no preferred direction while vessels within the framed constructs remained highly aligned. (b) The deformation field from the finite element models was interpolated to a skeletonized dataset of randomly-orientated microvessels in order to determine how the deformation would affect vascular alignment. (c) The anisotropy index measured using the fast Fourier transform was used to quantify alignment within the implants. An anisotropy index of 1.0 indicates high alignment, while an index of 0.0 indicates no preferred alignment. Anisotropy was measured prior to implantation at Day 7 (white bars) and after implantation at Day 30 (black bars). In the unframed implants, the anisotropy significantly decreased during implantation, indicating that vessels within these constructs lost alignment without the stainless steel boundary condition. Vessels within constructs implanted within the stainless steel frame not only maintained the original alignment but became more aligned during the implantation period. (d) Microvessel orientation predicted by the finite element models. The distribution of orientation angles with respect to the long-axis of the constructs was measured and collected in to 30 deg bins. An angle of 0 deg indicates alignment along the long-axis while an angle of 90 deg indicates alignment perpendicular to the long-axis. In the pre-implant simulation, microvessels were preferentially aligned along the long-axis of the construct. However, removal of the boundary condition in the unframed simulation caused microvessels to lose this alignment. Vessels in the framed simulation were subject to the boundary condition during the entire simulation and as a result were more highly aligned than vessels prior to implantation.

Grahic Jump Location
Fig. 8

(a) Vascularized gel cultures subjected to various mechanical boundary conditions (in order from least to most constrained): unconstrained (top left), long-axis constrained (top right), short-axis constrained (bottom right), and long-short-axis constrained (bottom left). Altering the boundary conditions of the gel had significant effects on the length, branching, and orientation of the vascular network during angiogenesis [31]. (b) The total length of the vascular network per unit of gel volume was measured in each of the boundary condition cases. No significant difference in total vascular length could be detected between UNC and LAC cultures, but there was a trend of reduced vascular length in the SAC and LSAC cultures, with length in the LSAC experiments significantly reduced. (c) The number of branching points per unit of vessel length decreased as the constructs became more constrained. (d) Microvessel orientation was measured in each boundary condition case by collecting the distribution of vessel orientation angles with respect of the long-axis of the culture. Microvessels in the LAC cultures were highly aligned along the long-axis, while vessels within the UNC, SAC, and LSAC exhibited no preference.

Grahic Jump Location
Fig. 9

Simulation of angiogenic microvessel fragments within a randomly-orientated collagen fibril network at Day 0, Day 3, and Day 6 of growth. The growth model, angio3d, uses local ECM properties to regulate neovessel elongation, branching, and anastomosis during growth.

Grahic Jump Location
Fig. 10

Our angiogenic growth model, angio3d, accurately simulates angiogenesis within numerous gel experiment conditions [32,33]. (a) (Top) Simulation of growth within an aligned fibril field [33]. Collagen fiber information was collection from long-axis constrained gel at Day 6 and used to seed a fibril field aligned along the constrained direction (horizontal). Since microvessels used local fibril orientation to determine the direction of growth, vessels grew along the constrained direction. (Bottom) The distribution of vessel orientation angle with respect to the horizontal demonstrates that vessels in these simulations were similar to vessels in long-axis constrained experiments. Scale bar 400 μm. (b) Angiogenesis is highly sensitive to the density of the matrix. Vascular networks grown in low density had more overall length, more branching, less free ends, and more interconnectivity compared networks grown in high density matrix [32]. Our growth model used local matrix density to determine the rate of growth and branching during angiogenesis and successfully simulated growth at different levels of matrix density (2.0, 3.0, and 4.0 mg/ml). Scale bar 350 μm.

Grahic Jump Location
Fig. 11

Our growth model was coupled to the nonlinear finite element software FEBio in order to simulate neovessel growth and matrix deformation [34,35]. (a) First, angio3d simulates a growth step using fibril orientation and matrix density information in the ECM field at time step n. (b) Then, a local directional sprout force field was applied to the mesh at each active sprout location. (c) FEBio was then called to solve for the deformation using the cell-generating loading scenario. (d) Lastly, the kinematics predicted by FEBio was used to update fibril orientation and matrix density in the ECM field for the next time step n + 1. This step involved displacing microvessels, reorienting ECM fibrils, and updating matrix density. The next growth step takes place in this updated ECM field. (e) Schematic of sprout force representation. To calculate force at a point inside the finite element mesh, a vector r was drawn from that point to the sprout location as described in Eq. (3). The force at that location due to the sprout was calculated using Eq. (4). This sprout force field included two terms: (1) a power cosine term that depends on the angle ψ between r and the sprout direction n, functioning to direct the force field in front of the sprout and (2) an exponential term that depends on the magnitude of r and localizes force around the sprout tip. (f) Fringe plot of a local directional sprout force field found in Eq. (4) in 2D. The parameters were a = 1.0, b = 1/250 μm−1, and N = 2. The force field is at its maximum directly in front of the sprout tip and force drops to zero away from the sprout.

Grahic Jump Location
Fig. 12

Simulation of a long-axis constrained gel using AngioFE. (a) and (b) A 1/8th symmetry model was used to simulate the vascularized construct. This figure shows the mesh viewed in the xy-plane at Day 0 (a) and Day 6 (b). The fully constrained edge of the mesh is on the right, while the geometric center of the gel where symmetry was applied is on the left. The presence of the boundary constraint causes the mesh to deform anistropically as neovessel contract the ECM, resulting in “necking,” with the most contraction occurring at the center of the gel, the furthest distance away from the boundary condition. (c) Full geometry reconstruction of a long-axis constrained gel at Day 6 using the results from the 1/8th symmetry mesh. The shape of the deformed gel predicted by the simulation closely resembled the shape of long-axis constrained gel culture at Day 6 (d).

Grahic Jump Location
Fig. 13

Simulations of long-axis constrained gels using AngioFE were able to make accurate predictions of gel contraction and microvessel alignment compared to experimental data. (a) Gel contraction was measured by calculating contractile engineering strain along the unconstrained axes at the geometric center of the gel (Eyy, Ezz). Predictions of gel contraction in the long-axis constrained simulations (gray bars) were within a single standard deviation of the experimental data (black bars). (b) and (c) Microvessel alignment was measured at the geometric center of the vascularized gels in both the simulations and experiments. The distribution of microvessel alignment angles indicate that vessels were highly aligned along the long-axis of the gel in both the experiments (triangular markers) and the simulations (square markers) (c). However, alignment predicted in the simulations was not as pronounced as seen in the experiments.

Grahic Jump Location
Fig. 14

Live imaging of angiogenic neovessels and the extracellular matrix using two-photon microscopy. Vessels were imaged using autofluorescence from green-fluorescent protein, show in red. The extracellular collagen matrix was imaged using second-harmonic generation, shown in green. Live imaging revealed that collagen is condensed into a layer surrounding the angiogenic microvessels, as demonstrated by the high intensity of SHG signal localized around the vessels. Scale bar 100 μm.

Grahic Jump Location
Fig. 15

Traction applied by neovessel sprouts remodels and condenses collagen during growth. This figure shows two-photon imaging of two neovessel sprouts, indicated by the white arrows, at culture times of 98 h (a) and 108 h (b). Neovessels were imaged via autofluorescence from green-fluorescent protein, shown in red, the extracellular collagen matrix was imaged via second harmonic generation and is shown in green. Collagen was condensed at the sprouts as neovessels applied traction and deformed the matrix as indicated by the higher collagen signal around the sprouts as time increased. Additionally, this figure demonstrates how most sprouts tended to produce multiple filopodia tips, but only elongated along a single direction. Scale bar 25 μm.

Grahic Jump Location
Fig. 16

Live two-photon imagine reveals neovessel behavior that would be difficult to investigate using traditional static culture experiments. Neovessels imaged using autofluorescence are shown in red; collagen matrix imaged using second-harmonic generation is shown in green. (a) This figure tracks the behavior of a particular neovessel sprout, indicated by the white arrow. The sprout initially branched from a parent vessel fragment. (b) The sprout elongated and extended into the matrix, its new position indicated by the arrow. However, sprouts also exhibited regression, during which the neovessel collapsed back toward the original sprouting location. A regressed sprout can be seen in this image, indicated by the asterisk. The sprout can be viewed in the previous image prior to the regression (a). (c) Sprouts exhibited both elongation and regression, and this spontaneous switching between elongation and regression caused episodic neovessel growth. In this image, the sprout has regressed back and has begun to elongate in a new direction. (d) Regression event were usually followed by elongation along a new direction, causing the neovessel to change its direction. In this image, a second neovessel sprout appeared, indicated by the second white arrow (point to the right). (e, f) Nearby neovessel sprouts seemed able to locate one another, leading to an anastomosis event indicated by the white dashed outline (f). Each of these neovessel behaviors was accompanied by condensation and remodeling of the ECM structure as seen in the second-harmonic generation data. Scale bar 50 μm. Supplemental material to be linked here April 2014 for video file of time series data.




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In