pmc.ncbi.nlm.nih.gov

Tissue-engineered 3D microvessel and capillary network models for the study of vascular phenomena

. Author manuscript; available in PMC: 2019 Sep 4.

Published in final edited form as: Microcirculation. 2017 Jul;24(5):10.1111/micc.12360. doi: 10.1111/micc.12360

Abstract

Advances in tissue-engineering, cell biology, microfabrication, and microfluidics, have led to the development of a wide range of vascular models. Here we review platforms based on templated microvessel fabrication to generate increasingly complex vascular models of: (1) the tumor microenvironment, (2) occluded microvessels, and (3) perfused capillary networks. We outline fabrication guidelines and demonstrate a number of experimental methods for probing vascular function such as permeability measurements, tumor cell intravasation, flow characterization, and endothelial cell morphology and proliferation.

Keywords: microvascular models, microfabrication, microfluidics, templating, self-organization, metastasis, vessel occlusion, capillary networks

Introduction

Microfabrication methods, microfluidic technology, and tissue engineering have been extensively utilized in recent decades to develop models to study human vascular function and pathology [7]. While there are no models that accurately recapitulate all aspects of the human vasculature, different models have been designed to study specific areas of vascular function such as angiogenesis [12,68], transport properties [18,64,79], shear stress [61], cancer metastasis [67], and hemodynamics [65]. By recapitulating the critical features relevant to a specific area of interest, in vitro models have enabled the study of vascular phenomena without the complicating factors associated with in vivo studies. Vascular engineering, which involves the fabrication of perfusable micovessels, is one of the most promising areas for generating models with physiological relevance that can better recapitulate in vivo conditions.

Many methods of vascular engineering rely on templating techniques to pattern empty channels or cylindrical tubes within three-dimensional gel scaffolds using removable wires/rods [9,11,51,56] or 3D printed structures [39,52,53]. Endothelial cells are then seeded into the pre-patterned channels to line and vascularized tubes within the gel. Template-based models that use thin rods to form straight cylindrical microvessels are low-cost, low-tech, and have been used by our lab and others to study vascular function, such as barrier properties [11,56,75], drug transport [8], angiogenesis [82], and shear stress [56,82].

Here we outline the templating method and adaptations thereof for creation of complex microvessel and capillary network models of vascular phenomena (Figure 1). We further discuss the significance of these models in terms of experimental and analytical techniques that can be used in each model. First we describe the templating process, and then show how this process can be modified to generate specific models.

Figure 1.

Figure 1.

Relevant design considerations for templated vascular models. VEGF-vascular endothelial growth factor, PMA- phorbol 12-myristate 13-acetate, bFGF- basic fibroblast growth factor growth factor, HGF- hepatocyte growth factor, cAMP-cyclic adenosine monophosphate.

Fabrication of Microvessels

Cylindrical microvessels fabricated by templating methods can be used to recapitulate a wide range of microenvironments and study various vascular phenomena [7,11,75]. In general, a transparent silicone elastomer, polydimethylsiloxane (PDMS), is cast in a suitable mold and plasma bonded to a glass coverslip. The rectangular compartment of the PDMS mold serves as the housing for the extracellular matrix and microvessel and is modified to connect with external tubing for perfusion (Figure 2). A cylindrical template rod, typically 75 – 250 µm in diameter, is positioned inside the housing approximately 200 – 500 µm above the glass substrate. The location of the microvessel above the glass substrate is a compromise between imaging resolution and minimizing interface effects. Locating the microvessel closer to the glass slide increases the imaging resolution by enabling the use of higher magnification microscope objectives with shorter working distances. Conversely, if the microvessel is too close to the glass slide, then interface effects from an anisotropic mechanical environment may dominate; for example, cells on a soft substrate can sense the presence of a stiff substrate within 50 – 100 µm [43].

Figure 2.

Figure 2.

Schematic illustration of template synthesis of a microvessel. A single cylindrical channel is formed by casting a solution form of extracellular matrix material (ECM) around a template rod. The rod is removed and the channel is perfused with media and endothelial cells to form a confluent vessel endothelium.

A solution form of the extracellular matrix (ECM) (e.g. collagen type I, fibrin, etc.) is introduced around the template rod and fills the PDMS housing. Optically transparent collagen gels can be made by increasing gelation temperature to 37 °C to produce uniformly smaller fiber diameters of less than 200 nm [59]. Alternatively, fibrin gels with similar optical properties can be formed by cross-linking fibrinogen with thrombin at room temperature. After gelation of the ECM material, the rod is physically removed by pulling it out from one side. Typical rod materials include stainless steel, nitinol, and glass [11,75]. Tearing of the gel during removal can be minimized using microscopically smooth template rods or by pre-coating with bovine serum albumin (BSA) to prevent adhesion to the gel [11,75]. After removal of the template rod, the empty cylindrical channel is perfused and seeded with endothelial cells that self-assemble into a continuous endothelial monolayer and ultimately, a functional microvessel. Both collagen type I and fibrin matrices are amenable to embedding live cells by mixing them into their neutralized solutions prior to injection.

Vascular flow, pressure, and stability

Unidirectional shear stress is known to induce significant morphological and biochemical changes in endothelial cells through mechanosensing [14], and improves the longevity and stability of 3D microvessels at high flow rates [56]. Pulsatile flow, as present in arterial circulation, induces reorganization of the cytoskeleton and focal adhesions in endothelial cells when compared to unidirectional shear stress [5,66]. The flow profile is dependent on the desired flow rate and hemodynamic waveform relevant for the tissue of interest. Constant flow, typical of many capillaries [54], can be applied through gravity flow where a height difference between inlet and outlet reservoirs is used to provide steady differential pressure [55]. More sophisticated recirculating fluid systems may be advantageous if flow rates are exceedingly high where liquid level sensing and computer controlled valves can precisely redirect the appropriate amount of media to maintain a certain pressure head and flow rate [75]. Pulsatile flow can also be adapted to gravity flow through the incorporation of solenoid pinch valves located either upstream or downstream of the vessel to turn vessel flow on and off electronically. The necessary flow rates to achieve typical shear stresses in arteries (10 – 70 dyne cm−2) and veins (1 – 6 dyne cm−2) can be calculated from Poiseuille’s equation assuming a straight cylindrical tube [14,46]. Upon applying flow and positive pressure at the inlet of the vessel, a transmural pressure will also be exerted across a continuous endothelium down the length of vessel. Transmural pressure improves vessel stability by increasing endothelial adhesion to the ECM and can be engineered through the addition of drainage channels [76]. In uniform straight channels, transmural pressure will decrease over the length of vessel, resulting in upstream portions exhibiting increased stability over downstream portions [56]. In the absence of drainage channels, increased stability can be achieved by tapering the vessels [56], which enhances transmural pressure upstream.

Endothelium formation

Endothelial cells (ECs) are seeded into the channels at a density of around 107 cells mL−1 and allowed to reach confluence. Commonly used human endothelial cell lines include human umbilical vein ECs (HUVECs) and dermal microvascular ECs (HMVEC), which form well-defined endothelial monolayers [11,75]. Enhanced endothelial seeding and faster monolayer formation (< 1 day) can be achieved by coating the interior of the ECM channels with solubilized fibronectin (50 µg mL−1) prior to introducing a suspension of ECs. After seeding, sub-confluent monolayers may require 1 – 3 days to proliferate and fully line the interior of the vessel. Formation of continuous cell-cell adherens junctions within the endothelium can be assessed by staining for proteins such as VE-cadherin (Figure 3A).

Figure 3.

Figure 3.

Microvessel functionality. (A) 3D projection from confocal z-stacks of a HUVEC microvessel fluorescently stained for VE-cadherin (green) and DAPI (blue). Maximum intensity projection of the lower half of the vessel (right). (B) Phase-contrast and fluorescence time-series depicting 70 kDa dextran (green) perfused through a HUVEC microvessel demonstrating retention of the tracer molecule and slow leakage into the surrounding ECM. (C) Corresponding fluorescence image intensity over time, showing a steady increase in tracer molecule accumulation in the ECM. The permeability can be calculated from the intensity profile. (D) Fluorescence time-series showing a focal leak of 70 kDa dextran from the microvessel that dissipates over 8 minutes.

Barrier function

After formation of the endothelium, vessel functionality can be ascertained by measuring the permeability to fluorescent tracer molecules. Tracer molecules of various sizes are introduced through the vessel lumen, and their rate of leakage into the surrounding ECM can be used to benchmark vessel barrier function (Figure 3B) [11]. Barrier function is usually described in terms of permeability, P (cm s−1), which is a measure of the global rate of transendothelial transport.

In a typical experiment, a tracer molecule, such as a fluorescently labeled dextran (3 kDa - 2 MDa) or small molecule dye (e.g. FITC, TxRed, Lucifer yellow) is introduced at a concentration of about 5 – 50 µg mL−1 for 60 minutes. The flow rate should be sufficiently high to ensure that the concentration in the microvessel remains constant during the experiment. For high permeabilities, higher flow rates are needed to maintain a constant concentration within the vessel. This is analogous to limitations in measuring the permeability of fast transporting small molecules used during in situ brain perfusion [76]. Using a sampling rate of 2 minutes usually provides sufficient resolution to determine the global permeability. Analysis typically involves identifying a region of interest that includes a section of the microvessel, and plotting the fluorescence intensity versus time (Figure 3C). For simple passive diffusion in a microvessel with an intact, continuous endothelium, the fluorescence intensity will show a steep increase, due to injection of the fluorescent probe, followed by a slower linear increase as the molecule diffuses radially into the surrounding matrix. Details for the determination of permeability have been reported elsewhere [55].

The permeability to fluorescent tracers is expected to be relatively low in a quiescent endothelium but higher in an activated endothelium, for example due to inflammation [11]. Similarly, tumor vasculature is leakier than normal tissue and exhibits a wide range of permeability coefficients due to heterogeneous vessel and tumor characteristics [7,80]. In some cases, barrier function may be almost non-existent resulting in the extravasation of large particles and red blood cells, or exhibit relatively normal functionality along the tumor periphery [31,80]. In vivo vessel permeability coefficients range from 10−8 to 10−6 cm s−1 for BSA or 70 kDa dextran depending on the state of the vessel [7].

The permeability is a measure of global transport across the endothelium, but provides little insight into the mechanism. Analysis of fluorescent tracer experiments can be used to extract more detailed information on transport mechanisms [8,55]. For example, transient focal leaks, or small plumes of tracer molecules, are frequent in activated microvasculature [11,19,30], and are revealed during time-lapse imaging of microvessels to characterized vessel functionality on a local scale (Figure 3D). Furthermore, some endothelial or epithelial barriers, such as those modeling the blood-brain barrier, incorporate polarized efflux pumps on the luminal surface, and more complex models are required to analyze apparent changes in permeability [4,8]; function blocking antibodies can be used to identify these contributions from efflux pumps [44].

Barrier function studies can also be adapted to study drug/gene delivery. For example by seeding target cells (e.g. tumor cells) into the surrounding matrix, uptake and cell fate can be visualized directly [49]. By quantifying the rate of cell division and/or apoptosis before and after the introduction of a drug into the flow loop, details of uptake in target cells and cell fate can be directly imaged in an environment that mimics systemic delivery and correlated to transport across the endothelium.

Quiescence and activation

In vivo vascular homeostasis describes a complex equilibrium state that incorporates numerous aspects of vascular function. A major challenge in developing vascular models is assessing whether relevant aspects of homeostasis have been achieved. In the literature, the term quiescence is often used to describe an inactive state, implying recapitulation of aspects of endothelium homeostasis. This is a complex issue since endothelial cells exhibit broad molecular heterogeneity and respond to a wide range of input stimuli including biochemical (e.g. small molecules, hormones, proteins, and cells) and physical cues (e.g. hemodynamic shear stress, oxygen, and curvature) [1,2]. Quiescence in confluent endothelial monolayers is usually described in terms of expression of key endothelial markers, contact inhibition, low activity, and low turnover. Activation of the endothelium is usually measured qualitatively through relative changes in these parameters. While relative changes in endothelial cell activity that mimic an injury response are easily measured, the baseline state itself is not well defined. Microvessel models enable creation of in vitro vasculature with the correct cylindrical geometry and flow, which is known to contribute to vascular homeostasis [1,2].

Recent developments in live cell microscopy and image analysis provide many tools for quantitative analysis of parameters associated with endothelium activity, including cell proliferation, apoptosis, and motility [61]. Time-lapse videos in phase and fluorescence obtained at 10 minute intervals provides sufficient resolution to enumerate mitosis and cell detachment events and track individual endothelial cells to quantify migration speeds (Figure 4).

Figure 4.

Figure 4.

Imaging endothelial proliferation, cell loss, and morphology within the vessel monolayer. (A) Illustration of the imaging plane focused on the bottom of the vessel and projected in 2D with a representative phase-contrast image (bottom). (B) Time-series images depicting mitosis and apoptosis of VeraVec-GFP endothelial cells in phase-contrast and GFP (green). Arrows indicate mitotic and apoptotic cells. (C) Time-series showing VeraVec alignment in the direction of flow in response to shear stress increased from 7 to 14 dyne cm−2 over 45 hours. Flow is from left to right in all images.

Measuring proliferation and apoptosis rates

Quiescence is often taken to mean low turnover, implying that endothelial cells are not responding to signals that promote an injury response. In assessing the quiescence of an endothelial monolayer, both cell proliferation and apoptosis are important, since together they define turnover and the net change in the number of cells in the monolayer. Fluorescence or staining assays (e.g. BrdU, propidium iodide) give the average rate over the experiment duration but multiple experiments are needed to obtain dynamic information [20,32,69,71,77]. For analysis of proliferation and apoptosis rates, phase contrast images from movies can be analyzed at the bottom of the microvessel away from projected edges (Figure 4A); thus avoiding errors associated with the curvature. From phase contrast movies, cell division and apoptosis events are readily visualized when sampling at 10 minute intervals (Figure 4B), and reliable quantitative data can be obtained by visually examining each frame and counting the number of cell division and apoptosis events. This method allows analysis of proliferation and apoptosis events in real time, in contrast to traditional fluorescence or staining assays. Typical values for proliferation and apoptosis rates in 2D, 3D, and in vivo are reported in Table 1. Cells are analyzed from a region approximately 4 – 5 cells wide. For a typical cell length of 20 – 50 µm, we can analyze ≥ 100 cells in a typical frame at 20X magnification (about 1 mm length of vessel). Statistics can be improved by imaging at multiple regions along the length of the vessel (total length 2 cm). Imaging and analysis is performed to quantify the spatial gradient in endothelial structure and function surrounding the occlusion. For analytical purposes, the vessel images are segmented into 500 µm long regions and parameters of interest will be determined individually for each region.

Table 1.

Endothelial proliferation and apoptosis rates within continuous monolayers in 2D flow chambers and 3D microvessels.

Proliferation Rate (% h−1) Apoptosis Rate (% h−1) Source
2D monolayer 0.2–0.4 0.05–0.6 HUVEC, 4–16 dyne cm−2
3D microvessels 0.4–0.7 0.1–0.2 VeraVec ECs, 2–6 dyne cm−2
In vivo 0.01–0.7 N/A [20,32,71]

Measuring cell motility and morphological parameters

Cell motility and morphology are also useful parameters in assessing the state of the endothelium. In EC monolayers, cell speed is typically in the range from 0.02 – 0.2 µm min−1 [61,75]. Cell speed increases in response to mitogenic factors (e.g. EGF, FGF, VEGF, etc.) and shear forces, and hence can be considered a measure of cell activity [28,66]. Cell speed can be evaluated using the MATLAB-based particle image velocimetry application, OpenPIV [72,75]. Based on validation experiments, the PIV speed is within about ±50% of values obtained from manually tracking the position of individual cells [61]. Other parameters associated with migration, such as persistence length and directedness, can also be obtained from live cell imaging [35].

Cell morphology is also an important parameter associated with endothelial homeostasis. In straight sections of larger vessels, endothelial cells are elongated and aligned in the direction of flow [2,14,15]. This property has been recapitulated in vitro, where many endothelial cell types in 2D monolayers show a transition from a cobblestone morphology to a spindle-like morphology in response to shear stress [5,15,40]. Interestingly, brain microvascular endothelial cells appear to be an exception [61]. Parameters associated with cell morphology, such as cell area, orientation with respect to flow, and inverse aspect ratio, can be obtained from live cell videos and automated image analysis [61]. We have observed VeraVec endothelial alignment in 3D microvessels within 48 hours under unidirectional flow from applied shear stresses increased from 7 to 14 dyne cm−2 (Figure 4C and Supplementary Video 1). Quantifying changes in cell morphology over time enables the understanding of the dynamic response of individual endothelial cells within a confluent monolayer to shear stress and vascular modulators for the study of quiescence and endothelial activation.

The Tumor Microenvironment

Metastasis is responsible for the majority of cancer-related deaths [48]. Metastatic tumor cells are able to achieve widespread dissemination in the body through critical steps, such as intravasation and extravasation, which are the entry and exit of tumor cells from the vascular system [29]. The intravasation of single tumor cells has been visualized in mice using intravital microscopy [30,78,83], and has implicated tumor associated macrophages (TAMs) in facilitating the invasion and transendothelial migration (TEM) of tumor cells [30,78]. Similar observations of single tumor cell intravasation and the influence of TAMs have been reported in vitro across orthogonally grown endothelial monolayers [81]. However, invasion and intravasation may proceed independently of TAMs [57,83], and the significant presence and contribution of circulating tumor cell clusters to cancer progression suggests the likelihood of multiple mechanisms of intravasation across different vascular beds and tumor microenvironments [6]. The physical and biological details that regulate intravasation have not been fully characterized and relatively few events are reported in detail, in part because these processes are dynamic and occur at the interface between the tumor and its local microvasculature, which is a complex 3D microenvironment interspersed throughout tumor tissue and downstream organs. Recent advances in the development of in vitro microvessel models provide the tools to recreate the essential components of the tumor microenvironment and enable visualization of the details of the metastatic cascade [7,36,75]. The ability to recapitulate perfusable human microvasculature within a precisely controlled tissue engineered model provides unique tools for visualizing and understanding the physical and biological factors governing metastasis [36].

Since tumor vasculature is formed relatively quickly through unregulated angiogenesis, it is often lacking components present in normal microvessels, such as smooth muscle cells, pericytes, and lymphatic drainage [29]. As a result, tumor microvasculature is typically leakier than normal vessels [23] and the majority of tumor cell intravasation is correlated with tumor microvessels larger than 30 µm [42]. Using the template microvessel model, tumor cells can be seeded into the extracellular matrix (ECM) surrounding the microvessel to simulate the tumor microenvironment (Figure 5) [75]. Tumor cells within the surrounding ECM can embedded as single or clusters of cells to model single cell or collective invasion (Figure 5A). To distinguish the interactions between tumor and endothelial cell types during live cell microscopy, fluorescently labeled proteins, such as histone-H2B-GFP enables identification and quantification of tumor cell proliferation in the ECM, while cytoplasmic RFP improves the observation of extended protrusions and invadopodia (Figure 5D) [33]. Images acquired with either confocal or wide-field microscopy are often focused in the midsection of the vessel, which is projected as a 2D channel with the vessel lumen exposed (Fig. 5E–F). The introduction of tumor cells into the ECM solution prior to injection will result in a random distribution of tumor cells at varying distances from the vessel wall. At a cell density of 5 × 105 cells mL−1 in the ECM, tumor cells often overrun the devices within 1 – 2 weeks. Typically, a small fraction of tumor cells will be in contact with the channel prior to endothelial formation of a functional vessel; therefore, permeability measurements along the length of vessel are important for verifying vessel functionality before any significant tumor-endothelium interactions occur.

Figure 5.

Figure 5.

Tumor-microvessel models of metastasis. (A) Illustration of platform observed from the side with tumor cells incorporated within the ECM. (B) Representative image of device connected to external tubing and flow. (C) Illustration of platform in 3D with the imaging plane focused in the middle of the vessel. (D) Representative 3D projection obtained from confocal fluorescence microscopy of a vessel lined with VeraVec-GFP (green) endothelial cells and a dual-labeled MDA-MB-231 breast cancer cell (nucleus (green), cytoplasm (red)) located on the periphery. (E) Illustration of a 2D projected image obtained from focusing at the middle of the vessel. (F) Wide field fluorescence images overlaid with phase-contrast focused at the middle of the vessel.

Using templated microvessels to model the tumor microenvironment, invasion and intravasation of single tumor cells can be visualized [75]. Intravasating tumor cells in the ECM first encounter the interface between the ECM and microvessel (Figure 6A and Supplementary Video 2) and may spend significant time residing at the interface and invading along the vessel before transendothelial migration and detachment occur. These observations of tumor cell dispersal along the ECM-vessel interface are consistent with the migration of glioblastoma cells along brain microvessels [10]. Intravasation is often preceded by cell-rounding that leads to sudden transendothelial migration and detachment into flow (Figure 6B,C and Supplementary Video 3) [75]; this is in contrast to conventional depictions of single cell intravasation, which show tumor cell processes extended across the vessel endothelium and diapedesis similar to neutrophils and macrophages [6]. Observed intravasation rates of single tumor cells across functional microvessels or endothelium are low [30,81], but when normalized to the number of microenvironments that promote or enable intravasation, such as the close proximity or contact of TAMs, tumor cells, and microvessels, termed tumor microenvironment of metastasis (TMEM), the frequency has been reported as 0.01 h−1 in mice [30]. Here, we demonstrate intravasation without the presence of TAMs across functional endothelium, and can similarly quantify an intravasation frequency pertaining to tumor cells located at the ECM-vessel interface.

Figure 6.

Figure 6.

Time-series images of invasion and intravasation within a tumor-microvessel model. (A) Dual labeled MDA-MB-231 breast cancer cell (nucleus (green), cytoplasm (red) invades into the ECM-microvessel interface. VeraVec-GFP (green) microvessel. (B) Breast cancer cell intravasation and detachment into vessel flow. HMVEC microvessel labeled with BSA-488 (green). (C) An orthogonal view of breast cancer cell invasion into the ECM-microvessel interface and detachment into flow. Image series adapted from reference [75]. HMVEC microvessel labeled with BSA-488 (green).

Occluded Microvessels

Various pathological conditions, such as stroke, are associated with vessel occlusion or turbulent flow [26,45,47,60,73,84]. Atherosclerosis is often associated with non-laminar flow dynamics at vessel bifurcations or in regions with the build-up of plaques, leading to activation and structural changes in endothelial cells [60,74]. The progression of atherosclerosis and thromboembolic events begins with an initial injury to the endothelial cell layer (AHA type I lesion) [63]. This initial injury can take many forms and results in macrophage infiltration through histologically normal endothelial cells [62,70]. We have developed a perfusable tissue engineered microvessel model which can be manually occluded, allowing detailed quantitative analysis of parameters of interest including endothelial cell morphology, activity, protein expression, and barrier function in real time These tools enable in vitro investigation of initial endothelial cell injury under occluded and non-occluded microvessels. This model is ideal for investigating endothelial cell homeostasis and dysfunction in occluded and non-occluded microvessels. Recapitulating atherosclerosis or its progression is more challenging and will require further increases in model complexity.

Recapitulating the 3D geometry of an occlusion in vitro permits direct observation and quantification of endothelial cell behavior in and around regions of non-laminar flow similar to in vivo conditions. The details of the pathogenesis of atherosclerosis are not completely understood but EC dysfunction is both a symptom and possible cause of the disease [21,62]. In vivo studies have shown EC dysfunction and complete endothelium retraction, exposing connective tissue and smooth muscle cells, and is a key symptom in the progression of atherosclerosis in primate models [22,27,74]. Accumulating evidence shows a link between regions with non-laminar flow patterns, as seen in occluded vessels, and changes in endothelium function [47]. It has also been shown that an activated endothelium, characterized by increased cell proliferation rates, leads to expression of factors that promote macrophage and platelet adhesion and penetration, furthering atherosclerotic symptoms [16,17,24,25].

Occlusions can be modeled by incorporating a rod with a rounded tip connected to an actuator (Eppendorf InjectMan) orthogonal to the microvessel model (Figure 7). Rods are cast within the ECM such that the rod tips are about 100 µm from either side of the template rod. Precise movement of the occlusion rod is achieved using a custom actuator head with fine motor actuation (20 nm step). The rod is then steadily moved towards the vessel to achieve the desired level of occlusion. By advancing one or both of the rods, the microvessel lumen can be compressed introducing a well-defined constriction and occluding flow. Disturbed flow patterns downstream of the occlusion are verified by introducing fluorescent microspheres into the media. This model can be used to study the influence of human plasma, such as components of the coagulation cascade, and anticoagulant drugs on endothelial cell parameters of interest in real-time.

Figure 7.

Figure 7.

Occluded microvessel model. (A) Schematic illustration of microvessel occlusion model. (B) Access rods actuated to occlude microvessel. (C) Phase contrast image of microvessel and access rod before occlusion. (D) Phase contrast image of partially occluded microvessel. (E) Fluorescence image of a VeraVec microvessel before occlusion. (F) Image of an occluded VeraVec microvessel. (G) Fluorescence image of microvessel perfusion with microspheres. (H) Fluorescence image of an occluded microvessel perfused with microspheres.

Capillary Networks

The templating method for microvessel formation is normally restricted to relatively large diameter vessels (> 20 μm) due to the difficulty in seeding cells in small cylindrical tubes without clogging [41,82]. However, capillary networks can be formed between two parallel microvessels by exploiting stimulated angiogenesis and anastomosis or self-assembly [12,34,38] (Figure 8A,B). The two microvessels can be independently perfused by connecting respective ports at each end. Following the formation of a confluent monolayer of endothelial cells, one of the channels is perfused with media supplemented with growth factors (e.g. VEGF, bFGF) at low shear stress (e.g. 2 dyne cm−2) while the second vessel is maintained under static conditions resulting in initiation and growth of angiogenic sprouts from both of the microvessels (Figure 8C,D) that develop into capillaries (Figure 8E–I).

Figure 8.

Figure 8.

Perfused capillary network model. (A) Schematic illustration showing a capillary network formed between two microvessels. (B) Schematic illustration showing flow conditions for optimizing capillary formation. (C) Fluorescence image showing two VeraVec microvessels (prior to capillary formation) with dextran supplemented media in one vessel to confirm independent perfusion. (D) Image showing capillary sprouting from a VeraVec microvessel. (E) Anastomosing capillary, and (F) DAPI stain of the same capillary comprised of ~9 cells. (G) Confocal image showing that capillaries form well defined lumens. (H) Fluorescence image of a long, branched capillary. (I) Fluorescence images showing capillary network (top) and perfusion of the same network with fluorescent microbeads (bottom) to identify perfused, functional capillaries.

The selection of ECM, concentration, and the extent of cross-linking all affect the degree of sprouting and capillary growth rate [3,13,37]. In general, low matrix concentrations promote angiogenic sprouting, however, higher concentrations are preferred to maintain robust microvessels that are stable under flow. Gel concentrations of about 4 mg mL−1 generally provide an acceptable compromise between these two factors. In addition, matrix composition and stiffness also modulate capillary structure and growth rate. Capillary networks formed in low stiffness gels tend to be thinner and denser than those formed in higher stiffness gels [58]. Capillary growth rates are often faster in fibrin gels compared to collagen gels of the same concentration [58]. The formation and stability of longer capillaries (≥ 500 µm) requires co-culture with fibroblasts (e.g. normal human lung fibroblasts) or supplementing with fibroblast growth factor conditioned media [39,49,50]. Following sprouting and growth, many of the capillaries that make contact with the second vessel become perfused (Figure 8I and Supplementary Video 4). In general, capillary diameter increases over time, particularly in short capillaries formed in collagen gels. The resulting hierarchical capillary network with upstream and downstream microvessels (Figure 8I) mimics the geometry of the microvasculature in many regions of the body (i.e. cerebral cortex and myocardium).

Capillary Perfusion

The perfusion of capillary networks can be assessed using a variation of the x-ray angiogram. In the in vitro angiogram, capillary function is assessed by perfusing a fluorescently-labeled dextran, as opposed to using a radio-opaque dye, through the upstream microvessel and recording fluorescence images every 2 minutes for up to 20 minutes. The capture rate is optimized to maximize the time-resolution while limiting photobleaching of the fluorescent probe. On introducing the fluorescently labeled dextran into the upstream microvessel, the dextran first fills the microvessel. In the absence of a capillary network, the dextran remains largely confined to the lumen of the vessel (Figure 9A) while a small amount diffuses into the surrounding ECM at a relatively slow rate (Figure 9A,C). In the capillary network model, the dextran begins to perfuse the tissue at a much faster rate (Figure 9B,C). Perfusion can be defined as the perfusion ratio, which is the ratio of the fluorescence intensity in vascularized tissue compared to un-vascularized tissue at a given time (Figure 9D). The perfusion ratio, quantified over time, is used to assess the functionality of capillary networks under various conditions. This technique and similar methods may allow greater insight into disorders involving capillary dysfunction (e.g. capillary no-reflow following ischemia).

Figure 9.

Figure 9.

In vitro capillary flow assay. (A) A region of the microvasculature lacking capillaries and sequential fluorescence time-lapse images of dextran perfusion. (B) A region containing a capillary bed. (C) Analysis of fluorescence images showing ~6.5X difference in dextran accumulation in vascularized tissue (blue) compared un-vascularized tissue (red) after 15 minutes. The capillary perfusion ratio (CPR) is about 6.5 at 15 minutes. (D) Analysis showing dextran intensity in the downstream microvessel in vascularized and un-vascularized tissue. When dextran is detected in the downstream microvessel, full perfusion of the capillary bed has occurred.

Conclusion and Perspectives

Tissue engineered microvessel and capillary network models enable in vitro imaging of vascular phenomena in physiologically relevant and customizable platforms. Here, we described how template-based microvessels can be used to model cancer metastasis, vessel occlusion, and capillary network perfusion. Across these microvessel models, vascular functionality (i.e. permeability, perfusion ratio), quiescence (e.g. proliferation, apoptosis, migration, morphology), and activation are key parameters for determining baseline conditions and response to modulators (e.g. growth factors, chemotherapeutics, targeted drug therapies) and the induction of a disease state. Understanding the influence of physical (e.g. geometry, fluid flow, matrix stiffness, etc.) and biological (e.g. co-culture, media components, etc.) parameters on microvessel and capillary formation within these models provides informative details for engineering perfused organs. As these models are refined to provide quantitative metrics for the biological and physical characteristics and regulators for vascular diseases, the incorporation of primary patient cell lines and iPSC derived lines will enhance our mechanistic understanding of vascular disease progression and paves the way for their potential use as prognostic tools to inform patient treatment.

Supplementary Material

Supp VideoS1

Supp VideoS2

Supp VideoS3

Supp VideoS4

Supp info

Acknowledgements

The authors gratefully acknowledge support from the American Heart Association (15GRNT25090122) and DTRA (HDTRA1-15-1-0046). MB acknowledges a fellowship from the American Heart Association (17PRE33460316).

Sources of support: American Heart Association (15GRNT25090122) and DTRA (HDTRA1-15-1-0046).

Abbreviations used:

ECM

extracellular matrix

EC

endothelial cells

PDMS

polydimethylsiloxane

HUVEC

human umbilical vein endothelial cells

HMVEC

human dermal microvascular endothelial cells

BSA

bovine serum albumin

BSA-488

bovine serum albumin conjugated to Alexa Fluor® 488 dye

GFP

green fluorescence protein

VeraVec

stably transfected HUVECs by Angiocrine Bioscience

TMEM

Tumor Microenvironment of Metastasis

Footnotes

Financial disclosure

None

Conflict of Interest

None

References

  • 1.Aird W. Spatial and temporal dynamics of the endothelium. Journal of Thrombosis and Haemostasis 3: 1392–1406, 2005. [DOI] [PubMed] [Google Scholar]
  • 2.Aird WC. Phenotypic heterogeneity of the endothelium II. Representative vascular beds. Circulation research 100: 174–190, 2007. [DOI] [PubMed] [Google Scholar]
  • 3.Allen P, Melero-Martin J, Bischoff J. Type I collagen, fibrin and PuraMatrix matrices provide permissive environments for human endothelial and mesenchymal progenitor cells to form neovascular networks. Journal of tissue engineering and regenerative medicine 5: e74–e86, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Bachmeier CJ, Trickler WJ, Miller DW. Comparison of drug efflux transport kinetics in various blood-brain barrier models. Drug metabolism and disposition: the biological fate of chemicals 34: 998–1003, 2006. [DOI] [PubMed] [Google Scholar]
  • 5.Blackman BR, Garcia-Cardena G, Gimbrone MA Jr. A new in vitro model to evaluate differential responses of endothelial cells to simulated arterial shear stress waveforms. J Biomech Eng 124: 397–407, 2002. [DOI] [PubMed] [Google Scholar]
  • 6.Bockhorn M, Jain RK, Munn LL. Active versus passive mechanisms in metastasis: do cancer cells crawl into vessels, or are they pushed? The Lancet Oncology 8: 444–448, 2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Bogorad MI, DeStefano J, Karlsson J, Wong AD, Gerecht S, Searson PC. Review: in vitro microvessel models. Lab on a Chip 15: 4242–4255, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Bogorad MI, Searson PC. Real-time imaging and quantitative analysis of doxorubicin transport in a perfusable microvessel platform. Integrative Biology 8: 976–984, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Buchanan CF, Voigt EE, Szot CS, Freeman JW, Vlachos PP, Rylander MN. Three-dimensional microfluidic collagen hydrogels for investigating flow-mediated tumor-endothelial signaling and vascular organization. Tissue Engineering Part C: Methods 20: 64–75, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Burden-Gulley SM, Qutaish MQ, Sullivant KE, Lu H, Wang J, Craig SE, Basilion JP, Wilson DL, Brady-Kalnay SM. Novel cryo-imaging of the glioma tumor microenvironment reveals migration and dispersal pathways in vivid three-dimensional detail. Cancer Res 71: 5932–5940, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Chrobak KM, Potter DR, Tien J. Formation of perfused, functional microvascular tubes in vitro. Microvascular research 71: 185–196, 2006. [DOI] [PubMed] [Google Scholar]
  • 12.Chung S, Sudo R, Mack PJ, Wan C-R, Vickerman V, Kamm RD. Cell migration into scaffolds under co-culture conditions in a microfluidic platform. Lab on a Chip 9: 269–275, 2009. [DOI] [PubMed] [Google Scholar]
  • 13.Critser P, Kreger S, Voytik-Harbin S, Yoder M. Collagen matrix physical properties modulate endothelial colony forming cell-derived vessels in vivo. Microvascular research 80: 23–30, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Davies PF. Flow-mediated endothelial mechanotransduction. Physiological reviews 75: 519–560, 1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Dewey C, Bussolari S, Gimbrone M, Davies PF. The dynamic response of vascular endothelial cells to fluid shear stress. Journal of Biomechanical Engineering 103: 177–185, 1981. [DOI] [PubMed] [Google Scholar]
  • 16.DiCorleto PE, Bowen-Pope DF. Cultured endothelial cells produce a platelet-derived growth factor-like protein. Proceedings of the National Academy of Sciences 80: 1919–1923, 1983. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Dicorleto PE, Gajdusek CM, Schwarth SM, Ross R. Biochemical properties of the endothelium-derived growth factor: Comparision to other growth factors. Journal of cellular physiology 114: 339–345, 1983. [DOI] [PubMed] [Google Scholar]
  • 18.Douville NJ, Tung Y-C, Li R, Wang JD, El-Sayed ME, Takayama S. Fabrication of two-layered channel system with embedded electrodes to measure resistance across epithelial and endothelial barriers. Analytical chemistry 82: 2505–2511, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Dreher MR, Liu W, Michelich CR, Dewhirst MW, Yuan F, Chilkoti A. Tumor vascular permeability, accumulation, and penetration of macromolecular drug carriers. Journal of the National Cancer Institute 98: 335–344, 2006. [DOI] [PubMed] [Google Scholar]
  • 20.Engerman RL, Pfaffenbach D, Davis M. Cell turnover of capillaries. Laboratory investigation; a journal of technical methods and pathology 17: 738–743, 1967. [PubMed] [Google Scholar]
  • 21.Epstein FH, Ross R. Atherosclerosis—an inflammatory disease. New England Journal of Medicine 340: 115–126, 1999. [DOI] [PubMed] [Google Scholar]
  • 22.Faggiotto A, Ross R. Studies of hypercholesterolemia in the nonhuman primate. II. Fatty streak conversion to fibrous plaque. Arteriosclerosis, thrombosis, and vascular biology 4: 341–356, 1984. [DOI] [PubMed] [Google Scholar]
  • 23.Folarin AA, Konerding MA, Timonen J, Nagl S, Pedley RB. Three-dimensional analysis of tumour vascular corrosion casts using stereoimaging and micro-computed tomography. Microvasc Res 80: 89–98, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Furchgott RF. Role of endothelium in responses of vascular smooth muscle. Circulation research 53: 557–573, 1983. [DOI] [PubMed] [Google Scholar]
  • 25.Gajdusek C, DiCorleto P, Ross R, Schwartz SM. An endothelial cell-derived growth factor. The Journal of cell biology 85: 467–472, 1980. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.General OotS, National Heart L, Institute B. SECTION I: Deep Vein Thrombosis and Pulmonary Embolism as Major Public Health Problems 2008. [Google Scholar]
  • 27.Goldstein JL, Kita T, Brown MS. Defective lipoprotein receptors and atherosclerosis. Lessons from an animal counterpart of familial hypercholesterolemia. The New England journal of medicine 309: 288, 1983. [DOI] [PubMed] [Google Scholar]
  • 28.Gospodarowicz D, Mescher AL, Birdwell CR. Stimulation of Corneal Endothelial Cell-Proliferation Invitro by Fibroblast and Epidermal Growth-Factors. Exp Eye Res 25: 75–89, 1977. [DOI] [PubMed] [Google Scholar]
  • 29.Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 144: 646–674, 2011. [DOI] [PubMed] [Google Scholar]
  • 30.Harney AS, Arwert EN, Entenberg D, Wang Y, Guo P, Qian BZ, Oktay MH, Pollard JW, Jones JG, Condeelis JS. Real-Time Imaging Reveals Local, Transient Vascular Permeability, and Tumor Cell Intravasation Stimulated by TIE2hi Macrophage-Derived VEGFA. Cancer discovery 5: 932–943, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Hashizume H, Baluk P, Morikawa S, McLean JW, Thurston G, Roberge S, Jain RK, McDonald DM. Openings between defective endothelial cells explain tumor vessel leakiness. The American journal of pathology 156: 1363–1380, 2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Hobson B, Denekamp J. Endothelial proliferation in tumours and normal tissues: continuous labelling studies. British journal of cancer 49: 405, 1984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Hoffman RM. The multiple uses of fluorescent proteins to visualize cancer in vivo. Nature reviews Cancer 5: 796–806, 2005. [DOI] [PubMed] [Google Scholar]
  • 34.Hsu Y-H, Moya ML, Hughes CC, George SC, Lee AP. A microfluidic platform for generating large-scale nearly identical human microphysiological vascularized tissue arrays. Lab on a Chip 13: 2990–2998, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Huang YJ, Samorajski J, Kreimer R, Searson PC. The influence of electric field and confinement on cell motility. PloS one 8: e59447, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Katt ME, Placone AL, Wong AD, Xu ZS, Searson PC. In Vitro Tumor Models: Advantages, Disadvantages, Variables, and Selecting the Right Platform. Front Bioeng Biotechnol 4: 12, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Kniazeva E, Kachgal S, Putnam AJ. Effects of extracellular matrix density and mesenchymal stem cells on neovascularization in vivo. Tissue engineering Part A 17: 905–914, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Lee H, Kim S, Chung M, Kim JH, Jeon NL. A bioengineered array of 3D microvessels for vascular permeability assay. Microvascular research 91: 90–98, 2014. [DOI] [PubMed] [Google Scholar]
  • 39.Lee VK, Lanzi AM, Ngo H, Yoo S-S, Vincent PA, Dai G. Generation of multi-scale vascular network system within 3D hydrogel using 3D bio-printing technology. Cellular and molecular bioengineering 7: 460–472, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Levesque M, Nerem R. The elongation and orientation of cultured endothelial cells in response to shear stress. Journal of Biomechanical Engineering 107: 341–347, 1985. [DOI] [PubMed] [Google Scholar]
  • 41.Linville RM, Boland NF, Covarrubias G, Price GM, Tien J. Physical and chemical signals that promote vascularization of capillary-scale channels. Cellular and molecular bioengineering 9: 73–84, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Liotta LA, Kleinerman J, Saidel GM. Quantitative relationships of intravascular tumor cells, tumor vessels, and pulmonary metastases following tumor implantation. Cancer Res 34: 997–1004, 1974. [PubMed] [Google Scholar]
  • 43.Lo CM, Wang HB, Dembo M, Wang YL. Cell movement is guided by the rigidity of the substrate. Biophysical journal 79: 144–152, 2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Loscher W, Potschka H. Blood-brain barrier active efflux transporters: ATP-binding cassette gene family. NeuroRx : the journal of the American Society for Experimental NeuroTherapeutics 2: 86–98, 2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Mahan CE, Borrego ME, Woersching AL, Federici R, Downey R, Tiongson J, Bieniarz MC, Cavanaugh BJ, Spyropoulos AC. Venous thromboembolism: annualised United States models for total, hospital-acquired and preventable costs utilising long-term attack rates. Thrombosis and haemostasis 108: 291–302, 2012. [DOI] [PubMed] [Google Scholar]
  • 46.Malek AM, Alper SL, Izumo S. Hemodynamic shear stress and its role in atherosclerosis. Jama 282: 2035–2042, 1999. [DOI] [PubMed] [Google Scholar]
  • 47.Mannino RG, Myers DR, Ahn B, Wang Y, Rollins M, Gole H, Lin AS, Guldberg RE, Giddens DP, Timmins LH. Do-it-yourself in vitro vasculature that recapitulates in vivo geometries for investigating endothelial-blood cell interactions. Scientific reports 5, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Mehlen P, Puisieux A. Metastasis: a question of life or death. Nature reviews Cancer 6: 449–458, 2006. [DOI] [PubMed] [Google Scholar]
  • 49.Moya M, Tran D, George SC. An integrated in vitro model of perfused tumor and cardiac tissue. Stem cell research & therapy 4 Suppl 1: S15, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Moya ML, Hsu Y-H, Lee AP, Hughes CC, George SC. In vitro perfused human capillary networks. Tissue Engineering Part C: Methods 19: 730–737, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Nguyen D-HT, Stapleton SC, Yang MT, Cha SS, Choi CK, Galie PA, Chen CS. Biomimetic model to reconstitute angiogenic sprouting morphogenesis in vitro. Proceedings of the National Academy of Sciences 110: 6712–6717, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Nishiyama Y, Nakamura M, Henmi C, Yamaguchi K, Mochizuki S, Nakagawa H, Takiura K. Development of a three-dimensional bioprinter: construction of cell supporting structures using hydrogel and state-of-the-art inkjet technology. Journal of Biomechanical Engineering 131: 035001, 2009. [DOI] [PubMed] [Google Scholar]
  • 53.Norotte C, Marga FS, Niklason LE, Forgacs G. Scaffold-free vascular tissue engineering using bioprinting. Biomaterials 30: 5910–5917, 2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Pawlik G, Rackl A, Bing RJ. Quantitative capillary topography and blood flow in the cerebral cortex of cats: an in vivo microscopic study. Brain research 208: 35–58, 1981. [DOI] [PubMed] [Google Scholar]
  • 55.Price GM, Tien J. Methods for forming human microvascular tubes in vitro and measuring their macromolecular permeability. Methods in molecular biology 671: 281–293, 2011. [DOI] [PubMed] [Google Scholar]
  • 56.Price GM, Wong KH, Truslow JG, Leung AD, Acharya C, Tien J. Effect of mechanical factors on the function of engineered human blood microvessels in microfluidic collagen gels. Biomaterials 31: 6182–6189, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Qian BZ, Pollard JW. Macrophage diversity enhances tumor progression and metastasis. Cell 141: 39–51, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Rao RR, Peterson AW, Ceccarelli J, Putnam AJ, Stegemann JP. Matrix composition regulates three-dimensional network formation by endothelial cells and mesenchymal stem cells in collagen/fibrin materials. Angiogenesis 15: 253–264, 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Raub CB, Suresh V, Krasieva T, Lyubovitsky J, Mih JD, Putnam AJ, Tromberg BJ, George SC. Noninvasive assessment of collagen gel microstructure and mechanics using multiphoton microscopy. Biophysical journal 92: 2212–2222, 2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Reidy MA, Langille BL. The effect of local blood flow patterns on endothelial cell morphology. Experimental and molecular pathology 32: 276–289, 1980. [DOI] [PubMed] [Google Scholar]
  • 61.Reinitz A, DeStefano J, Ye M, Wong AD, Searson PC. Human brain microvascular endothelial cells resist elongation due to shear stress. Microvascular research 99: 8–18, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Ross R The pathogenesis of atherosclerosis—an update. New England Journal of Medicine 314: 488–500, 1986. [DOI] [PubMed] [Google Scholar]
  • 63.Sakakura K, Nakano M, Otsuka F, Ladich E, Kolodgie FD, Virmani R. Pathophysiology of atherosclerosis plaque progression. Heart, Lung and Circulation 22: 399–411, 2013. [DOI] [PubMed] [Google Scholar]
  • 64.Shao J, Wu L, Wu J, Zheng Y, Zhao H, Lou X, Jin Q, Zhao J. A microfluidic chip for permeability assays of endothelial monolayer. Biomedical microdevices 12: 81–88, 2010. [DOI] [PubMed] [Google Scholar]
  • 65.Shevkoplyas SS, Yoshida T, Gifford SC, Bitensky MW. Direct measurement of the impact of impaired erythrocyte deformability on microvascular network perfusion in a microfluidic device. Lab on a Chip 6: 914–920, 2006. [DOI] [PubMed] [Google Scholar]
  • 66.Simmers MB, Pryor AW, Blackman BR. Arterial shear stress regulates endothelial cell-directed migration, polarity, and morphology in confluent monolayers. Am J Physiol-Heart C 293: H1937–H1946, 2007. [DOI] [PubMed] [Google Scholar]
  • 67.Song JW, Cavnar SP, Walker AC, Luker KE, Gupta M, Tung Y-C, Luker GD, Takayama S. Microfluidic endothelium for studying the intravascular adhesion of metastatic breast cancer cells. PloS one 4: e5756, 2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.Song JW, Munn LL. Fluid forces control endothelial sprouting. Proceedings of the National Academy of Sciences 108: 15342–15347, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69.Spaet T, Lejnieks I. Mitotic activity of rabbit blood vessels. Experimental Biology and Medicine 125: 1197–1201, 1967. [DOI] [PubMed] [Google Scholar]
  • 70.Stary HC, Chandler AB, Glagov S, Guyton JR, Insull W, Rosenfeld ME, Schaffer SA, Schwartz CJ, Wagner WD, Wissler RW. A definition of initial, fatty streak, and intermediate lesions of atherosclerosis. A report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Arteriosclerosis, thrombosis, and vascular biology 14: 840–856, 1994. [DOI] [PubMed] [Google Scholar]
  • 71.Tannock IF, Hayashi S. The proliferation of capillary endothelial cells. Cancer research 32: 77–82, 1972. [PubMed] [Google Scholar]
  • 72.Taylor ZJ, Gurka R, Kopp GA, Liberzon A. Long-Duration Time-Resolved PIV to Study Unsteady Aerodynamics. IEEE Transactions on Insttrumentation and Measurement 59: 3262–3269, 2010. [Google Scholar]
  • 73.Wilkens H, Regelson W, Hoffmeister F. The physiologic importance of pulsatile blood flow. New England Journal of Medicine 267: 443–446, 1962. [DOI] [PubMed] [Google Scholar]
  • 74.Wissler RW, Vesselinovitch D. Atherosclerosis in nonhuman primates. Advances in veterinary science and comparative medicine 21: 351–420, 1976. [PubMed] [Google Scholar]
  • 75.Wong AD, Searson PC. Live-cell imaging of invasion and intravasation in an artificial microvessel platform. Cancer research 74: 4937–4945, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 76.Wong AD, Ye M, Levy AF, Rothstein JD, Bergles DE, Searson PC. The blood-brain barrier: an engineering perspective. Frontiers in neuroengineering 6: 7, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 77.Wong KH, Truslow JG, Tien J. The role of cyclic AMP in normalizing the function of engineered human blood microvessels in microfluidic collagen gels. Biomaterials 31: 4706–4714, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 78.Wyckoff JB, Wang Y, Lin EY, Li JF, Goswami S, Stanley ER, Segall JE, Pollard JW, Condeelis J. Direct visualization of macrophage-assisted tumor cell intravasation in mammary tumors. Cancer Res 67: 2649–2656, 2007. [DOI] [PubMed] [Google Scholar]
  • 79.Young EW, Watson MW, Srigunapalan S, Wheeler AR, Simmons CA. Technique for real-time measurements of endothelial permeability in a microfluidic membrane chip using laser-induced fluorescence detection. Analytical chemistry 82: 808–816, 2010. [DOI] [PubMed] [Google Scholar]
  • 80.Yuan F, Leunig M, Huang SK, Berk DA, Papahadjopoulos D, Jain RK. Microvascular permeability and interstitial penetration of sterically stabilized (stealth) liposomes in a human tumor xenograft. Cancer Res 54: 3352–3356, 1994. [PubMed] [Google Scholar]
  • 81.Zervantonakis IK, Hughes-Alford SK, Charest JL, Condeelis JS, Gertler FB, Kamm RD. Three-dimensional microfluidic model for tumor cell intravasation and endothelial barrier function. Proceedings of the National Academy of Sciences of the United States of America 109: 13515–13520, 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 82.Zheng Y, Chen J, Craven M, Choi NW, Totorica S, Diaz-Santana A, Kermani P, Hempstead B, Fischbach-Teschl C, López JA. In vitro microvessels for the study of angiogenesis and thrombosis. Proceedings of the National Academy of Sciences 109: 9342–9347, 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 83.Zhou ZN, Sharma VP, Beaty BT, Roh-Johnson M, Peterson EA, Van Rooijen N, Kenny PA, Wiley HS, Condeelis JS, Segall JE. Autocrine HBEGF expression promotes breast cancer intravasation, metastasis and macrophage-independent invasion in vivo. Oncogene 33: 3784–3793, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 84.Zilberman-Rudenko J, Itakura A, Wiesenekker CP, Vetter R, Maas C, Gailani D, Tucker EI, Gruber A, Gerdes C, McCarty OJ. Coagulation factor XI promotes distal platelet activation and single platelet consumption in the bloodstream under shear flow. Arteriosclerosis, thrombosis, and vascular biology 36: 510–517, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp VideoS1

Supp VideoS2

Supp VideoS3

Supp VideoS4

Supp info