Vascularization in Microfluidics - Daniel Bell and Anna Comperchio

Introduction

It is extremely important to create in vitro models of vasculature in order to study cancer (including metastasis and tumor cell circulation), drug delivery, diseases, and to supply oxygen and remove waste to systems too large to depend on diffusion. Vascular diseases are the leading cause of death worldwide causing around 17 million deaths per year. There are limited treatments, and therefore a dire need to better understand the vascular system . Blood vessels are responsible for delivering nutrients and oxygen and for disposing wastes to and from all parts of the body. Vessels therefore end up controlling the microbiome. . Currently, the best experiments are conducted on animal models and/or cell culture, but even these methods are poor indicators for what occurs in vivo .

Cell culture experiments fail to consider the complex, diverse, vascular system. Rather than being a single type of cell, capillaries and arteries are composed with different layers of cells. The inner layer is made up of endothelial cells (ECs) which act as a semipermeable membrane between the vessel wall and the fluid from the outside. The middle layer is comprised of smooth muscle cells (SMCs), which give the vessels their mechanical properties and provide structure by producing collagen. Fibroblasts and connective tissue line the outer layer . Depending on what researchers are investigating, all of the layers of the blood vessels may need to be included. If the goal is to supply oxygen or deliver waste or maybe includes topics not associated with the actual biology of the blood vessel, then simpler fabrication methods can be used.

Figure 1. Schematic depiction of the general idea of 3D printing with multiple extruders necessary to create many vascularized microfluidic devices (Figure Adapted from Kolesky et al.).

The vascularization of microfluidic devices is the creation of channels within a device capable of transporting blood in one end and completely out the other. As blood vessels range in size from the smallest capillaries of 5 micrometers to the aorta of 25 mm, microfluidic devices are excellent tools in order to provide the proper channels for blood flow . The first generation of these devices relied on 2D manufactured devices stacked on top of one another in order to create a 3D vascularized device . Now technology allows for the creation of detailed 3D vascularized microfluidic devices with the proper channel size for the type of blood vessel being recreated and type of tissue being replaced.

The vascularization of microfluidic devices has a vast number of applications as the ability to successfully move blood in an artificial tissue construct is important in the effectiveness of the device. Vascularized microfluidic devices can be used in both the medical field as a tissue implant in a number of areas of the body or for more research applications as a way of studying blood flow in vitro . A number of considerations must be taken into account when creating these devices because of their use within the body. Biocompatibility of the materials used in these devices is very important as cells must adhere to the surface of the device and each device should accurately model the body . Over the past few decades, several different techniques for the construction of vascularized microfluidic devices have been developed including traditional methods of construction using photolithography[1] and soft lithography[2] as well as emerging technologies of 3D printing through both dissolvable sacrificial moldings and hybrid method bioprinting. As these technologies continue to develop, vascularized microfluidic devices will continue to grow in importance for organ and tissue repair.

There are several challenges that must be considered when using microfluidics to create a vascular system. For example, many microfluidic devices are inexpensive and easy to make because of the heat curing polymer polydimethylsiloxane (PDMS). It is typically considered a great material for biological applications because it is gas permeable, bio-compatible, and clear. However, PDMS adsorbs hydrophobic molecules, making it difficult to study drugs. Some decide to work with thermoplastic materials in place of PDMS for drug studies, but these are expensive, auto-fluorescent (bad for imaging), and not gas permeable (so cells will die quicker than in PDMS) .

Additionally, the geometry of the channels may not always be perfect with lithographic techniques. Even though chips can function on the micro and nano scale, the vessel walls in vivo are extremely thin, and consequently difficult to study. Soft lithography also typically will create rectangular channels by adhering a design to a flat surface . This may cause discrepancies between fluid interaction with cells in vivo and in vitro. Molding the channels a new way or taking advantage of 3D printing technologies can help create cylindrical channels.

Mimicking Geometry of Blood Vessels

Soft lithography techniques typically involve binding one side of the channels to a flat surface, creating rectangular or straight-edged channels. To properly model vascularization, standard fabrication methods need to change to make cylindrical channels that more accurately represent blood vessels. Sharp corners at intersections and bifurcations as well at the corners of a rectangular cross-section make it extremely difficult to properly seed them with endothelial cells. Non-cylindrical channels will have various shear stress depending on location, while a circular cross-sectional area will promote even shearing on all areas along the circumference. Correctly mimicking shear stress is critical when creating these microfluidic systems because shear stress determines endothelial cells functional phenotype and gene expression. Therefore, a microfluidic chip with a rectangular cross-section and uneven shearing will create a very different environment than what occurs in vivo .

Many research groups have developed interesting approaches to vascularize microfluidic systems. One lab developed microfluidic channels able to mimic blood vessels embedded in an extracellular matrix (ECM). They drew in sacrificial gelatin into a PDMS microfluidic channel, gelled it, and then removed the PDMS design. They then poured type I collagen (acting as the ECM) over the gelatin and allowed that to gel. The gelatin was then melted, leaving behind channels that can then be coated with endothelial cells to mimic blood vessels. With these types of channels, controllable spatial gradients of soluble molecules, such as growth factors, can be created . This device is easy to make, and offers a platform for studying morphogenesis. The same research group later implemented another device that also features channels completely immersed within an ECM. They polymerized rat tail collagen type I around two 400-μm-diameter needles. Once the needles were removed, they coated the channels with endothelial cells leaving cylindrical channels completely surrounded by ECM . This method solves the rectangular channels limitation that many microfluidic devices have. This also eliminates the issue of having part of the channel be silicon or glass, which cause devices to deviate from in vivo results further. This system was used to study gradient-driven angiogenic sprouting while allowing sprouting to occur in all directions from the preliminary channel .

Another interesting approach to achieving the correct geometry included electroplating wafers with copper and creating semi-circular ridges on silicon. The masters were embossed with polystyrene sheets with semi-circular channels and then aligned together to create circular channels.

One lab argued that microfluidic channels over simplified vascular systems and decided to use a template from nature as their template-a leaf. They believed that the complexity of the venation of the leaf was comparable to that of vascular systems and also followed the same Murray's law as blood flow . They first digested the soft tissues of the leaf, and then coated the remaining venation network with chrome. The leaf was used as a photomask over a UV-curing polymer that was spin-coated on a silicon wafer. The negative pattern remained on the silicon while the UV-irradiated area was dissolved with 0.5% wt NaOH. Chrome was coated over this silicon wafer with the design, creating a master. PDMS could then be layered over the wafer, producing a negative mold of the master, and a very complicated microfluidic design comparable to blood vessels . This technique exemplifies how creative researchers can be when developing methods on how to make something out of plastic mimic something that is alive.

Figure 2. Method for how to turn a leaf into a microfluidic device that can mimic blood vessels (Adapted from He et al.) .

In Vitro Methods for Creating Blood Vessels with Layers of Cells

Creating the complex tri-layer blood vessel wall mentioned previously is no easy task, but there is a strong need for an in vitro system to incorporate all layers. On one hand, the three layers need to be incorporated to properly study any diseases afflicting blood vessels. And on the other hand, microfluidic devices that include all layers could be used as a replacement for damaged arteries.

There are two main ways to create blood vessels in vivo: angiogenesis and prevascularization. Angiogenesis uses directional migration and self-organization of specific cell types to help create vascular structures. One way to do this is by sprouting endothelial cells on microbeads. Another possible method involves using angiogenic growth factors to promote vascularization in a matrix of materials that assist blood vein formation. A major drawback of using angiogenesis as the primary method for creating vascularization in vitro is its inability to create perfusable blood vessels . Angiogenesis can provide researchers with information on how veins develop and spread, but its limitations forced researchers to develop a new method to better create vascularized systems.

Prevascularization consists of encapsulating or seeding cells to channels’ surfaces, allowing perfusion and cell proliferation. This method promotes oxygen delivery and waste removal. Groups have successfully created a monolayer of ECs via prevascularization and others have also incorporated SMCs . These methods are steps up from angiogenesis procedures but still fail to incorporate the third layer. More recently, another group was able to add this critical third layer. They added fibroblasts, SMCs, and endothelial cells layer by layer concentrically to create a tri-layer microfluidic device in a photocrosslinkable 3D hydrogel . The three layers more closely resembles blood vessels in vivo, providing a platform to create advanced vascularized systems in vitro.

Figure 3. Depiction of healthy and unhealthy arteries. The image labels and highlights the different layers of an artery in vivo that in vitro models should strive to mimic. (Figure adapted from Gold et al.) .

3D Printing Technologies

Current 3D printing for microfluidic devices [3] has significant research investment and specifically for vascularization, the typical technologies used can be split into 2 main categories: sacrificial molding where a dissolvable material is 3D printed into the desired shape of the channel and later removed leaving the channels (Figure 4), and bioprinting which takes advantage of printing with many materials at once including inks containing cells .

Sacrificial Molding

Figure 4. 3D printed dissolvable channels that are then removed leaving a vascularized hydrogel (Figure adapted from Hasan et al.).

In order to create complex networks of channels in 3D devices, 3D printing using a dissolvable substance such as collagen can be used . First, the desired channel is created using traditional 3D printing techniques, followed by encapsulation in the desired 3D hydrogel, and finally the 3D printed dissolvable portion can be dissolved leaving a network of channels of any desired size. A typical example of the creation of a 3D device using a dissolvable substance is the creation of gelatin channels and printing the non-dissolvable scaffold in collagen or fibrin . Another example is the use of a fugitive ink composed of an coblock organic polymer that can be melted and removed under a vacuum (Figure 6). . This method provides additional structure when initially printing as the fugitive ink has more stability than gelatin, but it is more difficult to remove and complete the sacrificial molding process. Materials and processes for sacrificial molding must be kept under control as cells living within the additional portions of the 3D device must remain healthy . While the hydrogels required for the necessary mechanical properties in tissue and organs in the body are not very easy to 3D print in traditional ways, creating the channels as dissolvable structures offers a method to take advantage of affordable 3D printing for creation of vascularized devices . A new technology called omnidirectional printing offers additional improvements to the complexity of 3D printed devices by removing the need to create one layer of the material at a time . Dissolvable sacrificial technology has the potential to be very useful in creating vascularized microfluidic scale products through the use of 3D printing technology.

Hybrid Method Bioprinting

Figure 5. Depiction of the general idea of hybrid method bioprinting showing multiple extrudates with cells (Figure adapted from Kolesky et al.).

The use of bioprinting in order to create vascularized devices is a fairly new technology that uses the ability for 3D printers to extrude a number of components at once and extrudes an ink containing cells (Figure 5). Multiple extrudates containing materials such as poly(dimethyl siloxane) (PDMS) and an aqueous mixture of various polymers that allow the cells to survive for the short time necessary for injection. One commonly used polymer is a triblock polymer composed of one part poly(propylene oxide) and two parts poly(ethylene oxide) called Pluronic F127 . This material is inert to cells and can be washed away, leaving the healthy cells within the channels now created in the microfluidic device. One of the difficulties in 3D printed vascularized systems is the integration of cells after the printing, but the integration with the printing itself provides an exciting new method for successful vascularization of microfluidic systems . The potential limitation with this technology is the size of the channels, since expensive printers with advanced crosslinking capabilities can produce channels close to the smallest blood vessels, but slightly less expensive printers with less options for materials or crosslinking will only be able to study a smaller range of blood vessel types.

Figure 6. Demonstration of the use of fugitive ink as a sacrificial mold to create a vascularized 3D microfluidic device. . (Figure adapted from Wu et al.).

Studying Blood Flow Phenomena

Blood flow properties and flow phenomenon in blood vessels is poorly understood, and microfluidics could help researchers answer critical questions pertaining to blood flow. Researchers from Chosan University and Pusan National University both studied how platelet aggregation affected blood flow. In vivo, atherosclerosis is responsible for platelet accumulation along blood vessels’ walls. The groups investigated platelet aggregation and blood flow by measuring velocity and aggregation simultaneously with a velocimetry and a mapping technique. They analyzed how the flow changes from no platelet aggregation to once platelets began to cluster along the sites. Additionally, they could see how platelet aggregation rates change depending on the flow (Figure 7). The four-minute experiment ended when the channels were nearly blocked by platelets and only minimal amounts of blood could continue to flow through. This paper showcases an extremely useful tool to answer questions surrounding angiogenesis. It also provides a way to make in vitro systems more similar to in vivo models by creating platelet aggregation.

Figure 7. Optical micrographs demonstrating how the velocity of the blood flow changes over time as platelets aggregate (Figure adapted from Jung et al.) .

Applications

A huge number of applications are available for 3D printing vascularized microfluidic devices in both in vivo and in vitro. As a tool in the laboratory, accurately recreating blood vessels on 3D printed devices would allow for important research in vitro before transitions are made to in vivo experiments . Current technologies are fairly expensive when attempting to simulate systems like blood flow in small categories, but the low cost of 3D printing could help make more in vitro studies possible.

Microfluidic vascularized medical devices have major importance moving forward in organ and tissue engineering because of the need for blood to flow through implanted systems. These 3D printed devices need to have oxygen diffuse through them and creating blood vessels from the beginning throughout the chip can help with the integration of artificial tissue . PDMS has high gas diffusion and vascularized microfluidic devices made of PDMS have enough oxygen diffusion to support cell growth. Studies looking to research the effects of hypoxic environments can use materials such as poly(methacrylic acid) which do not allow oxygen diffusion . As this technology progresses and becomes cheaper and more accessible, 3D printed vascularized microfluidic devices will continue to grow in popularity and importance for these applications and more.

Figure 8 shows some of the potential results of 3D printing technology in order to create vascularized microfluidic devices . (Figure adapted from Hasan et al.).