|Home | About | Journals | Submit | Contact Us | Français|
Soft tissue engineering strategies targeting restoration of volume loss have inherent critical challenges as they relate to the problem of restoration of defects with a high volume to surface ratio. We outline the problems associated with the limitations of translational applications regarding soft tissue engineering strategies as follows: cell survival, mechanical challenges: macroenvironment (scaffold collapse and on-the-shelf availability), compositional considerations: microenvironment, inducing malignant behavior, cell migration, and cell exhaustion. These are discussed with our alternative suggestions for solutions.
The field of tissue engineering emerged as a distinct entity in the late 1980s. Prior to this time, the term “tissue engineering” was being mentioned in a few papers and discussions, but its precise definitions remained nebulous. The main impetus for tissue engineering arose from the challenges faced in the medical and surgical fields.
Before the concept of tissue engineering emerged, strategies at addressing tissue losses remained the domain of the surgeon. The removal or loss of tissue related to disease or trauma often resulted in functional losses, disfigurement, and, in cases in which the loss of tissue was incompatible with life, death. Options were limited to autologous transfer of the patient's own tissue, organ transplantation, or the use of artificial prostheses and implants. All these options continue to have critical intrinsic limitations. Autologous transfer of similar tissue is limited or impossible if the tissue loss too great, as in the case of severe burns, or if the tissue has specialized metabolic functions. As a result, the utility of surgical reconstruction using autologous tissue is usually greatest when the aim is to replace volume or structural deficits, not metabolic deficits. Autologous tissue transfers, even when indicated, introduce the additional problem of donor site morbidity. Organ tissue transplantation overcomes some of these limitations; it has been successful in replacing tissue with vital metabolic functions, such as liver and kidneys. However, it, too, has serious intrinsic limitations, including the limited availability of organs for transplantation and lifelong risk of immunologic complications. Finally, prosthetic devices and artificial implants, although becoming increasingly sophisticated, are still limited in their use to addressing volume or structural deficits accompanying tissue loss. Furthermore, implant distortion and rejection secondary to host immune reactions remain a major problem.
Tissue engineering emerged from the background of these surgical problems. The concept represented a major paradigm shift: manipulation of living cells and their extracellular products in the development of biological substitutes for replacements as opposed to the use of inert implants or tissue or organ transfer. Several meetings and symposia, most notably Granlibakken in 1988 (National Science Foundation), clarified the goals and definitions of this nascent field. Tissue engineering was defined as “the application of principles and methods of engineering and life sciences toward fundamental understanding of structure-function relationships in normal and pathological mammalian tissues and the development of biological substitutes to restore, maintain, or improve tissue function.”1
The landmark article published in Science in 1993 by Langer and Vacanti has since become the seminal work in the field of tissue engineering. This paper identified three main strategies for the development of biological substitutes: the use of (1) isolated cells or cell substitutes, (2) tissue-inducing substances, or (3) cells placed on or within matrices.2
All three strategies mentioned by Langer and Vacanti remain the primary methods employed in tissue engineering for the purposes of volume restoration. The challenges of tissue engineering as they relate to volume restoration are within the scope of the current discussion. Currently, the critical challenges in tissue engineering as they relate to the problem of volume restoration stem from the high volume to surface ratio of most volume defects.
The high volume to surface ratio severely limits long-term cell survival because of problem of adequate oxygenation and nutrition (i.e., vascular access). The goal of increasing the in vivo survival of engineered cells and allowing organ patterning (i.e., matching the three-dimensional structural topography of the lost tissue) has therefore been one of the primary challenges in tissue engineering thus far.
One approach to this problem of long-term survival of engineered tissue implants has been what is referred to as vascular network engineering: the use of computer and computed tomography–aided tissue engineering in constructing a three-dimensional scaffold to direct the growth of the vascular tree that will support the metabolic demands of engineered tissue. This has been attempted both ex vivo and in vivo. Tissue culture systems seeded with endothelial cells have been used, with limited success, in constructing vascular networks in vitro 3. In general, however, it appears to be more promising to introduce progenitor vascular cells in vivo or rely on modulating influences within the host organism to achieve neovascularization in vivo. In this regard, prevascular components such as stem cell–derived vascular cells or endothelial progenitor cells have been introduced in vivo to induce vasculogenesis in a manner resembling the embryological process in which “hemangioblasts” differentiate into blood cells.4 Most attempts have also involved the use of growth factors such as vascular endothelial growth factor, and such attempts have shown some success in creating new vasculature in animal models.5
Specific homing, vascular differentiation, and shape of the resulting vascular tree have yet to be controlled or studied. The effects of wound environment (i.e., inflammation, fibrosis, and extracellular matrix composition) on vasculogenesis need to be identified and optimized in a differential manner. In short, the concept of inducing a vascular support for engineered tissue has a long way to go before it can be used for clinical applications in tissue engineering.
One approach we propose to the problem of the volume/surface area restrictions in the implantation of engineered tissue is the concept of laminar construction of engineered tissue. Theoretically, this would require not the engineering of a complex vascular tree but the prefabrication of laminar tissue allowing shorter diffusion distances for oxygen and nutrients. Vascular supply could be established by various means, for example, through a comb-shaped vascular bed using silicon spacers that would support the laminar insertion of engineered tissue grafts or progenitor cells. We have used the concept of a negative template for tissue grafts for the prefabrication of the complex-shaped three-dimensional structures (Fig. 1).
A third way to address the problem of volume restoration is the in vivo insertion of prefabricated noncellular constructs. The main advantage of this method is that it activates progenitor cells from the recipient bed, thus inducing soft tissue regeneration in vivo. Although achieving vascular support is still a challenge with this method, the gradual induction of soft tissue replacement allows more time for the in vivo generation of an adequate vascular supply.
The concept of using noncellular constructs has its own challenges that must be addressed before it can be applied to the restoration of volume defects with soft tissue. The first challenge is the mechanical design parameters that the construct must satisfy. The scaffold would have to match the contours of the removed tissue and provide initial mechanical support to prevent collapse of the volume defect while being able to degrade completely and be replaced with soft tissue. The degradation properties of the scaffold would have to be optimized such that the kinetics match the de novo generation of soft tissue and the by-products do not harm the surrounding tissue or hamper de novo tissue generation.
We have worked on developing such a scaffold according to these design parameters. With the first design parameter in mind, we have devised a complex scaffold with three components. These are a soft fibrillar exterior region that interfaces with the native tissue and conforms to topographical variations in the volume defect (facilitates on-the-shelf availability), a rigid slowly degrading cage portion that supports load and maintains the geometric volume, and a highly porous central region housed within the rigid cage that contains the drug delivery vehicle and allows tissue ingrowth (Fig. 2).
In our construct, we have used an outer component composed of a thin layer of fibrillar matrix that is designed to aid in clotting and reducing stress profiles secondary to material mismatch.6 The rigid shell serves multiple functions as a porous conduit supporting drug delivery and tissue migration into the architecture and as a mechanical scaffold preventing tissue collapse.7 This layer can be fabricated from several slowly degrading polymer mixes such as poly(lactic-co-glycolic acid) (PLGA). Additional support will be added to the 2- to 3-mm layer by patterning the layer with a honeycomb architecture taken from cellular solids, which are architectures that optimize strength based on a specific material volume (Fig. 2).8 The honeycomb is a mechanically optimized structure that provides excellent in-plane compressive strength.9
Much attention must also be given to the microenvironment, that is, the environment “seen” by the cells, which encompasses both structural elements (i.e., porosity) and chemical makeup (i.e., extracellular matrix composition and cell adhesiveness characteristics). These elements would facilitate the adherence and infiltration of the newly generated tissue.
The patterned microenvironment of the construct involves many considerations. It must provide a degradable porous conduit for tissue infiltration, provide a substrate for cell adherence and migration, mimic the optimal extracellular environment for the engineered tissue, and house the delivery vehicles for any desired bioactive compounds.
The materials that have been used for the purpose of engineering tissues in vitro are plentiful and full into two main categories, biological materials and synthetic materials. Biological materials include collagen and fibrin matrices and contain an abundance of biological cell signaling domains, some of which are well characterized but most of which are uncharacterized. However, it is clear that extracellular matrix elements are essential participants in cell signaling, cell proliferation, cell migration, and cell adhesion. The use of biological materials, although necessary, therefore introduces additional variables in the bioactivity of the scaffold. The use of biomaterials may thus result in poorly controlled communication between the scaffold and the cells that leads to an unregulated result. Furthermore, the inherent structural characteristics of such biological scaffolds are not easily modified, thus limiting control over which physical cues are received by the cell.
Synthetic scaffolds are typically made from biologically inert materials, which reduces the unpredictability of physical and biological interactions between the cells and biological materials. Furthermore, the use of synthetic polymers permits high-resolution control over structural properties such as density, porosity, and compliance. However, these materials do not mimic the natural extracellular milieu of the desired tissue as well as natural materials, thus reducing the stimulatory potential of the extracellular environment necessary for tissue engineering aims.
The ability to create synthetic materials that can be covalently modified with biological materials has introduced a third category of hybrid materials, containing both synthetic and natural compounds. For example, polyethylene glycol (PEG) acrylates, covalently modified with the cell adhesive peptide arginine-glycine-aspartic acid (RGD) or the extracellular matrix component heparan sulfate, have been successfully incorporated within the scaffolds to facilitate cell attachment and to allow spatial sequestration of heparan-binding growth factors.
Several successful attempts at such hybrid materials exist in the literature. Luo and Shoichet have created agarose gels with an adhesive fibronectin fragment to facilitate and direct cell migration during neurite outgrowth.10 Kapur and Shoichet have designed similar scaffolds that have incorporated onto a synthetic backbone an immobilized concentration gradient of nerve growth factor.11 Hubbell has used a PEG backbone that is cross-linked with proteolytically sensitive oligopeptides, allowing cells mobility in the PEG gel through their own naturally secreted proteases.12
These studies have demonstrated the design concept of using synthetic polymers, for which the mechanical and structural properties can be manipulated, and transforming them into biologically responsive scaffolds for cellular remodeling and tissue engineering. Hybrid scaffolds such as these therefore combine the benefits of both synthetic and biological materials. They allow control over the structural properties of the scaffold at the microarchitectural level and permit the addition of biofunctionality to the scaffold through the incorporation of biological materials, which can be several and include not only extracellular matrix components but also any other desired biologically active components, including protein fragments, growth factors, or even biologically active peptide sequences. Furthermore, a precise, predesigned spatiotemporal distribution of these factors is possible within a specifically predesigned microenvironment.
However, our previous methods have not been able to control the spatiotemporal distribution of such bioactive compounds. Furthermore, we have not investigated the necessity of vasculogenic compounds and proteases and extracellular adhesive components to facilitate the increased demands on cell proliferation, migration, and vasculogenesis that would be required in larger volume defects.
The applied strategies of tissue engineering mimic the pathways that cancer cells naturally utilize, such as cell proliferation, cell migration, angiogenesis, and matrix turnover. These components of cell induction should be meticulously identified in tissue regeneration strategies, especially in postcancer reconstruction. Two behavioral models that do not follow or parallel cancer initiation or progression are cell differentiation and preventing inflammation. Therefore, anti-inflammatory and cell differentiative effects of utilized bioactive compounds should balance or overweigh the aggressive tissue growth-invasion vector.
In our research, we have use synthetic drug delivery vehicles such as PGLA-based microspheres to achieve local long-term delivery of bioactive compounds such as insulin-like growth factor 1 (IGF-1). We have developed scaffolds with a porous inner region designed to function as the drug delivery component. A fast-degrading 50:50 mix of PLGA with 90% porosity was designed. To offset the reduction in apparent properties related to the material volume, a strength-optimized architecture taken from a previous study evaluating different architectures resulting in varied apparent properties was used.13 The pore size of polymer was 500 to 800 μm, a size that has been shown to support tissue invasion in PLGA porous scaffolds in a rat model.14
We have demonstrated total soft tissue replacement of our biodegradable scaffold systems in 2.5-cm defects in a Sprague-Dawley rat model when combined with adipogenic factors such as insulin and IGF-1.15 Furthermore, we have shown that augmentation of adipofascial flaps and free fat grafts with insulin and IGF-1 delivery vehicles resulted in statistically significant increases in soft tissue regeneration (Fig. 3), which in both cases was secondary to adipocyte proliferation and differentiation and not hypertrophy of existing adipocytes.13,15,16 Considering the possible risks with IGF-1 of inducing malignant behavior (particularly in the reconstruction of defects of cancer resection), we have ceased to utilize these agents for postcancer reconstruction modalities. We have also turned our attention to the use of biodegradable scaffolds containing peroxisome proliferator-activated receptor γ (PPARγ) ligands as our long-release components in a biodegradable microsphere system. PPARγ receptors play a pivotal role in the terminal adipocytic differentiation of precursor cells. Like insulin and IGF-1, PPARγ ligands have been shown to stimulate adipocyte proliferation and differentiation; unlike insulin and IGF-1, PPARγ ligands do not have the same tumorigenic effect and indeed have demonstrated an antineoplastic effect on breast cancer cell lines, as mentioned previously.17 These effects are possibly due to the differentiative and anti-inflammatory properties of PPARγ ligands.
This ability to control precisely the spatiotemporal distribution of the biological factors is of tremendous import in tissue engineering and will allow the creation of pull vectors and gradients guiding cell migration. Matrix modification is an important component of this process as chemotactic signals in the matrix are required to stimulate inward movement of cells. In vivo design of matrix composition should be aimed at targeting the specialized cells of the target tissue (e.g., preadipocytes for soft tissue regeneration, endothelial cells or hemangioblasts for vasculogenesis).
In addition, micropatterning of biological factors, including synthetic backbones modified with alternating adhesive molecules, may make it possible for cells to “walk” on a scaffold by dynamic adhesion/contra-adhesion, thus maximizing tissue infiltration of even large-volume scaffolds.
Micropatterning of scaffolds can also encompass the controlled modification of synthetic backbones to create a gradient of pull for the target cell. This can include bioactive proteins incorporated into the synthetic backbone, the incorporation of synthetic drug delivery systems suspended within the scaffold in a gradient-like fashion, or the creation of an electrochemical gradient (Fig. 4).
Regeneration succeeds only to a certain depth. Several factors contribute to this limitation, including increasing resistance to cell migration secondary to fibrosis, mechanical properties of the scaffold, and inadequate vascular supply. These limitations have been the main difficulty in translating the results of rat studies to higher volume defects. We can utilize the strategy of the cancer cell to overcome this problem, such as releasing matrix degradation enzymes to clear the matrix wall. One possible solution is the delivery of tissue-type plasminogen activator (tPA) or other matrix degradation enzymes to facilitate the degradation of the extracellular environment necessary for cell migration and vasculogenesis. Theoretically, one can control the spatiotemporal distribution of such enzymatic activity along an increasing gradient of proteolytic activity normal to the noncellular scaffold. This can be done through selective modification of the synthetic backbone in a stratified manner to create shells of increasing tPA activity. A supraphysiologic concentration of tPA may be necessary to maximize cell migration past a critical distance.
The vascularization of engineered tissues in many cases does not keep up with the ingrowth of cells. Nutrient and oxygen supply are not sufficient, which ultimately leads to the death of the invading cells. The enhancement of the angiogenic capabilities of engineered tissues therefore represents a major challenge in the field of tissue engineering. The immobilization of angiogenic growth factors in the same manner as tPA may be useful for enhancing angiogenesis.