Education for Careers in Tissue Engineering and Regenerative Medicine White paper

Whitaker Education Summit White Paper Education for Careers in Tissue Engineering and Regenerative Medicine Linda Griffith, MIT, Biological Engineering and Mechanical Engineering Melody Swartz, Swiss Federal Institute of Technology Lausanne (EPFL), Integr ative Biosciences Institute Robert Tranquillo, University of Minnesota, Biomedical Engineering Overview of Tissue Engineering Broadly defined, tissue engineering is the process of creating living, functional, 3D tissues and organs starting with populations of individual cells (Griffith & Naughton 2002, Saltzman 2004). Coaxing cells to form tissue is inherently an engineering process, as they need physical support (typically in the form of some sort of 3D scaffold) as well as chemical and mechanical signals provided at appropriate times and places to form the structures and achieve the functions that characterize native tissue. Further, the process of forming tissues from cells is a highly orchestrated set of events that occur over time scales ranging fr om seconds to weeks and dimensions ranging from 0.0001 cm — 10 cm. The most well recognized applications of tissue engineering involve therapies for an array of clinical problems, from diabetes and heart failure to bone and cartilage repair. The goal is to create new tissue either directly in the patient via cell transplantation or directed growth from endogenous healthy tissue, or by growing tissue outside the body for transplantation into the patient. Cells may be derived from the patient or from a donor, and may be pure or mixed populations of differentiated cell types; stem cells derived from the adult tissue; or embryonic stem cells (capable of both self renewal and differentiation into a variety of cell lineages); or a mixture of cells at different stages of maturation (that would also include rare stem and progenitor cells). Although less heralded than the direct clinical applications, tissue engineering is emerging as a potentially powerful tool in another arena --- in vitro physiological models of human tissue that can be used to study disease pathogenesis, to develop molecular therapeutics, and to screen for toxic effects of drugs on human tissues. Animal models are adequate for some purposes, but good models do not exist for many diseases, and the toxicological response of animals to drugs is often not predictive of the human response. Better models of human diseases and human responses to interventions may ultimately eliminate the need for many therapeutic “engineered tissues.” For example, the leading single cause for liver transplants in the western world is hepatitis C, a disease for which no complete in vitro model exists. Creating better models for such diseases will enable rapid drug development to treat them – thus reducing (and perhaps one day eliminating) the need for organ transplants. One might draw an analogy with development of therapies for polio – an “engineered tissue” (the iron lung) was developed to treat the consequences of the disease in 1928, shortly after the disease outbreak. In the mid-50’s the iron lung was replaced with a vaccine to prevent the disease -- a molecular medicine. Finally, beyond creation of potential therapies and tools, many basic biologists and mechanobiologists are turning to tissue engineering as the y strive to understand more complex behaviors of cell- matrix and cell-cell interactions. This movement of biology towards more systems-oriented approaches parallels a growing appreciation for the importance of the coupled biophysical and biochemical environments on cell behavior. Thus, while traditionally applying known biology and biochemistry to create therapies and models, tissue engineers are coming full circle to help expand and refine that basic biology knowledge necessary to recapitulate tissue function. For example, cells can sense stresses through highly specific cell- matrix binding proteins, so it is often critical to the question being asked that the matrix be as physiological as possible. Often, the best way to do this is to let cells grow in 3-D matrices and remodel their own local environment to mimic physiological conditions, and to apply stresses to complex systems of cells. Therefore, sophisticated 3-D tissue models are seful not only as disease models for toxicology and response testing, but also for studying basic aspects of cell biology and mechanobiology, as they contain some of the complexity of real tissue and thus the cell-cell and cell- matrix interactions that dictate cell response. Tissue engineering holds strong appeal for students at the undergraduate and graduate level; the Whitaker Foundation database lists over 300 subjects in the category of tissue engineering. Student enthusiasm is not matched, however, by viable industry employment opportunities – a situation that is not likely to improve in the near future. Lysaght and coworkers have charted the evolution of the industry in a series of detailed analyses published in the journal Tissue Engineering (Lysaght & Hazelhurst, 2004 and references therein). In the most recent analysis, the authors noted that despite an aggregate of public and private investment of over $4.5 billion, the field has yet to produce a profitable product. Further, of the twenty tissueengineered products that had entered FDA trials by the end of 2002, only four were approved (none yet commercially successful), six were abandoned or not approved, and ten were still in trial. An interpretation offered by the authors for the slow success of the field is that most spending in the US has come from commercial entities, which focus on technology development, compared to only 5-10% of total funding from government, which goes more substantially to basic science that can push the field forward. Another possible interpretation is that too few scientists and engineers have yet been trained with the appropriate depth in both biology and engineering to tackle the most important problems in the field, or provide perspective on the most productive directions. We present here an analysis of the relevant basic sciences and engineering training needed to meet challenges in the field, and describe a draft syllabus of a specialized subject in tissue engineering. Foundational Science and Engineering Prerequisites for Tissue Engineering Students attracted to tissue engineering come from a range of disciplinary and interdisciplinary backgrounds, including materials science, chemistry, chemical engineering, biology, mechanical engineering, and biomedical engineering. Each of these areas individually contributes significantly to the tools and approaches used, but tissue engineering requires a synthesis of these within a curriculum. The fundamental process common to all tissue engineering approaches is manipulation of cells to create new functional tissue. Thus, students need a strong preparation in modern cell biology, which in turn requires several introductory biology subjects to set the stage for serious study of cell biology. Typically, an introductory biology course will cover the basic fundamental principles of biochemistry, genetics, molecular biology, and cell biology, and introduce students to molecular and cell function at various levels: the structure and regulation of genes, and the structure and synthesis of proteins; how these molecules are integrated into cells; and how cells are integrated into multicellular systems and organisms. Because biology remains a strongly experimental science, students who pursue tissue engineering at the graduate level should have a reasonably strong preparation in basic experimental methods used in modern molecular life sciences to evaluate protein chemistry and gene expression in eucaryotes. These methods include chromatography, spectrofluorimetry, gel electrophoresis and western blotting, FACS analysis, optical microscopy, microarray analysis, etc. that provide students with a basic understanding of how to measure molecular and cellular events at the level of DNA, RNA, and protein. Next, a course in Genetics is necessary to understand many of the basic tools used in tissue engineering, from the use of knockout mice to study pathogenesis to the use of microarrays to deduce changes in cell and system behavior, as well as to familiarize students with new frontiers such as cell fusion approaches to correcting many genetic diseases that are now the target of tissue engineering. This course should cover structure and function of genes, chromosomes and genomes; biological variation resulting from recombination, mutation, and selection; population genetics; and use of genetic methods to analyze protein function, gene regulation and inherited disease. Finally, an upper level subject in Cell Biology should be required which should, at a minimum, include structure, function, and biosynthesis of cellular membranes and organelles; cell growth and oncogenic transformation; transport, receptors and cell signaling; the cytoskeleton, the extracellular matrix, and cell movements; chromatin structure and RNA synthesis. Chemistry is the other key foundational science for tissue engineering and provides a basis for designing and synthesizing tools to modify cell behavior, where tools include natural biomolecules and their modifications as well as a range of synthetic biomaterials that control cell behavior. Most cell biology subjects require a strong foundation in chemistry, with an emphasis on organic chemistry and biochemistry, and thus students who take the appropriate foundational biology subjects may have adequate chemistry preparation as well, with the possible exception of Physical Chemistry. Thermodynamics and kinetics, including at least an introduction to statistical thermodynamics and behavior of macromolecules, is an essential foundation for understanding and controlling the quantitative aspects of molecular interactions in tissue engineered systems. The engineering prerequisites are probably not captured in one single discipline, but include, in addition to physics and math through differential equations and linear algebra, courses in mechanics (including molecular and continuum models, and some exposure to molecular, cell and tissue biomechanics), transport phenomena (laminar flow, diffusion, electrokinetic driving forces) at the level of solving multi-dimensional and transient problems, kinetics (including the basics of diffusion and reaction), basic tools of numerical analysis and statistical programming. Graduate Education in Tissue Engineering – Merging the Science and Engineering A typical course, or series of courses, can be broken down into three general areas; Biological Frontiers and Challenges; Cell Engineering; Integrative Tissue Engineering & Technologies. The more advanced topics and application areas (e.g. the mechanics of tissue growth) might be covered in a graduate seminar with a focus on current literature. An overview of 1-2 applications at the beginning of the course provides a context for the fundamental material and allows a framework for discussing how to identify feasible approaches in tissue engineering problems in general. 1. Biological Frontiers and Challenges The zeroth order starting point for tissue engineering is a source of cells – can the patient’s endogenous cells be used? Stem cells or mature cells? Embryonic stem cells or adult stem cells? Can the cells be stimulated in the body or must they be manipulated outside the body? What are the practical and ethical debates surrounding these sources? A baseline starting point is an overview of development and homeostasis normal tissues, including an overview of the cellular composition and structure of representative connective tissues and epithelial tissues, and the role of matrix (e.g. elastin, collagen, proteoglycans) and cells in the composite mechanical/transport properties of tissues. An overview of the normal wound healing process in skin and bone, for example, can be used to set the stage for controlling regeneration. A somewhat detailed description of hematopoiesis is useful at this point for several reasons. Hematopoietic stem cells are the first system characterized and are still the bestcharacterized stem cell system due to the robustness of animal models. Yet one cannot still with certainty identify a single stem cell from the population in marrow, and thus the “gold standard” methods used to characterize stem cells – repopulation of the tissue in a survival model – is wellrepresented by the hematopoietic stem cell system. Hematopoietic stem cell transplants are used clinically and thus there is somewhat extensive data on the translation of animal models to the human system, allowing important points to be made about comparative cell and tissue behaviors across species. Finally, reports that hematopoetic stem cells have the capacity to differentiate into, and thus potentially repair, tissues and organs outside of the blood system first emerged in the late 1990’s based on experiments conducted in mice with liver injuries. The apparent plasticity of stem cells is now hotly debated as data has emerged supporting the idea that these phenomena happen in humans in a wide variety of tissues, and as artifacts are identified in certain experimental models. At the least, it appears the endogenous stem cells in the adult bone marrow have some capacity for tissue repair regardless of whether repair occurs through cell fusion or through transdifferentiation. Thus serious students in tissue engineering should be able to read and interpret the current literature in stem cell biology (extending beyond the hematopoietic system), and understand the methods used to characterize these cells and the standards in the field for defining a “stem cell”. The details of embryonic stem cell derivation and culture should also be described, along with a review of current literature as a means to emphasize the barriers to controlling differentiation and how differentiation is assessed. Inherent in a discussion of wound healing and stem cells is discussion of the molecular regulators of these behaviors, cytokines, growth factors and matrix. Although these topics are covered in a standard cell biology course, a review of the classes of molecules that might reasonably be used to manipulate cell function, along with the properties of specific molecules such as VEGF and EGF that are potentially important in tissue engineering, is an essential component of setting the stage for cell engineering. Finally, the student needs to be cognizant of the ethical issues surrounding the use of stem cells and cell sources. This can either be covered in a separate bioengineering ethics course if available, or can be integrated throughout this and other graduate tissue engineering courses. Many excellent texts are available to guide students through these difficult questions and a debate format has been particularly successful in building students’ awareness and sensitivity to ethical issues in tissue engineering and stem cells. 2. Cell Engineering The design principles for tissue engineering lie in models for quantitative manipulation of cell function, primarily through receptor- mediated phenomena, and thus cell engineering principles are an important component for any tissue engineering subject or subject series. Many approaches in tissue engineering involve delivery of one or more growth factors to influence tissue function. Yet, apart from hematopoietic cytokines, successful use of growth factors for human tissue regeneration in vivo has been notoriously difficult. Thus a foundation is needed for the quantitative analysis of binding, internalization, and trafficking of growth factor ligands and their receptors, along with at least a simple analysis of signaling responses. Factors such as receptor down-regulation and desensitization may limit the effectiveness of growth factors delivered in a continuous mode. Further, the biophysical context of growth factor stimulation is important; many growth factors, including the angiogenic factors VEGF and FGF, are bound tightly to the extracellular matrix of normal tissues. A ligand for the epidermal growth factor receptor (EGFR) is immobilized within tenascin, a large extracellular matrix molecule. Presenting growth factors as part of an extracellular matrix rather than just releasing them in soluble form is a physiologically realistic mode for some growth factors, and may have strong consequences for signaling and downstream responses. The analysis of growth factor receptor dynamics also provides a foundation for analysis of how gene therapy vectors interact with cells, including emerging evidence for signaling pathways activated by viruses and other vectors. Quantitative analysis of adhesion and motility phenomena provides an essential foundation for two key areas in tissue engineering -- interpreting in vivo cell trafficking behavior of cells (e.g., homing of stem cells, seeding of tissue beds such as liver with cells injected for therapeutic purposes) and design principles for tissue engineering scaffolds and for strategies that may involve cellular self-assembly. A starting point for analysis of adhesion is discussion of the various ways of quantifying adhesion, including attachment and distraction assays, and what each can reveal about different modes of adhesion (e.g., cell rolling in blood vessels adhering by selectins; integrin- mediated adhesion to matrix; cadherin- mediated cell- cell adhesion). The concept that mechanical stress can lead to signaling, which can in turn lead to strengthening of cell adhesion, is the sort of phenomenon that can be discussed in terms of quantitative measurements of the magnitude of forces involved on a per-receptor basis. As with growth factors, the physical means of ligand presentation dictate cell responses to adhesive stimuli. For example, clustering of adhesion ligands can facilitate clustering of integrins. The mechanical properties of the matrix – compliant or stiff – also dictate response in ways that are critical to take into account when designing scaffolds. The area of mechanobiology is growing rapidly as new tools are developed to control both local mechanical properties as well as the overall mechanical stress/strain environment of cells. The now-classic studies of the effects of shear stress on gene expression in endothelia remain relevant but are just an example in the evergrowing body of data that the mechanical signals cells receive from the environment are key regulators of vast programs of gene expression. Several adhesion- mediated responses play a strong role in tissue engineering, including cell migration, contraction and remodeling of matrix, and cell-cell adhesion that can result in self-assembly of multi- cellular processes. Non- intuitive system responses are revealed by quantitative engineering models that incorporate biophysical representations of the individual components of the process; arguably the most well- recognized of these is the biphasic response of cell migration speed on cell-substrate adhesion strength, which has been borne out in many different 2-D and 3D in vitro systems. Although it is likely beyond the scope of most tissue engineering subjects to treat signaling networks in any depth, it may be appropriate to discuss factorial design of experiments to assess interactions between multiple components in the response of the system. 3. Integrative Tissue Engineering and Technologies Many of the difficult challenges in tissue engineering arise from scaling up responses at the cell level to the tissue level, and building models that explain (and predict) phenomena over multiple length scales and time scales. A key problem encountered in many if not most approaches in tissue engineering is diffusion and/or convection and reaction of nutrients and growth factors in the tissue environment. Many examples exist in the tissue engineering literature of approaches that worked well in a tiny mouse tissue but that failed on going to a larger scale. Although tissues vary widely in the degree of vascularity, from avascular cartilage to the highly vascularized structure of tissues such as endocrine glands and liver, where most cells are no more than one cell away from the blood flow, a typical transport distance is about 20-100 microns. Providing an adequate supply of nutrients on that length scale is incredibly challenging. Mass transfer limitations provide a context for discussion of the nature and operation of bioreactors and their use in ex vivo tissue engineering. Any cell-seeded scaffold or membrane-enclosed cell aggregate has possible mass (nutrient) transfer limitations. Thus one starting point for the integrative topics is analysis of transport limitations of nutrients (especially oxygen) and molecular regulators, including the potential role of convection and other driving forces for transport (and including the possible role of mechanical stress on tissues facilitating transport, as in cartilage). New experimental methods are being deployed to assess gradients, and although these might be beyond the scope of a single course, the data emerging from measurements obtained by methods such as two-photon microscopy are an important area to cover. While it is often tempting to focus on the role of molecules that are normally delivered from the blood, inclusion of at least some discussion on the role of local transport on the level of autocrine factors is important. Local accumulation of autocrine factors can have profound influences on cell motility, growth, and differentiation, and is emerging as a critical area of study in understanding cell populations. Likewise, the role of growth factor binding to matrix in development of gradients in tissues is also being addressed from quantitative experimental approaches. As the process of tissue growth proceeds, fundamental problems arise in linking local conditions to larger scale tissue processes. Dynamic changes in tissue mechanical properties that result from the net local production, degradation, and remodeling of matrix by cells, and the evolution of tissue structure due to migration and growth of cells, can be described by multiscale modeling methods and approaches to framing and modeling such processes are an essential foundation for a subject or series of subjects. Up to this point, the course has addressed fundamental issues that are common to almost any tissue engineering problem. In the last part of the course, the topics can diverge into a number of directions that focus on the role of various tools for accomplishing the requisite manipulations of cell systems, and into specific application areas that illustrate key points with respect to creating a product that works either in the clinic or as an in vitro screening model. Perhaps the most important single topic in this arena is a discussion of biomaterials with regard to general design and fabrication principles. Many biomaterials in common use in tissue engineering were not originally designed for biomedical purposes, and it is instructive to review successes and shortcomings of both biological molecules and of synthetic molecules, particularly those of biodegradable polyesters, the workhorses of tissue engineering. Uncontrolled adsorption of serum proteins that dictate cell adhesion to most biomaterials is an ext remely important concept, and one that draws heavily on a strong foundation in statistical thermodynamics. A common design paradigm has emerged of damping “non-specific” adhesion and building back in very controlled interactions with matrix. A few excellent examples exist where quantitative manipulation of receptor- mediated phenomena are integrated into biomaterial design to present defined ligands for cell adhesion, for stimulation of growth factors, and for degradation/remodeling due to proteolysis, where the system parameters are quantitatively varied. However, phenomenological approaches that are not yet understood at the mechanistic level can also yield remarkable phenomena and may still yield viable clinical products. Any number of specific applications in tissue engineering could be used in the last part of the course to illustrate the integration of the concepts discussed throughout the course, and to bring in the issues faced in the clinic in a more focused way. One approach is to compare the progress of a particular connective tissue (e.g. cartilage, bone, blood vessel) with that of a representative metabolic tissue (e.g., islets for diabetes, liver). References: Griffith, L. G. and Naughton, G., “Tissue Engineering: Current Challenges and Expanding Opportunities,” Science, 295, 1009 (2002). Lysaght, M. J., Hazlehurst, A. L., “Tissue engineering: the end of the beginning,” Tissue Eng., 10, 309-320 (2004) and references therein. Saltzman, W. Mark, “Tissue Engineering,” Oxford University Press, 2004.