Editor’s Note: Sharon Presnell discussed the use of bioprinting and 3D tissue to improve predictability in preclinical drug discovery this week at the BIO 2014 convention in San Diego. —BVB
The human race consists of a large population of genetically diverse organisms. This inherent diversity is compounded daily by the individualistic way we go about living – the air we breathe, the food we ingest, and the drugs, chemicals, micro-organisms, plants and animals that we contact. Studies that describe epigenetic mechanisms in the regulation of prevalent diseases such as diabetes, addiction, and cancer reveal the true complexity of homeostasis and pathogenesis.
Denis Noble, a renowned Oxford professor of cardiovascular physiology, elegantly described the genome as being “like an immense organ with 30,000 pipes,” pointing out that the music itself is not created by the organ alone, but is a product of the pipes that are available and the actions of an organist. The presence of a particular gene (pipe on the organ) establishes the potential for a particular event to occur (note being played), but it is the combination of genetic and epigenetic regulation (the organist) that determines the outcome (the music). Evidence continues to mount in support of these concepts, as genomic studies find that genes associated with a specific disease are absent in some people who have the disease, and present in some people who do not. So how do we accurately model physiology and disease in ways that emulate the inherent complexity and heterogeneity of the human being?
Beginning with the discovery that mammalian cells could be isolated and kept alive outside of the body (around the turn of the 20th century), scientists have endeavored to grow cells in a culture dish (in vitro) to model and predict outcomes within a living organism (in vivo). The reductionists in all of us have enthusiastically taken complex tissues, separated them into their cellular components, and undertaken experiments aimed at understanding how a particular cell type responds to a stimulus—painstakingly picking apart genomic, proteomic, and phenotypic responses so we can better understand the details. We use the data from these experiments to draw conclusions about the mechanisms of homeostasis and disease, to identify new drug targets, and to test the therapeutic or toxic effects of drugs. Results from experiments conducted with cell lines or purified primary cells are extrapolated to predict what would happen at the tissue level or in a living organism. When we have been informed by the in vitro experiments, we often leap to animal studies to validate hypotheses at the organism level, hoping the outcomes will predict human in vivo responses.
The inherent challenges with classical in vitro cell-based assays and in vivo animal models are twofold: In vitro cell cultures provide a great opportunity to study pathways and assess responses, but lack the broader context and structure of intact tissue as well as the systemic effects of a whole organism; In vivo animal studies enable the assessment of organism-level responses, but lack the context of a human system. In order to bridge the gap between current preclinical models and clinical trials, significant efforts are being expended to develop predictive models that can provide a greater degree of human tissue context.
Seminal work by Berkeley’s Mina Bissell and her collaborators provided clear evidence that three-dimensional cellular aggregates comprised of multiple breast tissue-relevant cell types yielded superior predictive results to traditional two-dimensional monocellular cultures. These results, and others, sparked a revolution aimed at developing next-generation tissue models that would allow in vitro experimentation within the context of tissue-level complexity.
Multi-cellular 3-D spheroids are used in a variety of applications beyond oncology and can be created by incubating cell suspensions on non-adherent surfaces or as hanging drops. Biomaterial scaffolds comprised of synthetic polymers, extracellular matrix proteins, or de-cellularized tissues can be combined with cells to create 3-D composite structures. With the advance of additive manufacturing strategies into the life sciences arena, it is now possible to use highly specialized 3-D printers (bioprinters) to reproducibly build compartmentalized 3-D tissues that mimic key aspects of native tissue architecture and function, with an extraordinary degree of user control in both tissue design and fabrication strategy. Although the field of 3-D bioprinting is young, the emerging capability to create new in vitro models that embrace the heterogeneity of native tissue is extremely compelling. We can expect a continuous flow of data over the next few years that reveals the benefits and predictive capabilities of these models.
Efforts have also been expended to build models that incorporate human context at the level of a whole organism. “Humanized” animals, predominantly rats and mice, are being generated that incorporate human cells or tissues to reconstitute some portion of the animal. Examples here include animals that bear human tumor xenografts, have been reconstituted with human bone marrow/immune systems, or contain “humanized” livers. These chimeras are providing new insights at the level of the whole organism, and it is exciting to think about the potential of combining complex tissue manufacturing strategies, like 3-D bioprinting, with this approach to enable these models to be more intricate and reproducible.
As with any new technology, success will create a host of additional challenges. Building systems that are reproducible, user-friendly, and provide true human correlation is key. To get the most out of these complex heterogeneous tissue models, we need to understand the tissue responses as well as the cellular, mechanistic responses. The development of analysis tools, in particular those that enable detailed temporal and spatial measurements at the cellular and molecular level without destructive testing, will be needed to maximize the value from experiments using 3-D tissue models.
The greatest challenge that lies ahead may be the drug discovery and development process, with a shift toward strategies that recognize and embrace the amazing complexity of human biology. If we are attempting to predict a human outcome, we should be asking the questions in models that approximate the complexity of a living human tissue, and deriving answers at the finest resolution achievable.
It is an exciting time to be working in predictive tissue modeling. Accelerating technologies, such as induced pluripotent stem cells and synthetic biology, are creating the opportunity to build even more sophisticated models that can be tailored to specific populations, disease states, or individuals. Advancements in engineering, cell biology, computing, and genomics will undoubtedly converge to create the kinds of integrated systems that our predecessors could only imagine.