Experimental Neurology 192 (2005) 1 – 6 www.elsevier.com/locate/yexnr
Commentary
A cultural renaissance: in vitro cell biology embraces three-dimensional context David B. Edelman, Edward W. Keefer* The Neurosciences Institute, 10640 John Jay Hopkins Drive, San Diego, CA 92121, USA Received 12 April 2004; accepted 13 October 2004 Available online 13 January 2005
Abstract Increasingly, researchers are recognizing the limitations of two-dimensional (2-D), monolayer cell culture and embracing more realistic three-dimensional (3-D) cell culture systems. Currently, 3-D culture techniques are being employed by neuroscientists to grow cells from the central nervous system. From this work, it has become clear that 3-D cell culture offers a more realistic milieu in which the functional properties of neurons can be observed and manipulated in a manner that is not possible in vivo. The implications of this technical renaissance in cell culture for both clinical and basic neuroscience are significant and far-reaching. D 2004 Elsevier Inc. All rights reserved.
Looking beyond the monolayer. A movement is afoot in the arcane world of tissue culture. Many of its practitioners are becoming ever more aware that cellular context is important. They are relinquishing their memberships in the bFlat Earth SocietyQ of two-dimensional (2-D) culture techniques to join the ranks of the bround-earthersQ by abandoning flat Petri dishes and flasks in favor of threedimensional matrices of natural and synthetic fiber scaffolding. This 3-D cell culture approach offers researchers a means to study cell growth, proliferation, and differentiation under conditions that emulate an in vivo environment and, to varying degrees, allow cell–cell and cell–extracellular matrix (ECM) interactions that might otherwise be severely constrained or precluded entirely in 2-D culturing of dissociated cells. The importance of spatiotemporal cellular context during development was appreciated early in the 20th century by such developmental biologists as Spemann and Mangold (1924), who showed, through a series of explant graft experiments, that interactions between grafted (socalled organizers) and host tissues determined tissue fate
* Corresponding author. Fax: +1 858 626 2199. E-mail addresses:
[email protected] (D.B. Edelman)8
[email protected] (E.W. Keefer). 0014-4886/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2004.10
and ultimately, the organization of animal form. Later, the work of Holtfreter (1943a,b, 1944), Holtfreter and Hamberger (1955), and Townes and Holtfreter (1947) demonstrated the importance of mechanical and chemical properties, such as surface tension, gravity, and cell adhesion and movement, in determining tissue and organ morphologies. From this early work came the observation that disaggregated cells from amphibian blastopores would form solid spheres in culture, exhibiting no particular organization until they were later grafted to a blastocoele, whereupon they would organize into recognizable tissues, such as brains with laterally disposed eyes and ear vesicles (Holtfreter, 1943b). Later experiments by Moscona (1952, 1963) demonstrated that mixed, dissociated cells cultured from other vertebrates, such as chickens, and marine invertebrates, such as sponges,1 also had the capacity for reaggregation and tissue differentiation. Furthermore, these studies were among the earliest to suggest that the properties of cell adhesion that are responsible for reaggregation and tissue differentiation might be determined by materials at the cell surface (Moscona, 1963). Since the early 1980s, an increasing number of cell adhesion molecules (CAMs) and 1
Reaggregation and differentiation was first observed in dissociated sponge tissue by Wilson (1907).
2
D.B. Edelman, E.W. Keefer / Experimental Neurology 192 (2005) 1–6
extracellular matrix proteins have been identified and characterized. Although the foregoing studies were clearly founded on a longstanding appreciation of the importance of cellular context, technical strides in three-dimensional cell culture were not manifest until relatively recently.2 The consequences of attempting to extrapolate 2-D findings into notions of cellular behavior in 3-D are becoming clearer. Insights into the differences in properties of cells in 2-D versus 3-D cell culture were first made by cancer researchers, who showed that when cancer cells grown in a 3-D environment were incubated with an antibody against h1-integrin, they lost their cancerous properties (Weaver et al., 1997). This effect was not reproducible in 2-D. Further work has shown that environmental influences have profound effects on the neoplastic properties of tumor cells, and such effects may determine the pharmacological responses of cancer cells to therapeutic drugs (Sahai and Marshall, 2003). The advantages conferred by 3-D cell culture are perhaps most significant for those interested in observing the behavior of cells from the central nervous system. In the living animal, cells such as neurons may project highly ramified processes over relatively great distances; conventional 2-D culture systems may not support such extensive growth. Moreover, 2-D cultures do not allow for the scaffolding that is required to both sustain cell densities that approximate those in the brain and promote heterogeneous cell–cell and cell–ECM interactions that yield functionally critical in situ features such as far-reaching and extensive myelinated projections. In contrast, 3-D cultures allow for the introduction of engineered matrices that offer adhesive and nutritive properties resembling in vivo conditions (Sasaki et al., 2002). Finally, many varieties of 3-D cell culture are amenable to repeated manipulation and sampling that would be far too disruptive to an ongoing monolayer culture. There have been five approaches most commonly employed to produce 3-D cultures: (1) organotypic explant cultures, in which whole organs or organ elements or slices are harvested and grown on a substrate in media; (2) stationary or rotating microcarrier cultures, in which dissociated cells aggregate around porous circular or cylindrical substrates with adhesive properties; (3) micromass cultures, in which cells are pelleted and suspended in media containing appropriate amounts of nutrients and differentiation factors; (4) free cells in a rotating vessel that adhere to one another and eventually form tissue- or organlike structures (so-called rotating wall vessels or microgravity bioreactors); and (5) gel-based techniques, in which cells are embedded in a substrate, such as agarose or 2 Although the effect of gravity on tissue differentiation and organization had been recognized for some time, the use of a rotationbased technique by Moscona (1961) to produce aggregates from cells in suspension was a notable technical advance.
matrigel, that may or may not contain a scaffolding of collagen or other organic or synthetic fiber which mimics the ECM. Cells in culture, particularly neurons, require a porous matrix with the specific adhesive properties for attachment and process elaboration, as well as the permeability to allow nutrient and waste transport. The matrix should provide some degree of structural support. In this regard, gel matrixbased systems appear to offer a number of advantages over the other four methodologies already mentioned. First, they are significantly less susceptible than other 3-D techniques to mechanical perturbations that could disrupt such key processes as cell aggregation. Microgravity-based systems utilizing a rotating culture vessel, for example, may be more prone to such disturbances. Second, gel matrices allow for the introduction of embedded scaffolding or the attachment of proteins engineered to simulate the known characteristics of the ECM, as well as the placement of microelectrode grids. Third, they can be designed to realistically emulate actual observed in vivo cell and surrounding tissue densities. Finally, they are amenable to repeated manipulations that would be difficult or even impossible to perform in other 3-D systems without compromising ongoing cultures. In the case of neuronal cultures, phenomena such as cell proliferation, cell movement, cell differentiation, neurite extension and pathfinding, axonal repair, cell survival, and the ECM properties in all of the foregoing can be readily examined in three-dimensional gel matrix systems. 3-D neuronal cultures exhibit realistic structural and functional properties. Along these lines, Ma et al. (2004) in this issue have demonstrated that E13 rat cortical progenitor cells can be expanded and differentiated when seeded both within and on the surface of a collagen gel matrix. In this study, the progenitors aggregated into clusters of clone-like cells within the collagen matrix and formed cellular networks at the surface of the collagen that were accessible for patch-clamping. Immunochemistry revealed a few neuronal marker- positive cells as early as day 2 in vitro; astrocytes and oligodendrocytes did not appear before day 10. At day 5 in vitro, most of the cells still stained for BrdU, a marker for actively dividing cells— and exhibited cytoplasmic Ca2+ oscillations in response to ATP or carbachol, but not GABA or glutamate. In contrast, by day 14 in vitro, 60% of the cells responded to GABA and glutamate and were positive for neuronal markers. These data show a developmental progression in 3-D collagen culture similar to that seen in vivo. The patchclamp data acquired from the surface-seeded progenitors resembled that observed in primary neurons (O’Shaughnessy et al., 2003). Previous work by the authors and their collaborators has demonstrated the superiority of collagen 3-D matrices over polyacrylate gels in promoting growth of rat cortical neurons, astrocytes, and E13 rat telencephalon neural progenitor cells (O’Connor et al., 2000a). The collagen
D.B. Edelman, E.W. Keefer / Experimental Neurology 192 (2005) 1–6
scaffolding in the gel promoted growth and process extension of cells cultured on its surface and embedded within it. Progenitor cells differentiated into both neuron and astrocyte morphologies when seeded and grown in the collagen matrix. A particular motivation for this research was the potential offered by neural progenitor cells grown in this manner as a renewable tissue source for biosensor modules. The structural stability of such 3-D systems represents a vast improvement over that of 2-D cultures used by other researchers (Keefer et al., 2001a; Stenger et al., 2001). Moreover, the same researchers showed that when neural precursor cells are allowed to grow in a collagen matrix for 7 days, the response of the cells to exogenous glutamate application can be monitored by Ca2+ imaging techniques (O’Connor et al., 2000b). The entrapped precursors divided in the presence of bFGF and formed clone-like spheres, each with a proliferative center and differentiated cells at the sphere surface. These cells expressed markers characteristic of neurons and astrocytes and migrated radially from the cell clusters. Electrophysiological studies in which E18 rat cortical cells were grown for 8–12 days on collagen gels and recorded by patch clamping revealed normal membrane properties, Na+ and K+ currents, spontaneous action potentials, and post-synaptic currents that could be blocked entirely by bicuculline, the GABAA receptor antagonist (O’Shaughnessy et al., 2003). Culturing the cells on a collagen surface instead of embedding them within the gel permitted access by patch-pipette. This study is a logical extension of earlier work by the researchers and their collaborators, and provides imaging, electrophysiological, pharmacological, and developmental data on neural progenitors in a more realistic 3-D context. Taken together, the body of work published by these researchers offers convincing proof that collagen gels support both primary neuronal and neural progenitor growth and function. Ma et al. (2004) propose this technology as a means of engineering nervous tissue replacements for repair of brain and spinal cord injuries, as well as for damage inflicted by neurodegenerative diseases. Neuronal cell culture has already provided a wealth of fundamental data on electrical properties, pharmacological responses, developmental dynamics, growth factor requirements, cell–cell and cell–substrate interactions, synaptic properties, and many other aspects of neuron and glial biology. However, 2-D culturing of dissociated neurons, while allowing proliferation and growth of electrically active neuronal networks for periods in excess of 12 months in both our laboratory and those of others, disregards the three-dimensional physical arrangement of neurons in vivo. This bflat-earthQ view of the cellular and histological worlds can lead to incorrect characterizations of the physiology of neurons and the networks they form. In comparisons of hippocampal neurons cultured in 2-D and 3-D (tissue slice) systems, differences have been identified in the induction and propagation of long-term potentiation (LTP), a commonly studied form of synaptic
3
plasticity that may be important in the formation and maintenance of memories (Tao et al., 2000). These differences might be attributable either to dilution of extracellular diffusible factors in the 2-D context or, intriguingly, the possible physical interactions of dendritic spines with the ECM that may constrain the remodeling of synaptic proteins in 2-D. In our laboratory, cultures of embryonic rat hippocampal cells on glass coverslips coated with poly-d-lysine and laminin typically produce neurons with extensive neuritic arborizations. However, these processes rarely exceed 1 mm in length. Hippocampal neurons that are seeded on coverslips and subsequently covered with Type I collagen at 48C and grown at 378C extend neurites through the entire thickness of the overlying collagen layer (approximately 5 mm) and arborize extensively at the collagen–culture medium interface. Studies of degenerative conditions such as Alzheimer’s and Parkinson’s disease may be greatly aided by this culture paradigm. Indeed, a common proximal event in the development of these and other neurodegenerative disorders appears to be the loss of innervations from projection axons extending many millimeters from the origination site. Constraints on axonal length imposed by 2-D substrates in vitro may actually obscure initial degenerative events in cellular compartments that are very remote from the cell body. In the human cortex, there are approximately 105 neurons per mm3, and each cell may synapse with 1000 or more distant neurons, but typically has only a few synaptic connections with each of its immediate neighbors (Kandel et al., 1991). This situation is entirely different from that observed in 2-D cultures, where cells are typically grown at low density to promote visibility, and each neuron may synapse several hundred times with adjacent neurons. Use of 3-D culturing techniques that incorporate some form of patterning to guide and align neurites may eventually yield high-density cell culture constructs with sufficient oriented dendritic growth to permit the generation of field potentials and other nonspiking oscillations observed both in vivo and in tissue slices. Growth of such constructs around a pre-existing microwire multi-electrode array would permit recording of both single unit and field potentials, thus enhancing future neuropharmacological studies. Although 2-D culture has yielded long-term action potential recordings of multiple drug applications (Keefer et al., 2001b), it does not support the generation of field potentials. In contrast, tissue slice cultures allow circuit-specific drug effects to be observed from both field and spike recordings (Heuschkel et al., 2002); unfortunately, the viability of the slice is typically only a few hours. High-density 3-D cultures may achieve the longevity of recordings seen in 2-D and, in addition, may produce field potential information, as well as spike data, thus providing both pre- and post-synaptic information which will greatly inform and expedite interpretation of pharmacological effects.
4
D.B. Edelman, E.W. Keefer / Experimental Neurology 192 (2005) 1–6
Ex vivo 3-D tissue engineering and its therapeutic applications. Three-dimensional growth in culture has been demonstrated in diverse tissue types, including cardiac myocytes (Akins et al., 1999; Dar et al., 2002; Evans et al., 2003), osteoblasts (Botchwey et al., 2003; Ferrera et al., 2002; Karp et al., 2002; Qiu et al., 2001), myoblasts (Bouten et al., 2003; Li et al., 2002; Stegemann and Nerem, 2003), chondrocytes (Dumas et al., 2000; Hung et al., 2003; Kisiday et al., 2002), hepatocytes (Richert et al., 2002; Selden et al., 1999), cerebral microvascular endothelial cells (Chow et al., 2001), mesothelial and endothelial cells (Bittinger et al., 1997), and odontoblasts (Camps et al., 2002). Most of these studies were directed toward providing exogenously grown replacement parts for injured and aged subjects. In this approach, stem cells of the appropriate lineage would be isolated from the patient, expanded in a bio-compatible, non-immunogenic matrix, and then implanted back into the donor/patient. Replacement of worn cartilage in the knee with an ex vivo engineered, pre-formed matrix containing autologous chondrocytes could keep a marathoner running for years longer than is possible today. Perhaps more significantly, such a therapy could provide relief and greater mobility to an ever-growing population of osteoarthritis sufferers. The type of neural engineering proposed by Ma et al. (2004) will entail many additional difficulties beyond those involved in cartilage replacement. Successful implantation and growth of neural cells in a damaged brain does not imply effective functionality. The ongoing business of neurons in the brain is to receive input stimuli from defined sources, filter this information, and propagate appropriate signals to other neurons. Neurons are connected in epigenetically specified networks of stupefying intricacy. Restoration of partial function in spinal cord-injured rats and cats (Houle and Tessler, 2003), some alleviation of experimentally induced and disease-induced Parkinson’s symptoms (Spector et al., 1993; Limke and Rao, 2002), and attenuation of Alzheimer’s progression (Tuszynski and Blesch, 2004) have been achieved, with varying degrees of success, through implantation of neural tissue. However, many of the improvements seen are due to trophic factors produced by the implanted cells, and not the integration of these cells into damaged neural circuits. Indeed, reports of transplanted stem cells integrating and beginning to function in brains are likely attributable to cell-fusion events rather than growth and incorporation of cells into pre-existing circuits (Weimann et al., 2003). Moreover, any assessment of the success or failure of attempts at neural tissue implantation must be tempered by both the reality of the rich heterogeneity of cell types represented in a given sample of brain tissue and the profound gaps in our understanding of the dynamics of, and interactions between, these diverse cell types. The degree of possible neural restoration envisioned by Ma et al. (2004) will require selective propagation of specific neuronal and glial subtypes in physiologically appropriate ratios. The success of procedures to implant
such constructs into injury sites, such as a bilaterally lesioned spinal cord, will depend upon the subsequent integration of the ex vivo construct into undamaged neural tissue on either side of the injury; maintenance of the plastic nature of the cells within the construct will be critical for full functional recovery. Since specific motor patterns generated before the lesion will have been lost, persistence of plasticity in the transplanted cells will be required in order for the patient to relearn approximations of these lost patterns, should the integration be successful. In addition, engineered constructs may require incorporation of cells that have been genetically altered to constitutively produce neurotrophins such as BDNF, which has been implicated in many forms of synaptic plasticity (Lessmann et al., 2003). 3-D cell culture and basic neuroscience: a more realistic view of the nervous system ex vivo. The deployment of 3D cell culture in neuroscience signals a real sea change in the way researchers view cell behavior. What has emerged is the practical realization that context cannot be ignored when we attempt to emulate aspects of the nervous system ex vivo. Although the use of 3-D cell culture techniques is not yet widespread in neuroscience, its impact on both basic and applied research will be tremendous. In basic neuroscience, well thought-out engineered matrices with ECM-like scaffolding will make it possible to emulate and dissect structural and dynamic behavioral attributes of neural systems at a level of resolution that cannot be achieved in the living animal. The degree of sophistication that will be required to develop realistic 3-D culture bmodelsQ of aspects of the nervous system should not be underestimated and certainly will not be realized in a vacuum; the effort will still require data gathered from ongoing in vivo neurophysiological, developmental neurobiological, and genetic studies. In applied neuroscience, innovations in 3-D cell culture will make it possible to both build a vastly more comprehensive picture of the mechanisms and progression of a number of neurological disease states and engineer neural structures ex vivo from autologous stem and neuroprogenitor cells with eventual transplantation into patients. The extracellular matrix of the CNS contains very little collagen or fibrillar matrix components. Therefore, extending results such as those obtained by Ma et al. (2004) with collagen into an in vivo context may prove difficult. In addition, the practical limitation to growing any tissue, not just the brain, is the distance-limited diffusion of nutrients and waste products. In vivo, vascular perfusion is the limiting factor for the growth of organs (Schmidt-Nielsen, 1994), and recognition of this fact has yielded great strides in solid tumor therapies (Folkman, 1971). Interesting results were reported earlier this year utilizing self-assembled nanofiber scaffolds, constructed of a bio-active pentapeptide sequence found in laminin attached to a hydrophobic alkyl tail group (Silva et al., 2004). When a 0.5% by weight solution of the peptide was seeded with a suspension of neural progenitor cells, within seconds, a gel formed. The neural progenitors survived up to 22 div. and differentiated
D.B. Edelman, E.W. Keefer / Experimental Neurology 192 (2005) 1–6
rapidly into neurons and astrocytes as determined by immunochemistry. Interestingly, the rate of neuronal differentiation was much faster in the 3-D scaffold than on coverslips coated with the same laminin-derived pentapeptide sequence, and the cell fate was profoundly altered by the dimensionality of their environment. The authors further show that it is the density of the laminin epitope that influences the rate of differentiation. Importantly, these results may also aid in resolving some of the diffusion limitations on organ growth, as the nanofiber scaffolding used was 99.5% water, a property that should enable perfusion of the embedded cells. Other concerns with utilizing progenitor cells to offset neuronal attrition is that neurons are but one cell type within the CNS, neurons are surrounded on all sides by a network of glial cells, working in concert with the neuronal cells to form a functional organ. Glial roles in neuronal nourishment, trophic support, neurotransmitter uptake and recycling, axon myelination, stress-signaling, etc. are very well studied, but much remains unknown. In typical 2-D neuronal culture, glial cells form a substrate on which neuronal somata lie. Neuronal/glial process interdigitation occurs to some extent, but not on the scale possible in 3-D. Can the appropriate ratios of neurons to glia, with appropriate timing of cell–cell interactions, be recapitulated by introducing progenitor cells into a post-mitotic milieu? Also, the observation of Silva et al. (2004) that matrix adhesion epitope density profoundly influences the cell fate and differentiation rate may be telling in the context of the living brain, where cell density greatly exceeds that demonstrated in any in vitro context to date. Do the studies that were already done support the idea that progenitor cells, even when differentiated into appropriate cells types, actually integrate into the existing neuronal circuitry, to the extent that they provide a functional benefit? It appears that such questions, if they can be addressed at all in vitro, can be best explored in a 3-D context. Comparisons of the influence of dimensionality on basic neuronal functions, such as intracellular transport, dendritic branching, synaptogenesis, cell migration, and axonal targeting will undoubtedly yield many insights. The work by Ma et al. (2004) is a necessary beginning.
Acknowledgment The work of the authors is supported by The Neurosciences Research Foundation.
References Akins, R.E., Boyce, R.A., Madonna, M.L., Schroedl, N.A., Gonda, S.R., McLaughlin, T.A., Hartzell, C.R., 1999. Cardiac organogenesis in vitro: reestablishment of three-dimensional tissue architecture by dissociated neonatal rat ventricular cells. Tissue Eng. 5 (2), 103 – 118.
5
Bittinger, F., Brochhausen, C., Skarke, C., Kohler, H., Kirkpatrick, C.J., 1997. Reconstruction of peritoneal-like structure in three-dimensional collagen gel matrix culture. Exp. Cell Res. 236, 155 – 160. Botchwey, E.A., Pollack, S.R., El-Amin, S., Levine, E.M., Tuan, R.S., Laurencin, C.T., 2003. Human osteoblast-like cells in three-dimensional culture with fluid flow. Biorheology 40, 299 – 306. Bouten, C.V.C., Breuls, R.G.M., Peeters, E.A.G., Oomens, C.W.J., Baaijens, F.P.T., 2003. In vitro models to study compressive straininduced muscle cell damage. Biorheology 40, 383 – 388. Camps, J., About, I., Thonneman, B., Mitsiadis, T.A., Schmaltz, G., Franquin, J.-C., 2002. Two- versus three-dimensional in vitro differentiation of human pulp cells into odontoblastic cells. Connect. Tissue Res. 43, 396 – 400. Chow, J., Ogunshola, O., Fan, S.-Y., Li, Y., Ment, L.R., Madri, J.A., 2001. Astrocyte-derived VEGF mediates survival and tube stabilization of hypoxic brain microvascular endothelial cells in vitro. Dev. Brain Res. 130, 123 – 132. Dar, A., Shachar, M., Leor, J., Cohen, S., 2002. Cardiac tissue engineering: optimization of cardiac cell seeding and distribution in 3D porous alginate scaffolds. Biotechnol. Bioeng. 80, 305 – 312. Dumas, D., Gigant, C., Presle, N., Cipolletta, C., Miralles, G., Payan, E., Jouzeau, J.Y., Mainard, D., Terlain, B., Netter, P., Stoltz, J.F., 2000. The role of 3D microscopy in the study of chondrocyte-matrix interaction (alginate bead or sponge, rat femoral head cap, human osteoarthritic cartilage) and pharmacological application. Biorheology 37, 165 – 176. Evans, H.J., Sweet, J.K., Price, R.L., Yost, M., Goodwin, R.L., 2003. Novel 3D culture system for study of cardiac myocyte development. Am. J. Physiol.: Heart Circ. Physiol. 285, H570 – H578. Ferrera, D., Poggi, S., Biassoni, C., Dickson, G.R., Astigiano, S., Barbieri, O., Favre, A., Franzi, A.T., Strangio, A., Federici, A., Manduca, P., 2002. Three-dimensional cultures of normal human osteoblasts: proliferation and differentiation potential in vitro and upon ectopic implantation in nude mice. Bone 30 (5), 718 – 725. Folkman, J., 1971. Tumor angiogenesis: therapeutic implications. N. Engl. J. Med. 285, 1182 – 1186. Heuschkel, M.O., Fejtl, M., Raggenbass, M., Bertrand, D., Renaud, P., 2002. A three-dimensional multi-electrode array for multi-site stimulation and recording in acute brain slices. J. Neurosci. Methods 114, 135 – 148. Holtfreter, J., 1943a. A study of the mechanics of gastrulation: part I. J. Exp. Zool. 94, 261 – 318. Holtfreter, J., 1943b. A study of the mechanics of gastrulation: part II. J. Exp. Zool. 95, 171 – 212. Holtfreter, J., 1944. Neural differentiation of ectoderm through exposure to saline solution. J. Exp. Zool. 95, 307 – 340. Holtfreter, J., Hamberger, V., 1955. Embryogenesis: progressive differentiation. Chap. 1: Amphibians. In: Willier, B.H., Weiss, P.A., Hamberger, V. (Eds.), Analysis of Development. Haffner, New York, pp. 230 – 296. Houle, J.D., Tessler, A., 2003. Repair of chronic spinal cord injury. Exp. Neurol. 182, 247 – 260. Hung, C.T., LeRoux, M.A., Palmer, G.D., Chao, P.-H.G., Lo, S., Valhmu, W.B., 2003. Disparate aggrecan gene expression in chondrocytes subjected to hypotonic and hypertonic loading in 2D and 3D culture. Biorheology 40, 61 – 72. Kandel, E.R., Schwartz, J.H., Jessel, T.M., 1991. Principles of Neural Science, third ed. Harlow Books, New York. Karp, J.M., Shoichet, M.S., Davies, J.E., 2002. Bone formation on twodimensional poly(dl-lactide-co-glycolide) (PLGA) films and threedimensional PLGA tissue engineering scaffolds in vitro. J. Biomed. Mater. Res. 64A, 388 – 396. Keefer, E.W., Gramowski, A., Stenger, D.A., Pancrazio, J.J., Gross, G.W., 2001a. Characterization of acute neurotoxic effects of trimethylolpropane phosphate via neuronal network biosensors. Biosens. Bioelectron. 16, 513 – 525. Keefer, E.W., Gramowski, A., Gross, G.W., 2001b. NMDA receptor-
6
D.B. Edelman, E.W. Keefer / Experimental Neurology 192 (2005) 1–6
dependent oscillations in cultured spinal cord networks. J. Neurophysiol. 86 (6), 3030 – 3042. Kisiday, J., Jin, M., Kurz, B., Hung, H., Semino, C., Zhang, S., Grodzinsky, A.J., 2002. Self-assembling peptide hydrogel fosters chondrocyte extracellular matrix production and cell division: implications for cartilage tissue repair. Proc. Natl. Acad. Sci. U. S. A. 99 (15), 9996 – 10001. Lessmann, V., Gottmann, K., Malcangio, M., 2003. Neurotrophin secretion: facts and future prospects. Prog. Neurobiol. 69 (5), 341 – 374. Li, S., Lao, J., Chen, B.P.C., Li, Y.-S., Zhao, Y., Chu, J., Chen, K.-D., Tsou, T.-C., Peck, K., Chien, S., 2002. Genomic analysis of smooth muscle cells in 3-dimensional collagen matrix. FASEB J. 17, 97 – 99. Limke, T.L., Rao, M.S., 2002. Neural stem cells in aging and disease. J. Cell. Mol. Med. 6, 475 – 496. Ma, W., Fitzgerald, W., Liu, Q.-Y., O’Shaughnessy, T.J., Maric, D., Lin, H.J., Alkon, D.L., Barker, J.L., 2004. CNS stem and progenitor cell differentiation into functional neuronal circuits in three-dimensional collagen gels. Exp. Neurol. 190, 276 – 288. Moscona, A.A., 1952. Cell suspension from organ rudiments of chick embryos. Exp. Cell Res. 3, 535 – 539. Moscona, A.A., 1961. Rotation-mediated histogenetic aggregation of dissociated cells: a quantifiable approach to cell interactions in vitro. Exp. Cell Res. 22, 455 – 475. Moscona, A.A., 1963. Studies on cell aggregation: demonstration of materials with selective cell-binding activity. Proc. Natl. Acad. Sci. U. S. A. 49, 742 – 747. O’Connor, S.M., Stenger, D.A., Shaffer, K.M., Maric, D., Barker, J.L., Ma, W., 2000a. Primary neural precursor cell expansion, differentiation and cytosolic Ca2+ response in three-dimensional collagen gel. J. Neurosci. Methods 102, 187 – 195. O’Connor, S.M., Andreadis, J.D., Shaffer, K.M., Ma, W., Pancrazio, J.J., Stenger, D.A., 2000b. Immobilization of neural cells in three-dimensional matrices for biosensor applications. Biosens. Bioelectron. 14, 871 – 881. O’Shaughnessy, T.J., Lin, H.J., Ma, W., 2003. Functional synapse formation among rat cortical neurons grown on three-dimensional collagen gels. Neurosci. Lett. 340, 169 – 172. Qiu, Q.-Q., Ducheyne, P., Ayyaswamy, P.S., 2001. 3D bone tissue engineered with bioactive microspheres in simulated microgravity. In Vitro Cell. Dev. Biol.: Anim. 37, 157 – 165. Richert, L., Binda, D., Hamilton, G., Viollon-Abadie, C., Alexandre, E., Bigot-Lasserre, D., Bars, R., Coassolo, P., LeCluyse, E., 2002. Evaluation of the effect of culture configuration on morphology, survival time, antioxidant status and metabolic capacities of cultured rat hepatocytes. Toxicol. in Vitro 16, 89 – 99. Sahai, E., Marshall, C.J., 2003. Differing modes of tumour cell invasion
have distinct requirements for Rho/ROCK signaling and extracellular proteolysis. Nat. Cell Biol. 5 (8), 711 – 719. Sasaki, T., Takagi, M., Soma, T., Yoshida, T., 2002. 3D culture of murine hematopoietic cells with spatial development of stromal cells in nonwoven fabrics. Cytotherapy 4 (3), 285 – 291. Schmidt-Nielsen, B., 1994. August Krogh and capillary physiology. Int. J. Microcirc.: Clin. Exp. 14, 104 – 110. Selden, C., Shariat, A., McCloskey, P., Ryder, T., Roberts, E., Hodgson, H., 1999. Three-dimensional in vitro cell culture leads to a marked upregulation of cell function in human hepatocyte cell lines—an important tool for the development of a bioartificial liver machine. Ann. N. Y. Acad. Sci. 875, 353 – 363. Silva, G.A., Czeisler, C., Niece, K.L., Beniash, E., Harrington, D.A., Kessler, J.A., Stupp, S.I., 2004. Selective differentiation of neural progenitor cells by high-epitope density nanofibers. Science 303, 1352 – 1355. Spemann, H., Mangold, H., 1924. eber induktion von embryonalanlagen durch implantation artfremder organisatoren. Arch. Mikrosk. Anat. Entwickl. Mech. 100, 599 – 638. Spector, D.H., Boss, B.D., Strecker, R.E., 1993. A model threedimensional culture system for mammalian dopaminergic precursor cells: application for functional intracerebral transplantation. Exp. Neurobiol. 124, 253 – 264. Stegemann, J.P., Nerem, R.M., 2003. Altered response of vascular smooth muscle cells to exogenous biochemical stimulation in two- and threedimensional culture. Exp. Cell Res. 283, 146 – 155. Stenger, D.A., Gross, G.W., Keefer, E.W., Shaffer, K.M., Andreadis, J.D., Ma, W., Pancrazio, J.J., 2001. Detection of physiologically active compounds using cell-based biosensors. Trends Biotechnol. 19 (8), 304 – 309. Tao, H-z.W., Zhang, L.I., Bi, G-q., Poo, M-m., 2000. Selective presynaptic propagation of long-term potentiation in defined neural networks. J. Neurosci. 20 (9), 3233 – 3243. Townes, P.L., Holtfreter, J., 1947. Directed movements and selective adhesion of embryonic amphibian cells. J. Exp. Zool. 128, 53 – 120. Tuszynski, M.H., Blesch, A., 2004. Nerve growth factor: from animal models of cholinergic neuronal degeneration to gene therapy in Alzheimer’s disease. Prog. Brain Res. 146, 441 – 449. Weaver, V.M., Petersen, O.W., Wang, F., Larabell, C.A., Briand, P., Damsky, C., Bissell, M.J., 1997. Reversion of the malignant phenotype of human breast cells in three-dimensional culture an in vivo by integrin blocking antibodies. J. Cell Biol. 137, 231 – 245. Weimann, J.M., Johansson, C.B., Trejo, A., Blau, H.M., 2003. Stable reprogrammed heterokaryons form spontaneously in Purkinje neurons after bone marrow transplant. Nat. Cell Biol. 5 (11), 959 – 966. Wilson, H.V.P., 1907. On some phenomena of coalescence and regeneration in sponges. J. Exp. Zool. 5, 245 – 258.
