- Email: [email protected]
Signaling in morphogenesis: transport cues in morphogenesis
547
Signaling in morphogenesis: transport cues in morphogenesis Melody A Swartz Extracellular transport processes play critical roles in morphogenesis. While diffusive transport effects on morphogenesis are well illustrated in examples like blood capillary architecture and in cell morphogenetic responses to the local extracellular protein environment, the effects of fluid convection, although important in many developing and regenerating tissues, are not well understood. Convective forces are present whenever a hydrated tissue undergoes dynamic mechanical strain, and so convection could not only dominate the transport of large molecules like proteins, but might also serve as a mechanism for mechanosensing. The complex interdependence of mechanical forces, protein transport and extracellular morphogen gradients needs to be elucidated in a comprehensive way in order for the importance of transport on morphogenesis to be fully appreciated. Addresses Institute for Biological and Chemical Engineering, Swiss Federal Institute of Technology (EPFL), CH-1015 Lausanne, Switzerland e-mail: [email protected]
Current Opinion in Biotechnology 2003, 14:547–550 This review comes from a themed issue on Tissue and cell engineering Edited by Jeffrey Hubbell 0958-1669/$ – see front matter ß 2003 Elsevier Ltd. All rights reserved. DOI 10.1016/j.copbio.2003.09.003
Abbreviations EGF epidermal growth factor EGFR EGF receptor VEGF vascular endothelial growth factor
Introduction Morphogenesis in cells and tissues is the result of a complex series of highly orchestrated events that lead to the organization of remarkably specialized cells and cell societies. These events are strongly regulated by the local biochemical environment: morphogenesis-inducing proteins (or morphogens), such as members of the transforming growth factor, wingless, hedgehog and bone morphogenetic protein families, affect embryonic patterning and morphogenesis (including regeneration) according to their extracellular concentration gradients. This article briefly surveys the ways in which transport processes can affect the spatial distribution of extracellular proteins, and therefore tissue morphogenesis, and www.current-opinion.com
speculates on the potential importance of convective transport on morphogenesis.
Diffusive transport and morphogenesis On a broad scale, blood vessel development is a wellstudied example of how diffusive transport affects tissue morphogenesis. The main function of the capillaries is to exchange oxygen and carbon dioxide with the tissue beds they serve, and thus capillary morphogenesis is fundamentally determined by the oxygen needs of the tissue. When tissue metabolism increases, either because of an increase in cell number or metabolic activity, new capillaries are formed in a process termed angiogenesis. This dependence of angiogenesis on oxygen transport can be seen both from estimates of the maximum capillary distance using diffusion and consumption of oxygen (e.g. [1–3]), and also by looking at the role of oxygen or hypoxia in gene regulation related to angiogenesis. For example, many of the polypeptide growth factors that govern angiogenesis, such as vascular endothelial growth factor (VEGF) and angiopoietin-2 (reviewed in [4]), are expressed under the influence of the oxygen-sensitive transcription factors like hypoxia-inducing factor-I [5]. Thus, the oxygen needs of the tissue — determined by the diffusive transport and consumption of oxygen — can directly induce transcription of the genes that control angiogenesis. When the production, relative concentrations or timing of such signals are off balance, transport function of the developing capillary network can change drastically. In a rapidly growing tumor, for example, VEGF and other growth factors can be upregulated to a large extent relative to other factors such that leaky vessels are formed with suboptimal architecture, including heterogeneity, increased tortuosity and redundancies like capillary loops [4]. The strongly coordinated interplay between physiological transport needs (e.g. oxygen diffusion) and molecular players (e.g. VEGF) is therefore necessary to drive the development of an optimally functional capillary bed. On the cellular level, morphogenesis is tightly controlled by extracellular concentrations and concentration gradients of morphogens, with different cell fates resulting from different morphogen concentrations [6–8]. In this way, gradients and/or thresholds in morphogenetic responses can result within groups of cells and give rise to pattern formation; for example, the graded distribution of Wingless (Wg) protein in Drosophila may be responsible for guiding cell polarity [9]. Surprisingly, there remains some debate in the literature about whether such gradients do indeed arise by passive diffusion (e.g. [10,11,12]) as opposed to active cellular propagation mechanisms. Current Opinion in Biotechnology 2003, 14:547–550
548 Tissue and cell engineering
These mechanisms include transcytosis (e.g. [6,13,14]) or a ‘bucket brigade’ mechanism, whereby the ligand is passed along a cell membrane by internalization and re-release and between cells by a series of receptor molecules [15]. Because of the complexity and nonlinearity of morphogen diffusion/reaction kinetics and the active cellular processes that have been shown to occur, it is almost certain that both passive and active mechanisms control extracellular morphogen distribution. For example, the interaction between Wg and one of its receptors, frizzled 2 (Dfz2), has been shown to slow Wg degradation and thus to allow build-up of its gradient; Dfz2 expression is in turn downregulated by Wg [16]. This example highlights the complexity and nonlinearity of the diffusion-reaction kinetics that govern even the simplest (i.e. those involving the fewest receptors and ligands) morphogenetic processes. The concept of morphogens as chemical ‘evocators’ whose net effect depends on extracellular concentration patterns formed by physical processes was first argued by the great mathematician Alan Turing in 1952 [17], long before the tools for demonstrating such phenomena were available. Using principles of diffusion and simple chemical reaction kinetics, he demonstrated how heterogeneity can spontaneously arise in extracellular spaces despite initially homogeneous conditions. The heterogeneous patterns are the result of different types of instabilities that depend on the kinetics of the morphogens present and their interactions with each other. He argued that such instabilities could potentially account for some of the developmental patterns seen in nature, such as standing waves on a ring giving rise to hydra tentacles and diffusion/reaction kinetics on a sphere giving rise to gastrulation. Although purely theoretical with no supporting experimental data, Turing provided the foundation for understanding the physical laws that govern morphogen gradients and give rise to embryonic patterning. With the current abundance of knowledge about specific morphogens and their receptors, binding kinetics, production and internalization rates, as well as the ability to visualize gradients in morphogens directly (e.g. [12]), conceptual and mathematical models based on some of Turing’s early ideas can give accurate predictions of morphogen gradients and broadcast signals as well as to the extent of autocrine versus paracrine regulation. In such models, the diffusion of morphogens away from a cell surface competes with receptor binding affinity and internalization rates to determine the local extracellular gradient and distance over which a signal can be broadcast. Such interactions have been extensively studied in epidermal growth factor receptor (EGFR) signaling, which can activate some fundamental morphogenetic processes including proliferation, mitogen-activated protein kinase (MAPK) signaling, endocytosis and oncogenesis [18]. The EGFR system can have autocrine effects in Current Opinion in Biotechnology 2003, 14:547–550
cells that depend on it for proliferation [19], such as fibroblasts and epithelial cells, as they produce both the receptor and a ligand. The higher the binding affinity of the ligand, the steeper the concentration gradient at the cell surface and the smaller the broadcast distance for paracrine regulation [20]. For example, transforming growth factor a (TGF-a), which has a high binding affinity for EGFR, is mostly consumed by the cell that produces it, whereas amphiregulin, which has a low binding affinity for EGFR, can act at longer distances and can thus affect neighboring cells or groups of cells. So, although diffusion creates the concentration gradients of morphogens, the kinetics of receptor binding, ligand production and internalization regulate the shape and size of that gradient. The cell cannot control passive diffusion, but it can regulate factors like receptor number and ligand type to control morphogen gradients and to affect neighboring cells. Diffusion does not generally depend on the local mechanical environment, with the exception of anisotropy that might arise in tissue structure due to tissue strain, which could, in turn, create anisotropy in the morphogen’s tissue diffusivity. This has not yet been addressed in models of morphogen gradients, although the general effects of anisotropic diffusion in heterogeneous porous media have been well characterized.
Convective transport and morphogenesis Although experimental and computational studies have demonstrated that diffusion-reaction mechanisms can give rise to the morphogen gradients responsible for patterning in morphogenesis, these studies are typically performed under static conditions. Experimental systems that could allow the identification and quantification of the contribution of interstitial convective transport to morphogen distribution do not currently exist. The importance of such phenomena may be realized, however, if one considers a single cell within a homogeneous extracellular matrix releasing a morphogen. Without interstitial convection, the morphogen gradient would be spherically symmetric and vary only radially according to diffusion and reaction, assuming that the receptors are equally distributed along the cell membrane. If interstitial flow were then introduced, this gradient would become skewed and lose its spherical symmetry; one would expect a more spread-out distribution on the downstream side of the cell giving rise to another dimension of spatial anisotropy in morphogen distribution. In turn, this asymmetry could direct different development or regeneration patterns. For example, it has been shown that autocrine epidermal growth factor (EGF) signaling can stimulate directionally persistent cell migration in static cultures [21]; one might then hypothetically speculate that the effects of interstitial flow on EGF gradients could further induce net directed migration. Of course, convective effects on EGF transport www.current-opinion.com
Transport cues in morphogenesis Swartz 549
would need to be of the same magnitude or greater than diffusive effects for convection to affect the extracellular EGF distribution. The ratio between convective and diffusive driving forces in any system, or the Peclet number (Pe), can be determined by comparing the approximate time td it would take for a molecule to travel a distance L by diffusion (td L2/D, where D is the diffusion coefficient) with the time tv it would take for that molecule to travel the same distance L by fluid convection (tv L/v, where v is the fluid velocity). Thus, Pe (¼ vL/D) can tell us when diffusion dominates the transport of a molecule (Pe << 1), when convection dominates (Pe >> 1) or when they both contribute (Pe 1). In a typical interstitial environment fed by a capillary bed, v might be on the order of 1 mm/s [22] and most large proteins like immunoglobulin G have a tissue diffusivity on the order of 107 cm2/s [23]. Thus, in this case diffusion would dominate the gradients of such a protein only within the first 1 mm, and convection would dominate distances on the order of 100 mm and greater. At intermediate distances (1–100 mm), both transport processes would need to be considered. Thus, where interstitial flow is present, the relevance of diffusion in governing the extracellular protein environment is likely to be low compared with that of protein convection, as most morphogenic proteins are quite large with small diffusivities in tissue (106– 107 cm2/s) compared with those of solutes like oxygen and glucose (for which capillary spacing was designed). (Note: Interstitial fluid velocities are very difficult to measure and can vary considerably with spatial location. Chary and Jain [22] give one of the few direct measurements; we might expect the actual range of velocities in a tissue to vary within an order of magnitude of this averaged value. The values used in this argument were only for the sake of demonstration.) Clearly, convection is a potentially important mechanism of interstitial protein transport. As the extracellular spatial distribution of morphogens strongly affects morphogenesis, convection is likely to be an important regulator of cell and tissue morphogenesis. Because the tissue environment is both dynamic and aqueous, fluid convection, however small, is usually present in most tissues: cilia movement, smooth muscle contractions, small interstitial pressure gradients, and tissue compression such as that seen in bone and cartilage all act to prevent fluid stagnation and drive fluid convection. Even in developing organisms, convective forces are often present; in the human embryo, for example, a beating heart develops in the first weeks and lung movements that mimic breathing are necessary for proper lung development. Despite its obvious importance, the interdependence between mechanical stress, fluid flow and the spatial distribution of morphogen concentration in the extracellular environment remains to be elucidated. Such convective effects www.current-opinion.com
are difficult to detect and to measure and, to date, they have not been considered or accounted for in any computational models of extracellular morphogen gradients or signaling kinetics. Finally, it is important to note that because fluid convection is either formed by stress (pressure) gradients within the tissue or causes extracellular matrix strain, the biological effects of interstitial fluid flow cannot be decoupled from those of mechanical stress. The importance of mechanical stress in tissue morphogenesis and regeneration is well appreciated; compressive strain in bone growth and calcification [24], shear stress in arterial wall remodeling (reviewed in [25,26]), and stretch in alveolar morphogenesis and repair [27–29] are but a few of the most well-studied examples. In any case, it is plausible that at least some ‘mechanosensing’ (e.g. the apparent ability of cells to sense acutely their mechanical environment) may actually be achieved through biochemical sensing of morphogen gradients. That is, rather than sense the stress or strain directly, the cell might only be sensitive to the changes in morphogen gradients that arise in response to interstitial fluid flow. In some instances it cannot be determined whether transport directly or indirectly affects a morphogenetic process. It was recently shown that lymphatic development in a regenerating skin model is at least partially directed by interstitial flow [30], which is consistent with the primary function of the lymphatic system (i.e. to provide interstitial convection and protein transport [31]). Because interstitial flow involves mechanical factors (e.g. shear stress on cells and the extracellular matrix) as well as transport effects (e.g. redistribution of growth factors, cytokines and proteases), these effects may result from either direct mechanical effects, direct convective transport effects or a combination of the two. Indeed, in a three-dimensional environment, it is difficult to determine whether cells are primarily responding to mechanical or biochemical cues, as interstitial flow is necessarily present in a mechanically active tissue and always affects local protein concentrations.
Conclusions Morphogenesis in development and regeneration is partly controlled by the extracellular spatial distribution patterns of morphogens, which are physically governed by transport processes of diffusion and convection. The role of diffusion coupled with production, internalization, and binding kinetics in governing the shapes and sizes of these morphogen gradients is well appreciated. However, the specific roles of convection have not yet been established. This important area of research warrants comprehensive consideration so that the interplay between the key factors that regulate morphogen distribution can be appreciated and more fully understood. Current Opinion in Biotechnology 2003, 14:547–550
550 Tissue and cell engineering
Acknowledgements The author is grateful to Vassily Hatzimanikatis for helpful comments.
References and recommended reading Papers of particular interest, published within the annual period of review, have been highlighted as: of special interest of outstanding interest 1.
Henquell L, Honig CR: Capillary spacing around coronary venules suggests that diffusion distance is controlled by local tissue pO2. Microvasc Res 1978, 15:363-366.
2.
Rakusan K, Hoofd L, Turek Z: The effect of cell size and capillary spacing on myocardial oxygen supply. Adv Exp Med Biol 1984, 180:463-475.
3.
Turek Z, Rakusan K, Olders J, Hoofd L, Kreuzer F: Computed myocardial PO2 histograms: effects of various geometrical and functional conditions. J Appl Physiol 1991, 70:1845-1853.
4.
Carmeliet P, Jain RK: Angiogenesis in cancer and other diseases. Nature 2000, 407:249-257.
5.
Bunn HF, Poyton RO: Oxygen sensing and molecular adaptation to hypoxia. Physiol Rev 1996, 76:839-885.
6.
Entchev EV, Schwabedissen A, Gonzalez-Gaitan M: Gradient formation of the TGF-b homolog Dpp. Cell 2000, 103:981-991.
7.
Tickle C: Morphogen gradients in vertebrate limb development. Semin Cell Dev Biol 1999, 10:345-351.
8.
Zecca M, Basler K, Struhl G: Direct and long-range action of a wingless morphogen gradient. Cell 1996, 87:833-844.
9.
Axelrod J, Miller J, Shulman J, Moon R, Perrimon N: Differential recruitment of Dishevelled provides signaling specificity in the planar cell polarity and Wingless signaling pathways. Genes Dev 1998, 12:2610-2622.
10. Lander AD, Nie Q, Wan FY: Do morphogen gradients arise by diffusion? Dev Cell 2002, 2:785-796. 11. McDowell N, Gurdon JB, Grainger DJ: Formation of a functional morphogen gradient by a passive process in tissue from the early Xenopus embryo. Int J Dev Biol 2001, 45:199-207.
17. Turing AM: The chemical basis of morphogenesis. Philos Trans R Soc Lond B Biol Sci 1952, 237:37-72. 18. Carpenter G: The EGF receptor: a nexus for trafficking and signaling. Bioessays 2000, 22:697-707. 19. Wiley HS, Shvartsman SY, Lauffenburger DA: Computational modeling of the EGF-receptor system: a paradigm for systems biology. Trends Cell Biol 2003, 13:43-50. 20. DeWitt A, Iida T, Lam HY, Hill V, Wiley HS, Lauffenburger DA: Affinity regulates spatial range of EGF receptor autocrine ligand binding. Dev Biol 2002, 250:305-316. Receptor binding affinity is shown to compete with diffusion to govern the gradient of EGF receptor ligands around a cell. When the affinity is strong and receptor-to-ligand ratio high, the morphogen binds before it can diffuse away and thus the higher affinity ligands act in autocrine fashion. 21. Maheshwari G, Wiley HS, Lauffenburger DA: Autocrine epidermal growth factor signaling stimulates directionally persistent mammary epithelial cell migration. J Cell Biol 2001, 155:1123-1128. 22. Chary SR, Jain RK: Direct measurement of interstitial convection and diffusion of albumin in normal and neoplastic tissues by fluorescence photobleaching. Proc Natl Acad Sci USA 1989, 86:5385-5389. 23. Berk DA, Yuan F, Leunig M, Jain RK: Direct in vivo measurement of targeted binding in a human tumor xenograft. Proc Natl Acad Sci USA 1997, 94:1785-1790. 24. Burger EH, Klein-Nulend J, Veldhuijzen JP: Mechanical stress and osteogenesis in vitro. J Bone Miner Res 1992, 7:S397-S401. 25. Kleinstreuer C, Hyun S, Buchanan JR Jr, Longest PW, Archie JP Jr, Truskey GA: Hemodynamic parameters and early intimal thickening in branching blood vessels. Crit Rev Biomed Eng 2001, 29:1-64. 26. Davies PF, Polacek DC, Shi C, Helmke BP: The convergence of haemodynamics, genomics, and endothelial structure in studies of the focal origin of atherosclerosis. Biorheology 2002, 39:299-306. 27. Liu M, Xu J, Tanswell AK, Post M: Stretch-induced growthpromoting activities stimulate fetal rat lung epithelial cell proliferation. Exp Lung Res 1993, 19:505-517.
12. Strigini M, Cohen SM: Wingless gradient formation in the Drosophila wing. Curr Biol 2000, 10:293-300. Direct evidence of diffusion-controlled morphogen gradients is shown here for the first time using fluorescent visualization techniques.
28. Wirtz HR, Dobbs LG: Calcium mobilization and exocytosis after one mechanical stretch of lung epithelial cells. Science 1990, 250:1266-1269.
13. Greco V, Hannus M, Eaton S: Argosomes: a potential vehicle for the spread of morphogens through epithelial. Cell 2001, 106:633-645.
29. Waters CM, Savla U: Keratinocyte growth factor accelerates wound closure in airway epithelium during cyclic mechanical strain. J Cell Physiol 1999, 181:424-432.
14. Narayanan R, Ramaswami M: Endocytosis in Drosophila: progress, possibilities, prognostications. Exp Cell Res 2001, 271:28-35.
30. Boardman KC, Swartz MA: Interstitial flow as a guide for lymphangiogenesis. Circ Res 2003, 92:801-808. One of the few studies that demonstrate a role for interstitial convection in morphogenesis. In regenerating skin, lymphatic regeneration only occurred in the direction of interstitial flow and in areas with greatly reduced interstitial flow, lymphatic capillaries never organized properly despite normal blood capillary development.
15. Kerszberg M: Morphogen propagation and action: towards molecular models. Semin Cell Dev Biol 1999, 10:297-302. 16. Cadigan KM, Fish MP, Rulifson EJ, Nusse R: Wingless repression of Drosophila frizzled 2 expression shapes the Wingless morphogen gradient in the wing. Cell 1998, 93:767-777.
Current Opinion in Biotechnology 2003, 14:547–550
31. Swartz MA: The physiology of the lymphatic system. Adv Drug Deliv Rev 2001, 50:3-20.
www.current-opinion.com
Signaling in morphogenesis: transport cues in morphogenesis Melody A Swartz Extracellular transport processes play critical roles in morphogenesis. While diffusive transport effects on morphogenesis are well illustrated in examples like blood capillary architecture and in cell morphogenetic responses to the local extracellular protein environment, the effects of fluid convection, although important in many developing and regenerating tissues, are not well understood. Convective forces are present whenever a hydrated tissue undergoes dynamic mechanical strain, and so convection could not only dominate the transport of large molecules like proteins, but might also serve as a mechanism for mechanosensing. The complex interdependence of mechanical forces, protein transport and extracellular morphogen gradients needs to be elucidated in a comprehensive way in order for the importance of transport on morphogenesis to be fully appreciated. Addresses Institute for Biological and Chemical Engineering, Swiss Federal Institute of Technology (EPFL), CH-1015 Lausanne, Switzerland e-mail: [email protected]
Current Opinion in Biotechnology 2003, 14:547–550 This review comes from a themed issue on Tissue and cell engineering Edited by Jeffrey Hubbell 0958-1669/$ – see front matter ß 2003 Elsevier Ltd. All rights reserved. DOI 10.1016/j.copbio.2003.09.003
Abbreviations EGF epidermal growth factor EGFR EGF receptor VEGF vascular endothelial growth factor
Introduction Morphogenesis in cells and tissues is the result of a complex series of highly orchestrated events that lead to the organization of remarkably specialized cells and cell societies. These events are strongly regulated by the local biochemical environment: morphogenesis-inducing proteins (or morphogens), such as members of the transforming growth factor, wingless, hedgehog and bone morphogenetic protein families, affect embryonic patterning and morphogenesis (including regeneration) according to their extracellular concentration gradients. This article briefly surveys the ways in which transport processes can affect the spatial distribution of extracellular proteins, and therefore tissue morphogenesis, and www.current-opinion.com
speculates on the potential importance of convective transport on morphogenesis.
Diffusive transport and morphogenesis On a broad scale, blood vessel development is a wellstudied example of how diffusive transport affects tissue morphogenesis. The main function of the capillaries is to exchange oxygen and carbon dioxide with the tissue beds they serve, and thus capillary morphogenesis is fundamentally determined by the oxygen needs of the tissue. When tissue metabolism increases, either because of an increase in cell number or metabolic activity, new capillaries are formed in a process termed angiogenesis. This dependence of angiogenesis on oxygen transport can be seen both from estimates of the maximum capillary distance using diffusion and consumption of oxygen (e.g. [1–3]), and also by looking at the role of oxygen or hypoxia in gene regulation related to angiogenesis. For example, many of the polypeptide growth factors that govern angiogenesis, such as vascular endothelial growth factor (VEGF) and angiopoietin-2 (reviewed in [4]), are expressed under the influence of the oxygen-sensitive transcription factors like hypoxia-inducing factor-I [5]. Thus, the oxygen needs of the tissue — determined by the diffusive transport and consumption of oxygen — can directly induce transcription of the genes that control angiogenesis. When the production, relative concentrations or timing of such signals are off balance, transport function of the developing capillary network can change drastically. In a rapidly growing tumor, for example, VEGF and other growth factors can be upregulated to a large extent relative to other factors such that leaky vessels are formed with suboptimal architecture, including heterogeneity, increased tortuosity and redundancies like capillary loops [4]. The strongly coordinated interplay between physiological transport needs (e.g. oxygen diffusion) and molecular players (e.g. VEGF) is therefore necessary to drive the development of an optimally functional capillary bed. On the cellular level, morphogenesis is tightly controlled by extracellular concentrations and concentration gradients of morphogens, with different cell fates resulting from different morphogen concentrations [6–8]. In this way, gradients and/or thresholds in morphogenetic responses can result within groups of cells and give rise to pattern formation; for example, the graded distribution of Wingless (Wg) protein in Drosophila may be responsible for guiding cell polarity [9]. Surprisingly, there remains some debate in the literature about whether such gradients do indeed arise by passive diffusion (e.g. [10,11,12]) as opposed to active cellular propagation mechanisms. Current Opinion in Biotechnology 2003, 14:547–550
548 Tissue and cell engineering
These mechanisms include transcytosis (e.g. [6,13,14]) or a ‘bucket brigade’ mechanism, whereby the ligand is passed along a cell membrane by internalization and re-release and between cells by a series of receptor molecules [15]. Because of the complexity and nonlinearity of morphogen diffusion/reaction kinetics and the active cellular processes that have been shown to occur, it is almost certain that both passive and active mechanisms control extracellular morphogen distribution. For example, the interaction between Wg and one of its receptors, frizzled 2 (Dfz2), has been shown to slow Wg degradation and thus to allow build-up of its gradient; Dfz2 expression is in turn downregulated by Wg [16]. This example highlights the complexity and nonlinearity of the diffusion-reaction kinetics that govern even the simplest (i.e. those involving the fewest receptors and ligands) morphogenetic processes. The concept of morphogens as chemical ‘evocators’ whose net effect depends on extracellular concentration patterns formed by physical processes was first argued by the great mathematician Alan Turing in 1952 [17], long before the tools for demonstrating such phenomena were available. Using principles of diffusion and simple chemical reaction kinetics, he demonstrated how heterogeneity can spontaneously arise in extracellular spaces despite initially homogeneous conditions. The heterogeneous patterns are the result of different types of instabilities that depend on the kinetics of the morphogens present and their interactions with each other. He argued that such instabilities could potentially account for some of the developmental patterns seen in nature, such as standing waves on a ring giving rise to hydra tentacles and diffusion/reaction kinetics on a sphere giving rise to gastrulation. Although purely theoretical with no supporting experimental data, Turing provided the foundation for understanding the physical laws that govern morphogen gradients and give rise to embryonic patterning. With the current abundance of knowledge about specific morphogens and their receptors, binding kinetics, production and internalization rates, as well as the ability to visualize gradients in morphogens directly (e.g. [12]), conceptual and mathematical models based on some of Turing’s early ideas can give accurate predictions of morphogen gradients and broadcast signals as well as to the extent of autocrine versus paracrine regulation. In such models, the diffusion of morphogens away from a cell surface competes with receptor binding affinity and internalization rates to determine the local extracellular gradient and distance over which a signal can be broadcast. Such interactions have been extensively studied in epidermal growth factor receptor (EGFR) signaling, which can activate some fundamental morphogenetic processes including proliferation, mitogen-activated protein kinase (MAPK) signaling, endocytosis and oncogenesis [18]. The EGFR system can have autocrine effects in Current Opinion in Biotechnology 2003, 14:547–550
cells that depend on it for proliferation [19], such as fibroblasts and epithelial cells, as they produce both the receptor and a ligand. The higher the binding affinity of the ligand, the steeper the concentration gradient at the cell surface and the smaller the broadcast distance for paracrine regulation [20]. For example, transforming growth factor a (TGF-a), which has a high binding affinity for EGFR, is mostly consumed by the cell that produces it, whereas amphiregulin, which has a low binding affinity for EGFR, can act at longer distances and can thus affect neighboring cells or groups of cells. So, although diffusion creates the concentration gradients of morphogens, the kinetics of receptor binding, ligand production and internalization regulate the shape and size of that gradient. The cell cannot control passive diffusion, but it can regulate factors like receptor number and ligand type to control morphogen gradients and to affect neighboring cells. Diffusion does not generally depend on the local mechanical environment, with the exception of anisotropy that might arise in tissue structure due to tissue strain, which could, in turn, create anisotropy in the morphogen’s tissue diffusivity. This has not yet been addressed in models of morphogen gradients, although the general effects of anisotropic diffusion in heterogeneous porous media have been well characterized.
Convective transport and morphogenesis Although experimental and computational studies have demonstrated that diffusion-reaction mechanisms can give rise to the morphogen gradients responsible for patterning in morphogenesis, these studies are typically performed under static conditions. Experimental systems that could allow the identification and quantification of the contribution of interstitial convective transport to morphogen distribution do not currently exist. The importance of such phenomena may be realized, however, if one considers a single cell within a homogeneous extracellular matrix releasing a morphogen. Without interstitial convection, the morphogen gradient would be spherically symmetric and vary only radially according to diffusion and reaction, assuming that the receptors are equally distributed along the cell membrane. If interstitial flow were then introduced, this gradient would become skewed and lose its spherical symmetry; one would expect a more spread-out distribution on the downstream side of the cell giving rise to another dimension of spatial anisotropy in morphogen distribution. In turn, this asymmetry could direct different development or regeneration patterns. For example, it has been shown that autocrine epidermal growth factor (EGF) signaling can stimulate directionally persistent cell migration in static cultures [21]; one might then hypothetically speculate that the effects of interstitial flow on EGF gradients could further induce net directed migration. Of course, convective effects on EGF transport www.current-opinion.com
Transport cues in morphogenesis Swartz 549
would need to be of the same magnitude or greater than diffusive effects for convection to affect the extracellular EGF distribution. The ratio between convective and diffusive driving forces in any system, or the Peclet number (Pe), can be determined by comparing the approximate time td it would take for a molecule to travel a distance L by diffusion (td L2/D, where D is the diffusion coefficient) with the time tv it would take for that molecule to travel the same distance L by fluid convection (tv L/v, where v is the fluid velocity). Thus, Pe (¼ vL/D) can tell us when diffusion dominates the transport of a molecule (Pe << 1), when convection dominates (Pe >> 1) or when they both contribute (Pe 1). In a typical interstitial environment fed by a capillary bed, v might be on the order of 1 mm/s [22] and most large proteins like immunoglobulin G have a tissue diffusivity on the order of 107 cm2/s [23]. Thus, in this case diffusion would dominate the gradients of such a protein only within the first 1 mm, and convection would dominate distances on the order of 100 mm and greater. At intermediate distances (1–100 mm), both transport processes would need to be considered. Thus, where interstitial flow is present, the relevance of diffusion in governing the extracellular protein environment is likely to be low compared with that of protein convection, as most morphogenic proteins are quite large with small diffusivities in tissue (106– 107 cm2/s) compared with those of solutes like oxygen and glucose (for which capillary spacing was designed). (Note: Interstitial fluid velocities are very difficult to measure and can vary considerably with spatial location. Chary and Jain [22] give one of the few direct measurements; we might expect the actual range of velocities in a tissue to vary within an order of magnitude of this averaged value. The values used in this argument were only for the sake of demonstration.) Clearly, convection is a potentially important mechanism of interstitial protein transport. As the extracellular spatial distribution of morphogens strongly affects morphogenesis, convection is likely to be an important regulator of cell and tissue morphogenesis. Because the tissue environment is both dynamic and aqueous, fluid convection, however small, is usually present in most tissues: cilia movement, smooth muscle contractions, small interstitial pressure gradients, and tissue compression such as that seen in bone and cartilage all act to prevent fluid stagnation and drive fluid convection. Even in developing organisms, convective forces are often present; in the human embryo, for example, a beating heart develops in the first weeks and lung movements that mimic breathing are necessary for proper lung development. Despite its obvious importance, the interdependence between mechanical stress, fluid flow and the spatial distribution of morphogen concentration in the extracellular environment remains to be elucidated. Such convective effects www.current-opinion.com
are difficult to detect and to measure and, to date, they have not been considered or accounted for in any computational models of extracellular morphogen gradients or signaling kinetics. Finally, it is important to note that because fluid convection is either formed by stress (pressure) gradients within the tissue or causes extracellular matrix strain, the biological effects of interstitial fluid flow cannot be decoupled from those of mechanical stress. The importance of mechanical stress in tissue morphogenesis and regeneration is well appreciated; compressive strain in bone growth and calcification [24], shear stress in arterial wall remodeling (reviewed in [25,26]), and stretch in alveolar morphogenesis and repair [27–29] are but a few of the most well-studied examples. In any case, it is plausible that at least some ‘mechanosensing’ (e.g. the apparent ability of cells to sense acutely their mechanical environment) may actually be achieved through biochemical sensing of morphogen gradients. That is, rather than sense the stress or strain directly, the cell might only be sensitive to the changes in morphogen gradients that arise in response to interstitial fluid flow. In some instances it cannot be determined whether transport directly or indirectly affects a morphogenetic process. It was recently shown that lymphatic development in a regenerating skin model is at least partially directed by interstitial flow [30], which is consistent with the primary function of the lymphatic system (i.e. to provide interstitial convection and protein transport [31]). Because interstitial flow involves mechanical factors (e.g. shear stress on cells and the extracellular matrix) as well as transport effects (e.g. redistribution of growth factors, cytokines and proteases), these effects may result from either direct mechanical effects, direct convective transport effects or a combination of the two. Indeed, in a three-dimensional environment, it is difficult to determine whether cells are primarily responding to mechanical or biochemical cues, as interstitial flow is necessarily present in a mechanically active tissue and always affects local protein concentrations.
Conclusions Morphogenesis in development and regeneration is partly controlled by the extracellular spatial distribution patterns of morphogens, which are physically governed by transport processes of diffusion and convection. The role of diffusion coupled with production, internalization, and binding kinetics in governing the shapes and sizes of these morphogen gradients is well appreciated. However, the specific roles of convection have not yet been established. This important area of research warrants comprehensive consideration so that the interplay between the key factors that regulate morphogen distribution can be appreciated and more fully understood. Current Opinion in Biotechnology 2003, 14:547–550
550 Tissue and cell engineering
Acknowledgements The author is grateful to Vassily Hatzimanikatis for helpful comments.
References and recommended reading Papers of particular interest, published within the annual period of review, have been highlighted as: of special interest of outstanding interest 1.
Henquell L, Honig CR: Capillary spacing around coronary venules suggests that diffusion distance is controlled by local tissue pO2. Microvasc Res 1978, 15:363-366.
2.
Rakusan K, Hoofd L, Turek Z: The effect of cell size and capillary spacing on myocardial oxygen supply. Adv Exp Med Biol 1984, 180:463-475.
3.
Turek Z, Rakusan K, Olders J, Hoofd L, Kreuzer F: Computed myocardial PO2 histograms: effects of various geometrical and functional conditions. J Appl Physiol 1991, 70:1845-1853.
4.
Carmeliet P, Jain RK: Angiogenesis in cancer and other diseases. Nature 2000, 407:249-257.
5.
Bunn HF, Poyton RO: Oxygen sensing and molecular adaptation to hypoxia. Physiol Rev 1996, 76:839-885.
6.
Entchev EV, Schwabedissen A, Gonzalez-Gaitan M: Gradient formation of the TGF-b homolog Dpp. Cell 2000, 103:981-991.
7.
Tickle C: Morphogen gradients in vertebrate limb development. Semin Cell Dev Biol 1999, 10:345-351.
8.
Zecca M, Basler K, Struhl G: Direct and long-range action of a wingless morphogen gradient. Cell 1996, 87:833-844.
9.
Axelrod J, Miller J, Shulman J, Moon R, Perrimon N: Differential recruitment of Dishevelled provides signaling specificity in the planar cell polarity and Wingless signaling pathways. Genes Dev 1998, 12:2610-2622.
10. Lander AD, Nie Q, Wan FY: Do morphogen gradients arise by diffusion? Dev Cell 2002, 2:785-796. 11. McDowell N, Gurdon JB, Grainger DJ: Formation of a functional morphogen gradient by a passive process in tissue from the early Xenopus embryo. Int J Dev Biol 2001, 45:199-207.
17. Turing AM: The chemical basis of morphogenesis. Philos Trans R Soc Lond B Biol Sci 1952, 237:37-72. 18. Carpenter G: The EGF receptor: a nexus for trafficking and signaling. Bioessays 2000, 22:697-707. 19. Wiley HS, Shvartsman SY, Lauffenburger DA: Computational modeling of the EGF-receptor system: a paradigm for systems biology. Trends Cell Biol 2003, 13:43-50. 20. DeWitt A, Iida T, Lam HY, Hill V, Wiley HS, Lauffenburger DA: Affinity regulates spatial range of EGF receptor autocrine ligand binding. Dev Biol 2002, 250:305-316. Receptor binding affinity is shown to compete with diffusion to govern the gradient of EGF receptor ligands around a cell. When the affinity is strong and receptor-to-ligand ratio high, the morphogen binds before it can diffuse away and thus the higher affinity ligands act in autocrine fashion. 21. Maheshwari G, Wiley HS, Lauffenburger DA: Autocrine epidermal growth factor signaling stimulates directionally persistent mammary epithelial cell migration. J Cell Biol 2001, 155:1123-1128. 22. Chary SR, Jain RK: Direct measurement of interstitial convection and diffusion of albumin in normal and neoplastic tissues by fluorescence photobleaching. Proc Natl Acad Sci USA 1989, 86:5385-5389. 23. Berk DA, Yuan F, Leunig M, Jain RK: Direct in vivo measurement of targeted binding in a human tumor xenograft. Proc Natl Acad Sci USA 1997, 94:1785-1790. 24. Burger EH, Klein-Nulend J, Veldhuijzen JP: Mechanical stress and osteogenesis in vitro. J Bone Miner Res 1992, 7:S397-S401. 25. Kleinstreuer C, Hyun S, Buchanan JR Jr, Longest PW, Archie JP Jr, Truskey GA: Hemodynamic parameters and early intimal thickening in branching blood vessels. Crit Rev Biomed Eng 2001, 29:1-64. 26. Davies PF, Polacek DC, Shi C, Helmke BP: The convergence of haemodynamics, genomics, and endothelial structure in studies of the focal origin of atherosclerosis. Biorheology 2002, 39:299-306. 27. Liu M, Xu J, Tanswell AK, Post M: Stretch-induced growthpromoting activities stimulate fetal rat lung epithelial cell proliferation. Exp Lung Res 1993, 19:505-517.
12. Strigini M, Cohen SM: Wingless gradient formation in the Drosophila wing. Curr Biol 2000, 10:293-300. Direct evidence of diffusion-controlled morphogen gradients is shown here for the first time using fluorescent visualization techniques.
28. Wirtz HR, Dobbs LG: Calcium mobilization and exocytosis after one mechanical stretch of lung epithelial cells. Science 1990, 250:1266-1269.
13. Greco V, Hannus M, Eaton S: Argosomes: a potential vehicle for the spread of morphogens through epithelial. Cell 2001, 106:633-645.
29. Waters CM, Savla U: Keratinocyte growth factor accelerates wound closure in airway epithelium during cyclic mechanical strain. J Cell Physiol 1999, 181:424-432.
14. Narayanan R, Ramaswami M: Endocytosis in Drosophila: progress, possibilities, prognostications. Exp Cell Res 2001, 271:28-35.
30. Boardman KC, Swartz MA: Interstitial flow as a guide for lymphangiogenesis. Circ Res 2003, 92:801-808. One of the few studies that demonstrate a role for interstitial convection in morphogenesis. In regenerating skin, lymphatic regeneration only occurred in the direction of interstitial flow and in areas with greatly reduced interstitial flow, lymphatic capillaries never organized properly despite normal blood capillary development.
15. Kerszberg M: Morphogen propagation and action: towards molecular models. Semin Cell Dev Biol 1999, 10:297-302. 16. Cadigan KM, Fish MP, Rulifson EJ, Nusse R: Wingless repression of Drosophila frizzled 2 expression shapes the Wingless morphogen gradient in the wing. Cell 1998, 93:767-777.
Current Opinion in Biotechnology 2003, 14:547–550
31. Swartz MA: The physiology of the lymphatic system. Adv Drug Deliv Rev 2001, 50:3-20.
www.current-opinion.com