Synthetic Morphogenesis

Abstract

Throughout biology, function is intimately linked with form. Across scales ranging from subcellular to multiorganismal, the identity and organization of a biological structure's subunits dictate its properties. The field of molecular morphogenesis has traditionally been concerned with describing these links, decoding the molecular mechanisms that give rise to the shape and structure of cells, tissues, organs, and organisms. Recent advances in synthetic biology promise unprecedented control over these molecular mechanisms; this opens the path to not just probing morphogenesis but directing it. This review explores several frontiers in the nascent field of synthetic morphogenesis, including programmable tissues and organs, synthetic biomaterials and programmable matter, and engineering complex morphogenic systems de novo. We will discuss each frontier's objectives, current approaches, constraints and challenges, and future potential.

Figure 1.

Synthetic gene networks drive cell-fate decisions using cell state sensors and extracellular signals. Pluripotent stem cells are directed to differentiate into β cells by sequential application of growth factors (Schiesser et al. 2014); the same differentiation program might be performed by a gene network that uses sequential expression of master cell-fate regulators such as Gata6 and Pdx1. Progression through the program could be controlled by endogenous sensors of cell state. This multistep sense-and-differentiate program could be part of a larger gene network, such as one that maintains tissue homeostasis (Miller et al. 2012). A gene network could sense the number of programmed stem cells, and terminally differentiated cells using diffusible signals then proliferate or differentiate to maintain a continuous supply of both.

Figure 2.

Engineered intercellular communication can coordinate spatial changes in gene expression. Communication channels are frequently based on diffusible signals such as acyl-homoserine lactone (AHL) (left) (Basu et al. 2005). The receiver gene network responds to intermediate concentrations of AHL, mimicking the “French flag” model of morphogen patterning (Wolpert 1969). Arranging senders in various starting configurations produces patterns similar to those found in natural development. Communication can also be based on juxtacrine signaling (right), in which a cell presenting a ligand on its surface activates signal transduction in adjacent cells (Sprinzak et al. 2010). A feedback loop between signal activation and ligand production leads to signal propagation (Matsuda et al. 2012). GFP, Green fluorescent protein.

Figure 3.

Chen et al. (2014) created a tunable biomaterial using curli fibers, a constituent of Escherichia coli biofilms. Curli fibers are made up of CsgA subunits, which self-assemble into higher-order structures. The investigators modified CsgA with a His tag that confers metal-binding capability; when they alternated expression of modified CsgA with unmodified CsgA (based on two different small-molecule inducers), the resulting curli fibers had regions that bound gold nanoparticles alternating with regions that did not, as shown with scanning electron microscopy. Scale bar, 100 nm. AHL, Acyl-homoserine lactone. (Image courtesy of Allen Chen and Tim Lu, both at the Massachusetts Institute of Technology.)

Figure 4.

Proposed synthetic morphogenic systems. Patterned biofilm formation (left) may be achieved by connecting a synthetic optogenetic sensor, such as CcaS/CcaR (Schmidl et al. 2014), to a master regulator of biofilm production, such as csgD (Brombacher et al. 2003). A model mammalian morphogenic system (right) may offer a richer palette of morphogenic behaviors. For example, hepatocyte growth factor (HGF) induces tubulogenesis in Madin–Darby canine kidney (MDCK) cells grown in three-dimensional culture (Zegers 2014). The HGF receptor, MET, activates a number of downstream signaling pathways; controlling one via synthetic receptor, such as STAT3 fused to an estrogen receptor (ER) (STAT3-ER) (Matsuda et al. 1999), may allow tunable control of some of the morphogenic modules that lead to tubulogenesis.