Michael Elowitz

Position Titles

Assistant Adjunct Professor


Programming cellular behaviors with biological circuits
Living cells use circuits of interacting genes and proteins to sense and respond to signals, communicate, remember information, and develop into multicellular organisms. These circuits often use non-intuitive designs. Making sense of those designs is essential for understanding, predicting, and controlling natural cellular behaviors, and for designing and building synthetic circuits that can program biomedically useful cellular behaviors.

Our strategy: naturally inspired synthetic circuit design. Our lab combines two approaches to the design and analysis of biological circuits: First, we create and analyze fully synthetic molecular circuits that provide new cellular capabilities. Second, we reconstitute core pathways in minimal cell culture systems and quantitatively analyze their behavior at the single cell level. A key premise of the lab is that these two approaches are synergistic: Natural pathways inform the design of their synthetic counterparts by providing elegant and unexpected solutions to synthetic design challenges. Conversely, for synthetic circuits to interact predictably with natural pathways we need to know how those natural pathways sense, process, and respond to their inputs. By combining these approaches, we aim to establish a foundation for programmable cell-based therapeutics and develop conceptual frameworks for understanding biological systems.

Focus: Circuits that compute, communicate, and remember. We currently focus on capabilities required for natural mammalian development and synthetic therapeutic and developmental circuits. These include intercellular communication to allow coordination among cells; computation to allow signal processing within the cell; and memory to store information. We combine synthetic biology, mathematical modeling, and dynamic single cell analysis. We also develop components and methods that broadly enable all of these areas. Recent projects include the following:

Synthetic computational circuits: We developed a framework for designing protein circuits that can sense, compute, and respond to cellular states. These circuits, termed CHOMP (Circuits of Hacked Orthogonal Modular Proteases) use engineered proteases that can regulate one another to perform binary logic, implement regulatory cascades, and carry out analog signal processing. We demonstrated a rationally designed, multi-protein circuit that selectively induces cell death in response to activation of the Ras oncogene, a major cancer driver that has been difficult to target with drugs. This and other related projects thus provide a platform for creating circuits that can act as smart therapeutics within cells (see X. Gao et al, Science 2018).

Communication: A handful of core molecular pathways, such as Notch, Bone Morphogenetic Protein (BMP), and Sonic Hedgehog (SHH), enable cell-cell communication across diverse developmental contexts. We still do not understand what specific signal processing capabilities each pathway provides, why different pathways use distinct molecular architectures, and how to predictively control the activity of these pathways in specific cell types. The lab currently focuses on a prevalent, mysterious, feature of many pathways: their use of many ligands that interact promiscuously (in a many-to-many fashion) with many corresponding receptors. We discovered that this seemingly ‘messy’ architecture functions as a powerful computational system that extracts information encoded across multiple ligands (Antebi et al, Cell, 2017), and could, counter-intuitively, allow precise addressing of signals to specific target cells (Su et al, Biorxiv 2020). We are also developing a framework for predictive understanding of how these promiscuous interactions lead to complex signal integration behaviors (Klumpe et al, Biorxiv 2020).

Memory: During development, cells proliferate and differentiate through sequences of states, driven by internal programs and external signals. The histories of individual cells are critical for understanding multicellular development, but difficult to follow directly in all but a few organisms. In collaboration with Long Cai’s lab, we developed a synthetic genetic recording system termed MEMOIR(Memory by Engineered Mutagenesis with Optical In-situ Readout). MEMOIR enables each cell to record its own lineage and molecular event history within its genome in a format that can be read out in situ by imaging. With Carlos Lois and Long Cai, we are now developing new generations of MEMOIR technology, based on 3-state memory elements as well as the ability to generate and image edits made by CRISPR base editors. These approaches enable the design of self-recording animals, such as the MEMOIR fly, Drosophila memoiphila, and provide new ways to address fundamental questions in development and disease. Meanwhile, a distinct and complementary effort in the lab aims to understand principles of epigenetic memory systems.