Synthetic RNA Networks for Sophisticated In Vivo Computation

Alexander A. Green1,2*, Jongmin Kim2, Mario Teichmann2, Pamela A. Silver2,3, James J. Collins2,4,5, Peng Yin2,3

1Center for Molecular Design and Biomimetics, The Biodesign Institute, Department of Chemistry and Biochemistry, Arizona State University, AZ 85287, USA. 2Wyss Institute for Biologically Inspired Engineering, Harvard University, Boston, MA 02115, USA.

3Department of Systems Biology, Harvard Medical School, Boston, MA 02115, USA.

4Institute for Medical Engineering and Science, Department of Biological Engineering, and Synthetic Biology Center, Massachusetts Institute of Technology, Cambridge, MA

02139, USA. 5Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA. *E-mail:

alexgreen@asu.edu.
Synthetic gene networks have generally employed well-characterized natural parts rewired to provide new functionality; however, the limited number and crosstalk between such parts has hindered the construction of more complex networks that can operate robustly in living cells. We have recently developed a new type of synthetic biological device called the toehold switch that addresses these limitations by providing dozens of orthogonal parts each with a wide dynamic range of gene expression. These validated new parts provide us with the opportunity to construct much more complex in vivo synthetic networks. Here, we present a new molecular programming strategy that employs long information-processing strands of RNA to evaluate the status of networks of short de-novo-designed RNAs being transcribed within the cell. These synthetic RNA computational networks, or ribocomputers, are enabled by sets of highly orthogonal toehold switches and self-assembling RNA complexes that act as conditional activators of gene expression. Ribocomputers can compute expressions combining AND, OR, and NOT operations for universal combinatorial logic in E. coli, and they exploit in silico RNA sequence design to greatly reduce the time and effort required to construct new functional systems. These systems can efficiently evaluate multi-input AND and OR logic operations using a single computational layer and routinely provide ON/OFF ratios greater than 20. Furthermore, multiple ribocomputers can be active within a cell at the same time providing multi-output capabilities. All these features make ribocomputers an important new addition to the synthetic biology toolkit and highlight the strengths of programmable RNA-based regulation for realizing sophisticated in vivo logic.