Enzymes R Us: BIOMOD at Rutgers University

Abstract

Multiple enzymes constituting a heterologously expressed metabolic pathway (“assembly line”) in microbial cells (“factories”), are now routinely used to biosynthesize various chemicals and drugs. The ability to control the nanoscale topologies of these molecular assembly lines will allow pathway optimization. However, in contrast with DNA, protein assembly design rules remain elusive and/or non-generalizable. Here, we demonstrate the capacity to build spatially controlled protein assemblies by creating supramolecular complexes of green and red fluorescent proteins (FPs) using a combined computational-experimental approach. Different types of assemblies were generated: terminating ring structures, self-propagating branched fractal formations, and irreversibly-bound scaffolding units mediated by an unnatural amino acid. Our method allows for both specificity and modularity, which enables us to substitute the FPs with enzymes of a biochemical pathway in the molecular assembly line. Furthermore, protein-protein interactions responsible for assembly can be controlled by phosphorylation to allow temporal control and responsiveness to external stimuli.

Introduction

Designing supramolecular assemblies with well-defined geomoetries has the potential to optimize many biological processes. By designing protein complexes to bring enzymes of a pathway together into one complex, we are able to physically channel intermediates by bringing them closer to the next enzyme to be used as substrates in the next step of the biochemical pathway. [1] Optimization of metabolic pathways will have enormous implications, most notably increasing efficiency for organic synthesis of drugs, synthesizing microorganisms for bioremediation, or elevating production of biofuels. [2]

Not only does this process decrease transit time of intermediates, but it also minimizes metabolite accumulation, which prevents diffusion of intermediates, protects unstable intermediates from solvent, restores equilibrium of metabolite concentrations, and prevents escape of potentially toxic intermediates. [3] Co-localizing pathway enzymes also increases efficiency by increasing reaction rates and reducing the necessary concentrations of enzymes.

Supramolecular assemblies consisting of macromolecules have been achieved using DNA and RNA, tiling, and origami; using proteins as building blocks instead of nucleic acids would allow us to organize enzymes of a pathway reaction in vivo. Thus far, supramolecular assemblies created through protein design have been limited to oligomeric proteins. [4] Since we want to create a modular system that will incorporate any enzyme on demand, we propose a novel approach that uses fused peptide-binding domains.

Our system

Peptide-binding domains are used in signal transduction to recruit pathway enzymes. Substrates in natural systems bind and disassociate with their binding domains to regulate on/off cell signaling. However, we aim to covalently tether pathway enzymes to the supramolecular complex and “staple” the enzymes together with a bioconjugation reaction that yields an irreversible bond.

The binding domains we have selected to use are the Src homology 2 (SH2) domain, PSD-95 / Discs-large / ZO-1 (PDZ) domains and Src homology 3 (SH3) domains that display high binding specificity and affinity to their cognate peptide ligands. [5,6] Our project goal was to create macromolecular assemblies by designing PDZ and SH3 domains that will bind irreversibly to their ligands through a thiol-bromoethyl SN2 reaction between an unnatural amino acid (UAA) and a cysteine. Covalent bonds between a cysteine and this particular UAA have been recently demonstrated to be resistant to denaturing conditions and reducing agents. [7,8] The bromoethyl group of (O-(2-bromoethyl)-tyrosine) (UAA) reacts with the thiol group of cysteine, so we can expect that inserting a cysteine residue on the PDZ domain and then placing the UAA on the corresponding ligand will result in a proximity-mediated bioconjugation reaction, yielding a stable thioether bond connecting the binding domains to their peptides, which will ultimately be incorporated into the targeted pathway enzymes. We also incorporated SH2 domains as these bind to phosphopeptides and may potentially allow toggle switching of the assembly using phosphorylation/de-phosphorylation

Prototyping Molecular Assemblies using Monomeric Florescent Proteins

To test the self-assembly capacities of the designed PDZ and SH3 domains, we plan to insert PDZ and SH3 domains into the loops of monomeric florescent proteins. We will insert PDZ and SH3 into certain loops of red florescent protein mCherry (RFP) and superfolder green florescent protein (sfGFP), and use loop-oriented domain insertion modeling algorithms to sample and determine linker sequences. The respective binding peptides of PDZ and SH3 would become C-termini appendages to RFP and sfGFP. RFP and sfGFP were chosen due to their high durability when domains are inserted into the protein. [9,10]

One type of assembly is a RFP-sfGFP linear chain that may form rings. This assembly involves SH2 inserted into sfGFP, and SH3 inserted into RFP, with SH3 peptides appended to the C terminus of sfGFP and SH2 peptides appended to the C terminus of RFP. The result is a linear chain of RFP and sfGFP units, which, depending on the rigidity of the connector peptides, may circularize by terminating onto itself. A second type of assembly is the fan model. In the fan model, an additional RFP is utilized. This RFP has a PDZ domain inserted and a SH2 binding peptide at its C terminus. The sfGFP with the SH2 insertion will have an additional PDZ binding peptide appended to the N terminus. The result is a fan-shaped assembly that propagates indefinitely and forms fractal patterns.

References

[1] Whitaker WR, Dueber JE. 2011. Metabolic pathway flux enhancement by synthetic protein scaffolding. Methods Enzymol 497: 447-468.

[2] Conrado RJ, Varner JD, DeLisa MP. 2008. Engineering the spatial organization of metabolic enzymes: mimicking nature's synergy. Curr Opin Biotechnol 19: 492-499.

[3] Dueber, JE, Wu GC, Malmirchegini GR, Moon TS, Petzold CJ, Ullal AV, Prather KLJ, Keasling JD. 2009. Synthetic protein scaffolds provide modular control over metabolic flux. Nat Biotechnol 27: 753-759.

[4] King NP, Bale JB, Sheffler W, McNamara DE, Gonen S, Gonen T, Baker D. 2014. Accurate design of co-assembling multi-component protein nanomaterials.Nature, 510: 103-108.

[5] Ernst A, Appleton BA, Ivarsson Y, Zhang Y, Gfeller D, Wiesmann C, Sidhu SS. 2014. A structural portrait of the PDZ domain family. J Mol Biol 426: 3509-3519.

[6] Cesareni C, Panni S, Nardelli G, Castagnoli L. 2002. Can we infer peptide recognition specificity mediated by SH3 domains? FEBS Lett 513: 38-44.

[7] Lu Y, Huang F, Wang J, Xia J. 2014. Affinity-guided covalent conjugation reactions based on PDZ–peptide and SH3–peptide interactions. Bioconjugate Chem 25: 989-999.

[8] Bionda N, Cryan AL, Fasan R. 2014. Bioinspired strategy for the ribosomal synthesis of thioether-bridged macrocyclic peptides in bacteria. ACS Chem Biol 9: 2008-2013.

[9] Abedi MR, Caponigro G, Kamb A. 1998. Green fluorescent protein as a scaffold for intracellular presentation of peptides. Nucleic Acid Res 26: 623-630.

[10] Li Y, Sierra AM, Ai HW, Campbell RE. 2007. Identification of sites within a monomeric red fluorescent protein that tolerate peptide insertion and testing of corresponding circular permutations. Photochem Photobiol 84: 111-119.

[10] Tong L, Warren TC, King J, Betageri R, Rose J, Jakes S. 1996. Crystal structures of the human p56lckSH2 domain in complex with two short phosphotyrosyl peptides at 1.0 Å and 1.8 Å resolution. J Mol Biol 256: 601-610.