Theoretical esterases

From Proteopedia

Background

De Novo protein design and synthesis has been a goal of biochemistry for decades and has wide sweeping implications in many fields. The advent of computers has given rise to computer programs such as RosettaMatch which are capable of identifying sequences compatible with given protein skeletons. In August of 2012, the Richter Lab and collaborators published an attempted to design an esterase from a theoretically functional protein backbone. This article is a summary of their contribution to the field of de Novo esterase design.

As described in Richter et al, esterases use serines and cysteines as nucleophilic donors in the first step of ester hydrolysis. Either the oxygen in the backbone or residues such as asparagine or glutamine deprotonate the nucleophile. Hydrogen bond donors such as the NH backbone or residues in the active site create an “oxyanion hole” to stabilize any oxyanion formed during catalysis ^[1].

Relevance

The general term esterases are used to categorize enzymes that split esters during hydrolysis into an acid and an alcohol. Esterases are a wide range of enzymes- from differences in substrate specificity, structure, to its biological function in downstream effects. They play an extremely important role in diseases such as cancer and diabetes and for this reason they are widely studied for their possibility in mimicking for drug development. Richter et al computationally studied the mechanism of esterases.

Theoretical protein design, while still a relatively unused field, has been used with success before. The video game FoldIt allows players to compete by altering existing protein structures and proposing new structures with points assigned for lowering the energy of folding associated with the molecule. Khatib et. Al were able to use the crowdsourcing capabilities of FoldIt to solve the structure of M-PMV retroviral protease. Much like the goal of Richter et. Al, Eiben et. Al were able to use FoldIt to increase the activity of a Diels-Alderase.

Design

Two hundred fourteen protein scaffolds collected from RosettaMatch that contained a Cys-His diad to function as the active site for a cysteine esterase were originally considered for synthesis. These 214 possibilities were narrowed to 207 based on the presence of a central Cys residue, a His capable of proton shuffling, and possible oxyanion stabilizers, or oxyanion holes. Of the 207 designs, 55 were synthesized and broken into 3 groups based on how the oxyanion intermediate is stabilized: 31 Theozyme I (backbone NHs), 12 Theozyme II (sidechain H-bonds), and 12 Theozyme III (explicit water molecules). Purified soluble protein could only be obtained for 19 Theozyme I, 5 Theozyme II, and 8 Theozyme III. Of these 32 purified designs, only 4 Theozyme I designs were active, alluding to the conclusion that backbone NHs are better stabilizers of the oxyanion intermediate than side chains or water molecules.

The 4 active proteins (ECH13, ECH19, ECH14, and FR29) were tested by knocking out either the catalytic Cys or His, rendering the protein either inactive or greatly decreasing catalysis, showing that the activity demonstrated was a result of the esterase tested. The crystal structures of all 4 proteins were then solved and processed, showing protein backbones similar to the design, but a unique problem in each active site. His100 in ECH13 rotated in a way not predicted in the design, hydrogen bonding with Asp10 instead of Cys45. ECH19 crystalizes in the open conformation rather than the catalytic closed conformation, with His226 interacting with Tyr250 and the backbone of Phe221 rather than Cys161. The catalytic dyad of ECH14 does not form, with the loop containing the catalytic Cys132 orienting away from the catalytic His104. The structure of FR29 folded as to be more similar to the open, unliganded conformation than the design model which was based on a bound ligand. This unpredicted structure causes the catalytic dyad not to form by shifting the helix-turn-helix motif containing His125 out and away from catalytic Cys9.

Implications

None of the proposed de novo structures were able to create a functional, efficient esterase as designed. While it is worth being appreciative of the fact that the Richter lab and collaborators were able to get as close as they were, both the technology of protein structure predication and our own knowledge of protein folding mechanisms must increase before we can reliably design proteins from the bottom up ^[2][3].

Structural highlights

A conserved element in esterases and ester hydrolysis are catalytic dyads and triads. Richter et al studied a class of esterases with a catalytic Cys-His in the active site, deemed a “catalytic dyad.” Figure 6 in Richter et al shows crystal structures (green) compared to the design model (purple) of the four active designs in (a) ECH 13, (b) ECH 19, (c) ECH 14, (d) FR 29.

In (a), the crystal structure, His100 interacts with C45. In the design model, His100 makes a hydrogen bond with Asp10. While the RMSD between the crystal structure and design model in (b) decreases, the catalytic His226 in the dyad does not interact with Cys161, but it does interact with Tyr250 and the oxygen in F221. In (c) the dyad was not formed. Cys132 loop with residues 127-140 moves up and away fom active site and His104 reorients and therefore rendering this design model inactive. In (d), resides 106-132 (containing the catalytis His125) moves outward towards Cys9 and creats a large shift in the backbond of His125 and Cys9, the catalytic dyad.

References

1. ↑ Richter F, Blomberg R, Khare SD, Kiss G, Kuzin AP, Smith AJ, Gallaher J, Pianowski Z, Helgeson RC, Grjasnow A, Xiao R, Seetharaman J, Su M, Vorobiev S, Lew S, Forouhar F, Kornhaber GJ, Hunt JF, Montelione GT, Tong L, Houk KN, Hilvert D, Baker D. Computational design of catalytic dyads and oxyanion holes for ester hydrolysis. J Am Chem Soc. 2012 Oct 3;134(39):16197-206. doi: 10.1021/ja3037367. Epub 2012, Sep 21. PMID:22871159 doi:http://dx.doi.org/10.1021/ja3037367
2. ↑ Eiben CB, Siegel JB, Bale JB, Cooper S, Khatib F, Shen BW, Players F, Stoddard BL, Popovic Z, Baker D. Increased Diels-Alderase activity through backbone remodeling guided by Foldit players. Nat Biotechnol. 2012 Jan 22;30(2):190-2. doi: 10.1038/nbt.2109. PMID:22267011 doi:http://dx.doi.org/10.1038/nbt.2109
3. ↑ Khatib F, Dimaio F, Cooper S, Kazmierczyk M, Gilski M, Krzywda S, Zabranska H, Pichova I, Thompson J, Popovic Z, Jaskolski M, Baker D. Crystal structure of a monomeric retroviral protease solved by protein folding game players. Nat Struct Mol Biol. 2011 Sep 18. doi: 10.1038/nsmb.2119. PMID:21926992 doi:10.1038/nsmb.2119