Journal:Molecular Cell:2

Automated design of efficient and functionally diverse enzyme repertoires

Olga Khersonsky, Rosalie Lipsh, Ziv Avizemer, Yacov Ashani, Moshe Goldsmith, Haim Leader, Orly Dym, Shelly Rogotner, Devin L. Trudeau, Jaime Prilusky, Pep Amengual-Rigo, Victor Guallar, Dan S. Tawfik, and Sarel J. Fleishman ^[1]

Molecular Tour
Substantial improvements in enzyme activity demand multiple mutations at spatially proximal positions in the active site. Such mutations, however, often exhibit unpredictable epistatic (non-additive) effects on activity. We describe FuncLib - an automated method to design multipoint mutants at enzyme active sites using phylogenetic analysis and Rosetta design calculations. FuncLib was applied to two unrelated enzymes, a phosphotriesterase (PTE) and an acetyl-CoA synthetase. All designs were active, and most showed activity profiles that significantly differed from wild type and from one another. Several dozen designs with only 3-6 active-site mutations nevertheless exhibited 10-4,000-fold higher efficiencies with a range of alternative substrates, including the hydrolys is of the toxic nerve agents soman and cyclosarin and the synthesis of butyryl-CoA - activities that are hardly detectable in the wild type enzymes. FuncLib designs included epistatic active-site mutations that are unlikely to be accessible to natural and laboratory evolution; the method circumvents high-throughput screens and opens the way to design highly efficient and diverse catalytic repertoires. FuncLib is implemented as a web-server (http://funclib.weizmann.ac.il).

The focal point of our study was the phosphotriesterase (PTE) from ​Pseudomonas diminuta. PTE is a promiscuous metalloenzyme: in addition to highly efficient hydrolysis of the organophosphate pesticide paraoxon (​k_cat/K_M approximately 10​⁸ M​^-1​s​^-1​), it promiscuously hydrolyzes esters, lactones, and diverse organophosphates, including toxic nerve agents, such as VX, Russian VX, soman (GD), and cyclosarin (GF), albeit with k_cat/K_M values that are orders-of-magnitude lower than for paraoxon. Effective organophosphate detoxification, however, demands high catalytic efficiency, with ​k_cat/K_M of 10​⁷ M​^-1​min^​-1 considered a minimum for​ in vivo protection, thereby motivating several recent enzyme-engineering efforts that targeted PTE​. Furthermore, the growing threat from a new generation of nerve agents, similar in structure to VX and GF​​, emphasizes the need for broad-spectrum nerve-agent hydrolases. FuncLib’s goal is to design a small set of stable, efficient,and functionally diverse multipoint active-site mutants suitable for low-throughput experimental testing. The design strategy is general and can be applied, in principle, to any natural enzyme starting from its molecular structure and adiverse set of homologous sequences.

(PDB entry 1hzy) comprises a bimetal center, typically of Zn​²⁺ ions (brown spheres), which are liganded by highly conserved residues (orange). Water molecules are shown as red spheres. (magenta) comprise the active-site wall and are less conserved. FuncLib starts by filtering single-point mutations according to evolutionary-conservation and atomistic-stability analyses, resulting in a subset of potentially tolerated mutations:

• 106 ICHLM
• 132 FL
• 254 HGR
• 257 HWY
• 271 LIR
• 303 LT
• 306 FI
• 317 ML

Eight active-site positions were select that comprise the PTE active-site wall (first-shell) for design (see the current scene). FuncLib starts by defining a sequence space comprising active-site point mutations that are predicted to be individually tolerated. First, it retains only mutations with at least a modest probability of occurrence in the natural diversity according to a multiple-sequence alignment of homologues. Second, it eliminates point mutations that substantially destabilize the wild-type protein according to Rosetta atomistic modeling. Applied to PTE, for instance, all the essential active-site positions - that is, the positions that interact with the metal ions - were not allowed any mutations, whereas other first-shell positions were allowed even radical mutations. Filtering drastically reduced the combinatorial space of multipoint mutants at the eight active-site positions from 10​10 mutants, if all 20 amino acids were allowed at each position, to <10​^5​. From this filtered set, we chose all the multipoint mutants that comprised 3-5 mutations relative to wild-type PTE for Rosetta modeling and refinement, including backbone and sidechain minimization. Hence, FuncLib explicitly models each combination of mutations, thereby exhaustively modeling andr anking multipoint mutants at potentially epistatic active-site positions according to their predicted stability. The top-ranked designs are therefore predicted to exhibit stable and preorganized active-site pockets - a prerequisite for high catalytic efficiency​​. Surprisingly, it was found that hundreds of unique active-site designs exhibited energies that were as favorable as or better than those of wild-type PTE, suggesting that a very large space of potentially tolerated multipoint mutants at the active site was accessible by computational design. Then the designs were clustered, eliminating ones that differed by fewer than two active-site mutations from one another or from wild-type PTE and selected the top 49 designs for experimental testing.

Note that in this implementation, FuncLib does not require a model oftheenzyme-transition state complex. Instead, it computes diverse yet stable networks of interacting residues at the active-site pocket, thereby encoding different stereochemical complementarities for alternative substrates that do not need to be defined​ a priori. Therefore it was anticipated that the designs would collectively form a repertoire, from which individual designs that efficiently hydrolyzed various target substrates could be isolated. In applications that target a specific substrate, by contrast, sequence space can be further constrained by designing the enzyme in the presence of the transition-state model, and this option is enabled in the FuncLib web-server, although recommended only when the model is likely to be accurate.

Modified shape and electrostatic properties of the active-site pocket in PTE designs:

• ​
• (PDB entry 6gbj). Mutations: H254R, H257W, L303T, and M317L.
• (PDB entry 6gbk). Mutations: I106L, H254G, and M317L.
• (PDB entry 6gbl). Mutations: I106L, H254G, H257W, and L303T.
• .

PTE_27 and PTE_28 exhibit a larger active-site pocket than PTE and high catalytic efficiency against bulky V- and G-type nerve agents:

• . Wild type PTE is in green, PTE_27 in magenta, and His254 in yellow.
• . Wild type PTE is in green, PTE_28 in magenta, His254 and Leu303 are in yellow.

Next catalytic efficiency was measured in the designs that retained high phosphotriesterase activity with the toxic nerve agents, VX, Russian VX (RVX), Soman (GD). PTE_27 exhibited 66-fold increase in VX hydrolysis efficiency relative to wild-type PTE, and PTE_28 exhibited remarkable gains in efficiency of 1,550 and 3,980-fold respectively, in hydrolyzing RVX and GF. Starting from PTE_27, a second round of design was tested, this time directing FuncLib to rank point mutations of PTE_27. 14 designs were experimentally tested, finding that designs PTE_27.14 and PTE_27.16 exhibited increased activities towards GD (32-fold and 122-fold, respectively), and both designs exhibited a 3,000-fold increase in hydrolyzing GF. These designs for the highly toxic nerve agents RVX, GD, and GF, may be suitable for ​in vivo detoxification.

Molecular docking simulations were used to model S-VX, S-RVX, and GD in the active-site pockets of PTE_27, PTE_28, and PTE_27.14, respectively. The resulting models indicated that the designed active-site pockets were large enough to accommodate the bulky nerve agents and form direct contacts with them, mostly due to two large-to-small mutations, His254Gly and Leu303Thr.These direct contacts may also underlie the high enantioselectivity observed in some designs (>10​⁴ for design PTE_28). Furthermore, several improved esterases and lactonases (PTE13-15, 30-34, and 36) encoded the His254Arg mutation,which changed the sterics and electrostatics of the active-site pocket, as also reported in laboratory-evolution studies that enhanced these activities​. Therefore it could be concluded that the FuncLib mutations only affected the structure of the active-site pocket,that improved efficiency for different substrates stemmed from different types of molecular changes, and that a handful of active-site mutations was sufficient to effect orders-of-magnitude improvements in catalytic efficiency and selectivity against several substrates.

Catalytic poses of nerve agents in designed active-site pockets

S-VX:

• . The active site highly conserved residues are in orange, The carbon atoms of S-VX are in green.
• .
• .
• . Wild type PTE is in cyan, PTE_27 in magenta.
• .

S-RVX:

• . The active site highly conserved residues are in orange, The carbon atoms of S-RVX are in salmon.
• .
• .
• .

• . Wild type PTE is in cyan, PTE_28 in magenta.
• .

Soman:

• . The active site highly conserved residues are in orange, The carbon atoms of soman are in olive.
• .
• .
• . Wild type PTE is in cyan, PTE_27.14 in magenta.
• .

These three models show high geometric complementarity between the designed pockets and the respective substrates.

PDB references: Repertoires of functionally diverse enzymes through computational design at epistatic active-site positions 6gbj; 6gbk; 6gbl.

References

1. ↑ Khersonsky O, Lipsh R, Avizemer Z, Ashani Y, Goldsmith M, Leader H, Dym O, Rogotner S, Trudeau DL, Prilusky J, Amengual-Rigo P, Guallar V, Tawfik DS, Fleishman SJ. Automated Design of Efficient and Functionally Diverse Enzyme Repertoires. Mol Cell. 2018 Oct 4;72(1):178-186.e5. doi: 10.1016/j.molcel.2018.08.033. Epub, 2018 Sep 27. PMID:30270109 doi:http://dx.doi.org/10.1016/j.molcel.2018.08.033