Barry L. Stoddard

De Novo Computational Design of Retro-Aldol Enzymes

The creation of enzymes capable of catalyzing any desired chemical reaction is a grand challenge for computational protein design. Using new algorithms that rely on hashing techniques to construct active sites for multistep reactions, we designed retro-aldolases that use four different catalytic motifs to catalyze the breaking of a carbon-carbon bond in a nonnatural substrate. Of the 72 designs that were experimentally characterized, 32, spanning a range of protein folds, had detectable retro-aldolase activity. Designs that used an explicit water molecule to mediate proton shuffling were significantly more successful, with rate accelerations of up to four orders of magnitude and multiple turnovers, than those involving charged side-chain networks. The atomic accuracy of the design process was confirmed by the x-ray crystal structure of active designs embedded in two protein scaffolds, both of which were nearly superimposable on the design model.

Computational design of an enzyme catalyst for a stereoselective bimolecular Diels-Alder reaction.

Biocatalytic Boost Enzymes tend to direct reactions toward specific products much more selectively than synthetic catalysts. Unfortunately, this selectivity has evolved for cellular purposes and may not promote the sorts of reactions chemists are seeking to enhance (see the Perspective by Lutz). Siegel et al. (p. 309) now describe the design of enzymes that catalyze the bimolecular Diels-Alder reaction, a carbon-carbon bond formation reaction that is central to organic synthesis but unknown in natural metabolism. The enzymes display high stereoselectivity and substrate specificity, and an x-ray structure of the most active enzyme confirms that the structure matches the design. Savile et al. (p. 305, published online 17 June) applied a directed evolution approach to modify an existing transaminase enzyme so that it recognized a complex ketone in place of its smaller native substrate, and could tolerate the high temperature and organic cosolvent necessary to dissolve this ketone. This biocatalytic reaction improved the production efficiency of a drug that treats diabetes. Synthetic enzymes catalyze a carbon-carbon bond-forming reaction with high stereoselectivity and substrate specificity. The Diels-Alder reaction is a cornerstone in organic synthesis, forming two carbon-carbon bonds and up to four new stereogenic centers in one step. No naturally occurring enzymes have been shown to catalyze bimolecular Diels-Alder reactions. We describe the de novo computational design and experimental characterization of enzymes catalyzing a bimolecular Diels-Alder reaction with high stereoselectivity and substrate specificity. X-ray crystallography confirms that the structure matches the design for the most active of the enzymes, and binding site substitutions reprogram the substrate specificity. Designed stereoselective catalysts for carbon-carbon bond-forming reactions should be broadly useful in synthetic chemistry.

Homing endonuclease structure and function

Homing endonucleases are encoded by open reading frames that are embedded within group I, group II and archael introns, as well as inteins (intervening sequences that are spliced and excised post-translationally). These enzymes initiate transfer of those elements (and themselves) by generating strand breaks in cognate alleles that lack the intervening sequence, as well as in additional ectopic sites that broaden the range of intron and intein mobility. Homing endonucleases can be divided into several unique families that are remarkable in several respects: they display extremely high DNA-binding specificities which arise from long DNA target sites (14-40 bp), they are tolerant of a variety of sequence variations in these sites, and they display disparate DNA cleavage mechanisms. A significant number of homing endonucleases also act as maturases (highly specific cofactors for the RNA splicing reactions of their cognate introns). Of the known homing group I endonuclease families, two (HNH and His-Cys box enzymes) appear to be diverged from a common ancestral nuclease. While crystal structures of several representatives of the LAGLIDADG endonuclease family have been determined, only structures of single members of the HNH (I-HmuI), His-Cys box (I-PpoI) and GIY-YIG (I-TevI) families have been elucidated. These studies provide an important source of information for structure-function relationships in those families, and are the centerpiece of this review. Finally, homing endonucleases are significant targets for redesign and selection experiments, in hopes of generating novel DNA binding and cutting reagents for a variety of genomic applications.

The Crystal Structure of TAL Effector PthXo1 Bound to Its DNA Target.

Wrapped DNA TAL effectors are proteins that bacterial pathogens inject into plant cells that bind to host DNA to activate expression of plant genes. The DNA-binding domain of TAL proteins is composed of tandem repeats within which a repeat-variable diresidue sequence confers nucleotide specificity. Deng et al. (p. 720, published online 5 January) report the structure of the TAL effector dHax3, containing 11.5 repeats, in DNA-free and DNA-bound states, and Mak et al. (p. 716, published online 5 January) report the structure of the PthXo1 TAL effector, containing 22 repeats, bound to its DNA target. Together, the structures reveal the conformational changes involved in DNA binding and provide the structural basis of DNA recognition. Structures show how a virulence factor in a plant pathogen recognizes and binds to host DNA. DNA recognition by TAL effectors is mediated by tandem repeats, each 33 to 35 residues in length, that specify nucleotides via unique repeat-variable diresidues (RVDs). The crystal structure of PthXo1 bound to its DNA target was determined by high-throughput computational structure prediction and validated by heavy-atom derivatization. Each repeat forms a left-handed, two-helix bundle that presents an RVD-containing loop to the DNA. The repeats self-associate to form a right-handed superhelix wrapped around the DNA major groove. The first RVD residue forms a stabilizing contact with the protein backbone, while the second makes a base-specific contact to the DNA sense strand. Two degenerate amino-terminal repeats also interact with the DNA. Containing several RVDs and noncanonical associations, the structure illustrates the basis of TAL effector–DNA recognition.

Capturing the Structure of a Catalytic RNA Intermediate: The Hammerhead Ribozyme

The crystal structure of an unmodified hammerhead RNA in the absence of divalent metal ions has been solved, and it was shown that this ribozyme can cleave itself in the crystal when divalent metal ions are added. This biologically active RNA fold is the same as that found previously for two modified hammerhead ribozymes. Addition of divalent cations at low pH makes it possible to capture the uncleaved RNA in metal-bound form. A conformational intermediate, having an additional Mg(II) bound to the cleavage-site phosphate, was captured by freeze-trapping the RNA at an active pH prior to cleavage. The most significant conformational changes were limited to the active site of the ribozyme, and the changed conformation requires only small additional movements to reach a proposed transition-state.

Computational redesign of endonuclease DNA binding and cleavage specificity.

The reprogramming of DNA-binding specificity is an important challenge for computational protein design that tests current understanding of protein–DNA recognition, and has considerable practical relevance for biotechnology and medicine. Here we describe the computational redesign of the cleavage specificity of the intron-encoded homing endonuclease I-MsoI using a physically realistic atomic-level forcefield. Using an in silico screen, we identified single base-pair substitutions predicted to disrupt binding by the wild-type enzyme, and then optimized the identities and conformations of clusters of amino acids around each of these unfavourable substitutions using Monte Carlo sampling. A redesigned enzyme that was predicted to display altered target site specificity, while maintaining wild-type binding affinity, was experimentally characterized. The redesigned enzyme binds and cleaves the redesigned recognition site ∼10,000 times more effectively than does the wild-type enzyme, with a level of target discrimination comparable to the original endonuclease. Determination of the structure of the redesigned nuclease-recognition site complex by X-ray crystallography confirms the accuracy of the computationally predicted interface. These results suggest that computational protein design methods can have an important role in the creation of novel highly specific endonucleases for gene therapy and other applications.

Design, Activity, and Structure of a Highly Specific Artificial Endonuclease

We have generated an artificial highly specific endonuclease by fusing domains of homing endonucleases I-DmoI and I-CreI and creating a new 1400 A(2) protein interface between these domains. Protein engineering was accomplished by combining computational redesign and an in vivo protein-folding screen. The resulting enzyme, E-DreI (Engineered I-DmoI/I-CreI), binds a long chimeric DNA target site with nanomolar affinity, cleaving it precisely at a rate equivalent to its natural parents. The structure of an E-DreI/DNA complex demonstrates the accuracy of the protein interface redesign algorithm and reveals how catalytic function is maintained during the creation of the new endonuclease. These results indicate that it may be possible to generate novel highly specific DNA binding proteins from homing endonucleases.

Homing Endonucleases: From Microbial Genetic Invaders to Reagents for Targeted DNA Modification

Homing endonucleases are microbial DNA-cleaving enzymes that mobilize their own reading frames by generating double strand breaks at specific genomic invasion sites. These proteins display an economy of size, and yet recognize long DNA sequences (typically 20 to 30 base pairs). They exhibit a wide range of fidelity at individual nucleotide positions in a manner that is strongly influenced by host constraints on the coding sequence of the targeted gene. The activity of these proteins leads to site-specific recombination events that can result in the insertion, deletion, mutation, or correction of DNA sequences. Over the past fifteen years, the crystal structures of representatives from several homing endonuclease families have been solved, and methods have been described to create variants of these enzymes that cleave novel DNA targets. Engineered homing endonucleases proteins are now being used to generate targeted genomic modifications for a variety of biotech and medical applications.

DNA binding and cleavage by the nuclear intron-encoded homing endonuclease I-PpoI

Homing endonucleases are a diverse collection of proteins that are encoded by genes with mobile, self-splicing introns. They have also been identified in self-splicing inteins (protein introns). These enzymes promote the movement of the DNA sequences that encode them from one chromosome location to another; they do this by making a site-specific double-strand break at a target site in an allele that lacks the corresponding mobile intron. The target sites recognized by these small endonucleases are generally long (14–44 base pairs). Four families of homing endonucleases have been identified, including the LAGLIDADG, the His–Cys box, the GIY–YIG and the H–N–H endonucleases. The first identified His–Cys box homing endonuclease was I-PpoI from the slime mould Physarum polycephalum,. Its gene resides in one of only a few nuclear introns known to exhibit genetic mobility. Here we report the structure of the I-PpoI homing endonuclease bound to homing-site DNA determined to 1.8 Å resolution. I-PpoI displays an elongated fold of dimensions 25 × 35 × 80 Å, with mixed α/β topology. Each I-PpoI monomer contains three antiparallel β-sheets flanked by two long α-helices and a long carboxy-terminal tail, and is stabilized by two bound zinc ions 15 Å apart. The enzyme possesses a new zinc-bound fold and endonuclease active site. The structure has been determined in both uncleaved substrate and cleaved product complexes.

Computational design of ligand-binding proteins with high affinity and selectivity.

The ability to design proteins with high affinity and selectivity for any given small molecule is a rigorous test of our understanding of the physiochemical principles that govern molecular recognition. Attempts to rationally design ligand-binding proteins have met with little success, however, and the computational design of protein–small-molecule interfaces remains an unsolved problem. Current approaches for designing ligand-binding proteins for medical and biotechnological uses rely on raising antibodies against a target antigen in immunized animals and/or performing laboratory-directed evolution of proteins with an existing low affinity for the desired ligand, neither of which allows complete control over the interactions involved in binding. Here we describe a general computational method for designing pre-organized and shape complementary small-molecule-binding sites, and use it to generate protein binders to the steroid digoxigenin (DIG). Of seventeen experimentally characterized designs, two bind DIG; the model of the higher affinity binder has the most energetically favourable and pre-organized interface in the design set. A comprehensive binding-fitness landscape of this design, generated by library selections and deep sequencing, was used to optimize its binding affinity to a picomolar level, and X-ray co-crystal structures of two variants show atomic-level agreement with the corresponding computational models. The optimized binder is selective for DIG over the related steroids digitoxigenin, progesterone and β-oestradiol, and this steroid binding preference can be reprogrammed by manipulation of explicitly designed hydrogen-bonding interactions. The computational design method presented here should enable the development of a new generation of biosensors, therapeutics and diagnostics.