Reactibodies Generated by Kinetic Selection Couple Chemical Reactivity with Favorable Protein Dynamics

Reactibodies generated by kinetic selection couple chemical reactivity with favorable protein dynamics

Ivan Smirnova,b,1, Eugénie Carlettic,1, Inna Kurkovaa,1, Florian Nachond, Yvain Nicoletc, Vladimir A. Mitkeviche, Hélène Débatf,g, Bérangère Avalleh, Alexey A. Belogurov, Jr.a,i, Nikita Kuznetsovj, Andrey Reshetnyakk, Patrick Massond, Alexander G. Tonevitskyl, Natalia Ponomarenkoa, Alexander A. Makarove, Alain Fribouleth, Alfonso Tramontanom, and Alexander Gabibova,b,i,2

aInstitute of Bioorganic Chemistry Russian Academy of Sciences, Moscow 117997, Russia; bChemistry Department, Moscow State University, Moscow 119991, Russia; cInstitut de Biologie Structurale (Commissariat à l’Energie Atomique-Centre National de la Recherche Scientifique - Université Joseph Fourier), 38027 Grenoble, France; dDepartement de Toxicologie, Institut de Recherche Biomédicale des Armées, 38702 La Tronche, France; eEngelhardt Institute of Molecular Biology, Russian Academy of Sciences, Moscow 119991, Russia; fUniversité Versailles Saint-Quentin, F-78035 Versailles, France; gInstitut de Génétique et Microbiologie, Unité Mixte de Recherche 8621 Centre National de la Recherche Scientifique, Université Paris-Sud, F-91405 Orsay, France; hUniversité de Technologie de Compiègne, Unité Mixte de Recherche 6022 Centre National de la Recherche Scientifique, 60205 Compiègne, France; iInstitute of Gene Biology, Moscow 117334, Russia; jInstitute of Chemical Biology and Fundamental Medicine, Siberian Branch of the Russian Academy of Sciences, Novosibirsk 630090, Russia; kDepartment of Pharmacology, Yale University School of Medicine, Yale University, New Haven, CT 06520-8066; lInstitute of General Pathology and Pathophysiology, Russian Academy of Medical Science, 125315 Moscow, Russia; and mSchool of Medicine, University of California, Davis, CA 95616

Edited* by Michael Sela, Weizmann Institute of Science, Rehovot, Israel, and approved August 1, 2011 (received for review May 30, 2011)

Igs offer a versatile template for combinatorial and rational design munization display among the most efficient de novo generated approaches to the de novo creation of catalytically active proteins. protein catalytic activity yet reported, and structural studies of these We have used a covalent capture selection strategy to identify antibodies have shown how their binding mechanism allows for biocatalysts from within a human semisynthetic antibody variable discontinuous evolution of a promiscuous catalytic site (9, 10). fragment library that uses a nucleophilic mechanism. Specific Selection by reversible covalent binding is compatible with fi phosphonylation at a single tyrosine within the variable light-chain conventional af nity maturation, where the protein environment framework was confirmed in a recombinant IgG construct. High- provides thermodynamic stabilization of the covalent adduct. Al-

resolution crystallographic structures of unmodified and phospho- ternatively, irreversible covalent binding allows for kinetic selec- IMMUNOLOGY nylated Fabs display a 15-Å-deep two-chamber cavity at the inter- tion, favoring sites that accelerate bond formation by transition state stabilization. In a classical example, serine hydrolases catalyze face of variable light (V ) and variable heavy (V ) fragments having L H phosphorylation of their active site as a result of the specifically a nucleophilic tyrosine at the base of the site. The depth and struc- enhanced reactivity of the nucleophilic serine-195 (chymotrypsin- ture of the pocket are atypical of antibodies in general but can be ogen numbering system) (11). We have sought to use such en- compared qualitatively with the catalytic site of cholinesterases. A hanced reactivity for the generation of a reactibody. It is, thus, structurally disordered heavy chain complementary determining evident that innate reactivity can be deployed to capture more region 3 loop, constituting a wall of the cleft, is stabilized after highly evolved biocatalysts that conserve an essential residue for covalent modification by hydrogen bonding to the phosphonate covalent catalysis (12). Mechanism-based kinetic selection may, tropinol moiety. These features and presteady state kinetics analysis thus, be differentiated from suicide substrate selection or similar indicate that an induced fit mechanism operates in this reaction. strategies that have successfully used covalent capture (13). We Mutations of residues located in this stabilized loop do not interfere previously applied this approach to selection of a diversified rep- with direct contacts to the organophosphate ligand but can interro- ertoire of human Ig variable light (VL) and variable heavy (VH) gate second shell interactions, because the H3 loop has a conforma- single-chain fragments (scFv) displayed on phage particles (14). tion adjusted for binding. Kinetic and thermodynamic parameters Using a biotinylated phosphonate ester (1a) (Fig. 1) for solution along with computational docking support the active site model, phase modification, we identified a set of nucleophilic scFv including plasticity and simple catalytic components. Although rela- encompassing sites predisposed for covalent catalysis. Sequence λ tively uncomplicated, this catalytic machinery displays both stereo- analysis of the reactive clones revealed preferred pairing of -VL and chemical selectivity. The organophosphate pesticide paraoxon is and VH chains with conserved heavy chain complementary de- hydrolyzed by covalent catalysis with rate-limiting dephosphoryla- termining region 3 (CDR-H3) sequences. Two structural motifs tion. This reactibody is, therefore, a kinetically selected protein tem- were seen using distinct tyrosine residues in VL as nucleophile. Of plate that has enzyme-like catalytic attributes. these clones, seven of eight clones had a tyrosine (Y-L33) residue within CDR1 as a putative nucleophile. However, the most nu- cleophilically active clone, A.17, used a tyrosine residue (Y-L37) in catalytic antibodies | crystal structure | insecticide | organophosphate a conserved framework (FR2), a clear expansion of the conven- hydrolysis tional concept of a binding site. The term reactibody is proposed nzymes achieve extraordinary catalytic efficiencies by co- Eordinating binding forces and reactive residues within their ac- Author contributions: I.S., A.A.B., N.P., A.F., A.T., and A.G. designed research; I.S., I.K., V.A.M., tive sites to alter the mechanism and energy landscape of a chemical N.K., and A.R. performed research; E.C., F.N., Y.N., V.A.M., H.D., B.A., N.K., and A.A.M. reaction (1, 2). This functional complexity poses a challenge for contributed new reagents/analytic tools; I.S., E.C., I.K., F.N., Y.N., V.A.M., H.D., B.A., A.A.B., approaches to create artificial enzyme-like catalysts. The interplay N.K., A.R., P.M., A.G.T., N.P., A.A.M., A.F., A.T., and A.G. analyzed data; and I.S., I.K., A.A.B., between binding and catalysis can be explored through directed P.M., A.F., A.T., and A.G. wrote the paper. evolution strategies (3). Simple enzymatic activities have been de- The authors declare no conflict of interest. veloped by adaptation of a protein template for binding to stable *This Direct Submission article had a prearranged editor. analogsofsubstratesortransitionstates(TSA),whichwasexem- Freely available online through the PNAS open access option. plified by the generation of a broad array of catalytic antibodies (4– Data deposition: The crystallography, atomic coordinates, and structure factors have been 6), and some of these anti-TSA antibodies use a nucleophilic residue deposited in the Protein Data Bank, www.pdb.org (PDB ID codes 2XZA and 2XZC). in their catalytic mechanism (7). This approach has only excep- 1I.S., E.C., and I.K. contributed equally to this work. tionally been directed to recruit residues that participate in covalent 2To whom correspondence should be addressed. E-mail: [email protected]. catalysis that can be considered essential for multistep catalytic ef- This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. ficiency (8). However, aldolase antibodies obtained by reactive im- 1073/pnas.1108460108/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1108460108 PNAS Early Edition | 1of6 Downloaded by guest on September 25, 2021 1 23dissociation constant (KD, step 1) and rate constant for A.17 NH2 ﬁ CH3 phosphonylation (k2, step 2) revealed a catalytic ef ciency similar N to the efﬁciency of serine proteases reacting with phosphonate O R Ph P ester inactivators (16). O O O O Although hydrolytic turnover was not observed, the reaction P O scheme can be extended to include a step for hydrolysis of the F intermediate to regenerate the nucleophilic tyrosine (step 3), and O S O NO structural studies can be implemented to provide a rational de- 2 F sign basis to implement this step (1):

465 : : − − þ 1 KD ! − −− 2 k2! − − þ − O O Re Ab Tyr OH OPX Re Ab Tyr OPX Re Ab Tyr OP X O 3:k ðNu− Þ OPO HO OP N Re Ab −Tyr − OP þ X − 3 ! Re Ab− Tyr −OH þ Nu − ; [1] OPO O O S N where Re Ab = reactibody, X = leaving group, and Nu = water/ NO2 NO2 nucleophile. To clarify the molecular mechanism of the machinery of the Fig. 1. Chemical structures of compounds used in the study: p-nitrophenyl 8- reactibody, crystal structures of the native A.17 Fab (Protein Data methyl-8-azabicyclo[3.2.1]octyl phenylphosphonate (1b; R = H) and its bio- Bank accession code 2XZA) and its product modified by reaction tinylated derivative (1a; R = biotinyl); diisopropyl fluorophosphate (DFP; 2); 1b fl with (Protein Data Bank accession code 2XZC) have been 4-(2-aminoethyl)benzenesulfonyl uoride (AEBSF; 3); 2-diethoxyphosphor- solved at 1.5- and 1.36-Å resolution, respectively (Table S1). These ylthioethyl-trimethylammonium iodide (echothiophate; 4); O,O-diethyl O-(4- structures reveal a deep cavity at the interface between V (three nitrophenyl) phosphate (paraoxon; 5); O-(4-nitrophenylphosphoryl) choline (6). L β-sheets) and VH (four β-sheets) with several striking features (Fig. 2 and Movie S1). Most notably, the antigen-binding site seems to be subdivided into upper and lower chambers (Fig. 2 A–C), each here to denote a protein template selected for chemical reactivity chamber formed from VH and VL residues. The upper chamber rather than ground state binding. (Fig. 2A), defined primarily by five tyrosines (Y-H33, Y-H53, Y- Results and Discussion H59, Y-L33, and Y-L50) plus tryptophan W-L92 (Fig. 2D), is analogous to a specific tyrosine cluster typical of antibodies inter- To facilitate efforts to elucidate structure–function correlation of A.17 acting with DNA or other phosphoryl esters (17, 18). An opening at the selected reactibodies, the VL and VH elements of were W-L92, Y-H34, and N-H105 leads to a lower chamber consisting reconstructed for expression as full-length IgG in Chinese Hamster primarily of framework residues. Reactive Y-L37 forms the base of Ovary (CHO) cells (Fig. S1). The apparent nucleophilic reactivity the chamber (Fig. 2B) with its side chain packed against F-L100, of the scFv A.17 was retained in the whole IgG construct, which was W-H48, and W-H109, which orientates the hydroxyl group of determined using kinetic studies with phosphonate substrate 1b Y-L37 into the cavity at a depth of 15 Å from the surface. Within (Fig. 1). Analysis of the kinetic data for reaction between IgG A.17 this subsite, main chain atoms of G-H90/T-H91 and side chains of and 1b, using DYNAFIT software (15), is compatible with a kinetic S-L35, W-L92, P-L98, and F-L100 form a niche for the phenyl model proceeding through a single noncovalent intermediate be- group of the phosphonate ester, whereas H-N105 provides an H fore the rate-determining covalent step (1). Calculation of the bond to the pseudotropine nitrogen in the OP-A.17 structure (Fig.

Y-H59 A Y-L33 B C Y-H53 Y-L92 Y-H33 Y-H34 Y-L50 N-H105

Y-L37

Fig. 2. Structural analysis of reactibody D E Y- H 5 9 FabA.17. Crystallographic snapshots of the Y-H59 two-chamber active center. α-Chain trace renderings on the upper (A) and lower (B) Y-H53 Y- H 5 3 chambers and merged (C) of the substrate binding site of the A.17 active site shown in Y- H 3 3 Y-H33 W-H48 space-filling mode. The light chain is shown W-H48 W-L92 in pale blue, and the heavy chain is shown S-H51 S-H51 – W-L92 in green for A C. A top view showing aromatic residues forming the walls of the upper cavity and lower chamber centered Y- H 3 4 on reactive side chain of Y-L37 in the un- Y-H34 modified (D) and phosphonate-modified 2.8Å (E) structures. The light chain is shown in F-L100 N-H105 F-L100 gray, the heavy chain is in blue, the VH loop 2.5Å is in magenta, and the OP molecule is in A-H1073.2Å green. Key residues are represented as sticks, with nitrogen atoms marked by deep W-H109 W-H109 blue, oxygen atoms marked by red, and Y-L37 Y-L37 phosphorus marked by orange. A bound chloride ion, indicated by the green sphere, is constant to modification (Fig. S3A).

2of6 | www.pnas.org/cgi/doi/10.1073/pnas.1108460108 Smirnov et al. Downloaded by guest on September 25, 2021 TEPC-15 AChE residues A-H107 and W-H109 oriented into the cavity (Fig. 2E). In the unmodified structure, this solvent-exposed loop is highly dis- 2 18Å ordered, with a B factor of 40–50 Å compared with an overall B 6Å factor of 19 Å2. By contrast, the electron density map at these residues in OP-A.17 is well-defined, with B factors ranging from 15 2 49G7 BChE to 25 Å (Fig. S2). Additional displacements are required at H-H104 (3.05 Å), N-H105 (3.63 Å), A-H107 (0.70 Å), and W- 12Å 16Å H109 (0.86 Å) to accommodate the ligand and permit strong H bonding (2.79 Å) between the main-chain oxygen of N-H105 and the pseudotropine nitrogen (Figs. 2E and 4A). Significant 2 33F12 A.17 displacements in the L-chain, including a 20° χ angle rotation and 0.4-Å shift at W-L92 and a 10° rotation at Cα of F-L100, 13Å 15Å enable nesting of the phenyl ring of 1b with T-stacking interactions with the ligand. These motions may be concerted with a 5° rotation of the Y-L37 side chain. These observed conformational fi Fig. 3. Comparison of active site cavities of natural and de novo created rearrangements support the existence of an induced t-like biocatalysts. Chemically selected reactibody FabA.17 has a deep substrate mechanism during the ligand–antibody interaction. A cloud of binding niche. Cross-section views of the active center of esterolytic anti- electron density just above the face opposite to the Y-L37-OP bodies 49G7, TEPC15, aldolase antibody 33F12, choline esterases AChE and bond was modeled as a cluster of mobile water molecules. Y-H34 BChE, and antibody A.17 complexed with their ligands. In each case, the participates in an H-bonding network stabilizing this cluster (Fig. distance measured is the height of a pyramid with a triangle base con- 4B). The phosphonyl oxygen is strongly H-bonded to water w614 structed on the three residues nearest to the entrance of the active site, and (2.55 Å), suggesting polarization of the P–O bond to enhance apexes are the residue nearest to the ligand. electrophilicity (Fig. 2E). In unmodified A.17, a network of water molecules (w689, w688, and w274) bridges the hydroxyl group of S-L35toN-H105(Fig. S4). Conversely, transfer of the OP sub- 2E). The dimensions and topology of the site are atypical of anti- strate into the lower compartment must break this network and bodies in general, including catalytic antibodies, which are illus- displace these water molecules. trated in Fig. 3 (19–21). The reactive nucleophile is farther from In order to further assess interactions in the protein environ-

the protein surface than the corresponding group in an aldolase ment, thermodynamic parameters for the reaction of A.17 and IMMUNOLOGY antibody (15 vs. 13 Å). The depth and structure of the pocket may selected mutants with 1b were measured by isothermal titration be compared qualitatively with the catalytic sites of cholinesterases calorimetry (ITC) (Fig. S5A). The data were fitted to theoretical (Fig. 3) (22–24). binding curves assuming a single ligand site. In case of A.17 both Differences in the structure of native and 1b-modified reacti- entropy and enthalpy of reaction were favorable, with ΔH de- body suggest that considerable conformational rearrangement creasing threefold as temperature increased from 10 °C to 37 °C † accompanies ligand migration into the site (Fig. 4A). Basically, (Table 1). The observed enthalpy changes and apparent KD (KD ) there are two primary models of the mechanism of conformational were in line with ITC-estimated parameters for antibodies binding change because of antibody–ligand interactions: the existence of to their haptens (28). Furthermore, a theoretical ΔH of −2.56 kcal/ preequilibrium conformers and an induced fit mechanism (25– mol for computational docking of 1b from the surface to the lower 27). Conformational rearrangements of reactibody A.17 have been chamber is in good correspondence with the experimental value examined by both structural methods and kinetic approaches to of −2.78 kcal/mol (Fig. S6 and Movie S2). obtain a model that best fits one of these modes. Mutants Y-L33F and Y-L37F of A.17 IgG probe two possible Structural analyses revealed significant differences between nucleophilic residues identified in scFv reactibodies (14). Consis- crystal structures of unmodified and phosphonylated antibodies. tent with previous studies and the X-ray structure, only Y-L37F † Residues L99-G110 of HCDR3 seem to define a conformationally results in loss of reactivity. The ΔH and KD of A.17 and ΔH and KD mobile element that spans the upper and lower chambers, with key of the inactive mutant A.17 Y-L37F interacting with 1b (Fig. S5B) were nearly identical and within a factor of two of the respective factors for the BChE reaction (Table 1 and Fig. S5C). A very small enthalpy change might be expected for the ester exchange reaction A B where a P–OAr bond is broken and another formed to a tyrosine Y-H34 residue (or to a serine residue in the case of BChE). A similar effect N-H105 has been described in calorimetric studies of horse serum BChE H-H104 5 A.17 fl W-L92 inhibition by (29). The covalent reaction of must also re ect w363 free energies caused by noncovalent interactions stabilizing the w648 w668 adduct and free energies caused by release of the leaving group. 2.79Å w614 The latter energy change seemed to be negligible and was esti- A-H107 w588 2.55Å mated in reference experiments of p-nitrophenol binding to native w646 or chemically modified A.17 (Fig. S5D). These observations sug- W-H109 gest that the thermodynamic changes are sensitive primarily to

F-L100 Y-L37 noncovalent interactions. Accordingly, changes in heat capacity Y- L 3 7 because of binding, calculated from the temperature dependence of enthalpy changes (Fig. S5E), were used to estimate the change in fi solvent accessible area (30) using the following empirical formula Fig. 4. Superposition of active sites of native (green) and OP-modi ed (blue) 2 FabA.17 (A), illustrating the conformational changes of Y-L37, W- L92, F-L100, (Eq. ): and residues of the V loop (H-H104, N-H105, A-H107, and W-H109). Phos- H cal phonate moiety is in red. (B) Representative view of the active site of 1b- ΔCp ¼ − 70 ¼ 0:27ΔA þ 0:4ΔA ; [2] modified A.17 Fab showing a strong H bond between a nonbridging phos- mol$K ar nonar phonyl oxygen and water molecule w614 (2.55 Å). H-Y34 stabilizes a cluster of Δ Δ six water molecules by participating in their H-bond network. The light chain where Aar and Anonar are the protected areas in angstroms is in gray, the heavy chain is in blue, and the OP molecule is in green. Hydrogen squared because of formed by aromatic and nonaromatic residues, bonds are represented by gray dashes. A bound chloride ion, indicated by the respectively (31). Because covalently bound OP is effectively bur- green sphere, is independent of modification (Fig. S4). ied in the cavity, these changes can be attributed to conformational

Smirnov et al. PNAS Early Edition | 3of6 Downloaded by guest on September 25, 2021 Table 1. Thermodynamic parameters of A.17 WT, A.17 Y-L37F, 1 and 2 of equation 1 (Fig. 5A). The conformational fluctuations of A.17 H-H104A, and BChE binding to 1b determined by isothermal Ab were monitored through intrinsic tryptophan fluorescence titration calorimetry changing during binding of phosphonate molecule to A.17.We † observed a single increase in tryptophan fluorescence in a 100-ms Abzyme/ Temp KD(KD ) ΔH TΔS ΔG μ time range (Fig. 5B Left). The direct dependence of kobs on sub- enzyme (°C) ( M) (kcal/mol) (kcal/mol) (kcal/mol) strate concentration suggests the prevalence of an induced fit A.17 WT 10 2.9 −0.91 6.26 −7.17 mechanism in the reaction of A.17 with phosphonate 1b (Fig. 5B A.17 WT 15 3.5 −1.18 6.02 −7.20 Right) (25). Overall, our balanced view is that the evidence favors fi A.17 WT 20 2.8 −1.45 6.00 −7.45 an induced t-like model, whereas we can accept that, over the − − wide range of conditions that we have used, a significant fraction of A.17 WT 25 2.3 1.89 5.78 7.67 fl A.17 WT 37 1.7 −2.78 5.43 −8.21 the ux can occur through both pathways, resulting in a mixed A.17 Y-L37F 10 1.4 −1.49 6.08 −7.57 mechanism for this reaction (32). The efficiency of A.17 reacting with 1b is more than an order of A.17 Y-L37F 15 3.2 −1.37 5.88 −7.25 − − magnitude greater than the corresponding BChE reaction (Table A.17 Y-L37F 25 1.0 1.50 6.65 8.15 2), and it compares favorably with typical rates of serine protease A.17 Y-L37F 37 0.72 −1.35 7.28 −8.63 fi 1b − − modi cation by phosphonate (16, 33, 34). This kinetic advan- A.17 H-H104A 25 3.8 3.6 3.72 7.32 tage is a clear demonstration that the complementary matching of − − BChE 15 7.9 2.40 4.29 6.69 the shape and chemical reactivity of substrate 1b with the selected BChE 37 6.1 −4.35 2.93 −7.28 reactibody exceeds the performance of an enzyme such as BChE with a classical phosphorylating agent. Enzyme selectivity for the phosphonate structure may also distort these comparisons (33– 36). Reactive phosphoesters diisopropyl fluorophosphate (DFP) Δ − − 2 changes at the protein surface. The SAA of 175 to 260 Å (2), paraoxon (5), and 4-(2-aminoethyl)benzenesulfonyl fluoride Δ − 2 calculated from ITC is comparable with SAA of 240 Å calcu- (AEBSF) (3) inhibited A.17 reaction with 1a, whereas echothio- A.17 Δ lated from the crystal structures. In contrast to WT, the H of phate (4) had no effect at concentrations up to 10 mM (Fig. 6A and the Y-L37F interaction with 1b is independent of temperature, Figs. S3B and S8A). These molecules may be excluded from the suggesting slight conformational changes in case of incubation of reactive center while binding with varying affinities to the outer 1b the Y-L37F antibody with (Fig. S5E). site. Stereoselective SN2 reaction with the RP enantiomer of 1b was Mutants N-H105A and H-H104A were also generated to in- deduced from the configuration at phosphorus in the A.17-OP fl vestigate the role of the conformationally exible CDR-H3 loop on adduct. Rate constants could be higher by a factor of two if the SP reactivity. These mutations do not interfere with direct contacts to enantiomer does not react or even more if it is an inhibitor. Over the OP ligand but can interrogate second shell interactions as the 5 † a longer time span, hydrolysis of paraoxon was detected, sug- CDR-H3 loop adjusts for binding. ITC data show similar KD but gesting that a smaller OP molecule can access the lower chamber in differing contribution of ΔH and ΔS for the reactions of 1b with a productive orientation. Moreover, p-nitrophenol product was A.17WT and A.17 H-H104A, where the latter reaction is en- released in molar excess to active sites (Fig. S8B), and the hydro- thalpically more favorable (Table 1 and Fig. S5F). The effect of lysis rate was linearly dependent on A.17 concentration (Fig. S3C). both mutations on the kinetically determined KD is also minimal, The reaction rate also increased linearly with hydroxylamine whereas the rate constant k2 decreases 2.4-fold in H-H104A and concentration (Fig. 6B), whereas mutant Y-L37F had no activity increases 1.3-fold in N-H105A (Table 2). Thus, mutations within (Fig. S3D). These observations strongly support a reaction mech- the H3 loop influence the pseudofirst order reaction rate without anism that proceeds through a phosphotyrosine covalent inter- significantly altering the binding step 2 [1]. These results support an mediate at Y-L37. Elementary rate constants of the reaction (k = − − − − 2 active site model where H3 loop dynamics allow penetration of 1.1 ± 0.1 × 10 1 min 1, k =1.6± 0.2 × 10 2 min 1,andk =1.26± − − − 3 4 reactant from outer to inner chambers. The H-H104 side chain 0.09 × 10 3 mM 1 min 1) were estimated where dephosphorylation toggles between intrachain H bonds to D-H106 in unmodified A.17 is rate limiting (Fig. 6B). Interaction between A.17 and paraoxon and to T-H100 in A.17-OP (Fig. S7). The H-H104A substitution leads to the accumulation of the covalent intermediate according may increase H3 conformational freedom but also can impair in- to the kinetic data. The existence of the covalent intermediate, duced fit by destabilizing a conformation favored in the product. supporting the covalent catalysis mechanism for paraoxon hydro- The N-H105 side chain participates in H bonding to the water lysis by the A.17 reactibody, was also proved by MS analysis show- bridge with Y-L37 in the unmodified structure (Fig. S4). N-H105A ing the existence of phosphorylated antibody (Fig. S8C). weakens this lattice, which could facilitate displacements for in- By contrast, the A.17Y-L37-OP ester obtained from reaction sertion of substrate into the reactive center. Based on these with 1b was stable to hydrolysis under both native and denaturing observations and data from crystallographic analysis, we suggest conditions, whereas hydroxylamine (1–100 mM) had no influence that covalent modification of reactibody A.17 by 1b can be de- on the observed rate (Fig. S3E). Monoisonitrosoacetone, a classi- scribed by an induced fit model. To clarify this model, we carried cal cholinesterase reactivator, also failed to regenerate A.17 re- out presteady state kinetic investigations of fluorescence changing activity. Thus, notwithstanding their chemical similarity, the two during the phosphonylation reaction. According to steady state A.17 phosphoesters show quite different P–O bond stability, sug- kinetics, the minimal kinetic scheme of the reaction includes stages gesting that the active site assists hydrolysis of the intermediate generated from 5. Although the mechanism remains speculative and the rate of the phosphate release is slow, these data show that Table 2. Titration of the active sites of A.17, its mutants, and a pathway is available for catalytic turnover, which is described by BChE with 1b (Kitz–Wilson) equation 1. The enhanced reactivity of amino acid residues used in covalent k2/KD −1 μ −1 −1 × −4 catalysis is an inherent function of the structured environment Reactibody/enzyme k2 (min ) KD ( M) (min M 10 ) within the enzyme active site. Reactive selection shows how A.17 WT 0.25 ± 0.02 6.2 ± 0.4 4.0 ± 0.6 architectures that support efficient chemistry can arise from rela- A.17 Y-L37F Inactive Inactive Inactive tively minor alterations of the protein interior. Studies of aldolase A.17 H-H104A 0.14 ± 0.01 8.5 ± 2.5 1.6 ± 0.6 antibodies obtained by reactive immunization suggested the re- A.17 N-H105A 0.32 ± 0.01 7.7 ± 1.8 4.2 ± 1.1 active nucleophile could emerge from a single mutation releasing BChE* 0.066 ± 0.006 24 ± 60.3± 0.1 a lysine side chain constrained in a hydrophobic environment by a salt bridge or H bond (10, 21). The A.17 reactibody reveals an *The data obtained for BChE are consistent with previously reported data (34). alternative solution to reorganizing available structures for re-

4of6 | www.pnas.org/cgi/doi/10.1073/pnas.1108460108 Smirnov et al. Downloaded by guest on September 25, 2021 A 12 A A.17+1a

M 10 Experimental trace µ

, Fitted by model 1 8 Fitted by model 2 Model 1 0.030 k2 Ab + OPX Ab-OP 6 DFP (2)AEBSF (3) EchothiophateParaoxon (4) (5) 0.000 a-Fc-HRP -0.030Residuals 4 0 400 800 1,200 1,600 [p-nitrophenol] Model 2 C 2 0.030 KD k2 Ab + OPX Ab--OPX Ab-OP Strept-HRP 0.000 0 -0.030Residuals 0 400 800 1,200 1,600 Ab + P2 B K k k 3 0 200 400 600 800 1,000 1,200 1,400 1,600 1,800 D 2 Ab + OPX Ab--OPX Ab-OP k 4 [ Time, s NH Ab + P P1 2 OH 3 5.0 ] KD k2 B Ab + OPX (Ab--OPX)’ (Ab--OPX)’’ k-2 -1 4.0 OPX, µM 55 , min

1.0 2 3.0

50 10 2.2 3.0 . k4 2.0 45 cat k k3 10.0 40 1.0 35 2.0 20.0 0

obs 30 0 5 10 15 20 25 k

25 K = 17.4±3.3 µM 40.0 D Fig. 6. Description of multistep covalent catalysis by A.17 reactibody. Co- Trp fluorescence, a.u. k = 54.9±2.7 s-1 2 valent reactivity of A.17 and its mutants with biotinylated phosphonate 1a. 20 k = 12.1±1.3 s-1

-2 IMMUNOLOGY 1.8 50.0 Antibodies were incubated for 1 h at 37 °C with PBS buffer alone or 1 mM each 15 of 2 (DFP) and 3 (AEBSF) and for 16 h with PBS buffer alone or 1 mM each 4 10 (echothiophate) and 5 (paraoxon). All samples were then incubated for 1 h at 0.01 0.1 01020304050 37 °C with 100 μMof1a and analyzed by Western blot (A). Concentrations of Time, s [OPX], µM antibodies were normalized as confirmed by comparable staining of heavy chains on the same blots (A Upper). Three-step reaction of A.17 antibody with Fig. 5. Discrimination of the modification mechanism between A.17 reac- 5 in the presence of NH2OH (B). tibody and phosphonate 1b. Experimental and fitted kinetic curves of interaction of A.17 with 1b. The quality of fit of kinetic models to the experimental data was assessed by monitoring residuals against time for fi netic selection may exploit protein dynamics, exposing reactive different scheme ts (A). Presteady state kinetics curves of the intrinsic Trp residues in the protein interior. It is remarkable that motions deep fluorescence changing of A.17 depending on concentration of phosphonate within the pocket also enhance substrate complementarity. Al- 1b (B Left). Dependence of kobs of the 1b interaction with reactibody A.17 on substrate concentration (B Right). AInsetshows the reaction mechanism though it is possible that a favorable enthalpy of chemical reaction used for data fitting and elementary constant calculation. drives these displacements, the similar free energy of binding to a chemically inert Y-L37F mutant suggested that noncovalent interactions may promote conformational changes for a produc-

activity. In antibody structures, Y-L37 is invariably buried at the VL- tive Michaelian complex. Collectively, our observations provide a VH interface, where it packs against W-H109 (103) and adjacent compelling case for coupling between dynamics of the cavity and residues at the base of CDR-H3 (e.g., Cr6361, Protein Data Bank reactivity at the tyrosine side chain. 3GBM, 85% VL homology; E51, Protein Data Bank 1RZF, 84% VL Details of the structure capable of revealing interactions with a homology). By contrast, a long and flexible H3 loop in A.17 allows putative trigonal bipyramidal transition state for ester exchange sufficient displacement at its base to open a chamber directly above and a trajectory for the leaving group are not evident from the the Y-L37 side chain. It is worth noting that H3 sequences were structures analyzed herein. However, small adjustments in contact synthetically diversified in the scFv library from which A.17 was residues are clearly capable of accommodating such a transition derived (37). It is conceivable that this diversity encompasses state. The water cluster opposite the new P–O bond in the complex sequences that are rare or prohibited in a natural repertoire. may also play a role, and one water molecule is poised to hydrolyze The disordered H3 loop in unmodified A.17 seems to act as the phosphotyrosine ester to achieve turnover. Although there is a flexible wall of the cavity to facilitate migration of ligand into the no evidence that the site can activate this water for such nucleo- lower chamber. Moreover, compression and stabilization of the philic attack, a mechanism must be available to accomplish the phosphonylated protein, evident from both the structure analysis observed hydrolysis of the intermediate formed by reaction with and ITC studies, were indicative of a correlation between reactivity paraoxon 5. Possibly, the site is locked in the case of reaction with and protein dynamics, and they suggest that structural plasticity is 1b, whereas the interaction with 5 allows dynamic displacements a selected feature. Conformational changes in antibody binding for the second-step reaction. Additional studies to clarify this are not uncommon and may be seen with protein antigens where question may uncover possible means of improving the observed flexibility can optimize multiple interactions over a large antibody– catalytic efficiency. antigen interface. However, antibodies elicited against haptens or We set out to show that kinetic selection for covalent reactivity other small ligands show evidence of diminishing V-region plas- can identify protein sites correctly configured for development of ticity, because somatic mutations serve to anchor topologies suit- artificial biocatalysts. Our thermodynamic and structural analyses able for ligand complementarity (38, 39). Plasticity encoded in the unexpectedly revealed an associated feature with broader impli- V germline is thought to amplify combinatorial diversity, because cations. Recently, the question of how enzymes harness protein various shapes could be imprinted on each template. Accordingly, dynamics for catalytic efficiency has been revisited through affinity maturation favors more rigid protein structures and may structural and computational approaches (1, 40). Coupled con- preclude dynamic features in enzymatic function. By contrast, ki- formational substates optimized to admit or release substrates or

Smirnov et al. PNAS Early Edition | 5of6 Downloaded by guest on September 25, 2021 organize the active site for catalysis have been described (41). carried out in 0.1 M phosphate buffer (pH 7.4) at 25 °C. Stopped-flow meas- Reactibodies can present unique cavities that facilitate simple urements with fluorescence detection were carried out using a model SX.18MV chemical processes and provide a useful model for how proteins stopped-flow spectrometer (Applied Photophysics). The Ab concentration was can acquire these adaptations. Strategies to deploy antibodies as fixed at 3 μM, and the concentration of 1b was varied from 1 to 50 μM. Nu-

enzymes have been criticized on the grounds that an Ig template cleophile competition in the presence of NH2OH at concentrations ranging cannot deploy residues for covalent catalysis and that it lacks the from 0 to 100 mM is carried out under the same reaction conditions with flexibility required for dynamic mechanisms. Reactive immuni- background correction. zation and kinetic selection have provided evidence to dispel these concerns. At the same time, our reactibody template has proved ESI-FTICR-MS (Top Down). The A.17 antibody (10 μM) was incubated with 5 adaptable in ways that may apply also to other protein scaffolds. (paraoxon; 2 mM in 0.1 M phosphate buffer, pH 7.4) at 37 °C for 16 h. Samples were analyzed directly using ESI-MS with a Bruker APEX Ultra FTICR Methods mass spectrometer (Bruker Daltonics). Crystallization of FabA.17 and OP-FabA.17. The purified A.17 Fab fragment (9 mg/mL) was crystallized using the hanging drop method. Diffraction data ACKNOWLEDGMENTS. We thank Professor Bernard Green for reading the were collected at the European Synchrotron Radiation Facility on ID14-eh4 manuscript and providing valuable comments and Professor Evgeny Nikolaev, beam line using λ = 0.979-Å wavelength with ADSC Quantum 4 detector. Dr. Alexey Kononikhin, and Dr. Oleg Kharybin for performing ESI-FTICR-MS analysis. Highly purified BChE from human plasma was kindly provided by ITC. The thermodynamic parameters of A.17WT, A.17 mutants, or BChE from Professor Anikienko. This research was performed in frames of the Skolkovo human plasma reacting with 1b or p-nitrophenol were measured on a program and supported by North Atlantic Treaty Organization SfP#982833, MicroCal iTC200 instrument (MicroCal). Aliquots of ligands (2.5 μL) were Russian Education Agency Contract N P1371, PICS Centre National de la Recherche Scientifique N 4238 France-Russia, Russian Foundation for Basic injected into the 0.2-mL cell containing the protein solution in PBS buffer Research Grants 10.04.00673a and 07.04.92168, Federal Special Purpose Pro- (pH 7.4) to achieve a complete binding isotherm. The resulting titration gram “Research and Development in Priority Directions of Growth of Scien- fi curves were tted using MicroCal Origin software. tific-Technological Complex of Russia for 2007–2012”, an International Center for Genetic Engineering and Biotechnology (Italy) grant, Scientific Schools Evaluation of Kinetic Mechanism and Processing of Primary Data. Reactions of 64658.2010 “Chemical Basis of Biocatalysis,” and Programs of the Russian A.17WT and its mutants (3–32 μM) or highly purified BChE (3–10 μM) from Academy of Sciences “Fundamental Science for Medicine” and “Molecular human plasma with 1b (or 5) at concentrations ranging from 10 to 500 μMwere and Cellular Biology.”

1. Benkovic SJ, Hammes-Schiffer S (2003) A perspective on enzyme catalysis. Science 301: 22. Carletti E, et al. (2009) Structure-activity analysis of aging and reactivation of human 1196e1202. butyrylcholinesterase inhibited by analogues of tabun. Biochem J 421:97e106. 2. Tawfik DS (2010) Messy biology and the origins of evolutionary innovations. Nat 23. Sanson B, et al. (2009) Crystallographic snapshots of nonaged and aged conjugates of Chem Biol 6:692e696. soman with acetylcholinesterase, and of a ternary complex of the aged conjugate 3. Gupta RD, et al. (2011) Directed evolution of hydrolases for prevention of G-type with pralidoxime. J Med Chem 52:7593e7603. nerve agent intoxication. Nat Chem Biol 7:120e125. 24. Dvir H, Silman I, Harel M, Rosenberry TL, Sussman JL (2010) Acetylcholinesterase: From 4. Belogurov A, Jr., Kozyr A, Ponomarenko N, Gabibov A (2009) Catalytic antibodies: 3D structure to function. Chem Biol Interact 187:10e22. Balancing between Dr. Jekyll and Mr. Hyde. Bioessays 31:1161e1171. 25. Foote J, Milstein C (1994) Conformational isomerism and the diversity of antibodies. 5. Lerner RA, Benkovic SJ, Schultz PG (1991) At the crossroads of chemistry and immu- Proc Natl Acad Sci USA 91:10370e10374. nology: Catalytic antibodies. Science 252:659e667. 26. Lindner AB, Eshhar Z, Tawfik DS (1999) Conformational changes affect binding and 6. Savinov SN, Hirschmann R, Benkovic SJ, Smith AB, 3rd (2003) Investigation of an an- catalysis by ester-hydrolysing antibodies. J Mol Biol 285:421e430. tibody-ligase. Evidence for strain-induced catalysis. Bioorg Med Chem Lett 13: 27. James LC, Roversi P, Tawfik DS (2003) Antibody multispecificity mediated by confor- e 1321 1324. mational diversity. Science 299:1362e1367. 7. Stewart JD, Roberts VA, Thomas NR, Getzoff ED, Benkovic SJ (1994) Site-directed 28. Wade H, Scanlan TS (2003) Binding and catalysis: A thermodynamic study on a cata- mutagenesis of a catalytic antibody: An arginine and a histidine residue play key lytic antibody system. ChemBioChem 4:537e540. e roles. Biochemistry 33:1994 2003. 29. Debord J, Harel M, Verneuil B, Bollinger JC, Dantoine T (2009) Microcalorimetric study fi 8. Wagner J, Lerner RA, Barbas CF, 3rd (1995) Ef cient aldolase catalytic antibodies that of the inhibition of butyrylcholinesterase by paraoxon. Anal Biochem 389:97e101. e use the enamine mechanism of natural enzymes. Science 270:1797 1800. 30. Connelly PR, Thomson JA (1992) Heat capacity changes and hydrophobic interactions 9. Zhu X, et al. (2004) The origin of enantioselectivity in aldolase antibodies: Crystal struc- in the binding of FK506 and rapamycin to the FK506 binding protein. Proc Natl Acad ture, site-directed mutagenesis, and computational analysis. JMolBiol343:1269e1280. Sci USA 89:4781e4785. 10. Barbas CF, 3rd, et al. (1997) Immune versus natural selection: Antibody aldolases with 31. Jelesarov I, Bosshard HR (1999) Isothermal titration calorimetry and differential enzymic rates but broader scope. Science 278:2085e2092. scanning calorimetry as complementary tools to investigate the energetics of bio- 11. Demuth HU, et al. (1988) Molecular interactions in intermediate and transition states molecular recognition. J Mol Recognit 12:3e18. in the self-stimulated inhibition of enzymes. Pharmazie 43:262e264. 32. Hammes GG, Chang YC, Oas TG (2009) Conformational selection or induced fit: A flux 12. Soumillion P, Fastrez J (2001) Novel concepts for selection of catalytic activity. Curr description of reaction mechanism. Proc Natl Acad Sci USA 106:13737e13741. Opin Biotechnol 12:387e394. 33. Becker EL (1967) The relationship of the structure of phosphonate esters to their 13. Cesaro-Tadic S, et al. (2003) Turnover-based in vitro selection and evolution of bio- ability to inhibit chymotrypsin, trypsin, acetylcholinesterase and C’Ia. Biochim Biophys catalysts from a fully synthetic antibody library. Nat Biotechnol 21:679e685. Acta 147:289e296. 14. Reshetnyak AV, et al. (2007) Routes to covalent catalysis by reactive selection for 34. Tramontano A, Ivanov B, Gololobov G, Paul S (2000) Inhibition and labeling of nascent protein nucleophiles. J Am Chem Soc 129:16175e16182. 15. Kuzmic P (1996) Program DYNAFIT for the analysis of enzyme kinetic data: Applica- enzymes and abzymes by phosphonate diesters. Appl Biochem Biotechnol 83: e tion to HIV proteinase. Anal Biochem 237:260e273. 233 242. 16. Oleksyszyn J, Boduszek B, Kam CM, Powers JC (1994) Novel amidine-containing 35. Jackson DS, et al. (1998) Synthesis and evaluation of diphenyl phosphonate esters as peptidyl phosphonates as irreversible inhibitors for blood coagulation and related inhibitors of the trypsin-like granzymes A and K and mast cell tryptase. J Med Chem e serine proteases. J Med Chem 37:226e231. 41:2289 2301. 17. Kim YR, et al. (2006) Heavy and light chain variable single domains of an anti-DNA 36. Boduszek B, et al. (1994) Dipeptide phosphonates as inhibitors of dipeptidyl pepti- binding antibody hydrolyze both double- and single-stranded DNAs without se- dase IV. J Med Chem 37:3969e3976. quence specificity. J Biol Chem 281:15287e15295. 37. Griffiths AD, et al. (1994) Isolation of high affinity human antibodies directly from 18. Schuermann JP, Prewitt SP, Davies C, Deutscher SL, Tanner JJ (2005) Evidence for large synthetic repertoires. EMBO J 13:3245e3260. structural plasticity of heavy chain complementarity-determining region 3 in anti- 38. Yin J, Beuscher AE, 4th, Andryski SE, Stevens RC, Schultz PG (2003) Structural plasticity body-ssDNA recognition. J Mol Biol 347:965e978. and the evolution of antibody affinity and specificity. J Mol Biol 330:651e656. 19. Patten PA, et al. (1996) The immunological evolution of catalysis. Science 271: 39. Wedemayer GJ, Patten PA, Wang LH, Schultz PG, Stevens RC (1997) Structural insights 1086e1091. into the evolution of an antibody combining site. Science 276:1665e1669. 20. Scotti C, Gherardi E (2006) Structural basis of affinity maturation of the TEPC15/ 40. Henzler-Wildman KA, et al. (2007) Intrinsic motions along an enzymatic reaction Vkappa45.1 anti-2-phenyl-5-oxazolone antibodies. J Mol Biol 359:1161e1169. trajectory. Nature 450:838e844. 21. Zhu X, Tanaka F, Lerner RA, Barbas CF, 3rd, Wilson IA (2009) Direct observation of 41. Jackson CJ, et al. (2009) Conformational sampling, catalysis, and evolution of the an enamine intermediate in amine catalysis. J Am Chem Soc 131:18206e18207. bacterial phosphotriesterase. Proc Natl Acad Sci USA 106:21631e21636.

6of6 | www.pnas.org/cgi/doi/10.1073/pnas.1108460108 Smirnov et al. Downloaded by guest on September 25, 2021