1.85 A structure of anti-fluorescein 4-4-20 Fab

Engineered cytokine/antibody fusion proteins improve delivery of IL-2 to pro-inflammatory cells and promote antitumor activity

Progress in cytokine engineering is driving therapeutic translation by overcoming the inherent limitations of these proteins as drugs. The interleukin-2 (IL-2) cytokine harbors great promise as an immune stimulant for cancer treatment. However, the cytokine's concurrent activation of both pro-inflammatory immune effector cells and anti-inflammatory regulatory T cells, its toxicity at high doses, and its short serum half-life have limited clinical application. One promising approach to improve the selectivity, safety, and longevity of IL-2 is complexation with anti-IL-2 antibodies that bias the cytokine towards the activation of immune effector cells (i.e., effector T cells and natural killer cells). Although this strategy shows therapeutic potential in preclinical cancer models, clinical translation of a cytokine/antibody complex is complicated by challenges in formulating a multi-protein drug and concerns about complex stability. Here, we introduce a versatile approach to designing intramolecularly assembled single-agent fusion proteins (immunocytokines, ICs) comprising IL-2 and a biasing anti-IL-2 antibody that directs the cytokine's activities towards immune effector cells. We establish the optimal IC construction and further engineer the cytokine/antibody affinity to improve immune biasing function. We demonstrate that our IC preferentially activates and expands immune effector cells, leading to superior antitumor activity compared to natural IL-2 without inducing toxicities associated with IL-2 administration. Collectively, this work presents a roadmap for the design and translation of immunomodulatory cytokine/antibody fusion proteins.

Structural and biochemical basis of the formation of isoaspartate in the complementarity-determining region of antibody 64M-5 Fab

The formation of the isoaspartate (isoAsp) is one of spontaneous degradation processes of proteins, affecting their stability and activity. Here, we report for the first time the crystal structures of an antibody Fab that contains isoAsp in the complementarity-determining region (CDR), along with biochemical studies to detect isoAsp. By comparing the elution profiles of cation-exchange chromatography, it was clarified that the antibody 64M-5 Fab is converted from the normal form to isoAsp form spontaneously and time-dependently under physiological conditions. The isoAsp residue was identified with tryptic peptide mapping, N-terminal sequencing, and the protein isoaspartyl methyltransferase assay. Based on the fluorescence quenching method, the isoAsp form of 64M-5 Fab shows a one order of magnitude lower binding constant for its dinucleotide ligand dT(6-4)T than the normal form. According to the structure of the isoAsp form, the conformation of CDR L1 is changed from the normal form to isoAsp form; the loss of hydrogen bonds involving the Asn28L side-chain, and structural conversion of the β-turn from type I to type II'. The formation of isoAsp leads to a large displacement of the side chain of His27dL, and decreased electrostatic interactions with the phosphate group of dT(6-4)T. Such structural changes should be responsible for the lower affinity of the isoAsp form for dT(6-4)T than the normal form. These findings may provide insight into neurodegenerative diseases (NDDs) and related diseases caused by misfolded proteins.

Integrin β1 Promotes the Interaction of Murine IgG3 with Effector Cells

Abs exert several of their effector functions by binding to cell surface receptors. For murine IgG3 (mIgG3), the identity of its receptors (and the very existence of a receptor) is still under debate, as not all mIgG3 functions can be explained by interaction with FcγRI. This implies the existence of an alternate receptor, whose identity we sought to pinpoint. We found that blockage of integrin β1 selectively hampered binding of mIgG3 to macrophages and mIgG3-mediated phagocytosis. Manganese, an integrin activator, increased mIgG3 binding to macrophages. Blockage of FcγRI or Itgb1 inhibited binding of different mIgG3 Abs to variable extents. Our results are consistent with the notion that Itgb1 functions as part of an IgG receptor complex. Given the more ancient origin of integrins in comparison with FcγR, this observation could have far-ranging implications for our understanding of the evolution of Ab-mediated immunity as well as in immunity to microorganisms, pathogenesis of autoimmune diseases, and Ab engineering.

Measuring the sequence-affinity landscape of antibodies with massively parallel titration curves

Despite the central role that antibodies play in the adaptive immune system and in biotechnology, much remains unknown about the quantitative relationship between an antibody’s amino acid sequence and its antigen binding affinity. Here we describe a new experimental approach, called Tite-Seq, that is capable of measuring binding titration curves and corresponding affinities for thousands of variant antibodies in parallel. The measurement of titration curves eliminates the confounding effects of antibody expression and stability that arise in standard deep mutational scanning assays. We demonstrate Tite-Seq on the CDR1H and CDR3H regions of a well-studied scFv antibody. Our data shed light on the structural basis for antigen binding affinity and suggests a role for secondary CDR loops in establishing antibody stability. Tite-Seq fills a large gap in the ability to measure critical aspects of the adaptive immune system, and can be readily used for studying sequence-affinity landscapes in other protein systems.

DOI:http://dx.doi.org/10.7554/eLife.23156.001

Antibodies are proteins produced by cells of the immune system to tag or neutralize potential threats to the body, such as foreign substances and disease-causing microbes. Antibodies do this by binding to target molecules called antigens. An antibody’s ability to bind to an antigen depends on the sequence of amino acids – the building blocks of proteins – that make up the antibody. Through a process that randomizes this sequence of amino acids, the immune system generates a vast pool of antibodies that are able to target almost any foreign antigen that exists in nature.

Currently, little is understood about how the sequence of amino acids in an antibody determines how strongly that antibody binds to its antigen target – a property referred to as the antibody’s binding affinity. Answering this fundamental question requires techniques that can measure the affinities of many different antibodies at the same time. However, previous high-throughput methods have been unable to provide quantitative measurements of binding affinities. These kinds of measurements are difficult because an antibody’s amino acid sequence governs more than just binding affinity: it also affects how easy it is to produce that antibody, and what fraction of antibody molecules work properly.

Adams et al. now describe a new method, named “Tite-Seq”, that overcomes these issues. First, thousands of different antibodies are displayed on the surface of yeast cells, with each cell carrying a single kind of antibody. These cells are then incubated with fluorescently labeled antigen at a wide range of different concentrations. Next, the yeast cells are sorted based on how brightly they glow; brighter cells have more antigen bound to them, and so it is possible to calculate how much of the antigen is bound to each kind of antibody at each concentration. Plotting these data provides a “binding curve” for each antibody, which is then used to read off the antibody’s binding affinity in a way that is not affected by the factors that have plagued other high-throughput methods.

Tite-Seq is thus able to measure the binding affinities for thousands of different antibodies at the same time. This will potentially allow researchers to address many fundamental and yet unanswered questions about how the immune system works. Tite-Seq can also be used to measure how amino acid sequence affects the binding affinity of proteins other than antibodies.

DOI:http://dx.doi.org/10.7554/eLife.23156.002

Three-dimensional structure, binding, and spectroscopic characteristics of the monoclonal antibody 43.1 directed to the carboxyphenyl moiety of fluorescein

Unlike other known anti-fluorescein antibodies, the monoclonal antibody 43.1 is directed toward the fluorescein's carboxyl phenyl moiety. It demonstrates a very high affinity (KD ∼ 70 pM) and a fast association rate (kon ∼ 2 × 10(7) M(-1 ) s(-1) ). The three-dimensional structure of the Fab 43.1-fluorescein complex was resolved at 2.4 Å resolution. The antibody binding site is exclusively assembled by the CDR loops. It is comprised of a 14 Å groove-shaped entrance leading to a 9 Å by 7 Å binding pocket. The highly polar binding pocket complementary encloses the fluorescein's carboxyphenyl moiety and tightly fixes it by multiple hydrogen bonds. The fluorescein's xanthene ring is embedded in the more hydrophobic groove and stacked between the side chains of Tyr37L and of Arg99H providing conditions for an excited state electron transfer process. In comparison to fluorescein, the absorption spectrum of the complex in the visible region is shifted to the "red" by 23 nm. The complex demonstrates a very weak fluorescence (Φc  = 0.0018) with two short lifetime components: 0.03 ns (47%) and 0.8 ns (24%), which reflects a 99.8% fluorescein emission quenching effect upon complex formation. The antibody 43.1 binds fluorescein with remarkable affinity, fast association rate, and strongly quenches its emission. Therefore, it may present a practical interest in applications such as molecular sensors and switches.

Direct and indirect interactions in the recognition between a cross-neutralizing antibody and the four serotypes of dengue virus

Dengue fever is the most important vector-borne viral disease. Four serotypes of dengue virus, DENV1 to DENV4, coexist. Secondary infection by a different serotype is a risk factor for severe dengue. Monoclonal antibody mAb4E11 neutralizes the four serotypes of DENV with varying efficacies by recognizing an epitope located within domain-III (ED3) of the viral envelope (E) protein. To better understand the cross-reactivities between mAb4E11 and the four serotypes of DENV, we constructed mutations in both Fab4E11 fragment and ED3, and we searched for indirect interactions in the crystal structures of the four complexes. According to the serotype, 7 to 12 interactions are mediated by one water molecule, 1 to 10 by two water molecules, and several of these interactions are conserved between serotypes. Most interfacial water molecules make hydrogen bonds with both antibody and antigen. Some residues or atomic groups are engaged in both direct and water-mediated interactions. The doubly-indirect interactions are more numerous in the complex of lowest affinity. The third complementarity determining region of the light chain (L-CDR3) of mAb4E11 does not contact ED3. The structures and double-mutant thermodynamic cycles showed that the effects of (hyper)-mutations in L-CDR3 on affinity were caused by conformational changes and indirect interactions with ED3 through other CDRs. Exchanges of residues between ED3 serotypes showed that their effects on affinity were context dependent. Thus, conformational changes, structural context, and indirect interactions should be included when studying cross-reactivity between antibodies and different serotypes of viral antigens for a better design of diagnostics, vaccine, and therapeutic tools against DENV and other Flaviviruses.

Plug-and-play pairing via defined divalent streptavidins

Streptavidin is one of the most important hubs for molecular biology, either multimerizing biomolecules, bridging one molecule to another, or anchoring to a biotinylated surface/nanoparticle. Streptavidin has the advantage of rapid ultra-stable binding to biotin. However, the ability of streptavidin to bind four biotinylated molecules in a heterogeneous manner is often limiting. Here, we present an efficient approach to isolate streptavidin tetramers with two biotin-binding sites in a precise arrangement, cis or trans. We genetically modified specific subunits with negatively charged tags, refolded a mixture of monomers, and used ion-exchange chromatography to resolve tetramers according to the number and orientation of tags. We solved the crystal structures of cis-divalent streptavidin to 1.4 Å resolution and trans-divalent streptavidin to 1.6 Å resolution, validating the isolation strategy and explaining the behavior of the Dead streptavidin variant. cis- and trans-divalent streptavidins retained tetravalent streptavidin's high thermostability and low off-rate. These defined divalent streptavidins enabled us to uncover how streptavidin binding depends on the nature of the biotin ligand. Biotinylated DNA showed strong negative cooperativity of binding to cis-divalent but not trans-divalent streptavidin. A small biotinylated protein bound readily to cis and trans binding sites. We also solved the structure of trans-divalent streptavidin bound to biotin-4-fluorescein, showing how one ligand obstructs binding to an adjacent biotin-binding site. Using a hexaglutamate tag proved a more powerful way to isolate monovalent streptavidin, for ultra-stable labeling without undesired clustering. These forms of streptavidin allow this key hub to be used with a new level of precision, for homogeneous molecular assembly.

•

Streptavidin is a widely used nanohub: often, tetravalency makes assembly imprecise.

•

We isolated divalent streptavidins with biotin-binding sites in cis or in trans.

•

cis- and trans-divalent streptavidins retain exceptional biotin binding and thermostability.

•

Binding large biotinylated ligands to adjacent binding sites is greatly disfavored.

•

cis- and trans-divalent streptavidins represent a simple way to bridge biomolecules precisely.

Malachite green mediates homodimerization of antibody VL domains to form a fluorescent ternary complex with singular symmetric interfaces

We report that a symmetric small-molecule ligand mediates the assembly of antibody light chain variable domains (VLs) into a correspondent symmetric ternary complex with novel interfaces. The L5* fluorogen activating protein is a VL domain that binds malachite green (MG) dye to activate intense fluorescence. Crystallography of liganded L5* reveals a 2:1 protein:ligand complex with inclusive C2 symmetry, where MG is almost entirely encapsulated between an antiparallel arrangement of the two VL domains. Unliganded L5* VL domains crystallize as a similar antiparallel VL/VL homodimer. The complementarity-determining regions are spatially oriented to form novel VL/VL and VL/ligand interfaces that tightly constrain a propeller conformer of MG. Binding equilibrium analysis suggests highly cooperative assembly to form a very stable VL/MG/VL complex, such that MG behaves as a strong chemical inducer of dimerization. Fusion of two VL domains into a single protein tightens MG binding over 1000-fold to low picomolar affinity without altering the large binding enthalpy, suggesting that bonding interactions with ligand and restriction of domain movements make independent contributions to binding. Fluorescence activation of a symmetrical fluorogen provides a selection mechanism for the isolation and directed evolution of ternary complexes where unnatural symmetric binding interfaces are favored over canonical antibody interfaces. As exemplified by L5*, these self-reporting complexes may be useful as modulators of protein association or as high-affinity protein tags and capture reagents.

Antigen-antibody interactions and structural flexibility of a femtomolar-affinity antibody

The femtomolar-affinity mutant antibody (4M5.3) generated by directed evolution is interesting because of the potential of antibody engineering. In this study, the mutant and its wild type (4-4-20) were compared in terms of antigen-antibody interactions and structural flexibility to elucidate the effects of directed evolution. For this purpose, multiple steered molecular dynamics (SMD) simulations were performed. The pulling forces of SMD simulations elucidated the regions that form strong attractive interactions in the binding pocket. Structural analysis in these regions showed two important mutations for improving attractive interactions. First, mutation of Tyr102(H) to Ser (sequence numbering of Protein Data Bank entry 1FLR ) played a role in resolving the steric hindrance on the pathway of the antigen in the binding pocket. Second, mutation of Asp31(H) to His played a role in resolving electrostatic repulsion. Potentials of mean force (PMFs) of both the wild type and the mutant showed landscapes that do not include obvious intermediate states and go directly to the bound state. These landscapes were regarded as funnel-like binding free energy landscapes. Furthermore, the structural flexibility based on the fluctuations of the positions of atoms was analyzed. It was shown that the fluctuations in the positions of the antigen and residues in contact with antigen tend to be smaller in the mutant than in the wild type. This result suggested that structural flexibility decreases as affinity is improved by directed evolution. This suggestion is similar to the relationship between affinity and flexibility for in vivo affinity maturation, which was suggested by Romesberg and co-workers [Jimenez, R., et al. (2003) Proc. Natl. Acad. Sci. U.S.A.100, 92-97]. Consequently, the relationship was found to be applicable up to femotomolar affinity levels.

Spacer length effects on in vitro imaging and surface accessibility of fluorescent inhibitors of prostate specific membrane antigen

Prostate-specific membrane antigen (PSMA), a type II transmembrane protein, has been becoming an active target for imaging and therapeutic applications for prostate cancer. Recently, the development of its various chemical inhibitor scaffolds has been explored to serve as carriers for therapeutic or diagnostic payloads targeted to PSMA-positive tumor cells. However, there have been few efforts to definitively determine the optimal length of linker between PSMA inhibitor cores and their payload molecules with regard to the affinity to PSMA and in vitro performance. In our present model study, three spacer-length varied fluorescent inhibitors (FAM-CTT-54, FAM-X-CTT-54 and FAM-PEG(8)-CTT-54) were synthesized, and further enzymatic inhibition studies displayed linker length-dependent changes in: inhibitory potency (IC(50)=0.41 nM, 0.35 nM, 1.93 nM), modes of binding (reversible, slowly reversible, irreversible), respectively. Furthermore, cell-labeling imaging revealed the spacer length-related change of fluorescence intensity (FAM-X-CTT-54>FAM-PEG(8)-CTT-54>FAM-CTT-54). These results suggest that selection of linkers and their lengths will be important considerations in the development of next-generation prostate tumor-targeted imaging probes and therapeutic agents that specifically home to PSMA on tumor cells.

OptCDR: a general computational method for the design of antibody complementarity determining regions for targeted epitope binding

Antibodies are an important class of proteins with many biomedical and biotechnical applications. Although there are a plethora of experimental techniques geared toward their efficient production, there is a paucity of computational methods for their de novo design. OptCDR is a general computational method to design the binding portions of antibodies to have high specificity and affinity against any targeted epitope of an antigen. First, combinations of canonical structures for the antibody complementarity determining regions (CDRs) that are most likely to be able to favorably bind the antigen are selected. This is followed by the simultaneous refinement of the CDR structures' backbones and optimal amino acid selection for each position. OptCDR is applied to three computational test cases: a peptide from the capsid of hepatitis C, the hapten fluorescein and the protein vascular endothelial growth factor. The results demonstrate that OptCDR can efficiently generate diverse antibody libraries of a pre-specified size with promising antigen affinity potential as exemplified by computationally derived binding metrics.

Molecular description of flexibility in an antibody combining site

Mature antibodies (Abs) that are exquisitely specific for virtually any foreign molecule may be produced by affinity maturation of naïve (or germline) Abs. However, the finite number of germline Abs available suggests that, in contrast to mature Abs, germline Abs must be broadly polyspecific so that they are able to recognize a wide range of ligands. Thus, affinity maturation must play a role in mediating Ab specificity. One biophysical property that distinguishes polyspecificity from specificity is protein flexibility; a flexible combining site is able to adopt different conformations that recognize different foreign molecules (or antigens), while a rigid combining site is locked into a conformation that is specific for a given antigen. Recent studies (Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 8821-8826) have examined, at the atomic level, the structural properties that mediate changes in flexibility at four stages of affinity maturation in the 4-4-20 Ab. These studies employed molecular dynamics simulations to reveal a network of residue interactions that mediate the flexibility changes accompanying maturation. The flexibility of the Ab combining sites in these molecular systems was originally measured using three-pulse photon echo spectroscopy (3PEPS). The present investigation extends this work by providing a concrete link between structural properties of the Ab molecules and features of the spectroscopic measurements used to characterize their flexibility. Results obtained from the simulations are in good qualitative agreement with the experimental measurements and indicate that the spectroscopic signal is sensitive to protein dynamics distributed throughout the entire combining site. Thus, the simulations provide a molecular-level interpretation of the changes induced by affinity maturation of the Ab. The results suggest that 3PEPS spectroscopy in combination with molecular dynamics simulations can provide a detailed description of protein dynamics and, in this case, how it is evolved for biological function.

Decreased secretion and unfolded protein response up-regulation are correlated with intracellular retention for single-chain antibody variants produced in yeast

Heterologous protein expression can easily overwhelm a cell's capacity to properly fold protein, initiating the unfolded protein response (UPR), and resulting in a loss of protein expression. In the current model of the UPR, the chaperone BiP modulates the activation of the UPR due to its interactions with the signaling protein Ire1p and newly synthesized proteins. In this research, 4-4-20 scFv variants were generated by rational design to alter BiP binding to newly synthesized scFv proteins or via directed evolution aimed at improved secretion. Interestingly, the predicted BiP binding ability did not correlate significantly with the UPR. However, pulse-chase analysis of scFv fate revealed that mutants with a decreased ER residence time were more highly secreted, indicating that improved protein folding was more likely the cause for improved secretion. In fact, decreased secretion correlated with increased binding by BiP, as determined by co-immune precipitation studies. This suggests that the algorithm is not useful for in vivo prediction of variants, and that in vivo screens are more effective for finding variants with improved properties.

Application of the PM6 semi-empirical method to modeling proteins enhances docking accuracy of AutoDock

BACKGROUND

Molecular docking methods are commonly used for predicting binding modes and energies of ligands to proteins. For accurate complex geometry and binding energy estimation, an appropriate method for calculating partial charges is essential. AutoDockTools software, the interface for preparing input files for one of the most widely used docking programs AutoDock 4, utilizes the Gasteiger partial charge calculation method for both protein and ligand charge calculation. However, it has already been shown that more accurate partial charge calculation - and as a consequence, more accurate docking- can be achieved by using quantum chemical methods. For docking calculations quantum chemical partial charge calculation as a routine was only used for ligands so far. The newly developed Mozyme function of MOPAC2009 allows fast partial charge calculation of proteins by quantum mechanical semi-empirical methods. Thus, in the current study, the effect of semi-empirical quantum-mechanical partial charge calculation on docking accuracy could be investigated.

RESULTS

The docking accuracy of AutoDock 4 using the original AutoDock scoring function was investigated on a set of 53 protein ligand complexes using Gasteiger and PM6 partial charge calculation methods. This has enabled us to compare the effect of the partial charge calculation method on docking accuracy utilizing AutoDock 4 software. Our results showed that the docking accuracy in regard to complex geometry (docking result defined as accurate when the RMSD of the first rank docking result complex is within 2 A of the experimentally determined X-ray structure) significantly increased when partial charges of the ligands and proteins were calculated with the semi-empirical PM6 method. Out of the 53 complexes analyzed in the course of our study, the geometry of 42 complexes were accurately calculated using PM6 partial charges, while the use of Gasteiger charges resulted in only 28 accurate geometries. The binding affinity estimation was not influenced by the partial charge calculation method - for more accurate binding affinity prediction development of a new scoring function for AutoDock is needed.

CONCLUSION

Our results demonstrate that the accuracy of determination of complex geometry using AutoDock 4 for docking calculation greatly increases with the use of quantum chemical partial charge calculation on both the ligands and proteins.

Exploring the energy landscape of antibody-antigen complexes: protein dynamics, flexibility, and molecular recognition

The production of antibodies that selectively bind virtually any foreign compound is the hallmark of the immune system. While much is understood about how sequence diversity contributes to this remarkable feat of molecular recognition, little is known about how sequence diversity impacts antibody dynamics, which is also expected to contribute to molecular recognition. Toward this goal, we examined a panel of antibodies elicited to the chromophoric antigen fluorescein. On the basis of isothermal titration calorimetry, we selected six antibodies that bind fluorescein with diverse binding entropies, suggestive of varying contributions of dynamics to molecular recognition. Sequencing revealed that two pairs of antibodies employ homologous heavy chains that were derived from common germline genes, while the other two heavy chains and all six of the light chains were derived from different germline genes and are not homologous. Interestingly, more than half of all the somatic mutations acquired during affinity maturation among the six antibodies are located in positions unlikely to contact fluorescein directly. To quantify and compare the dynamics of the antibody-fluorescein complexes, three-pulse photon echo peak shift and transient grating spectroscopy were employed. All of the antibodies exhibited motions on three distinct time scales, ultrafast motions on the <100 fs time scale, diffusive motions on the picosecond time scale, and motions that occur on time scales longer than nanoseconds and thus appear static. However, the exact frequency of the picosecond time scale motion and the relative contribution of the different motions vary significantly among the antibody-chromophore complexes, revealing a high level of dynamic diversity. Using a hierarchical model, we relate the data to features of the antibodies' energy landscapes as well as their flexibility in terms of elasticity and plasticity. In all, the data provide a consistent picture of antibody flexibility, which interestingly appears to be correlated with binding entropy as well as with germline gene use and the mutations introduced during affinity maturation. The data also provide a gauge of the dynamic diversity of the antibody repertoire and suggest that this diversity might contribute to molecular recognition by facilitating the recognition of the broadest range of foreign molecules.

Dissecting NGF interactions with TrkA and p75 receptors by structural and functional studies of an anti-NGF neutralizing antibody

The anti-nerve growth factor (NGF) monoclonal antibody alphaD11 is a potent antagonist that neutralizes the biological functions of its antigen in vivo. NGF antagonism is expected to be a highly effective and safe therapeutic approach in many pain states. A comprehensive functional and structural analysis of alphaD11 monoclonal antibody was carried out, showing its ability to neutralize NGF binding to either tropomyosine receptor kinase A (TrkA) or p75 receptors. The 3-D structure of the alphaD11 Fab fragment was solved at 1.7 A resolution. A computational docking model of the alphaD11 Fab-NGF complex, based on epitope mapping using a pool of 44 NGF mutants and experimentally validated by small-angle X-ray scattering, provided the structural basis for identifying the residues involved in alphaD11 Fab binding. The present study pinpoints loop II of NGF to be an important structural determinant for NGF biological activity mediated by TrkA receptor.

Molecular evolution of affinity and flexibility in the immune system

The immune system responds to the introduction of foreign antigens by rapidly evolving antibodies with increasing affinity for the antigen (i.e., maturation). To investigate the factors that control this process at the molecular level, we have assessed the changes in flexibility that accompany ligand binding at four stages of maturation in the 4-4-20 antibody. Our studies, based on molecular dynamics, indicate that increased affinity for the target ligand is associated with a decreased entropic cost to binding. The entropy of binding is unfavorable, opposing favorable enthalpic contributions that arise during complex formation. Computed binding free energies for the various antibody-ligand complexes qualitatively reproduce the trends observed in the experimentally derived values, although the absolute magnitude of free-energy differences is overestimated. Our results support the existence of a correlation between high-affinity interactions and decreased protein flexibility in this series of antibody molecules. This observation is likely to be a general feature of molecular association processes and key to the molecular evolution of antibody responses.

The consequences of scoring docked ligand conformations using free energy correlations

Ligands from a set of 19 protein-ligand complexes were re-docked with AutoDock, GOLD and FlexX using the scoring algorithms native to these programs supplemented by analysis using the HINT free energy force field. A HINT scoring function was calibrated for this data set using a simple linear regression of total HINT score for crystal-structure complexes vs. measured free energy of binding. This function had an r(2) of 0.84 and a standard error of +/-0.42 kcal mol(-1). The free energies of binding were calculated for the best poses using the AutoDock, GOLD and FlexX scoring functions. The AutoDock and GoldScore algorithms estimated more than half of the binding free energies within the reported calibration standard errors for these functions, while that of FlexX did not. In contrast, the calibrated HINT scoring function identified optimized poses with standard errors near +/-0.5 kcal mol(-1). When the metric of success is minimum RMSD (vs. crystallographic coordinates) the three docking programs were more successful, with mean RMSDs for the top-ranking poses in the 19 complexes of 3.38, 2.52 and 2.62 A for AutoDock, GOLD and FlexX, respectively. Two key observations in this study have general relevance for computational medicinal chemistry: first, while optimizing RMSD with docking score functions is clearly of value, these functions may be less well optimized for free energy of binding, which has broader applicability in virtual screening and drug discovery than RMSD; second, scoring functions uniquely calibrated for the data set or sets under study should nearly always be preferable to universal scoring functions. Due to these advantages, the poses selected by the HINT score also required less post-docking structure optimization to produce usable molecular models. Most of these features may be achievable with other scoring functions.

AID-initiated purposeful mutations in immunoglobulin genes

Exposure brings risk to all living organisms. Using a remarkably effective strategy, higher vertebrates mitigate risk by mounting a complex and sophisticated immune response to counter the potentially toxic invasion by a virtually limitless army of chemical and biological antagonists. Mutations are almost always deleterious, but in the case of antibody diversification there are mutations occurring at hugely elevated rates within the variable (V) and switch regions (SR) of the immunoglobulin (Ig) genes that are responsible for binding to and neutralizing foreign antigens throughout the body. These mutations are truly purposeful. This chapter is centered on activation-induced cytidine deaminase (AID). AID is required for initiating somatic hypermutation (SHM) in the V regions and class switch recombination (CSR) in the SR portions of Ig genes. By converting C --> U, while transcription takes place, AID instigates a cascade of mutational events involving error-prone DNA polymerases, base excision and mismatch repair enzymes, and recombination pathways. Together, these processes culminate in highly mutated antibody genes and the B cells expressing antibodies that have achieved optimal antigenic binding undergo positive selection in germinal centers. We will discuss the biological role of AID in this complex process, primarily in terms of its biochemical properties in relation to SHM in vivo. The chapter also discusses recent advances in experimental methods to characterize antibody dynamics as a function of SHM to help elucidate the role that the AID-induced mutations play in tailoring molecular recognition. The emerging experimental techniques help to address long-standing conundrums concerning evolution-imposed constraints on antibody structure and function.

In vitro evolution of single-chain antibodies using mRNA display

Here we describe the application of the in vitro virus mRNA display method, which involves covalent linkage of an in vitro-synthesized antibody (phenotype) to its encoding mRNA (genotype) through puromycin, for in vitro evolution of single-chain Fv (scFv) antibody fragments. To establish the validity of this approach to directed antibody evolution, we used random mutagenesis by error-prone DNA shuffling and off-rate selection to improve the affinity of an anti-fluorescein scFv as a model system. After four rounds of selection of the library of mRNA-displayed scFv mutants, we obtained six different sequences encoding affinity-matured mutants with five consensus mutations. Kinetic analysis of the mutant scFvs revealed that the off-rates have been decreased by more than one order of magnitude and the dissociation constants were improved approximately 30-fold. The antigen-specificity was not improved by affinity maturation, but remained similar to that of the wild type. Although the five consensus mutations of the high-affinity mutants were scattered over the scFv sequence, analysis by site-directed mutagenesis demonstrated that the critical mutations for improving affinity were the two that lay within the complementarity determining regions (CDRs). Thus, mRNA display is expected to be useful for rapid artificial evolution of high-affinity diagnostic and therapeutic antibodies by optimizing their CDRs.

Antibody evolution constrains conformational heterogeneity by tailoring protein dynamics

The evolution of proteins with novel function is thought to start from precursor proteins that are conformationally heterogeneous. The corresponding genes may be duplicated and then mutated to select and optimize a specific conformation. However, testing this idea has been difficult because of the challenge of quantifying protein flexibility and conformational heterogeneity as a function of evolution. Here, we report the characterization of protein heterogeneity and dynamics as a function of evolution for the antifluorescein antibody 4-4-20. Using nonlinear laser spectroscopy, surface plasmon resonance, and molecular dynamics simulations, we demonstrate that evolution localized the Ab-combining site from a heterogeneous ensemble of conformations to a single conformation by introducing mutations that act cooperatively and over significant distances to rigidify the protein. This study demonstrates how protein dynamics may be tailored by evolution and has important implications for our understanding of how novel protein functions are evolved.

Distance versus energy fluctuations and electron transfer in single protein molecules

Stochastic nature due to distance and energy fluctuations of single protein molecules involved in electron-transfer (ET) reactions is studied. Distance fluctuations have been assumed previously for causing the slow fluctuations in the ET rates between a donor-acceptor pair constrained to a native protein. Although the observed t(-1/2) power law can be derived using Langevin dynamics with a simple chain model, some discrepancies exist. The friction coefficient and the Rouse segment time constant deduced from experimental data are several orders of magnitude too large, even though the extracted force constant is reasonable. Therefore, questions are raised about the distance-fluctuation mechanism and the activationless ET hypothesis. As an alternative mechanism, we considered fluctuations in activation energy and analyzed the data from two different single protein experiments to determine spectral distribution of energy fluctuations.

Context-dependent mutations predominate in an engineered high-affinity single chain antibody fragment

A mutational analysis of the femtomolar-affinity anti-fluorescein antibody 4M5.3, compared to its wild-type progenitor, 4-4-20, indicates both context-dependent and -independent mutations are responsible for the 1800-fold affinity improvement. 4M5.3 was engineered from 4-4-20 by directed evolution and contains 14 mutations. The seven mutations identified as present in each of 10 final round affinity maturation clones were studied here. Affinities of the 4-4-20 single mutant addition and 4M5.3 single site reversion mutants were compared. These experiments identified four mutations, of these seven, that were context-dependent in their contribution to higher affinity. A simplified mutant containing only these seven mutations was created to analyze complete double mutant cycles of selected sets of mutations. Specific mutational sets studied included the ligand contact mutations, the heavy chain CDR3 mutations, the heavy chain CDR3 mutations plus the neighboring residue at site H108, and the early and late acquired mutations on the directed evolution pathway. The heavy chain CDR3 mutational set and the ligand-contacting mutations were shown to provide -1.4 and -2.0 kcal/mol, respectively, of the total -3.5 kcal/mol change in free energy of binding of the seven-site consensus mutant. The mutations acquired late in the directed evolution rounds provided much of the change in free energy without the earlier acquired mutations (-3.1 kcal/mol of the total -3.5 kcal/mol). Prior structural data and electrostatic calculations presented several hypotheses for the higher affinity contributions, some of which are supported by these mutational data.

A mutation designed to alter crystal packing permits structural analysis of a tight-binding fluorescein-scFv complex

The structure of the scFv fragment FITC-E2, obtained from a naive phage antibody scFv library derived from human donors, was determined at 2.1 A resolution in the free form and at 3.0 A in the complexed form. The wild-type (wt) scFv binds fluorescein with a K(D) of 0.75 nM. The free scFv readily crystallizes by compacting its 18 amino acid-long CDR-H3, partially occluding the binding site and further blocking access by binding to the "bottom" of a neighboring scFv molecule with a cluster of exposed aromatic residues within CDR-H3. Only upon mutating one of the residues involved in this dominant crystal contact, an exposed tryptophan in the middle of CDR-H3, crystals of the complex could be obtained. A series of alanine mutants within the putative antigen binding site, covering a range of binding affinities, were used to relate macroscopic thermodynamic and kinetic binding parameters to single-molecule disruption forces measured by AFM. The effects of the mutations on the binding properties, particularly on the fraction of binding-competent molecules within the population, cannot be fully explained by changes in the strength of local interactions. The significant conformational change of CDR-H3 between the free and the liganded form illustrates the plasticity of the binding site. An accompanying study in this issue by Curcio and colleagues presents the molecular dynamics simulation of the forced unbinding experiments and explores possible effects of the mutations on the unbinding pathway of the hapten.

Fluctuating enzymes: lessons from single-molecule studies

Recent single-molecule enzymology measurements with improved statistics have demonstrated that a single enzyme molecule exhibits large temporal fluctuations of the turnover rate constant at a broad range of time scales (from 1 ms to 100 s). The rate constant fluctuations, termed as dynamic disorder, are associated with fluctuations of the protein conformations observed on the same time scales. We discuss the unique information extractable from these experiments and the reconciliation of these observations with ensemble-averaged Michaelis-Menten equation. A theoretical model based on the generalized Langevin equation (GLE) treatment of Kramers' barrier crossing problem for chemical reactions accounts naturally for the observation of dynamic disorder and highly dispersed kinetics.

Observation of a power-law memory kernel for fluctuations within a single protein molecule

The fluctuation of the distance between a fluorescein-tyrosine pair within a single protein complex was directly monitored in real time by photoinduced electron transfer and found to be a stationary, time-reversible, and non-Markovian Gaussian process. Within the generalized Langevin equation formalism, we experimentally determine the memory kernel K(t), which is proportional to the autocorrelation function of the random fluctuating force. K(t) is a power-law decay, t(-0.51 +/- 0.07) in a broad range of time scales (10(-3)-10 s). Such a long-time memory effect could have implications for protein functions.

Structural Studies of a Stabilized Phosphoenzyme Intermediate of Ca2+-ATPase

Ca(2+)-ATPase belongs to the family of P-type ATPases and maintains low concentrations of intracellular Ca(2+). Its reaction cycle consists of four main intermediates that alternate ion binding in the transmembrane domain with phosphorylation of an aspartate residue in a cytoplasmic domain. Previous work characterized an ultrastable phosphoenzyme produced first by labeling with fluorescein isothiocyanate, then by allowing this labeled enzyme to establish a maximal Ca(2+) gradient, and finally by removing Ca(2+) from the solution. This phosphoenzyme is characterized by very low fluorescence and has specific enzymatic properties suggesting the existence of a high energy phosphoryl bond. To study the structural properties of this phosphoenzyme, we used cryoelectron microscopy of two-dimensional crystals formed in the presence of decavanadate and determined the structure at 8-A resolution. To our surprise we found that at this resolution the low fluorescence phosphoenzyme had a structure similar to that of the native enzyme crystallized under equivalent conditions. We went on to use glutaraldehyde cross-linking and proteolysis for independent structural assessment and concluded that, like the unphosphorylated native enzyme, Ca(2+) and vanadate exert a strong influence over the global structure of this low fluorescence phosphoenzyme. Based on a structural model with fluorescein isothiocyanate bound at the ATP site, we suggest that the stability as well as the low fluorescence of this phosphoenzyme is due to a fluorescein-mediated cross-link between two cytoplasmic domains that prevents hydrolysis of the aspartyl phosphate. Finally, we consider the alternative possibility that phosphate transfer to fluorescein itself could explain the properties of this low fluorescence species.

Macromolecular accessibility of fluorescent taxoids bound at a paclitaxel binding site in the microtubule surface

The macromolecular accessibility of the paclitaxel binding site in microtubules has been investigated using a fluorescent taxoid and antibodies against fluorescein, which cannot diffuse into the microtubule lumen. The formation of a specific ternary complex of microtubules, Hexaflutax (7-O-{N-[6-(fluorescein-4'-carboxamido)-n-hexanoyl]-l-alanyl}paclitaxel) and 4-4-20 IgG (a monoclonal antibody against fluorescein) has been observed by means of sedimentation and electron microscopy methods. The kinetics of binding of the antibody to microtubule-bound Hexaflutax has been measured. The quenching of the observed fluorescence is fast (k+ 2.26 +/- 0.25 x 10(6) m(-1) s(-1) at 37 degrees C), indicating that the fluorescein groups of Hexaflutax are exposed to the outer solvent. The velocity of the reaction is linearly dependent on the antibody concentration, indicating that a bimolecular reaction is being observed. Another fluorescent taxoid (Flutax-2) bound to microtubules has also been shown to be rapidly accessible to polyclonal antibodies directed against fluorescein. A reduced rate of Hexaflutax quenching by the antibody is observed in microtubule-associated proteins containing microtubules or in native cellular cytoskeletons. It can be concluded that the fluorescent taxoids bind to an outer site on the microtubules that is shared with paclitaxel. Paclitaxel would be internalized in a further step of binding to reach the known luminal site, this step being blocked in the case of the fluorescent taxoids. Because the fluorescent ligands are able to induce microtubule assembly, binding to the outer site should be enough to induce assembly by a preferential binding mechanism.

Substantial Energetic Improvement with Minimal Structural Perturbation in a High Affinity Mutant Antibody

Here, we compare an antibody with the highest known engineered affinity (K(d)=270 fM) to its high affinity wild-type (K(d)=700 pM) through thermodynamic, kinetic, structural, and theoretical analyses. The 4M5.3 anti-fluorescein single chain antibody fragment (scFv) contains 14 mutations from the wild-type 4-4-20 scFv and has a 1800-fold increase in fluorescein-binding affinity. The dissociation rate is approximately 16,000 times slower in the mutant; however, this substantial improvement is offset somewhat by the association rate, which is ninefold slower in the mutant. Enthalpic contributions to binding were found by calorimetry to predominate in the differential binding free energy. The crystal structure of the 4M5.3 mutant complexed with antigen was solved to 1.5A resolution and compared with a previously solved structure of an antigen-bound 4-4-20 Fab fragment. Strikingly, the structural comparison shows little difference between the two scFv molecules (backbone RMSD of 0.6A), despite the large difference in affinity. Shape complementarity exhibits a small improvement between the variable light chain and variable heavy chain domains within the antibody, but no significant improvement in shape complementarity of the antibody with the antigen is observed in the mutant over the wild-type. Theoretical modeling calculations show electrostatic contributions to binding account for -1.2 kcal/mol to -3.5 kcal/mol of the binding free energy change, of which -1.1 kcal/mol is directly associated with the mutated residue side-chains. The electrostatic analysis reveals several mechanistic explanations for a portion of the improvement. Collectively, these data provide an example where very high binding affinity is achieved through the cumulative effect of many small structural alterations.

Protein dynamics and the immunological evolution of molecular recognition

While it is accepted that protein flexibility plays a role in protein folding, catalysis, and molecular recognition, few techniques are capable of the rigorous measurement of protein motions required to quantify flexibility. Three-pulse photon echo shift spectroscopy can be used to measure the time scale of protein motions, and we have used this technique, along with steady-state spectroscopy and binding and structural data, to examine the immunological evolution of protein flexibility in an anti-fluorescein antibody. Two light chain somatic mutations increase affinity for fluorescein by 12-fold but also significantly affect flexibility. Specifically, a rigidification of the protein is seen in each of three observable motions; two slower motions undergo decreased amplitudes of displacement, by 3- and 20-fold, respectively, in response to an applied force, and the distribution associated with the amplitude of a faster motion is narrowed upon somatic mutation. The somatic mutations appear to rigidify the antibody-fluorescein complex by more strongly anchoring fluorescein to the protein and by more tightly packing the complex. The data demonstrate that in addition to affinity, antibody dynamics are systematically manipulated during affinity maturation, and they imply that the evolution of protein flexibility may be a central component of the immune response. The results also reflect the type of protein rigidification that may be important for other biological interactions, such as protein-protein, protein-ligand or protein-drug, and enzyme-substrate recognition.

Alteration of amino acid residues at the L-chain N-terminus and in complementarity-determining region 3 increases affinity of a recombinant F(ab) for the human N blood group antigen

BACKGROUND

To examine the fine specificity of glycopeptide-specific antibodies, this study focused on the human MN blood group system. F(ab) phage display methods were previously used to construct an F(ab) family in which the H-chain Fd fragment was held constant whereas the L chains were "shuffled." This yielded two related F(ab), N92 and NNA7, with low and high affinity for N, respectively. Although their L-chain sequences are very similar, sharing 92 percent amino acid identity, there are intriguing differences at the N-terminus and in complementarity-determining region 3 (CDR3) at positions 89, 91, 92, and 96.

STUDY DESIGN AND METHODS

Site-directed mutagenesis, ELISA, and hemagglutination were used to examine the contributions of these variations to antibody affinity.

RESULTS

Studies with the N92-S91G and NNA7-G91S mutants demonstrated that the Gly at position 91 was critically important for ensuring high affinity. Indeed, the affinity of N92-S91G was almost as high as N92TM, in which all four CDR3 residues were changed to match NNA7. N-terminal L-chain differences were surprisingly important in determining affinity. For example, when the N-terminus of N92 was changed to match that of NNA7, affinity increased approximately 30-fold.

CONCLUSION

Specific residues at the L-chain N-terminus and in CDR3 significantly affected F(ab) affinity for N. Future structural studies of these F(ab), alone and complexed with this glycopeptide antigen, will provide further insights into these phenomena.

Modeling the binding sites of anti-hen egg white lysozyme antibodies HyHEL-8 and HyHEL-26: an insight into the molecular basis of antibody cross-reactivity and specificity

Three antibodies, HyHEL-8 (HH8), HyHEL-10 (HH10), and HyHEL-26 (HH26) are specific for the same epitope on hen egg white lysozyme (HEL), and share >90% sequence homology. Their affinities vary by several orders of magnitude, and among the three antibodies, HH8 is the most cross-reactive with kinetics of binding that are relatively invariable compared to HH26, which is highly specific and has quite variable kinetics. To investigate structural correlates of these functional variations, the Fv regions of HH8 and HH26 were homology-modeled using the x-ray structure of the well-characterized HH10-HEL complex as template. The binding site of HH26 is most charged, least hydrophobic, and has the greatest number of intramolecular salt bridges, whereas that of HH8 is the least charged, most hydrophobic and has the fewest intramolecular salt bridges. The modeled HH26-HEL structure predicts the recently determined x-ray structure of HH26, (Li et al., 2003, Nat. Struct. Biol. 10:482-488) with a root-mean-square deviation of 1.03 A. It is likely that the binding site of HH26 is rendered rigid by a network of intramolecular salt bridges whereas that of HH8 is flexible due to their absence. HH26 also has the most intermolecular contacts with the antigen whereas HH8 has the least. HH10 has these properties intermediate to HH8 and HH26. The structurally rigid binding site with numerous specific contacts bestows specificity on HH26 whereas the flexible binding site with correspondingly fewer contacts enables HH8 to be cross-reactive. Results suggest that affinity maturation may select for high affinity antibodies with either "lock-and-key" preconfigured binding sites, or "preconfigured flexibility" by modulating combining site flexibility.

Crystallographic analysis of an "anticalin" with tailored specificity for fluorescein reveals high structural plasticity of the lipocalin loop region

The artificial lipocalin FluA with novel specificity toward fluorescein was derived via combinatorial engineering from the bilin-binding protein, BBP by exchange of 16 amino acids in the ligand pocket. Here, we describe the crystal structure of FluA at 2.0 A resolution in the space group P2(1) with two protein-ligand complexes in the asymmetric unit. In both molecules, the characteristic beta-barrel architecture with the attached alpha-helix is well preserved. In contrast, the four loops at one end of the beta-barrel that form the entrance to the binding site exhibit large conformational deviations from the wild-type protein, which can be attributed to the sidechain replacements. Specificity for the new ligand is furnished by hydrophobic packing, charged sidechain environment, and hydrogen bonds with its hydroxyl groups. Unexpectedly, fluorescein is bound in a much deeper cavity than biliverdin IX(gamma) in the natural lipocalin. Triggered by the substituted residues, unmutated sidechains at the bottom of the binding site adopt conformations that are quite different from those observed in the BBP, illustrating that not only the loop region but also the hydrophobic interior of the beta-barrel can be reshaped for molecular recognition. Particularly, Trp 129 participates in a tight stacking interaction with the xanthenolone moiety, which may explain the ultrafast electron transfer that occurs on light excitation of the bound fluorescein. These structural findings support our concept of using lipocalins as a scaffold for the engineering of so-called "anticalins" directed against prescribed targets as an alternative to recombinant antibody fragments.

Flexibility and molecular recognition in the immune system

Photon echo spectroscopy has been used to measure the response of three antibody-binding sites to perturbation from electronic excitation of a bound antigen, fluorescein. The three antibodies show motions that range in time scale from tens of femtoseconds to nanoseconds. Relative to the others, one antibody, 4-4-20, possesses a rigid binding site that likely results from a short and inflexible heavy chain complementarity-determining region 3 (HCDR3) loop and a critical Tyr that acts as a "molecular splint," rigidifying the antigen across its most flexible internal degree of freedom. The remaining two antibodies, 34F10 and 40G4, despite being generated against the same antigen, possess binding sites that are considerably more flexible. The more flexible combining sites likely result from longer HCDR3 loops and a deletion in the light chain complementarity-determining region 1 (LCDR1) that removes the critical Tyr residue. The binding site flexibilities may result in varying mechanisms of antigen recognition including lock-and-key, induced-fit, and conformational selection.

Forces and energetics of hapten-antibody dissociation: a biased molecular dynamics simulation study

The unbinding of fluorescein from the single-chain Fv fragment of the 4D5Flu antibody is investigated by biased molecular dynamics with an implicit solvation model. To obtain statistically meaningful results, a large number of unbinding trajectories are calculated; they involve a total simulation time of more than 200 ns. Simulations are carried out with a time-dependent perturbation and in the presence of a constant force. The two techniques, which provide complementary information, induce unbinding by favoring an increase in the distance between the ligand and the antibody. This distance is an appropriate progress variable for the dissociation reaction and permits direct comparison of the unbinding forces in the simulations with data from atomic force microscopy (AFM). The time-dependent perturbation generates unfolding pathways that are close to equilibrium and can be used to reconstruct the mean force; i.e. the derivative of the potential of mean force, along the reaction coordinate. This is supported by an analysis of the overall unbinding profile and the magnitude of the mean force, which are similar to those of the unbinding force (i.e. the external force due to the time-dependent perturbation) averaged over several unbinding events. The multiple simulations show that unbinding proceeds along a rather well-defined pathway for a broad range of effective pulling speeds. Initially, there is a distortion of the protein localized in the C-terminal region followed by the fluorescein exit from the binding site. This occurs in steps that involve breaking of specific electrostatic and van der Waals interactions. It appears that the simulations do not explore the same barriers as those measured in the AFM experiments because of the much higher unfolding speed in the former. The dependence of the force on the logarithm of the loading rate is linear and the slope is higher than in the AFM, in agreement with experiment in other systems, where different slopes were observed for different regimes. Based on the unbinding events, mutations in the 4D5Flu antigen binding site are predicted to result in significant changes in the unbinding force.

The influence of the buried glutamine or glutamate residue in position 6 on the structure of immunoglobulin variable domains

Immunoglobulin V(H) domain frameworks can be grouped into four distinct types, depending on the main-chain conformation of framework 1. Based on the analysis of over 200 X-ray structures representing more than 100 non-redundant V(H) domain sequences, we have come to the conclusion that the marked structural variability of the V(H) framework 1 region is caused by three residues: the buried side-chain of H6, which can be either a glutamate or a glutamine residue, the residue in position H7, which may be proline only if H6 is glutamine, and by H9 (H10 according to a new consensus nomenclature), which has to be either glycine or proline if H6 is a glutamate residue. In natural antibodies, these three residues are encoded in combinations that are compatible with each other and with the rest of the structure and therefore will yield functional molecules. However, the degenerate primer mixtures commonly used for PCR cloning of antibody fragments can and frequently do introduce out-of-context mutations to combinations that can lead to severe reduction of stability, production yield and antigen affinity.

Yet another numbering scheme for immunoglobulin variable domains: an automatic modeling and analysis tool

A common residue numbering scheme for all immunoglobulin variable domains (immunoglobulin light chain lambda (V(lambda)) and kappa (V(kappa)) variable domains, heavy chain variable domains (V(H)) and T-cell receptor alpha (V(alpha)), beta (V(beta)), gamma (V(gamma)) and delta (V(delta)) variable domains) has been devised. Based on the spatial alignment of known three-dimensional structures of immunoglobulin domains, it places the alignment gaps in a way that minimizes the average deviation from the averaged structure of the aligned domains. This residue numbering scheme was applied to the immunoglobulin variable domain structures in the PDB database to automate the extraction of information on structural variations in homologous positions of the different molecules. A number of methods are presented that allow the automated projection of information derived from individual structures or from the comparison of multi-structure alignments onto a graphical representation of the sequence alignment.

'Anticalins': a new class of engineered ligand-binding proteins with antibody-like properties

The development of soluble receptor proteins that recognise given target molecules--ranging from small chemical compounds to macromolecular structures at a cell surface, for example--is of ever increasing importance in the life sciences and biotechnology. For the past century this area of application was dominated by antibodies, which were traditionally generated via immunisation of animals but have recently also become available by means of protein engineering methods. The so-called 'anticalins' offer an alternative type of ligand-binding proteins, which has been constructed on the basis of lipocalins as a scaffold. The central element of this protein architecture is a beta-barrel structure of eight antiparallel strands, which supports four loops at its open end. These loops form the natural binding site of the lipocalins and can be reshaped in vitro by extensive amino acid replacement, thus creating novel binding specificities. The bilin-binding protein (BBP) was employed as a model system for the preparation of a random library with 16 selectively mutagenized residues. Using bacterial phagemid display and colony screening techniques, several lipocalin variants--termed anticalins--have been selected from this library, exhibiting binding activity for compounds like fluorescein or digoxigenin. Anticalins possess high affinity and specificity for their prescribed ligands as well as fast binding kinetics, so that their functional properties are similar to those of antibodies. Compared with them, they exhibit however several advantages, including a smaller size, composition of a single polypeptide chain, and a simple set of four hypervariable loops that can be easily manipulated at the genetic level. Apart from haptenic compounds as targets, anticalins should also be able to recognise macromolecular antigens, provided that the random library is accordingly designed. Hence, they should not only serve as valuable reagents for bioanalytical purposes, but may also have a potential in replacing antibodies for medical therapy.

Covariance analysis of protein families: the case of the variable domains of antibodies

A nonrestrictive method for identifying covariance in protein families is described and applied to human and mouse germline Vkappa and VH sequence alignments. Amino acids that occur at each position in a sequence alignment are divided into two sets, called a word, by generating all possible combinations of alternative amino acids. Each word is associated with a pattern of changes. Words with identical patterns identify covariant positions. In antibody variable domains, the number of words generated ranged between 1103 and 2195 depending on the alignment, of which 4 to 12 % occurred in covariant pairs. Despite the nonrestrictive character of pattern generation, covariant residues did not reflect a random selection with respect to the nature of amino acid changes and/or their spatial proximity in a reference crystallographic structure. This approach allowed the identification of a covariance signal for positions with high variability, mostly located in the outer part of the common structural framework of antibody variable domains. Covariance in these regions may reflect the existence of alternative and mutually exclusive atomic arrangements that are compatible with antibody function. The method may be of general applicability to rationalize residue variability in protein families.

Lipocalins as a scaffold

The concept of scaffolds that can be equipped with artificial biochemically active sites has gained recent interest in the field of protein design. Members of the lipocalin protein family represent promising model systems in this respect. Especially prototypic lipocalins, such as the retinol-binding protein or the bilin-binding protein (BBP), exhibit a structurally simple one-domain fold with a conformationally well conserved beta-barrel as their central motif. This type of supersecondary structure is made of a cylindrically closed beta-sheet of eight antiparallel strands. At the open end of the barrel the beta-strands are connected by four loops in a pairwise manner so that a pocket for the ligand is formed. In a rational protein design study a metal-binding site was functionally grafted on the solvent-exposed surface of the beta-barrel, whereby the rigid backbone conformation permitted the spatially defined arrangement of three His side chains. In a combinatorial protein design approach, the natural ligand pocket of a lipocalin was reshaped. In this manner variants of the BBP were engineered which exhibit high affinity and remarkable specificity for haptens like fluorescein and digoxigenin. The so-called 'anticalins', i.e. artificial lipocalins recognizing prescribed ligands, could provide an interesting alternative to recombinant antibody fragments. Consequently, the use of lipocalins as a scaffold opens new applications for members of this functionally diverse protein family in biotechnology and medicine.

Directed evolution of antibody fragments with monovalent femtomolar antigen-binding affinity

Single-chain antibody mutants have been evolved in vitro with antigen-binding equilibrium dissociation constant K(d) = 48 fM and slower dissociation kinetics (half-time > 5 days) than those for the streptavidin-biotin complex. These mutants possess the highest monovalent ligand-binding affinity yet reported for an engineered protein by over two orders of magnitude. Optimal kinetic screening of randomly mutagenized libraries of 10(5)-10(7) yeast surface-displayed antibodies enabled a >1,000-fold decrease in the rate of dissociation after four cycles of affinity mutagenesis and screening. The consensus mutations are generally nonconservative by comparison with naturally occurring mouse Fv sequences and with residues that do not contact the fluorescein antigen in the wild-type complex. The existence of these mutants demonstrates that the antibody Fv architecture is not intrinsically responsible for an antigen-binding affinity ceiling during in vivo affinity maturation.

Unbinding forces of single antibody-antigen complexes correlate with their thermal dissociation rates

Point mutants of three unrelated antifluorescein antibodies were constructed to obtain nine different single-chain Fv fragments, whose on-rates, off-rates, and equilibrium binding affinities were determined in solution. Additionally, activation energies for unbinding were estimated from the temperature dependence of the off-rate in solution. Loading rate-dependent unbinding forces were determined for single molecules by atomic force microscopy, which extrapolated at zero force to a value close to the off-rate measured in solution, without any indication for multiple transition states. The measured unbinding forces of all nine mutants correlated well with the off-rate in solution, but not with the temperature dependence of the reaction, indicating that the same transition state must be crossed in spontaneous and forced unbinding and that the unbinding path under load cannot be too different from the one at zero force. The distance of the transition state from the ground state along the unbinding pathway is directly proportional to the barrier height, regardless of the details of the binding site, which most likely reflects the elasticity of the protein in the unbinding process. Atomic force microscopy thus can be a valuable tool for the characterization of solution properties of protein-ligand systems at the single molecule level, predicting relative off-rates, potentially of great value for combinatorial chemistry and biology.

Anticalins versus antibodies: made-to-order binding proteins for small molecules

Engineering proteins to bind small molecules presents a challenge as daunting as drug discovery, for both hinge upon our understanding of receptor-ligand molecular recognition. However, powerful techniques from combinatorial molecular biology can be used to rapidly select artificial receptors. While traditionally researchers have relied upon antibody technologies as a source of new binding proteins, the lipocalin scaffold has recently emerged as an adaptable receptor for small molecule binding. 'Anticalins', engineered lipocalin variants, offer some advantages over traditional antibody technology and illuminate features of molecular recognition between receptors and small molecule ligands.

Crystal structure of the 64M-2 antibody Fab fragment in complex with a DNA dT(6-4)T photoproduct formed by ultraviolet radiation

DNA photoproducts with (6-4) pyrimidine-pyrimidone adducts formed by ultraviolet radiation are implicated in mutagenesis and cancer, particularly skin cancer. The crystal structure of the Fab fragment of the murine 64M-2 antibody specific to DNA T(6-4)T photoproducts is determined as a complex with dT(6-4)T, a (6-4) pyrimidine-pyrimidone photodimer of dTpT, at 2.4 A resolution to a crystallographic R-factor of 0.199 and an R(free) value of 0.279. The 64M-2 Fab molecule is in an extended arrangement with an elbow angle of 174 degrees, and its five complementarity-determining regions, except L2, are involved in the ligand binding. The bound dT(6-4)T ligand adopting a ring structure with (6-4) linked 5' thymine-3' pyrimidone bases is fully accommodated in an antigen-binding pocket of about 15 Ax10 A. The 5'-thymine and 3'-pyrimidone bases are in half-chair and planar conformations, respectively, and are nearly perpendicular to each other. The 5'-thymine base is hydrogen-bonded to Arg95H and Ser96H, and is in van der Waals contact with Tyr100iH. The 3'-pyrimidone base is hydrogen-bonded to His35H, and is in contact with Trp33H. Three water molecules are located at the interface between the bases and the Fab residues. Hydrogen bonds involving these water molecules also contribute to Fab recognition of the dT(6-4)T bases. The sugar-phosphate backbone connecting the bases is surrounded by residues His27dL, Tyr32L, Ser92L, Trp33H, and Ser58H, but is not hydrogen-bonded to these residues.

Exploring the feasibility of an anti-idiotypic cocaine vaccine: analysis of the specificity of anticocaine antibodies (Ab1) capable of inducing Ab2beta anti-idiotypic antibodies

Conventional vaccination with the cocaine molecule conjugated to a protein carrier is a new approach in the treatment of addiction. Experimentally, this strategy has been shown to alter the pharmacokinetics as well as the psychostimulant effect of a cocaine challenge. The purpose of this study was to investigate whether a more stable and less controversial molecule, an anti-idiotypic antibody, which mimics the configuration of the cocaine molecule (Ab2beta), could be successfully used instead of cocaine. Two cocaine conjugates that presented different areas of the cocaine molecule to the immune system were used to produce monoclonal antibodies specific for cocaine (Ab1). Several anti-idiotypic antibodies were then produced. Four were identified as Ab2beta, or internal images of the antigen; when injected into BALB/c mice, they elicited an anticocaine response. The anticocaine response elicited by one of the four Ab2beta (K1-4c) was sufficient to significantly reduce the level of cocaine that targeted the brain following cocaine challenge, compared with the level of cocaine found in the brain of control animals immunized with irrelevant antibody. In conclusion, the possibility of an anti-idiotypic vaccine seems to be worth pursuing.

Selection for improved protein stability by phage display

A library of mutants of a single-chain Fv fragment (scFv) was generated by a combination of directed and random mutagenesis, using oligonucleotides randomized at defined positions and two rounds of DNA shuffling. The library was based on the already well folding and stable scFv fragment 4D5Flu. In order to further improve this framework and test the efficiency of various selection strategies, phage display selection was carried out under different selective pressures for higher thermodynamic stability. Incubation of the display phages at elevated temperatures was compared to exposure of the phages to high concentrations of guanidinium chloride. Temperature stress-guided selection yielded the most stable scFv mutant after two rounds of mutagenesis and selection, due to the irreversibility of the unfolding process. It possessed only two mutations (His(L27d)Asn and Phe(L55)Val) and showed a thermodynamic stability improved by roughly 4 kcal/mol, threefold better expression yields in Escherichia coli as well as a 20-fold better binding constant than the 4D5Flu wild-type. The selection results obtained in this study delineate the advantages, disadvantages and limitations of different stability stress selection methods in phage display.