Solving the mechanism of a chemical reaction requires determining the structures of all the ground states on the pathway and the elusive transition states linking them. 2024 is the centenary of Brønsted’s landmark paper that introduced the β-value and structure-activity studies as the only experimental means to infer the structures of transition states. It involves making systematic small changes in the covalent structure of the reactants and analysing changes in activation and equilibrium-free energies. Protein engineering was introduced for an analogous procedure, Φ-value analysis, to analyse the noncovalent interactions in proteins central to biological chemistry. The methodology was developed first by analysing noncovalent interactions in transition states in enzyme catalysis. The mature procedure was then applied to study transition states in the pathway of protein folding – ‘part (b) of the protein folding problem’. This review describes the development of Φ-value analysis of transition states and compares and contrasts the interpretation of β- and Φ-values and their limitations. Φ-analysis afforded the first description of transition states in protein folding at the level of individual residues. It revealed the nucleation-condensation folding mechanism of protein domains with the transition state as an expanded, distorted native structure, containing little fully formed secondary structure but many weak tertiary interactions. A spectrum of transition states with various degrees of structural polarisation was then uncovered that spanned from nucleation-condensation to the framework mechanism of fully formed secondary structure. Φ-analysis revealed how movement of the expanded transition state on an energy landscape accommodates the transition from framework to nucleation-condensation mechanisms with a malleability of structure as a unifying feature of folding mechanisms. Such movement follows the rubric of analysis of classical covalent chemical mechanisms that began with Brønsted. Φ-values are used to benchmark computer simulation, and Φ and simulation combine to describe folding pathways at atomic resolution.

We selected 50 crystal (Cry) toxins specific to lepidopteran insects. Phylogenetic relationships of whole toxins as well as individual domains revealed that Cry1Jb and Cry1Ac are distantly placed but related to reduced cross-resistance in Helicoverpa armigera. Multiple alignments of Cry1Jb and Cry1Ac amino acid sequences showed significant differences in the composition and length of the loops of domains II and III. This was further confirmed by the superpositioning of 3D structures. cry1Jb and cry1Ac genes cloned in expression vectors were overexpressed in Escherichia coli Castellani & Chalmers, and proteins were harvested. Insect bioassays revealed that the wild-type Cry1Jb and Cry1Ac proteins showed differential specificity to H. armigera (Hübner), Spodoptera litura (Fab.) and Earias vittella F. Chimeric genes were constructed by exchanging the domains between cry1Jb and cry1Ac and overexpressed in BL21(DE3). Substitution of domain III of Cry1Jb with that of Cry1Ac enhanced its activity against H. armigera 7.8-fold. Bioassay of the parental and chimeric toxins against S. litura revealed no activity even at the discriminative dose of 100 μg/ml, but a significant difference in growth was observed. The present results, along with previous domain swapping experiments, suggest that protein engineering not only reveals the mechanism by which endotoxins work, but also generates novel toxins with enhanced toxicity/broader specificity.

I wish to thank the Society for inviting me to give the Introductory Talk in this Symposium. It is going to centre around what might be called ‘epitope specific’ immunization and I want to draw your attention to a number of factors which will be bound to affect any new approach to the development of vaccines. To begin with, it is now clear that complex antigens, such as those associated with parasites, present many different epitopes to the immune system. These epitopes are ‘seen’ by different cell types using different receptors and epitope recognition can activate a number of different effector systems.

We have characterized the thermodynamic stability of the SH3 domain from the Saccharomyces cerevisiae Abp1p protein and found it to be relatively low compared to most other SH3 domains, with a Tm of 60 °C and a ΔGu of 3.08 kcal/mol. Analysis of a large alignment of SH3 domains led to the identification of atypical residues at eight positions in the wild-type Abp1p SH3 domain sequence that were subsequently replaced by the residue seen most frequently at that position in the alignment. Three of the eight mutants constructed in this way displayed increases in Tm ranging from 8 to 15 °C with concomitant increases in ΔGu of up to 1.4 kcal/mol. The effects of these substitutions on folding thermodynamics and kinetics were entirely additive, and a mutant containing all three was dramatically stabilized with a Tm greater than 90 °C and a ΔGu more than double that of the wild-type domain. The folding rate of this hyperstable mutant was 10-fold faster than wild-type, while its unfolding rate was fivefold slower. All of the stabilized mutants were still able to bind a target peptide with wild-type affinity. We have analyzed the stabilizing amino acid substitutions isolated in this study and several other similar sequence alignment based studies. In approximately 25% of cases, increased stability can be explained by enhanced propensity of the substituted residue for the local backbone conformation at the mutagenized site.

By using a novel consensus approach, we have previously managed to generate a fully synthetic phytase, consensus phytase-1, that was 15–26 °C more thermostable than the parent fungal phytases used in its design (Lehmann et al., 2000). We now sought to use the backbone of consensus phytase-1 and to modify its catalytic properties. This was done by replacing a considerable part of the active site (i.e., all the divergent residues) with the corresponding residues of Aspergillus niger NRRL 3135 phytase, which displays pronounced differences in specific activity, substrate specificity, and pH-activity profile. For the new protein termed consensus phytase-7, a major—although not complete—shift in catalytic properties was observed, demonstrating that rational transfer of favorable catalytic properties from one phytase to another is possible by using this approach. Although the exchange of the active site was associated with a 7.6 °C decrease in unfolding temperature (Tm) as measured by differential scanning calorimetry, consensus phytase-7 still was >7 °C more thermostable than all wild-type ascomycete phytases known to date. Thus, combination of the consensus approach with the selection of a “preferred” active site allows the design of a thermostabilized variant of an enzyme family of interest that (most closely) matches the most favorable catalytic properties found among its family members.

The potential for engineering stable proteins with multiple amino acid substitutions was explored. Eleven lysine, five methionine, two tryptophan, one glycine, and three threonine substitutions were simultaneously made in barley chymotrypsin inhibitor-2 (CI-2) to substantially improve the essential amino acid content of the protein. These substitutions were chosen based on the three-dimensional structure of CI-2 and an alignment of homologous sequences. The initial engineered protein folded into a wild-type-like structure, but had a free energy of unfolding of only 2.2 kcal/mol, considerably less than the wild-type value of 7.5 kcal/mol. Restoration of the lysine mutation at position 67 to the wild-type arginine increased the free energy of unfolding to 3.1 kcal/mol. Subsequent cysteine substitutions at positions 22 and 82 resulted in disulfide bond formation and a protein with nearly wild-type thermodynamic stability (7.0 kcal/mol). None of the engineered proteins retained inhibitory activity against chymotrypsin or elastase, and all had substantially reduced inhibitory activity against subtilisin. The proteolytic stabilities of the proteins correlated with their thermodynamic stabilities. Reduction of the disulfide bond resulted in substantial loss of both thermodynamic and proteolytic stabilities, confirming that the disulfide bond, and not merely the cysteine substitutions, was responsible for the increased stability. We conclude that it is possible to replace over a third of the residues in CI-2 with minimal disruption of stability and structural integrity.

The activation domain of human procarboxypeptidase A2 (ADA2h) aggregates following thermal or chemical denaturation at acidic pH. The aggregated material contains well-defined ordered structures with all the characteristics of the fibrils associated with amyloidotic diseases. Variants of ADA2h containing a series of mutations designed to increase the local stability of each of the two helical regions of the protein have been found to have a substantially reduced propensity to form fibrils. This arises from a reduced tendency of the denatured species to aggregate rather than from a change in the overall stability of the native state. The reduction in aggregation propensity may result from an increase in the stability of local relative to longer range interactions within the polypeptide chain. These findings show that the intrinsic ability of a protein to form amyloid can be altered substantially by protein engineering methods without perturbing significantly its overall stability or activity. This suggests new strategies for combating diseases associated with the formation of aggregated proteins and for the design of novel protein or peptide therapeutics.

We have reported variants of the M13 bacteriophage major coat protein (P8) that enable high copy display of monomeric and oligomeric proteins, such as human growth hormone and steptavidin, on the surface of phage particles (Sidhu SS, Weiss GA, Wells JA. 2000. High copy display of large proteins on phage for functional selections. J Mol Biol 296:487–495). Here, we explore how an optimized P8 variant (opti-P8) could evolve the ability to efficiently display a protein fused to its N-terminus. Reversion of individual opti-P8 residues back to the wild-type P8 residue identifies a limited set of hydrophobic residues responsible for the high copy protein display. These hydrophobic amino acids bracket a conserved hydrophobic face on the P8 alpha helix thought to be in contact with the phage coat. Mutations additively combine to promote high copy protein display, which was further enhanced by optimization of the linker between the phage coat and the fusion protein. These data are consistent with a model in which protein display-enhancing mutations allow for better packing of the fusion protein into the phage coat. The high tolerance for phage coat protein mutations observed here suggests that filamentous phage coat proteins could readily evolve new capabilities.

An analysis of the pairwise side-chain packing geometries of cysteine residues observed in high-resolution protein crystal structures indicates that cysteine pairs have pronounced orientational preferences due to the geometric constraints of disulfide bond formation. A potential function was generated from these observations and used to evaluate models for novel disulfide bonds in human carbonic anhydrase II (HCAII). Three double-cysteine variants of HCAII were purified and the effective concentrations of their thiol groups were determined by titrations with glutathione and dithiothreitol. The effects of the cysteine mutations on the native state structure and stability were characterized by circular dichroism, enzymatic activity, sulfonamide binding, and guanidine hydrochloride titration. These analyses indicate that the PAIRWISE potential is a good predictor of the strength of the disulfide bond itself, but the overall structural and thermodynamic effects on the protein are complicated by additional factors. In particular, the effects of cysteine substitutions on the native state and the stabilization of compact nonnative states by the disulfide can override any stabilizing effect of the cross-link.

The crystal structure of the fibrinolytic enzyme tissue plasminogen activator (tPA) shows that the bulky side chain of Y99 hinders access to the active site by partially occluding the S2 site and may be responsible for the low catalytic activity of tPA toward plasminogen. We have tested the role of Y99 by replacing it with Leu, the residue found in more proficient proteases like trypsin and thrombin. The Y99L replacement results in an increase in the kcat/Km for chromogenic substrates due to enhanced diffusion into the active site. The increase is modest (threefold) for substrates specific for tPA that carry Pro or Gly at P2, but reaches 80-fold for less specific substrates carrying Arg at P2. On the other hand, the Y99L mutation has no effect on the activity of tPA toward the natural substrate plasminogen, that carries Gly at P2, and reduces more than 10-fold the inhibition of tPA by plasminogen activator inhibitor-1 (PAI-1), that carries Ala at P2. We conclude that the steric hindrance provided by Y99 in the crystal structure affects mostly nonphysiological substrates with bulky residues at P2. In addition, residue Y99 plays an active role in the recognition of PAI-1, but not plasminogen. Mutations of Y99 could therefore afford a resistance to inhibition by PAI-1 without compromising the fibrinolytic potency of tPA, a result of potential therapeutic relevance.

The mechanism of β-sheet formation remains a fundamental issue in our understanding of the protein folding process, but is hampered by the often encountered kinetic competition between folding and aggregation. The role of local versus nonlocal interactions has been probed traditionally by mutagenesis of both turn and strand residues. Recently, rigid organic molecules that impose a correct chain reversal have been introduced in several small peptides to isolate the importance of the long-range interactions. Here, we present the incorporation of a well-studied β-turn mimic, designated as the dibenzofuran-based (DBF) amino acid, in the B1 domain of streptococcal protein G (B1G), and compare our results with those obtained upon insertion of the same mimic into the N-terminal β-hairpin of B1G (O Melnyk et al., 1998, Lett Pept Sci 5:147–150). The DBF-B1G domain conserves the structure and the functional and thermodynamical properties of the native protein, whereas the modified peptide does not adopt a native-like conformation. The nature of the DBF flanking residues in the modified B1G domain prevents the β-turn mimic from acting as a strong β-sheet nucleator, which reinforces the idea that the native β-hairpin formation is not driven by the β-turn formation, but by tertiary interactions.

The folding of immunoglobulin domains requires the formation of a conserved structural disulfide. Therefore, as a general rule, they cannot be functionally expressed in the reducing environment of the cellular cytoplasm. We have previously reported that stability engineering can lead to the cytoplasmic expression of functional immunoglobulin VL domains. Here we apply rational stability engineering by consensus sequence analysis to VH domains. Isolated VH domains tend to aggregate more easily than VL domains; they do not refold quantitatively and are generally more difficult to handle in vitro. To overcome these problems, we successfully predicted and experimentally verified several stabilizing point mutations in the VH domain of a designed, catalytic Fv fragment. The effect of single mutations was additive, and they could be combined in a prototype domain with significantly improved stability against chemical denaturation and a 20-fold increased half time of irreversible thermal denaturation, at physiological temperature. This stabilized, isolated VH domain could be expressed solubly in the reducing cellular cytoplasm of Escherichia coli, at a yield of approximately 1.2 mg/L of shake flask culture. It remains fully functional, as evidenced by the successful reconstitution of an esterolytic Fv fragment with the VL domain. This success provides further evidence that consensus sequence engineering is a rational, plannable route to the construction of intrabodies.

GroEL minichaperones have potential in the biotechnology industry for the refolding of recombinant proteins. With the aim of enhancing and widening their use, we have created two highly stable functional variants of minichaperone GroEL(193–345). A sequence alignment of 130 members of the chaperonin 60 (Cpn60) family was used to design 37 single mutations. Two small-to-large mutations, A223T, A223V and one similar-size mutation, M233L, all located in the hydrophobic core were found to stabilize the protein by more than 1 kcal mol−1 each. Six stabilizing mutations were combined, yielding two multiple mutants that were 6.99 and 6.15 kcal mol−1 more stable than wild-type protein. Even though some of the substituted residue pairs are close to each other in the protein structure, the energetic effects of mutation are approximately additive. In particular, the stabilizing substitution A223T is unexpected and would have been missed by purely structural analysis. In the light of previously reported successes employing similar methods with several other proteins, our results show that a homology based approach is a simple and efficient method of increasing the stability of a protein.

Cellulases are increasingly being used for industrial purposes, particularly in washing powders, yet little is known of the factors governing the stability of proteins in detergent solutions. We present a comparative analysis of the behavior of the cellulase Cel45 from Humicola insolens in the presence of the denaturant guanidinium chloride and the anionic detergent C12-LAS. Although Cel45 unfolds in GdmCl according to a simple two-state model under equilibrium conditions, it accumulates a transient intermediate during refolding. The four disulfide bonds do not contribute detectably to the stability of the native state. Cel45 is unfolded by very low concentrations of C12-LAS (1–4 mM). An analysis of 16 mutants of Cel45 shows a very weak correlation between unfolding rates in denaturant and detergent; mutants that have the same unfolding rate in GdmCl (within a factor of 1.5) vary 1,000-fold in their unfolding rates in C12-LAS. The data support a simple model for unfolding by detergent, in which the introduction of positive charges or removal of negative charges greatly increases detergent sensitivity, while interactions with the hydrophobic detergent tail contribute to a smaller extent. This implies that different detergent-mediated unfolding pathways exist, whose accessibilities depend on individual residues. Double-mutant cycles reveal that mutations in two proximal residues lead to repulsion and a destabilization greater than the sum of the individual mutations as measured by GdmCl denaturation, but they also reduce the affinity for LAS and therefore actually stabilize the protein relative to wild-type. Ligands that interact strongly with the denatured state may therefore alter the unfolding process.

The lipases from Rhizopus and Rhizomucor are members of the family of Mucorales lipases. Although they display high sequence homology, their stereoselectivity toward triradylglycerols (sn-2 substituted triacylglycerols) varies. Four different triradylglycerols were investigated, which were classified into two groups: flexible substrates with rotatable O′-C1′ ether or ester bonds adjacent to C2 of glycerol and rigid substrates with a rigid N′-C1′ amide bond or a phenyl ring in sn-2. Although Rhizopus lipase shows opposite stereopreference for flexible and rigid substrates (hydrolysis in sn-1 and sn-3, respectively), Rhizomucor lipase hydrolyzes both groups of triradylglycerols preferably in sn-1. To explain these experimental observations, computer-aided molecular modeling was applied to study the molecular basis of stereoselectivity. A generalized model for both lipases of the Mucorales family highlights the residues mediating stereoselectivity: (1) L258, the C-terminal neighbor of the catalytic histidine, and (2) G266, which is located in a loop contacting the glycerol backbone of a bound substrate. Interactions with triradylglycerol substrates are dominated by van der Waals contacts. Stereoselectivity can be predicted by analyzing the value of a single substrate torsion angle that discriminates between sn-1 and sn-3 stereopreference for all substrates and lipases investigated here. This simple model can be easily applied in enzyme and substrate engineering to predict Mucorales lipase variants and synthetic substrates with desired stereoselectivity.