Structural analysis of a set of proteins resulting from a bacterial genomics project.

The targets of the Structural GenomiX (SGX) bacterial genomics project were proteins conserved in multiple prokaryotic organisms with no obvious sequence homolog in the Protein Data Bank of known structures. The outcome of this work was 80 structures, covering 60 unique sequences and 49 different genes. Experimental phase determination from proteins incorporating Se-Met was carried out for 45 structures with most of the remainder solved by molecular replacement using members of the experimentally phased set as search models. An automated tool was developed to deposit these structures in the Protein Data Bank, along with the associated X-ray diffraction data (including refined experimental phases) and experimentally confirmed sequences. BLAST comparisons of the SGX structures with structures that had appeared in the Protein Data Bank over the intervening 3.5 years since the SGX target list had been compiled identified homologs for 49 of the 60 unique sequences represented by the SGX structures. This result indicates that, for bacterial structures that are relatively easy to express, purify, and crystallize, the structural coverage of gene space is proceeding rapidly. More distant sequence-structure relationships between the SGX and PDB structures were investigated using PDB-BLAST and Combinatorial Extension (CE). Only one structure, SufD, has a truly unique topology compared to all folds in the PDB.

A novel quaternary structure of the dimeric alpha-crystallin domain with chaperone-like activity.

alphaB-crystallin, a member of the small heat-shock protein family and a major eye lens protein, is a high molecular mass assembly and can act as a molecular chaperone. We report a synchrotron radiation x-ray solution scattering study of a truncation mutant from the human alphaB-crystallin (alphaB57-157), a dimeric protein that comprises the alpha-crystallin domain of the alphaB-crystallin and retains a significant chaperone-like activity. According to the sequence analysis (more than 23% identity), the monomeric fold of the alpha-crystallin domain should be close to that of the small heat-shock protein from Methanococcus jannaschii (MjHSP16.5). The theoretical scattering pattern computed from the crystallographic model of the dimeric MjHSP16.5 deviates significantly from the experimental scattering by the alpha-crystallin domain, pointing to different quaternary structures of the two proteins. A rigid body modeling against the solution scattering data yields a model of the alpha-crystallin domain revealing a new dimerization interface. The latter consists of a strand-turn-strand motif contributed by each of the monomers, which form a four-stranded, antiparallel, intersubunit composite beta-sheet. This model agrees with the recent spin labeling results and suggests that the alphaB-crystallin is composed by flexible building units with an extended surface area. This flexibility may be important for biological activity and for the formation of alphaB-crystallin complexes of variable sizes and compositions.

Role of GM1 binding in the mucosal immunogenicity and adjuvant activity of the Escherichia coli heat-labile enterotoxin and its B subunit.

Escherichia coli (E. coli) heat-labile toxin (LT) is a potent mucosal immunogen and immunoadjuvant towards co-administered antigens. LT is composed of one copy of the A subunit, which has ADP-ribosylation activity, and a homopentamer of B subunits, which has affinity for the toxin receptor, the ganglioside GM1. Both the ADP-ribosylation activity of LTA and GM1 binding of LTB have been proposed to be involved in immune stimulation. We investigated the roles of these activities in the immunogenicity of recombinant LT or LTB upon intranasal immunization of mice using LT/LTB mutants, lacking either ADP-ribosylation activity, GM1-binding affinity, or both. Likewise, the adjuvant properties of these LT/LTB variants towards influenza virus subunit antigen were investigated. With respect to the immunogenicity of LT and LTB, we found that GM1-binding activity is essential for effective induction of anti-LTB antibodies. On the other hand, an LT mutant lacking ADP-ribosylation activity retained the immunogenic properties of the native toxin, indicating that ADP ribosylation is not critically involved. Whereas adjuvanticity of LTB was found to be directly related to GM1-binding activity, adjuvanticity of LT was found to be independent of GM1-binding affinity. Moreover, a mutant lacking both GM1-binding and ADP-ribosylation activity, also retained adjuvanticity. These results demonstrate that neither ADP-ribosylation activity nor GM1 binding are essential for adjuvanticity of LT, and suggest an ADP-ribosylation-independent adjuvant effect of the A subunit.

Mutational analysis of the role of ADP-ribosylation activity and GM1-binding activity in the adjuvant properties of the Escherichia coli heat-labile enterotoxin towards intranasally administered keyhole limpet hemocyanin.

The Escherichia coli heat-labile enterotoxin (LT) is known for its potent mucosal immunoadjuvant activity towards co-administered antigens. LT is composed of one A subunit, which has ADP-ribosylation activity, and a homopentameric B subunit, which has high affinity for the toxin receptor, ganglioside GM1. In previous studies, we have investigated the role of the LTA and LTB subunits in the adjuvanticity of LT towards influenza virus hemagglutinin (HA), administered intranasally to mice. We now studied the adjuvant properties of LT and LT variants towards keyhole limpet hemocyanin (KLH), which, in contrast to HA, does not bind specifically to mucosal surfaces. It is demonstrated that LT mutants without ADP-ribosylation activity, as well as LTB, retain mucosal immunoadjuvant activity when administered intranasally to mice in conjunction with KLH. As with influenza HA, adjuvanticity of LTB required GM1-binding activity, whereas GM1-binding was not essential for adjuvant activity of LT. Furthermore, we found that also recombinant LTA alone acts as a potent mucosal adjuvant, and that this adjuvanticity is independent of ADP-ribosylation activity. It is concluded that binding of the antigen to mucosal surfaces does not play an essential role in the immunostimulation by LT and LT variants, and that both recombinant LTA and LTB represent powerful nontoxic mucosal adjuvants.

Stepwise transplantation of an active site loop between heat-labile enterotoxins LT-II and LT-I and characterization of the obtained hybrid toxins.

Members of the cholera toxin family, including Escherichia coli heat-labile enterotoxins LT-I and LT-II, catalyze the covalent modification of intracellular proteins by transfer of ADP-ribose from NAD to a specific arginine of the target protein. The ADP-ribosylating activity of these toxins is located in the A-subunit, for which LT-I and LT-II share a 63% sequence identity. The flexible loop in LT-I, ranging from residue 47 to 56, closes over the active site cleft. Previous studies have shown that point mutations in this loop have dramatic effects on the activity of LT-I. Yet, in LT-II the sequence of the equivalent loop differs at four positions from LT-I. Therefore five mutants of the active site loop were created by a stepwise replacement of the loop sequence in LT-I with virtually all the corresponding residues in LT-II. Since we discovered that LT-II had no activity versus the artificial substrate diethylamino-benzylidine-aminoguanidine (DEABAG) while LT-I does, our active site mutants most likely probe the NAD binding, not the arginine binding region of the active site. The five hybrid toxins obtained (Q49A, F52N, V53T, Q49V/F52N and Q49V/F52N/V53T) show (i) great differences in holotoxin assembly efficiency; (ii) decreased cytotoxicity in Chinese hamster ovary cells; and (iii) increased in vitro enzymatic activity compared with wild type LT-I. Specifically, the three mutants containing the F52N substitution display a greater Vmax for NAD than wild type LT-I. The enzymatic activity of the V53T mutant is significantly higher than that of wild type LT-I. Apparently this subtle variation at position 53 is beneficial, in contrast to several other substitutions at position 53 which previously had been shown to be deleterious for activity. The most striking result of this study is that the active site loop of LT-I, despite great sensitivity for point mutations, can essentially be replaced by the active site loop of LT-II, yielding an active 'hybrid enzyme' as well as 'hybrid toxin'.

Modified substrate specificity of L-hydroxyisocaproate dehydrogenase derived from structure-based protein engineering.

L-2-Hydroxyisocaproate dehydrogenase (L-HicDH) is characterized by a broad substrate specificity and utilizes a wide range of 2-oxo acids branched at the C4 atom. Modifications have been made to the sequence of the NAD(H)-dependent L-HicDH from Lactobacillus confusus in order to define and alter the region of substrate specificity towards various 2-oxocarbonic acids. All variations were based on a 3D-structure model of the enzyme using the X-ray coordinates of the functionally related L-lactate dehydrogenase (L-LDH) from dogfish as a template. This protein displays only 23% sequence identity to L-HicDH. The active site of L-HicDH was modelled by homology to the L-LDH based on the conservation of catalytically essential residues. Substitutions of the active site residues Gly234, Gly235, Phe236, Leu239 and Thr245 were made in order to identify their unique participation in substrate recognition and orientation. The kinetic properties of the L239A, L239M, L236V and T245A enzyme variants confirmed the structural model of the active site of L-HicDH. The substrates 2-oxocaproate, 2-oxoisocaproate, phenylpyruvate, phenylglyoxylate, keto-tert-leucine and pyruvate were fitted into the active site of the subsequently refined model. In order to design dehydrogenases with an improved substrate specificity towards keto acids branched at C3 or C4, amino acid substitutions at positions Leu239, Phe236 and Thr245 were introduced and resulted in mutant enzymes with completely different substrate specificities. The substitution T245A resulted in a relative shift of substrate specificity for keto-tert-leucine of more than 17000 compared with the 2-oxocaproate (kcat/KM). For the substrates branched at C4 a relative shift of up to 500 was obtained for several enzyme variants. A total of nine mutations were introduced and the kinetic data for the set of six substrates were determined for each of the resulting mutant enzymes. These were compared with those of the wild-type enzyme and rationalized by the active site model of L-HicDH. An analysis of the enzyme variants provided new insight into the residues involved in substrate binding and residues of importance for the differences between LDHs and HicDH. After the protein design project was complete the X-ray structure of the enzyme was solved in our group. A comparison between the model and the experimental 3D structure proved the quality of the model. All the variants were designed, expressed and tested before the 3D structure became available.

Structural foundation for the design of receptor antagonists targeting Escherichia coli heat-labile enterotoxin.

BACKGROUND: Escherichia coli heat-labile enterotoxin (LT) is the causative agent of traveller's diarrhoea, and it is also responsible for the deaths of hundreds of thousands of children per year in developing countries. LT is highly homologous in sequence, structure and function to cholera toxin (CT). Both toxins attack intestinal epithelial cells via specific binding to the branched pentasaccharide of ganglioside GM1 at the cell surface. A receptor-binding antagonist which blocked this interaction would potentially constitute a prophylactic drug conferring protection both against the severe effects of cholera itself and against the milder but more common disease caused by LT. RESULTS: Four derivatives of the simple sugar galactose, members of a larger series of receptor antagonists identified by computer modeling and competitive binding studies, have been co-crystallized with either the full LT AB5 holotoxin or the LT B pentamer. These crystal structures have provided detailed views of the toxin in complex with each of the four antagonists: melibionic acid at 2.8 A resolution, lactulose at 2.65 A resolution, metanitrophenylgalactoside (MNPG) at 2.2 A resolution and thiodigalactoside (TDG) at 1.7 A resolution. The binding mode of each galactose derivative was observed 5-15 times, depending on the number of crystallographically independent toxin B pentamers per asymmetric unit. There is a remarkable consistency, with one important exception, in the location and hydrogen-bonding involvement of well-ordered water molecules at the receptor-binding site. CONCLUSIONS: The bound conformations of these receptor antagonist compounds preserve the toxin-galactose interactions previously observed for toxin-sugar complexes, but gain additional favorable interactions. The highest affinity compound, MNPG, is notable in that it displaces a water molecule that is observed to be well-ordered in all other previous and current crystal structures of toxin-sugar complexes. This could be a favorable entropic factor contributing to the increased affinity. The highest affinity members of the present set of antagonists (MNPG and TDG) bury roughly half (400 A2) of the binding-site surface covered by the full receptor GM1 pentasaccharide, despite being considerably smaller. This provides an encouraging basis for the creation of subsequent generations of derived compounds that can compete effectively with the natural receptor.

Crystal structure of heat-labile enterotoxin from Escherichia coli with increased thermostability introduced by an engineered disulfide bond in the A subunit.

Cholera toxin (CT) produced by Vibrio cholerae and heat-labile enterotoxin (LT-I), produced by enterotoxigenic Escherichia coli, are AB5 heterohexamers with an ADP-ribosylating A subunit and a GM1 receptor binding B pentamer. These toxins are among the most potent mucosal adjuvants known and, hence, are of interest both for the development of anti-diarrheal vaccines against cholera or enterotoxigenic Escherichia coli diarrhea and also for vaccines in general. However, the A subunits of CT and LT-I are known to be relatively temperature sensitive. To improve the thermostability of LT-I an additional disulfide bond was introduced in the A1 subunit by means of the double mutation N40C and G166C. The crystal structure of this double mutant of LT-I has been determined to 2.0 A resolution. The protein structure of the N40C/G166C double mutant is very similar to the native structure except for a few local shifts near the new disulfide bond. The introduction of this additional disulfide bond increases the thermal stability of the A subunit of LT-I by 6 degrees C. The enhancement in thermostability could make this disulfide bond variant of LT-I of considerable interest for the design of enterotoxin-based vaccines.

Mutants of the Escherichia coli heat-labile enterotoxin with reduced ADP-ribosylation activity or no activity retain the immunogenic properties of the native holotoxin.

The Escherichia coli heat-labile enterotoxin (LT) is a potent inducer of mucosal immune responses. In a previous study (L. DeHaan, W. R. Verweij, M. Holtrop, E. Agsteribbe, and J. Wilschut, Vaccine 14:620-626, 1996), we have shown that efficient induction of an LTB-specific mucosal immune response by LT requires the presence of the LTA chain, suggesting a possible role of the enzymatic activity of LTA in the induction of these responses. In the present study, we generated LT mutants with altered ADP-ribosylation activities and evaluated their immunogenicity upon intranasal administration to mice. The results demonstrate that the mucosal immunogenicity of LT is not dependent on its ADP-ribosylation activity.

Protein engineering studies of A-chain loop 47-56 of Escherichia coli heat-labile enterotoxin point to a prominent role of this loop for cytotoxicity.

Heat-labile enterotoxin (LT), produced by enterotoxigenic Escherichia coli, is a close relative of cholera toxin (CT). These two toxins share approximately 80% sequence identity, and consists of one 240-residue A chain and five 103-residue B subunits. The B pentamer is responsible for GM1 receptor recognition, whereas the A subunit carries out an ADP-ribosylation of an arginine residue in the G protein, Gs alpha, in the epithelial target cell. This paper explores the importance of specific amino acids in loop 47-56 of the A subunit. This loop was observed to be highly mobile in the inactive R7K mutant of the A subunit. The position of the loop in wild-type protein is such that it might require considerable reorganization during substrate binding and is likely to have a crucial role in substrate binding. Five single-site substitutions have been made in the LT-A subunit 47-56 loop to investigate its possible role in the enzymatic activity and toxicity of LT and CT. The wild-type residues Thr-50 and Val-53 were replaced either by a glycine or by a proline. The glycine substitutions were intended to increase the mobility of this active-site loop, and the proline substitutions were intended to decrease the mobility of this same loop by restricting the accessible conformational space. Under the hypothesis that mobility of the loop is important for catalysis, the glycine-substitution mutants T50G and V53G would be expected to exhibit activity equal to or greater than that of the wild-type A subunit, while the proline substitution mutants T50P and T53P would be less active. Cytotoxicity assays showed, however, that all four of these mutants were considerably less active than wild-type LT. These results lend support for assignment of a prominent role to loop 47-56 in catalysis by LT and CT.

Crystal structure of glycosomal glyceraldehyde-3-phosphate dehydrogenase from Leishmania mexicana: implications for structure-based drug design and a new position for the inorganic phosphate binding site.

The structure of glycosomal glyceraldehyde-3-phosphate dehydrogenase (GAPDH) from the trypanosomatid parasite Leishmania mexicana has been determined by X-ray crystallography. The protein crystallizes in space group P2(1)2(1)2(1) with unit cell parameters a = 99.0 A, b = 126.5 A, and c = 138.9 A. There is one 156,000 Da protein tetramer per asymmetric unit. The model of the protein with bound NAD+s and phosphates has been refined against 86% complete data from 10.0 to 2.8 A to a crystallographic Rfactor of 0.198. Density modification by noncrystallographic symmetry averaging was used during model building. The final model of the L. mexicana GAPDH tetramer shows small deviations of less than 0.5 degrees from ideal 222 molecular symmetry. The structure of L. mexicana GAPDH is very similar to that of glycosomal GAPDH from the related trypanosomatid Trypanosoma brucei. A significant structural difference between L. mexicana GAPDH and most previously determined GAPDH structures occurs in a loop region located at the active site. This unusual loop conformation in L. mexicana GAPDH occludes the inorganic phosphate binding site which has been seen in previous GAPDH structures. A new inorganic phosphate position is observed in the L. mexicana GAPDH structure. Model building studies indicate that this new anion binding site is well situated for nucleophilic attack of the inorganic phosphate on the thioester intermediate in the GAPDH-catalyzed reaction. Since crystals of L. mexicana GAPDH can be grown reproducibly and diffract much better than those of T. brucei GAPDH, L. mexicana GAPDH will be used as a basis for structure-based drug design targeted against trypanosomatid GAPDHs.

Protein crystallography and infectious diseases.

The current rapid growth in the number of known 3-dimensional protein structures is producing a database of structures that is increasingly useful as a starting point for the development of new medically relevant molecules such as drugs, therapeutic proteins, and vaccines. This development is beautifully illustrated in the recent book, Protein structure: New approaches to disease and therapy (Perutz, 1992). There is a great and growing promise for the design of molecules for the treatment or prevention of a wide variety of diseases, an endeavor made possible by the insights derived from the structure and function of crucial proteins from pathogenic organisms and from man. We present here 2 illustrations of structure-based drug design. The first is the prospect of developing antitrypanosomal drugs based on crystallographic, ligand-binding, and molecular modeling studies of glycolytic glycosomal enzymes from Trypanosomatidae. These unicellular organisms are responsible for several tropical diseases, including African and American trypanosomiases, as well as various forms of leishmaniasis. Because the target enzymes are also present in the human host, this project is a pioneering study in selective design. The second illustrative case is the prospect of designing anti-cholera drugs based on detailed analysis of the structure of cholera toxin and the closely related Escherichia coli heat-labile enterotoxin. Such potential drugs can be targeted either at inhibiting the toxin's receptor binding site or at blocking the toxin's intracellular catalytic activity. Study of the Vibrio cholerae and E. coli toxins serves at the same time as an example of a general approach to structure-based vaccine design. These toxins exhibit a remarkable ability to stimulate the mucosal immune system, and early results have suggested that this property can be maintained by engineered fusion proteins based on the native toxin structure. The challenge is thus to incorporate selected epitopes from foreign pathogens into the native framework of the toxin such that crucial features of both the epitope and the toxin are maintained. That is, the modified toxin must continue to evoke a strong mucosal immune response, and this response must be directed against an epitope conformation characteristic of the original pathogen.

Deletion variants of L-hydroxyisocaproate dehydrogenase. Probing substrate specificity.

The substrate specificity and catalytic activity of the dinucleotide-dependent L-2-hydroxyisocaproate dehydrogenase from Lactobacillus confusus (L-HicDH) have been altered by modifying an enzyme region which is assumed to be involved in substrate recognition. The design of the variant enzymes was based on an amino acid alignment of the modified region with the functionally related L-lactate dehydrogenases. The best absolute sequence similarity for a protein with known tertiary structure was found for L-lactate dehydrogenase from dogfish (23%). In this study, the coenzyme loop, a functional element which is essential for catalysis and substrate specificity, was modified in order to identify the residues involved in the catalytic reaction and observe the effect on the substrate specificity. Deletions were introduced into the L-hydroxyisocaproate gene by site-directed mutagenesis. Several deletion-variant enzymes Ile100A delta, Lys100B delta, Leu101 delta, Asn105A delta and Pro105B delta showed an altered substrate specificity. For the variant enzyme with the deletion of Asn/Pro105A/B, 2-oxo carboxylic acids branched at C4 proved to be better substrates than 2-oxocaproate, the substrate with the best kcat/KM ratio known for the wild-type enzyme. The mutation resulted in a 5.2-fold increased catalytic efficiency towards 2-oxoisocaproate compared to the wild-type enzyme. After deleting Ile/Lys100A/B, 2-phenylpyruvate is the only substrate which is still converted at a significant catalytic rate. The kcat ratios of 2-oxocaproate versus 2-phenylpyruvate changed by a factor of 6500 when comparing wild-type enzyme and deletion-variant enzyme data. The single amino acid deletions in position 100A and 100B caused drastic reductions in the catalytic activity for all tested substrates, whereas the deletion of Lys100B, Leu101, Asn105A as well as Pro105B showed more specific modifications in catalytic rates and substrate recognition for each tested substrate.