High order quaternary arrangement confers increased structural stability to Brucella sp. lumazine synthase

The penultimate step in the pathway of riboflavin biosynthesis is catalyzed by the enzyme lumazine synthase (LS). One of the most distinctive characteristics of this enzyme is the structural quaternary divergence found in different species. The protein exists as pentameric and icosahedral forms, built from practically the same structural monomeric unit. The pentameric structure is formed by five 18-kDa monomers, each extensively contacting neighboring monomers. The icosahedrical structure consists of 60 LS monomers arranged as 12 pentamers giving rise to a capsid exhibiting icosahedral 532 symmetry. In all lumazine synthases studied, the topologically equivalent active sites are located at the interfaces between adjacent subunits in the pentameric modules. The Brucella sp. lumazine synthase (BLS) sequence clearly diverges from pentameric and icosahedric enzymes. This unusual divergence prompted us to further investigate its quaternary arrangement. In the present work, we demonstrate by means of solution light scattering and x-ray structural analyses that BLS assembles as a very stable dimer of pentamers, representing a third category of quaternary assembly for lumazine synthases. We also describe by spectroscopic studies the thermodynamic stability of this oligomeric protein and postulate a mechanism for dissociation/unfolding of this macromolecular assembly. The higher molecular order of BLS increases its stability 20 degrees C compared with pentameric lumazine synthases. The decameric arrangement described in this work highlights the importance of quaternary interactions in the stabilization of proteins.     

Crystal structure analysis of a pentameric fungal and an icosahedral plant lumazine synthase reveals the structural basis for differences in assembly

Lumazine synthase catalyzes the penultimate step in the synthesis of riboflavin in plants, fungi, and microorganisms. The enzyme displays two quaternary structures, the pentameric forms in yeast and fungi and the 60-meric icosahedral capsids in plants and bacteria. To elucidate the structural features that might be responsible for differences in assembly, we have determined the crystal structures of lumazine synthase, complexed with the inhibitor 5-nitroso-6-ribitylamino-2,4-pyrimidinedione, from spinach and the fungus Magnaporthe grisea to 3.3 and 3.1 A resolution, respectively. The overall structure of the subunit and the mode of inhibitor binding are very similar in these enzyme species. The core of the subunit consists of a four-stranded parallel beta-sheet sandwiched between two helices on one side and three helices on the other. The packing of the five subunits in the pentameric M. grisea lumazine synthase is very similar to the packing in the pentameric substructures in the icosahedral capsid of the plant enzyme. Two structural features can be correlated to the differences in assembly. In the plant enzyme, the N-terminal beta-strand interacts with the beta-sheet of the adjacent subunit, thus extending the sheet from four to five strands. In fungal lumazine synthase, an insertion of two residues after strand beta1 results in a completely different orientation of this part of the polypeptide chain and this conformational difference prevents proper packing of the subunits at the trimer interface in the icosahedron. In the spinach enzyme, the beta-hairpin connecting helices alpha4 and alpha5 participates in the packing at the trimer interface of the icosahedron. Another insertion of two residues at this position of the polypeptide chain in the fungal enzyme disrupts the hydrogen bonding in the hairpin, and the resulting change in conformation of this loop also interferes with proper intrasubunit contacts at the trimer interface.     

Divergence in macromolecular assembly: X-ray crystallographic structure analysis of lumazine synthase from Brucella abortus

We have determined the three-dimensional structure of 6, 7-dimethyl-8-ribityllumazine synthase (lumazine synthase) from Brucella abortus, the infectious organism of the disease brucellosis in animals. This enzyme catalyses the formation of 6, 7-dimethyl-8-ribityllumazine, the penultimate product in the synthesis of riboflavin. The three-dimensional X-ray crystal structure of the enzyme from B. abortus has been solved and refined at 2.7 A resolution to a final R-value of 0.18 (R(free)=0.23). The macromolecular assembly of the enzyme differs from that of the enzyme from Bacillus subtilis, the only other lumazine synthase structure known. While the protein from B. subtilis assembles into a 60 subunit icosahedral capsid built from 12 pentameric units, the enzyme from B. abortus is pentameric in its crystalline form. Nonetheless, the active sites of the two enzymes are virtually identical indicating inhibitors to theses enzymes could be effective pharmaceuticals across a broad species range. Furthermore, we compare the structures of the enzyme from B. subtilis and B. abortus and describe the C teminus structure which accounts for the differences in quaternary structure.     

X-ray structure analysis and crystallographic refinement of lumazine synthase from the hyperthermophile Aquifex aeolicus at 1.6 A resolution: determinants of thermostability revealed from structural comparisons

An open reading frame optimized for expression of 6,7-dimethyl-8-ribityl-lumazine synthase of the hyperthermophilic bacterium Aquifex aeolicus in Escherichia coli was synthesized and expressed in a recombinant E. coli strain to a level of around 15 %. The recombinant protein was purified by heat-treatment and gel-filtration. The protein was crystallized in the cubic space group I23 with the cell dimensions a = b = c = 180.8 A, and diffraction data were collected to 1.6 A resolution. The structure was solved by molecular replacement using lumazine synthase from Bacillus subtilis as search model. The structure of the A. aeolicus enzyme was refined to a resolution of 1.6 A. The spherical protein consists of 60 identical subunits with strict icosahedral 532 symmetry. The subunit fold is closely related to that of the B. subtilis enzyme (rmsd 0.80 A). The extremely thermostable lumazine synthase from A. aeolicus has a melting temperature of 119.9 degrees C. Compared to other icosahedral and pentameric lumazine synthases, the A. aeolicus enzyme has the largest accessible surface presented by charged residues and the smallest surface presented by hydrophobic residues. It also has the largest number of ion-pairs per subunit. Two ion-pair networks involving two, respectively three, stacking arginine residues assume a distinct role in linking adjacent subunits. The findings indicate the influence of the optimization of hydrophobic and ionic contacts in gaining thermostability.     

Multiple assembly states of lumazine synthase: a model relating catalytic function and molecular assembly

Lumazine synthases have been observed in the form of pentamers, dimers of pentamers, icosahedral capsids consisting of 60 subunits and larger capsids with unknown molecular structure. Here we describe the analysis of the assembly of native and mutant forms of lumazine synthases from Bacillus subtilis and Aquifex aeolicus at various pH values and in the presence of different buffers using small angle X-ray scattering and electron microscopy. Both wild-type lumazine synthases are able to form capsids with a diameter of roughly 160 A and larger capsids with diameters of around 300 A. The relative abundance of smaller and larger capsids is strongly dependent on buffer and pH. Both forms can co-exist and are in some cases accompanied by other incomplete or deformed capsids. Several mutants of the B. subtilis lumazine synthase, in which residues in or close to the active site were replaced, as well as an insertion mutant of A. aeolicus lumazine synthase form partially or exclusively larger capsids with a diameter of about 300 A. The mutations also reduce or inhibit enzymatic activity, suggesting that the catalytic function of the enzyme is tightly correlated with its quaternary structure. The data show that multiple assembly forms are a general feature of lumazine synthases.     

Crystal structure of lumazine synthase from Mycobacterium tuberculosis as a target for rational drug design: binding mode of a new class of purinetrione inhibitors

The enzymes involved in the biosynthesis of riboflavin represent attractive targets for the development of drugs against bacterial pathogens, because the inhibitors of these enzymes are not likely to interfere with enzymes of the mammalian metabolism. Lumazine synthase catalyzes the penultimate step in the riboflavin biosynthesis pathway. A number of substituted purinetrione compounds represent a new class of highly specific inhibitors of lumazine synthase from Mycobacterium tuberculosis. To develop potent antibiotics for the treatment of tuberculosis, we have determined the structure of lumazine synthase from M. tuberculosis in complex with two purinetrione inhibitors and have studied binding via isothermal titration calorimetry. The structures were determined by molecular replacement using lumazine synthase from Saccharomyces cerevisiae as a search model and refined at 2 and 2.3 A resolution. The R-factors were 14.7 and 17.4%, respectively, and the R(free) values were 19.3 and 26.3%, respectively. The enzyme was found to be a pentamer consisting of five subunits related by 5-fold local symmetry. The comparison of the active site architecture with the active site of previously determined lumazine synthase structures reveals a largely conserved topology with the exception of residues Gln141 and Glu136, which participate in different charge-charge interactions in the core space of the active site. The impact of structural changes in the active site on the altered binding and catalytic properties of the enzyme is discussed. Isothermal titration calorimetry measurements indicate highly specific binding of the purinetrione inhibitors to the M. tuberculosis enzyme with dissociation constants in micromolar range.     

Evolutionary Morphing of Tryptophan Synthase: Functional Mechanisms for the Enzymatic Channeling of Indole

Tryptophan synthase (TrpS) is a heterotetrameric αββα enzyme that exhibits complex substrate channeling and allosteric mechanisms and is a model system in enzymology. In this work, we characterize proposed early and late evolutionary states of TrpS and show that they have distinct quaternary structures caused by insertions–deletions of sequence segments (indels) in the β-subunit. Remarkably, indole hydrophobic channels that connect α and β active sites have re-emerged in both TrpS types, yet they follow different paths through the β-subunit fold. Also, both TrpS geometries activate the α-subunit through the rearrangement of loops flanking the active site. Our results link evolutionary sequence changes in the enzyme subunits with channeling and allostery in the TrpS enzymes. The findings demonstrate that indels allow protein quaternary architectures to escape “minima” in the evolutionary landscape, thereby overcoming the conservational constraints imposed by existing functional interfaces and being free to morph into new mechanistic enzymes.     

Plant riboflavin biosynthesis. Cloning, chloroplast localization, expression, purification, and partial characterization of spinach lumazine synthase.

Lumazine synthase, which catalyzes the penultimate step of riboflavin biosynthesis, has been cloned from three higher plants (spinach, tobacco, and arabidopsis) through functional complementation of an Escherichia coli auxotroph. Whereas the three plant proteins exhibit some structural similarities to known microbial homologs, they uniquely possess N-terminal polypeptide extensions that resemble typical chloroplast transit peptides. In vitro protein import assays with intact chloroplasts and immunolocalization experiments verify that higher plant lumazine synthase is synthesized in the cytosol as a larger molecular weight precursor protein, which is post-translationally imported into chloroplasts where it is proteolytically cleaved to its mature size. The authentic spinach enzyme is estimated to constitute <0.02% of the total chloroplast protein. Recombinant "mature" spinach lumazine synthase is expressed in E. coli at levels exceeding 30% of the total soluble protein and is readily purified to homogeneity using a simple two-step procedure. Apparent V(max) and K(m) values obtained with the purified plant protein are similar to those reported for microbial lumazine synthases. Electron microscopy and hydrodynamic studies reveal that native plant lumazine synthase is a hollow capsid-like structure comprised of 60 identical 16.5-kDa subunits, resembling its icosahedral counterparts in E. coli and Bacillus subtilis.     

Conserved patterns and interactions in the unfolding transition state across SH3 domain structural homologues

Proteins with similar structures are generally assumed to arise from similar sequences. However, there are more cases than not where this is not true. The dogma is that sequence determines structure; how, then, can very different sequences fold to the same structure? Here, we employ high temperature unfolding simulations to probe the pathways and specific interactions that direct the folding and unfolding of the SH3 domain. The SH3 metafold in the Dynameomics Database consists of 753 proteins with the same structure, but varied sequences and functions. To investigate the relationship between sequence and structure, we selected 17 targets from the SH3 metafold with high sequence variability. Six unfolding simulations were performed for each target, transition states were identified, revealing two general folding/unfolding pathways at the transition state. Transition states were also expressed as mathematical graphs of connected chemical nodes, and it was found that three positions within the structure, independent of sequence, were consistently more connected within the graph than any other nearby positions in the sequence. These positions represent a hub connecting different portions of the structure. Multiple sequence alignment and covariation analyses also revealed certain positions that were more conserved due to packing constraints and stabilizing long‐range contacts. This study demonstrates that members of the SH3 domain with different sequences can unfold through two main pathways, but certain characteristics are conserved regardless of the sequence or unfolding pathway. While sequence determines structure, we show that disparate sequences can provide similar interactions that influence folding and lead to similar structures.     

Enzyme catalysis via control of activation entropy: site-directed mutagenesis of 6,7-dimethyl-8-ribityllumazine synthase

6,7-Dimethyl-8-ribityllumazine synthase (lumazine synthase) catalyses the penultimate step in the biosynthesis of riboflavin. In Bacillus subtilis, 60 lumazine synthase subunits form an icosahedral capsid enclosing a homotrimeric riboflavin synthase unit. The ribH gene specifying the lumazine synthase subunit can be expressed in high yield. All amino acid residues exposed at the surface of the active site cavity were modified by PCR assisted mutagenesis. Polar amino acid residues in direct contact with the enzyme substrates, 5-amino-6-ribitylamino-2,4(1H,3H)-pyrimidinedione and 3,4-dihydroxy-2-butanone 4-phosphate, could be replaced with relative impunity with regard to the catalytic properties. Only the replacement of Arg127, which forms a salt bridge with the phosphate group of 3,4-dihydroxy-2-butanone 4-phosphate, reduced the catalytic rate by more than one order of magnitude. Replacement of His88, which is believed to assist in proton transfer reactions, reduced the catalytic activity by about one order of magnitude. Surprisingly, the activation enthalpy deltaH of the lumazine synthase reaction exceeds that of the uncatalysed reaction. On the other hand, the free energy of activation deltaG of the uncatalysed reaction is characterised by a large entropic term (TdeltaS) of -37.8 kJmol(-1), whereas the entropy of activation (TdeltaS) of the enzyme-catalysed reaction is -6.7 kJmol(-1). This suggests that the rate enhancement by the enzyme is predominantly achieved by establishing a favourable topological relation of the two substrates, whereas acid/base catalysis may play a secondary role.     

A structure-based model of the reaction catalyzed by lumazine synthase from Aquifex aeolicus

6,7-Dimethyl-8-ribityllumazine is the biosynthetic precursor of riboflavin, which, as a coenzyme, plays a vital role in the electron transfer process for energy production in all cellular organisms. The enzymes involved in lumazine biosynthesis have been studied in considerable detail. However, the conclusive mechanism of the reaction catalyzed by lumazine synthase has remained unclear. Here, we report four crystal structures of the enzyme from the hyperthermophilic bacterium Aquifex aeolicus in complex with different inhibitor compounds. The structures were refined at resolutions of 1.72 A, 1.85 A, 2.05 A and 2.2 A, respectively. The inhibitors have been designed in order to mimic the substrate, the putative reaction intermediates and the final product. Structural comparisons of the native enzyme and the inhibitor complexes as well as the kinetic data of single-site mutants of lumazine synthase from Bacillus subtilis showed that several highly conserved residues at the active site, namely Phe22, His88, Arg127, Lys135 and Glu138 are most likely involved in catalysis. A structural model of the catalytic process, which illustrates binding of substrates, enantiomer specificity, proton abstraction/donation, inorganic phosphate elimination, formation of the Schiff base and cyclization is proposed.     

Flexible DNA target site recognition by divergent homing endonuclease isoschizomers I-CreI and I-MsoI

Homing endonucleases are highly specific catalysts of DNA strand breaks that induce the transposition of mobile intervening sequences containing the endonuclease open reading frame. These enzymes recognize long DNA targets while tolerating individual sequence polymorphisms within those sites. Sequences of the homing endonucleases themselves diversify to a great extent after founding intron invasion events, generating highly divergent enzymes that recognize similar target sequences. Here, we visualize the mechanism of flexible DNA recognition and the pattern of structural divergence displayed by two homing endonuclease isoschizomers. We determined structures of I-CreI bound to two DNA target sites that differ at eight of 22 base-pairs, and the structure of an isoschizomer, I-MsoI, bound to a nearly identical DNA target site. This study illustrates several principles governing promiscuous base-pair recognition by DNA-binding proteins, and demonstrates that the isoschizomers display strikingly different protein/DNA contacts. The structures allow us to determine the information content at individual positions in the binding site as a function of the distribution of direct and water-mediated contacts to nucleotide bases, and provide an evolutionary snapshot of endonucleases at an early stage of divergence in their target specificity.     

Distinguishing structural and functional restraints in evolution in order to identify interaction sites

Structural genomics projects are producing many three-dimensional structures of proteins that have been identified only from their gene sequences. It is therefore important to develop computational methods that will predict sites involved in productive intermolecular interactions that might give clues about functions. Techniques based on evolutionary conservation of amino acids have the advantage over physiochemical methods in that they are more general. However, the majority of techniques neither use all available structural and sequence information, nor are able to distinguish between evolutionary restraints that arise from the need to maintain structure and those that arise from function. Three methods to identify evolutionary restraints on protein sequence and structure are described here. The first identifies those residues that have a higher degree of conservation than expected: this is achieved by comparing for each amino acid position the sequence conservation observed in the homologous family of proteins with the degree of conservation predicted on the basis of amino acid type and local environment. The second uses information theory to identify those positions where environment-specific substitution tables make poor predictions of the overall amino acid substitution pattern. The third method identifies those residues that have highly conserved positions when three-dimensional structures of proteins in a homologous family are superposed. The scores derived from these methods are mapped onto the protein three-dimensional structures and contoured, allowing identification clusters of residues with strong evolutionary restraints that are sites of interaction in proteins involved in a variety of functions. Our method differs from other published techniques by making use of structural information to identify restraints that arise from the structure of the protein and differentiating these restraints from others that derive from intermolecular interactions that mediate functions in the whole organism.     

Structural and kinetic properties of lumazine synthase isoenzymes in the order Rhizobiales

6,7-Dimethyl-8-ribityllumazine synthase (lumazine synthase; LS) catalyzes the penultimate step in the biosynthesis of riboflavin in plants and microorganisms. This protein is known to exhibit different quaternary assemblies between species, existing as free pentamers, decamers (dimers of pentamers) and icosahedrally arranged dodecamers of pentamers. A phylogenetic analysis on eubacterial, fungal and plant LSs allowed us to classify them into two categories: Type I LSs (pentameric or icosahedral) and Type II LSs (decameric). The Rhizobiales represent an order of alpha-proteobacteria that includes, among others, the genera Mesorhizobium, Agrobacterium and Brucella. Here, we present structural and kinetic studies on several LSs from Rhizobiales. Interestingly, Mesorhizobium and Brucella encode both a Type-I LS and a Type-II LS called RibH1 and RibH2, respectively. We show that Type II LSs appear to be almost inactive, whereas Type I LSs present a highly variable catalytic activity according to the genus. Additionally, we have solved four RibH1/RibH2 crystallographic structures from the genera Mesorhizobium and Brucella. The relationship between the active-site architecture and catalytic properties in these isoenzymes is discussed, and a model that describes the enzymatic behavior is proposed. Furthermore, sequence alignment studies allowed us to extend our results to the genus Agrobacterium. Our results suggest that the selective pressure controlling the riboflavin pathway favored the evolution of catalysts with low reaction rates, since the excess of flavins in the intracellular pool in Rhizobiales could act as a negative factor when these bacteria are exposed to oxidative or nitrosative stress.     

Analysis of covariation in an SH3 domain sequence alignment: applications in tertiary contact prediction and the design of compensating hydrophobic core substitutions

We have analyzed sequence covariation in an alignment of 266 non-redundant SH3 domain sequences using chi-squared statistical methods. Artifactual covariations arising from close evolutionary relationships among certain sequence subgroups were eliminated using empirically derived sequence diversity thresholds. This covariation detection method was able to predict residue-residue contacts (side-chain centres of mass within 8 A) in the structure of the SH3 domain with an accuracy of 85 %, which is greater than that achieved in many previous covariation studies. In examining the positions involved most frequently in covariations, we discovered a dramatic over-representation of a subset of five hydrophobic core positions. This covariation information was used to design second and third site substitutions that could compensate for highly destabilizing hydrophobic core substitutions in the Fyn SH3 domain, thus providing experimental data to validate the covariation analysis. The testing of our covariation detection method on 15 other alignments showed that the accuracy of contact prediction is highly variable depending on which sequence alignment is used, and useful levels of prediction accuracy were obtained with only approximately one-third of alignments. The results presented here provide insight into the difficulties inherent in covariation analysis, and suggest that it may have limited usefulness in tertiary structure prediction. On the other hand, our ability to use covariation analysis to design stabilizing combinations of hydrophobic core substitutions attests to its potential utility for gaining deeper insight into the stability determinants and functional mechanisms of proteins with known three-dimensional structures.     

Crystal structure of the Bse634I restriction endonuclease: comparison of two enzymes recognizing the same DNA sequence

Crystal structures of Type II restriction endonucleases demonstrate a conserved common core and active site residues but diverse structural elements involved in DNA sequence discrimination. Comparative structural analysis of restriction enzymes recognizing the same nucleotide sequence might therefore contribute to our understanding of the structural diversity of specificity determinants within restriction enzymes. We have solved the crystal structure of the Bacillus stearothermophilus restriction endonuclease Bse634I by the multiple isomorphous replacement technique to 2.17 Å resolution. Bse634I is an isoschisomer of the Cfr10I restriction enzyme whose crystal structure has been reported previously. Comparative structural analysis of the first pair of isoschisomeric enzymes revealed conserved structural determinants of sequence recognition and catalysis. However, conformations of the N-terminal subdomains differed between Bse634I/Cfr10I, suggesting a rigid body movement that might couple DNA recognition and catalysis. Structural similarities extend to the quaternary structure level: crystal contacts suggest that Bse634I similarly to Cfr10I is arranged as a tetramer. Kinetic analysis reveals that Bse634I is able to interact simultaneously with two recognition sites supporting the tetrameric architecture of the protein. Thus, restriction enzymes Bse634I, Cfr10I and NgoMIV, recognizing overlapping nucleotide sequences, exhibit a conserved tetrameric architecture that is of functional importance.     

Evolutionary information hidden in a single protein structure

The knowledge of conserved sequences in proteins is valuable in identifying functionally or structurally important residues. Generating the conservation profile of a sequence requires aligning families of homologous sequences and having knowledge of their evolutionary relationships. Here, we report that the conservation profile at the residue level can be quantitatively derived from a single protein structure with only backbone information. We found that the reciprocal packing density profiles of protein structures closely resemble their sequence conservation profiles. For a set of 554 nonhomologous enzymes, 74% (408/554) of the proteins have a correlation coefficient > 0.5 between these two profiles. Our results indicate that the three-dimensional structure, instead of being a mere scaffold for positioning amino acid residues, exerts such strong evolutionary constraints on the residues of the protein that its profile of sequence conservation essentially reflects that of its structural characteristics.     

The sequence determinants of cadherin molecules

The sequence and structural analysis of cadherins allow us to find sequence determinants-a few positions in sequences whose residues are characteristic and specific for the structures of a given family. Comparison of the five extracellular domains of classic cadherins showed that they share the same sequence determinants despite only a nonsignificant sequence similarity between the N-terminal domain and other extracellular domains. This allowed us to predict secondary structures and propose three-dimensional structures for these domains that have not been structurally analyzed previously. A new method of assigning a sequence to its proper protein family is suggested: analysis of sequence determinants. The main advantage of this method is that it is not necessary to know all or almost all residues in a sequence as required for other traditional classification tools such as BLAST, FASTA, and HMM. Using the key positions only, that is, residues that serve as the sequence determinants, we found that all members of the classic cadherin family were unequivocally selected from among 80,000 examined proteins. In addition, we proposed a model for the secondary structure of the cytoplasmic domain of cadherins based on the principal relations between sequences and secondary structure multialignments. The patterns of the secondary structure of this domain can serve as the distinguishing characteristics of cadherins.     

Lumazine synthase from Candida albicans as an anti-fungal target enzyme: structural and biochemical basis for drug design

Lumazine synthase is an enzyme involved in riboflavin biosynthesis in many plants and microorganisms, including numerous human pathogens. The fact that the enzymes of the riboflavin biosynthesis pathway are not present in the human or animal host makes them potential targets for anti-infective agents. The crystal structure of lumazine synthase from Candida albicans was solved by molecular replacement and refined at 2.5-Angstrom resolution. The results of crystallographic investigations and sedimentation equilibrium experiments clearly indicated the presence of pentameric assemblies of the enzyme either in crystals or in solution. Isothermal titration calorimetry measurements of the binding reactions of four different inhibitors revealed high affinity for all four compounds with binding constants in the micromolar range. Structural comparison with previously determined structures of the enzyme.ligand complexes of other orthologue allowed modeling of the binding of four different inhibitors into the active site of lumazine synthase from Candida albicans.     

Extreme functional sensitivity to conservative amino acid changes on enzyme exteriors

Mutagenesis studies and alignments of homologous sequences have demonstrated that protein function typically is compatible with a variety of amino-acid residues at most exterior non-active-site positions. These observations have led to the current view that functional constraints on sequence are minimal at these positions. Here, it is shown that this inference assumes that the set of acceptable residues at each position is independent of the overall sequence context. Two approaches are used to test this assumption. First, highly conservative replacements of exterior residues, none of which would cause significant functional disruption alone, are combined until roughly one in five have been changed. This is found to cause complete loss of function in vivo for two unrelated monomeric enzymes: barnase (a bacterial RNase) and TEM-1 beta-lactamase. Second, a set of hybrid sequences is constructed from the 50 %-identical TEM-1 and Proteus mirabilis beta-lactamases. These hybrids match the TEM-1 sequence except for a region at the C-terminal end, where they are random composites of the two parents. All of these hybrids are biologically inactive. In both experiments, complete loss of activity demonstrates the importance of sequence context in determining whether substitutions are functionally acceptable. Contrary to the prevalent view, then, enzyme function places severe constraints on residue identities at positions showing evolutionary variability, and at exterior non-active-site positions, in particular. Homologues sharing less than about two-thirds sequence identity should probably be viewed as distinct designs with their own sets of optimising features.