pH-profile crystal structure studies of C-terminal despentapeptide nitrite reductase from Achromobacter cycloclastes

Crystal structures of C-terminal despentapeptide nitrite reductase (NiRc-5) from Achromobacter cycloclastes were determined from 1.9 to 2.3A at pH 5.0, 5.4, and 6.2. NiRc-5, that has lost about 30% activity, is found to possess quite similar trimeric structures as the native enzyme. Electron density and copper content measurements indicate that the activity loss is not caused by the release of type 2 copper (T2Cu). pH-profile structural comparisons with native enzyme reveal that the T2Cu active center in NiRc-5 is perturbed, accounting for the partial loss of enzyme activity. This perturbation likely results from the less constrained conformations of two catalytic residues, Asp98 and His255. Hydrogen bonding analysis shows that the deletion of five residues causes a loss of more than half the intersubunit hydrogen bonds mediated by C-terminal tail. This study shows that the C-terminal tail plays an important role in controlling the conformations around the T2Cu site at the subunit interface, and helps keep the optimum microenvironment of active center for the full enzyme activity of AcNiR.     

Preliminary crystallographic studies of two C-terminally truncated copper-containing nitrite reductases from Achromobacter cycloclastes: changed crystallizing behaviors caused by residue deletion

The C-terminal segment of copper-containing nitrite reductase from Achromobacter cycloclastes (AcNiR) has been found essential for maintaining both the quaternary structure and the enzyme activity of AcNiR. C-terminal despentapeptide AcNiR (NiRc-5) and desundecapeptide AcNiR (NiRc-11) are two important truncated mutants whose activities and stability have been affected by residue deletion. In this study, the two mutants were crystallized using the hanging drop vapor diffusion method. Crystals of NiRc-5 obtained at pH 5.0 and 6.2 both belonged to the P2(1)2(1)2(1) space group with unit cell parameters a=99.0 A, b=117.4 A, c=122.8 A (pH 5.0) and a=98.9A, b=117.7A, c=123.0A (pH 6.2). NiRc-11 was crystallized in two crystal forms: the tetragonal form belonged to the space group P4(1) with a=b=96.0A and c=146.6A; the monoclinic form belonged to the space group P2(1) with a=86.0A, b=110.1A, c=122.7A, and beta=101.9 degrees. The crystallizing behaviors of the two mutants differed from that of the native enzyme. Such change in combination with residue deletion is also discussed here.     

Alternate substrate binding modes to two mutant (D98N and H255N) forms of nitrite reductase from Alcaligenes faecalis S-6: structural model of a transient catalytic intermediate

High-resolution nitrite soaked oxidized and reduced crystal structures of two active site mutants, D98N and H255N, of nitrite reductase (NIR) from Alcaligenes faecalis S-6 were determined to better than 2.0 A resolution. In the oxidized D98N nitrite-soaked structures, nitrite is coordinated to the type II copper via its oxygen atoms in an asymmetric bidentate manner; however, elevated B-factors and weak electron density indicate that both nitrite and Asn98 are less ordered than in the native enzyme. This disorder likely results from the inability of the N delta 2 atom of Asn98 to form a hydrogen bond with the bound protonated nitrite, indicating that the hydrogen bond between Asp98 and nitrite in the native NIR structure is essential in anchoring nitrite in the active site for catalysis. In the oxidized nitrite soaked H255N crystal structure, nitrite does not displace the ligand water and is instead coordinated in an alternative mode via a single oxygen to the type II copper. His255 is clearly essential in defining the nitrite binding site despite the lack of direct interaction with the substrate in the native enzyme. The resulting pentacoordinate copper site in the H255N structure also serves as a model for a proposed transient intermediate in the catalytic mechanism consisting of a hydroxyl and nitric oxide molecule coordinated to the copper. The formation of an unusual dinuclear type I copper site in the reduced nitrite soaked D98N and H255N crystal structures may represent an evolutionary link between the mononuclear type I copper centers and dinuclear Cu(A) sites.     

The thioesterase domain of the fengycin biosynthesis cluster: a structural base for the macrocyclization of a non-ribosomal lipopeptide

Many secondary metabolic peptides from bacteria and fungi are produced by non-ribosomal peptide synthetases (NRPS) where the final step of biosynthesis is often catalysed by designated thioesterase domains. Here, we report the 1.8A crystal structure of the fengycin thioesterase (FenTE) from Bacillus subtilis F29-3, which catalyses the regio- and stereoselective release and macrocyclization of the antibiotic fengycin from the NRPS template. A structure of the PMSF-inactivated FenTE domain suggests the location of the oxyanion hole and the binding site of the C-terminal residue l-Ile11 of the lipopeptide. Using a combination of docking, molecular dynamics simulations and in vitro activity assays, a model of the FenTE-fengycin complex was derived in which peptide cyclization requires strategic interactions with residues lining the active site canyon.     

Crystal structure of a deubiquitinating enzyme (human UCH-L3) at 1.8 A resolution.

Ubiquitin C-terminal hydrolases catalyze the removal of adducts from the C-terminus of ubiquitin. We have determined the crystal structure of the recombinant human Ubiquitin C-terminal Hydrolase (UCH-L3) by X-ray crystallography at 1.8 A resolution. The structure is comprised of a central antiparallel beta-sheet flanked on both sides by alpha-helices. The beta-sheet and one of the helices resemble the well-known papain-like cysteine proteases, with the greatest similarity to cathepsin B. This similarity includes the UCH-L3 active site catalytic triad of Cys95, His169 and Asp184, and the oxyanion hole residue Gln89. Papain and UCH-L3 differ, however, in strand and helix connectivity, which in the UCH-L3 structure includes a disordered 20 residue loop (residues 147-166) that is positioned over the active site and may function in the definition of substrate specificity. Based upon analogy with inhibitor complexes of the papain-like enzymes, we propose a model describing the binding of ubiquitin to UCH-L3. The UCH-L3 active site cleft appears to be masked in the unliganded structure by two different segments of the enzyme (residues 9-12 and 90-94), thus implying a conformational change upon substrate binding and suggesting a mechanism to limit non-specific hydrolysis.     

Conserved active site residues limit inhibition of a copper-containing nitrite reductase by small molecules

The interaction of copper-containing dissimilatory nitrite reductase from Alcaligenes faecalis S-6 ( AfNiR) with each of five small molecules was studied using crystallography and steady-state kinetics. Structural studies revealed that each small molecule interacted with the oxidized catalytic type 2 copper of AfNiR. Three small molecules (formate, acetate and nitrate) mimic the substrate by having at least two oxygen atoms for bidentate coordination to the type 2 copper atom. These three anions bound to the copper ion in the same asymmetric, bidentate manner as nitrite. Consistent with their weak inhibition of the enzyme ( K i >50 mM), the Cu-O distances in these AfNiR-inhibitor complexes were approximately 0.15 A longer than that observed in the AfNiR-nitrite complex. The binding mode of each inhibitor is determined in part by steric interactions with the side chain of active site residue Ile257. Moreover, the side chain of Asp98, a conserved residue that hydrogen bonds to type 2 copper-bound nitrite and nitric oxide, was either disordered or pointed away from the inhibitors. Acetate and formate inhibited AfNiR in a mixed fashion, consistent with the occurrence of second acetate binding site in the AfNiR-acetate complex that occludes access to the type 2 copper. A fourth small molecule, nitrous oxide, bound to the oxidized metal in a side-on fashion reminiscent of nitric oxide to the reduced copper. Nevertheless, nitrous oxide bound at a farther distance from the metal. The fifth small molecule, azide, inhibited the reduction of nitrite by AfNiR most strongly ( K ic = 2.0 +/- 0.1 mM). This ligand bound to the type 2 copper center end-on with a Cu-N c distance of approximately 2 A, and was the only inhibitor to form a hydrogen bond with Asp98. Overall, the data substantiate the roles of Asp98 and Ile257 in discriminating substrate from other small anions.     

The role of hydrogen bonding at the active site of a cupredoxin: the Phe114Pro azurin variant

The Phe114Pro mutation to the cupredoxin azurin (AZ) leads to a number of structural changes at the active site attributed to deletion of one of the hydrogen bonds to the Cys112 ligand, removal of the bulky phenyl group from the hydrophobic patch of the protein, and steric interactions made by the introduced Pro. The remaining hydrogen bond between the coordinating thiolate and the backbone amide of Asn47 is strengthened. At the type-1 copper site, the Cu(II)-O(Gly45) axial interaction decreases, while the metal moves out of the plane formed by the equatorial His46, Cys112, and His117 ligands, shortening the bond to the axially coordinating Met121. The resulting distorted tetrahedral geometry is distinct from the trigonal bipyramidal arrangement in the wild-type (WT) protein. The unique position of the main S(Cys) --> Cu(II) ligand-to-metal charge-transfer transition in AZ (628 nm) has shifted in the Phe114Pro variant to a value that is more typical for cupredoxins (599 nm). This probably occurs because of the removal of the Phe114-Cys112 hydrogen bond. The Phe114Pro mutation results in a 90 mV decrease in the reduction potential of AZ, and removal of the second hydrogen bond to the Cys ligand seems to be the major cause of this change. The C-terminal His117 ligand does not protonate in the reduced Phe114Pro AZ variant, which suggests that none of the structural features altered by the mutation are responsible for the absence of this effect in the WT protein. Upon reduction, the copper displaces further from the equatorial ligand plane and the Cu-S(Met121) bond length decreases. These changes are larger than those seen in the WT protein and contribute to the order of magnitude decrease in the intrinsic electron-transfer capabilities of the Phe114Pro variant.     

Crystal structure of a NO-forming nitrite reductase mutant: an analog of a transition state in enzymatic reaction

I257E was obtained by site directed mutagenesis of nitrite reductase from Achromobacter cycloclastes. The mutant has no enzyme activity. Its crystal structure determined at 1.65A resolution shows that the side-chain carboxyl group of the mutated residue, Glu257, coordinates with the type 2 copper in the mutant and blocks the contact between the type 2 copper and its solvent channel, indicating that the accessibility of the type 2 copper is essential for maintaining the activity of nitrite reductase. The carboxylate is an analog of the substrate, nitrite, but the distances between the type 2 copper and the two oxygen atoms of the side-chain carboxyl group are reversed in comparison to the binding of nitrite to the native enzyme. In the mutant, both the type 2 copper and the N epsilon atom on the imidazole ring of its coordinated residue His135 move in the substrate binding direction relative to the native enzyme. In addition, an EPR study showed that the type 2 copper in the mutant is in a reduced state. We propose that mutant I257E is in a state corresponding to a transition state in the enzymatic reaction.     

Primary structure homologies between two zinc metallopeptidases, the neutral endopeptidase 24.11 ("enkephalinase") and thermolysin, through clustering analysis

Analogies in the sequences of two related zinc metallopeptidases, the bacterial thermolysin (316 amino acids) and the recently cloned neutral endopeptidase 24.11 ("enkephalinase", 749 amino acids), have been demonstrated by a hydrophobic cluster analysis method derived from the Lim theory. Two sequence alignments are proposed for the entire primary structure of thermolysin and the C-terminal part of endopeptidase 24.11. Except for an arginine residue, all the amino acids involved in the active site of thermolysin have been retrieved in both models of endopeptidase 24.11 within conserved clustered structures. The first model is characterized by a deletion of the Ca2+-binding coil present in thermolysin and the second by replacement of this coil by two alpha-helices. In both models an Arg residue can be located in the active site of the neutral endopeptidase.     

Rigidity and flexibility of dipeptidyl peptidase IV: crystal structures of and docking experiments with DPIV

Dipeptidyl peptidase IV (DPIV) is an alpha,beta-hydrolase-like serine exopeptidase, which removes dipeptides, preferentially with a C-terminal l-Pro residue, from the N terminus of longer peptide substrates. Previously, we determined the tetrameric 1.8A crystal structure of native porcine DPIV. Each monomer is composed of a beta-propeller and a catalytic domain, which together embrace an internal cavity housing the active centre. This cavity is connected to the bulk solvent by a "propeller opening" and a "side opening". Here, we analyse DPIV complexes with a t-butyl-Gly-Pro-Ile tripeptide, Pro-boroPro, a piperazine purine compound, and aminoethyl phenyl sulfonylfluoride. The latter two compounds bind to the active-site groove in a compact and a quite bulky manner, respectively, causing considerable shifts of the catalytic Ser630 side-chain and of the Tyr547 phenolic group, which forms the oxyanion hole. The tripeptide, mimicking a peptide substrate, is clamped to the active site through tight interactions via its N-terminal alpha-ammonium group, the P2 carbonyl group, the P1-l-Pro side-chain, the C-terminal carboxylate group, and the stable orthoacid ester amide formed between the scissile peptide carbonyl group and Ser630 O(gamma). This stable trapping of the tripeptide could be due to stabilization of the protonated His740 imidazolium cation by the adjacent negatively charged C-terminal carboxylate group, preventing proton transfer to the leaving group nitrogen atom. Docking experiments with the compact rigid 58 residue protein aprotinin, which had been shown to be processed by DPIV, indicate that the Arg1-Pro2 N terminus can access the DPIV active site only upon widening of its side openings, probably by separation of the first and the last propeller blades, and/or of the catalytic and the propeller domain.     

Influencing the monophenolase/diphenolase activity ratio in tyrosinase

Tyrosinase is a type 3 copper enzyme with great potential for production of commercially valuable diphenols from monophenols. However, the use of tyrosinase is limited by its further oxidation of diphenols to quinones. We recently determined the structure of the Bacillus megaterium tyrosinase revealing a residue, V218, which we proposed to take part in positioning of substrates within the active site. In the structure of catechol oxidase from Ipomoea batatas, the lack of monophenolase activity was attributed to the presence of F261 near CuA. Consequently, we engineered two variants, V218F and V218G. V218F was expected to have a decreased monophenolase activity, due to the bulky residue extending into the active site. Surprisingly, both V218F and V218G exhibited a 9- and 4.4-fold higher monophenolase/diphenolase activity ratio, respectively. X-ray structures of variant V218F display a flexibility of the phenylalanine residue along with an adjacent histidine, which we propose to be the source of the change in activity ratio.     

Crystal structure and site-directed mutagenesis of a nitroalkane oxidase from Streptomyces ansochromogenes

Nitroalkane oxidase (NAO) catalyzes neutral nitroalkanes to their corresponding aldehydes or ketones, hydrogen peroxide and nitrite. The crystal structure of NAO from Streptomyces ansochromogenes was determined; it consists of two domains, a TIM barrel domain bound to FMN and C-terminal domain with a novel folding pattern. Site-directed mutagenesis of His¹⁷⁹, which is spatially adjacent to FMN, resulted in the loss of enzyme activity, demonstrating that this amino acid residue is important for catalysis. The crystal structure of mutant H179D-nitroethane was also analyzed. Interestingly, Sa-NAO shows the typical function as nitroalkane oxidase but its structure is similar to that of 2-nitropropane dioxygenase. Overall, these results suggest that Sa-NAO is a novel nitroalkane oxidase with TIM barrel structure.     

Crystallographic studies on the role of the C-terminal segment of human angiogenin in defining enzymatic potency

Human angiogenin (Ang) is an RNase in the pancreatic RNase superfamily that induces angiogenesis. Its catalytic activity is comparatively weak, but nonetheless critical for biological activity. The crystal structure of Ang has shown that enzymatic potency is attenuated in part by the obstructive positioning of Gln117 within the B(1) pyrimidine binding pocket, and that the C-terminal segment of residues 117-123 must reorient for Ang to bind and cleave RNA. The native closed conformation appears to be stabilized by Gln117-Thr44 and Asp116-Ser118 hydrogen bonds, as well as hydrophobic packing of Ile119 and Phe120. Consistent with this view, Q117G, D116H, and I119A/F120A variants are 4-30-fold more active than Ang. Here we have determined crystal structures for these variants to examine the structural basis for the activity increases. In all three cases, the C-terminal segment remains obstructive, demonstrating that none of the residues that has been replaced is essential for maintaining the closed conformation. The Q117G structure shows no changes other than the loss of the side chain of residue 117, whereas those of D116H and I119A/F120A reveal C-terminal perturbations beyond the replacement site, suggesting that the native closed conformation has been destabilized. Thus, the interactions of Gln117 seem to be less important than those of residues 116, 119, and 120 for stabilization. In D116H, His116 does not replicate either of the hydrogen bonds of Asp116 with Ser118 and instead forms a water-mediated interaction with catalytic residue His114; residues 117-121 deviate significantly from their positions in Ang. In I119A/F120A, the segment of residues 117-123 has become highly mobile and all of the interactions thought to position Gln117 have been weakened or lost; the space occupied by Phe120 in Ang is partially filled by Arg101, which has moved several angstroms. A crystal structure was also determined for the deletion mutant des(121-123), which has 10-fold reduced activity toward large substrates. The structure is consistent with the earlier proposal that residues 121-123 form part of a peripheral substrate binding subsite, but also raises the possibility that changes in the position of another residue, Lys82, might be responsible for the decreased activity of this variant.     

Catalytic roles for two water bridged residues (Asp-98 and His-255) in the active site of copper-containing nitrite reductase

Two active site residues, Asp-98 and His-255, of copper-containing nitrite reductase (NIR) from Alcaligenes faecalis have been mutated to probe the catalytic mechanism. Three mutations at these two sites (D98N, H255D, and H255N) result in large reductions in activity relative to native NIR, suggesting that both residues are involved intimately in the reaction mechanism. Crystal structures of these mutants have been determined using data collected to better than 1. 9-A resolution. In the native structure, His-255 Nepsilon2 forms a hydrogen bond through a bridging water molecule to the side chain of Asp-98, which also forms a hydrogen bond to a water or nitrite oxygen ligated to the active site copper. In the D98N mutant, reorientation of the Asn-98 side chain results in the loss of the hydrogen bond to the copper ligand water, consistent with a negatively charged Asp-98 directing the binding and protonation of nitrite in the native enzyme. An additional solvent molecule is situated between residues 255 and the bridging water in the H255N and H255D mutants and likely inhibits nitrite binding. The interaction of His-255 with the bridging water appears to be necessary for catalysis and may donate a proton to reaction intermediates in addition to Asp-98.     

Crystal structure of a theta-class glutathione transferase.

Glutathione S-transferases (GSTs) are a family of enzymes involved in the cellular detoxification of xenotoxins. Cytosolic GSTs have been grouped into four evolutionary classes for which there are representative crystal structures of three of them. Here we report the first crystal structure of a theta-class GST. So far, all available GST crystal structures suggest that a strictly conserved tyrosine near the N-terminus plays a critical role in the reaction mechanism and such a role has been convincingly demonstrated by site-directed mutagenesis. Surprisingly, the equivalent residue in the theta-class structure is not in the active site, but its role appears to have been replaced by either a nearby serine or by another tyrosine residue located in the C-terminal domain of the enzyme.     

Localization of the C-terminus of rat glutathione S-transferase A1-1: crystal structure of mutants W21F and W21F/F220Y

Twelve C-terminal residues of human glutathione S-transferase A1-1 form a helix in the presence of glutathione-conjugate, or substrate alone, and partly cover the active site. According to X-ray structures, the helix is disordered in the absence of glutathione, but it is not known if it is helical and delocalized, or in a random-coil conformation. Mutation to a tyrosine of residue 220 within this helix was previously shown to affect the pK(a) of Tyr-9 at the active site, in the apo form of the enzyme, and it was proposed that an on-face hydrogen bond between Tyr-220 and Tyr-9 provided a means for affecting this pK(a). In the current study, X-ray structures of the W21F and of the C-terminal mutation, W21F/F220Y, with glutathione sulfonate bound, show that the C-terminal helix is disordered (or delocalized) in the W21F crystal but is visible and ordered in a novel location, a crystal packing crevice, in one of three monomers in the W21F/F220Y crystal, and the proposed hydrogen bond is not formed. Fluorescence spectroscopy studies using an engineered F222W mutant show that the C-terminus remains delocalized in the absence of glutathione or when only the glutathione binding site is occupied, but is ordered and localized in the presence of substrate or conjugate, consistent with these and previous crystallographic studies. Proteins 2001;42:192-200.     

Carboxy-terminus recruitment induced by substrate binding in eukaryotic fructose bis-phosphate aldolases

The crystal structures of Leishmania mexicana fructose-1,6-bis(phosphate) aldolase in complex with substrate and competitive inhibitor, mannitol-1,6-bis(phosphate), were solved to 2.2 A resolution. Crystallographic analysis revealed a Schiff base intermediate trapped in the native structure complexed with substrate while the inhibitor was trapped in a conformation mimicking the carbinolamine intermediate. Binding modes corroborated previous structures reported for rabbit muscle aldolase. Amino acid substitution of Gly-312 to Ala, adjacent to the P1-phosphate binding site and unique to trypanosomatids, did not perturb ligand binding in the active site. Ligand attachment ordered amino acid residues 359-367 of the C-terminal region (353-373) that was disordered beyond Asp-358 in the unbound structure, revealing a novel recruitment mechanism of this region by aldolases. C-Terminal peptide ordering is triggered by P1-phosphate binding that induces conformational changes whereby C-terminal Leu-364 contacts P1-phosphate binding residue Arg-313. C-Terminal region capture synergizes additional interactions with subunit surface residues, not perturbed by P1-phosphate binding, and stabilizes C-terminal attachment. Amino acid residues that participate in the capturing interaction are conserved among class I aldolases, indicating a general recruitment mechanism whereby C-terminal capture facilitates active site interactions in subsequent catalytic steps. Recruitment accelerates the enzymatic reaction by using binding energy to reduce configurational entropy during catalysis thereby localizing the conserved C-terminus tyrosine, which mediates proton transfer, proximal to the active site enamine.     

Influence of substrate modification and C-terminal truncation on the active site structure of substrate-bound heme oxygenase from Neisseriae meningitidis. A 1H NMR study

Heme oxygenase (HO), from the pathogenic bacterium N. meningitidis(NmHO), which secures host iron, shares many properties with mammalian HOs but also exhibits some key differences. The crystal structure appears more compact, and the crystal-undetected C-terminus interacts with substrate in solution. The unique nature of substrate-protein, specifically pyrrole-I/II-helix-2, peripheral interactions in NmHO are probed by 2D (1)H NMR to reveal unique structural features controlling substrate orientation. The thermodynamics of substrate orientational isomerism are mapped for substrates with individual vinyl → methyl → hydrogen substitutions and with enzyme C-terminal deletions. NmHO exhibits significantly stronger orientational preference, reflecting much stronger and selective pyrrole-I/II interactions with the protein matrix, than in mammalian HOs. Thus, replacing bulky vinyls with hydrogens results in a 180° rotation of substrate about the α,γ-meso axis in the active site. A "collapse" of the substrate pocket as substrate size decreases is reflected in movement of helix-2 toward the substrate as indicated by significant and selective increased NOESY cross-peak intensity, increase in steric Fe-CN tilt reflected in the orientation of the major magnetic axis, and decrease in steric constraints controlling the rate of aromatic ring reorientation. The active site of NmHO appears "stressed" for native protohemin, and its "collapse" upon replacing vinyls by hydrogen leads to a factor ~10(2) increase in substrate affinity. Interaction of the C-terminus with the active site destabilizes the crystallographic protohemin orientation by ~0.7 kcal/mol, which is consistent with optimizing the His207-Asp27 H-bond. Implications of the active site "stress" for product release are discussed.     

Structure determinants of substrate specificity of hydroxynitrile lyase from Manihot esculenta

Tryptophan 128 of hydroxynitrile lyase of Manihot esculenta (MeHNL) covers a significant part of a hydrophobic channel that gives access to the active site of the enzyme. This residue was therefore substituted in the mutant MeHNL-W128A by alanine to study its importance for the substrate specificity of the enzyme. Wild-type MeHNL and MeHNL-W128A showed comparable activity on the natural substrate acetone cyanohydrin (53 and 40 U/mg, respectively). However, the specific activities of MeHNL-W128A for the unnatural substrates mandelonitrile and 4-hydroxymandelonitrile are increased 9-fold and approximately 450-fold, respectively, compared with the wild-type MeHNL. The crystal structure of the MeHNL-W128A substrate-free form at 2.1 A resolution indicates that the W128A substitution has significantly enlarged the active-site channel entrance, and thereby explains the observed changes in substrate specificity for bulky substrates. Surprisingly, the MeHNL-W128A--4-hydroxybenzaldehyde complex structure at 2.1 A resolution shows the presence of two hydroxybenzaldehyde molecules in a sandwich type arrangement in the active site with an additional hydrogen bridge to the reacting center.     

Annexin XII E105K crystal structure: identification of a pH-dependent switch for mutant hexamerization

Annexins are a family of calcium- and phospholipid-binding proteins involved with numerous cellular processes including membrane fusion, ion channel activity, and heterocomplex formation with other proteins. The annexin XII (ANXB12) crystal structure presented evidence that calcium mediates the formation of a hexamer through a novel intermolecular calcium-binding site [Luecke et al. (1995) Nature 378, 512-515]. In an attempt to disrupt hexamerization, we mutated a conserved key ligand in the intermolecular calcium-binding site, Glu105, to lysine. Despite its occurrence in a new spacegroup, the 1.93 A resolution structure reveals a hexamer with the Lys105 epsilon-amino group nearly superimposable with the original intermolecular calcium position. Our analysis shows that the mutation is directly involved in stabilizing the hexamer. The local residues are reoriented to retain affinity between the two trimers via a pH-dependent switch residue, Glu76, which is now protonated, allowing it to form tandem hydrogen bonds with the backbone carbonyl and nitrogen atoms of Thr103 located across the trimer interface. The loss of the intermolecular calcium-binding site is recuperated by extensive hydrogen bonding favoring hexamer stabilization. The presence of this mutant structure provides further evidence for hexameric annexin XII, and possible in vivo roles are discussed.