X-ray analysis (1. 4-A resolution) of avian pancreatic polypeptide: Small globular protein hormone

The crystal structure of avian pancreatic polypeptide (aPP), a 36-residue polypeptide with some hormonal properties, has been determined by using single isomorphous replacement and anomalous scattering to 2.1-A resolution. The phases were extended to 1.4-A resolution by using a modified tangent formula. The molecule contains two regions of secondary structure-an extended polyproline-like helix (residues 1-8) and an alpha-helix (residues 14-31)-that run roughly antiparallel. The packing together of nonpolar groups from these regions gives the molecule a hydrophobic core in spite of its small size. The aPP molecules form a symmetrical dimer in the crystal stabilized principally by interlocking of nonpolar groups from the alpha-helices. The aPP dimers are crosslinked by coordination of Zn(2+); three aPP molecules contribute ligands to each zinc. The coordination geometry is a distorted trigonal bipyramid. The properties of the aPP molecule in solution are consistent with expectations based on the crystal structure. The aPP molecule has several general features in common with the pancreatic hormones insulin and glucagon. All three hormones have complex mechanisms for self-association. Like insulin, aPP seems to have a stable monomeric structure but its biological activity seems to depend on the more flexible COOH-terminal region analogous to the flexible NH(2)-terminal region of glucagon.

Polypeptide hormone-receptor interactions: the structure and receptor binding of insulin and glucagon

Insulin is a small globular protein with a well defined tertiary structure which is closely similar in all species with the exception of certain hystricomorphs such as the guinea pig. Insulin-like growth factor is homologous with insulin and probably has an insulin-like tertiary structure. In contrast glucagon is not a globular protein. It exists as an equilibrium population of conformers with low helix content at physiological concentrations but attains a largely helical conformation on association to trimers. The receptor binding of insulin depends critically on the correct three-dimensional juxtaposition of groups (A1, A21, B25, etc) and involves both hydrophobic and polar interactions. In insulin-like growth factor part of the insulin receptor region is thought to be buried in extra peptide, so explaining its weak binding to insulin receptors. In contrast the glucagon receptor complex probably involves largely hydrophobic contacts which are possible when a helical conformer is formed.

X-ray structures of five renin inhibitors bound to saccharopepsin: exploration of active-site specificity

Saccharopepsin is a vacuolar aspartic proteinase involved in activation of a number of hydrolases. The enzyme has great structural homology to mammalian aspartic proteinases including human renin and we have used it as a model system to study the binding of renin inhibitors by X-ray crystallography. Five medium-to-high resolution structures of saccharopepsin complexed with transition-state analogue renin inhibitors were determined. The structure of a cyclic peptide inhibitor (PD-129,541) complexed with the proteinase was solved to 2.5 A resolution. This inhibitor has low affinity for human renin yet binds very tightly to the yeast proteinase (K(i)=4 nM). The high affinity of this inhibitor can be attributed to its bulky cyclic moiety spanning P(2)-P(3)' and other residues that appear to optimally fit the binding sub-sites of the enzyme. Superposition of the saccharopepsin structure on that of renin showed that a movement of the loop 286-301 relative to renin facilitates tighter binding of this inhibitor to saccharopepsin. Our 2.8 A resolution structure of the complex with CP-108,420 shows that its benzimidazole P(3 )replacement retains one of the standard hydrogen bonds that normally involve the inhibitor's main-chain. This suggests a non-peptide lead in overcoming the problem of susceptible peptide bonds in the design of aspartic proteinase inhibitors. CP-72,647 which possesses a basic histidine residue at P(2), has a high affinity for renin (K(i)=5 nM) but proves to be a poor inhibitor for saccharopepsin (K(i)=3.7 microM). This may stem from the fact that the histidine residue would not bind favourably with the predominantly hydrophobic S(2) sub-site of saccharopepsin.

Rfree and the rfree ratio. II. Calculation Of the expected values and variances of cross-validation statistics in macromolecular least-squares refinement

The free R factor is used routinely as a cross-validation tool in macromolecular crystallography. However, without any means of deriving quantitative estimates of its expected value and variance, its application has been rather subjective and its usefulness therefore somewhat limited. In the first part of this series, estimates of the expected value of the ratio of the free R factor to the standard R factor at the convergence of the structure refinement were given. Here, estimates of the variance of this ratio are given and are compared with the observed deviations from the expected values for a selection of refined structures. It is discussed how errors in the functional form of the structure-factor model as well as other types of errors might influence this ratio.

MAD analyses of yeast 5-aminolaevulinate dehydratase: their use in structure determination and in defining the metal-binding sites

MAD experiments attempting to solve the structure of 5--aminolaevulinic acid dehydratase using Zn and Pb edges are described. The data obtained proved insufficient for a complete structure solution but were invaluable in subsequent identification of metal-binding sites using anomalous difference Fourier analyses once the structure of the enzyme had been solved. These sites include the highly inhibitory substitution of an enzymic cofactor Zn(2+) ion by Pb(2+) ions, which represents a major contribution towards understanding the molecular basis of lead poisoning. The MAD data collected at the Pb edge were also used with isomorphous replacement data from the same Pb co-crystal and a Hg co-crystal to provide the first delineation of the enzyme's quaternary structure. In this MADIR analysis, the Hg co-crystal data were treated as native data. Anomalous difference Fouriers were again used, revealing that Hg(2+) had substituted for the same Zn(2+) cofactor ion as had Pb(2+), a finding of fundamental importance for the understanding of mercury poisoning. In addition, Pt(2+) ions were found to bind at the same place in the structure. The refined structures of the Pb- and the Hg-complexed enzymes are presented at 2.5 and 3.0 A resolution, respectively.

Two structural transitions in membrane pore formation by pneumolysin, the pore-forming toxin of Streptococcus pneumoniae

The human pathogen Streptococcus pneumoniae produces soluble pneumolysin monomers that bind host cell membranes to form ring-shaped, oligomeric pores. We have determined three-dimensional structures of a helical oligomer of pneumolysin and of a membrane-bound ring form by cryo-electron microscopy. Fitting the four domains from the crystal structure of the closely related perfringolysin reveals major domain rotations during pore assembly. Oligomerization results in the expulsion of domain 3 from its original position in the monomer. However, domain 3 reassociates with the other domains in the membrane pore form. The base of domain 4 contacts the bilayer, possibly along with an extension of domain 3. These results reveal a two-stage mechanism for pore formation by the cholesterol-binding toxins.

Rfree and the rfree ratio. I. Derivation of expected values of cross-validation residuals used in macromolecular least-squares refinement

The last five years have seen a large increase in the use of cross validation in the refinement of macromolecular structures using X-ray data. In this technique a test set of reflections is set aside from the working set and the progress of the refinement is monitored by the calculation of a free R factor which is based only on the excluded reflections. This paper gives estimates for the ratio of the free R factor to the R factor calculated from the working set for both unrestrained and restrained refinement. It is assumed that both the X-ray and restraint observations have been weighted correctly and that there is no correlation of errors between the test and working sets. It is also shown that the least-squares weights that minimize the variances of the refined parameters, also approximately minimize the free R factor. The estimated free R-factor ratios are compared with those reported for structures in the Protein Data Bank.

Error estimates of protein structure coordinates and deviations from standard geometry by full-matrix refinement of gammaB- and betaB2-crystallin

Faster workstations with larger memories are making error estimation from full-matrix least-squares refinement a more practicable technique in protein crystallography. Using minimum variance weighting, estimated standard deviations of atomic positions have been calculated for two eye lens proteins from the inverse of a least-squares normal matrix which was full with respect to the coordinate parameters. gammaB-crystallin, refined at 1.49 A yielded average errors in atomic positions which ranged from 0.05 A for main-chain atoms to 0.27 A for unrestrained water molecules. The second structure used in this work was that of betaB2-crystallin refined at 2.1 A resolution where the corresponding average errors were 0.08 and 0.35 A, respectively. The relative errors in atomic positions are dependent on the number and kinds of restraints used in the refinements. It is also shown that minimum variance weighting leads to mean-square deviations from target geometry in the refined structures which are smaller than the variances used in the distance weighting.

Engineering a change in metal-ion specificity of the iron-dependent superoxide dismutase from Mycobacterium tuberculosis-- X-ray structure analysis of site-directed mutants

We have refined the X-ray structures of two site-directed mutants of the iron-dependent superoxide dismutase (SOD) from Mycobacterium tuberculosis. These mutations which affect residue 145 in the enzyme (H145Q and H145E) were designed to alter its metal-ion specificity. This residue is either Gln or His in homologous SOD enzymes and has previously been shown to play a role in active-site interactions since its side-chain helps to coordinate the metal ion via a solvent molecule which is thought to be a hydroxide ion. The mutations were based on the observation that in the closely homologous manganese dependent SOD from Mycobacterium leprae, the only significant difference from the M. tuberculosis SOD within 10 A of the metal-binding site is the substitution of Gln for His at position 145. Hence an H145Q mutant of the M. tuberculosis (TB) SOD was engineered to investigate this residue's role in metal ion dependence and an isosteric H145E mutant was also expressed. The X-ray structures of the H145Q and H145E mutants have been solved at resolutions of 4.0 A and 2.5 A, respectively, confirming that neither mutation has any gross effects on the conformation of the enzyme or the structure of the active site. The residue substitutions are accommodated in the enzyme's three-dimensional structure by small local conformational changes. Peroxide inhibition experiments and atomic absorption spectroscopy establish surprisingly the H145E mutant SOD has manganese bound to it whereas the H145Q mutant SOD retains iron as the active-site metal. This alteration in metal specificity may reflect on the preference of manganese ions for anionic ligands.

X-ray structure of 5-aminolaevulinate dehydratase, a hybrid aldolase

5-Aminolaevulinate dehydratase (ALAD) is a homo-octameric metallo-enzyme that catalyses the formation of porphobilinogen from 5-aminolaevulinic acid. The structure of the yeast enzyme has been solved to 2.3 A resolution, revealing that each subunit adopts a TIM barrel fold with a 39 residue N-terminal arm. Pairs of monomers wrap their arms around each other to form compact dimers and these associate to form a 422 symmetric octamer. All eight active sites are on the surface of the octamer and possess two lysine residues (210 and 263), one of which, Lys 263, forms a Schiff base link to the substrate. The two lysine side chains are close to two zinc binding sites one of which is formed by three cysteine residues (133, 135 and 143) while the other involves Cys 234 and His 142. ALAD has features at its active site that are common to both metallo- and Schiff base-aldolases and therefore represents an intriguing combination of both classes of enzyme. Lead ions, which inhibit ALAD potently, replace the zinc bound to the enzyme's unique triple-cysteine site.

Crystallization of 5-aminolaevulinic acid dehydratase from Escherichia coli and Saccharomyces cerevisiae and preliminary X-ray characterization of the crystals

5-Aminolaevulinic acid dehydratase (ALAD) catalyzes the formation of porphobilinogen from two molecules of 5-aminolaevulinic acid. Both Escherichia coli and Saccharomyces cerevisiae ALADs are homo-octameric enzymes which depend on Zn2+ for catalytic activity and are potently inhibited by lead ions. The E. coli enzyme crystallized in space group I422 (unit cell dimensions a = b = 130.7 A, c = 142.4 A). The best crystals were obtained in the presence of the covalently bound inhibitor laevulinic acid. The yeast enzyme (expressed in E. coli) crystallized in the same space group (I422) but with a smaller unit cell volume (a = b = 103.7 A, c = 167.7 A). High resolution synchrotron data sets were obtained from both E. coli and yeast ALAD crystals by cryocooling to 100 K.

Crystal structure of momordin, a type I ribosome inactivating protein from the seeds of Momordica charantia

A type I ribosome-inactivating protein, extracted and purified from M. charantia seeds, was crystallised by vapour diffusion with polyethylene glycol at pH 7.2. X-ray data were collected to 2.1 A resolution and the structure solved by molecular replacement using the A-chain coordinates of ricin. The overall fold of the protein is similar to ricin but there are differences in secondary structure, on the surface and in the active site cleft. These differences are probably due in part to the evolution of the protein without a B-chain partner. The most extensive reorganisation occurs at the C-terminus whereas Tyr70 shows the greatest change in the active site cleft.

X-ray analysis at 2.0 A resolution of mouse submaxillary renin complexed with a decapeptide inhibitor CH-66, based on the 4-16 fragment of rat angiotensinogen

The structure of mouse submaxillary renin complexed with a decapeptide inhibitor, CH-66 (Piv-His-Pro-Phe-His-Leu-OH-Leu-Tyr-Tyr-Ser-NH2), where Piv denotes a pivaloyl blocking group, and -OH- denotes a hydroxyethylene (-(S)CHOH-CH2-) transition state isostere as a scissile bond surrogate, has been refined to an agreement factor of 0.18 at 2.0 A resolution. The positions of 10,038 protein atoms and 364 inhibitor atoms (4 independent protein inhibitor complexes), as well as of 613 solvent atoms, have been determined with an estimated root-mean-square (r.m.s.) error of 0.21 A. The r.m.s. deviation from ideality for bond distances is 0.026 A, and for angle distances is 0.0543 A. We have compared the three-dimensional structure of mouse renin with other aspartic proteinases, using rigid-body analysis with respect to shifts involving the domain comprising residues 190 to 302. In terms of the relative orientation of domains, mouse submaxillary renin is closest to human renin with only a 1.7 degrees difference in domain orientation. Porcine pepsin (the molecular replacement model) differs structurally from mouse renin by a 6.9 degrees domain rotation, whereas endothiapepsin, a fungal aspartic proteinase, differs by 18.8 degrees. The triple proline loop (residues 292 to 294), which is structurally opposite the active-site "flap" (residues 72 to 83), gives renin a superficial resemblance to the fold of the retroviral proteinases. The inhibitor is bound in an extended conformation along the active-site cleft, and the hydroxyethylene moiety forms hydrogen bonds with both catalytic aspartate carboxylates. The complex is stabilized by hydrogen bonds between the main chain of the inhibitor and the enzyme. All side-chains of the inhibitor are in van der Waals contact with groups in the enzyme and define ten specificity sub-sites. This study shows how renin has compact sub-sites due to the positioning of secondary structure elements, to complementary substitutions and to the residue composition of its loops close to the active site, leading to extreme specificity towards its prohormone substrate, angiotensinogen. We have analysed the micro-environment of each of the buried charged groups in order to predict their ionization states.

Structure of pentameric human serum amyloid P component

The three-dimensional structure of pentameric human serum amyloid P component at high resolution, the first reported for a pentraxin, reveals that the tertiary fold is remarkably similar to that of the legume lectins. Carboxylate and phosphate compounds bind directly to two calcium ions; interactions with a carboxyethylidene ring are mediated by Asn 59 and Gln 148 ligands of the calcium ions. These X-ray results indicate the probable modes of binding of the biologically important ligands, DNA and amyloid fibrils.

X-ray analyses of aspartic proteinases. V. Structure and refinement at 2.0 A resolution of the aspartic proteinase from Mucor pusillus

The structure of mucor pusillus pepsin (EC 3.4.23.6), the aspartic proteinase from Mucor pusillus, has been refined to a crystallographic R-factor of 16.2% at 2.0 A resolution. The positions of 2638 protein atoms, 221 solvent atoms and a sulphate ion have been determined with an estimated root-mean-square (r.m.s.) error of 0.15 to 0.20 A. In the final model, the r.m.s. deviation from ideality for bond distances is 0.022 A, and for angle distances it is 0.050 A. Comparison of the overall three-dimensional structure with other aspartic proteinases shows that mucor pusillus pepsin is as distant from the other fungal enzymes as it is from those of mammalian origin. Analysis of a rigid body shift of residues 190 to 302 shows that mucor pusillus pepsin displays one of the largest shifts relative to other aspartic proteinases (14.4 degrees relative to endothiapepsin) and that changes have occurred at the interface between the two rigid bodies to accommodate this large shift. A new sequence alignment has been obtained on the basis of the three-dimensional structure, enabling the positions of large insertions to be identified. Analysis of secondary structure shows the beta-sheet to be well conserved whereas alpha-helical elements are more variable. A new alpha-helix hN4 is formed by a six-residue insertion between positions 131 and 132. Most insertions occur in loop regions: -5 to 1 (five residues relative to porcine pepsin): 115 to 116 (six residues); 186 to 187 (four residues); 263 to 264 (seven residues); 278 to 279 (four residues); and 326 to 332 (six residues). The active site residues are highly conserved in mucor pusillus pepsin; r.m.s. difference with rhizopuspepsin is 0.37 A for 25 C alpha atom pairs. However, residue 303, which is generally conserved as an aspartate, is changed to an asparagine in mucor pusillus pepsin, possibly influencing pH optimum. Substantial changes have occurred in the substrate binding cleft in the region of S1 and S3 due to the insertion between 115 and 116 and the rearrangement of loop 9-13. Residue Asn219 necessitates a shift in position of substrate main-chain atoms to maintain hydrogen bonding pattern. Invariant residues Asp11 and Tyr14 have undergone a major change in conformation apparently due to localized changes in molecular structure. Both these residues have been implicated in zymogen stability and activation.

The use of protein homologues in the rotation function

The success of molecular replacement depends, in part, on the degree of similarity of the target and search molecules. We have systematically investigated this effect in cross-rotation functions for members of the aspartic proteinase family of enzymes. The influence of various parameters on peak heights was investigated for six search models using magnitude of F(obs) data for two target enzymes. The beneficial effects of high-resolution data and a large radius of integration are most pronounced when target and search molecules have high-percentage identities. Correction for small differences in domain-domain orientation (typically 4-8 degrees) between search and target structures leads to only a marginal improvement in the rotation-function peak height. There is an almost linear relationship between the structural distance, D, a parameter used in cluster analysis to define differences between three-dimensional protein structures, and the height of the cross-rotation-function peaks.

X-ray analyses of peptide-inhibitor complexes define the structural basis of specificity for human and mouse renins

X-ray analyses have defined the three-dimensional structures of crystals of mouse and human renins complexed with peptide inhibitors at resolutions of 1.9 and 2.8 A, respectively. The exquisite specificity of renin arises partly from ordered loop regions at the periphery of the binding cleft. Although the pattern of main-chain hydrogen bonding in other aspartic proteinase inhibitor complexes is conserved in renins, differences in the positions of secondary structure elements (particularly helices) also lead to improved specificity in renins for angiotensinogen substrates.

Structure of oligomeric beta B2-crystallin: an application of the T2 translation function to an asymmetric unit containing two dimers

The molecular structure of the main subunit of the beta-crystallins, components of the vertebrate eye lens, has recently been solved by molecular replacement at 2.1 A resolution [Bax, Lapatto, Nalini, Driessen, Lindley, Mahadevan, Blundell & Slingsby (1990). Nature (London), 347, 776-780]. The protein, beta B2, is a dimer in solution, but a tetramer in the crystal with one subunit in the asymmetric unit of space group I222. Using the crystallographic dimer from this I-centred form the structure of a C222 crystal form of the beta B2 protein with four subunits in the asymmetric unit has now been solved by molecular replacement at 3.3 A. The solution involved the use of a new translation function for non-crystallographic symmetry, based on the T2 function of Crowther & Blow [Acta Cryst. (1967), 23, 544-548].

X-ray analyses of aspartic proteinases. IV. Structure and refinement at 2.2 A resolution of bovine chymosin

The structure of calf chymosin (EC 3.4.23.3), the aspartic proteinase from the gastric mucosa, was solved using the technique of molecular replacement. We describe the use of different search models based on distantly related fungal aspartic proteinases and investigate the effect of using only structurally conserved regions. The structure has been refined to a crystallographic R-factor of 17% at 2.2 A resolution with an estimated co-ordinate error of 0.21 A. In all, 136 water molecules have been located of which eight are internal. The structure of chymosin resembles that of pepsin and other aspartic proteinases. However, there is a considerable rearrangement of the active-site "flap" and, in particular, Tyr75 (pepsin numbering), which forms part of the specificity pockets S1 and S1'. This is probably a consequence of crystal packing. Electrostatic interactions on the edge of the substrate binding cleft appear to account for the restricted proteolysis of the natural substrate kappa-casein by chymosin. The local environment of invariant residues is examined, showing that structural constraints and side-chain hydrogen bonding can play an important role in the conservation of particular amino acids.

MOLPACK: molecular graphics for studying the packing of protein molecules in the crystallographic unit cell

A graphics program, MOLPACK, has been developed on the Silicon Graphics IRIS-4D computer system for displaying the packing of proteins in the crystallographic unit cell. In addition to the normal viewing operations of rotation, translation and scaling, the program has the ability to translate molecules along the cell axes while maintaining their crystallographic equivalent positions within the unit cell. This allows the user to observe the packing of protein molecules generated by molecular replacement, to create a new packing model or to locate an unknown molecule. A special feature of the program is that up to four independent molecules can be manipulated in the asymmetric unit.

X-ray analyses of aspartic proteinases. II. Three-dimensional structure of the hexagonal crystal form of porcine pepsin at 2.3 A resolution

The molecular structure of the hexagonal crystal form of porcine pepsin (EC 3.4.23.1), an aspartic proteinase from the gastric mucosa, has been determined by molecular replacement using the fungal enzyme, penicillopepsin (EC 3.4.23.6), as the search model. This defined the space group as P6522 and refinement led to an R-factor of 0.190 at 2.3 A resolution. The positions of 2425 non-hydrogen protein atoms in 326 residues have been determined and the model contains 371 water molecules. The structure is bilobal, consisting of two predominantly beta-sheet lobes which, as in other aspartic proteinases, are related by a pseudo 2-fold axis. The strands of the mixed beta-sheets (1N and 1C) of each lobe are related by an intra-lobe topological 2-fold symmetry. Two further beta-sheets, 2N and 2C, are each composed of two topologically related beta-hairpins folded below the 1N and 1C sheets. A further six-stranded sheet (3) spans the two lobes and forms a structure resembling an arch upon which the four other sheets reside. The interface between sheets 1N and 1C forms the catalytic centre consisting of absolutely conserved aspartate residues 32 and 215, which are shielded from solvent by a beta-hairpin loop (75 to 78). The crystal structure of a mammalian aspartic proteinase indicates that interactions with substrate may be more extensive on the prime side of the active site cleft than in the fungal enzymes and involve Tyr189 and the loop 290 to 295, perhaps contributing to the transpeptidase activity of pepsin and the specificity of the renins. Comparison with the high-resolution structure of pepsinogen gives a root-mean-square deviation of 0.9 A and reveals that, in addition to local rearrangement at the active site, there appears to be a rigid group movement of part of the C-terminal lobe of pepsin towards the cleft on activation. A large proportion of the absolutely conserved residues in aspartic proteinases are polar and buried. An examination of the pepsin structure reveals that these side-chains are involved in hydrogen-bond interactions with either the main chain of the protein or other conserved side-chains of the enzyme or propart.

The conformation of deamino-oxytocin: X-ray analysis of the 'dry' and 'wet' forms

Two crystal structures of (1 beta-mercaptopropionic acid) deamino-oxytocin are reported. The 'dry form' in space group C2 has cell dimensions a = 27.08 +/- 0.03, b = 9.06 +/- 0.01, c = 22.98 +/- 0.02 A, beta = 102.06 +/- 0.03 with one deamino-oxytocin and six water molecules per asymmetric unit. The 'wet form' in space group P2(1) has cell dimensions a = 27.27 +/- 0.02, b = 9.04 +/- 0.01, c = 23.04 +/- 0.02 A, beta = 102.24 +/- 0.02, with two deamino-oxytocin and 13 water molecules per asymmetric unit. A local twofold parallel to the monoclinic axis gives a pseudo C2 packing. Initial phases of the 'dry form' were calculated by the heavy-atom method from the isomorphous and anomalous difference Pattersons and anomalous difference Fouier synthesis. The structure was refined by using restrained least-squares at 1.2 A resolution to a crystallographic R = 0.10. The molecular replacement method yielded the P2(1) structure that was refined with geometric restraints to R less than 0.09, by using all data to 1.09 A resolution. Deamino-oxytocin consists of a cyclic tocin ring formed by six amino acids, closed by a disulphide bridge, S1-S6, and held by two trans-annular hydrogen bonds N2-O5 and N5-O2 with a type II turn at residues 3 and 4. A flexible tripeptide tail has a loosely hydrogen-bonded type I beta-turn between N9 and O6. The sulphur of cysteine at position 1 is disordered in all the molecules leading to alternative hands of disulphide. The conformational flexibility of Ile 3, Asn 5, Pro 7 side chains and the disulphide bridge is consistent with previous models of oxytocin in which flexibility is necessary for biological activity.

X-ray analyses of aspartic proteinases. The three-dimensional structure at 2.1 A resolution of endothiapepsin

The molecular structure of endothiapepsin (EC 3.4.23.6), the aspartic proteinase from Endothia parasitica, has been refined to a crystallographic R-factor of 0.178 at 2.1 A resolution. The positions of 2389 protein non-hydrogen atoms have been determined and the present model contains 333 solvent molecules. The structure is bilobal, consisting of two predominantly beta-sheet domains that are related by an approximate 2-fold axis. Of approximately 170 residues, 65 are topologically equivalent when one lobe is superimposed on the other. Twenty beta-strands are arranged as five beta-sheets and are connected by regions involving 29 turns and four helices. A central sheet involves three antiparallel strands from each lobe organized around the dyad axis. Each lobe contains a further local dyad that passes through two sheets arranged as a sandwich and relates two equivalent motifs of four antiparallel strands (a, b, c, d) followed by a helix or an irregular helical region. Sheets 1N and 1C, each contain two interpenetrating psi structures contributed by strands c,d,d' and c',d',d, which are related by the intralobe dyad. A further sheet, 2N or 2C, is formed from two extended beta-hairpins from strands b,c and b',c' that fold above the sheets 1N and 1C, respectively, and are hydrogen-bonded around the local intralobe dyad. Asp32 and Asp215 are related by the interlobe dyad and form an intricate hydrogen-bonded network with the neighbouring residues and comprise the most symmetrical part of the structure. The side-chains of the active site aspartate residues are held coplanar and the nearby main chain makes a "fireman's grip" hydrogen-bonding network. Residues 74 to 83 from strands a'N and b'N in the N-terminal lobe form a beta-hairpin loop with high thermal parameters. This "flap" projects over the active site cleft and shields the active site from the solvent region. Shells of water molecules are found on the surface of the protein molecule and large solvent channels are observed within the crystal. There are only three regions of intermolecular contacts and the crystal packing is stabilized by many solvent molecules forming a network of hydrogen bonds. The three-dimensional structure of endothiapepsin is found to be similar to two other fungal aspartic proteinases, penicillopepsin and rhizopuspepsin. Even though sequence identities of endothiapepsin with rhizopuspepsin and penicillopepsin are only 41% and 51%, respectively, a superposition of the three-dimensional structures of these three enzymes shows that 237 residues (72%) are within a root-mean-square distance of 1.0 A.

Crystal structure analysis of deamino-oxytocin: conformational flexibility and receptor binding

Two crystal structures of deamino-oxytocin have been determined at better than 1.1A resolution from isomorphous replacement and anomalous scattering x-ray measurements. In each of two crystal forms there are two closely related conformers with disulfide bridges of different chirality, which may be important in receptor recognition and activation.

The crystal structures of [Met5]enkephalin and a third form of [Leu5]enkephalin: observations of a novel pleated beta-sheet

The structures of [Met5]enkephalin (Tyr-Gly-Gly-Phe-Met) and [Leu5]enkephalin (Tyr-Gly-Gly-Phe-Leu) have been determined from single crystal x-ray diffraction data and refined to residuals of 0.100 and 0.092, respectively. The [Met5]enkephalin structure consists of dimers forming antiparallel beta-sheets extending in the monoclinic ac plane with 10.6 water molecules per dimer. The two molecules, related by pseudo two-fold axes, have similar backbone conformations and similar tyrosine and phenylalanine side-chain conformations. Both methionine residues are disordered and the disorder is different in the two independent molecules. Additional hydrogen bonds connect adjacent dimers to form infinite sheets normal to the b axis. The water molecules are found mainly in the interstices between the sheets. [Leu5]Enkephalin crystallizes as a monohydrate that is isomorphous with the [Met5]enkephalin structure with respect to the beta-sheet but different with respect to the tyrosine and phenylalanine side-chain conformations and water content. The peptide chains in both structures are fully extended and more nearly planar than pleated. The planes of the peptide chains in the dimers form an angle of 143.3 degrees with one another in [Met5]enkephalin and 156.0 degrees in [Leu5]enkephalin. This produces a zigzag pattern or pleat in the beta-sheets perpendicular to the direction of the peptide chains and, therefore, perpendicular to the normal beta-sheet pleat. The average repeat distance between Ni and Ni+2 in the peptide chains of both structures is 7.10 A, versus an ideal value of 6.68 A.

Refinement of the solution structure of the B DNA hexamer 5'd(C-G-T-A-C-G)2 on the basis of inter-proton distance data

A restrained least-squares refinement of the solution structure of the self-complementary B DNA hexamer 5'd(C-G-T-A-C-G)2 is presented. The structure is refined on the basis of 190 inter-proton distances determined by pre-steady-state nuclear Overhauser enhancement measurements. Two refinements were carried out starting from two initial B DNA structures differing by an overall root-mean-square (r.m.s.) difference of 0.32 A. In both cases, the final r.m.s. difference between the experimental and calculated inter-proton distances was 0.12 A compared to 0.61 A and 0.58 A for the two initial structures. The difference between the two refined structures is small, with an overall r.m.s. difference of 0.16 A, and represents the error in the refined co-ordinates. The refined structures have a B-type conformation with local structural variations in backbone and glycosidic bond torsion angles, and base-pair propellor twist, base roll, base tilt and local helical twist angles.

Anisotropic thermal motion and polypeptide secondary structure studied by X-ray analysis at 0.98A resolution

aPP is a 36-amino acid polypeptide which forms a stable globular structure stabilised by hydrophobic interactions between a polyproline-like helix and an alpha-helix. Crystals contain dimers and are crosslinked by coordination through zinc ions leading to a well-ordered lattice which diffracts X-rays to a resolution of 0.98A. This gives a 5:1 ratio of observations-to-parameters even when anisotropic thermal ellipsoids defined by six parameters for each non-hydrogen atom were refined using least-squares techniques. Rigid body refinement of groups within the polypeptide was also undertaken. The relationship of the principal axes of individual thermal ellipsoids and the librations of rigid side groups to features of secondary, tertiary, and quaternary structures of aPP and its interactions with water molecules are described.

Conformational studies on the pancreatic polypeptide hormone family

Pancreatic polypeptide has been extracted and sequenced from a wide range of species. The 36-residue polypeptides have some hormonal characteristics, and show a high degree of sequence homology. Two recently isolated polypeptides, from porcine gut and brain, also show a high degree of sequence homology with the pancreatic polypeptides. It was proposed that these polypeptides were members of a related family. The X-ray determined structure of one member of the family, turkey pancreatic polypeptide, is known to high resolution, but there is no structural information for the others. Studies designed to give an insight into the tertiary structure of these related molecules have been carried out, including model building using interactive computer graphics, circular dichroic spectroscopy and secondary structure prediction using a variety of algorithms. The results indicate that a compact globular conformation, similar to that observed in turkey pancreatic polypeptide may be adopted by all molecules and that this may be more highly conserved than the individual amino acid sequences.

The crystal structures of three non-pancreatic human insulins

X-ray studies on semi-synthetic human insulin have shown that it crystallizes in the rhombohedral space group R3 and is nearly isomorphous with 2 Zn pig insulin. Precession photographs of crystals of human and pig insulins show observable changes in the intensity patterns. Crystallographic analysis and refinement of semi-synthetic human insulin at 1.9 A resolution have shown that its molecular structure is very like that of pig insulin except at the C-terminus of the B chain where the change in sequence occurs. We also report the results of a high resolution crystallographic study of human insulins from different origins. The X-ray diffraction patterns of three non-pancreatic human insulins are indistinguishable from each other and from pancreatic human insulin. Refinement of the structures of the non-pancreatic human insulins has shown that they are identical within the limits of experimental error.

Adaptation of plasminogen activator sequences to known protease structures

The sequences of urokinase (UK) and tissue-type plasminogen activator (TPA) were aligned with those of chymotrypsin, trypsin, and elastase according to their 'structurally conserved regions'. In spite of its trypsin-like specificity UK was model-built on the basis of the chymotrypsin structure because of a corresponding disulfide pattern. The extra disulfide bond falls to cysteines 50 and 111d. Insertions can easily be accommodated at the surface. As they occur similarly in both, UK and TPA, a role in plasminogen recognition may be possible. Of the functional positions known to be involved in substrate or inhibitor binding, Asp 97, Lys 143 and Arg 217 (Leu in TPA) may contribute to plasminogen activating specificity. PTI binding may in part be impaired by structural differences at the edge of the binding pocket.