Substrates of DNA polymerases with planar conformation of sugar: model of substrate transition state?

Several years ago, we published an hypothesis concerning conformation of the glycone moiety of different substrates in active centers of several DNA metabolizing enzymes (Nucleosides & Nucleotides 1993, 12, 649-670). This hypothesis prompted us to further study the subtle conformational changes on substrates of DNA polymerases. Data collected in our, as well as other laboratories, have been analyzed, and models of active centers of different DNA polymerases are discussed below. Based on the model of substrate requirements, we now can divide DNA polymerases into two distinguished classes.     

Influence of cofactor binding and active site occupancy on the conformation of the macromolecular substrate exosite of factor VIIa

The catalytic activity of the trypsin-like serine protease coagulation factor VIIa is allosterically regulated. In this work, we employed monoclonal antibodies as probes to analyze conformational changes in the VII protease domain that are induced by zymogen activation, cofactor tissue factor (TF) binding, and active site occupancy. The epitopes of three monoclonal antibodies were mapped using a panel of 57 individual alanine replacement mutants in the protease domain. Two of the antibodies had typical "hot spot" epitopes in a basic cluster above the active site cleft and antibody binding to these epitopes was not affected by zymogen activation, TF binding, or active site occupancy. In contrast, the binding kinetics of VII/VIIa to a monoclonal antibody that mapped to an extended epitope overlapping with the macromolecular substrate exosite was affected by each of the conformational transitions of the VIIa protease domain. The changes in antibody affinity are consistent with a transition from zymogen VII to the TF.VIIa complex, with free enzyme VIIa as an intermediate that retains some zymogen-like features responsible for its low catalytic activity. In contrast, active site occupancy resulted in effects that were qualitatively different from the effects of zymogen activation on the antibody epitope. This provides novel insight into the conformational interdependence between the active site, the region for macromolecular substrate recognition, and the cofactor binding exosite of this allosterically regulated serine protease.     

Conformational changes of active sites during refolding of urea-denatured creatine kinase

The course of the recovery of the active site conformation during refolding of urea-denatured creatine kinase (ATP:creatine N-phosphotransferase, EC 2.7.3.2) has been studied. The recovery of the active site conformation has been followed by changes in probe fluorescence of active site of the enzyme labeled by o-phthaldehyde and has been shown to be a triphasic process. A comparison of the rate constants for the conformational recovery of the active site with those for the recovery of enzyme catalytic activity shows that these are synchronized. The results obtained show clearly that, although recovery of partial activity and conformation of the active site occur in reactions with nearly the same rate as those of the refolding processes observed, the complete recovery of the enzyme activity and the active site conformation can only be obtained long after any detectable conformational change of whole enzyme molecule.     

A ligand-gated, hinged loop rearrangement opens a channel to a buried artificial protein cavity

Conformational changes that gate the access of substrates or ligands to an active site are important features of enzyme function. In this report, we describe an unusual example of a structural rearrangement near a buried artificial cavity in cytochrome c peroxidase that occurs on binding protonated benzimidazole. A hinged main-chain rotation at two residues (Pro 190 and Asn 195) results in a surface loop rearrangement that opens a large solvent-accessible channel for the entry of ligands to an otherwise inaccessible binding site. The trapping of this alternate conformational state provides a unique view of the extent to which protein dynamics can allow small molecule penetration into buried protein cavities.     

Structural mapping of the catalytic mechanism for a mammalian phosphoinositide-specific phospholipase C

The crystal structures of various ternary complexes of phosphoinositide-specific phospholipase C-delta 1 from rat with calcium and inositol phosphates have been determined at 2.30-2.95 A resolution. The inositol phosphates used in this study mimic the binding of substrates and the reaction intermediate and include D-myo-inositol-1,4,5-trisphosphate, D-myo-inositol-2,4, 5-trisphosphate. D-myo-inositol-4,5-bisphosphate, and D,1-myo-inositol-2-methylene-1,2-cycli?monophosphonate. The complexes exhibit an almost invariant mode of binding in the active site, each fitting edge-on into the active site and interacting with both the enzyme and the catalytic calcium at the bottom of the active site. Most of the active site residues do not undergo conformational changes upon binding either calcium or inositol phosphates. The structures are consistent with bidentate liganding of the catalytic calcium to the inositol phosphate intermediate and transition state. The complexes suggest explanations for substrate preference, pH optima, and ratio of cyclic to acyclic reaction products. A reaction mechanism is derived that supports general acid/base catalysis in a sequential mechanism involving a cyclic phosphate intermediate and rules out a parallel mechanism where acyclic and cyclic products are simultaneously generated.     

Molecular modelling studies of lysozyme catalysed hydrolysis of synthetic substrates

Kinetic data for the lysozyme catalysed hydrolysis of aryl chitooligosides were surveyed. Both electron-donating and electron-withdrawing substituents on the departing aryl aglycones enhance the rate of hydrolyses. The parallel pH-rate profiles implicate that identical catalytic residues are involved in the hydrolytic fission of the glycosyl-aryloxy bond of these two groups of synthetic substrates. Molecular modelling studies of lysozyme complexes with aryl diN-acetyl chitobiosides and their intermediates were performed. The two synthetic substrates bearing aryl aglycones with opposite electronic effects bind to the active site of lysozyme in different conformations. Based on the energetic and geometric considerations, the oxocarbonium ion whose pyranose ring D in a sofa conformation is the most plausible reaction intermediate for the lysozyme catalysed hydrolysis of the synthetic substrates. The modelling study also suggests that considerable conformational changes of both the lysozyme binding site and the chitobiosyl group accompany the formation of the glycosyl enzyme intermediate. In particular, the chitobiosyl group undergoes a dislocation of the pyranose ring C from the subsite C and a constraint of the pyranose ring D to form a boat conformer.     

Mechanism of pancreatic lipase action. 1. Interfacial activation of pancreatic lipase

Hydrolysis of dissolved p-nitrophenyl acetate by pancreatic lipase follows the classical acyl enzyme pathway already proposed for other esterases. Kinetic parameters of the hydrolysis have been determined. The turnover rate of the reaction is many orders of magnitude slower than that for the natural emulsified substrates. Nevertheless, several arguments are in favor of the specificity of this hydrolysis: (1) triacetin, which resembles the usual substrates for the enzyme, is also hydrolyzed very slowly in solution; (2) dissolved triacetin and tripropionin are competitive inhibitors for the p-nitrophenyl acetate hydrolysis; (3) the same chemical structural features which are required in the case of emulsified substrates are also necessary to promote hydrolysis of dissolved p-nitrophenyl esters. This suggests that the same active site (or part of the same active site) is responsible for hydrolysis of both p-nitrophenyy acetate and specific emulsified substrates. Since deacylation is the rate-limiting step in the catalysis of p-nitrophenyl acetate, the intermediate acetyl enzyme can be isolated by trapping it at pH 5.0. Kinetic competence of this intermediate has been demonstrated. Hydrolysis by pancreatic lipase of dissolved monomeric p-nitrophenyl acetate and triacetin is considerably enhanced (100- to 500-fold) by various interfaces. This suggests that at least the deacylation step, which is rate limiting in absence of interface, is accelerated by the presence of inert interfaces. Siliconized glass beads were directly shown to accelerate the deacylation of isolated [3H]acetyl lipase by at least a hundred times. This step does not directly involve the ester substrate.Thus, it is suggested that a part of the activation of lipase at interfaces may be due to a conformational change resulting from adsorption.     

Metal-substrate interactions facilitate the catalytic activity of the bacterial phosphotriesterase

The bacterial phosphotriesterase from Pseudomonas diminuta is a zinc metalloenzyme which catalyzes the hydrolysis of a variety of organophosphorus nerve agents with high efficiency. The active site of the enzyme consists of a coupled binuclear metal center embedded within a cluster of histidine residues. Potential protein-substrate interactions at the active site were probed by a systematic variation of metal identity, leaving group potential, phosphate host, and amino acid replacement. In order to determine the roles of these metal ions in binding and catalysis, the microscopic rate constants and kinetic parameters were obtained with various divalent cations. The divalent cations that were utilized in this investigation consisted of Co2+, Ni2+, Cd2+, Zn2+, Mn2+, and the mixed-metal Zn2+/Cd2+ hybrid. The leaving group potential and phosphate host were varied by altering the pKa of the departing substituted phenol or thiophenol in either a diethyl phosphate or a diethyl thiophosphate substrate. The Br?nsted plots for the nonenzymatic hydroxide catalyzed hydrolysis of these substrates showed a linear dependence between the pseudo-first-order rate constant and the pKa of the leaving group. Enzymatic activities of the wild-type enzyme with these same substrates varied by over 7 orders of magnitude over the entire experimental pKa range (4.1-10.3), and the corresponding Br?nsted plots were nonlinear. Those substrates with leaving groups with high pKa values were limited by the rate of bond cleavage while those substrates having leaving groups with low pKa values were limited by a conformational change or binding event. Thiophosphate substrates having leaving groups with high pKa values were better substrates than the corresponding phosphate analogues. These results are consistent with the direct coordination of one or both metal ions with the phosphoryl sulfur or oxygen atom of the substrate. A large dependence of the rate on the leaving group rules out the possibility of protonation of the leaving group or electrostatic interaction of the leaving group oxygen (or sulfur) with a metal ion or cationic group at the active site. The large differences in the size of the beta lg over the range of metal ions utilized by the enzyme indicate that the metal ions polarize the phosphoryl group and alter the structure of the transition state. The values of V/K(m) for the enzyme-catalyzed hydrolysis for a series of substituted thiophenol analogues were 10(2)-10(3)-fold smaller than those obtained for the hydrolysis of the corresponding phenolic substrates, suggesting that the bulkier sulfur substituent in the leaving group may induce conformational restrictions at the active site. With the zinc-substituted H201N mutant enzyme, there was a large decrease in the rate of phosphotriester hydrolysis but essentially no change in the rate of thiophosphotriester hydrolysis relative to the values observed for the zinc-substituted wild-type enzyme. These results suggest that a direct perturbation in the ligand structure of the binuclear metal center induces alterations in the mechanism of substrate hydrolysis.     

Pig heart fumarase contains two distinct substrate-binding sites differing in affinity

A eukaryotic fumarase is for the first time unequivocally shown to contain two distinct substrate-binding sites. Pig heart fumarase is a tetrameric enzyme consisting of four identical subunits of 50 kDa each. Besides the true substrates L-malate and fumarate, the active sites (sites A) also bind their analogs D-malate and oxaloacetate, as well as the competitive inhibitor glycine. The additional binding sites (sites B) on the other hand also bind the substrates and their analogs D-malate and oxaloacetate, as well as L-aspartate which is not an inhibitor. Depending on the pH, the affinity of sites B for ligands (Kd being in the millimolar range) is 1-2 orders of magnitude lower than the affinity of sites A (of which Kd is in the micromolar range). However, saturating sites B results in an increase in the overall activity of the enzyme. The benzenetetracarboxyl compound pyromellitic acid displays very special properties. One molecule of this ligand is indeed able to bind into a site A and a site B at the same time. Four molecules of pyromellitic acid were found to bind per molecule fumarase, and the affinity of the enzyme for this ligand is very high (Kd = 0.6 to 2.2 microM, depending on the pH). Experiments with this ligand turned out to be crucial in order to explain the results obtained. An essential tyrosine residue is found to be located in site A, whereas an essential methionine residue resides in or near site B. Upon limited proteolysis, a peptide of about 4 kDa is initially removed, probably at the C-terminal side; this degradation results in inactivation of the enzyme. Small local conformational changes in the enzyme are picked up by circular dichroism measurements in the near-UV region. This spectrum is built up of two tryptophanyl triplets, the first one of which is modified upon saturating the active sites (A), and the second one upon saturating the low affinity binding sites (B).     

Structures of native and complexed complement factor D: implications of the atypical His57 conformation and self-inhibitory loop in the regulation of specific serine protease activity

Factor D is a serine protease essential for the activation of the alternative pathway of complement. The structures of native factor D and a complex formed with isatoic anhydride inhibitor were determined at resolution of 2.3 and 1.5 A, respectively, in an isomorphous monoclinic crystal form containing one molecule per asymmetric unit. The native structure was compared with structures determined previously in a triclinic cell containing two molecules with different active site conformations. The current structure shows greater similarity with molecule B in the triclinic cell, suggesting that this may be the dominant factor D conformation in solution. The major conformational differences with molecule A in the triclinic cell are located in four regions, three of which are close to the active site and include some of the residues shown to be critical for factor D catalytic activity. The conformational flexibility associated with these regions is proposed to provide a structural basis for the previously proposed substrate-induced reversible conformational changes in factor D. The high-resolution structure of the factor D/isatoic anhydride complex reveals the binding mode of the mechanism-based inhibitor. The higher specificity towards factor D over trypsin and thrombin is based on hydrophobic interactions between the inhibitor benzyl ring and the aliphatic side-chain of Arg218 that is salt bridged with Asp189 at the bottom of the primary specificity (S1) pocket. Comparison of factor D structural variants with other serine protease structures revealed the presence of a unique "self-inhibitory loop". This loop (214-218) dictates the resting-state conformation of factor D by (1) preventing His57 from adopting active tautomer conformation, (2) preventing the P1 to P3 residues of the substrate from forming anti-parallel beta-sheets with the non-specific substrate binding loop, and (3) blocking the accessibility of Asp189 to the positive1y charged P1 residue of the substrate. The conformational switch from resting-state to active-state can only be induced by the single macromolecular substrate, C3b-bound factor B. This self-inhibitory mechanism is highly correlated with the unique functional properties of factor D, which include high specificity toward factor B, low esterolytic activity toward synthetic substrates, and absence of regulation by zymogen and serpin-like or other natural inhibitors in blood.     

Activation of protein kinase C induces cortical granule exocytosis in a Ca(2+)-independent manner, but not the resumption of cell cycle in porcine eggs

N-Acetylneuraminic acid (1) is a common sugar in many biological recognition processes. Neuraminidase enzymes recognize and cleave terminal sialic acids from cell surfaces. Viral entry into host cells requires neuraminidase activity, thus inhibition of neuraminidase is a useful strategy for development of drugs for viral infections. A recent crystal structure for influenza viral neuraminidase with sialic acid bound shows that the sialic acid is in a boat conformation [Prot Struct Funct Genet 14: 327 (1992)]. Our studies seek to determine if structural pre-organization can be achieved through the use of sialyllactones. Determination of whether siallylactones are pre-organized in a binding conformation requires conformational analysis. Our inability to find a systematic study comparing the results obtained by various computational methods for carbohydrate modeling led us to compare two different conformational analysis techniques, four different force fields, and three different solvent models. The computational models were compared based on their ability to reproduce experimental coupling constants for sialic acid, sialyl-1,4-lactone, and sialyl-1,7-lactone derivatives. This study has shown that the MM3 forcefield using the implicit solvent model for water implemented in Macromodel best reproduces the experimental coupling constants. The low-energy conformations generated by this combination of computational methods are pre-organized toward conformations which fit well into the active site of neuraminidase.     

Immunoreactivity for LN2 and LN3 distinguishes small cell carcinomas from non-small cell carcinomas in the lung

The hydrolyzing activities of O-hexyl O-2,5-dichlorophenyl phosphoramidate (HDCP) and p-nitrophenyl butyrate (p-NPB) in chicken serum had been found to copurify in the same protein, identified as albumin. The hydrolyzing activities of both chicken serum and commercial serum albumins from different species were inhibited in a dose-dependent manner by short chain fatty acids. On simultaneous incubation of chicken serum with HDCP and p-NPB, a competitive interaction was detected between the two substrates. This behavior suggests that both are hydrolyzed in the same albumin active site. When chicken serum was preincubated with one of the substrates, and the latter were withdrawn by large dilution, the hydrolyzing activities with both substrates were found to be reduced. This reduction was in turn dependent upon the time of preincubation with the first substrate. These results suggest that HDCP and p-NPB are hydrolyzed by the same albumin active site, via a mechanism based on transient phosphorylation/acylation of the active site. The proposed hydrolysis mechanism would account for the hydrolytic kinetics of both substrates.     

Enhanced expression of GTP-binding proteins in differentiated U937 monocytic cells: possible involvement of tyrosine kinase and protein kinase C

Seven Arxula adeninivorans strains originating from The Netherlands (CSIR 1136, CSIR 1138 and CBS 8244T), South Africa (CSIR 1147, CSIR 1148 and CSIR 1149) and Siberia (Ls3) were compared concerning their secretory glucoamylase. Within the spectrum of secretory polypeptides obtained by SDS-polyacrylamide gel electrophoresis, the glucoamylase corresponding polypeptide, could be identified by means of specific antibodies directed against the glucoamylase of strain Ls3. The molecular masses of the glucoamylases vary from 84 to 95 kDa. After endoglycosidase F treatment of the secretory proteins, the N-linked carbohydrate content for each glucoamylase could be quantified at about 25%. Glucoamylase activity staining of secretory proteins separated under nondenaturing conditions revealed one glucoamylase active protein for the Dutch strains and two active forms for the strains originating from South Africa and Siberia. Highest activities of glucoamylase can be measured at the end of the exponential growth phase for each strain. The Dutch strain CSIR 1138 secretes the highest activity. Optimum values for temperature and pH were determined and compared. By means of pulsed field gel electrophoresis and Southern analysis, the glucoamylase corresponding gene could be localized on the second of the four chromosomes of each yeast strain.     

Appearance of T cells in the bursa of Fabricius and cecal tonsils during the acute phase of infectious bursal disease virus infection in chickens

Presteady and steady-state kinetic results on the interactions of a wild-type, and the mutant glucoamylases Trp52-->Phe and Trp317-->Phe, from Aspergillus niger with maltose, maltotriose and maltotetraose have been obtained and analyzed. The results are compared with previous ones on the mutants, Trp120-->Phe and Glu180-->Gln, and with results obtained from structure energy minimization calculations based on known three-dimensional structural data. All results are in accordance with a three-step reaction model involving two steps in the substrate binding and a rate-determining catalytic step. Trp317 and Glu180 belong to different subsites, but are placed on the same flank of the active site (beta-flank). The Trp317-->Phe and the Glu180-->Gln mutants show almost identical kinetic results: weakening of the substrate binding, mainly caused by changes in the second reaction step, and practically no change of the catalytic rate. Structure energy minimization calculations show that the same loss of Arg305 and Glu180 hydrogen bonds to the substrate occurs in the Michaelis complexes of each of these mutants. These results indicate that important interactions of the active site may be better understood from a consideration of its flanks rather than of its subsites. The results further indicate differences in the substrate binding mode of maltose and of longer substrates. Trp52 and Trp120 each interact with the catalytic acid, Glu179, and are placed on the flank (alpha-flank) of the active site opposite to Trp317, Arg305 and Glu180. Also the Trp52-->Phe and Trp120-->Phe mutants show kinetic results similar to each other. The catalytic rates are strongly reduced and the substrates are bound more strongly, mainly as a result of the formation of a more stable complex in the second reaction step. All together, the substrate binding mechanism seems to involve an initial enzyme-substrate complex, in which the beta-flank plays a minor role, except for maltose binding; this is followed by a conformational change, in which hydrogen bonds to Arg305 and Glu180 of the beta-flank are established and the correct alignment on the alpha-flank of Glu179, the general acid catalyst, governed by its flexible interactions with Trp52 and Trp120, occurs.     

Solution structure of porcine pancreatic phospholipase A2

The lipolytic enzyme phospholipase A2 (PLA2) is involved in the degradation of high-molecular weight phospholipid aggregates in vivo. The enzyme has very high catalytic activities on aggregated substrates compared with monomeric substrates, a phenomenon called interfacial activation. Crystal structures of PLA2s in the absence and presence of inhibitors are identical, from which it has been concluded that enzymatic conformational changes do not play a role in the mechanism of interfacial activation. The high-resolution NMR structure of porcine pancreatic PLA2 free in solution was determined with heteronuclear multidimensional NMR methodology using doubly labeled 13C, 15N-labeled protein. The solution structure of PLA2 shows important deviations from the crystal structure. In the NMR structure the Ala1 alpha-amino group is disordered and the hydrogen bonding network involving the N-terminus and the active site is incomplete. The disorder observed for the N-terminal region of PLA2 in the solution structure could be related to the low activity of the enzyme towards monomeric substrates. The NMR structure of PLA2 suggests, in contrast to the crystallographic work, that conformational changes do play a role in the interfacial activation of this enzyme.     

Expression of a plant protein by Neurospora crassa

Heterologous expression of plant genes may serve as an important alternative for producing plant proteins. We have investigated the ability of the fungus Neurospora crassa to secrete zeamatin, a protein produced by Zea mays. Zeamatin was induced after being fused to glucoamylase, an extracellular hydrolase produced by N. crassa. Glucoamylase induction and other culture parameters were monitored in untransformed N. crassa grown in shaken liquid culture. A DNA plasmid, pGEZ, was constructed by inserting zeamatin-encoding cDNA into an expression cassette containing the promoter, a truncated open reading frame, and the terminator sequence of the N. crassa glucoamylase gene. Zeamatin-encoding cDNA was modified at the N terminus to include a kex-2 protease site, allowing cleavage of the chimeric product in the secretory pathway. Strains containing the chimeric gene construct were grown in liquid culture and induced for glucoamylase and zeamatin production. Zeamatin antibody detected a protein in a Western blot of concentrated culture supernatants that comigrated with authentic zeamatin. Secreted zeamatin was active in inhibiting the growth of Candida albicans in an agar diffusion assay, indicating that zeamatin had been correctly synthesized, processed, and secreted by N. crassa.     

The P-glycoprotein efflux pump: how does it transport drugs?

Pgp is an atypical translocating ATPase, with low affinity for ATP and high constitutive ATPase activity. Pgp also has an unusually broad specificity for hydrophobic substrates, including many chemotherapeutic drugs. Transport studies in reconstituted systems indicate that drug transport requires ATP hydrolysis and is active, generating a drug concentration gradient. Binding of drugs and ATP to Pgp induces conformational changes in the protein, and the drug binding site is conformationally coupled to the NBDs. Evidence accumulated to date suggests that the transporter interacts directly with nonpolar substrates within the membrane environment, and may act as a drug flippase, moving drugs from the inner to the outer leaflet of the bilayer. Chemosensitizers that block the action of Pgp are proposed to act as alternative substrates, but their high rate of spontaneous flip-flop across the membrane results in futile cycling of the transporter.     

Pancreatic cancer in rats and hamsters does not induce IAPP-related hyperglycaemia

The X-ray crystallographic structure of the branched-chain amino acid aminotransferase from Escherichia coli was determined by means of isomorphous replacement using the selenomethionyl enzyme as one of the heavy atom derivatives. The enzyme is a homo hexamer with D3 symmetry, and the polypeptide chain of the subunit is folded into two domains (small and large domains). The coenzyme, pyridoxal 5'-phosphate, resides at the domain interface, its re-face facing toward the protein. The active site structure shows that the following sites can recognize branched-chain amino acids and glutamate as substrates: (1) a hydrophobic core formed by Phe36, Tyr164, Tyr31*, and Val109* for a branched-chain; (2) Arg97 for an acidic side chain of glutamate; and (3) Tyr95 and two main chain NH groups of Thr257 and Ala258 for the alpha-carboxylate of substrates. Although the main chain conformation of the active site is homologous to that of D-amino acid aminotransferase, many of the active site residues are different between them.     

Conformational changes due to vicinal glycosylation: the branched alpha-L-Rhap(1-2)[beta-D-Galp(1-3)]-beta-D-Glc1-OMe trisaccharide compared with its parent disaccharides

Conformations of the alpha-L-Rhap(1-2)-beta-D-Glc1-OMe and beta-D-Galp(1-3)-beta-D-Glc1-OMe disaccharides and the branched title trisaccharide were examined in DMSO-d6 solution by 1H-nmr. The distance mapping procedure was based on rotating frame nuclear Overhauser effect (NOE) constraints involving C- and O-linked protons, and hydrogen-bond constraints manifested by the splitting of the OH nmr signals for partially deuteriated samples. An "isotopomer-selected NOE" method for the unequivocal identification of mutually hydrogen-bonded hydroxyl groups was suggested. The length of hydrogen bonds thus detected is considered the only one motionally nonaveraged nmr-derived constraint. Molecular mechanics and molecular dynamics methods were used to model the conformational properties of the studied oligosaccharides. Complex conformational search, relying on a regular phi, psi-grid based scanning of the conformational space of the selected glycosidic linkage, combined with simultaneous modeling of different allowed orientations of the pendant groups and the third, neighboring sugar residue, has been carried out. Energy minimizations were performed for each member of the phi, psi grid generated set of conformations. Conformational clustering has been done to group the minimized conformations into families with similar values of glycosidic torsion angles. Several stable syn and anti conformations were found for the 1-->2 and 1-->3 bonds in the studied disaccharides. Vicinal glycosylation affected strongly the occupancy of conformational states in both branches of the title trisaccharide. The preferred conformational family of the trisaccharide (with average phi, psi values of 38 degrees, 17 degrees for the 1-->2 and 48 degrees, 1 degree for the 1-->3 bond, respectively) was shown by nmr to be stabilized by intramolecular hydrogen bonding between the nonbonded Rha and Gal residues.