Structure of the Sulfolobus solfataricus alpha-glucosidase: implications for domain conservation and substrate recognition in GH31

The crystal structure of alpha-glucosidase MalA from Sulfolobus solfataricus has been determined at 2.5Angstrom resolution. It provides a structural model for enzymes representing the major specificity in glycoside hydrolase family 31 (GH31), including alpha-glucosidases from higher organisms, involved in glycogen degradation and glycoprotein processing. The structure of MalA shows clear differences from the only other structure known from GH31, alpha-xylosidase YicI. MalA and YicI share only 23% sequence identity. Although the two enzymes display a similar domain structure and both form hexamers, their structures differ significantly in quaternary organization: MalA is a dimer of trimers, YicI a trimer of dimers. MalA and YicI also differ in their substrate specificities, as shown by kinetic measurements on model chromogenic substrates. In addition, MalA has a clear preference for maltose (Glc-alpha1,4-Glc), whereas YicI prefers isoprimeverose (Xyl-alpha1,6-Glc). The structural origin of this difference occurs in the -1 subsite where MalA residues Asp251 and Trp284 could interact with OH6 of the substrate. The structure of MalA in complex with beta-octyl-glucopyranoside has been determined. It reveals Arg400, Asp87, Trp284, Met321 and Phe327 as invariant residues forming the +1 subsite in the GH31 alpha-glucosidases. Structural comparisons with other GH families suggest that the GH31 enzymes belong to clan GH-D.     

A remote but significant sequence homology between glycoside hydrolase clan GH-H and family GH31

Although both the alpha-amylase super-family, i.e. the glycoside hydrolase (GH) clan GH-H (the GH families 13, 70 and 77), and family GH31 share some characteristics, their different catalytic machinery prevents classification of GH31 in clan GH-H. A significant but remote evolutionary relatedness is, however, proposed for clan GH-H with GH31. A sequence alignment, based on the idea that residues equivalent in the primordial catalytic GH-H/GH31 (beta/alpha)(8)-barrel may not be found in the present-day GH-H and GH31 structures at strictly equivalent positions, shows remote sequence homologies covering beta3, beta4, beta7 and beta8 of the GH-H and GH31 (beta/alpha)(8)-barrels. Structure comparison of GH13 alpha-amylase and GH31 alpha-xylosidase guided alignment of GH-H and GH31 members for construction of evolutionary trees. The closest sequence relationship displayed by GH31 is to GH77 of clan GH-H.     

Overexpression and characterization of two unknown proteins, YicI and YihQ, originated from Escherichia coli

The proteins encoded in the yicI and yihQ gene of Escherichia coli have similarities in the amino acid sequences to glycoside hydrolase family 31 enzymes, but they have not been detected as the active enzymes. The functions of the two proteins have been first clarified in this study. Recombinant YicI and YihQ produced in E. coli were purified and characterized. YicI has the activity of alpha-xylosidase. YicI existing as a hexamer shows optimal pH at 7.0 and is stable in the pH range of 4.7-10.1 with incubation for 24h at 4 degrees C and also is stable up to 47 degrees C with incubation for 15 min. The enzyme shows higher activity against alpha-xylosyl fluoride, isoprimeverose (6-O-alpha-xylopyranosyl-glucopyranose), and alpha-xyloside in xyloglucan oligosaccharides. The alpha-xylosidase catalyzes the transfer of alpha-xylosyl residue from alpha-xyloside to xylose, glucose, mannose, fructose, maltose, isomaltose, nigerose, kojibiose, sucrose, and trehalose. YihQ exhibits the hydrolysis activity against alpha-glucosyl fluoride, and so is an alpha-glucosidase, although the natural substrates, such as alpha-glucobioses, are scarcely hydrolyzed. alpha-Glucosidase has been found for the first time in E. coli.     

A tightly packed hydrophobic cluster directs the formation of an off-pathway sub-millisecond folding intermediate in the alpha subunit of tryptophan synthase, a TIM barrel protein

Protein misfolding is now recognized as playing a crucial role in both normal and pathogenic folding reactions. An interesting example of misfolding at the earliest state of a natural folding reaction is provided by the alpha-subunit of tryptophan synthase, a (beta/alpha)(8) TIM barrel protein. The molecular basis for the formation of this off-pathway misfolded intermediate, I(BP), and a subsequent on-pathway intermediate, I1, was probed by mutational analysis of 20 branched aliphatic side-chains distributed throughout the sequence. The elimination of I(BP) and the substantial destabilization of I1 by replacement of a selective set of the isoleucine, leucine or valine residues (ILV) with alanine in a large ILV cluster external-to-the-barrel and spanning the N and C termini (cluster 2) implies tight-packing at most sites in both intermediates. Differential effects on I(BP) and I1 for replacements in alpha3, beta4 and alpha8 at the boundaries of cluster 2 suggest that their incorporation into I1 but not I(BP) reflects non-native folds at the edges of the crucial (beta/alpha)(1-2)beta(3) core in I(BP). The retention of I(BP) and the smaller and consistent destabilization of both I(BP) and I1 by similar replacements in an internal-to-the-barrel ILV cluster (cluster 1) and a second external-to-the-barrel ILV cluster (cluster 3) imply molten globule-like packing. The tight packing inferred, in part, for I(BP) or for all of I1 in cluster 2, but not in clusters 1 and 3, may reflect the larger size of cluster 2 and/or the enhanced number of isoleucine, leucine and valine self-contacts in and between contiguous elements of secondary structure. Tightly packed ILV-dominated hydrophobic clusters could serve as an important driving force for the earliest events in the folding and misfolding of the TIM barrel and other members of the (beta/alpha)(n) class of proteins.     

Altering the substrate chain-length specificity of an alpha-glucosidase

Dextran glucosidases show high sequence identity (50%) to Bacillus sp. SAM1606 alpha-glucosidase, which is more specific for short-chain substrates. Sequence comparison of these enzymes as well as molecular modeling studies predicted that the extension of loop 4 of the (beta/alpha)(8)-barrel fold may be responsible for the narrower specificity of SAM1606 alpha-glucosidase with respect to substrate chain length. Indeed, deletion mutants of SAM1606 alpha-glucosidase that lack this extension showed higher relative activities toward dextran and long-chain isomaltooligosaccharides. Kinetic and thermodynamic analyses of oligosaccharide hydrolysis catalyzed by SAM1606 alpha-glucosidase and its deletion mutants suggested that the loss of such extension(s) in loop 4 should energetically destabilize the Michaelis complexes with long-chain substrates to result in smaller differences between the activation free energies for the enzymatic hydrolyses of isomaltoheptaose and isomaltose than those observed for the wild-type enzyme. This is the reason that dextran glucosidase, whose loop 4 is shorter in length, shows broader substrate chain-length specificity than does SAM1606 alpha-glucosidase.     

Crystal structure of intracellular family 1 beta-glucosidase BGL1A from the basidiomycete Phanerochaete chrysosporium

The white-rot fungus Phanerochaete chrysosporium has two intracellular beta-glucosidases (BGL1A and BGL1B) belonging to glycoside hydrolase (GH) family 1. BGL1B effectively hydrolyzes cellobiose and cellobionolactone, but BGL1A does not. We have determined the crystal structure of BGL1A in substrate-free and gluconolactone complexed forms. The overall structure and the characteristic of subsite -1 (glycone site) were similar to those of other known GH1 enzymes. The loop regions covering on the (beta/alpha)(8) barrel was significantly deviated, and they form a unique subsite +1 (aglycone site) of BGL1A.     

Mechanistic and structural analysis of a family 31 alpha-glycosidase and its glycosyl-enzyme intermediate

We have determined the first structure of a family 31 alpha-glycosidase, that of YicI from Escherichia coli, both free and trapped as a 5-fluoroxylopyranosyl-enzyme intermediate via reaction with 5-fluoro-alpha-D-xylopyranosyl fluoride. Our 2.2-A resolution structure shows an intimately associated hexamer with structural elements from several monomers converging at each of the six active sites. Our kinetic and mass spectrometry analyses verified several of the features observed in our structural data, including a covalent linkage from the carboxylate side chain of the identified nucleophile Asp(416) to C-1 of the sugar ring. Structure-based sequence comparison of YicI with the mammalian alpha-glucosidases lysosomal alpha-glucosidase and sucrase-isomaltase predicts a high level of structural similarity and provides a foundation for understanding the various mutations of these enzymes that elicit human disease.     

Role of a single amino acid in the evolution of glycans of invertebrates and vertebrates

Structures of glycoconjugate N-glycans and glycolipids of invertebrates show significant differences from those of vertebrates. These differences are due largely to the vertebrate beta1,4-galactosyltransferase-1 (beta4Gal-T1), which is found as a beta1,4-N-acetylgalactosaminyltransferase (beta4GalNAc-T1) in invertebrates. Mutation of Tyr285 to Ile or Leu in human beta4Gal-T1 converts the enzyme into an equally efficient beta4GalNAc-T1. A comparison of all the human beta4Gal-T1 ortholog enzymes shows that this Tyr285 residue in human beta4Gal-T1 is conserved either as Tyr or Phe in all vertebrate enzymes, while in all invertebrate enzymes it is conserved as an Ile or Leu. We find that mutation of the corresponding Ile residue to Tyr in Drosophila beta4GalNAc-T1 converts the enzyme to a beta4Gal-T1 by reducing its N-acetylgalactosaminyltransferase activity by nearly 1000-fold, while enhancing its galactosyltransferase activity by 80-fold. Furthermore, we find that, similar to the vertebrate/mammalian beta4Gal-T1 enzymes, the wild-type Drosophila beta4GalNAc-T1 enzyme binds to a mammary gland-specific protein, alpha-lactalbumin (alpha-LA). Thus, it would seem that, during the evolution of vertebrates from invertebrates over 500 million years ago, beta4Gal-T1 appeared as a result of the single amino acid substitution of Tyr or Phe for Leu or Ile in the invertebrate beta4GalNAc-T1. Subsequently, the pre-existing alpha-LA-binding site was utilized during mammalian evolution to synthesize lactose in the mammary gland during lactation.     

The crystal structure of human cytosolic beta-glucosidase unravels the substrate aglycone specificity of a family 1 glycoside hydrolase

Human cytosolic beta-glucosidase (hCBG) is a xenobiotic-metabolizing enzyme that hydrolyses certain flavonoid glucosides, with specificity depending on the aglycone moiety, the type of sugar and the linkage between them. In this study, the substrate preference of this enzyme was investigated by mutational analysis, X-ray crystallography and homology modelling. The crystal structure of hCBG was solved by the molecular replacement method and refined at 2.7 A resolution. The main-chain fold of the enzyme belongs to the (beta/alpha)(8) barrel structure, which is common to family 1 glycoside hydrolases. The active site is located at the bottom of a pocket (about 16 A deep) formed by large surface loops, surrounding the C termini of the barrel of beta-strands. As for all the clan of GH-A enzymes, the two catalytic glutamate residues are located on strand 4 (the acid/base Glu165) and on strand 7 (the nucleophile Glu373). Although many features of hCBG were shown to be very similar to previously described enzymes from this family, crucial differences were observed in the surface loops surrounding the aglycone binding site, and these are likely to strongly influence the substrate specificity. The positioning of a substrate molecule (quercetin-4'-glucoside) by homology modelling revealed that hydrophobic interactions dominate the binding of the aglycone moiety. In particular, Val168, Trp345, Phe225, Phe179, Phe334 and Phe433 were identified as likely to be important in determining substrate specificity in hCBG, and site-directed mutagenesis supported a key role for some of these residues.     

Key aromatic residues at subsites + 2 and + 3 of glycoside hydrolase family 31 α-glucosidase contribute to recognition of long-chain substrates

Glycoside hydrolase family 31 α-glucosidases (31AGs) show various specificities for maltooligosaccharides according to chain length. Aspergillus niger α-glucosidase (ANG) is specific for short-chain substrates with the highest k_cat/K_m for maltotriose, while sugar beet α-glucosidase (SBG) prefers long-chain substrates and soluble starch. Multiple sequence alignment of 31AGs indicated a high degree of diversity at the long loop (N-loop), which forms one wall of the active pocket. Mutations of Phe236 in the N-loop of SBG (F236A/S) decreased k_cat/K_m values for substrates longer than maltose. Providing a phenylalanine residue at a similar position in ANG (T228F) altered the k_cat/K_m values for maltooligosaccharides compared with wild-type ANG, i.e., the mutant enzyme showed the highest k_cat/K_m value for maltotetraose. Subsite affinity analysis indicated that modification of subsite affinities at + 2 and + 3 caused alterations of substrate specificity in the mutant enzymes. These results indicated that the aromatic residue in the N-loop contributes to determining the chain-length specificity of 31AGs.     

The structure of 3-deoxy-d-arabino-heptulosonate 7-phosphate synthase from Mycobacterium tuberculosis reveals a common catalytic scaffold and ancestry for type I and type II enzymes

The shikimate pathway, responsible for the biosynthesis of aromatic compounds, is essential for the growth of Mycobacterium tuberculosis and is a potential target for the design of new anti-tuberculosis drugs. The first step of this pathway is catalyzed by 3-deoxy-d-arabino-heptulosonate 7-phosphate synthase (DAH7PS). The DAH7PSs have been classified into two apparently unrelated types and, whereas structural data have been obtained for the type I DAH7PSs, no structural information is available for their type II counterparts. The type II DAH7PS from M.tuberculosis has been expressed in Escherichia coli, purified, functionally characterized and crystallized. It is found to be metal ion-dependent and subject to feedback inhibition by phenylalanine, tryptophan, tyrosine and chorismate, with a significant synergistic effect when tryptophan is used in combination with phenylalanine. The crystal structure of M.tuberculosis DAH7PS has been determined by single-wavelength anomalous diffraction and refined at 2.3A in complex with substrate phosphoenolpyruvate and Mn(2+). The structure reveals a tightly associated dimer of (beta/alpha)(8) TIM barrels. The monomer fold, the arrangement of key residues in the active site, and the binding modes of PEP and Mn(2+), all match those of the type I enzymes, and indicate a common ancestry for the type I and type II DAH7PSs, despite their minimal sequence identity. In contrast, the structural elements that decorate the core (beta/alpha)(8) fold differ from those in the type I enzymes, consistent with their different regulatory and oligomeric properties.     

Molecular cloning and heterologous expression of beta1,2-xylosyltransferase and core alpha1,3-fucosyltransferase from maize

Maize is considered a promising alternative production system for pharmaceutically relevant proteins. However, like in all other plant species asparagine-linked oligosaccharides of maize glycoproteins are modified with beta1,2-xylose and core alpha1,3-fucose sugar residues, which are considered to be immunogenic in mammals. This altered N-glycosylation when compared to mammalian cells may reduce the potential of maize as a production system for heterologous glycoproteins. Here we report the cloning and characterization of the cDNA sequences coding for the maize enzymes beta1,2-xylosyltransferase (XylT) and core alpha1,3-fucosyltransferase (FucT). The cloned XylT and FucT cDNAs were shown to encode enzymatically active proteins, which were independently able to convert a mammalian acceptor glycoprotein into an antigen binding anti-plant N-glycan antibodies. The complete sequence of the XylT gene was determined. Evidence for the presence of at least three XylT and FucT gene loci in the maize genome was obtained. The identification of the two enzymes and their genes will allow the targeted downregulation or even elimination of beta1,2-xylose and core alpha1,3-fucose addition to recombinant glycoproteins produced in maize.     

Evolution of function in (beta/alpha)8-barrel enzymes

The (beta/alpha)(8)-barrel is the most common fold in structurally characterized enzymes. Whether the functionally diverse enzymes that share this fold are the products of either divergent or convergent evolution (or both) is an unresolved question that will probably be answered as the sequence databases continue to expand. Recent work has examined natural, designed, and directed evolution of function in several superfamilies of (beta/alpha)(8)-barrel containing enzymes.     

A common evolutionary origin of two elementary enzyme folds

The (beta alpha)(8)-barrel is the most frequent and most versatile fold among enzymes [H?cker et al., Curr. Opin. Biotechnol. 12 (2001) 376-381; Wierenga, FEBS Lett. 492 (2001) 193-198]. Structural and functional evidence suggests that (beta alpha)(8)-barrels evolved from an ancestral half-barrel, which consisted of four (beta alpha) units stabilized by dimerization [Lang et al., Science 289 (2000) 1546-550; H?cker et al., Nat. Struct. Biol. 8 (2001) 32-36; Gerlt and Babbitt, Nat. Struct. Biol. 8 (2001) 5-7]. Here, by performing a comprehensive database search, we detect a striking and unexpected structural and amino acid sequence similarity between (beta alpha)(4) half-barrels and members of the (beta alpha)(5) flavodoxin-like fold. These findings provoke the hypothesis that a large fraction of the modern-day enzymes evolved from a basic structural building block, which can be identified by a combination of sequence and structural analyses.     

Crystal structure of a secretory signalling glycoprotein from sheep at 2.0A resolution

A 40kDa glycoprotein from dry secretion of sheep is implicated as a signaling factor and is named as SPS-40. This protein is secreted only during the early phase of involution when the drastic tissue remodeling occurs in the mammary gland. SPS-40 was purified from sheep dry secretions and crystallized using hanging drop vapour diffusion method. The crystals belong to orthorhombic space group P2(1)2(1)2(1) with cell dimensions, a=62.7A, b=66.4A, c=107.5A. The protein was also cloned for the determination of its complete amino acid sequence. The three-dimensional structure of SPS-40 was determined by X-ray crystallographic method at 2.0A resolution. The structure revealed the presence of an N-linked glycan chain at Asn39. The protein adopts a conformation with a classical (beta/alpha)(8)-barrel fold of triosephosphate isomerase (TIM) (residues 1-237 and 310-360) with an insertion of a small (alpha+beta) domain (residues 240-307) similar to that observed in chitinases. However, the Leu substitution for Glu in the consensus catalytic sequence in SPS-40 causes a loss of chitinase activity. Furthermore, the sugar-binding groove in SPS-40 is distorted considerably from the standard chitin-binding site in chitinase enzymes and hence the binding of chitin-like oligosaccharides is considerably hampered. Three surface loops, His188-His197, Phe202-Arg212 and Phe244-Pro260 have exceptionally high values of B-factors (average=70.5A(2)), indicating the presence of a less defined region.     

The collagen-binding A-domains of integrins alpha(1)beta(1) and alpha(2)beta(1) recognize the same specific amino acid sequence, GFOGER, in native (triple-helical) collagens

We have previously assigned an integrin alpha(2)beta(1)-recognition site in collagen I to the sequence, GFOGERGVEGPOGPA (O = Hyp), corresponding to residues 502-516 of the alpha(1)(I) chain and located in the fragment alpha(1)(I)CB3 (Knight, C. G., Morton, L. F., Onley, D. J., Peachey, A. R., Messent, A. J., Smethurst, P. A., Tuckwell, D. S., Farndale, R. W., and Barnes, M. J. (1998) J. Biol. Chem. 273, 33287-33294). In this study, we show that recognition is entirely contained within the six-residue sequence GFOGER. This sequence, when in triple-helical conformation, readily supports alpha(2)beta(1)-dependent cell adhesion and exhibits divalent cation-dependent binding of isolated alpha(2)beta(1) and recombinant alpha(2) A-domain, being at least as active as the parent collagen. Replacement of E by D causes loss of recognition. The same sequence binds integrin alpha(1) A-domain and supports integrin alpha(1)beta(1)-mediated cell adhesion. Triple-helical GFOGER completely inhibits alpha(2) A-domain binding to collagens I and IV and alpha(2)beta(1)-dependent adhesion of platelets and HT 1080 cells to these collagens. It also fully inhibits alpha(1) A-domain binding to collagen I and strongly inhibits alpha(1)beta(1)-mediated adhesion of Rugli cells to this collagen but has little effect on either alpha1 A-domain binding or adhesion of Rugli cells to collagen IV. We conclude that the sequence GFOGER represents a high-affinity binding site in collagens I and IV for alpha(2)beta(1) and in collagen I for alpha(1)beta(1). Other high-affinity sites in collagen IV mediate its recognition of alpha(1)beta(1).     

CD4 and CD8 may adopt a similar mode of binding to their MHC ligands.

A theoretical scheme is proposed by which the type-specific cell surface receptors of T-lymphocytes, CD8 and CD4, bind class I and II MHC proteins in a similar manner. The scheme has equivalent residues in the C'/C' loop-C' strand-C'/D loop region in domain 1 of CD4 and CD8 alpha binding to equivalent residues in the C and D beta-strands and C/D loops in HLA-DR beta 2 (class II) and HLA-A2 alpha 3 (class I) respectively through a series of electrostatic, hydrogen and hydrophobic bonds.     

CT domain of CCN2/CTGF directly interacts with fibronectin and enhances cell adhesion of chondrocytes through integrin alpha5beta1

Searching for CCN family protein 2/connective tissue growth factor (CCN2/CTGF) interactive proteins by yeast-two-hybrid screening, we identified fibronectin 1 gene product as a major binding partner of CCN2/CTGF in the chondrosarcoma-derived chondrocytic cell line HCS-2/8. Only the CT domain of CCN2/CTGF bound directly to fibronectin (FN). CCN2/CTGF and its CT domain enhanced the adhesion of HCS-2/8 cells to FN in a dose-dependent manner. The CCN2/CTGF-enhancing effect on cell adhesion to FN was abolished by a blocking antibody against alpha5beta1 integrin (alpha5beta1), but not by one against anti-alphavbeta3 integrin. These findings suggest for the first time that CCN2/CTGF enhances chondrocyte adhesion to FN through direct interaction of its C-terminal CT domain with FN, and that alpha5beta1 is involved in this adhesion.     

PTHrP contributes to the anti-proliferative and integrin alpha6beta4-regulating effects of 1,25-dihydroxyvitamin D(3)

Parathyroid hormone-related protein (PTHrP) increases the growth and metastatic potential of prostate cancer cells, making it important to control PTHrP expression in these cells. 1,25-Dihydroxyvitamin D(3) [1,25(OH)(2)D(3)] suppresses PTHrP expression and exerts an anti-proliferative effect in prostate carcinoma cells. We used the human prostate cancer cell line C4-2 as a model system to ask whether down-regulation of PTHrP expression by 1,25(OH)(2)D(3) plays a role in the anti-proliferative effects of 1,25(OH)(2)D(3). Since PTHrP increases the expression of the pro-invasive integrin alpha6beta4, we also asked whether 1,25(OH)(2)D(3) decreases integrin alpha6beta4 expression in C4-2 cells, and whether modulation of PTHrP expression by 1,25(OH)(2)D(3) plays a role in the effects of 1,25(OH)(2)D(3) on integrin alpha6beta4 expression. Two strategies were utilized to modulate PTHrP levels: overexpression of PTHrP (-36 to +139) and suppression of endogenous PTHrP expression using siRNAs. We report a direct correlation between PTHrP expression, C4-2 cell proliferation and integrin alpha6beta4 expression at the mRNA and cell surface protein level. Treatment of parental C4-2 cells with 1,25(OH)(2)D(3) decreased cell proliferation and integrin alpha6 and beta4 expression. These 1,25(OH)(2)D(3) effects were significantly attenuated in cells with suppressed PTHrP expression. 1,25(OH)(2)D(3) regulates PTHrP expression via a negative vitamin D response element (nVDRE) within the noncoding region of the PTHrP gene. The effects of 1,25(OH)(2)D(3) on cell proliferation and integrin alpha6beta4 expression were significantly attenuated in cells overexpressing PTHrP (-36 to +139), which lacks the nVDRE. These findings suggest that one of the pathways via which 1,25(OH)(2)D(3) exerts its anti-proliferative effects is through down-regulation of PTHrP expression.