Protein Science
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pinotsis, N.
Right arrow Articles by Mavridis, I. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pinotsis, N.
Right arrow Articles by Mavridis, I. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?
Protein Science (2003), 12:1914-1924.
Copyright © 2003 The Protein Society

The binding of ß- and {gamma}-cyclodextrins to glycogen phosphorylase b: Kinetic and crystallographic studies

Nikos Pinotsis1, Demetres D. Leonidas2, Evangelia D. Chrysina2, Nikos G. Oikonomakos2,3 and Irene M. Mavridis1

1 Institute of Physical Chemistry, National Center for Scientific Research "Demokritos," Athens, Greece
2 Institute of Organic and Pharmaceutical Chemistry, and
3 Institute of Biological Research and Biotechnology, The National Hellenic Research Foundation, Athens 11635, Greece

Reprints requests to: Nikos G. Oikonomakos, Institute of Organic and Pharmaceutical Chemistry, The National Hellenic Research Foundation, 48 Vas. Constantinou Avenue, Athens 11635, Greece; e-mail: ngo{at}eie.gr; fax: 30-210-7273758(831).

(RECEIVED April 21, 2003; FINAL REVISION May 29, 2003; ACCEPTED May 30, 2003)

Article and publication are at http://www.proteinscience.org/cgi/doi/10.1110/ps.03149503.


   Abstract
TOP
 Abstract
Introduction
 Results and Discussion
 Materials and methods
 References
 
A number of regulatory binding sites of glycogen phosphorylase (GP), such as the catalytic, the inhibitor, and the new allosteric sites are currently under investigation as targets for inhibition of hepatic glycogenolysis under high glucose concentrations; in some cases specific inhibitors are under evaluation in human clinical trials for therapeutic intervention in type 2 diabetes. In an attempt to investigate whether the storage site can be exploited as target for modulating hepatic glucose production, {alpha}-, ß-, and {gamma}-cyclodextrins were identified as moderate mixed-type competitive inhibitors of GPb (with respect to glycogen) with Ki values of 47.1, 14.1, and 7.4 mM, respectively. To elucidate the structural basis of inhibition, we determined the structure of GPb complexed with ß- and {gamma}-cyclodextrins at 1.94 Å and 2.3 Å resolution, respectively. The structures of the two complexes reveal that the inhibitors can be accommodated in the glycogen storage site of T-state GPb with very little change of the tertiary structure and provide a basis for understanding their potency and subsite specificity. Structural comparisons of the two complexes with GPb in complex with either maltopentaose (G5) or maltoheptaose (G7) show that ß- and{gamma}-cyclodextrins bind in a mode analogous to the G5 and G7 binding with only some differences imposed by their cyclic conformations. It appears that the binding energy for stabilization of enzyme complexes derives from hydrogen bonding and van der Waals contacts to protein residues. The binding of {alpha}-cyclodextrin and octakis (2,3,6-tri-O-methyl)-{gamma}-cyclodextrin was also investigated, but none of them was bound in the crystal; moreover, the latter did not inhibit the phosphorylase reaction.

Keywords: Glycogen phosphorylase; {alpha}-cyclodextrin; ß-cyclodextrin; {gamma}-cyclodextrin; oligosaccharide binding; protein-carbohydrate interactions; X-ray crystallography

Abbreviations: GP, glycogen phosphorylase • GPb, muscle glycogen phosphorylase b • GPa, muscle glycogen phosphorylase a • glucose, {alpha}-d-glucose • Glc-1-P, {alpha}-d-glucose 1-phosphate • {alpha}-,ß-,{gamma}-CD, {alpha}-,ß-,{gamma}-cyclodextrin • TM{gamma}CD, octakis (2,3,6-tri-O-methyl)-{gamma}-cyclodextrin • G5, maltopentaose • G7, maltoheptaose • RMSD, root-mean-square deviation


   Introduction
TOP
 Abstract
 Introduction
Results and Discussion
 Materials and methods
 References
 
Glycogen phosphorylase (EC 2.4.1.1 [EC] ; GP) has been a molecular target for structure-assisted design of inhibitors that might prevent unwanted glycogenolysis under high glucose conditions. Several binding sites in GP, such as the catalytic, the allosteric, the inhibitor, and the new allosteric sites have been explored as specific targets in kinetic, crystallographic, and physiological studies (reviewed in McCormack et al. 2001; Somsak et al. 2001; Treadway et al. 2001; Oikonomakos 2002; Kurukulasuriya et al. 2003). However, the glycogen storage site has not been experimentally validated as a target for modulating hepatic glucose production in type 2 diabetes, and no medicinal chemistry studies have been reported on this target yet. This site was identified from crystallographic binding studies with maltopentaose (G5) and maltoheptaose (G7), that also bind to the catalytic site and act as substrates (Johnson et al. 1988). In GPb, G5 can bind to the glycogen storage site by filling five major and two minor subsites. The precise role of the glycogen storage site is not fully understood, but it might serve as a region through which the mammalian enzyme is attached to glycogen particles in vivo, to produce an effective high concentration of end groups and act as additional regulatory site whereby occupation by glycogen is a prerequisite for catalysis. Also, model building studies (McLaughlin et al. 1984) have shown that the minor site might be consistent with the position of oligosaccharide at an {alpha}-1,6 branch position, providing an explanation for the increased affinity of GP for branched polysaccharides like glycogen over linear glucans like maltodextrins (Hu and Gold 1975). Inhibitors, specific for this site, would be therefore of most interest.

Acarbose, a pseudo-maltotetraose with the cyclitol unit of {alpha}-1,4-linked to a 4-amino-4,6-dideoxyglucose residue, is a potent inhibitor of many {alpha}-glucosidases, {alpha}-amylases, and sucrase-isomaltase, and is considered to be an analog of a glucosyl cation like transition state (Truscheit et al. 1981). Binding studies in the crystal of GPa showed that it bound at the major site of glycogen storage site in an analogous way to maltopentaose inhibitor. Kinetic studies have shown that acarbose is a reasonable inhibitor of GPa (Ki = 26 mM) in competition experiments with 30 mM G5 (Goldsmith et al. 1987). Apart from acarbose, specific inhibitors of the enzyme that bind at the glycogen storage site have not been previously described.

Given the structural analogy of cyclic oligosaccharides to linear ones, we investigated the effect of {alpha}-, ß-, and {gamma}-cyclodextrins (CDs) on the catalytic and structural properties of GPb. The cyclodextrins are torus-like macrorings built up from {alpha}-D-glucopyranose units connected with an {alpha}-1,4 glycosidic linkage (Fig. 1Go). The {alpha}-CD molecule comprises 6 glucopyranose units, while ß- and {gamma}-CD comprise 7 and 8 units, respectively. As a consequence of the 4C1 conformation of the glucopyranose units, primary and secondary hydroxyl groups are situated on either side of two edges of the ring. The ring is a truncated cone whose cavity is lined by the hydrogen atoms and the glycosidic oxygen bridges, respectively. The secondary O2 hydroxyl group of one glucopyranose unit forms a hydrogen bond with the O3 hydroxyl group of the preceding glucopyranose. As a result, a complete belt of hydrogen bonds is formed in the secondary side that is stronger in ß-CD (Makedonopoulou and Mavridis 2000), making it the most rigid and the less water-soluble of all three CDs. In contrast, this hydrogen bond belt is not complete in {alpha}-CD (one of the glucose moieties is distorted) and in {gamma}-CD (more distorted conformation), resulting in a higher flexibility and aqueous solubility compared to that of ß-CD (Szejtli 1998).



View larger version (16K):
[in this window]
[in a new window]
  		Figure 1. Chemical structure of {gamma}-CD and numbering scheme used in a glucose unit (IUPAC-IUB 1983).

 
In this work we present a detailed description of the binding of {alpha}-, ß-, {gamma}-CDs and TM{gamma}CD to GPb through kinetic and X-ray binding studies. The kinetic results show that {alpha}-, ß-, and {gamma}-CDs are moderate inhibitors of the enzyme, with {gamma}-CD being the best (Ki = 7.4 mM), while TM{gamma}CD did not inhibit GPb. To elucidate the key interactions responsible for inhibition we performed crystallographic binding studies with the above compounds and analyzed the structures of those that showed binding, namely GPb-ß-CD and GPb-{gamma}-CD complexes at 1.94 Å and 2.3 Å resolution, respectively. The detailed interactions of ß- and {gamma}-CDs with the protein provide a structural explanation for the kinetic properties and show that the cyclic oligosaccharides bind at the glycogen storage site of GPb in a mode analogous to the G5 and G7 binding, with some changes imposed by differences in their conformations. To improve the precision of our understanding of the molecular basis of linear oligosaccharide recognition, we have also determined the crystal structures of the GPb-G5 and GPb-G7 complexes at 2.2 Å resolution.


   Results and Discussion
TOP
 Abstract
 Introduction
 Results and Discussion
Materials and methods
 References
 
The kinetic parameters (Kis) of {alpha}-, ß-, and {gamma}-CDs are summarized in Table 1Go. The kinetic experiments with GPb, in the direction of glycogen synthesis, showed that all three cyclodextrins exhibited mixed-type inhibition with respect to glycogen substrate at constant concentrations of AMP (1 mM) and Glc-1-P (4 mM). Lineweaver-Burk plots (1/v versus 1/[glycogen]) yielded straight lines of which the slopes and the vertical axis intercept increased with increasing concentrations of inhibitor. {gamma}-CD (Ki = 7.4 mM) was found to be a better inhibitor than ß-CD (Ki = 14.1 mM) and {alpha}-CD (Ki = 47.1 mM). TM{gamma}CD had no effect on the enzymic activity of GPb when added in concentrations varied from 10–50 mM, in the presence of 1 mM AMP, 4 mM Glc-1-P, and 0.035% (w/v) glycogen.


View this table:
[in this window]
[in a new window]
  		Table 1. Kinetic parameters for maltoheptaose and cyclodextrins
 
Crystallographic data collection, processing, and refinement statistics for the structure determinations of GPb complexed with G5, G7, ß-CD, and {gamma}-CD, respectively, are summarized in Table 2Go. The overall architecture of the native T-state GPb showing the location of the catalytic, the inhibitor, the allosteric, the new allosteric, and the glycogen storage site is presented in Figure 2Go. For both the G5 and G7 complexes, electron density maps indicated that five glucose residues (S3 to S7) bound at the major site of the glycogen storage site, and there was no binding at the minor site. For the GPb–{gamma}-CD complex, all glucose residues (S3 to S10) were visible in the electron density map, while for the GPb–ß-CD complex the electron density maps suggested partial occupancy of ß-CD of the glycogen storage site. No binding was observed at the catalytic site. The electron density maps for GPb crystals soaked with {alpha}-CD and TM{gamma}CD indicated no binding at the glycogen storage site. We describe briefly the G5 and G7 interactions at the glycogen storage site and in more detail the {gamma}-CD and ß-CD interactions at this site.


View this table:
[in this window]
[in a new window]
  		Table 2. Diffraction data and refinement statistics
 


View larger version (98K):
[in this window]
[in a new window]
  		Figure 2. A schematic diagram of the muscle GPb dimeric molecule viewed down the twofold for residues 13 to 838. One subunit is colored gray and the other subunit light gray. The positions are shown for the catalytic, the allosteric, the inhibitor, the new allosteric inhibitor site, and the glycogen storage site. The catalytic site, which includes the essential cofactor pyridoxal 5'-phosphate (not shown), is buried at the center of the subunit accessible to the bulk solvent through a 15 Å long channel. Glucose (shown in ball-and-stick representation), a competitive inhibitor of the enzyme that promotes the less active T state through stabilization of the closed position of the 280s loop, binds at this site. The allosteric site, which binds the activator AMP, is situated at the subunit–subunit interface some 30 Å from the catalytic site. The inhibitor site, which binds purine compounds, nucleosides, or nucleotides at high concentrations, and flavopiridol (shown bound; Oikonomakos et al. 2000) is located on the surface of the enzyme some 12 Å from the catalytic site and, in the T state, obstructs the entrance to the catalytic site tunnel. The new allosteric inhibitor site, located inside the central cavity, formed on association of the two subunits, binds N-Benzoyl-N'-ß-D-glucopyranosyl urea molecule (shown; Oikonomakos et al. 2002). The glycogen storage site (with {gamma}-CD bound) is on the surface of the molecule some 30 Å from the catalytic site, 40 Å from the allosteric site, and 50 Å from the new allosteric site.

 
The binding of maltopentaose and maltoheptaose to glycogen phosphorylase b
A major and a minor glycogen storage site have been defined from the crystal structures of the GPb–G5 complex at 2.5 Å resolution (Johnson et al. 1988). The major site is defined by the binding of five glucose units in subsites S3-S4-S5-S6-S7, with the reducing end of the oligosacharide bound in subsite S3. This site is made up of helices {alpha}12 and {alpha}13 (residues 396 to 418 and 420 to 429, respectively) and the loop connecting the two antiparallel strands, ß15 (residues 430 to 432) and ß16 (residues 437 to 411; Fig. 2Go). The minor site, defined by the binding of glucosyl residues S8 and S9, lies above the nonreducing end of the major site. It involves GP residues from the top of helix {alpha}12, the loop connecting the antiparallel ß-sheets ß8 (residues 198 to 209) and ß9 (residues 212 to 223) and Val354 from helix {alpha}9 (residues 344 to 355). In the complex structures of GPa–G7–caffeine–glucose and GPa–G5–phosphate, seven glucosyl moieties (S1 to S7) were found at the major site and four (S8 to S11) at the minor site (Goldsmith et al. 1982Goldsmith et al. 1989; Goldsmith and Fletterick 1983). In contrast, in the complex GPa–G5–caffeine–glucose, only three glucosyl residues (S4 to S6) and one glucose (S10) were found in the minor site (Goldsmith et al. 1989). In the GPb–heptulose–2-P–G7–AMP complex (Johnson et al. 1990), the minor oligosaccharide site is not occupied.

One molecule of either G5 or G7 bound at the GPb storage site occupying the major site, while no binding was observed at the minor site. The mode of binding and the interactions that G5 and G7 exhibit with GPb are almost identical with those for the GPa–G5–phosphate, GPa–G5– caffeine–glucose, and GPa–G7–caffeine–glucose complexes (Goldsmith et al. 1982 Goldsmith et al. 1989; Goldsmith and Fletterick 1983) derived from medium-resolution X-ray crystallographic analyses, and that for the GPb–heptulose–2-P–G7–AMP complex (Johnson et al. 1990) determined at a resolution of 2.86 Å resolution. The resolution of the present structures allows us to describe the binding site for G5 and G7 in some detail. Both structures indicated five well-ordered structural waters that form a network of hydrogen bonds that link oligosaccharide to protein residues; these waters were not observed in the previous medium-resolution structures.

The bound G5 has a regular helical structure stabilized by the O2–O3 hydrogen bonds formed between successive sugars with a break between glucose residues S4 and S5 (Fig. 3AGo). The temperature factors are lower (50–62 Å2) at glucose residues S4, S5, and S6, and increase (65–80 Å2) toward the two ends of the oligosaccharide chain. The O3 hydroxyl group of glucose residue S4 forms a hydrogen bond with Ser429 from helix {alpha}13, and O2 hydroxyl is involved in a water-mediated interaction with Val431 O, Arg426 O, and Gln433 N through another water molecule. The O2 hydroxyl group of glucose residue S5 is hydrogen bonded to N{zeta} of Lys437 from ß-sheet ß16 and a water molecule; O5 and O6 hydroxyl groups are hydrogen bonded to N{delta}2 of Asn407 from helix {alpha}12; moreover, O6 hydroxyl groups is hydrogen bonded to Gln401 O through a water molecule (Fig. 3BGo). Finally, O5 and O6 hydroxyl groups of S3 make a water-mediated hydrogen bond with Phe197 N from a symmetry-related molecule (1/2 + x, 1/2 - y, 1/4 - z).








View larger version (395K):
[in this window]
[in a new window]
  		Figure 3. (A, C, and E) Diagrams of the 2.2 Å and 2.3 Å sigmaA 2|Fo| - |Fc| electron density map in the vicinity of the glycogen storage site for G5, G7 and {gamma}-CD bound to GPb, respectively. Electron density maps were calculated using the standard protocol as implemented in CNS v1.1 (Brünger et al. 1998) from the GPb model before incorporating the coordinates of G5, G7, and {gamma}-CD, respectively. The maps are contoured at the 1.0{sigma} level. (B, D, and F) Diagrams showing the interactions of G5, G7, and {gamma}-CD with GPb, respectively. GPb residues are drawn as ball-and-stick models, water molecules as cyan spheres, and the oligosaccharide molecules are shown in standard colors. Dashed lines indicate hydrogen bonds.

 
On forming the complex with GPb, G5 becomes buried. The solvent accessibilities of the free and bound G5 molecules are 937 Å2 and 532 Å2, indicating that a surface area of 405 Å2 becomes inaccessible to water or that G5 becomes 57% buried in the enzyme complex. Both polar and nonpolar groups are buried; nonpolar groups contribute 175 Å2 (43%), and polar groups contribute 230 Å2 (57%) of the surface that becomes inaccessible. On the protein surface, a total of 311 Å2 solvent-accessible surface area becomes inaccessible on binding G5, of which 180 Å2 (58%) is contributed by nonpolar groups. The total buried surface area (protein + ligand) is 716 Å2.

GP displays a high affinity for G7 at the glycogen storage site, with a dissociation constant of 1 mM (Kasvinsky et al. 1978), a value similar to that for glycogen. Only five (S3 to S7) out of the seven glucose residues of G7 were defined in the electron density map of the GPb–G7 complex (Fig. 3CGo). The first two glucose residues could not be observed in the electron density map; similarly, there was no density for these residues in the GPb–heptulose–2-P–G7–AMP complex (Johnson et al. 1990). The conformation of the oligosaccharide and its interactions within the glycogen storage site are almost identical to those of G5 (Fig. 3DGo).

The binding of {gamma}-cyclodextrin to glycogen phosphorylase b
{gamma}-CD on binding to GPb occupies the major site of the glycogen storage site (Fig. 4Go). For reasons of comparison, the numbering scheme used for the glucose residues of {gamma}-CD corresponds to that for the linear oligosaccharides G5 and G7. Examination of the electron density in this region indicates that there is poor density for O2 and O6 of S3 and S8, O6 of S7 and S9, O2, O6, and C6 of S10, and there is also a break in the electron density for glycosidic bond between S9 and S10. Glucose residues S4, S5, S6, and S7 of {gamma}-CD are well defined within the electron density map (Fig. 3EGo).



View larger version (92K):
[in this window]
[in a new window]
  		Figure 4. The molecular surface of the GPb dimer calculated using the program GRASP (Nicholls and Honig 1991). One subunit is colored green and the other subunit is cyan. The {gamma}-CD molecule is shown as a cpk model and GPb residues interacting with {gamma}-CD are in yellow.

 
The interactions between the bound {gamma}-CD and GPb are very similar to those made by G5 and G7 to GPb. There are four direct hydrogen bonds and one water-mediated hydrogen bond between {gamma}-CD and the protein (Fig. 3FGo). These include a hydrogen bond of the O6 hydroxyl group of S4 to the O{gamma} of Ser429, and three hydrogen bonds of residue S5 (two of the O2 and O6 hydroxyl groups and one of the O5 oxygen) to the side chain atoms of Lys 437 (N{zeta}) and Asn 407 (N{delta}2). In addition, O6 of glucose residue S6 makes an indirect contact to Gln401 (O) via a water molecule. The most characteristic feature for {gamma}-CD binding to GPb is the parallel stacking interactions of the glucose plane S5 with the aromatic residue Tyr404 so that the H atom of carbon atom C4 interacts with the aromatic ring. In the GPb–{gamma}-CD complex, the buried area of {gamma}-CD is 467 Å2 and the buried surface in GPb is 391 Å2, making a total decrease in solvent-accessible area of 858 Å2 compared with the total buried surface area in the GPb–G5 complex of 716 Å2.

The geometry of the bound {gamma}-CD is exhibited in Table 3Go along with that of the uncomplexed hydrated {gamma}-CD (Harata 1987). Glucose numbering has been changed in the latter to be consistent with the present structure (initial numbering of glucose residues in parentheses). The pyranose rings in both {gamma}-CDs have the usual 4C1 conformations and the distance and angle values between the glucosidic oxygen atoms O4 are similar. However, there are significant differences both in the deviations of the O4 atoms from their mean plane and in the tilt angles of {gamma}-CD bound to GPb, compared to those of the free {gamma}-CD (Fig. 5Go). The tilt angles correspond to the dihedral angles formed between each glucose residue and the mean O4 plane, tilt angle close to zero corresponding to a glucose residue perpendicular to the O4 plane, whereas high tilt angle indicates tilting of the primary side of the glucose residues towards the cavity. Table 3Go shows that the residues interacting with the protein via hydrogen bonds and their immediate neighbors exhibit maximum tilt angles. The large deviations from the mean O4 plane mentioned above are probably the results of excessive tilting. To the same reason we must attribute the deviations in the intramolecular hydrogen bonds between glucose residues, O3n. . .O2(n+1), that stabilise the cyclodextrin ring, also observed for residues S3–S7. While in the free {gamma}-CD the distance between O3 and O2 hydroxyls range between 2.76 and 2.91 Å (although less symmetrical than in ß-CD), in the GPb-{gamma}-CD complex the O3–O2 distances range between 2.33 and 3.72 Å, indicating that some of the intramolecular hydrogen bonds have been broken. Therefore, it seems that binding of glucose residues to GPb is stabilized in conformations that distort the overall shape of the macrocycle. Table 3Go also shows the conformations of the primary hydroxyl groups O6, which differ also from those of free {gamma}-CD. The torsion angles O5–C5–C6–O6 indicate that residues S3, S4, and S6 exhibit gauche–gauche conformation and the hydroxyl groups point away from the cavity. The torsion angles in residues S5 and S8 are gauche–trans, and point towards the interior of the cavity, whereas in residues S7, S9, and S10 exhibit the rarely observed trans–gauche conformation. {gamma}-CD presents some additional close contacts in the crystal: C1 and C2 of glucose S3, and C6 of glucose S10 make van der Waals contacts with Asn595, while O2 hydroxyl of S3 contacts Lys596 from a symmetry-related molecule (1/2 + x, 1/2 - y, 1/4 - z). Because these contacts are not expected to provide much energy for stabilization, we do not think that the deformations described above for {gamma}-CD can be attributed to crystal packing forces.


View this table:
[in this window]
[in a new window]
  		Table 3. Conformations of {gamma}-CD
 


View larger version (44K):
[in this window]
[in a new window]
  		Figure 5. Stereoview of the superimposed structures of the free (white) and GPb bound {gamma}-CD (gray).

 
There are no significant conformational changes of GPb residues upon binding of {gamma}-CD. The comparison of the native GPb structure with the GPb–{gamma}-CD complex (residues 18–249, 262–312, and 326–837) gave an RMSD value of 0.22 Å for C{alpha} atoms, indicating that the two structures are similar in their overall conformation within the limits of the resolution of the study. Some minor conformational changes in the protein involve side chains of residues Glu433 and Lys437, which move to allow optimal contacts with the cyclic oligosaccharide in subsites S4 and S5. A small movement is also observed for the side chain of residue Gln408, which is displaced by glucose residue S6. These conformational changes are almost identical to those induced by the binding of G5 and G7. After superimposing the protein C{alpha} atoms of the GPb–{gamma}-CD structure onto the GPb–G5 structure, RMSDs values of 4.79, 3.25, 0.67, 0.42, and 1.16 Å were calculated for all atoms of residues S3, S4, S5, S6, and S7, respectively. The corresponding values for the pair GPb–{gamma}-CD/GPb–G7 were 4.73, 3.33, 0.62, 0.40, and 1.50 Å, respectively. Residues S5 and S6 exhibited the best superimposition whereas residues S3 and S4 show the largest RMSDs between {gamma}-CD and G5 and G7 (Fig. 6Go). These values are consistent with the differences in the conformation between {gamma}-CD and the linear oligosaccharides because, due to the cyclic conformation of {gamma}-CD, S3 and S4 cannot reach the positions of the corresponding residues in G5 or G7. This is also evident from the differences between the glycosidic angle parameters of {gamma}-CD and G5 (Tables 4Go). Kinetic experiments showed that {gamma}-CD is a mixed-type competitive inhibitor of GPb with respect to glycogen with a Ki value of 7.4 ± 1.3 mM (Table 1Go). These data suggest that {gamma}-CD is a less potent inhibitor of GPb than G7, and are also consistent with the crystallographic data.



View larger version (58K):
[in this window]
[in a new window]
  		Figure 6. Superimposed structures of GPb-G7 (white) and GPb-{gamma}-CD (gray) complexes in the glycogen storage site.

 

View this table:
[in this window]
[in a new window]
  		Table 4. Glycosidic angle parameters of maltopentaose and {gamma}-cyclodextrin when bound to the glycogen storage site of GPb
 
The crystal structures of several bacterial cyclodextrin glucosyltranferases (CGTases; EC 2.4.1.19 [EC] ), the CD producing enzymes, in complex with either linear maltosaccharides or with CD derivatives, have revealed the specific interactions of CDs in the active site of these enzymes (Knegtel et al. 1995; Strokopytov et al. 1995, 1996; Schmidt et al. 1998). In the crystal structures of different CGTases complexed with either {alpha}- or ß-CD derivatives, the glucose planes of the cyclic oligosaccharide show parallel stacking with the aromatic residues of the active site. The polar contacts between the protein and the cyclic oligosaccharide involve only four glucose units that form direct or water-mediated hydrogen bonds with residues of the active site of the enzyme. Similar mode of binding has also been observed for ß-CD bound to the maltodextrin binding protein (Sharff et al. 1993), and {alpha}-CD complexed with pig pancreatic {alpha}-amylase (Larson et al. 1994) and soybean ß-amylase (Mikami et al. 1993). This mode of binding is analogous to that observed in the present study for {gamma}-CD in GP as expected because the macrocycles are bound on the surface of the proteins mentioned and the round shape does not allow for tight contacts of all glucose units with the proteins.

The binding of ß-cyclodextrin to glycogen phosphorylase b
In the electron density maps calculated from the coordinates of the free GPb for the GPb–ß-CD complex structure there was enough density in the glycogen storage site to allow fitting of glucose residues S5 and S6 according to the G5/G7 nomenclature. A model of the ß-CD molecule was then fitted in the glycogen storage site following the electron density map in these positions, and the complex structure was subjected to one round of refinement (positional and B-factor). Successive refinement cycles significantly improved the quality of the sigmaA 2|Fo| - |Fc| electron density map for glucose residues S5 and S6, but the rest of the map around the ß-CD molecule was very poor, and most of the atoms had high temperature factors, indicating partial binding of ß-CD to GPb. These crystallographic results are consistent with (1) kinetic data showing that ß-CD is a rather poor inhibitor of GPb with a Ki value of 14.1 mM, and (2) the relatively low concentration of ß-CD (15 mM) used in the soaking experiment due to limitations in the solubility of ß-CD in aqueous solutions. Despite the partial binding of ß-CD to GPb observed in the crystal its binding mode is similar to that of {gamma}-CD. The RMSDs between the C{alpha} atoms of the ß-CD and {gamma}-CD complexes is 0.22 Å. Furthermore, after superimposition of the GPb–{gamma}-CD and GPb–ß-CD complex structures, glucose residues S5 and S6 of {gamma}-CD superimpose to their counterparts in ß-CD with RMSD values of 0.54 and 0.73 Å, respectively.

In conclusion, four different cyclodextrins were tested for inhibition of GPb. Of these, {gamma}-CD is the best inhibitor with a Ki value of 7.4 mM, which is in the same order of magnitude as that reported for the linear oligosaccharide G7 (Kasvinsky et al. 1978). The crystallographic binding studies on GPb with {gamma}-CD and ß-CD revealed that the cyclic oligosaccharides bind at the glycogen storage site by anchoring two of their glucose residues to subsites S4 and S5 in a manner similar to the binding of the linear oligosaccharides G5 and G7. However, the inhibition constant of {gamma}-CD is almost seven times higher than that of G7, and that of ß-CD is even higher (Table 1Go). The lower potency of {gamma}-CD is explained by the crystallographic results, which show that G5 or G7 are involved in a more extensive network of interactions with GPb than {gamma}-CD. This is mostly due to additional interactions made by glucose residues S3 and S6 that are not observed in the GPb–{gamma}-CD complex. Residues S3 and S6 cannot interact with the protein due to the shape of {gamma}-CD itself. {gamma}-CD binds at the glycogen storage site of T-state GPb at approximately the same position as that of G5 (or G7), and there is very little change in the structure of the enzyme. The structure of the {gamma}-CD complex helps to explain the kinetic properties of the inhibitor. On binding to the glycogen storage site, {gamma}-CD is able to inhibit glycogen binding to this site directly and glycogen binding to the catalytic site indirectly. Further kinetic and structural studies with oligosaccharide analogs are in progress, with the aim to identify the structural requirements of the glycogen storage site for the design of more effective enzyme inhibitors.


   Materials and methods
TOP
 Abstract
 Introduction
 Results and Discussion
 Materials and methods
References
 
Kinetic experiments
AMP, Glc-1-P (dipotassium salt), glycogen, maltopentaose, maltoheptaose, and other chemicals were obtained from Sigma-Aldrich Chemie GmbH. Cyclodextrins were obtained from Cyclolab. Rabbit muscle GPb was isolated, purified, recrystallized, and assayed as described (Oikonomakos et al. 2002). Care was taken to ensure that the average increase in glycogen chain length did not exceed two glucose units (Kasvinsky et al. 1978).

Crystallization and data collection
T-state tetragonal (P43212) GPb crystals were grown as described previously (Oikonomakos et al. 2000). The {gamma}-CD, G5, and G7 complexes were obtained by soaking native GPb crystals with 150 mM {gamma}-CD, 70 mM maltopentaose, or 70 mM maltoheptaose, respectively, in fresh solutions of mother liquor (10 mM Bes, 0.1 mM EDTA, pH 6.7), for at least 2 h prior to data collection. Data to 2.3 Å resolution for the {gamma}-CD, G5, and G7 complexes were collected at room temperature on an Image Plate RAXIS IV mounted on a Rigaku Ru-H3RHB rotating anode generator with a belt drive rotating anode ({lambda} = 1.5418 Å). Data for the {alpha}-CD and the ß-CD complexes, obtained by soaking native GPb crystals with 15 mM ß-CD (11 h) or with 120 mM {alpha}-CD (1 h) in mother liquor, were collected on Station X11 of the EMBL Outstation, on a MAR Research Image Plate. Data for TM{gamma}CD complex obtained by soaking native GPb crystals with 100 mM TM{gamma}CD in mother liquor, for 1 h, were collected from a single crystal using a rotating anode source ({lambda} = 1.5418 Å) and an 18-cm small Mar detector. Crystal orientation and integration of reflections, interframe scaling, partial reflection summation, data reduction, and postrefinement were all performed using DENZO and SCALEPACK (Otwinowski and Minor 1997).

Structure refinement
All crystallographic refinements were carried out with CNS version 1.1 (Brünger et al. 1998), using bulk solvent corrections. All data were included with no sigma cutoff. The starting protein structure was the refined model of the room temperature GPb–caffeine complex (code 1GFZ [PDB] ; Oikonomakos et al. 2000) with caffeine removed. Alternating cycles of manual building with the program O (Jones et al. 1991), and torsion angle dynamics, conjugate gradient minimization, and restrained individual B-factor refinement using the maximum likelihood target function as implemented in CNS were performed until the Rfree value for every model could not be improved any further. A summary of the data processing statistics and refinement parameters are given in Table 2Go. Analysis of the Ramachandran plots (CCP4 1994) showed that all residues lie in the allowed regions. Geometric parameters were determined by programs SHELXL (Sheldrick and Schneider 1997).

The structures were analyzed with the graphics program O (Jones et al. 1991). Solvent-accessible areas have been calculated for oligosaccharides in isolation and when bound to provide an estimate of the molecular surface area involved in binding to the enzyme, by using the program NACCESS (Hubbard and Thornton 1993). GP structures were superimposed over well-defined residues using LSQKAB (CCP4 1994). Coordinates for the 2.2 Å resolution GPb–maltopentaose, 2.2 Å resolution GPb–maltoheptaose, 1.94 Å resolution GPb–ß-cyclodextrin, and 2.3 Å resolution GPb–{gamma}-cyclodextrin complexes have been deposited with the RCSB Protein Data Bank (http://www.rcsb.org/; codes 1P29, 1P2B, 1P2D, and 1P2G, respectively ). All figures were prepared with the program MOLSCRIPT (Kraulis 1991) and rendered with Raster3D (Merritt and Bacon 1997).


   Acknowledgments
 
This work was supported by the General Secretariat of Research and Technology of the Greek Ministry of Development (PENED-204/2001), a Royal Society Joint Project (to L.N. Johnson and N.G.O.), NCSR Demokritos (fellowship to N.P.), EMBL Hamburg Outstation through the IHPP (HPRI-CT-1999-00017), SRS Daresbury Laboratory through the IHPP (HPRI-CT-1999-00012), and EU contract HPRI-CT-1999-00097 for work at CDSB, Jena, Germany, through minor grants to N.G.O. We thank K.E. Tsitsanou for help in kinetic experiments. We also wish to acknowledge the assistance of the staff at EMBL, Hamburg, SRS Daresbury Laboratory, CDSB, Jena, Minaksi Gosh and Elspeth Garman at the LMB, University of Oxford, for providing excellent facilities for X-ray data collection.

The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.


   References
TOP
 Abstract
 Introduction
 Results and Discussion
 Materials and methods
 References
 
Brünger, A.T., Adams, P.D., Clore, G.M., DeLano, W.L., Gros, P., GrosseKunstleve, R.W., Jiang, J.S., Kuszewski, J., Nilges, M., Pannu, N.S., et al. 1998. Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr. D 54: 905–921.[CrossRef][Medline]

CCP4. 1994. The CCP4 suite: Programs for protein crystallography. Acta Crystallogr. D 50: 760–763.[CrossRef][Medline]

Goldsmith, E.J. and Fletterick, R.J. 1983. Oligosacharide conformation and protein saccharide interactions in solution. Pure Appl. Chem. 55: 577–588.

Goldsmith, E.J., Sprang, S.R., and Fletterick, R.J. 1982. Structure of maltoheptaose by difference Fourier methods and a model for glycogen. J. Mol. Biol. 156: 411–427.[CrossRef][Medline]

———. 1989. Alternative binding models for maltopentaose in the activation site of glycogen phosphorylase a. In Protein carbohydrate interactions (ed. H.M. Einspahr), pp. 87–108. Trans. Am. Crystallogr. Assoc.

Goldsmith, E.J., Fletterick, R.J., and Withers, S.G. 1987. The three-dimensional structure of Acarbose bound to glycogen phosphorylase. J. Biol. Chem. 262: 1449–1455.[Abstract/Free Full Text]

Harata, K. 1987. The structure of the cyclodextrin complex.20. Crystal-structure of uncomplexed hydrated {gamma}-cyclodextrin. Bull. Chem. Soc. Jpn. 60: 2763–2767.

Hu, H.Y. and Gold, A.M. 1975. Kinetics of glycogen phosphorylase a with a series of semisynthetic branched saccharides. A model for binding of polysaccharide substrates. Biochemistry 14: 2224–2230.[Medline]

Hubbard, S.J. and Thornton, J.M. 1993. NACCESS. Department of Biochemistry and Molecular Biology, University College, London.

IUPAC-IUB. 1983. IUPAC-IUB, Joint Commision on Biochemical Nomenclature. Eur. J. Biochem. 131: 9–15.[Medline]

Johnson, L.N., Cheetham, J., McLaughlin, P.J., Acharya, K.R., Barford, D., and Phillips, D.C. 1988. Protein–oligosacharide interactions: Lysozyme, phosphorylase, amylases. Curr. Top. Microbiol. Immunol. 139: 81–134.[Medline]

Johnson, L.N., Acharya, K.R., Jordan, M.D., and McLaughlin, P.J. 1990. Refined crystal structure of the phosphorylase-heptulose 2-phosphate-oligosaccharide-AMP complex. J. Mol. Biol. 211: 645–661.[CrossRef][Medline]

Jones, T.A., Zou, J.Y., Cowan, S.W., and Kjeldgaard, M. 1991. Improved methods for building models in electron density maps and the location of errors in these models. Acta Crystallogr. A 47: 110–119.

Kasvinsky, P.J., Madsen, N.B., Fletterick, R.J., and Sygusch, J. 1978. X-ray crystallographic and kinetic studies of oligosaccharide binding to phosphorylase. J. Biol. Chem. 253: 1290–1296.[Free Full Text]

Knegtel, R.M., Strokopytov, B., Penninga, D., Faber, O.G., Rozeboom, H.J., Kalk, K.H., Dijkhuizen, L., and Dijkstra, B.W. 1995. Crystallographic studies of the interaction of cyclodextrin glycosyltransferase from Bacillus circulans strain 251 with natural substrates and products. J. Biol. Chem. 270: 29256–29264.[Abstract/Free Full Text]

Kraulis, P.J. 1991. MOLSCRIPT—A program to produce both detailed and schematic plots of protein structures. J. Appl. Crystallogr. 24: 946–950.[CrossRef]

Kurukulasuriya, R., Link, J.T., Madar, D.J., Pei, Z., Richards, S.J., Rohde, J.J., Souers, A.J., and Szczepankiewicz, B.G. 2003. Potential drug targets and progress towards pharmacologic inhibition of hepatic glucose production. Curr. Med. Chem. 10: 123–153.[Medline]

Larson, S.B., Greenwood, A., Cascio, D., Day, J., and McPherson, A. 1994. Refined molecular structure of pig pancreatic {alpha}-amylase at 2.1 Å resolution. J. Mol. Biol. 235: 1560–1584.[CrossRef][Medline]

Leatherbarrow, R.J. 1992. GraFit version 3.0. Erithacus Software Ltd. Staines, UK.

Makedonopoulou, S. and Mavridis, I.M. 2000. Structure of the inclusion complex of ß-cyclodextrin with 1,12-dodecanedioic acid using synchrotron radiation data; A detailed dimeric ß-cyclodextrin structure. Acta Crystallogr. B 56: 322–331.

McCormack, J.G., Westergaard, N., Kristiansen, M., Brand, C.L., and Lau, J. 2001. Pharmacological approaches to inhibit endogenous glucose production as a means of anti-diabetic therapy. Curr. Pharm. Des. 7: 1451–1474.[CrossRef][Medline]

McLaughlin, P.J., Stuart, D.I., Klein, H.W., Oikonomakos, N.G., and Johnson, L.N. 1984. Substrate–cofactor interactions for glycogen phosphorylase b: A binding study in the crystal with heptenitol and heptulose 2-phosphate. Biochemistry 23: 5862–5873.[CrossRef][Medline]

Merritt, E.A. and Bacon, D.J. 1997. Raster3D: Photorealistic molecular graphics. Methods Enzymol. B277: 505–524.

Mikami, B., Hehre, E.J., Sato, M., Katsube, Y., Hirose, M., Morita, Y., and Sacchettini, J.C. 1993. The 2.0 Å resolution structure of soybean ß-amylase complexed with {alpha}-cyclodextrin. Biochemistry 32: 6836–6845.[CrossRef][Medline]

Nicholls, A. and Honig, B. 1991. A rapid finite difference algorithm, utilising successive over relaxation to solve the Poisson-Boltzmann equation. J. Comput. Chem. 12: 435–445.[CrossRef]

Oikonomakos, N.G. 2002. Glycogen phosphorylase as a molecular target for type 2 diabetes therapy. Curr. Protein Pept. Sci. 3: 561–586.[CrossRef][Medline]

Oikonomakos, N.G., Schnier, J.B., Zographos, S.E., Skamnaki, V.T., Tsitsanou, K.E., and Johnson, L.N. 2000. Flavopiridol inhibits glycogen phosphorylase by binding at the inhibitor site. J. Biol. Chem. 275: 34566–34573.[Abstract/Free Full Text]

Oikonomakos, N.G., Kosmopoulou, M., Zographos, S.E., Leonidas, D.D., Chrysina, E.D., Somsák, L., Nagy, V., Hadady, Z., Docsa, T., Tóth, B., et al. 2002. Binding of N-acetyl-N'-ß-D-glucopyranosyl urea N-benzoyl-N'-ß-D-glucopyranosyl urea to glycogen phosphorylase b: Kinetic and crystallographic studies. Eur. J. Biochem. 269: 1684–1696.[Medline]

Otwinowski, Z. and Minor, W. 1997. Processing of X-ray diffraction data collected in oscillation mode. In Methods in enzymology (eds. C.W.J. Carter and R.M. Sweet), Vol. 276, pp. 307–326. Academic Press, New York.

Schmidt, A.K., Cottaz, S., Driguez, H., and Schulz, G.E. 1998. Structure of cyclodextrin glycosyltransferase complexed with a derivative of its main product ß-cyclodextrin. Biochemistry 37: 5909–5915.[CrossRef][Medline]

Segel, I.H. 1975. Enzyme kinetics, pp. 170–176. Wiley Interscience, New York.

Sharff, A.J., Rodseth, L.E., and Quiocho, F.A. 1993. Refined 1.8 Å structure reveals the mode of binding of ß-cyclodextrin to the maltodextrin binding protein. Biochemistry 32: 10553–10559.[CrossRef][Medline]

Sheldrick, G.M. and Schneider, T.R. 1997. SHELXL: High resolution-refinement. Methods Enzymol. 277: 319–343.

Somsak, L., Kovacs, L., Toth, M., Osz, E., Szilagyi, L., Gyorgydeak, Z., Dinya, Z., Docsa, T., Toth, B., and Gergely, P. 2001. Synthesis of and a comparative study on the inhibition of muscle and liver glycogen phosphorylases by epimeric pairs of d-gluco- and d-xylopyranosylidene-spiro-(thio)hydantoins and N-(d-glucopyranosyl) amides. J. Med. Chem. 44: 2843–2848.[CrossRef][Medline]

Strokopytov, B., Penninga, D., Rozeboom, H.J., Kalk, K.H., Dijkhuizen, L., and Dijkstra, B.W. 1995. X-ray structure of cyclodextrin glycosyltransferase complexed with acarbose. Implications for the catalytic mechanism of glycosidases. Biochemistry 34: 2234–2240.[CrossRef][Medline]

Strokopytov, B., Knegtel, R.M., Penninga, D., Rozeboom, H.J., Kalk, K.H., Dijkhuizen, L., and Dijkstra, B.W. 1996. Structure of cyclodextrin glycosyltransferase complexed with a maltononaose inhibitor at 2.6 Å resolution. Implications for product specificity. Biochemistry 35: 4241–4249.[CrossRef][Medline]

Szejtli, J. 1998. Introduction and general overview of cyclodextrin chemistry. Chem. Rev. 98: 1743–1753.[CrossRef][Medline]

Treadway, J.L., Mendys, P., and Hoover, D.J. 2001. Glycogen phosphorylase inhibitors for treatment of type 2 diabetes mellitus. Expert Opin. Investig. Drugs 10: 439–454.[CrossRef][Medline]

Truscheit, E., Frommer, W., Junge, B., Muller, L., Schmidt, D.D., and Wingender, W. 1981. Chemistry and biochemistry of microbial {alpha}-glucosidase inhibitors. Angew. Chem. Int. Ed. 20: 744–761.[CrossRef]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pinotsis, N.
Right arrow Articles by Mavridis, I. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pinotsis, N.
Right arrow Articles by Mavridis, I. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS