In Situ Extension as an Approach for Identifying Novel α-Amylase Inhibitors

A new approach for the discovery and subsequent structural elucidation of oligosaccharide-based inhibitors of α-amylases based upon autoglucosylation of known α-glucosidase inhibitors is presented. This concept, highly analogous to what is hypothesized to occur with acarbose, is demonstrated with the known α-glucosidase inhibitor, d-gluconohydroximino-1,5-lactam. This was transformed from an inhibitor of human pancreatic α-amylase with a Ki value of 18 mm to a trisaccharide analogue with a Ki value of 25 μm. The three-dimensional structure of this complex was determined by x-ray crystallography and represents the first such structure determined with this class of inhibitors in any α-glycosidase. This approach to the discovery and structural analysis of amylase inhibitors should be generally applicable to other endoglucosidases and readily adaptable to a high throughput format. A new approach for the discovery and subsequent structural elucidation of oligosaccharide-based inhibitors of α-amylases based upon autoglucosylation of known α-glucosidase inhibitors is presented. This concept, highly analogous to what is hypothesized to occur with acarbose, is demonstrated with the known α-glucosidase inhibitor, d-gluconohydroximino-1,5-lactam. This was transformed from an inhibitor of human pancreatic α-amylase with a Ki value of 18 mm to a trisaccharide analogue with a Ki value of 25 μm. The three-dimensional structure of this complex was determined by x-ray crystallography and represents the first such structure determined with this class of inhibitors in any α-glycosidase. This approach to the discovery and structural analysis of amylase inhibitors should be generally applicable to other endoglucosidases and readily adaptable to a high throughput format. α-Amylases (EC 3.2.1.1) are endoglycosidases that hydrolyze α(1,4) glucosidic linkages with net retention of configuration at the anomeric center. From their primary structures, these enzymes have been classified into glycosyl hydrolase family 13 (1Henrissat B. Biochem. J. 1991; 280: 309-316Crossref PubMed Scopus (2623) Google Scholar, 2Henrissat B. Bairoch A. Biochem. J. 1993; 293: 781-788Crossref PubMed Scopus (1771) Google Scholar, 3Henrissat B. Callebaut I. Fabrega S. Lehn P. Mornon J.P. Davies G. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 7090-7094Crossref PubMed Scopus (516) Google Scholar), a family that also includes α-glucosidases and cyclodextrin glucanotransferases. Analysis of tertiary structures reveals that the catalytic domain of family 13 enzymes, especially the active site, is very well conserved (4Aghajari N. Feller G. Gerday C. Haser R. Protein Sci. 1998; 7: 564-572Crossref PubMed Scopus (161) Google Scholar, 5Brayer G.D. Luo Y. Withers S.G. Protein Sci. 1995; 4: 1730-1742Crossref PubMed Scopus (292) Google Scholar, 6Brzozowski A.M. Davies G.J. Biochemistry. 1997; 36: 10837-10845Crossref PubMed Scopus (196) Google Scholar, 7Machius M. Wiegand G. Huber R. J. Mol. Biol. 1995; 246: 545-559Crossref PubMed Scopus (308) Google Scholar, 8Kadziola A. Abe J. Svensson B. Haser R. J. Mol. Biol. 1994; 239: 104-121Crossref PubMed Scopus (230) Google Scholar, 9Uitdehaag J.C. Mosi R. Kalk K.H. van der Veen B.A. Dijkhuizen L. Withers S.G. Dijkstra B.W. Nat. Struct. Biol. 1999; 6: 432-436Crossref PubMed Scopus (368) Google Scholar, 10Ramasubbu N. Paloth V. Luo Y.G. Brayer G.D. Levine M.J. Acta Crystallogr. Sect. D Biol. Crystallogr. 1996; 52: 435-446Crossref PubMed Scopus (227) Google Scholar). Indeed, a number of studies have shown that members of this family of glycosidases utilize a common double displacement catalytic mechanism (9Uitdehaag J.C. Mosi R. Kalk K.H. van der Veen B.A. Dijkhuizen L. Withers S.G. Dijkstra B.W. Nat. Struct. Biol. 1999; 6: 432-436Crossref PubMed Scopus (368) Google Scholar, 11McCarter J.D. Withers S.G. J. Biol. Chem. 1996; 271: 6889-6894Abstract Full Text PDF PubMed Scopus (121) Google Scholar, 12Rydberg E.H. Li C. Maurus R. Overall C.M. Brayer G.D. Withers S.G. Biochemistry. 2002; 41: 4492-4502Crossref PubMed Scopus (103) Google Scholar, 13Tao B.Y. Reilly P.J. Robyt J.F. Biochim. Biophys. Acta. 1989; 995: 214-220Crossref PubMed Scopus (60) Google Scholar), in which a glycosyl-enzyme intermediate is formed and hydrolyzed with acid/base catalysis via oxocarbenium ion-like transition states (14Sinnott M.L. Chem. Rev. 1990; 90: 1171-1202Crossref Scopus (1491) Google Scholar, 15Rye C.S. Withers S.G. Curr. Opin. Chem. Biol. 2000; 4: 573-580Crossref PubMed Scopus (435) Google Scholar, 16Zechel D.L. Withers S.G. Acc. Chem. Res. 2000; 33: 11-18Crossref PubMed Google Scholar). In humans, the pancreatic α-amylase (HPA) 1The abbreviations used are: HPA, human pancreatic α-amylase; CGTase, cyclodextrin glucanotransferase; CNP-G3, 2-chloro-4-nitrophenyl α-maltotrioside; G3F, α-maltotriosyl fluoride; G2-GHIL, maltosyl-α(1,4)-d-gluconohydroximino-1,5-lactam; GHIL, d-gluconohydroximino-1,5-lactam; MeG2F, 4′-O-methyl α-maltosyl fluoride; MeG2-GHIL, 4″-O-methyl-maltosyl-α(1,4)-d-gluconohydroximino-1,5-lactam; MALDI-TOF, matrix-assisted laser desorption ionization time-of-flight. is responsible for cleaving large malto-oligosaccharides to smaller oligosaccharides, which are then substrates for intestinal α-glucosidases (Fig. 1). This digestion process is important for glucose absorption from the intestine to the blood, and in principle, control of HPA activity can be used as a means of controlling blood glucose levels. In fact, HPA activity has been correlated to post-prandial blood glucose levels (17Jenkins D.J. Taylor R.H. Goff D.V. Fielden H. Misiewicz J.J. Sarson D.L. Bloom S.R. Alberti K.G. Diabetes. 1981; 30: 951-954Crossref PubMed Google Scholar, 18Meyer B.H. Muller F.O. Kruger J.B. Clur B.K. Grigoleit H.G. S. Afr. Med. J. 1984; 66: 222-223PubMed Google Scholar, 19Taylor R.H. Jenkins D.J. Barker H.M. Fielden H. Goff D.V. Misiewicz J.J. Lee D.A. Allen H.B. MacDonald G. Wallrabe H. Diabetes Care. 1982; 5: 92-96Crossref PubMed Scopus (33) Google Scholar), and inhibitors of α-amylase have been successfully used in the treatment of diseases such as diabetes or obesity where control of the blood glucose level is essential (20Bailey C.J. Chem. Ind. 1998; : 53-57Google Scholar). Arguably, the most studied inhibitor of α-amylase is the naturally occurring and commercially available drug, acarbose (Fig. 2a), which has a Ki value in the low nanomolar range. This pseudo-tetrasaccharide is composed of a valienamine (unsaturated cyclitol) linked “α(1,4)” by an amine linkage to 6″-deoxy-maltotriose. The valienamine portion mimics the flattened sugar ring of the oxocarbenium ion-like transition state, whereas the exocyclic nitrogen places a proton acceptor in a position to interact with the acid/base catalyst at the active site. Indeed, with another family 13 enzyme, the CGTase from Bacillus circulans, this compound has been demonstrated to be a transition state analogue (21Mosi R. Sham H. Uitdehaag J.C.M. Ruiterkamp R. Dijkstra B.W. Withers S.G. Biochemistry. 1998; 37: 17192-17198Crossref PubMed Scopus (59) Google Scholar). As would be expected, structural studies of acarbose bound to HPA and other family 13 glycosidases have shown the valienamine moiety binding to the -1 subsite (22Brayer G.D. Sidhu G. Maurus R. Rydberg E.H. Braun C. Wang Y.L. Nguyen N.T. Overall C.M. Withers S.G. Biochemistry. 2000; 39: 4778-4791Crossref PubMed Scopus (198) Google Scholar), where sugar distortion takes place in the presumed catalytic mechanism. Interestingly, when acarbose is bound to HPA in this structure, it is modified by the enzyme through the addition of a maltosyl unit to the nonreducing end of the valienamine ring and the loss of a glucose moiety at the reducing end (Fig. 2a). The inhibitor therefore occupies all five subsites of HPA (–3 to +2). Similar, although not identical, modifications have also been seen in structural studies of other α-amylases in complexes with acarbose (6Brzozowski A.M. Davies G.J. Biochemistry. 1997; 36: 10837-10845Crossref PubMed Scopus (196) Google Scholar, 23Qian M.X. Haser R. Buisson G. Duee E. Payan F. Biochemistry. 1994; 33: 6284-6294Crossref PubMed Scopus (285) Google Scholar, 24Brzozowski A.M. Lawson D.M. Turkenburg J.P. Bisgaard-Frantzen H. Svendsen A. Borchert T.V. Dauter Z. Wilson K.S. Davies G.J. Biochemistry. 2000; 39: 9099-9107Crossref PubMed Scopus (124) Google Scholar). Because increased subsite occupancy is often associated with an increase in the overall binding affinity of an inhibitor, such a modification presumably results in a tighter binding molecule. To date, however, beyond the crystallographic data for the complex, there has been no direct evidence for such a product being formed in solution with α-amylase. Despite the considerable interest in α-amylases, there are relatively few other known inhibitors of this group of enzymes. From kinetic and structural studies, one would expect that the greatest transformation of the substrate on going from the ground state to the transition state will take place on the sugar occupying the -1 subsite. As such, the interactions at the -1 subsite should be particularly optimized for transition state binding, and therefore transition state mimics should bind particularly well at this site. Because the active sites of other family 13 α-glucosidases are thought to be structurally similar to that of HPA, it is likely that transition state analogues of such enzymes will interact similarly with the -1 subsite residues of α-amylase. Furthermore, it seems reasonable that the affinity of the inhibitor would correlate, in part, with the number of subsites with which the compound interacts. Therefore, an α-glucosidase inhibitor extended on the reducing and/or nonreducing end by a malto-oligosaccharide so as to occupy the other subsites should make a good inhibitor of α-amylases. In the past, the main difficulty in testing this hypothesis has been in the synthesis of such extended α-glucosidase inhibitors, although some work to this end has been carried out (25Takada M. Ogawa K. Saito S. Murata T. Usui T. J. Biochem. (Tokyo). 1998; 123: 508-515Crossref PubMed Scopus (8) Google Scholar, 26Takada M. Ogawa K. Murata T. Usui T. J. Carbohydr. Chem. 1999; 18: 149-163Crossref Scopus (2) Google Scholar, 27Uchida R. Nasu A. Tokutake S. Kasai K. Tobe K. Yamaji N. Chem. Pharm. Bull. 1999; 47: 187-193Crossref PubMed Scopus (15) Google Scholar, 28Yoon S.H. Robyt J.F. Carbohydr. Res. 2003; 338: 1969-1980Crossref PubMed Scopus (59) Google Scholar). In this paper, we describe a method to extend α-glucosidase inhibitors in situ, both in solution and within α-amylase crystals, by co-addition of an activated substrate. In doing so, we have been able to show that a potent α-glucosidase inhibitor, d-gluconohydroximino-1,5-lactam (GHIL; Fig. 2b) (29Hoos R. Vasella A. Rupitz K. Withers S.G. Carbohydr. Res. 1997; 298: 291-298Crossref Scopus (45) Google Scholar), which is a HPA inhibitor, is into a inhibitor of this it should be to a of known α-glucosidase inhibitors a kinetic for HPA by the and then the structure of the of the and from HPA was to E.H. Sidhu G. J. Wang Y. S. Overall C.M. Brayer G.D. Withers S.G. Protein Sci. 1999; PubMed Scopus (59) Google Scholar). was a from and can be from M. S. R. Chem. 1984; : Google and S. 1990; Scopus Google Scholar, R. Vasella A. Rupitz K. Withers S.G. Acta. 1993; Scopus Google to of and 4′-O-methyl α-maltosyl will be of the studies carried out at in mm mm of by HPA was the addition of enzyme by the increase in at a to a control The of by HPA was by the increase in upon the addition of enzyme an to a the of the data the Scholar). and studies by HPA with and acarbose or and the activity of HPA was by the addition of and the increase in at in the of the or in the of with was by and with HPA for at in mm mm A solution of in was used as the The and in a and of this was on the and in The a in with an of Ki and of determined by the of of or in the of a of inhibitor The data the and in the of a to of the of and in 25 of mm was by the in the of CGTase at for The was by the was and the was by a solution as to in of the product was in The was in and the was in The solution was with and and then of HPA (22Brayer G.D. Sidhu G. Maurus R. Rydberg E.H. Braun C. Wang Y.L. Nguyen N.T. Overall C.M. Withers S.G. Biochemistry. 2000; 39: 4778-4791Crossref PubMed Scopus (198) Google Scholar). The inhibitor complex was for structural analysis by a HPA in a solution mm for A of HPA with both and was by first in a solution mm for by a in a solution mm A similar approach was used in the of the complex with and data for all complex on a at by a at and data and to structure with the of Z. 1997; Scopus Google Scholar). are in I. The of all of the complexes with that of HPA (22Brayer G.D. Sidhu G. Maurus R. Rydberg E.H. Braun C. Wang Y.L. Nguyen N.T. Overall C.M. Withers S.G. Biochemistry. 2000; 39: 4778-4791Crossref PubMed Scopus (198) Google Scholar), and as such, this structure was used as the for these was carried out with the P. J. M. T. Acta Crystallogr. 1998; PubMed Scopus (8) Google Scholar). In these of and with with M. Acta Crystallogr. Sect. A. 1991; 47: PubMed Scopus Google Scholar). In the of this the of the was based on a of the residues the was with and the of the of HPA, bound inhibitor on the of both and inhibitor then to the overall In an moiety bound to the of was in the structural and this from a and in the based on the to and the of a of In a the inhibitor, and to The are in of in to the or in to the or in to the or of within in to the or of of and of the data was for the The in to the or of the data was for the in a new first out of to inhibitors of α-glucosidases would also as inhibitors of As we the potent α-glucosidase inhibitor with α-glucosidase (29Hoos R. Vasella A. Rupitz K. Withers S.G. Carbohydr. Res. 1997; 298: 291-298Crossref Scopus (45) Google The with the substrate α-maltotriosyl a In this a Ki value of mm was this is to be it that and other such be inhibitors by a for the of extended and potent inhibitors that also occupy sugar binding subsites in the enzyme active site. Indeed, that of these inhibitors with glucose residues to in affinity from the for substrates to increase upon the addition of glucose to the nonreducing end of a substrate. This to transition state from these sugar residues the increase from to on going from maltosyl to (22Brayer G.D. Sidhu G. Maurus R. Rydberg E.H. Braun C. Wang Y.L. Nguyen N.T. Overall C.M. Withers S.G. Biochemistry. 2000; 39: 4778-4791Crossref PubMed Scopus (198) Google Scholar). studies have shown that oligosaccharide-based are inhibitors of the pancreatic α-amylase (25Takada M. Ogawa K. Saito S. Murata T. Usui T. J. Biochem. (Tokyo). 1998; 123: 508-515Crossref PubMed Scopus (8) Google Scholar, 26Takada M. Ogawa K. Murata T. Usui T. J. Carbohydr. Chem. 1999; 18: 149-163Crossref Scopus (2) Google Scholar, 27Uchida R. Nasu A. Tokutake S. Kasai K. Tobe K. Yamaji N. Chem. Pharm. Bull. 1999; 47: 187-193Crossref PubMed Scopus (15) Google Scholar). To α-amylase inhibitors, a the commercially available was To this compound as a a Ki value of 18 mm was for GHIL, which is that as substrate. This analysis other data that this was not an of kinetic analysis and that there was a in the affinity of HPA for on the substrate. evidence for such a in activity is when for HPA activity with at of GHIL, when a in was a of such was in the of inhibitor, substrate as the of this of HPA with to of activity with not both and on as of this The most likely for both the in Ki and the is that the kinetic analysis of with CNP-G3, of a moiety to is This would an of that the analogous to what is presumed to occur in of HPA with is a substrate for HPA is value is for the moiety will likely be to the inhibitor at a when with with this the the of is and at levels when when CNP-G3, for the Ki value the CNP-G3, of the state of the and is seen as a in the catalytic activity To this hypothesis HPA was with and for to an extended inhibitor is then HPA should be these the is with The determined these was of the of a control in which HPA was with or that HPA is a the inhibitor and substrate and that the product of this is a inhibitor of HPA (Fig. A was also carried out with acarbose in place of and similar results with with for the control in which no acarbose was not Because structural studies have shown acarbose to an product in the of HPA, this for the that inhibitors are the in the kinetic analysis as one expect from this of HPA with and glucose not in a in the of of the substrate (Fig. was to the in activity by an Ki value for the inhibitor formed The carried out of of with a of and of at the HPA activity was these then in the of a the of to a and an Ki value of was from the this and the to is the of in the of mm not In with the Ki value of 18 mm determined for this Ki value represents an increase in affinity as a of The these when with that and/or of are to with The Ki value however, represents a of the Ki value for the inhibitor it all of the inhibitor is to the binding which is highly and which analysis reveals not to be the The Ki value therefore be to an formed the by a of the and of the is being formed at well the level of the of both and therefore such that was at and was at the Ki value of the Analysis of such by of and to the of linked to a maltosyl or (Fig. that HPA will the of of an for that is a inhibitor of HPA, an extended of was a approach (Fig. To both the synthesis and subsequent kinetic analysis of the a was in which the group at the nonreducing end was an inhibitor should not at the and therefore should a kinetic of this was by with in the of CGTase from B. circulans, which has been shown to be highly in sugar to analogues R. Nasu A. Tokutake S. Kasai K. Tobe K. Yamaji N. Chem. Pharm. Bull. 1999; 47: 187-193Crossref PubMed Scopus (15) Google the product that would be seen when α-amylases. the the group in a This product was by in Analysis by with the product and of the by with in the α(1,4) linkage formed the maltosyl moiety and the the and inhibitor in kinetic analysis of was The of the should kinetic analysis and of a Ki As shown in Fig. as a was shown to be a inhibitor of HPA with a Ki value of 25 μm. This Ki value is of that for the inhibitor, GHIL, as a that of has binding affinity for the active of HPA, it has to bound structure crystallographic (Fig. this analogue is bound in the subsite with a ring not for inhibitors bound at this subsite. This of results in the inhibitor interactions and from with studied inhibitors such as acarbose (Fig. and G.D. Sidhu G. Maurus R. Rydberg E.H. Braun C. Wang Y.L. Nguyen N.T. Overall C.M. Withers S.G. Biochemistry. 2000; 39: 4778-4791Crossref PubMed Scopus (198) Google Scholar). The main interactions the group with the of the group with the of the acid/base and a the group and the of the catalytic In with kinetic studies, is not bound in the active of HPA, as is from the for of this inhibitor in with of the enzyme as a the bound of the inhibitors product and acarbose in the active of active residues are shown for the interactions formed in the active of HPA by formed on with and and acarbose The subsites of HPA by bound moiety are and are with their A of in which HPA with both and in a modified inhibitor being bound in the active of this enzyme and The presumably a product of and by the enzyme, is a trisaccharide analogue that the to -1 The in the -1 subsite is a bound in the whereas the and subsites are by and This product to one of the by (Fig. and is to the inhibitor that was and studied (Fig. The for all of the sugar is well and the overall of the are that for (Fig. is also seen for another and of GHIL, bound the and subsites (Fig. the for the of this group are very that binding in this is very The interactions formed by the glucose of in subsites and are very similar to the interactions for acarbose, as are the of these sugar and and and The the of the group in subsite which results in this group with a whereas in the acarbose structure, the was to is the of an the group in subsite and in the The structure of the complex formed when HPA is with and was also determined as a and Fig. the bound has a very similar to that of the complex, the of HPA to a binding inhibitor and an activated substrate to a binding bound was in this The the complexes with and is the of the group the with can be into the activated substrate moiety used to by This an new for end to the HPA active to inhibitor binding of the most from these structural is the for the moiety of and In has been shown to a whereas the structure of a R. Vasella A. Rupitz K. Withers S.G. Acta. 1993; Scopus Google Scholar). is bound to HPA, it to a similar to the by the analogue bound to the from V. R. Withers S.G. Biochemistry. 2000; 39: PubMed Scopus (59) Google Scholar). In the moiety of and MeG2-GHIL, bound to the -1 subsite of HPA, an A of the is that the group and is in the acid/base catalyst with the exocyclic nitrogen being within of of the group (Fig. This is as for an Vasella Chem. 1999; PubMed Google Scholar). In fact, such interactions have also been in the C. an from family V. R. Withers S.G. Biochemistry. 2000; 39: PubMed Scopus (59) Google Scholar, Vasella Chem. 1999; PubMed Google Scholar). In C. this is the this catalytic and the whereas in HPA the of is also in to the From studies it can be seen should the moiety of the bound a as was seen in the structure, these interactions would not be Furthermore, of the position of in the structure, with position in the acarbose structure, that it has to this with the has been that the structurally similar is not a good inhibitor of α-glucosidases the group is not for such interactions (29Hoos R. Vasella A. Rupitz K. Withers S.G. Carbohydr. Res. 1997; 298: 291-298Crossref Scopus (45) Google Scholar, Vasella Chem. 1999; PubMed Google Scholar). the was also to be a inhibitor, presumably for the (29Hoos R. Vasella A. Rupitz K. Withers S.G. Carbohydr. Res. 1997; 298: 291-298Crossref Scopus (45) Google Scholar). Indeed, it has been that the increased of the exocyclic nitrogen on it by the nitrogen should the the and catalytic acid/base so that would be The structural such a in the of the binding of to the -1 subsite. is that the catalytic and the exocyclic where the is to the of the with from the exocyclic of not interact with the as have been on the of the of the with the glycosyl-enzyme intermediate in CGTase (9Uitdehaag J.C. Mosi R. Kalk K.H. van der Veen B.A. Dijkhuizen L. Withers S.G. Dijkstra B.W. Nat. Struct. Biol. 1999; 6: 432-436Crossref PubMed Scopus (368) Google Scholar). it with the an that is also in the complex where no is of the and of the moiety with HPA are also very similar to in the HPA is an important for controlling blood glucose very few have been on the of inhibitors of this enzyme, in large of the in the synthesis of oligosaccharide-based of the of HPA to it is to analogues in by the compound with an activated substrate The of such and potent inhibitors from binding analogues can be readily by with a substrate and with The structure of the and interactions with the enzyme can be determined with the analogue and This is adaptable to a and is for the of HPA Vasella for the of and for the of

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.