- Email: [email protected]
The acbH gene of Actinoplanes sp. encodes a solute receptor with binding activities for acarbose and longer homologs
Research in Microbiology 156 (2005) 322–327 www.elsevier.com/locate/resmic
The acbH gene of Actinoplanes sp. encodes a solute receptor with binding activities for acarbose and longer homologs Claudia Brunkhorst a , Udo F. Wehmeier b , Wolfgang Piepersberg b , Erwin Schneider a,∗ a Humboldt Universität zu Berlin, Institut für Biologie/Bakterienphysiologie, Chausseestr. 117, 10115 Berlin, Germany b Bergische Universität Wuppertal, Chemische Mikrobiologie, Gauß-Str. 20, 42097 Wuppertal, Germany
Received 7 September 2004; accepted 25 October 2004 Available online 22 December 2004
Abstract Acarbose, a pseudomaltotetraose, is produced by strains of the genus Actinoplanes and is a potent inhibitor of α-glucosidases, including those from the human intestine. Therefore, it is used in the treatment of patients suffering from type 2 diabetes. The benefits of acarbose for the producer are not known; however, besides acting as an inhibitor of α-amylases secreted by competitors, a role as a ‘carbophor’ has been proposed. This would require a transport system mediating its uptake into the cytoplasm of Actinoplanes sp. A putative sugar ATP binding cassette (ABC) transport system, the genes of which are included within the biosynthetic gene cluster for acarbose, was suggested to be a possible candidate. The genes acbHFG encode a possible sugar binding protein (AcbH) and two membrane integral subunits (AcbFG). A gene coding for an ATPase component is missing. Since Actinoplanes sp. cannot yet be genetically manipulated we performed experiments to identify the substrate(s) of the putative transporter by assessing the substrate specificity of AcbH. The protein was overproduced in Escherichia coli as His10 -fusion protein, purified under denaturating conditions and renatured. Refolding was verified by circular dichroism spectroscopy. Surface plasmon resonance studies revealed that AcbH binds acarbose and longer derivatives, but not maltodextrins, maltose or sucrose. Immunoblot analysis revealed the association of AcbH with the membrane fraction of Actinoplanes cells that were grown in the presence of maltose, maltodextrins or acarbose. Together, these findings suggest that the AcbHFG complex might be involved in the uptake of acarbose and are consistent with a role for acarbose as a ‘carbophor’. 2004 Elsevier SAS. All rights reserved. Keywords: Acarbose; ABC transport; Solute receptor; Actinoplanes sp.
1. Introduction The α-glucosidase inhibitor acarbose (Fig. 1), a pseudomaltotetraose produced by strains of the genera Actinoplanes and Streptomyces, is a member of an unusual group of bacterial secondary metabolites, all of which inhibit various α-glucosidases, especially from the intestine [11,17]. Acarbose is produced industrially in media containing maltose and maltotriose using developed strains of Actinoplanes sp. SE50. It is used in the treatment of patients suffering from type 2 diabetes mellitus. Besides acarbose, the organism produces an extensive number of homologs dependent on the * Corresponding author.
E-mail address: [email protected] (E. Schneider). 0923-2508/$ – see front matter 2004 Elsevier SAS. All rights reserved. doi:10.1016/j.resmic.2004.10.016
carbon source that differ in the number and chemical nature of sugars added to the core unit and in their inhibitory potential. The possible benefits that the bacteria derive from the production of acarbose and homologs in their natural environment remain unclear. As a potent inhibitor of α-amylases [18] and maltose/maltodextrin transport systems [2,8] of other microorganisms, acarbose can be envisaged as a means to restrict access of competitors to carbon sources like starch and other α-1,4-glucans. This notion is further supported by the observation that Actinoplanes sp. contains at least one maltose/maltodextrin transporter that is insensitive to acarbose (Brunkhorst and Schneider, manuscript in preparation). Recently, a cluster comprising 25 genes (acb) was identified in Actinoplanes sp. SE50/110 the products of which
C. Brunkhorst et al. / Research in Microbiology 156 (2005) 322–327
323
2. Materials and methods 2.1. Chemicals
Fig. 1. Structure of acarbose. Acarbose is a pseudo-oligosaccharide that consists of an unsaturated cyclitol moiety (A), an aminodeoxyhexose (B) (together also called acarviosine), and a normal maltose (rings C and D).
Acarbose (acarviosyl-1,4-maltose), component 5C (acarviosyl-1,4-maltose-1,4-glucose-1,1-glucose, component mixture 6AB (acarviosyl-1,4-maltose-1,4-maltose-1,4-fructose and acarviosyl-1,4-maltose-1,4-maltose-1,4-glucose) and MD50 (mixture of maltose and maltotriose, containing trace amounts of longer dextrins) were kind gifts from H. Wehlmann (Bayer AG, Wuppertal, Germany). Maltodextrins were purchased from Pfanstiehl Laboratories (IL, USA). 2.2. Bacterial strains, media and growth conditions
are involved in the biosynthesis of acarbose [18]. Biochemical characterization of several of the gene products has subsequently led to the proposal of a biosynthetic pathway [6,15,18,19] and of an additional function for acarbose [18]. Accordingly, the biosynthetic reactions in the cytoplasm result in either acarbose-7-phosphate or a precursor, e.g., the monoglucosylated acarviosine-7-phosphate that is secreted into the medium, possibly by an ABC exporter (AcbWXY). Here, acarbose and/or derivatives are formed by action of an acarviosyltransferase, AcbD, that catalyzes the transfer of the acarviosyl moiety (Fig. 1) to maltooligosaccharides and dextrins [6]. Re-uptake of acarbose and derivatives would then result in a net gain of carbon and energy. Thus, acarbose or its core subunit might function as a ‘carbophor’ that is secreted and charged with additional glucose moieties thereby extracting glucose from the extracellular pool and simultaneously tagging it for import. Consistent with this hypothesis is the finding that Actinoplanes sp. can utilize acarbose as sole source of carbon and energy (Brunkhorst and Schneider, manuscript in preparation). Within the acb cluster a putative operon (acbHFG) encoding components of an ATP-binding cassette (ABC) import system was found that consists of an extracellular solute binding protein (AcbH) and two membrane-spanning subunits (AcbFG). As often observed for sugar ABC transport systems of Gram-positive bacteria, a gene translating into an ATPase component is lacking [8,12]. BLAST searches revealed homology to members of the CUT 1 (carbohydrate uptake transporter) subfamily [13] which makes the transporter a candidate for an acarbose uptake system. Unfortunately, Actinoplanes sp. cannot yet be genetically manipulated. Thus, elucidating gene functions requires the biochemical characterization of their products. Since the solute receptors of ABC import systems are the major determinants of substrate specificity [1], we have focused on AcbH to elucidate the role of the putative AcbHFG transporter. We report on the overproduction in Escherichia coli, purification and characterization of AcbH. Our data are consistent with the notion that the AcbHFG proteins are involved in the transport of acarbose and derivatives in the context of the carbophor model.
E. coli strain BL21(DE3)(pLysS) (Stratagene, Heidelberg, Germany) harboring plasmid pCB10 (expressing acbH under control of the pT7lac promoter) was grown in LB medium supplemented with ampicillin (100 µg/ml) and chloramphenicol (20 µg/ml), at 30 ◦ C. At OD650 = 1, 0.1 mM isopropyl thio-β-galactoside (IPTG) was added and growth continued for 3 h. Streptomyces lividans TK23 harboring plasmid pCB12 (expressing acbH under control of the ermEp promotor) was grown in trypton soybean broth (TSB, Oxoid), supplemented with thiostrepton (25 µg/ml) for 48 h at 30 ◦ C [19]. Actinoplanes sp. SE50/110 (ATCC 31044) was grown for 72 h at 30 ◦ C in TSB or in minimal salt medium (containing 5.7 g Tris–HCl, 1 g NaCl, 2 g (NH4 )2 SO4 , 0.5 g K2 SO4 , 0.2 g MgSO4 , 0.1 g CaCl2 , 0.34 g KH2 PO4 per liter of deionized water, adjusted to pH 7.2 with H2 SO4 (20%, v/v); 10 ml of a trace element solution [7] were added after autoclaving) supplemented with the indicated carbon sources (0.5%, w/v). 2.3. Cloning and subcloning of acbH The acbH gene lacking the signal sequence was amplified by PCR from genomic DNA of Actinoplanes sp. SE50/110 using the oligonucleotide primers acbH12 (5 -C GAG CGG CAT ATG GGC AGT GAC GAC AAG AGC GGG ACG-3 ) and acbH13 (3 -CCC GCC GGA TCC CCG GTC GAA TGG TCG GGG TCA TCA-5 ). The primers were designed such that the resulting fragment could be ligated with plasmid pET19b (Novagene, Madison, WI, USA) via NdeI and BamHI restriction sites, yielding plasmid pCB10. As a consequence the translated polypeptide has 10 consecutive histidine residues fused to the N-terminal amino acid of mature AcbH which was changed from cysteine to methionine. Plasmid pCB12 is a derivative of pCB10 that carries his10 –acbH on the E. coli/Streptomycetes shuttle vector pPWW49 [4]. 2.4. Purification and renaturation of His10 –AcbH Cells of strain BL21(DE3)(pCB10, pLysS) were harvested, resuspended in buffer 1 (20 mM Tris–HCl, pH 7.2,
324
C. Brunkhorst et al. / Research in Microbiology 156 (2005) 322–327
containing 5% (v/v) glycerol, 300 mM NaCl, 20 mM 2-mercaptoethanol, 0.1 mM phenylmethylsulfonyl fluoride; 1/10 of the culture volume), and disrupted by passage through a French press. After a low speed spin, the pellet containing His10 –AcbH was dissolved in buffer 1 containing 8 M urea (buffer 2) at 5 mg/ml. After gentle stirring for 30 min at room temperature insoluble material was removed by ultracentrifugation and the supernatant was mixed with Ni-NTA matrix equilibrated in buffer 2. The resin was washed with buffer 2 containing 20 mM imidazole followed by elution of His10 –AcbH with buffer 2 containing 150 mM imidazole. The peak fraction was adjusted to 1 M L-arginine and passed through a desalting column (PD10, AmershamPharmacia) to remove urea [3,10]. The eluent was subsequently passed through a second PD10 column, equilibrated with 20 mM Tris–HCl, pH 7.2, 5% glycerol, 300 mM NaCl, and the resulting protein containing fractions were ultracentrifuged to remove insoluble material. 2.5. Surface plasmon resonance analysis Binding of sugars to AcbH was detected by surface plasmon resonance analysis using a Biacore X optical biosensor (Biacore Uppsala, Sweden). Freshly renatured His10 –AcbH (0.273 pmol) was immobilized on a NTA sensor chip with an efficiency of 13 000 resonance units. The sugars tested were applied at a flow rate of 10 µl/min. Dissociation from and association of the sugars to the immobilized protein were monitored for 2 min each. Purified His10 –MalK (0.15 pmol) of Salmonella typhimurium (gift of B. Blüschke, HU Berlin) was used as reference. Dissociation constants were calculated using the manufacturer’s software package (fitted according to Langmuir).
Fig. 2. Steps in the purification and renaturation of His10 –AcbH from inclusion bodies. His10 –AcbH was purified by Ni-NTA chromatography in the presence of 8 M urea and subsequently renatured in the presence of L-arginine as described in Section 2. Lanes: 1, low speed pellet fraction (uninduced); 2, molecular mass standards (KDa); 3, low speed pellet fraction (induced); 4, cytosolic fraction; 5, supernatant after treatment of the low speed pellet fraction with 8 M urea and subsequent ultracentrifugation; 6, eluate fraction from Ni-NTA chromatography (150 mM imidazole); 7, renatured His10 –AcbH after ultracentrifugation (2.3 µg).
3. Results
When cells of E. coli strain BL21(DE3)(pLysS) harboring pCB10 were grown in rich medium and induced for gene expression, a protein of the expected molecular size of about 47 kDa was found in substantial amounts in whole cell extracts. After cell fractionation, the recombinant protein was recovered almost exclusively with the pellet obtained after a low speed centrifugation step. Attempts to increase the amount of soluble AcbH failed, including variations in medium composition, growth temperature or co-expression of E. coli chaperones or tRNAs for translation of GC rich mRNAs. Since overproduction of other gene products from the acb cluster was successfully achieved in S. lividans [6], a similar experiment was performed with acbH. Although the protein could be detected by immunoblotting in the cytoplasmic fraction of strain TK23(pCB12) the yield was insufficient for further biochemical characterization. Thus, His10 – AcbH was purified from the insoluble fraction of E. coli BL21(DE3)(pCB10, pLysS) under denaturing conditions in the presence of 8 M urea by Ni-NTA affinity chromatography and subsequently renatured as described in Section 2 (Fig. 2). Refolding of His10 –AcbH was successful, as judged by circular dichroism spectroscopy which indicated a substantial portion of α-helical secondary structure (Fig. 3). Routinely, 14 mg of purified refolded protein were obtained from 1-l culture.
3.1. Purification of an N-terminal His10 –AcbH fusion protein
3.2. Synthesis of AcbH in Actinoplanes sp. in the presence of different sugars
In order to explore the possible function of the putative extracellular solute receptor, AcbH, the acbH gene lacking the signal peptide-encoding sequence was amplified from genomic DNA of Actinoplanes sp. SE50/110 by PCR and subcloned in expression vector pET19b, yielding pCB10.
The purified protein was used to raise a polyclonal antiserum which then allowed us to study the carbon sourcedependent synthesis of AcbH in Actinoplanes sp. SE50/110. As shown in Fig. 4, AcbH was detected after 72 h in the membrane fraction of cells that were grown in minimal
2.6. Miscellaneous methods SDS–polyacrylamide gel electrophoresis (SDS–PAGE) and immunoblotting were performed as in [14]. The antiserum raised against purified AcbH was obtained from Pineda antibody service (Berlin). Protein was determined with the BCA-protein assay kit from Pierce (Germmany). The CD spectrum was recorded with a Jasco 720 (UK) in 50 mM Tris–HCl, pH 7.2, 5%, v/v) glycerol, 150 mM NaCl, at a protein concentration of 0.15 mg/ml.
C. Brunkhorst et al. / Research in Microbiology 156 (2005) 322–327
325
Fig. 3. Synthesis of acbH in Actinoplanes sp. in the presence of different sugars. Cells were grown in minimal medium supplemented with the indicated carbon sources or in TSB alone for 72 h at 30 ◦ C. Aliquots from membrane fractions (30 µg protein) of each culture were subjected to SDS–PAGE followed by immunoblotting. The blot was probed with an anti-AcbH antiserum. Control: 0.1 µg purified His10 –AcbH.
medium supplemented with either MD50, maltose or acarbose. Some AcbH was also observed in TSB. In contrast, AcbH was absent in cells grown in the presence of glucose or glycerol. Association of AcbH with the membrane was further confirmed by the observation that treatment with the non-ionic detergent Triton X-100 resulted in solubilization of the protein. Together, these data are consistent with a role of AcbH as solute receptor which in Gram-positive bacteria is anchored to the cytoplasmic membrane by an N-terminal lipid modification [16]. Moreover, AcbH is synthesized under conditions that were also shown to induce expression of other acb genes ([6]; Y. Rockser and U.F. Wehmeier, unpublished).
Fig. 4. Far-UV-CD spectrum of purified, renatured AcbH. The spectrum was recorded as described in Section 2. The minima in ellipticity ([]) at 222 and 208 nm are characteristic of α-helical content.
3.3. Determination of substrate specificity of His10 –AcbH Homology searches indicated that AcbH is a component of an ABC import system belonging to the CUT 1 subfamily [13], the members of which predominantly transport di- and oligosaccharides. Since maltose and maltotriose were found to induce AcbH synthesis, a function of the acbHFG gene products as maltose/maltodextrin transporter could not be excluded. Thus we first studied a possible role of AcbH as a maltodextrin receptor by subjecting the cytoplasmic fraction of S. lividans TK23(pCB12) to amylose affinity chromatography. Immobilized amylose is recognized as substrate by solute receptors displaying maltodextrin binding activity [5,8]. However, binding of His10 –AcbH could not be detected by immunoblotting (not shown). In order to test other oligosaccharides, including acarbose and longer homologs, as possible substrates of His10 –AcbH we then performed binding studies by using surface plasmon resonance analysis. Freshly purified and refolded His10 –AcbH protein was coupled to a Ni-NTA-sensor chip and, subsequently, solutions of various substrates (10 mM) were allowed to flow over the immobilized protein. As shown in Fig. 4, the strongest binding signal was obtained with acarbose followed by homologs 5C and the mixture 6AB. In contrast no signals were observed with maltodextrins (Fig. 5). Maltose or sucrose were also ineffective (not shown). None of the substrates tested exhibited an interaction with the immo-
Fig. 5. Analysis of substrate specificity of His10 –AcbH by surface plasmon resonance. Interactions of His10 –AcbH with various sugars are shown. For details see Section 2. The sensorgrams represent the binding responses of His10 –AcbH in response units (RUs) as a function of time. The period during which the substrates were applied is indicated by arrows. A representative experiment is shown.
bilized His10 –MalK protein of S. typhimurium that served as a control for non-specific binding. Taking into account that 1000 RU correspond to 1 ng of protein and the molecular masses of AcbH (47.5 kDa) and acarbose (645 Da), one can calculate that acarbose bound to about 74% of immobilized AcbH molecules. This further indicates a successful refolding of the protein. By varying the sugar concentrations between 1 µM and 10 mM dissociation constants of 3.8, 1.2, and 1.8 mM were calculated for acarbose, 5C and 6AB, respectively. Together, we conclude that acarbose and longer derivatives but not maltodextrins are substrates of AcbH. The observation that the addition of 20 mM acarbose to the renaturation buffer increased the yield in refolded protein might be taken as further evidence in favor of this notion.
326
C. Brunkhorst et al. / Research in Microbiology 156 (2005) 322–327
4. Discussion The data presented in this communication strengthen the notion that the products of the acbHFG genes constitute an ABC import system for acarbose-like compounds. We have shown that (i) the product of the acbH gene is induced under the same conditions as other genes of the acb cluster, (ii) AcbH is attached to the cytoplasmic membrane, and (iii) a purified His10 –AcbH fusion protein binds acarbose and its longer derivatives 5C and 6AB but not maltodextrins. Together with the finding that His10 –AcbH failed to bind to an amylose resin, these data exclude a role in maltodextrin transport but are consistent with the view that AcbH is an extracellular, membrane-anchored solute receptor with specificity for acarbose-like compounds that likely delivers its substrates to an ABC import system comprising AcbFG and an as yet to be identified ATPase subunit. Thus, the substrate binding site of AcbH is apparently designed in such a way that it can discriminate maltooligosaccharides from pseudooligosaccharides. This is in contrast to maltose/maltodextrin binding proteins from E. coli and Alicyclobacillus acidocaldarius which also bind acarbose [2,8]. The observed dissociation constants for acarbose and derivatives are a hundredfold to a thousandfold higher than those typically measured for sugar binding proteins [1,8]. Thus, the values may not reflect the affinities existing under physiological conditions, possibly due to the refolding process. However, the actual concentration of acarbose in the natural environment is unknown, and therefore it is noteworthy that the Michaelis constant for acarbose of the extracellular acarviosyl transferase (AcbD) is in a similar range [6]. Clearly, it would be desirable to demonstrate a direct interaction of AcbH with the membrane-associated transport complex, e.g., by assaying AcbH-dependent stimulation of ATPase activity in proteoliposomes [13]. Such experiments are hampered by the lack of a gene within the acb cluster encoding a cognate ATPase component. However, preliminary results demonstrating that an ABC protein from Actinoplanes, displaying homology to MsiK of S. lividans [9], forms a complex with AcbFG (D. Elvers, G. Schäfer, C. Brunkhorst and E. Schneider, unpublished) are promising in this respect. Unfortunately, biochemical characterization of this complex is currently not feasible due to poor expression of acbFGmsiK in E. coli.
Acknowledgements We are grateful to Bayer AG for supporting this project in many respects, especially the collaboration with researchers from the Biochemical Development Division, and particularly A. Crueger and H. Wehlmann. We thank M. Dathe for recording the CD spectrum and B. Blüschke for assistance with surface plasmon resonance experiments. This work was
supported by the ‘Berliner Programm zur Förderung der Chancengleichheit für Frauen in Forschung und Lehre.’
References [1] W. Boos, J.M. Lucht, Periplasmic binding-protein-dependent ABCtransporters, in: F.C. Neidthard, R. Curtiss III, J.L. Ingraham, E.C.C. Lin, K.B. Low, B. Magasanik, W.S. Reznikoff, M. Riley, M. Schaechter, H.E. Umbarger (Eds.), Escherichia coli and Salmonella: Cellular and Molecular Biology, second ed., American Society for Microbiology, Washington, DC, 1996, pp. 1175–1209. [2] C. Brunkhorst, C. Andersen, E. Schneider, Acarbose, a pseudooligosaccharide, is transported but not metabolized by the maltose– maltodextrin system of Escherichia coli, J. Bacteriol. 181 (1999) 2612–2619. [3] E. De Bernardez Clark, Refolding of recombinant proteins, Curr. Opin. Biotechnol. 9 (1998) 157–163. [4] M. Doumith, P. Weingarten, U.F. Wehmeier, K. Salah-Bey, B. Benhamou, C. Capdevilla, J.-M. Michel, W. Piepersberg, M.-C. Raynal, Analysis of genes involved in 6-deoxyhexose biosynthesis and transfer in Saccharopolyspora erythraea, Mol. Gen. Genet. 264 (2000) 477– 485. [5] T. Ferenci, U. Klotz, Affinity chromatography of the periplasmic maltose binding protein of Escherichia coli, FEBS Lett. 94 (1978) 213– 217. [6] M. Hemker, A. Stratmann, K. Goeke, W. Schroder, J. Lenz, W. Piepersberg, H. Pape, Identification, cloning, expression, and characterization of the extracellular acarbose-modifying glycosyltransferase, AcbD, from Actinoplanes sp. strain SE50, J. Bacteriol. 183 (2001) 4484–4492. [7] D.A. Hopwood, Genetic manipulation of Streptomyces: A Laboratory Manual, The John Innes Foundation, Norwich, England, 1995. [8] A. Hülsmann, R. Lurz, F. Scheffel, E. Schneider, Maltose and maltodextrin transport in the thermoacidophilic Gram-positive bacterium Alicyclobacillus acidocaldarius is mediated by a high-affinity transport system that includes a maltose-binding protein tolerant to low pH, J. Bacteriol. 182 (2000) 6292–6301. [9] Y. Hurtubise, F. Shareck, D. Kluepfel, R. Morosoli, A cellulase/ xylanase-negative mutant of Streptomyces lividans 1326 defective in cellobiose and xylobiose uptake is mutated in a gene encoding a protein homologous to ATP-binding proteins, Mol. Microbiol. 17 (1995) 367–377. [10] A.P.J. Middelberg, Preparative protein folding, Trends Biotechnol. 20 (2002) 437–443. [11] L. Müller, Chemistry, biochemistry and therapeutic potential of microbial α-glucosidase inhibitors, in: A.L. Demain, G.A. Somkuti, J.C. Hunter-Creva, H.W. Rossmoore (Eds.), Novel Microbial Products for Medicine and Agriculture, Elsevier, Amsterdam, 1989, pp. 109– 116. [12] A. Schlösser, T. Kampers, H. Schrempf, The Streptomyces ATPbinding component MsiK assists in cellobiose and maltose transport, J. Bacteriol. 179 (1997) 2092–2095. [13] E. Schneider, ABC transporters catalyzing carbohydrate uptake, Res. Microbiol. 152 (2001) 303–310. [14] E. Schneider, C. Walter, A chimeric nucleotide-binding protein, encoded by a hisPmalK hybrid gene, is functional in maltose transport in Salmonella typhimurium, Mol. Microbiol. 5 (1991) 1375–1383. [15] A. Stratmann, T. Mahmud, S. Lee, J. Distler, H.G. Floss, W. Piepersberg, The AcbC protein from Actinoplanes species is a C7-cyclitol synthase related to 3-dehydroquinate synthases and is involved in the biosynthesis of the alpha-glucosidase inhibitor acarbose, J. Biol. Chem. 274 (1999) 10889–10896. [16] I.C. Sutcliffe, R.B. Russel, Lipoproteins of Gram-positive bacteria, J. Bacteriol. 177 (1995) 1123–1128.
C. Brunkhorst et al. / Research in Microbiology 156 (2005) 322–327
[17] E. Truscheit, W. Frommer, B. Junge, L. Müller, D. Schmidt, W. Wingender, Chemistry and biochemistry of α-glucosidase inhibitors, Angew. Chem. Int. Ed. 20 (1981) 744–761. [18] U.F. Wehmeier, W. Piepersberg, Biotechnology and molecular biology of the alpha-glucosidase inhibitor acarbose, Appl. Microbiol. Biotechnol. 63 (2004) 613–625.
327
[19] C.S. Zhang, A. Stratmann, O. Block, R. Bruckner, M. Podeschwa, H.J. Altenbach, U.F. Wehmeier, W. Piepersberg, Biosynthesis of the C(7)-cyclitol moiety of acarbose in Actinoplanes species SE50/110. 7-O-phosphorylation of the initial cyclitol precursor leads to proposal of a new biosynthetic pathway, J. Biol. Chem. 277 (2002) 22853– 22862.
The acbH gene of Actinoplanes sp. encodes a solute receptor with binding activities for acarbose and longer homologs Claudia Brunkhorst a , Udo F. Wehmeier b , Wolfgang Piepersberg b , Erwin Schneider a,∗ a Humboldt Universität zu Berlin, Institut für Biologie/Bakterienphysiologie, Chausseestr. 117, 10115 Berlin, Germany b Bergische Universität Wuppertal, Chemische Mikrobiologie, Gauß-Str. 20, 42097 Wuppertal, Germany
Received 7 September 2004; accepted 25 October 2004 Available online 22 December 2004
Abstract Acarbose, a pseudomaltotetraose, is produced by strains of the genus Actinoplanes and is a potent inhibitor of α-glucosidases, including those from the human intestine. Therefore, it is used in the treatment of patients suffering from type 2 diabetes. The benefits of acarbose for the producer are not known; however, besides acting as an inhibitor of α-amylases secreted by competitors, a role as a ‘carbophor’ has been proposed. This would require a transport system mediating its uptake into the cytoplasm of Actinoplanes sp. A putative sugar ATP binding cassette (ABC) transport system, the genes of which are included within the biosynthetic gene cluster for acarbose, was suggested to be a possible candidate. The genes acbHFG encode a possible sugar binding protein (AcbH) and two membrane integral subunits (AcbFG). A gene coding for an ATPase component is missing. Since Actinoplanes sp. cannot yet be genetically manipulated we performed experiments to identify the substrate(s) of the putative transporter by assessing the substrate specificity of AcbH. The protein was overproduced in Escherichia coli as His10 -fusion protein, purified under denaturating conditions and renatured. Refolding was verified by circular dichroism spectroscopy. Surface plasmon resonance studies revealed that AcbH binds acarbose and longer derivatives, but not maltodextrins, maltose or sucrose. Immunoblot analysis revealed the association of AcbH with the membrane fraction of Actinoplanes cells that were grown in the presence of maltose, maltodextrins or acarbose. Together, these findings suggest that the AcbHFG complex might be involved in the uptake of acarbose and are consistent with a role for acarbose as a ‘carbophor’. 2004 Elsevier SAS. All rights reserved. Keywords: Acarbose; ABC transport; Solute receptor; Actinoplanes sp.
1. Introduction The α-glucosidase inhibitor acarbose (Fig. 1), a pseudomaltotetraose produced by strains of the genera Actinoplanes and Streptomyces, is a member of an unusual group of bacterial secondary metabolites, all of which inhibit various α-glucosidases, especially from the intestine [11,17]. Acarbose is produced industrially in media containing maltose and maltotriose using developed strains of Actinoplanes sp. SE50. It is used in the treatment of patients suffering from type 2 diabetes mellitus. Besides acarbose, the organism produces an extensive number of homologs dependent on the * Corresponding author.
E-mail address: [email protected] (E. Schneider). 0923-2508/$ – see front matter 2004 Elsevier SAS. All rights reserved. doi:10.1016/j.resmic.2004.10.016
carbon source that differ in the number and chemical nature of sugars added to the core unit and in their inhibitory potential. The possible benefits that the bacteria derive from the production of acarbose and homologs in their natural environment remain unclear. As a potent inhibitor of α-amylases [18] and maltose/maltodextrin transport systems [2,8] of other microorganisms, acarbose can be envisaged as a means to restrict access of competitors to carbon sources like starch and other α-1,4-glucans. This notion is further supported by the observation that Actinoplanes sp. contains at least one maltose/maltodextrin transporter that is insensitive to acarbose (Brunkhorst and Schneider, manuscript in preparation). Recently, a cluster comprising 25 genes (acb) was identified in Actinoplanes sp. SE50/110 the products of which
C. Brunkhorst et al. / Research in Microbiology 156 (2005) 322–327
323
2. Materials and methods 2.1. Chemicals
Fig. 1. Structure of acarbose. Acarbose is a pseudo-oligosaccharide that consists of an unsaturated cyclitol moiety (A), an aminodeoxyhexose (B) (together also called acarviosine), and a normal maltose (rings C and D).
Acarbose (acarviosyl-1,4-maltose), component 5C (acarviosyl-1,4-maltose-1,4-glucose-1,1-glucose, component mixture 6AB (acarviosyl-1,4-maltose-1,4-maltose-1,4-fructose and acarviosyl-1,4-maltose-1,4-maltose-1,4-glucose) and MD50 (mixture of maltose and maltotriose, containing trace amounts of longer dextrins) were kind gifts from H. Wehlmann (Bayer AG, Wuppertal, Germany). Maltodextrins were purchased from Pfanstiehl Laboratories (IL, USA). 2.2. Bacterial strains, media and growth conditions
are involved in the biosynthesis of acarbose [18]. Biochemical characterization of several of the gene products has subsequently led to the proposal of a biosynthetic pathway [6,15,18,19] and of an additional function for acarbose [18]. Accordingly, the biosynthetic reactions in the cytoplasm result in either acarbose-7-phosphate or a precursor, e.g., the monoglucosylated acarviosine-7-phosphate that is secreted into the medium, possibly by an ABC exporter (AcbWXY). Here, acarbose and/or derivatives are formed by action of an acarviosyltransferase, AcbD, that catalyzes the transfer of the acarviosyl moiety (Fig. 1) to maltooligosaccharides and dextrins [6]. Re-uptake of acarbose and derivatives would then result in a net gain of carbon and energy. Thus, acarbose or its core subunit might function as a ‘carbophor’ that is secreted and charged with additional glucose moieties thereby extracting glucose from the extracellular pool and simultaneously tagging it for import. Consistent with this hypothesis is the finding that Actinoplanes sp. can utilize acarbose as sole source of carbon and energy (Brunkhorst and Schneider, manuscript in preparation). Within the acb cluster a putative operon (acbHFG) encoding components of an ATP-binding cassette (ABC) import system was found that consists of an extracellular solute binding protein (AcbH) and two membrane-spanning subunits (AcbFG). As often observed for sugar ABC transport systems of Gram-positive bacteria, a gene translating into an ATPase component is lacking [8,12]. BLAST searches revealed homology to members of the CUT 1 (carbohydrate uptake transporter) subfamily [13] which makes the transporter a candidate for an acarbose uptake system. Unfortunately, Actinoplanes sp. cannot yet be genetically manipulated. Thus, elucidating gene functions requires the biochemical characterization of their products. Since the solute receptors of ABC import systems are the major determinants of substrate specificity [1], we have focused on AcbH to elucidate the role of the putative AcbHFG transporter. We report on the overproduction in Escherichia coli, purification and characterization of AcbH. Our data are consistent with the notion that the AcbHFG proteins are involved in the transport of acarbose and derivatives in the context of the carbophor model.
E. coli strain BL21(DE3)(pLysS) (Stratagene, Heidelberg, Germany) harboring plasmid pCB10 (expressing acbH under control of the pT7lac promoter) was grown in LB medium supplemented with ampicillin (100 µg/ml) and chloramphenicol (20 µg/ml), at 30 ◦ C. At OD650 = 1, 0.1 mM isopropyl thio-β-galactoside (IPTG) was added and growth continued for 3 h. Streptomyces lividans TK23 harboring plasmid pCB12 (expressing acbH under control of the ermEp promotor) was grown in trypton soybean broth (TSB, Oxoid), supplemented with thiostrepton (25 µg/ml) for 48 h at 30 ◦ C [19]. Actinoplanes sp. SE50/110 (ATCC 31044) was grown for 72 h at 30 ◦ C in TSB or in minimal salt medium (containing 5.7 g Tris–HCl, 1 g NaCl, 2 g (NH4 )2 SO4 , 0.5 g K2 SO4 , 0.2 g MgSO4 , 0.1 g CaCl2 , 0.34 g KH2 PO4 per liter of deionized water, adjusted to pH 7.2 with H2 SO4 (20%, v/v); 10 ml of a trace element solution [7] were added after autoclaving) supplemented with the indicated carbon sources (0.5%, w/v). 2.3. Cloning and subcloning of acbH The acbH gene lacking the signal sequence was amplified by PCR from genomic DNA of Actinoplanes sp. SE50/110 using the oligonucleotide primers acbH12 (5 -C GAG CGG CAT ATG GGC AGT GAC GAC AAG AGC GGG ACG-3 ) and acbH13 (3 -CCC GCC GGA TCC CCG GTC GAA TGG TCG GGG TCA TCA-5 ). The primers were designed such that the resulting fragment could be ligated with plasmid pET19b (Novagene, Madison, WI, USA) via NdeI and BamHI restriction sites, yielding plasmid pCB10. As a consequence the translated polypeptide has 10 consecutive histidine residues fused to the N-terminal amino acid of mature AcbH which was changed from cysteine to methionine. Plasmid pCB12 is a derivative of pCB10 that carries his10 –acbH on the E. coli/Streptomycetes shuttle vector pPWW49 [4]. 2.4. Purification and renaturation of His10 –AcbH Cells of strain BL21(DE3)(pCB10, pLysS) were harvested, resuspended in buffer 1 (20 mM Tris–HCl, pH 7.2,
324
C. Brunkhorst et al. / Research in Microbiology 156 (2005) 322–327
containing 5% (v/v) glycerol, 300 mM NaCl, 20 mM 2-mercaptoethanol, 0.1 mM phenylmethylsulfonyl fluoride; 1/10 of the culture volume), and disrupted by passage through a French press. After a low speed spin, the pellet containing His10 –AcbH was dissolved in buffer 1 containing 8 M urea (buffer 2) at 5 mg/ml. After gentle stirring for 30 min at room temperature insoluble material was removed by ultracentrifugation and the supernatant was mixed with Ni-NTA matrix equilibrated in buffer 2. The resin was washed with buffer 2 containing 20 mM imidazole followed by elution of His10 –AcbH with buffer 2 containing 150 mM imidazole. The peak fraction was adjusted to 1 M L-arginine and passed through a desalting column (PD10, AmershamPharmacia) to remove urea [3,10]. The eluent was subsequently passed through a second PD10 column, equilibrated with 20 mM Tris–HCl, pH 7.2, 5% glycerol, 300 mM NaCl, and the resulting protein containing fractions were ultracentrifuged to remove insoluble material. 2.5. Surface plasmon resonance analysis Binding of sugars to AcbH was detected by surface plasmon resonance analysis using a Biacore X optical biosensor (Biacore Uppsala, Sweden). Freshly renatured His10 –AcbH (0.273 pmol) was immobilized on a NTA sensor chip with an efficiency of 13 000 resonance units. The sugars tested were applied at a flow rate of 10 µl/min. Dissociation from and association of the sugars to the immobilized protein were monitored for 2 min each. Purified His10 –MalK (0.15 pmol) of Salmonella typhimurium (gift of B. Blüschke, HU Berlin) was used as reference. Dissociation constants were calculated using the manufacturer’s software package (fitted according to Langmuir).
Fig. 2. Steps in the purification and renaturation of His10 –AcbH from inclusion bodies. His10 –AcbH was purified by Ni-NTA chromatography in the presence of 8 M urea and subsequently renatured in the presence of L-arginine as described in Section 2. Lanes: 1, low speed pellet fraction (uninduced); 2, molecular mass standards (KDa); 3, low speed pellet fraction (induced); 4, cytosolic fraction; 5, supernatant after treatment of the low speed pellet fraction with 8 M urea and subsequent ultracentrifugation; 6, eluate fraction from Ni-NTA chromatography (150 mM imidazole); 7, renatured His10 –AcbH after ultracentrifugation (2.3 µg).
3. Results
When cells of E. coli strain BL21(DE3)(pLysS) harboring pCB10 were grown in rich medium and induced for gene expression, a protein of the expected molecular size of about 47 kDa was found in substantial amounts in whole cell extracts. After cell fractionation, the recombinant protein was recovered almost exclusively with the pellet obtained after a low speed centrifugation step. Attempts to increase the amount of soluble AcbH failed, including variations in medium composition, growth temperature or co-expression of E. coli chaperones or tRNAs for translation of GC rich mRNAs. Since overproduction of other gene products from the acb cluster was successfully achieved in S. lividans [6], a similar experiment was performed with acbH. Although the protein could be detected by immunoblotting in the cytoplasmic fraction of strain TK23(pCB12) the yield was insufficient for further biochemical characterization. Thus, His10 – AcbH was purified from the insoluble fraction of E. coli BL21(DE3)(pCB10, pLysS) under denaturing conditions in the presence of 8 M urea by Ni-NTA affinity chromatography and subsequently renatured as described in Section 2 (Fig. 2). Refolding of His10 –AcbH was successful, as judged by circular dichroism spectroscopy which indicated a substantial portion of α-helical secondary structure (Fig. 3). Routinely, 14 mg of purified refolded protein were obtained from 1-l culture.
3.1. Purification of an N-terminal His10 –AcbH fusion protein
3.2. Synthesis of AcbH in Actinoplanes sp. in the presence of different sugars
In order to explore the possible function of the putative extracellular solute receptor, AcbH, the acbH gene lacking the signal peptide-encoding sequence was amplified from genomic DNA of Actinoplanes sp. SE50/110 by PCR and subcloned in expression vector pET19b, yielding pCB10.
The purified protein was used to raise a polyclonal antiserum which then allowed us to study the carbon sourcedependent synthesis of AcbH in Actinoplanes sp. SE50/110. As shown in Fig. 4, AcbH was detected after 72 h in the membrane fraction of cells that were grown in minimal
2.6. Miscellaneous methods SDS–polyacrylamide gel electrophoresis (SDS–PAGE) and immunoblotting were performed as in [14]. The antiserum raised against purified AcbH was obtained from Pineda antibody service (Berlin). Protein was determined with the BCA-protein assay kit from Pierce (Germmany). The CD spectrum was recorded with a Jasco 720 (UK) in 50 mM Tris–HCl, pH 7.2, 5%, v/v) glycerol, 150 mM NaCl, at a protein concentration of 0.15 mg/ml.
C. Brunkhorst et al. / Research in Microbiology 156 (2005) 322–327
325
Fig. 3. Synthesis of acbH in Actinoplanes sp. in the presence of different sugars. Cells were grown in minimal medium supplemented with the indicated carbon sources or in TSB alone for 72 h at 30 ◦ C. Aliquots from membrane fractions (30 µg protein) of each culture were subjected to SDS–PAGE followed by immunoblotting. The blot was probed with an anti-AcbH antiserum. Control: 0.1 µg purified His10 –AcbH.
medium supplemented with either MD50, maltose or acarbose. Some AcbH was also observed in TSB. In contrast, AcbH was absent in cells grown in the presence of glucose or glycerol. Association of AcbH with the membrane was further confirmed by the observation that treatment with the non-ionic detergent Triton X-100 resulted in solubilization of the protein. Together, these data are consistent with a role of AcbH as solute receptor which in Gram-positive bacteria is anchored to the cytoplasmic membrane by an N-terminal lipid modification [16]. Moreover, AcbH is synthesized under conditions that were also shown to induce expression of other acb genes ([6]; Y. Rockser and U.F. Wehmeier, unpublished).
Fig. 4. Far-UV-CD spectrum of purified, renatured AcbH. The spectrum was recorded as described in Section 2. The minima in ellipticity ([]) at 222 and 208 nm are characteristic of α-helical content.
3.3. Determination of substrate specificity of His10 –AcbH Homology searches indicated that AcbH is a component of an ABC import system belonging to the CUT 1 subfamily [13], the members of which predominantly transport di- and oligosaccharides. Since maltose and maltotriose were found to induce AcbH synthesis, a function of the acbHFG gene products as maltose/maltodextrin transporter could not be excluded. Thus we first studied a possible role of AcbH as a maltodextrin receptor by subjecting the cytoplasmic fraction of S. lividans TK23(pCB12) to amylose affinity chromatography. Immobilized amylose is recognized as substrate by solute receptors displaying maltodextrin binding activity [5,8]. However, binding of His10 –AcbH could not be detected by immunoblotting (not shown). In order to test other oligosaccharides, including acarbose and longer homologs, as possible substrates of His10 –AcbH we then performed binding studies by using surface plasmon resonance analysis. Freshly purified and refolded His10 –AcbH protein was coupled to a Ni-NTA-sensor chip and, subsequently, solutions of various substrates (10 mM) were allowed to flow over the immobilized protein. As shown in Fig. 4, the strongest binding signal was obtained with acarbose followed by homologs 5C and the mixture 6AB. In contrast no signals were observed with maltodextrins (Fig. 5). Maltose or sucrose were also ineffective (not shown). None of the substrates tested exhibited an interaction with the immo-
Fig. 5. Analysis of substrate specificity of His10 –AcbH by surface plasmon resonance. Interactions of His10 –AcbH with various sugars are shown. For details see Section 2. The sensorgrams represent the binding responses of His10 –AcbH in response units (RUs) as a function of time. The period during which the substrates were applied is indicated by arrows. A representative experiment is shown.
bilized His10 –MalK protein of S. typhimurium that served as a control for non-specific binding. Taking into account that 1000 RU correspond to 1 ng of protein and the molecular masses of AcbH (47.5 kDa) and acarbose (645 Da), one can calculate that acarbose bound to about 74% of immobilized AcbH molecules. This further indicates a successful refolding of the protein. By varying the sugar concentrations between 1 µM and 10 mM dissociation constants of 3.8, 1.2, and 1.8 mM were calculated for acarbose, 5C and 6AB, respectively. Together, we conclude that acarbose and longer derivatives but not maltodextrins are substrates of AcbH. The observation that the addition of 20 mM acarbose to the renaturation buffer increased the yield in refolded protein might be taken as further evidence in favor of this notion.
326
C. Brunkhorst et al. / Research in Microbiology 156 (2005) 322–327
4. Discussion The data presented in this communication strengthen the notion that the products of the acbHFG genes constitute an ABC import system for acarbose-like compounds. We have shown that (i) the product of the acbH gene is induced under the same conditions as other genes of the acb cluster, (ii) AcbH is attached to the cytoplasmic membrane, and (iii) a purified His10 –AcbH fusion protein binds acarbose and its longer derivatives 5C and 6AB but not maltodextrins. Together with the finding that His10 –AcbH failed to bind to an amylose resin, these data exclude a role in maltodextrin transport but are consistent with the view that AcbH is an extracellular, membrane-anchored solute receptor with specificity for acarbose-like compounds that likely delivers its substrates to an ABC import system comprising AcbFG and an as yet to be identified ATPase subunit. Thus, the substrate binding site of AcbH is apparently designed in such a way that it can discriminate maltooligosaccharides from pseudooligosaccharides. This is in contrast to maltose/maltodextrin binding proteins from E. coli and Alicyclobacillus acidocaldarius which also bind acarbose [2,8]. The observed dissociation constants for acarbose and derivatives are a hundredfold to a thousandfold higher than those typically measured for sugar binding proteins [1,8]. Thus, the values may not reflect the affinities existing under physiological conditions, possibly due to the refolding process. However, the actual concentration of acarbose in the natural environment is unknown, and therefore it is noteworthy that the Michaelis constant for acarbose of the extracellular acarviosyl transferase (AcbD) is in a similar range [6]. Clearly, it would be desirable to demonstrate a direct interaction of AcbH with the membrane-associated transport complex, e.g., by assaying AcbH-dependent stimulation of ATPase activity in proteoliposomes [13]. Such experiments are hampered by the lack of a gene within the acb cluster encoding a cognate ATPase component. However, preliminary results demonstrating that an ABC protein from Actinoplanes, displaying homology to MsiK of S. lividans [9], forms a complex with AcbFG (D. Elvers, G. Schäfer, C. Brunkhorst and E. Schneider, unpublished) are promising in this respect. Unfortunately, biochemical characterization of this complex is currently not feasible due to poor expression of acbFGmsiK in E. coli.
Acknowledgements We are grateful to Bayer AG for supporting this project in many respects, especially the collaboration with researchers from the Biochemical Development Division, and particularly A. Crueger and H. Wehlmann. We thank M. Dathe for recording the CD spectrum and B. Blüschke for assistance with surface plasmon resonance experiments. This work was
supported by the ‘Berliner Programm zur Förderung der Chancengleichheit für Frauen in Forschung und Lehre.’
References [1] W. Boos, J.M. Lucht, Periplasmic binding-protein-dependent ABCtransporters, in: F.C. Neidthard, R. Curtiss III, J.L. Ingraham, E.C.C. Lin, K.B. Low, B. Magasanik, W.S. Reznikoff, M. Riley, M. Schaechter, H.E. Umbarger (Eds.), Escherichia coli and Salmonella: Cellular and Molecular Biology, second ed., American Society for Microbiology, Washington, DC, 1996, pp. 1175–1209. [2] C. Brunkhorst, C. Andersen, E. Schneider, Acarbose, a pseudooligosaccharide, is transported but not metabolized by the maltose– maltodextrin system of Escherichia coli, J. Bacteriol. 181 (1999) 2612–2619. [3] E. De Bernardez Clark, Refolding of recombinant proteins, Curr. Opin. Biotechnol. 9 (1998) 157–163. [4] M. Doumith, P. Weingarten, U.F. Wehmeier, K. Salah-Bey, B. Benhamou, C. Capdevilla, J.-M. Michel, W. Piepersberg, M.-C. Raynal, Analysis of genes involved in 6-deoxyhexose biosynthesis and transfer in Saccharopolyspora erythraea, Mol. Gen. Genet. 264 (2000) 477– 485. [5] T. Ferenci, U. Klotz, Affinity chromatography of the periplasmic maltose binding protein of Escherichia coli, FEBS Lett. 94 (1978) 213– 217. [6] M. Hemker, A. Stratmann, K. Goeke, W. Schroder, J. Lenz, W. Piepersberg, H. Pape, Identification, cloning, expression, and characterization of the extracellular acarbose-modifying glycosyltransferase, AcbD, from Actinoplanes sp. strain SE50, J. Bacteriol. 183 (2001) 4484–4492. [7] D.A. Hopwood, Genetic manipulation of Streptomyces: A Laboratory Manual, The John Innes Foundation, Norwich, England, 1995. [8] A. Hülsmann, R. Lurz, F. Scheffel, E. Schneider, Maltose and maltodextrin transport in the thermoacidophilic Gram-positive bacterium Alicyclobacillus acidocaldarius is mediated by a high-affinity transport system that includes a maltose-binding protein tolerant to low pH, J. Bacteriol. 182 (2000) 6292–6301. [9] Y. Hurtubise, F. Shareck, D. Kluepfel, R. Morosoli, A cellulase/ xylanase-negative mutant of Streptomyces lividans 1326 defective in cellobiose and xylobiose uptake is mutated in a gene encoding a protein homologous to ATP-binding proteins, Mol. Microbiol. 17 (1995) 367–377. [10] A.P.J. Middelberg, Preparative protein folding, Trends Biotechnol. 20 (2002) 437–443. [11] L. Müller, Chemistry, biochemistry and therapeutic potential of microbial α-glucosidase inhibitors, in: A.L. Demain, G.A. Somkuti, J.C. Hunter-Creva, H.W. Rossmoore (Eds.), Novel Microbial Products for Medicine and Agriculture, Elsevier, Amsterdam, 1989, pp. 109– 116. [12] A. Schlösser, T. Kampers, H. Schrempf, The Streptomyces ATPbinding component MsiK assists in cellobiose and maltose transport, J. Bacteriol. 179 (1997) 2092–2095. [13] E. Schneider, ABC transporters catalyzing carbohydrate uptake, Res. Microbiol. 152 (2001) 303–310. [14] E. Schneider, C. Walter, A chimeric nucleotide-binding protein, encoded by a hisPmalK hybrid gene, is functional in maltose transport in Salmonella typhimurium, Mol. Microbiol. 5 (1991) 1375–1383. [15] A. Stratmann, T. Mahmud, S. Lee, J. Distler, H.G. Floss, W. Piepersberg, The AcbC protein from Actinoplanes species is a C7-cyclitol synthase related to 3-dehydroquinate synthases and is involved in the biosynthesis of the alpha-glucosidase inhibitor acarbose, J. Biol. Chem. 274 (1999) 10889–10896. [16] I.C. Sutcliffe, R.B. Russel, Lipoproteins of Gram-positive bacteria, J. Bacteriol. 177 (1995) 1123–1128.
C. Brunkhorst et al. / Research in Microbiology 156 (2005) 322–327
[17] E. Truscheit, W. Frommer, B. Junge, L. Müller, D. Schmidt, W. Wingender, Chemistry and biochemistry of α-glucosidase inhibitors, Angew. Chem. Int. Ed. 20 (1981) 744–761. [18] U.F. Wehmeier, W. Piepersberg, Biotechnology and molecular biology of the alpha-glucosidase inhibitor acarbose, Appl. Microbiol. Biotechnol. 63 (2004) 613–625.
327
[19] C.S. Zhang, A. Stratmann, O. Block, R. Bruckner, M. Podeschwa, H.J. Altenbach, U.F. Wehmeier, W. Piepersberg, Biosynthesis of the C(7)-cyclitol moiety of acarbose in Actinoplanes species SE50/110. 7-O-phosphorylation of the initial cyclitol precursor leads to proposal of a new biosynthetic pathway, J. Biol. Chem. 277 (2002) 22853– 22862.