maltose and maltodextrin utilization by bacillus subtilis · (11, 34, 35, 44). in contrast to the...

JOURNAL OF BACTERIOLOGY, June 2006, p. 3911–3922 Vol. 188, No. 110021-9193/06/$08.00�0 doi:10.1128/JB.00213-06Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Maltose and Maltodextrin Utilization by Bacillus subtilisStefan Schonert,* Sabine Seitz, Holger Krafft, Eva-Anne Feuerbaum,† Iris Andernach,

Gabriele Witz, and Michael K. DahlDepartment of Biology, University of Konstanz, D-78457 Konstanz, Germany

Received 8 February 2006/Accepted 15 March 2006

Bacillus subtilis can utilize maltose and maltodextrins that are derived from polysaccharides, like starch orglycogen. In this work, we show that maltose is taken up by a member of the phosphoenolpyruvate-dependentphosphotransferase system and maltodextrins are taken up by a maltodextrin-specific ABC transporter.Uptake of maltose by the phosphoenolpyruvate-dependent phosphotransferase system is mediated by maltose-specific enzyme IICB (MalP; synonym, GlvC), with an apparent Km of 5 �M and a Vmax of 91 nmol · min�1 ·(1010 CFU)�1. The maltodextrin-specific ABC transporter is composed of the maltodextrin binding proteinMdxE (formerly YvdG), with affinities in the low micromolar range for maltodextrins, and the membrane-spanning components MdxF and MdxG (formerly YvdH and YvdI, respectively), as well as the energizingATPase MsmX. Maltotriose transport occurs with an apparent Km of 1.4 �M and a Vmax of 4.7 nmol · min�1 ·(1010 CFU)�1.

The gram-positive soil bacterium Bacillus subtilis can utilizeglycogen, starch, and amylose as carbon sources. Prior to trans-port through the cell membrane, these polysaccharides arehydrolyzed by the extracellular �-amylase AmyE into smallermaltodextrins (15). The resulting disaccharides, maltose andisomaltose, as well as maltodextrins, are secondary metabolitesthat can serve as sole carbon and energy sources in B. subtilis(11, 34, 35, 44).

In contrast to the very well-known maltose and maltodextrinuptake of the Escherichia coli maltose ABC transporter (re-viewed in reference 3), the uptake of these carbohydrates in B.subtilis is not very well understood. Recently, uptake of malt-ose via the phosphoenolpyruvate-dependent phosphotransfer-ase system (PTS) was postulated based on the identificationof an NAD(H)-dependent phospho-�-1,4-glucosidase (MalA;synonym, GlvA) (46). In general, substrates that are taken upinto the cell by the PTS become phosphorylated when theyenter the cell. During this process, the required phosphorylgroup is transferred from phosphoenolpyruvate via the generalcytoplasmic proteins enzyme I and HPr to a substrate-specificmembrane bound enzyme II and finally to the transportedsubstrate (for detailed information, see references 26 and 32).

The malA gene is located in an operon composed of threegenes (Fig. 1): (i) malA (synonym, glvA) (46); (ii) glvR (syn-onym, yfiA), encoding the potential activator of the operon(47); and (iii) malP (synonym, glvC), encoding the putativeenzyme IICB (EIICBMal) specific for maltose (29). Inactiva-tion of the putative EIICBMal resulted in a sevenfold-longergeneration time on maltose minimal medium than that of thewild type (29).

However, these results contradict previous reports that con-

cluded that uncouplers negatively affect maltose uptake in B.subtilis (44). This observation led to the conclusion that maltosetransport in B. subtilis is proton motive force dependent and doesnot occur via the PTS, but is regulated by the PTS (44).

In addition, several genes encoding ATP binding cassette(ABC) transporters can be found on the B. subtilis chromo-some, which might encode maltose and/or maltodextrin uptakesystems (41). ABC transporters in general consist of four do-mains. Two of these domains are located in the cytoplasmicmembrane and form a canal. The other two domains areATPases that energize the transport of the substrate throughthe canal. In bacteria, ABC importers can be distinguishedfrom exporters in that importers possess a high specific sub-strate binding protein that delivers the substrate to the trans-membrane domains. These substrate binding proteins are sol-uble in the periplasm of gram-negative bacteria (2) and areanchored to the membrane by lipid modifications in gram-positive bacteria (42). ABC transporters do not phosphorylateor otherwise modify their substrates during transport (8, 14).

In addition to the potential maltose system mentionedabove, B. subtilis contains a maltose-inducible �-glucosidaseactivity that is associated with MalL (34). MalL activity is alsoinduced by exogenous amylose, starch, and glycogen (35). ThemalL gene is located in a gene cluster consisting of nine genes(41) (Fig. 1). The last gene of this cluster, pgcM, encodes a�-phosphoglucomutase/glucose-1-phosphosphate phosphodis-mutase acting on phosphorylated glucose molecules presum-ably resulting from the degradation of glucose oligomers (30).Based on similarities of the amino acid sequences deducedfrom the open reading frames, the other genes encode a tran-scriptional regulator (yvdE), a substrate binding protein(mdxE; formerly yvdG), and membrane-spanning components(mdxF and mdxG; formerly yvdH and yvdI, respectively) of anABC transporter, a cytoplasmic maltogenic amylase or neopul-lulanase (yvdF) (5), and a maltose phosphorylase (yvdK), re-spectively (41). No prediction of the function of YvdJ could bemade (41). Since uptake and utilization systems in bacteria areoften encoded by genes located in the same operon, the region

* Corresponding author. Mailing address: Lehrstuhl fur Mikrobiologie,Fachbereich Biologie der Universitat Konstanz, Universitatsstrasse 10,M605, D-78457 Konstanz, Germany. Phone: (49-7531) 882041. Fax: (49-7531) 883356. E-mail: [email protected].

† Present address: Lehrstuhl fur Mikrobiologie, Fachbereich Biologie/Chemie der Universitat Osnabruck, D-49069 Osnabruck, Germany.

3911

on April 1, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

around malL is a prime candidate to encode a maltose and/ormaltodextrin utilization system. However, based on computer-aided analysis, other potential ABC transporters can be foundin B. subtilis that might encode potential maltose and malto-dextrin uptake systems (28, 41). As in 11 of the 78 ABC trans-porter-encoding gene clusters of B. subtilis, the ABC-encodingopen reading frame is missing in the above-described genecluster (28, 41). However, the B. subtilis genome encodes threepotential ATPases presumably involved in substrate import,namely, ylmA, yusV, and msmX, which are probably organizedin monocystronic operons (28, 41). This raises the possibilitythat one ATP binding cassette serves different kinds of ABCtransporters (28).

The aim of this study was to answer the question of howmaltose or maltodextrins are taken up by B. subtilis. We showthat in contrast to the gram-negative bacterium Escherichiacoli, maltose is taken up by the maltose-specific enzyme IICB(MalP) of the PTS in B. subtilis and that maltotriose andpresumably maltodextrins up to at least maltoheptaose are

taken up by a specific ABC transporter. The latter consists ofthe maltodextrin-binding protein MdxE, with high affinities formaltodextrins and a low affinity for maltose, as well as themembrane components MdxF and MdxG. Furthermore, ourdata show that transport via this transporter is energized byMsmX as the cognate ABC domain.

MATERIALS AND METHODS

Reagents and enzymes. Restriction enzymes, Taq DNA polymerase, and T4ligase were used as recommended by the manufacturers. Maltose, maltotriose,maltotetraose, maltopentaose, maltohexaose, maltoheptaose, para-nitrophenyl-�-D-glucopyranoside, and all other sugars used were purchased from Sigma(Munich, Germany). [14C]maltose (610 mCi/mmol; 200 �Ci/ml) was obtainedfrom Amersham, Braunschweig, Germany; [14C]maltotriose (800 mCi/mmol; 100�Ci/ml) was from Biotrend Chemikalien GmbH, Koln, Germany. [14C]-malto-triose was contaminated with a significant amount of a radioactive substrate thatshowed the same behavior as maltose on thin-Layer chromatography (data notshown). All other reagents were of analytical grade.

Bacterial strains, plasmids, media, and selection of recombinants. The plas-mids and strains used in this study are listed in Table 1 and Table 2, respectively.Standard procedures were used to transform E. coli, extract plasmids, and ma-

FIG. 1. The mal operon and the yvdE-pgcM region of B. subtilis. The numbers indicate the locations of the first bases of malA and yvdE andthe last bases of malP and pgcM on the B. subtilis chromosome. Putative terminators are shown as stem loops. Encoded proteins and predictedproteins, based on similarities of the derived amino acid sequences, are indicated below. ?, no similarities to proteins with known functions.

TABLE 1. Plasmids used in this work

Plasmid Descriptiona Source orreference

pAC6 Integration vector carrying amyE::cat 40pBlueHI pBluescript II SK(�) derivative containing the mdxF-mdxG region of B. subtilis This workpBlueHIK1 pBlueHI derivative containing a mdxF-mdxG region disruption by the aphA3 gene in mdxF-mdxG direction This workpBlueHIK2 pMBP derivative containing a mdxF-mdxG region disruption by the aphA3 gene in opposite direction to

the mdxF-mdxG regionThis work

pBlueMBPkurz pMBP derivative carrying �mdxE483 (mdxE missing an internal 483-bp fragment) This workpBluemsmX pBluescript II SK(�) derivative carrying msmX This workpBluescript II SK(�) Cloning vector 1pDG792 pMTL23 derivative carrying the aphA3 antibiotic cassette 13pGP109 pBluescript II SK(�) derivative carrying malP::spec 29pMalEkurzts pWH1509C derivative carrying �mdxE483 (mdxE missing an internal 483-bp fragment) This workpMBP pBluescript II SK(�) derivative carrying mdxE of B. subtilis missing the first 65 bp of mdxE This workpMBP-His6x pQE-9 derivative carrying mdxE fused N terminally to a His tag coding region This workpMBPK1 pMBP derivative containing a mdxE gene disruption by the aphA3 gene in mdxE direction This workpmsmXK pBluescript II SK(�) derivative carrying msmX::aphA3 This workpQE-9 Expression vector for His tag fusions under T5 promoter control 27pWH1509C Shuttle vector (E. coli-B. subtilis) carrying genes for tetracycline and ampicillin resistance and the pBR327

ori for E. coli and a chloramphenicol resistance gene and the temperature-sensitive pE194(ts) ori forB. subtilis

31

a For a detailed description of construction, see Materials and Methods.

3912 SCHONERT ET AL. J. BACTERIOL.


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

nipulate DNA (33). PCRs (23) were performed with Taq polymerase (Boehr-inger Mannheim GmbH, Germany). Plasmids carrying fragments that were madeby PCR were verified by sequencing.

Plasmid pMBP was constructed by ligating an XbaI/EcoRI-restricted 1,227-bpDNA fragment into XbaI/EcoRI-restricted pBluescript II SK(�). The fragmentwas obtained by PCR using primers EcoRI-yvdG (5�-GGA ATT CTG CTC AAGTTC AAA AAA TCC AGC-3�) and XbaI-yvdG (5�-CCC TCT AGA CTT CCGGAC GCT ATC-3�), introducing an EcoRI site 5� to mdxE and an XbaI site 3�to the gene. Chromosomal DNA of wild-type B. subtilis 168 was used as atemplate. Plasmid pMBPHis6x was constructed by ligating a fragment harboringmdxE without the leader peptide-encoding sequence with vector pQE-9. Boththe fragment and pQE-9 were restricted with BamHI and PstI prior to ligation.The fragment was amplified by PCR using chromosomal DNA of B. subtilis 168as a template and the primers yvdG-BamHI (5�-CGG GAT CCT GCT CAAGTT CAA AAA ATC C-3�) and yvdG-PstI (5�-CCT CTA TTC TTC CGC TGCAGA TCC C-3�), introducing a 5� BamHI site and a 3� PstI site of the truncatedmdxE gene.

Plasmids pMBPK1 and pMBPK2 were obtained by inserting a 1,497-bp SmaI/StuI DNA fragment of plasmid pDG792 carrying the aphA3 kanamycin resis-tance cassette (13) into HpaI-restricted pMBP. Transformants of E. coli strainTG1 (12) were selected on Luria-Bertani plates supplemented with ampicillin(100 �g/ml) and kanamycin (25 �g/ml). The direction of the aphA3 gene withrespect to mdxE was determined by restriction of the plasmids with HindIII.

Plasmid pBlueMBPkurz was constructed by ligation of the 3,320-bp XbaI/BglII-restricted fragment of pMBP with an XbaI/BglII-restricted 453-bp frag-ment obtained by PCR using primer �malE-BglII (5�-GAA ATA TAC AAAAGA TCT CGA GCT GG-3�) and a reverse primer (5�-AAC AGC TAT GACCAT G-3�) and plasmid pMBP as a template.

Plasmid pMalEkurzts was achieved by ligation of the 768-bp fragment result-ing from the EagI/SalI restriction of pBlueMBPkurz with the 6,991-bp fragmentfrom the EagI/SalI cleavage of pWH1509C (31).

Plasmid pBlueHI was constructed by ligation of the 2,241-bp fragment result-ing from the ClaI/EheI restriction of an mdxF-mdxG-carrying PCR product(primers, a, yvdH-5� [5�-GCA AGG AAG AAA GCC GAT GAG CA-3�], and b,yvdH-3� [5�-CCC GCC TGT TAG TTT TCC GTT CCT-3�]; template, chromo-somal DNA of B. subtilis 168) with the 2,927-bp fragment of a ClaI/SmaI restric-tion of pBluescript II SK(�).

Plasmids pBlueHIK1 and pBlueHIK2 were created following a strategy similarto that described for pMBPK1 and pMBPK2 (see above). pBlueHI was restrictedwith Tth111I and Eco47III, followed by a fill-in modification of the Tth111I,resulting in sticky ends with T4 polymerase. The direction of the aphA3 resis-

tance cassette with respect to the mdxF and mdxG genes was proven by BlpI/SacII restriction.

Plasmid pBluemsmX was constructed so that a 1,123-bp fragment resultingfrom a BamHI/SalI restriction of the PCR product obtained with primers msmX-BamHI (5�-GGG AGG ATC CAT GGC TGA ATT GCG GAT G-3�) andmsmX-SalI (5�-CAT GTC GAC ATG TCC GGT TTT TTT GAT CTT ATCG-3�) and chromosomal DNA of wild-type B. subtilis as a template was ligatedinto BamHI/SalI-restricted pBluescript II SK(�).

Plasmid pmsmXK was obtained by inserting a 1,537-bp NruI/EcoRI DNAfragment of plasmid pDG792 carrying the aphA3 kanamycin resistance cassette(13) into NruI/EcoRI-restricted pBluemsmX.

B. subtilis wild-type strain 168 (1A1) was obtained from the Bacillus GeneticStock Center (Ohio State University). For B. subtilis strain constructions (Table2), transformation was carried out using a one-step procedure (19). StrainMD215 was constructed by transforming B. subtilis 168 with linearized plasmidpMBPK1, followed by selection on kanamycin (25 �g/ml). To obtain a nonpolardeletion of the mdxE gene, MD215 was transformed at 30°C with the circularplasmid pMalEkurzts, followed by selection on kanamycin (25 �g/ml) and chlor-amphenicol (5 �g/ml). The resulting mutant was used to inoculate 4 ml LB/chloramphenicol (5 �g/ml). The culture was grown at 42°C overnight; 1 ml ofthat culture was again transferred to 4 ml LB/chloramphenicol (5 �g/ml). Growthwas allowed at 42°C for an additional 16 h. One hundred microliters of theculture was plated on LB/chloramphenicol (5 �g/ml) and incubated at 42°Covernight. The resulting colonies were used to inoculate 4 ml minimal mediumM9 (22 mM potassium dihydrogen phosphate, 16.8 mM disodium hydrogenphosphate, 7.6 mM ammonium sulfate, 8.6 mM sodium chloride, 1 mM magne-sium sulfate, 0.05% [wt/vol] Casamino Acids, 0.05% [wt/vol] yeast extract, pH7.4) without antibiotics. The culture was incubated at 28°C for 2 days and dilutedto 10�6. Aliquots of 100 �l were plated on LB. About 20,000 colonies werescreened for kanamycin- and chloramphenicol-sensitive mutants by replica plat-ing, resulting in one positive clone (strain MD235).

Mutants in malP were constructed by transformation with ScaI-linearizedplasmid pGP109 (29) and selection on spectinomycin (100 �g/ml). Mutantsof the mdxF-mdxG region were constructed by transformation with plasmidpBlueHIK1 that was linearized with DraIII, followed by selection on kanamycin(25 �g/ml). The amyE gene was disrupted by transformation of B. subtilis withSapI-linearized plasmid pAC6 (40), followed by selection on chloramphenicol (5�g/ml). Mutants in msmX were constructed by transformation with FspI-linear-ized plasmid pmsmXK (29) and selection on kanamycin (25 �g/ml).

To verify correct mutagenesis of the malP, msmX, mdxE, and mdxF-mdxGgenes, chromosomal DNA of the mutants was isolated (25) and used as a

TABLE 2. Bacterial strains used in this work

Strain Genotype Source/construction orreference

Bacillus subtilis168 (wild type) trpC2 BGSC,a 1A1GP110 trpC2 malP::spec 29MD153 trpC2 ptsG::cat 6MD195 trpC2 amyE::cat 34MD215 trpC2 mdxE::aphA3 pMBPK1 tf �b 168MD235 trpC2 �mdxE483 �c

MD246 trpC2 malP::spec amyE::cat pAC6 tf � GP110MD247 trpC2 �mdxE483 malP::spec pGP109 tf � MD235MD248 trpC2 �mdxE483 amyE::cat pAC6 tf � MD235MD252 trpC2 �mdxE483 malP::spec amyE::cat pAC6 tf � MD247MD274 trpC2 (mdxF-mdxG)::aphA3 pBlueHIK1 tf � 168MD276 trpC2 (mdxF-mdxG)::aphA3 amyE::cat pAC6 tf � MD274MD278 trpC2 (mdxF-mdxG)::aphA3 malP::spec pBlueHIK1 tf � GP110MD280 trpC2 (mdxF-mdxG)::aphA3 malP::spec amyE::cat pBlueHIK1 tf � MD246MD285 trpC2 msmX::aphA3 pmsmXK tf � 168MD286 trpC2 msmX::aphA3 malP::spec pmsmXK tf � GP110

Escherichia coliRB791 F� lacIq L8 hsdR� hsdM 4TG1 supE thi hsdD �(lac-proAB) F� traD36 proAB� lacIq lacZ�M15 12

a BGSC, Bacillus Genetic Stock Center, Ohio State University, Columbus.b tf � indicates transformation with the DNA mentioned.c For a description of construction, see Materials and Methods.

VOL. 188, 2006 MALTOSE AND MALTODEXTRIN UTILIZATION BY B. SUBTILIS 3913


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

template for PCR (for the primer sets, see above). The resulting DNA fragmentswere compared to that received using chromosomal DNA of the parental strainas a template DNA under the same conditions. Correct amyE mutants werechecked for loss of amylase activity on starch plates (36).

Computer analysis. For sequence analysis, we used programs from the Uni-versity of Wisconsin Genetics Computer Group (9) and the BLAST server (45)of the National Center for Biotechnology Information at the National Institutesof Health, Bethesda, Maryland (http://www.ncbi.nlm.nih.gov), as well as CloneManager 5 (Scientific and Educational Software, Durham, NC) and Lasergene(DNAstar, Inc., Madison, WI) software.

Overproduction and purification of the maltodextrin-binding protein. Over-production of the mdxE gene product was achieved with plasmid pMBPHis6xcarrying the His tag coding region of plasmid pQE-9 fused in frame to mdxElacking the first 23 codons of the open reading frame, leading to expression underthe control of the IPTG (isopropyl-�-D-thiogalactopyranoside)-inducible T5 pro-moter in E. coli strain RB791 (4, 27).

A 1-liter culture of E. coli strain RB791/pMBPHis6x was grown at 37°C inLuria-Bertani broth (33) containing ampicillin (100 �g/ml). Expression was in-duced by the addition of 2 mM IPTG as the culture reached an A600 of 0.5.Growth was allowed to proceed for another 4 h after induction before the cellswere harvested by centrifugation at 5,000 � g. The resulting cell pellet waswashed once in lysis buffer (10 mM imidazole, 50 mM potassium dihydrogenphosphate, 50 mM potassium chloride, pH 7.0), resuspended in 10 ml of thesame buffer, and used immediately. Cell extracts were prepared as describedpreviously (24), with minor modifications. The cells were sonicated eight timesfor 30 s each time at 40 W with 0.9-s pulse intervals, using a Labsonic U sonicator(B. Braun, Melsungen, Germany). After centrifugation for 50 min at 30,000 � g,the supernatant was collected. Maltodextrin-binding protein was purified with anNi2�-loaded 1-ml HiTrap Chelating column and a Pharmacia �kta Purifierapparatus following the instructions of the manufacturer (Pharmacia, Freiburg,Germany). The elution buffer was 50 mM NaH2PO4, 50 mM NaCl, 500 mMimidazole, pH 7.

For further purification and to remove imidazole, the eluate was passed overa size fractionation column (HiLoad Superdex 75; buffer, 50 mM NaH2PO4, 50mM NaCl, pH 7) following the instructions of the manufacturer (Pharmacia,Freiburg, Germany). Protein-containing fractions were analyzed by sodium do-decyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

SDS-PAGE. Crude extracts of E. coli RB791/pMBPHis6x cells induced with 2mM IPTG and lysed by SDS were prepared according to the method of Silhavyet al. (37). Aliquots of fractions from protein purification were mixed withloading buffer prior to SDS-PAGE on 10% gels (20), using a minigel system(Bio-Rad Laboratories, Richmond, CA.). The molecular size reference marker

70L was obtained from Sigma (Munich, Germany). Proteins were visualized withCoomassie blue R250.

Surface plasmon resonance spectroscopy. Interaction of the purified MdxEprotein with potential substrates was analyzed using surface plasmon resonanceperformed on a BIAcoreX (Pharmacia Biosensor AB, Uppsala, Sweden) (21,39). Research grade Ni2�-loaded nitrilotriacetic acid chips were coated withpurified His6-tagged MdxE to about 3,000 to 4,000 resonance units. The sameamount of His6-tagged GlcK (38) or Ni2�-loaded nitrilotriacetic acid chips with-out protein was used in the reference flow cell. Reaction temperatures were setto 25°C. Eluent buffer (10 mM HEPES, 150 mM NaCl, 50 �M EDTA, 0.005%Surfactant P20, pH 7.4) was used as a running buffer (Pharmacia). For furtheranalysis, sugars (as mentioned above) were diluted in eluent buffer and passedover the experimental and reference flow cells at a flow rate of 5 �l/min. Todetermine the interactions between the coupled protein and the different sub-strates, signals obtained from the reference flow cell were subtracted from signalsobtained from the experimental flow cell. To determine substrate Kd values,concentrations of 1 �M, 5 �M, 15 �M, 50 �M, 100 �M, 500 �M, 1 mM, 5 mM,and 10 mM of the indicated sugars were used. The values obtained were used tocalculate the substrate affinities from Scatchard plots.

Transport assays. B. subtilis strains were grown in Luria-Bertani broth sup-plemented with the sugars indicated. Cells were harvested by centrifugation for5 min at 5,000 � g, washed three times in transport buffer (50 mM Tris-HCl, 20mM MgCl2, pH 7.2), and resuspended in transport buffer to an A600 of 1. Underthe conditions used, a 1-ml cell suspension with an A600 of 1 led to 2.2 � 108

CFU. All washing steps were carried out at room temperature.To determine transport activity, 500 �l of cell suspension was preincubated at

37°C for 5 min. After preincubation, 14C-labeled sugars (final concentration,

FIG. 2. Overlay of several sensorgrams derived from the interactions of MdxE with different carbohydrates. The sugars used are listed on theright. The order of the sugars corresponds to the order of the responses. Responses are given in resonance units (RU) on the ordinate, and timeis given on the abscissa. The start of injection of the sugar solutions is set to 0 s. Time of injection, 360 s.

TABLE 3. Substrate binding affinities of Bacillus subtilis MdxEa

Substrate Kd (mM)

Maltose ..............................................................................................1.045Maltotriose ........................................................................................0.007Maltotetraose ....................................................................................0.007Maltopentaose...................................................................................0.004Maltohexaose ....................................................................................0.003Maltoheptaose...................................................................................0.005

a Substrate binding affinities were determined using surface plasmon reso-nance spectroscopy.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

1 �M) were added. The cells were further incubated at 37°C. After 15, 30, 45, 60,and 120 seconds, 70-�l aliquots were taken, filtered through 0.45-�m-pore-sizeNC 45 filters (diameter, 25 mm; Schleicher & Schuell GmbH, Dassel, Germany),and washed three times with 5 ml transport buffer. The radioactivity retained onthe filters was determined in a scintillation counter (LS 1801; Beckmann, Mu-nich, Germany). Uptake rates were calculated based on the maximal gradients ofthe resulting curves when retained radioactivity was plot over time.

In the case of titration experiments, 490-�l preincubated cells (see above) wereplaced in a 10-�l mixture of 14C (final concentration, 1 �M) and 12C sugars (forthe sugars and final concentrations, see Results). Uptake rates were determinedas described above.

For kinetic studies, the 10-�l sugar mixture was composed of 2.3 pmol[14C]maltose (1.4 nCi) or 7.8 pmol [14C]maltotriose (7 nCi) and differentamounts of the appropriate 12C sugar, leading to final overall sugar concentra-tions from 0.1 �M to 50 �M.

Growth experiments. To study the phenotypes of different mutants forgrowth on maltodextrins, we used minimal medium (44 mM potassium dihy-drogen phosphate, 60 mM dipotassium hydrogen phosphate, 2.9 mM triso-dium citrate, 15 mM ammonium sulfate, 245 �M tryptophan, 33 �M iron(III)citrate, 2 mM magnesium sulfate, 1 mM calcium chloride, 17 mM potassiumL-glutamate) containing 0.1% maltodextrins (ICN). Cells from overnight cul-tures (grown in LB containing the relevant antibiotics) were washed three

FIG. 3. Cytoplasmic �-glucosidase activities in different B. subtilis mutants compared to the wild type. All strains were grown in LB (uninduced),LB containing 1% maltose (induced), or LB containing 1% maltose and 1% glucose (repressed). The error bars represent root mean square errors.

FIG. 4. [14C]maltose transport in different B. subtilis mutants. Uptake rates are given in nmol maltose taken up per minute by 1010 CFU. Allmutant strains were grown in LB containing 1% maltose (gray columns). The wild type (wt) was grown in LB containing 1% maltose (induced;gray column), LB containing 1% maltose and 1% glucose (repressed; white column) or LB (uninduced; black column). The relevant genotypes areindicated below. The error bars represent standard deviations.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

times with minimal medium before they were used to inoculate fresh minimalmedium. Absorption was monitored at 600 nm.

�-Glucosidase assays (MalL activity assay). MalL activity in crude cell extractof B. subtilis was measured according to the �-galactosidase method of Miller(22) with slight modifications. Instead of o-nitrophenyl-�-D-galactopyranoside,p-nitrophenyl-�-D-glucopyranoside was used as a substrate. Cells grown in theindicated media were harvested by centrifugation for 5 min at 5,000 � g. Toprepare crude cell extracts, cells were resuspended in Z buffer (22) and treatedwith lysozyme/DNase I (final concentrations, lysozyme, 40 �g/ml, and DNase I,6 �g/ml) at 37°C until lysis of the cells occurred. Cell debris was removed bycentrifugation for 5 min at 10,000 � g and 4°C. The resulting supernatant wasused further. Protein concentrations of the cell extracts were determined usingthe Bio-Rad protein assay (Bio-Rad, Munich, Germany). For the assay, 200 �lcell extract or less (when high activity was expected) was brought to 800 �l withZ buffer. The mixture was incubated at 28°C for 5 min. The reaction at 28°C wasstarted by the addition of 200 �l p-nitrophenyl-�-D-glucopyranoside (4 mg/mldissolved in 0.1 M phosphate buffer, pH 7.0) (22). After the mixture becameslightly yellow, the reaction was stopped by the addition of 0.5 ml 1 M Na2CO3.The resulting p-nitrophenol was measured at 420 nm. As a reference, 800 �l Zbuffer treated in the same way was used. Under these conditions, 1 nmolp-nitrophenol has an optical density at 420 nm of 0.0097. The specific activity isgiven in nmol · min�1 · mg crude cell extract�1.

RESULTS

Overproduction and purification of the His6-tagged MdxEprotein. The primary sequence deduced from the mdxE nucle-otide sequence exhibits similarities to maltose-binding proteinsof enteric bacteria (7, 10) and to several other binding proteinsin B. subtilis (28). Several attempts to clone the complete mdxEgene in high-copy-number expression vectors replicating in E.

coli failed (data not shown). Therefore, we constructed a vari-ant in which the signal sequence (the first 22 amino acids [7, 28,10]) was exchanged for a His6 tag, followed by Cys23 of thenative protein. This protein has a 12-residue N-terminal exten-sion that includes the affinity tag (underlined): Met-Arg-Gly-Ser-His-His-His-His-His-His-Gly-Ser-Cys (Cys23 of the nativeprotein). The modified protein was overproduced by E. colistrain RB791/pMBPHis6x and purified to homogeneity byNi2� affinity gel chromatography and gel filtration (data notshown).

Characterization of MdxE substrate specificity. To investi-gate the substrate binding properties of the purified protein,substrate interactions were analyzed by surface plasmon reso-nance spectroscopy at sugar concentrations of 1 mM. Maltoseand maltodextrins up to maltoheptaose were able to triggerresponse signals (Fig. 2). As expected, the number of responseunits increased with the size of the interacting substrate. Nosignals were observed under the same experimental conditionsusing isomaltose, isomaltotriose, isomaltotetraose, lactose, pal-atinose, sucrose, or trehalose (some of the controls are shownin Fig. 2). To determine the dissociation constants, we variedthe substrate concentrations in the range between 1 �M and 1mM (maximum, 10 mM for maltose) (Table 3). These datashow that MdxE functions as a maltodextrin-binding proteinthat exhibits micromolar affinities for maltodextrins (e.g., malto-hexaose; Kd, 3 �M) but only a low affinity for maltose (Kd, 1

FIG. 5. Uptake of [14C]maltose by B. subtilis strain MD274 (mdxF-mdxG::aphA3) in the presence of different concentrations of 12C sugars. The12C sugars used are indicated below the columns. For comparison, uptake of [14C]maltose (1 �M) without additional 12C sugar is shown as apositive control (hatched column). Uptake rates are given in nmol [14C]maltose taken up per minute by 1010 CFU. The error bars represent rootmean square errors.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

mM). We therefore propose that the yvdG gene be renamedmdxE and its product MdxE.

Determination of �-glucosidase (MalL) activity. To testwhether mutations reduce the expression of downstream genesin the yvdE-pgcM gene cluster, we measured MalL-mediated�-glucosidase activity in cell extracts obtained from culturesgrown in Luria-Bertani broth supplemented with the indicatedsugars at 1%. Insertion of the aphA3 resistance cassette (whichcarries its own constitutive promoter) in mdxE and in the sameorientation as mdxE resulted in constitutive MalL activity. Inthis mutant, MalL activity is independent of an exogenousinducer. In a mutant which carried the �mdxE483 deletion,MalL activity was comparable to that of the wild type (Fig. 3).These data showed that the �mdxE483 mutation is not polaron the distal genes.

On the other hand the mdxF-mdxG insertion showed almostno activity and is therefore polar on malL.

Maltose and maltotriose uptake. To test whether B. subtilisMdxE and the mdxF and mdxG gene-encoded products play asignificant role in transport, we measured [14C]maltose and[14C]maltotriose uptake in several mutants and compared it tothat of the wild type after growth in Luria-Bertani broth inthe presence and absence of 1% maltose, as well as 1% maltoseand 1% glucose, respectively. As seen in Fig. 4, maltose uptakewas induced by maltose, but induction was reduced by glucose.No significant differences in maltose uptake rates betweenthe wild type and the mdxE or mdxF-mdxG mutant were detect-able. In contrast, in a malP single mutant or in a malP mdxE ora malP mdxF-mdxG double mutant, no maltose uptake wasdetectable after growth in Luria-Bertani broth in the presenceof 1% maltose. These data demonstrate that maltose uptake isMalP-dependent (as indicated by Reizer et al. [29]) and inde-pendent of MdxE, MdxF, and MdxG. We also measured malt-ose transport in a ptsG mutant to test whether the requiredphosphoryl group for transport is delivered by enzyme IIGlc

(PtsG). Since maltose transport in this strain is elevated evencompared to the wild type, PtsG is not required for maltosetransport, confirming previous observations by Reizer et al.(29), but might have influence on maltose transport regulation.

To further characterize maltose transport, we analyzed[14C]maltose uptake in the mdxF-mdxG mutant in the presenceof different amounts (10 �M, 100 �M, and 1 mM) of severalunlabeled sugars. The mdxF-mdxG mutant was used to preventpossible interference between the maltodextrin ABC trans-porter and [14C]maltose uptake when maltodextrins were usedin the competition experiment. As shown in Fig. 5, unlabeledmaltose exhibited the most significant effect on [14C]maltose

FIG. 6. [14C]maltotriose transport in different B. subtilis mutants. Uptake rates are given in nmol maltose taken up per minute by 1010 CFU.All mutant strains were grown in LB containing 1% maltose (gray columns). The wild type (wt) was grown in LB containing 1% maltose (induced;gray column), LB containing 1% maltose and 1% glucose (repressed; white column), or LB (uninduced; black column). The relevant genotypesare indicated below. The error bars represent standard deviations.

TABLE 4. Transport affinities a of Bacillus subtilis

Parameter

Value for strain:

Maltoseb Maltotriose

MD195 MD235 GP110 MD246

Relevant genotype amyE �mdxE malP amyE malPApparent Km(�M) 5 7.5 2.2 1.4Vmax nmol · min�1 ·

(1010 cfu)�191 114 3.2 4.7

a Transport affinities (given as apparent Km) were determined assuming thattransport follows Michaelis-Menten kinetics.

b Substrate.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

uptake. Weaker effects were obtained with glucose and treha-lose. Maltotriose resulted only in slight effects at high concen-trations. Nearly no effects were detected when maltotetraose,maltopentaose, maltohexaose, or maltoheptaose was used(data for maltohexaose and maltoheptaose not shown). Withthe exception of maltose, none of the sugars at 1 mM concen-tration blocked transport completely. Whether the interfer-ence of glucose and trehalose with maltose transport is director indirect (e.g., by competing for phosphorylated PTS pro-teins) cannot be determined based on these data.

We also determined the kinetics for maltose uptake in anamyE and an mdxE mutant in order to avoid problems withextracellular hydrolytic activity or interference with the poten-tial maltodextrin system (Table 4). An apparent Km of 5 �Mand a Vmax of 91 nmol · min�1 · (1010 CFU)�1 for the amyEmutant, as well as an apparent Km of 7.5 �M and a Vmax of 114nmol · min�1 · (1010 CFU)�1 for the mdxE mutant, wasdetermined.

Maltotriose uptake in the wild type is induced by maltoseand reduced by glucose (Fig. 6). No uptake of maltotriose wasdetectable in malP mdxE, malP mdxF-mdxG, or malP msmXdouble mutants (Fig. 6), leading to the conclusion that malto-triose uptake is mediated by the ABC transporter composed ofMdxE, MdxF, MdxG, and MsmX. The observation that mal-totriose uptake was only slightly or not at all affected in thecorresponding single mutants (Fig. 6) was most likely causedby the contamination of [14C]maltotriose by [14C]maltose (datanot shown).

Since the purity of the [14C]maltotriose used was not satis-factory, we chose the malP mutant for further analysis of mal-totriose uptake. As shown in Fig. 7, the presence of 10 �M

maltotriose, maltotetraose, maltopentaose, maltohexaose, ormaltoheptaose was able to reduce maltotriose uptake in thisstrain significantly. Glucose, maltose, and lactose lower malto-triose uptake, whereas trehalose shows no effect on it. We alsodetermined the transport kinetics for maltotriose (Table 4),revealing that the mdxE-, mdxF-, mdxG-, and msmX-encodedABC transporter is a high-affinity transporter with an apparentKm of 2.2 �M and a Vmax of 3.2 nmol · min�1 · (1010 CFU)�1.The same analysis in a malP amyE double mutant yielded anapparent Km of 1.4 �M and a Vmax of 4.7 nmol · min�1 · (1010

CFU)�1 for maltotriose uptake.These data demonstrate that maltose is taken up by MalP

and that maltotriose, and probably maltodextrin, transport ismediated via the mdxE-, mdxF-, mdxG-, and msmX-encodedABC transporter.

Growth in minimal medium with maltodextrins or maltose.To get more information about utilization of maltodextrins byB. subtilis, we analyzed the phenotypes of several mutantsgrown in minimal medium with potassium glutamate as a car-bon source in the presence of maltose or maltodextrins. Wild-type cells grown in minimal medium with potassium glutamateas a sole carbon source served as a control.

When grown in minimal medium containing 0.1% maltose,all mutants defective in malP showed prolonged doublingtimes (about 74 min) and reached a final optical density (A600)of about 1.0 to 1.4 (Table 5). This growth behavior was com-parable to that of the wild type when grown in minimal me-dium lacking an additional carbon source (doubling time, 84min; final absorption, 1.3). All mutants defective in mdxF-mdxG but carrying wild-type malP showed doubling times likethat of the wild type grown in minimal medium containing

FIG. 7. Uptake of [14C]maltotriose by B. subtilis strain GP110 (malP::spec) in the presence of different 12C sugars [10 �M]. The 12C sugars usedare indicated below the columns. For comparison, uptake of [14C]maltotriose (1 �M) without additional 12C sugar is shown as a positive control.Uptake rates are given in nmol [14C]maltotriose taken up per minute by 1010 CFU. The error bars represent standard deviations.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

0.1% maltose but reached a final absorption at 600 nm ofabout 2.5. All other mutants showed growth behavior like thatof the wild type when grown in minimal medium containing0.1% maltose (doubling time, 59 min; final absorption, 3.4)(Table 5). These data show that neither extracellular amylaseAmyE nor the maltodextrin ABC transporter is required formaltose uptake.

When the cells were grown in minimal medium containing0.1% maltodextrins, the mdxE, mdxF-mdxG, and amyE singlemutants showed doubling times comparable to that of the wildtype (Table 6). However, the mdxF-mdxG mutant reached anoptical density at 600 nm of only 3.0. When the amyE mutationwas combined with either mdxE or mdxF-mdxG, the mutantsshowed prolonged doubling times and reached final absorp-tions of about 2.7. The malP single and malP amyE, malPmdxE, and malP mdxF-mdxG double mutants showed ex-tremely prolonged doubling times but reached optical densitiesof about 2. When triple mutants (malP amyE in combinationwith either mdxE or mdxF-mdxG) were tested, they showedgrowth behavior like that of the wild type when grown inminimal medium without an additional carbon source (Table6). Thus, maltodextrins can either be transported inside thecell by the maltodextrin ABC transporter or, after degradationto maltose by extracellular amylase AmyE, transported as malt-ose by MalP.

DISCUSSION

Bacillus subtilis contains at least 37 open reading framesencoding putative solute-binding proteins (18, 28, 41), but thespecificities of many of these transporter systems are not yetknown. Three of them showed significant homology to maltose/

maltodextrin-binding proteins, namely, YesO, YvfK, and MdxE(YvdG) (18, 41). Recently, it was shown that YvfK interactswith linear and cyclic maltodextrins (16). The mdxE (yvdG)gene is located in a cluster of nine genes. One of the othergenes, malL, encodes an �-glucosidase. Its activity is inducibleby maltose (34). This raised the question of whether this pu-tative transport system that mdxE belongs to is responsible formaltose uptake.

To analyze the function of MdxE, we purified a lipidless,N-terminally His6-tagged variant of the putative maltose bind-ing protein. Using surface plasmon resonance spectroscopy,maltose (Kd, 1 mM) and longer maltodextrins (Kd, 3 to 6 �M)were found to bind to MdxE (Table 3). Based on the lowaffinity for maltose, it is not clear whether maltose is an actualsubstrate. Comparison of the affinities of B. subtilis MdxE withmaltose binding proteins of E. coli (Kd, 1 �M for maltose) orAeromonas hydrophila (Kd, 1.6 �M for maltose) or the affinityof glycine/betaine binding protein of B. subtilis for glycine/betaine (Kd, 6 �M) (17) suggests that B. subtilis MdxE is amaltodextrin-binding protein rather than a maltose bindingprotein.

To test whether the in vitro results obtained can be con-firmed by in vivo experiments, we measured maltose and mal-totriose transport. The data obtained show that maltose trans-port is inducible by maltose and underlies carbon cataboliterepression mediated by glucose. Comparing the wild typegrown in Luria-Bertani medium (supplemented with maltose)with the different mutants, all mutants defective in malP lostmaltose transport. These data are in agreement with previousdata, which demonstrated that a malP mutant showed seven-fold-increased doubling times on maltose minimal medium(29). The mdxE or mdxF-mdxG (yvdH-yvdI) single mutations

TABLE 5. Phenotype analysis of Bacillus subtilis strains for growthbehavior on minimal mediuma plus maltose

Strain Relevantgenotype

Carbonsource

Doublingtime

(min)d

Finalabsorptione

wt168 Wild type Maltoseb 59 3.4MD195 amyE::cat Maltose 58 3.6MD235 �mdxE483 Maltose 57 3.6MD274 mdxF-mdxG::aphA3 Maltose 60 2.6MD248 �mdxE483 amyE::cat Maltose 60 3.8MD276 mdxF-mdxG::aphA3

amyE::catMaltose 59 2.4

GP110 malP::spec Maltose 73 1MD246 malP::spec amyE::cat Maltose 70 1.3MD247 malP::spec �mdxE483 Maltose 73 1.1MD278 malP::spec mdxF-

mdxG::aphA3Maltose 76 1.4

MD252 malP::spec �mdxE483amyE::cat

Maltose 75 1.2

MD280 malP::spec mdxF-mdxG::aphA3 amyE::cat

Maltose 73 1.2

wt168 Wild type �c 84 1.3

a For composition of minimal medium, see Materials and Methods.b Cells were grown in minimal medium containing K-glutamate and 0.1%

maltose.c Cells were grown in minimal medium with K-glutamate as the sole carbon

source.d Doubling times were calculated based on midlog phase.e Optical density was measured at 600 nm. Final absorption means optical

density that was measured after all cultures had entered stationary phase for 3hours (time point, 14 h).

TABLE 6. Phenotype analysis of Bacillus subtilis strains for growthbehavior on minimal mediuma plus maltodextrins

Strain Relevantgenotype

Carbonsource

Doublingtime

(min)d

Finalabsorptione

wt168 Wild type Maltodextrinsb 69 3.6MD195 amyE::cat Maltodextrins 71 4MD235 �mdxE483 Maltodextrins 71 3.8MD274 mdxF-mdxG::aphA3 Maltodextrins 71 3MD248 �mdxE483 amyE::cat Maltodextrins 76 2.8MD276 mdxF-mdxG::aphA3

amyE::catMaltodextrins 75 2.6

GP110 malP::spec Maltodextrins 87 2.2MD246 malP::spec amyE::cat Maltodextrins 87 2MD247 malP::spec �mdxE483 Maltodextrins 81 1.6MD278 malP::spec

mdxF-mdxG::aphA3Maltodextrins 86 1.8

MD252 malP::spec �mdxE483amyE::cat

Maltodextrins 87 1.1

MD280 malP::specmdxF-mdxG::aphA3amyE::cat

Maltodextrins 88 1.4

wt168 Wild type �c 84 1.3

a For composition of minimal medium, see Materials and Methods.b Cells were grown in minimal medium containing K-glutamate and 0.1%

maltodextrins.c Cells were grown in minimal medium with K-glutamate as the sole carbon

source.d Doubling times were calculated based on midlog phase.e Optical density was measured at 600 nm. Final absorption means optical

density that was measured after all cultures had entered stationary phase for 3hours (time point, 14 h).



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

showed no effect on maltose transport (Fig. 4). These datademonstrate that maltose uptake occurs only via the malP-encoded specific EIICBMal and not via another transportmechanism. A ptsG mutant showed even higher maltose trans-port rates than wild-type cells. Thus, PtsG is not the phospho-ryl group donating enzyme II, as is known to be the case insucrose or trehalose uptake (6, 43).

When a 14C-labeled maltose/maltotriose mixture was used asa substrate in transport assays, it could be shown that the malPmutant transported the maltotriose portion, whereas themdxE, malF-malG, or msmX mutant transported the maltoseportion of the labeled substrate mixture used (Fig. 6). In com-bination with the data obtained from competition experiments,these data show that maltotriose is taken up by the mdxE-,

mdxF-, mdxG-, and msmX-encoded transport system, and theyindicate that larger maltodextrins are also taken up by thissystem. The mechanism by which glucose and lactose inhibitmaltotriose transport is unclear. The competition achieved bymaltose might occur at the binding-protein level, since MdxEexhibits a low affinity for maltose (Table 3). Interestingly, mal-totriose uptake in the wild type and in the malP mutant can beinduced by exogenous maltose. Like maltose transport, malto-triose transport underlies carbon catabolite repression medi-ated by glucose (Fig. 6).

The described transport data were corroborated by growthanalysis with wild-type and mutant strains. The mdxE andmdxF-mdxG single mutants showed no significant phenotypefor growth on maltose. These data show that maltose is taken

FIG. 8. Model of starch and glycogen utilization in B. subtilis. The proteins involved are shown as circles. Protein names are indicated. Starchand glycogen are hydrolyzed by AmyE extracellularly (outside) in maltodextrins and maltose. The resulting oligosaccharides are than transportedinto the cytoplasm (inside) over the membrane by the maltodextrin ABC transporter or the maltose-specific EIICB. Further degradation of thesugars finally leads to glucose-6-P, which can enter glycolysis. The phosphoryl moiety delivering EIIA is unknown. For further explanation,see the text.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

up by the PTS-dependent EIICBMal (MalP), as was suggestedby Reizer et al. (29), and that inactivation of extracellular �-amy-lase AmyE has no influence on the utilization of maltose.

In contrast to growth on maltose, utilization of maltodex-trins showed an additional intermediate phenotype. Only whenmutations of the ABC transporter were combined with muta-tions in amyE, or when amyE mutations were combined withthe mutation in malP, were doubling times prolonged and theabsorptions finally reached reduced. When mutations of theABC transporter were combined with the mutation in amyEand the mutation in malP, the strains showed growth behaviorsimilar to that of the wild type in minimal medium without anadditional carbon source.

These data led to our current model of starch or glycogenutilization by B. subtilis (Fig. 8). First, these polysaccharidesare hydrolyzed extracellularly by AmyE, resulting in maltoseand maltodextrins (15). Maltose is then taken up by the PTSand becomes phosphorylated. Cytoplasmic maltose-P is hydro-lyzed by MalA, resulting in glucose and glucose-6-P (46). Mal-todextrins are taken up by the maltodextrin-specific ABCtransporter composed of MdxE, MdxF, MdxG, and MsmXwithout phosphorylation. After they enter the cell, they aredegraded in the concerted action of cytoplasmic maltogenicamylase or neopullulanase YvdF (5), maltose phosphorylaseYvdK, and �-glucosidase MalL (34, 35), resulting in glucoseand glucose-1-P. Glucose-1-P is than converted to glucose-6-Pby PgcM (30). Free glucose resulting from maltodextrin deg-radation, as well as from maltose-P hydrolysis, is finally phos-phorylated by glucose kinase GlcK, leading to glucose-6-P (38).By this concerted action, exogenous starch and glycogen areconverted to glucose-6-P that can enter glycolysis. Transcrip-tion of the mal operon is regulated by GlvR (47), whereastranscription of the maltodextrin operon remains unclear butcould be regulated by YvdE. The EIIA domain that deliversthe phosphoryl group to EIICBMal remains unknown, but it isclearly not the EIIA domain of PtsG, as in sucrose or trehaloseuptake (6, 43).

ACKNOWLEDGMENTS

We thank Winfried Boos, in whose laboratory work was done.Financial support was from the Deutsche Forschungsgemeinschaft

(TR-SFB11) and the BMBF Knoll AG.Dedicated in loving memory to Michael K. Dahl, who died unex-

pectedly on 4 May 2003.

REFERENCES

1. Alting-Mees, M. A., and J. M. Short. 1989. pBluescript II: gene mappingvectors. Nucleic Acids Res. 17:9494.

2. Boos, W., and J. M. Lucht. 1996. Periplasmic binding protein-dependentABC transporters, p. 1175–1209. In C. C. Lin, R. B. Loe, B. Magasanik, W. S.Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichiacoli and Salmonella: cellular and molecular biology, 2nd ed. ASM Press,Washington, D.C.

3. Boos, W., and H. Shuman. 1998. Maltose/maltodextrin system of Escherichiacoli: transport, metabolism, and regulation. Microbiol. Mol. Biol. Rev. 62:204–229.

4. Brent, R., and M. Ptashne. 1981. Mechanism of action of the lexA geneproduct. Proc. Natl. Acad. Sci. USA 78:4202–4208.

5. Cho, H. Y., Y. W. Kim, T. J. Kim, H. S. Lee, D. Y. Kim, J. W. Kim, Y. W. Lee,S. Leed, and K. H. Park. 2000. Molecular characterization of a dimericintracellular maltogenic amylase of Bacillus subtilis SUH4-2. Biochim. Bio-phys. Acta 1478:333–340.

6. Dahl, M. K. 1997. Enzyme IIGlc contributes to trehalose metabolism inBacillus subtilis. FEMS Microbiol. Lett. 148:233–238.

7. Dahl, M. K., E. Francoz, W. Saurin, W. Boos, M. D. Manson, and M.Hofnung. 1989. Comparison of sequences from the malB regions of Salmo-

nella typhimurium and Enterobacter aerogenes with Escherichia coli K12: apotential new regulatory site in the interoperonic region. Mol. Gen. Genet.218:199–207.

8. Davidson, A. L., and J. Chen. 2004. ATP-binding cassette transporters inbacteria. Annu. Rev. Biochem. 73:241–268.

9. Deveraux, J., P. Haeberli, and O. Smithies. 1984. A compressive set ofsequence analysis programs for the VAX. Nucleic Acids Res. 12:387.

10. Duplay, P., H. Bedouelle, A. Fowler, I. Zabin, W. Saurin, and M. Hofnung.1984. Sequences of the malE gene and of its product, the maltose-bindingprotein of Escherichia coli K12. J. Biol. Chem. 259:10606–10613.

11. Durand, M., F. Pichinoty, C. Job, and M. Mandel. 1979. Carbon assimilationand taxonomic study of Bacillus subtillis and B. licheniformis. Can. J. Micro-biol. 25:491–498.

12. Gibson, T. G. 1984. Studies on the Epstein-Barr virus genome. Ph.D. thesis.Cambridge University, Cambridge, United Kingdom.

13. Guerout-Fleury, A.-M., K. Shazand, N. Frandsen, and P. Stragier. 1995.Antibiotic-resistance cassettes for Bacillus subtilis. Gene 167:335–336.

14. Higgins, C. F. 1992. ABC transporters: from microorganisms to man. Annu.Rev. Cell Biol. 8:67–113.

15. Ishikura, K., Y. Nitta, and T. Watanabe. 1977. Hydrolysis of phenyl beta-maltoside catalyzed by saccharifying alpha-amylase from Bacillus subtilis.J. Biochem. 81:1187–1192.

16. Kamionka, A., and M. K. Dahl. 2001. Bacillus subtilis contains a cyclodextrin-binding protein which is part of a putative ABC-transporter. FEMS Micro-biol. Lett. 204:55–60.

17. Kempf, B., J. Gade, and E. Bremer. 1997. Lipoprotein from the osmoregu-lated ABC transport system OpuA of Bacillus subtilis: purification of theglycine betaine binding protein and characterization of a functional lipidlessmutant. J. Bacteriol. 179:6213–6220.

18. Kunst, F., N. Ogasawara, I. Moszer, A. M. Albertini, G. Alloni, V. Azevedo,M. G. Bertero, P. Bessieres, A. Bolotin, S. Borchert, R. Borriss, L. Boursier,A. Brans, M. Braun, S. C. Brignell, S. Bron, S. Brouillet, C. V. Bruschi, B.Caldwell, V. Capuano, N. M. Carter, S. K. Choi, J. J. Codani, I. F. Connerton,A. Danchin, et al. 1997. The complete genome sequence of the gram-positivebacterium Bacillus subtilis. Nature390:249–256.

19. Kunst, F., and G. Rapoport. 1995. Salt stress is an environmental signalaffecting degradative enzyme synthesis in Bacillus subtilis. J. Bacteriol. 177:2403–2407.

20. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly ofthe head of bacteriophage T4. Nature 227:680–685.

21. Malmquist, M. 1993. Biospecific interaction analysis using biosensor tech-nology. Nature 361:186–187.

22. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring HarborLaboratory, Cold Spring Harbor, N.Y.

23. Mullis, K. B., and F. A. Faloona. 1987. Specific synthesis of DNA in vitro viaa polymerase-catalyzed chain reaction. Methods Enzymol. 155:335–350.

24. Perego, M., S. P. Cole, D. Burbulys, K. Trach, and J. A. Hoch. 1989. Char-acterization of the gene for a protein kinase which phosphorylates the sporu-lation-regulatory proteins Spo0A and Spo0F of Bacillus subtilis. J. Bacteriol.171:6187–6196.

25. Pospiech, A., and B. Neumann. 1995. A versatile quick-prep of genomicDNA from gram-positive bacteria. Trends Genet. 11:217–218.

26. Postma, P. W., J. W. Lengeler, and G. R. Jacobson. 1993. Phosphoenolpyru-vate:carbohydrate phosphotransferase systems of bacteria. Microbiol. Rev.57:543–594.

27. Qiagen. 2001. The QIAexpressionistTM, 5th ed. Qiagen Inc., Hilden, Germany.28. Quentin, Y., G. Fichant, and F. Denizot. 1999. Inventory, assembly and

analysis of Bacillus subtilis ABC transport systems. J. Mol. Biol. 287:467–484.29. Reizer, J., S. Bachem, A. Reizer, M. Arnaud, M. H. Saier, Jr., and J. Stulke.

1999. Novel phosphotransferase system genes revealed by genome analysis—the complete complement of PTS proteins encoded within the genome ofBacillus subtilis. Microbiology 145:3419–3429.

30. Rosana-Mesak, L., and M. K. Dahl. 2000. Purification and enzymatic char-acterization of PgcM, a �-phosphoglucomutase and glucose-1-phosphos-phate phosphodismutase of Bacillus subtilis. Arch. Microbiol. 174:256–264.

31. Rygus, T., and W. Hillen. 1992. Catabolite repression of the xyl operon inBacillus megaterium. J. Bacteriol. 174:3049–3055.

32. Saier, M. H., Jr., and J. Reizer. 1994. The bacterial phosphotransferasesystem: new frontiers 30 years later. Mol. Microbiol. 13:755–764.

33. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: alaboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold SpringHarbor, N.Y.

34. Schonert, S., T. Buder, and M. K. Dahl. 1998. Identification and enzymaticcharacterization of the maltose inducible �-glucosidase MalL (sucrase-iso-maltase-maltase) of Bacillus subtilis. J. Bacteriol. 180:2574–2578.

35. Schonert, S., T. Buder, and M. K. Dahl. 1999. Properties of maltose induc-ible �-glucosidase in Bacillus subtilis: evidence for its contribution in malto-dextrin utilization. Res. Microbiol. 150:167–177.

36. Shimotsu, H., and D. J. Henner. 1986. Construction of a single-copy inte-gration vector and its use in analysis of regulation of the trp operon ofBacillus subtilis. Gene 43:85–94.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

37. Silhavy, T. J., M. L. Berman, and L. W. Enquist. 1984. Experiments withgene fusions. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

38. Skarlatos, P., and M. K. Dahl. 1998. The glucose kinase of Bacillus subtilis.J. Bacteriol. 180:3222–3226.

39. Stenberg, E., B. Persson, H. Roos, and C. Urbaniczky. 1991. Quantitativedetermination of surface concentration of protein with surface plasmonresonance using radiolabeled proteins. J. Cell. Interface Sci. 143:513–526.

40. Stulke, J., I. Martin-Verstrarte, M. Zagorec, M. Rose, A. Klier, and G.Rapoport. 1997. Induction of the Bacillus subtilis ptsGHI operon by glucoseis controlled by a novel antiterminator, GlcT. Mol. Microbiol. 25:65–78.

41. SubtiList. 1999. Data release R15.1. In I. Mozer and A. Danchin (ed.),http://www.pasteur.fr./Bio/Subtilist.html. Institut Pasteur, Paris, France, http://genolist.pasteur.fr/SubtiList/.

42. Sutcliffe, I. C., and R. R. Russell. 1995. Lipoproteins of gram-positive bac-teria. J. Bacteriol. 177:1123–1128.

43. Sutrina, S. L., P. Reddy, M. H. Saier, Jr., and J. Reizer. 1990. The glucose

permease of Bacillus subtilis is a single polypeptide chain that functions toenergize the sucrose permease. J. Biol. Chem. 265:18581–18589.

44. Tangney, M., C. J. Buchanan, F. G. Priest, and W. J. Mitchell. 1992. Maltoseuptake and its regulation in Bacillus subtilis. FEMS Microbiol. Lett. 97:191–196.

45. Tatusova, T. A., and T. L. Madden. 1999. Blast2 sequences—a new tool forcomparing protein and nucleotide sequences. FEMS Microbiol. Lett. 174:247–250.

46. Thompson, J., A. Pikis, S. B. Ruvinov, B. Henrissat, H. Yamamoto, and J.Sekiguchi. 1998. The gene glvA of Bacillus subtilis 168 encodes a metal-requiring, NAD(H)-dependent 6-phospho-�-glucosidase. J. Biol. Chem. 273:27347–27356.

47. Yamamoto, H., M. Serizawa, J. Thompson, and J. Sekiguchi. 2001. Regula-tion of the glv operon in Bacillus subtilis: YfiA (GlvR) is a positive regulatorof the operon that is repressed through CcpA and cre. J. Bacteriol. 183:5110–5121.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

maltose and maltodextrin utilization by bacillus subtilis · (11, 34, 35, 44). in contrast to the...

Documents