Journal of Bacteriology, March 2004, p . 1287-1296, Vol . 186, No . 5

Transcriptional Regulation of Genes Encoding Arabinan-Degrading Enzymes in Bacillus subtilis

Maria Paiva Raposo,^1, José Manuel Inácio,¹ Luís Jaime Mota,^1, and Isabel de Sá-Nogueira^1,2*

Instituto de Tecnologia Química e Biológica, Universidade Nova de Lisboa, 2781-901 Oeiras,¹ Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, Quinta da Torre, 2829-516 Caparica, Portugal²

Received 25 August 2003/ Accepted 14 November 2003

 
Bacillus subtilis produces hemicellulases capable of releasing arabinosyl oligomers and arabinose from plant cell walls . Inthis work, we characterize the transcriptional regulation ofthree genes encoding arabinan-degrading enzymes that are clusteredwith genes encoding enzymes that further catabolize arabinose.The abfA gene comprised in the metabolic operon araABDLMNPQ-abfA and the xsa gene located 23 kb downstream most probably encode -L-arabinofuranosidases [EC 3.2.1.55] . Here, we show that theabnA gene, positioned immediately upstream from the metabolicoperon, encodes an endo--1,5-arabinanase [EC 3.2.1.99] . Furthermore,by in vivo RNA studies, we inferred that abnA and xsa are monocistronicand are transcribed from ^A-like promoters . Transcriptional fusionanalysis revealed that the expression of the three arabinasesis induced by arabinose and arabinan and is repressed by glucose.The levels of induction by arabinose and arabinan are higherduring early postexponential growth, suggesting a temporal regulation.Moreover, the induction mechanism of these genes is mediatedthrough negative control by the key regulator of arabinose metabolism,AraR . Thus, we analyzed AraR-DNA interactions by in vitro quantitativeDNase I footprinting and in vivo analysis of single-base-pairsubstitutions within the promoter regions of xsa and abnA . Theresults indicate that transcriptional repression of the abfAand xsa genes is achieved by a tightly controlled mechanismbut that the regulation of abnA is more flexible . We suggestthat the expression of genes encoding extracellular degradingenzymes of arabinose-containing polysaccharides, transport systems,and intracellular enzymes involved in further catabolism isregulated by a coordinate mechanism triggered by arabinose viaAraR.

 
Hemicellulose is the second-most abundant renewable biomasspolymer, next to cellulose . This fraction of plant cell wallscomprises a complex mixture of polysaccharides that includesxylans, arabinans, galactans, mannans, and glucans . Enzymesresponsible for degrading plant cell wall polysaccharides havemany agroindustrial applications, such as biobleaching of pulpsin the pulp and paper industry, improving digestibility of animalfeedstock, processing of flour in the baking industry, and clarifyingjuices [references 6, 26, and 27, and references therein] . Althoughmany hemicellulases have been purified and characterized fromboth fungi and bacteria, including mesophilic and thermophilicBacillus spp., knowledge concerning regulation at the molecularlevel of hemicellulolytic genes is scarce [reference 34 andreferences therein].

The saprophytic endospore-forming gram-positive bacterium Bacillus subtilis participates in enzymatic dissolution of plant cell walls in its natural reservoir, the soil . L-Arabinose is distributedin hemicelluloses and is present at high concentrations in arabinoxylans,arabinogalactans, and arabinan . The latter is composed of -1,5-linked L-arabinofuranosyl units, some of which are replaced with -1,3- and -1,2-linked chains of L-arabinofuranosyl residues [2] . Thetwo major enzymes that hydrolyze arabinan are -L-arabinofuranosidases [AFs] [EC 3.2.1.55] and endo--1,5-arabinanases [ABNs] [EC 3.2.1.99].AFs remove arabinose side chains, allowing ABNs to attack theglycosidic bonds of the arabinan backbone and releasing a mixtureof arabinooligosaccharides and L-arabinose [9] . B . subtilissynthesizes at least three enzymes, an ABN and two AFs, capableof releasing arabinosyl oligomers and L-arabinose from plantcell walls [12, 13, 28, 39].

Previous work by our group characterized the genes involvedin the utilization of L-arabinose that belong to the araABDLMNPQ-abfAoperon [32] and the divergently arranged araE and araR genes[31, 33], located in distinct regions of the B . subtilis chromosome.The first three genes of the L-arabinose metabolic operon, araA,araB, and araD, encode the enzymes required for the intracellularconversion of L-arabinose into D-xylulose 5-phosphate, whichis further catabolized through the pentose phosphate pathway[30] . The product of the araE gene is a permease, the main transporterof L-arabinose into the cell [33] . The araR gene encodes the regulatory protein of L-arabinose metabolism in B . subtilis,negatively controlling the expression from the L-arabinose-induciblepromoters of the ara genes [22, 23] . Additionally, the ara regulonis subjected to carbon catabolite repression by glucose and glycerol [11] . The last gene of the L-arabinose metabolic operon,abfA, and the xsa gene located 23 kb downstream from the operon[32, 40] [Fig . 1] most probably encode AFs belonging to the glycosyl hydrolase [GH] family G51 [see the Carbohydrate-Active Enzymes website [http://afmb.cnrs-mrs.fr/cazy/CAZY]] . The geneabnA, located immediately upstream from the metabolic operon[32, 40] [Fig. 1], most likely encodes an ABN grouped in theGH43 family [http://afmb.cnrs-mrs.fr/cazy/CAZY].

 
Our work focuses on the regulation of expression of the abfA, xsa, and abnA genes . Additionally, functional analysis of abnA revealed that this gene encodes an ABN . In vivo RNA studies demonstrated the monocistronic nature of xsa and abnA and allowed us to characterize their promoter regions . We show that the expression of the abfA, xsa, and abnA genes is positively controlledat the transcriptional level by arabinose and arabinan, repressedby glucose, and most likely subjected to temporal regulation.Moreover, in vivo and in vitro studies indicate that the transcriptionfactor AraR plays a major role in the control of the expressionof the arabinan-degrading genes . It is hypothesized that coordinateexpression of genes involved in the degradation of arabinose-containingpolysaccharides is triggered by arabinose and mediated by AraR.

 
Bacterial strains and growth conditions. The B . subtilis strains used in this study are listed in Table1. Escherichia coli DH5 [Gibco BRL] was used for routine molecularcloning work and was grown on Luria-Bertani [LB] medium [20].Ampicillin [75 µg ml^-1], X-Gal [5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside; 40 µg ml^-1], or IPTG [isopropyl-ß-D-thiogalactopyranoside; 1 mM] was added as appropriate . The B . subtilis strains were grown on LB medium [20] or C minimal medium [24], and chloramphenicol[5 µg ml^-1] or kanamycin [10 µg ml^-1] was addedas appropriate . Solid medium was made with LB or C minimal mediumcontaining 1.6% [wt/vol] Bacto Agar [Difco] . The abnA phenotypewas tested in C minimal medium [24] plates supplemented with0.4% [wt/vol] debranched arabinan [Megazyme] . The amyE phenotypewas tested by plating strains on tryptose blood agar base medium[Difco] containing 1% [wt/vol] potato starch, and after overnightincubation, the plates were flooded with a solution of 0.5%[wt/vol] I₂ and 5.0% [wt/vol] KI for the detection of starchhydrolysis . For the ß-galactosidase assays and RNApreparation, the B . subtilis strains were grown in liquid Cminimal medium supplemented with 1% [wt/vol] casein hydrolysate.When necessary, 0.4% [wt/vol] L-arabinose, 0.4% [wt/vol] arabinan[sugar beet; Megazyme], and 0.4% [wt/vol] glucose were addedto the cultures . The transformation of E . coli and B . subtilis strains was performed as previously described [23].

 
DNA manipulations and sequencing. DNA manipulations were carried out as described by Sambrooket al . [29] . Restriction enzymes were purchased from MBI Fermentasand New England Biolabs and used according to the manufacturer'sinstructions . DNA was eluted from agarose gels with a GENECLEANIIkit [Bio101] . DNA sequencing was performed with a Sequenaseversion 2.0 kit [USB] or an ABI PRIS BigDye terminator readyreaction cycle sequencing kit [Applied Biosystems] . PCR amplificationswere done by using high-fidelity native Pfu DNA polymerase [Stratagene],and the products were purified by using a QIAquick PCR purificationkit [QIAGEN].

Construction of plasmids and strains. The xsa and abnA promoter regions were amplified by PCR of chromosomalDNA of wild-type strain B . subtilis 168T⁺ with oligonucleotides ARA87 and ARA88 and oligonucleotides ARA85 and ARA86 [Table 2], respectively . Plasmid pRIT1 was obtained by cloning a 281-bpEcoRI-EcoRV DNA fragment, obtained from the PCR product bearingthe xsa promoter region, into pSN32 [22] digested with EcoRIand SmaI . An insertion of the same fragment into pJM783 [25]restricted with EcoRI and SmaI yielded plasmid pRIT3 . To constructplasmid pSN40, the PCR product bearing the abnA promoter regionwas digested with EcoRI and XmnI, and the resulting 292-bp product was inserted into pSN32 restricted with EcoRI and SmaI . Plasmid pMPR1 was obtained by ligation of a 300-bp EcoRI-BamHI DNA fragmentfrom pSN40, containing the abnA promoter region, to the pJM783EcoRI-BamHI sites . A PCR product bearing the abnA promoter region,amplified from chromosomal DNA of wild-type strain B . subtilis168T⁺ with oligonucleotides ARA85 and ARA89 [Tables 1 and 2] and digested with EcoRI-SacI, was inserted into the pBluescriptII SK[+] [Stratagene] EcoRI-SacI sites, to yield pMPR2 . To constructplasmid pSA1, a 687-bp DNA fragment from the araABDLMNPQ-abfAoperon carrying the 3' end of the araQ gene and the 5' end ofthe abfA gene [Fig. 1], obtained by EcoRI-HincII digestion of plasmid pTN13 [32], was ligated to the pJM783 [25] EcoRI-SmaIsites . The same DNA fragment inserted into pSN32 restrictedwith EcoRI and SmaI yielded pSA2 . Plasmid pSA3 was constructedby insertion of a 1,024-bp PCR product, amplified from chromosomalDNA of wild-type strain B . subtilis 168T⁺ with oligonucleotidesARA85 and ARA92 [Tables 1 and 2] and digested with EcoRI-BglII, into the pSN32 EcoRI-BamHI sites . The sequences of all the insertsobtained by PCR were confirmed by DNA sequencing.

 
To construct pZI15, which carries a single-base-pair substitutionin the AraR binding site OR_B1 [-38 GT], plasmid pZI13, a pBluescriptII SK[+] [Stratagene] derivative containing a 302-bp EcoRI-BamHIinsert from pSN40, was used as the target DNA for site-directedmutagenesis with the QuikChange kit [Stratagene] and the overlappingoligonucleotides ARA129 and ARA130 [Table 2] . An EcoRI-BamHI DNA fragment from the resulting plasmid, pZI14, was subcloned into those sites of pSN32 to yield pZI15 . To perform a single-nucleotide substitution in the AraR binding site OR_X2 [-27 GT], plasmidpZI17, a pBluescript II SK[+] [Stratagene] derivative containinga 291-bp EcoRI-BamHI insert from pRIT1, was used as target DNAfor site-directed mutagenesis and for the overlapping primersARA113 and ARA114 [Table 2], as described above . An EcoRI-BamHIDNA fragment from the resulting plasmid, pZI18, was ligatedto the pSN32 EcoRI-BamHI sites to obtain pZI19 . The single pointmutations were confirmed by DNA sequencing.

Linearized plasmid DNA from pSN40, pRIT1, pSA2, pSA3, pZI15,and pZI19 [Fig . 1], carrying the different promoter-lacZ transcriptionalfusions, was used to transform B . subtilis strains [Table 1],and the fusions were integrated into the chromosome via doublerecombination with the back and front sequences of the amyEgene . This event led to the disruption of the amyE locus andwas confirmed as described above . Plasmids pRIT3, pMPR1, andpSA1 [Fig . 1] were integrated into the host chromosome by meansof a single-crossover [Campbell-type] recombinational eventthat occurred in the region of homology [Table 1].

A PCR product containing the entire abnA gene, amplified from chromosomal DNA of wild-type strain B . subtilis 168T⁺ with oligonucleotidesARA32 and ARA85 [Tables 1 and 2], was digested with DdeI-BamHI,and this fragment bearing the 3' end of abnA was inserted inthe pSN32 SmaI-BamHI sites to yield pSN41 . Plasmid pSN42 was the result of the insertion of a 535-bp EcoRI-BamHI DNA fragmentfrom pSN41 into the pBluescript II SK[+] [Stratagene] EcoRI-BamHI sites . To construct plasmid pMPR3, a 1.5-kb SphI-SmaI DNA fragmentfrom pAH248 [31] containing a kanamycin resistance [Km^r] genewas inserted into the pSN42 SphI-SmaI site . By subcloning a1,626-bp SalI DNA fragment from pSN40 [see above] at the uniqueSalI site of pMPR3, pMPR4 was obtained . This plasmid was used,after linearization, to delete the abnA gene from the wild-typeB . subtilis 168T⁺ chromosome.

ß-Galactosidase assays. Strains of B . subtilis harboring the transcriptional lacZ fusionswere grown as described above . Samples of cell culture werecollected 2 h [exponential growth phase] and 4 h [late exponentialgrowth phase] after induction, and the level of ß-galactosidaseactivity was determined as previously described [32] . The ratio of ß-galactosidase activity from cultures grown inthe presence or absence of an inducer [arabinose or arabinan]was taken as a measure of AraR repression in each strain analyzed[regulation factor] . The ratio of ß-galactosidaseactivity from cultures grown in the presence or absence of glucosewas taken as a measure of glucose repression [glucose repressionfactor].

RNA preparation, Northern blot analysis, and primer extension analysis. B . subtilis strains were grown as described above, and cellswere harvested 2 h after induction . Total RNA was prepared by using an RNeasy kit [QIAGEN] according to the manufacturer's instructions . For Northern blot analysis, 10 µg of totalRNA was run in a 1.2% [wt/vol] agarose formaldehyde denaturinggel and transferred to positively charged Hybond-N⁺ [Amersham] nylon membranes according to standard procedures [29] . A size determination was done by using an RNA ladder [9 to 0.5 kb [New England Biolabs] or 6 to 0.2 kb [MBI Fermentas]] . A DNA fragmentof 763 bp used as an abnA probe was obtained by PCR amplification of chromosomal DNA with primers ARA85 and ARA89 followed byPstI digestion . PCR amplification with chromosomal DNA as atemplate and primers ARA87 and ARA91 [Table 2] yielded a DNA fragment that, after digestion with BclI, resulted in a 1.4-kb xsa DNA probe . The DNA probes were labeled with a Megaprime DNA labeling system [Amersham] and [-³²P]dCTP [3,000 Ci/mmol;Amersham].

Primer extension analysis was performed essentially as described by Sambrook et al . [29] . Primer ARA86, complementary to the abnA sequence [Table 2], and primer ARA90, complementary tothe xsa sequence [Table 2], were end labeled with [-³²P]ATP [3,000 Ci/mmol] by using T4 polynucleotide kinase [NEB] . A total of 2.5 mol of each labeled primer was mixed with 50 to 100 µgof RNA in separate experiments, denatured by heating to 85°Cfor 10 min, and annealed by incubation at 45°C overnight.The extension reaction was conducted for 2 h at 37°C byusing 50 U of avian Moloney murine leukemia virus reverse transcriptase[RevertAid; MBI Fermentas] . Analysis of the extended productswas carried out on 6% [wt/vol] polyacrylamide urea gels.

DNase I footprinting. The target DNA fragments from the xsa and abnA promoters wereobtained by PCR amplification with oligonucleotides ARA1 andARA72 [Table 2] using pRIT1 and pSN40 [see above] as templatesand yielding 339- and 350-bp DNA fragments, respectively . Thelabeling of the fragments and the DNase I footprinting experimentswere performed by using purified native AraR as previously describedby Mota et al . [22] . The apparent dissociation constant [K_app]for the different operators was determined as the total concentrationof AraR required for half-maximal site protection.

 
The abfA and xsa genes and functional analysis of the abnA gene The abfA and the xsa genes most probably encode AFs [EC 3.2.1.55].The amino acid sequence of AbfA displays a high level of identity[71%] to characterized AbfA from Geobacillus stearothermophilusT-6 [8], and AF activity was reported for the B . subtilis abfAgene product [40] . Xsa is highly homologous to characterizedAFs from Thermobacillus xylanilyticus AbfD3 [64% identity] [3], Clostridium cellulovorans ArfA [60% identity] [15], and Clostridiumstercorarium ArfB [56% identity] [42] . Based on primary aminoacid sequence analysis, the abnA gene most likely encodes anABN [EC 3.2.1.99], with 52% identity to a characterized thermostableABN from Bacillus thermodinitrificans [36] and 38% identityto ArbA from Cellvibrio japonicus [19] . Previously, Sakamotoet al . [28] reported the cloning of the gene ppc from B . subtilis strain IFO 3134 that encodes an ABN displaying 94% identityto the product of the abnA gene from B . subtilis 168T⁺ . However,it is unclear whether the arabinan-degrading activity measuredin that strain reflects the expression of the ppc gene [28].To characterize the function of the abnA gene, we constructedan insertion-deletion mutation in the abnA region [see Materialsand Methods] . The abnA-null mutant strain [IQB413] [Table 1]was able to grow on minimal medium plates supplemented withdebranched arabinan [see Materials and Methods], although moreslowly than the wild-type strain . However, the clear halo ofhydrolysis observed for the wild-type strain was absent in themutant [Fig . 2], indicating that the product of the abnA geneis an ABN.

 
abnA and xsa transcript analysis Previous work showed that the abfA gene encodes a 500-amino-acidpolypeptide and belongs to the araABDLMNPQ-abfA operon, a polycistronic transcriptional unit responsive to arabinose [32] [Fig. 1].The abnA and xsa genes encode 323- and 495-amino-acid polypeptides,respectively, and both potential open reading frame terminatorswere found downstream [32, 40] [Fig . 1] . To study transcription of the abnA and xsa genes, total RNA isolated from the wild-typestrain grown for 2 h in the absence or presence of arabinoseand arabinan [potential inducers] was annealed separately toDNA probes for abnA and xsa . Arabinose-inducible abnA- and xsa-specifictranscripts of about 0.9 and 1.6 kb, respectively, were detected[Fig . 3] . Weaker hybridization signals of the same size werealso visible with RNA from cells grown in the presence of arabinan.No hybridization signals were detected in the absence of sugars,suggesting that both arabinose and arabinan might function asinducers . The extent of both abnA and xsa mRNA signals closelymatched the expected sizes [1 and 1.6 kb, respectively] andconfirm their monocistronic nature . However, when arabinan wasused as an inducer, a weak high-molecular-weight RNA signalwas visible with the abnA DNA probe . One possibility for thisfinding is that this weak message corresponds to cotranscriptionwith upstream genes and/or the downstream araABDLMNPQ-abfA operon.

 
Expression of abnA and xsa is driven from ^A-like promoters Primer extension analysis of total RNA isolated from cells grownin the presence of arabinose showed that the 5' end of the abnAmessage corresponds to a G residue 117 bp upstream from the initiation TTG codon [Fig . 4 and 5A] . Centered at -35 and -10bp upstream from the abnA transcription start site are two sequences,TGTACA and TACAAT, respectively [Fig . 5A], that are similarto the consensus sequences for recognition by B . subtilis ^A-containing RNA polymerase [TTGACA-17bp-TATAAT] [10, 21] . By using the sametechnique, the apparent transcriptional start point of xsa wasassigned to a T nucleotide 100 bp upstream from the initiationATG codon [Fig . 4 and 5B] . The potential -35 and -10 regions [TTGACA-17bp-TATGGT] [Fig . 5B] closely match the ^A consensus[see above] . No abnA- or xsa-specific extension products wereseen with RNA extracted from cells grown in the absence of sugar,results parallel to those observed by Northern blot analysis.

 
abnA, xsa, and abfA transcription is responsive to arabinose and arabinan and is repressed by glucose. To study the functionality of the abnA and xsa promoters, theywere fused to the lacZ gene of E . coli and integrated at the amyE locus of the B . subtilis wild-type chromosome [strains IQB410 and IQB405, respectively] [Fig . 1 and Table 1] . Sincetranscription of the abfA gene is driven from a ^A-like promoterlocated upstream from the araA gene of the metabolic operon,expression of abfA was analyzed by the construction of a transcriptionallacZ fusion at the abfA locus [strain IQB450] [Fig . 1 and Table1] . The same DNA fragment, harboring the 3' end of the araQgene and the 5' end of the abfA gene, was also fused to lacZin a different vector and integrated at the amyE locus of the B . subtilis wild-type chromosome [strain IQB451] [Fig . 1 andTable 1] . As expected, this strain did not show promoter activityin the experiments described below [data not shown] . The levelsof accumulated ß-galactosidase activity of the strainsbearing the various transcriptional fusions were examined in the absence of sugars and in the presence of arabinose, arabinan, and arabinose plus glucose . Samples were collected 2 and 4 h after induction [t₂ and t₄, respectively], which correspondsto the exponential growth phase and early postexponential phase,respectively . In the presence of arabinose, expression fromthe abfA'-lacZ, xsa'-lacZ, and abnA'-lacZ fusions [strains IQB450,IQB405, and IQB410] [Table 3] increased during exponential growth about 96-, 24-, and 3-fold, respectively, and a small increment in expression was observed at early postexponential phase [t₄] [Table 3] . The presence of arabinan also stimulated expressionfrom the same abfA'-lacZ, xsa'-lacZ, and abnA'-lacZ fusions,about five-, three-, and fivefold, respectively [t₂] [Table3] . However, these lower-level responses, compared to thoseobserved in the presence of arabinose, increased more dramaticallyat the end of the exponential growth phase [t₄] [Table 3] . Theseresults confirm the Northern blot analysis indicating that both arabinose and arabinan function as inducers of the abnA, xsa, and abfA genes . The possibility that other regulatory regions upstream from the DNA fragments used to construct the xsa'-lacZ and abnA'-lacZ fusions integrated at the amyE locus might influenceexpression from xsa and abnA was examined . We constructed transcriptionallacZ fusions at the xsa and abnA loci [strains IQB407 and IQB412][Table 1], and the regulation factor calculated for these strains was similar to that observed for strains IQB405 and IQB410, respectively [data not shown].

 
The addition of glucose caused a 21.8-fold repression of theabfA'-lacZ fusion expression, a 19.3-fold repression of xsa'-lacZ expression, and a 6.5-fold repression of abnA'-lacZ expression [t₂] [Table 3] . No significant differences in the levels ofglucose repression were observed during early postexponentialphase [t₄] [Table 3] . Previously, it has been shown that glucose repression of the araABDLMNPQ-abfA metabolic operon is mainly regulated by CcpA via binding to two catabolite responsive elements [CREs], one located between the promoter region of the operon and the araA gene and one located 2 kb downstream within the araB gene [11] . Mutagenesis studies of B . subtilis revealedthe CRE consensus sequence TGWAARCGYTWNCW [W = A or T, R = A or G, Y = C or T, N = any base] [14, 18, 38, 41] . Based on sequence homology studies, potential CRE sequences were detected in the promoter region of the abnA and xsa genes . The CRE for abnA is positioned between the promoter and the TTG initiation codon [⁺⁷⁹TGTAAGCGCTTTCT⁺⁹²] [Fig . 5A], and the CRE for xsa overlapsthe transcription start site of the gene [⁺¹TAAAAGCGCTTACA⁺¹⁴][Fig . 5B].

AraR plays a major role in transcriptional control of abnA, xsa, and abfA. In previous work, a search of the B . subtilis database was done[SubtiList] for sequences similar to the AraR consensus operator[ATTTGTACGTACAAAT] [22], the key regulator of arabinose utilization.Among the potential AraR binding sites detected, two were locatedin the promoter region of the xsa and abnA genes [22] . Thus, we investigated the expression from the abfA'-lacZ, xsA'-lacZ, and abnA'-lacZ fusions in an araR-null mutant background . Thelevels of accumulated ß-galactosidase activity of the resulting strains [IQB453, IQB406, and IQB411, respectively] were examined as described above, and the results are shownin Table 3 . The degree of AraR repression [regulation factor] was determined indirectly by the ratio of the values obtained in induced and noninduced cultures . Disruption of the araR gene led to a total derepression of the expression from the abfA'-lacZ, xsA'-lacZ, and abnA'-lacZ fusions [strains IQB453, IQB406, andIQB411, respectively] in comparison to that from the wild type.These results suggest that AraR plays a major role in the transcriptionalcontrol of the abnA, xsa, and abfA genes . Interestingly, wefound another sequence within the abnA coding region [AAACAGTACGTACAAAA,at a position +831 relative to the transcription start site]similar to that of the AraR consensus operator [see above].We constructed an abnA'-lacZ fusion bearing this element todetermine its involvement in the regulation of abnA expression[Fig . 1] . The transcriptional fusion was integrated in a singlecopy at the amyE locus of the wild-type and araR-null mutantbackgrounds, and the resulting strains IQB448 and IQB449 [Table1], respectively, were analyzed as described above . In the presenceof arabinose and arabinan, strain IQB448 showed a twofold increasein the level of regulation by AraR relative to that of strainIQB410 [Table 3], indicating that this putative operator might contribute to the regulation of abnA expression at the transcriptionallevel.

Binding of AraR to the promoter region of xsa and abnA genes. The ability of AraR to bind to the promoter region of xsa andabnA genes was determined by quantitative DNase I footprintingwith DNA fragments from the plasmids harboring the transcriptionalfusions as targets [see Materials and Methods] . Two AraR bindingsites were detected by DNase I footprinting in the xsa promoterregion [Fig . 6] . In the xsa coding strand, AraR protects theregions between positions -72 and -52 [OR_X1] and between positions-32 and -11 [OR_X2]; similar sequences are protected in the noncodingstrand [Fig. 6] . A pattern of DNase I-enhanced and -diminished cleavage was observed between OR_X1 and OR_X2 [Fig. 5 and 6] thatresembled the DNase I footprintings of the araABDLMNPQ-abfAoperon and araE [22] . AraR binding to the two in-phase operators of the xsa metabolic operon and of the transport gene promoter regions thus seems to occur in similar ways, producing in both cases a distortion of the DNA helix . Binding of AraR to OR_X1 and OR_X2 was also inhibited by the presence of arabinose [Fig. 6], indicating that arabinose is the effector which modulatesAraR binding to DNA . The K_app for each individual binding sitewas determined as the repressor concentration at which half-maximalsite occupancy was observed [Fig . 6] . Although these valueswere calculated for a single experiment, the relative affinityof AraR to OR_X1 and OR_X2 is comparable to that observed forthe AraR binding sites in the promoter regions of the metabolicoperon and the araE gene [22] . Binding of AraR to the putativeOR_B1 operator identified by sequence analysis within the abnApromoter was not detected in the same range of protein concentrationsused for the xsa promoter [data not shown].

 
To assess the functionality of the AraR operators in vivo, we introduced the same single-base-pair substitution in both OR_X2 and the putative OR_B1 [Fig . 5] . This mutation in a highly conservedposition of the AraR target sequence [22] was designed to preventthe binding of AraR to OR_X2 in the xsa promoter and to OR_B1in the abnA promoter . The two mutant promoters were fused tothe lacZ gene of E . coli and were analyzed in a B . subtiliswild-type background as described above . Both mutations resultedin a loss of regulation by AraR [Table 4] . These results indicate that just one base pair change in one of the two operators [OR_X1 and OR_X2] is sufficient to abolish repression and suggest cooperativebinding to the two in-phase operators in the xsa promoter region.Furthermore, OR_B1 is an active cis-acting element for the regulationof the abnA promoter.

 
In B . subtilis, the genes encoding hemicellulolytic enzymes are clustered with genes encoding enzymes that further catabolize these carbon sources [reference 35 and references therein]. In this study, we analyzed the mechanisms that regulate the expression of three arabinan-degrading genes, abfA, xsa, and abnA, which are assembled with genes involved in arabinose catabolism[Fig . 1] . The abfA and xsa genes most probably encode AFs [EC3.2.1.55] belonging to the GH51 family [http://afmb.cnrs-mrs.fr/cazy/CAZY]. Although the exact subcellular localization of Xsa and AbfAis unknown, these enzymes are believed to be intracellular [1, 37] . Nonetheless, AFs from G . stearothermophilus and B . subtiliswere purified from supernatants of cultures at the end of thestationary phase [8, 13, 39] . Based on primary amino acid sequence analysis, the abnA gene most likely encodes an ABN [EC 3.2.1.99] grouped in the GH43 family [http://afmb.cnrs-mrs.fr/cazy/CAZY], which was shown to be extracellular [1] . Here, we determinedthe function of the abnA gene by showing that an insertion-deletionmutation in this gene led to a loss of arabinanase activity[Fig . 2] . However, the abnA-null mutant is still able to growon minimal medium with arabinan as the sole carbon source . Thisresult might be due to the activity of the yxiA gene product,a hypothetical arabinanase displaying 27% identity to AbnA [17].

During exponential growth, the expression from abfA'-lacZ, xsA'-lacZ,and abnA'-lacZ transcriptional fusions revealed that abfA andxsa are strongly induced by arabinose, whereas the inductionof abnA is weak [Table 3] . The levels of induction in responseto arabinan are very similar for the three genes . However, thelevel of arabinan-mediated induction of abfA and xsa is considerably lower than that observed with arabinose, while the induction levels of abnA are similar in both cases . This result is most probably due to different kinetics of induction by arabinoseamong the promoters [see below] . A disruption of the araR gene abolished the regulation of the three arabinan-degrading genes, suggesting that AraR plays a major role in the transcriptional control of these genes . Previously, it has been reported thatthe addition of arabinose to the medium causes an immediatecessation of growth in an araR-null mutant background, whichis probably due to an intracellular increase of arabinose andconsequently an increase of the metabolic sugar phosphate intermediatesthat are toxic to the cell [31] . This effect is not observed when arabinan is added to the cultures, indicating that the reduced levels of expression observed in the presence of thiscarbon source might be due to lower levels of intracellulararabinose during the exponential growth phase . Since the arabinan-mediatedinduction of the three promoters is strictly dependent on AraR,and arabinan itself is not the molecular inducer [data not shown],the observed arabinan induction should be the result of lowlevels of intracellular arabinose . As mentioned above, the arabinose-mediated induction of xsa and abfA promoters is higher than that of the abnA promoter, while the arabinan induction levels are identical in the three cases . Therefore, a likely explanation for these results is that a lower level of arabinose is enough to rapidlyand fully induce the abnA promoter but not the abfA and xsa promoters . Taken together, these observations indicate thatthe AraR protein exerts a tight control of arabinose- and arabinan-inducible transcription of the abfA and xsa genes but that repression of the abnA gene is more flexible.

One of the mechanisms involved in the synthesis of many degradative enzymes in B . subtilis is mediated by transition phase regulation [7] . Accordingly, when the transcriptional fusions are analyzedat early postexponential phase, the levels of expression of abfA, xsa, and abnA in response to arabinose and arabinan arehigher than those observed during exponential growth [Table3] . This finding suggests that the arabinan-degrading genesare subject to temporal regulation . In agreement with our observations,by extracellular proteome analysis, Antelmann et al . [1] detectedAbnA at higher levels during stationary phase . Interestingly,the temporal differences among the induced levels of expressionthat we noticed are more striking in the case of arabinan thanfor arabinose . This finding suggests that arabinan, or one ofits degradation products, may play an important role in thisprocess . However, it may also be due simply to a higher amountof intracellular free arabinose as a result of a higher levelof AbnA during the stationary phase . The mechanisms underlyingthe temporal regulation of the arabinan-degrading genes areunknown, but studies are currently in progress to address this question . Nonetheless, the data presented here indicate that the arabinose repressor, AraR, also plays a crucial role inthe control of abfA, xsa, and abnA expression during early postexponentialphase . Additionally, glucose repression previously characterizedfor the araABDLMNPQ-abfA operon [11] seems to be mediated bya similar mechanism in the case of the abnA and xsa genes.

The DNase I footprinting analysis of the xsa promoter suggests that this gene should be regulated by AraR by a mechanism similar to that proposed for the araABDLMNPQ-abfA operon and the araEtransport gene [22, 23] . As in these two cases, the AraR bindingsites are separated by approximately four turns of the DNA helix[41 bp], and a pattern of DNase I hypersensibility was observedin the interoperator region [Fig . 5 and 6] . Furthermore, a single-base-pairchange in one of the two operators [OR_X1 and OR_X2] is sufficientto abolish AraR repression in vivo . By analogy, this findingsuggests that the binding of AraR to the operators in the xsapromoter is cooperative, resulting in a distortion of the DNAhelix that may be in the form of a small DNA loop . In contrast,noncooperative binding of AraR to one operator in the promoterregion of the abnA gene, and possibly to a second operator locateddownstream within the abnA coding region [Tables 3 and 4], isless effective, as observed in the case of autoregulation ofaraR expression [22, 23] . The fact that we could not detectin vitro binding of AraR to the promoter region of abnA mightindicate a low affinity of the regulator to its operator site.However, one cannot exclude the possibility of additional trans-actingfactors involved in the regulation of abnA expression whichmay contribute to AraR binding or which may directly controlabnA expression . Together, these observations might explainthe different mode of response to arabinose and arabinan ofabnA expression compared to those of xsa and abfA during exponentialgrowth [Table 3], which may reflect distinct physiological requirements. A tight control of the xsa and abfA genes ensures the expressionof these intracellular enzymes solely when the arabinose induceris present . On the other hand, a weak control of abnA allowsfor a low level of basal transcription of this extracellular enzyme.

Bacilli secrete a vast number of polysaccharide backbone-degrading enzymes, which produce relatively large oligosaccharide products. These units, disaccharides, trisaccharides, and oligosaccharides, enter the cell by specific transport systems and are furtherbroken down by intracellular enzymes [4, 35] . We have shownthat in B . subtilis, arabinan is degraded by at least one extracellularhemicellulase, AbnA . The resulting products, arabinose, arabinobiose,arabinotriose, and arabinooligosaccharides, are transportedby different systems . Arabinose enters the cell mainly throughthe AraE permease [33], and the uptake of arabinose oligomersmost likely occurs via AraNPQ, an ABC-type transporter [32].These latter products might be further digested intracellularlyby AbfA and Xsa . Interestingly, the AraE permease is also responsiblefor the transport of xylose and galactose into the cell [16]. These three structurally different sugars, arabinose, xylose, and galactose, are frequently found associated in hemicelluloses. Furthermore, xylan- and xylose-utilizing genes are controlledby the XylR repressor, and no regulatory protein specificallycontrolling galactose utilization has been found [reference35 and references therein] . These observations suggest a coordinated expression, triggered by arabinose and mediated by AraR, ofgenes encoding enzymes responsible for extracellular degradationof arabinose-containing polysaccharides and transport systemsand intracellular catabolism of arabinose, xylose, and galactose. Concerted regulation of the production of all pectin side-chain-cleaving enzymes in response to arabinose seems likely to occur in Aspergillus spp . [5] . Thus, it will be interesting to know how this regulatorycircuitry in response to arabinose is disseminated among hemicellulase-producingmicroorganisms.

 
We thank Rita Teodoro and Susana S . Silva for constructing some plasmids and strains.

This work was supported by grant no . POCTI/AGR/36212/00 from Fundação para a Ciência e Tecnologia andFEDER.

 
* Corresponding author . Mailing address: Instituto de Tecnologia Química e Biológica, Universidade Nova de Lisboa, Avenida da República, Apartado 127, 2781-901 Oeiras, Portugal . Phone: [351] 21-4469524 . Fax: [351] 21-4411277 . E-mail: sanoguei@itqb.unl.pt .

Present address: Department of Biochemistry and Molecular Biology, University of Miami, School of Medicine, Miami, FL 33136.

Present address: Biozentrum der Universität Basel, 50-70CH-Basel, Switzerland.

1. Antelmann, H., H . Tjalsma, B . Voigt, S . Ohlmeier, S . Bron, J . M . van Dijl, and M . Hecker. 2001 . A proteomic view on genome-based signal peptide predictions . Genome Res . 11:1484-1502 .
2. Beldman, G., H . A . Schols, S . M . Piston, M . J . F . Searl-van Leewen, and A . G . J . Vorangen. 1997 . Arabinans and arabinan degrading enzymes . Adv . Macromol . Carbohydr . Res . 1:1-64.
3. Debeche, T., N . Cummings, I . Connerton, P . Debeire, and M . J . O'Donohue. 2000 . Genetic and biochemical characterization of a highly thermostable -L-arabinofuranosidase from Thermobacillus xylanilyticus. Appl . Environ . Microbiol . 66:1734-1736 .
4. Deutscher, J., A . Galinier, and I . Martin-Verstraete. 2002 . Carbohydrate uptake and metabolism, p . 129-150 . In A . L . Sonenshein, J . A . Hoch, and R . Losick [ed.], Bacillus subtilis and its closest relatives: from genes to cells . ASM Press, Washington, D.C.
5. de Vries, R . P. 2003 . Regulation of Aspergillus genes encoding plant cell wall polysaccharide-degrading enzymes; relevance for industrial production . Appl . Microbiol . Biotechnol . 61:10-20.
6. de Vries, R . P., and J . Visser. 2001 . Aspergillus enzymes involved in degradation of plant cell wall polysaccharides . Microbiol . Mol . Biol . Rev . 65:497-522 .
7. Ferrari, E., A . S . Jarnagin, and B . F . Schmidt. 1993 . Commercial production of extracellular enzymes, p . 917-937 . In A . L . Sonenshein, J . A . Hoch, and R . Losick [ed.], Bacillus subtilis and other gram-positive bacteria: biochemistry, physiology, and molecular genetics . American Society for Microbiology, Washington, D.C.
8. Gilead, S., and Y . Shoham. 1995 . Purification and characterization of an -L-arabinofuranosidase from Bacillus stearothermophilus T-6 . Appl . Environ . Microbiol . 61:170-174.
9. Hazlewood, G . P., and H . J . Gilbert. 1998 . Structure and function analysis of Pseudomonas plant cell wall hydrolases . Prog . Nucleic Acid Res . Mol . Biol . 61:211-241.
10. Helmann, J . D., and C . P . Moran, Jr. 2002 . RNA polymerase and sigma factors, p . 289-312 . In A . L . Sonenshein, J . A . Hoch, and R . Losick [ed.], Bacillus subtilis and its closest relatives: from genes to cells . ASM Press, Washington, D.C.
11. Inácio, J . M., C . Costa, and I . de Sá-Nogueira. 2003 . Distinct molecular mechanisms involved in carbon catabolite repression of the arabinose regulon in Bacillus subtilis. Microbiology 149:2345-2355 .
12. Kaji, A., and T . Saheki. 1975 . Endo-arabanase from Bacillus subtilis F-11 . Biochim . Biophys . Acta 410:354-360.
13. Kaneko, S., M . Sano, and I . Kusakabe. 1994 . Purification and some properties of -L-arabinofuranosidase from Bacillus subtilis 3-6 . Appl . Environ . Microbiol . 60:3425-3428.
14. Kim, J . H., and G . H . Chambliss. 1997 . Contacts between Bacillus subtilis catabolite regulatory protein CcpA and amyO target site . Nucleic Acids Res . 25:3490-3496 .
15. Kosugi, A., K . Murashima, and R . H . Doi. 2002 . Characterization of two noncellulosomal subunits, ArfA and BgaA, from Clostridium cellulovorans that cooperate with the cellulosome in plant cell wall degradation . J . Bacteriol . 184:6859-6865 .
16. Krispin, O., and R . Allmansberger. 1998 . The Bacillus subtilis AraE protein displays a broad substrate specificity for several different sugars . J . Bacteriol . 180:3250-3252.
17. Kunst, F., N . Ogasawara, I . Moszer, A . M . Albertini, G . Alloni, V . Azevedo, et al. 1997 . The complete genome sequence of the Gram-positive bacterium Bacillus subtilis. Nature 390:249-256.
18. Martin-Verstraete, I., J . Stülke, A . Klier, and G . Rapoport. 1995 . Two different mechanisms mediate catabolite repression of the Bacillus subtilis levanase operon . J . Bacteriol . 177:6919-6927.
19. McKie, V . A., G . W . Black, S . J . Millward-Sadler, G . P . Hazlewood, J . I . Laurie, and H . J . Gilbert. 1997 . Arabinanase A from Pseudomonas fluorescens subsp . cellulosa exhibits both an endo- and an exo- mode of action . Biochem . J . 323:547-555.
20. Miller, J . H. 1972 . Experiments in molecular genetics . Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
21. Moran, C . P., Jr., N . Lang, S . F . J . LeGrice, G . Lee, M . Stephens, A . L . Sonenshein, J . Pero, and R . Losick. 1982 . Nucleotide sequences that signal the initiation of transcription in Bacillus subtilis. Mol . Gen . Genet . 186:339-346.
22. Mota, L . J., P . Tavares, and I . Sá-Nogueira. 1999 . Mode of action of AraR, the key regulator of L-arabinose metabolism in Bacillus subtilis. Mol . Microbiol . 33:476-489.
23. Mota, L . J., L . M . Sarmento, and I . de Sá-Nogueira. 2001 . Control of the arabinose regulon in Bacillus subtilis by AraR in vivo: crucial roles of operators, cooperativity, and DNA looping . J . Bacteriol . 183:4190-4201 .
24. Pascal, M., F . Kunst, J . A . Lepesant, and R . Dedonder. 1971 . Characterization of two sucrase activities in Bacillus subtilis Marburg . Biochimie 53:1059-1066.
25. Perego, M. 1993 . Integrational vectors for genetic manipulation in Bacillus subtilis, p . 615-624 . In A . L . Sonenshein, J . A . Hoch, and R . Losick [ed.], Bacillus subtilis and other gram-positive bacteria: biochemistry, physiology, and molecular genetics . American Society for Microbiology, Washington, D.C.
26. Saha, B . C. 2000. -L-Arabinofuranosidases: biochemistry, molecular biology and application in biotechnology . Biotechnol . Adv . 18:403-423.
27. Saha, B . C. 2003 . Hemicellulose bioconversion . J . Ind . Microbiol . Biotechnol . 5:279-291.
28. Sakamoto, T., M . Yamada, H . Kawasaki, and T . Sakai. 1997 . Molecular cloning and nucleotide sequence of an endo-1,5--L-arabinase gene from Bacillus subtilis. Eur . J . Biochem . 245:708-714.
29. Sambrook, J., E . F . Fritsch, and T . Maniatis. 1989 . Molecular cloning: a laboratory manual, 2nd ed . Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
30. Sá-Nogueira, I., and H . de Lencastre. 1989 . Cloning and characterization of araA, araB, and araD, the structural genes for L-arabinose utilization in Bacillus subtilis. J . Bacteriol . 171:4088-4091.
31. Sá-Nogueira, I., and L . J . Mota. 1997 . Negative regulation of L-arabinose metabolism in Bacillus subtilis: characterization of the araR [araC] gene . J . Bacteriol . 179:1598-1608.
32. Sá-Nogueira, I., T . V . Nogueira, S . Soares, and H . Lencastre. 1997 . The L-arabinose [ara] operon of Bacillus subtilis: nucleotide sequence, genetic organization and expression . Microbiology 143:957-969.
33. Sá-Nogueira, I., and S . S . Ramos. 1997 . Cloning, functional analysis, and transcriptional regulation of the Bacillus subtilis araE gene involved in L-arabinose utilization . J . Bacteriol . 179:7705-7711.
34. Shallom, D., and Y . Shoham. 2003 . Microbial hemicellulases . Curr . Opin . Microbiol . 3:219-228.
35. Stülke, J., and W . Hillen. 2000 . Regulation of carbon catabolism in Bacillus species . Annu . Rev . Microbiol . 54:849-880.
36. Takao, M., A . Yamaguchi, K . Yoshikawa, T . Terashita, and T . Sakai. 2002 . Molecular cloning of the gene encoding thermostable endo-1,5-alpha-L-arabinase of Bacillus thermodenitrificans TS-3 and its expression in Bacillus subtilis. Biosci . Biotechnol . Biochem . 66:430-433.
37. Tjalsma, H., A . Bolhuis, J . D . Jongbloed, S . Bron, and J . M . van Dijl. 2000 . Signal peptide-dependent protein transport in Bacillus subtilis: a genome-based survey of the secretome . Microbiol . Mol . Biol . Rev . 64:515-547 .
38. Weickert, M . J., and G . H . Chambliss. 1990 . Site-directed mutagenesis of a catabolic repression operator sequence in Bacillus subtilis. Proc . Natl . Acad . Sci . USA 87:6238-6242.
39. Weinstein, L., and P . Albersheim. 1979 . Structure of plant cell walls . IX . Purification and partial purification of a wall-degrading endoarabanase and an arabinosidase from Bacillus subtilis. Plant Physiol . 63:425-432.
40. Wipat, A., N . Carter, S . C . Brignell, B . J . Guy, K . Piper, J . Sanders, P . T . Emmerson, and C . R . Harwood. 1996 . The dnaB-pheA [256 degrees-240 degrees] region of the Bacillus subtilis chromosome containing genes responsible for stress responses, the utilization of plant cell walls and primary metabolism . Microbiology 142:3067-3078.
41. Zalieckas, J . M., L . V . Wray, Jr., and S . H . Fisher. 1998 . Expression of the Bacillus subtilis acsA gene: position and sequence context affect cre-mediated carbon catabolite repression . J . Bacteriol . 180:6649-6654 .
42. Zverlov, V . V., W . Liebl, M . Bachleitner, and W . H . Schwarz. 1998 . Nucleotide sequence of arfB of Clostridium stercorarium, and prediction of catalytic residues of -L-arabinofuranosidases based on local similarity with several families of glycosyl hydrolases . FEMS Microbiol . Lett . 164:337-343.

Free Online Full-text Article

What Is Prokaryote?, What Is Bioengineering?, What Is Amino Acid?, What Is Molecular Microbiology?, What Is Salmonella?, e, Microbe, o, Bacterium, c, Microorganisms, r, Bacteriology, c, Microorganism, c, Phage, n, Bacillus, c, Escherichia coli, n, Lactobacillus, i, Schizosaccharomyces, n, Streptococci, i, Salmonella typhimurium, r, Microorganisms, s, Bacillus subtilis, i, Campylobacter, a, Rhizobacterium, r, Cell cultures, e, Salmonella typhimurium, a, Bacillus, r, Campylobacter, c, Staphylococcus, a, Antibiotics, e, Enterobacters, e, Bacteroides, n, Streptococcal, s, Ps. putida




 

© 2005 Transgalactic Ltd (manufacturer of Bioscreen C software) | Privacy Statement | P.O. Box 1393, 00101 Helsinki, Finland, phone: +358 9 85172920, fax: +358 9 8749481, e-mail: microbiology@bionewsonline.com
 

Last modified: May 25, 2005