HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Inácio, J. M. Articles by de Sá-Nogueira, I. Search for Related Content PubMed PubMed Citation Articles by Inácio, J. M. Articles by de Sá-Nogueira, I. Agricola Articles by Inácio, J. M. Articles by de Sá-Nogueira, I.

Two distinct arabinofuranosidases contribute to arabino-oligosaccharide degradation in Bacillus subtilis

¹ Laboratory of Microbial Genetics, Instituto de Tecnologia Química e Biológica, Universidade Nova de Lisboa, Av. da República, Apt 127, 2781-901 Oeiras, Portugal
² Departamento de CiÁncias da Vida, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, Quinta da Torre, 2829-516 Caparica, Portugal

Correspondence
Isabel de Sá-Nogueira
sanoguei{at}itqb.unl.pt

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

 
Bacillus subtilis produces -L-arabinofuranosidases (EC 3.2.1.55; AFs) capable of releasing arabinosyl oligomers and L-arabinose from plant cell walls. Here, we show by insertion-deletion mutational analysis that genes abfA and xsa(asd), herein renamed abf2, encode AFs responsible for the majority of the intracellular AF activity in B. subtilis. Both enzyme activities were shown to be cytosolic and functional studies indicated that arabino-oligomers are natural substrates for the AFs. The products of the two genes were overproduced in Escherichia coli, purified and characterized. The molecular mass of the purified AbfA and Abf2 was about 58 kDa and 57 kDa, respectively. However, native PAGE gradient gel analysis and cross-linking assays detected higher-order structures (>250 kDa), suggesting a multimeric organization of both enzymes. Kinetic experiments at 37 °C, with p-nitrophenyl--L-arabinofuranoside as substrate, gave an apparent K_m of 0.498 mM and 0.421 mM, and V_max of 317 U mg^–1 and 311 U mg^–1 for AbfA and Abf2, respectively. The two enzymes displayed maximum activity at 50 °C and 60 °C, respectively, and both proteins were most active at pH 8.0. AbfA and Abf2 both belong to family 51 of the glycoside hydrolases but have different substrate specificity. AbfA acts preferentially on (15) linkages of linear -1,5-L-arabinan and -1,5-linked arabino-oligomers, and is much less effective on branched sugar beet arabinan and arabinoxylan and arabinogalactan. In contrast, Abf2 is most active on (12) and (13) linkages of branched arabinan and arabinoxylan, suggesting a concerted contribution of these enzymes to optimal utilization of arabinose-containing polysaccharides by B. subtilis.

The GenBank/EMBL/DDBJ accession number for the nucleotide sequence of the abf2 gene reported in this paper is EU073712.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

 
-L-Arabinofuranosidases (-L-arabinofuranoside arabinofuranohydrolases, EC 3.2.1.55; AFs) catalyse the hydrolysis of terminal -1,2-, -1,3- or -1,5-L-arabinofuranoside bonds in different hemicellulosic homopolysaccharides (branched and debranched arabinans) and heteropolysaccharides (arabinoxylans, arabinogalactans, etc.). The complete degradation of hemicellulose and arabinose-containing polysaccharides requires the concerted action of many different enzymes, and AFs play a key role in this synergistic process. To hydrolyse arabinan, in addition to AFs that act in an exo- fashion and cleave arabinose side chains, -1,5-arabinanases (EC 3.2.1.99; ABNs) act in an endo- fashion, i.e. they attack the -1,5-linked L-arabinofuranose backbone of the homopolysaccharide (Beldman et al., 1997). Arabinan-degrading enzymes have several biotechnological applications: improvement of wine flavours, pulp treatment, juice clarification, quality of animal feedstock, production of important medicinal compounds, and production of bioethanol (Numan & Bhosle, 2006; Saha, 2000; Shallom & Shoham, 2003).

Bacillus subtilis, an aerobic, mesophilic, endospore-forming bacterium, produces a vast number of polysaccharolytic enzymes, including ABNs and AFs, capable of releasing arabinosyl oligomers and L-arabinose from plant cell walls (Kaji & Saheki, 1975; Kaneko et al., 1994; Leal & Sá-Nogueira, 2004; Sakai & Sakamoto, 1990; Weinstein & Albersheim, 1979). In a previous study, we characterized the transcriptional regulation of three B. subtilis genes involved in arabinan degradation (abnA, xsa, and abfA), which respond to arabinose (Raposo et al., 2004). The abnA gene was shown to encode an extracellular endo-ABN that hydrolyses sugar beet arabinan and linear -1,5-L-arabinan (Leal & Sá-Nogueira, 2004; Raposo et al., 2004). To further characterize this B. subtilis hemicellulolytic system we have now analysed the function of the abfA and xsa genes by mutagenic studies and determination of the capacity of the different mutants to utilize either sugar beet arabinan or -1,5-linked arabino-oligosaccharides. The determination of the subcellular localization of their products indicates that the two enzymes are scattered throughout the cytosol and do not localize in particular foci. The two potential AFs were expressed in Escherichia coli and the biochemical properties of the recombinant proteins determined. Both enzymes exhibited AF activity; however, they displayed different substrate specificity. Since the product of xsa does not show β-xylosidase activity we propose to rename this gene abf2.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

 
Substrates.
Sugar beet arabinan, debranched arabinan (linear -1,5-L-arabinan, purity 95 %), -1,5-linked arabino-oligosaccharides (arabinobiose, arabinotriose, arabinotetraose, purity 95 %), and wheat arabinoxylan were purchased from Megazyme International Ireland, and larch wood arabinogalactan, p-nitrophenyl--L-arabinofuranoside (pNPAf), p-nitrophenyl--L-arabinopyranoside (pNPAp), and p-nitrophenyl-β-D-xylopyranoside (pNPXp) from Sigma.

Bacterial strains and growth conditions.
The B. subtilis strains used in this study are listed in Table 1. E. coli DH5 (Gibco-BRL) was used for routine molecular cloning work and E. coli BL21(DE3) pLysS (Studier et al., 1990) as the host for production of native and recombinant AbfA and Abf2. E. coli strains were grown in Luria–Bertani (Miller, 1972) medium and kanamycin (20 µg ml^–1), chloramphenicol (25 µg ml^–1) or IPTG were added as appropriate. For growth kinetics experiments, overnight cultures of the different B. subtilis strains, grown in minimal medium C (Pascal et al., 1971) supplemented with L-tryptophan (100 µg ml^–1), potassium glutamate (8 g l^–1) and potassium succinate (6 g l^–1) (CSK medium; Débarbouillé et al., 1990), were washed and resuspended in 200 µl (to an OD₆₀₀ of 0.05) of the same medium without potassium succinate (CE-minimal medium), supplemented with one of the various carbon sources: arabinose (26 mM), glucose (22 mM), arabinobiose (13 mM), arabinotriose (8.6 mM), arabinotetraose (6.5 mM) and sugar beet arabinan (0.26 mM). The cultures were incubated in Multiple Well Plate 96-Well Round Bottom with Lid plates (Sarstedt), at 37 °C and 180 r.p.m., and growth was followed by periodic reading in a Molecular Devices Spectra max Plus Microplate Spectrophotometer at 600 nm. Transformation of E. coli and B. subtilis strains was performed as previously described (Inácio et al., 2003).

Construction of plasmids and strains.
To construct the B. subtilis integrative plasmid pTL2, encoding a fusion of the C-terminus of AbfA to a His₆-tag, the abfA C-terminal-encoding region was amplified by PCR of chromosomal DNA from B. subtilis 168T⁺ using primers ARA124 and ARA234 (Table 1). The resulting 1215 bp DNA fragment was digested with KpnI and XhoI and cloned into pET30a(+) (Novagen), yielding pTL1. This plasmid was then digested with XbaI and BglII, and a chloramphenicol resistance (Cm^R) cassette from pMS38 (Zilhão et al., 2004) was inserted between those sites, generating pTL2. For the construction of pTL4, encoding a fusion of the C terminus of Abf2 to a His₆-tag, the C-terminal-encoding region of the abf2 gene was amplified by PCR of chromosomal DNA using primers ARA112 and ARA233 (Table 1), and the resulting 1380 bp PCR product digested with EcoRI and XhoI was cloned into pET30a(+), yielding pTL3. To generate pTL4, the pMS38 (Zilhão et al., 2004) XbaI–EcoRI DNA fragment, containing the Cm^R cassette, was subcloned into pTL3. Plasmids pTL2 and pTL4 were integrated into the host chromosome by means of a single-crossover (Campbell-type) recombinational event that occurred in the region of homology (Table 1).

Plasmids pTL11 and pTL12 are pET30a(+) derivatives encoding recombinant AbfA-His₆ and Abf2-His₆, respectively, under the control of a T7 inducible promoter. To construct pTL11, the coding sequence of abfA was amplified by PCR with primers ARA121 and ARA226, which introduced a unique XbaI site in the 5'-end region of the gene (Table 1). The resulting 1479 bp DNA fragment digested with XbaI/KpnI was cloned into pTL1. The same strategy was used in the construction of pTL12: the coding sequence of the abf2 gene was amplified by PCR of chromosomal DNA using primers ARA220 and ARA227, which created a unique NdeI site at the start of the codon. The resulting 994 bp PCR product digested with NdeI/EcoRI was subcloned in pTL3.

Linearized plasmids pMPR7 and pZI12 were used separately to delete the abfA gene and the abf2 gene, respectively, in the wild-type B. subtilis 168T⁺ chromosome. Plasmid pMPR7 was obtained by subcloning the 1441 bp SspI DNA fragment from pTN13 (Sá-Nogueira et al., 1997) into pSN21 (Sá-Nogueira & Mota, 1997) digested with SmaI. The construction of pZI12 was achieved by subcloning an 1106 bp NsiI–BamHI DNA fragment (containing the Cm^R cassette) from pMS38 (Zilhão et al., 2004) into pMPR5 digested with NsiI and BamHI. This latter plasmid, pMPR5, was constructed by subcloning a 1700 bp DNA fragment amplified by PCR of chromosomal DNA using primers ARA87 and ARA91 (Table 1) into pBluescript II KS(–) (Stratagene) digested with SmaI.

Construction of green fluorescent protein (GFP) fusions.
GFPmut2 fused to the N-terminus of Abf2 under the control of the abf2 promoter was engineered in three successive steps using the splicing by overlay extension (SOE) technique (Horton et al., 1989). First, the coding region of the abf2 gene was amplified by PCR, with primers ARA350 and ARA270 (Table 1), using chromosomal DNA of wild-type strain B. subtilis 168T⁺ as template. A linker encoding four asparagines was engineered in the primer ARA350 (Glaser et al., 1997). Second, a 726 bp fragment comprising the coding region of the gfpmut2 gene was PCR amplified from plasmid pEA18 (Quisel et al., 1999) by using the primers ARA349 and gfp30D (Table 1). Third, the promoter region of abf2 was amplified by PCR, with primers ARA87 and ARA348 (Table 1), using chromosomal DNA. P_abf2-gfpmut2-abf2 was amplified with primers ARA87 and ARA270 (Table 1), using an equimolar mix of abf2, gfpmut2 and abf2 promoter region PCR products as templates. The resulting fragment was inserted into pGEM-TEasy vector (Promega), generating pZI51. The same strategy was used to obtain an N-terminal fusion of AbfA to GFPmut2 under the control of the araABDLMNPQ-abfA promoter. First, the coding region of abfA was amplified with primers ARA347 and ARA269 (Table 1). A linker in the primer ARA347 was engineered as described above. Second, the araQ C-terminal coding region was amplified using primers ARA345 and ARA346 (Table 1). araQ-gfpmut2-abfA was amplified with primers ARA345 and ARA269 (Table 1) from the three PCR fragments, as described above, and inserted into pGEM-TEasy, yielding pZI50.

The P_abf2-gfpmut2-abf2 and araQ-gfpmut2-abfA fusions were integrated, in separate experiments, into the B. subtilis chromosome of the abf2 and abfA null mutants (IQB460 and IQB419, Table 1) by congression, using linearized pZI51 and pZI50, respectively, together with linearized pMLK83 (Karow & Piggot, 1995). Selection was made for the neomycin-resistance cassette (from pLMK83), integrated at the amyE locus by a double recombination event. The transformants were then screened for sensitivity to chloramphenicol or sensitivity to spectinomycin, indicative of the integration of P_abf2-gfpmut2-abf2 or araQ-gfpmut2-abfA fusions at the abf2 or abfA locus, respectively, of the recipient strains IQB460 and IQB419. The correct insertion of the fusions in the resulting strains, IQB493 and IQB494 (Table 1), respectively, was confirmed by PCR.

AF activity assays in B. subtilis.
To measure the AF activity in B. subtilis, the strains were grown on minimal medium C (Pascal et al., 1971) supplemented with 1 % (w/v) casein hydrolysate. During the early exponential phase (OD₆₀₀ 0.11–0.15), L-arabinose at a final concentration of 0.4 % (w/v) was added. Samples were collected 2 h after induction. The harvested cells were suspended in 1 ml Z buffer (Miller, 1972), and two drops of chloroform and one drop of 0.1 % (w/v) SDS were added and mixed vigorously for 10 s on a tabletop vortex apparatus. The enzyme reaction was started by adding 200 µl pNPAf [4 mg ml^–1 in P buffer (Miller, 1972)], and incubated for 20 min at 28 °C. Adding 250 µl of 2 M Na₂CO₃ then stopped the reaction and the A₄₀₀ was measured after a 5 min centrifugation to pellet cell debris. The level of accumulated AF activity was calculated as described by Miller (1972).

Fluorescence microscopy.
B. subtilis strains harbouring GFP fusions, and the wild-type strain 168T⁺ (used as negative control), were grown as described above for the AF assays. Two hours after induction, 0.5 ml aliquots were collected, harvested and washed three times with PBS, to eliminate possible traces of tryptophan present in the culture medium, and resuspended in 0.1 ml of the same buffer. Samples were then applied to agarose-coated microscope slides, and images were acquired under a Leica DMRA2 microscope coupled with a CoolSNAP HQ Photometrics camera (Roper Scientific).

Production and purification of recombinant AFs.
E. coli BL21(DE3) pLysS cells harbouring pTL11 or pTL12 were grown at 37 °C and 160 r.p.m. in 1 litre of LB with appropriate antibiotic selection. When the OD₆₀₀ reached 0.6 the expression of AbfA or Abf2 was induced by the addition of 1 mM IPTG. The culture was grown for an additional 3 h at 37 °C and 160 r.p.m. Cells were harvested by centrifugation at 4 °C, 8000 g, 10 min. All subsequent steps were carried out at 4 °C. The harvested cells were resuspended in Start Buffer [20 mM sodium phosphate, pH 7.4, 10 mM imidazole, 50 mM NaCl and 1 mM PMSF] and lysed by passing twice through a French pressure cell. The lysate was centrifuged for 30 min at 15 000 g and the proteins from the supernatant were loaded onto a 1 ml Histrap column (Amersham Phamarcia Biotech). The bound proteins were eluted with a discontinuous imidazole gradient and the fractions containing AbfA or Abf2 that were more than 95 % pure were dialysed overnight against storage buffer (20 mM sodium phosphate, pH 7.4, 50 mM NaCl, 10 % glycerol) and then frozen in liquid nitrogen and kept at –80 °C until further use.

To prepare the cell-free extracts, for small-scale analysis, the cells were resuspended in lysis buffer (50 mM sodium phosphate, pH 8.0, 300 mM NaCl, 10 mM imidazole) and disrupted in the presence of lysozyme (1 mg ml^–1) by three cycles of freezing in liquid nitrogen and thawing for 5 min at 37 °C, followed by incubation with benzonase (Invitrogen). After 15 min centrifugation at 16 000 g and 4 °C the soluble and insoluble fractions of the crude extract were obtained.

Protein analysis.
The analysis of production, homogeneity and molecular mass of the enzymes was done by SDS-PAGE, using broad-range molecular mass markers (Bio-Rad) as standards. Recombinant AbfA-His₆ and Abf2-His₆ were also analysed on native gradient (5–12.5 %) gels using a high-molecular-mass calibration kit for electrophoresis (Amersham) as standards. The degree of purification was determined by densitometric analysis of Coomassie-blue-stained SDS-PAGE gels. The protein content was determined using Bradford reagent (Bio-Rad) with BSA as standard.

Cross-linking was performed with purified His₆-tagged proteins. The reaction was initiated by adding glutaraldehyde, freshly prepared from stock solution in distilled water, to a final concentration of 2.5 mM followed by overnight incubation on ice.

Biochemical characterization.
For the determination of substrate specificities, various pNP-glycosides (pNPAf, pNPAp and pNPXp) and non-chromogenic substrates (arabinobiose, arabinotriose, sugar beet arabinan, linear arabinan, arabinoxylan and arabinogalactan) were used. To measure the activity of the enzymes against arabinose-containing polysaccharides and oligosaccharides, the reducing sugar content after hydrolysis was determined by the Nelson–Somogyi method (Somogyi, 1952), with L-arabinose as standard, as previously described (Leal & Sá-Nogueira, 2004). One unit of activity was defined as the amount of enzyme that produces 1 µmol arabinose equivalents per minute. The enzyme activities towards pNP-glycosides were determined using a reaction mixture containing 0.4 mM substrate in PC buffer (72.8 mM sodium phosphate, 13.6 mM citric acid), pH 6.6, and appropriately diluted enzyme. The reaction was incubated at 37 °C and the increase in A₄₀₀ was measured against time. The amount of product generated was calculated using a molar absorption coefficient for p-nitrophenol of 10 500 M^–1 cm^–1 at 400 nm. All assays were performed in triplicate.

Temperature and pH values for maximum enzymic activity of the AFs were determined by incubation for 5 min using 0.4 mM pNPAf as substrate in PC buffer, and calculated as described above. Enzymic activity was also determined in the presence of 1 mM EDTA using the same conditions. The effect of temperature was tested in PC buffer, pH 6.6, at temperatures ranging from 30 °C to 80 °C. The effect of pH on the activity was assayed at 37 °C in a series of Britton–Robinson buffers from pH 3.0 to 9.0 (0.1 M boric acid, 0.1 M acetic acid, 0.1 M phosphoric acid, adjusted to the desired pH with NaOH) (Britton & Robinson, 1931). Thermal stability of the enzymes was estimated by incubating the diluted solution of enzymes in PC buffer, pH 6.6, at 55 °C for AbfA and 70 °C for Abf2. Samples were removed after 10–120 min, kept in ice for 5 min, and residual enzyme activity was determined, at pH 6.6 and 37 °C, using 0.4 mM pNPAf in PC buffer (pH 6.6) as substrate.

Nucleotide sequence accession numbers.
The nucleotide sequence of the abf2 gene reported in this paper has been submitted to GenBank under accession number EU073712. The nucleotide sequence of the abfA gene was previously assigned GenBank accession number X89810.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

 
Functional analysis of AbfA and Abf2 in B. subtilis
By primary amino acid sequence analysis, the abfA and abf2 genes from B. subtilis most probably encode AFs (EC 3.2.1.55) (Raposo et al., 2004; Sá-Nogueira et al., 1997; Wipat et al., 1996). The primary structure of B. subtilis AbfA is very similar to that of the characterized AbfA from Geobacillus stearothermophilus T-6 [71 % identity (Gilead & Shoham, 1995)], AbfATK4 from Geobacillus caldoxylolyticus TK4 [70 % identity (Canakci et al., 2007)] and Araf51 from Clostridium thermocellum [63 % identity (Taylor et al., 2006)], and AF activity was reported for the B. subtilis abfA gene product (Wipat et al., 1996). The amino acid sequence of Abf2 displays high identity to characterized AFs from Bacillus pumilus, ArfA [65 % identity (Degrassi et al., 2003)], Thermobacillus xylanilyticus, AbfD3 [64 % identity (Debeche et al., 2000)], Clostridium cellulovorans, ArfA [60 % identity (Kosugi et al., 2002)] and Clostridium stercorarium ArfB [56 % identity (Zverlov et al., 1998)]. Although AbfA (500 aa) and Abf2 (495 aa) share only 23 % identity, they both belong to family 51 of the glycoside hydrolases (GH 51) from different bacteria (http://www.cazy.org/fam/GH51.html).

To characterize the function of these genes in B. subtilis we constructed, by insertion-deletion mutations, single abfA and abf2 mutants (IQB419 and IQB460; Table 1) and the double abfA abf2 mutant (IQB462; Table 1). The wild-type 168T⁺, and abf2 and abfA null mutant strains were grown on minimal medium C supplemented with 1 % casein hydrolysate in the presence of arabinose. Samples were collected 2 h after induction and the intracellular level of accumulated AF activity was measured, using pNPAf as substrate. The results showed that the single abf2 and abfA B. subtilis null mutants retained 70 % and 38 % of AF activity, respectively, relative to the wild-type (data not shown). In the abfA abf2 double null mutant a complete loss of AF activity was observed (data not shown). In conclusion, both genes encode AFs and are responsible for the majority of the intracellular AF activity in B. subtilis. Interestingly, the difference in AF activity observed in the single abf2 and abfA null mutants might reflect the distinct levels of abfA and abf2 gene expression observed previously (Raposo et al., 2004). Since the two enzymes displayed similar kinetic parameters for pNPAf (see below), the different expression observed at the transcriptional level might lead to dissimilar synthesis of the two enzymes.

The physiological effect of the mutations on the utilization of sugar beet arabinan and -1,5-linked arabino-oligosaccharides as the sole carbon and energy source was determined; the results are summarized in Table 2. The double mutant was unable to grow on minimal medium with either arabinan or -1,5-linked arabino-oligosaccharides (arabinobiose, arabinotriose and arabinotetraose), but it utilized arabinose like the wild-type strain, which indicates that the products of arabinan degradation, i.e. arabinose oligomers, are the natural substrates for the two AFs. The inactivation of the abfA gene in the single mutant IQB419 has a much more drastic effect on the growth on -1,5-linked arabino-oligosaccharides as sole carbon source when compared to the abf2 single mutant, strain IQB460. On the other hand, the abfA and abf2 single mutations had a more similar effect on growth in the presence of sugar beet arabinan (Table 2). These observations strongly suggest a different contribution of the two enzymes to utilization of arabino-polysaccharides (discussed below).

View larger version (98K): [in this window] [in a new window]   Fig. 1. Fluorescence microscopy of B. subtilis expressing GFP-AbfA and GFP-Abf2: phase-contrast (a, c), and GFP localization (b, d) images of cells in exponential phase grown in minimal medium C supplemented with 1 % (w/v) casein hydrolysate in the presence of arabinose (see Methods).

View larger version (21K): [in this window] [in a new window]   Fig. 2. Purification and characterization of the oligomeric state of AbfA and Abf2. (a, b) Purified recombinant AbfA (a) and Abf2 (c) (1 µg of each protein) were separated by SDS-PAGE on 12.5 % gels and stained with Coomassie blue. The positions of the recombinant proteins are shown by open arrowheads. The sizes of the broad-range molecular mass markers (Bio-Rad) are indicated. (b, d) The analysis of oligomer formation by the recombinant proteins (4 µg of each protein) was performed under native conditions. The proteins were separated on native gradient 5–12.5 % gels and stained with Coomassie blue. The AbfA (b) and Abf2 (d) oligomeric forms are shown by arrowheads. The sizes of the proteins in the high molecular mass calibration kit for electrophoresis (Amersham Biosciences) are indicated. (e) Glutaraldehyde cross-linking of AbfA and Abf2. Proteins (3.8 µg) were separated by SDS-PAGE (12.5 %) after cross-linking with 2.5 mM glutaraldehyde [lane 2, Abf2; lane 3, AbfA; lane 4, negative control (MBP 2* maltose-binding protein; Biolabs)] or in the absence of cross-linker (lane 5, MBP2*; lane 6, Abf2; lane 7, AbfA). The sizes of the broad-range molecular mass markers (Bio-Rad) are indicated (lane 1).

Enzymic characterization of purified AbfA and Abf2
The biochemical and biophysical properties of purified recombinant AbfA-His₆ and Abf2-His₆ were determined as described in Methods. The ability of the two AFs to hydrolyse different substrates was assayed; the results are summarized in Table 3. AbfA was able to release arabinose from arabinobiose, arabinotriose, linear -1,5-L-arabinan, sugar beet arabinan (branched), larch wood arabinogalactan and wheat arabinoxylan but was not active towards pNPAp or pNPXp. Although the enzyme cleaved larch wood arabinogalactan and wheat arabinoxylan the determination of the individual kinetic constants was not possible, most probably due to both the high K_m of AbfA and low solubility of the polysaccharides. Abf2 was able to hydrolyse linear -1,5-L-arabinan, sugar beet arabinan (branched) and wheat arabinoxylan but was not active towards larch wood arabinogalactan, pNPAp or pNPXp (data not shown). In addition, Abf2 was able to cleave arabinobiose and arabinotriose; however, most probably due to the high K_m of the enzyme for arabinobiose the precise determination of the kinetic parameters was not possible. In sum, the substrate-specificity assays indicated that AbfA has the following decreasing order of reactivity on arabinose-containing polysaccharides: debranched arabinan>sugar beet arabinan>wheat arabinoxylan=larch wood arabinogalactan (weak activity). Recombinant Abf2 was able to hydrolyse linear -1,5-L-arabinan, sugar beet arabinan and wheat arabinoxylan. Moreover, it displayed higher activity towards branched arabinan, a molecule comprising a backbone of -1,5-linked L-arabinofuranosyl residues decorated with -1,2-, and -1,3-linked L-arabinofuranosyl units, than towards debranched arabinan. In addition, arabinoxylan, which has L-arabinofuranose residues attached to the main chain by -1,2- and/or -1,3-glycosidic linkages, is preferred to linear -1,5-L-arabinan. Taken together, these results highlight the preference of AbfA to hydrolyse (15) linkages and corroborate the data obtained on the growth kinetics of the wild-type and abfA and abf2 mutant strains on different arabino-polysaccharides as sole carbon sources (see above; Table 2). The activity of the two enzymes towards sugar beet arabinan resembles that of AF I and II, respectively, partially purified from the culture supernatant of B. subtilis F-11 (Weinstein & Albersheim, 1979). AbfA's substrate specificity is similar to that of an AF purified from the culture supernatant of B. subtilis 3-6 (Kaneko et al., 1994). The kinetic parameters determined by measuring the hydrolysis of the synthetic substrate pNPAf were very similar for both enzymes (Table 3) and comparable to the values determined for other bacterial AFs (Beldan et al., 1997; Beylot et al., 2001b; Debeche et al., 2000; Degrassi et al., 2003; Gilead & Shoham, 1995; Kosugi et al., 2002; Taylor et al., 2006; Zverlov et al., 1998).

AbfA and Abf2: two different AFs for concerted action on arabino-oligosaccharides by B. subtilis
Although Abf2 and AbfA have similar biochemical and biophysical properties the two enzymes showed different substrate specificity. AbfA acts preferentially on (15) linkages, and Abf2 on (12) and (13) linkages, which suggest a concerted contribution of these enzymes to optimal utilization of arabinose-containing polysaccharides by B. subtilis and some other Bacillus species. The functional and subcellular localization analyses performed in this work indicated that both AFs are retained within the cytoplasm, where presumably their targets are oligosaccharides with (15), (12), and (13) linkages, such as arabino-oligomers, arabinoxylo-oligosaccharides and arabinogalacto-oligosaccharides. In a previous study we showed that arabinose is transported into the cell mainly by a permease, AraE (Sá-Nogueira & Ramos, 1997). Additionally, the predicted products of the genes araN, araP and araQ, belonging to the metabolic operon araABDLMNPQ-abfA, are homologous to components of bacterial binding-protein-dependent transport systems involved in the transport of malto-oligosaccharides and multiple sugars. An insertion-deletion mutation in the region downstream of the araD gene caused only a small decrease in doubling time of the mutant strain (IQB206) in a minimal medium with arabinose as the sole carbon source. Therefore a possible role of the AraNPQ proteins in the transport of L-arabinose and/or arabinose oligomers was suggested (Sá-Nogueira et al., 1997). Thus, we determined the physiological effect of this mutation (strain IQB206) on the utilization of arabinan and -1,5-linked arabino-oligosaccharides as the sole carbon and energy source (Table 2). This mutant strain was unable to grow on arabinan and -1,5-linked arabino-oligosaccharides (arabinobiose, arabinotriose, and arabinotetraose) as the sole carbon sources (Table 2), which strongly suggests the involvement of the AraNPQ system in the transport of arabinose oligomers into the cell. The slight decrease in the doubling time with arabinose observed in strain IQB206 (Table 2) is consistent with previous results (Sá-Nogueira et al., 1997).

Taken together, these observations lead us to propose the following pathway for the depolymerization of arabinose-containing polysaccharides in B. subtilis (Fig. 3). The extracellular degradation of arabinan, arabinoxylan and arabinogalactan, is accomplished by arabinanases, xylanases and galactanases. The resulting arabino-oligosaccharides (arabinose oligomers, arabinoxylo-oligosaccharides and arabinogalacto-oligosaccharides) enter the cell by specific transport systems, namely the AraNPQ ABC-type transporter. Once inside the cell, arabino-oligosaccharides with (15), (12) and (13) linkages are degraded by the concerted action of the two GH 51 AFs, AbfA and Abf2, releasing arabinose. A complete depolymerization to monosaccharides, arabinose, xylose and galactose, is accomplished together with β-xylosidases and β-galactosidases. At the level of gene expression abfA and abf2 are induced by arabinose and repressed by glucose (Raposo et al., 2004). AraR, the key regulator of arabinose utilization in B. subtilis, in the absence of the inducer (arabinose) represses and tightly controls the transcription of both abfA and abf2 by binding to two in-phase operators in both the promoter region of the araABDLMNPQ-abfA operon and the promoter region of abf2 (Mota et al., 1999; Raposo et al., 2004). Glucose repression of the expression of the two AF genes is achieved by CcpA, the master regulator of carbon catabolite repression, associated mainly with co-effector HPr(Ser-P) (Inácio & de Sá-Nogueira, 2007). In the presence of glucose and arabinose plus glucose CcpA binds to one cre element located downstream near the promoter region abf2 and to two cre elements present in the araABDLMNPQ-abfA operon, cre araA located near the promoter and cre araB positioned within the araB gene (Inácio et al., 2003; Inácio & de Sá-Nogueira, 2007).

View larger version (39K): [in this window] [in a new window]   Fig. 3. The non-redundant roles of AbfA and Abf2 in arabino-oligosaccharide degradation in B. subtilis. Arabino-oligosaccharides resulting from the action of extracellular enzymes in different hemicellulosic homopolysaccharides (branched and debranched arabinans) and heteropolysaccharides (arabinoxylans, arabinogalactans) are transported into the cell by specific transport systems. Arabinose oligomers enter the cell most probably via AraNPQ, an ABC-type transporter (Sá-Nogueira et al., 1997). Mixed oligomers, arabinoxylo-oligosaccharides and arabinogalacto-oligosaccharides may be transported into the cell also via AraNPQ or by other unidentified transport systems (?). The transport of the monosaccharides, arabinose, xylose and galactose, is accomplished by the same transport system, the AraE permease (Sá-Nogueira & Ramos, 1997; Krispin & Allmansberger, 1998). Once inside the cell, arabino-oligosaccharides with -1,5, -1,2, and -1,3 linkages are degraded by the concerted action of the two GH 51 AFs, AbfA and Abf2. L-Arabinose is then converted, by the action of the products of the araA, araB and araD genes (Sá-Nogueira et al., 1997), to D-xylulose 5-phosphate, which is further catabolized through the pentose phosphate pathway (PPP).

	ACKNOWLEDGEMENTS

 
We thank Adriano O. Henriques for critically reading the manuscript and helpful suggestions, and Maria P. Raposo and Teresa F. Leal for their contribution in the early stages of this study. This work was partially supported by grant no. POCI/AGR/60236/2004 from the Fundação para a Ciência e Tecnologia (FCT) and FEDER to I. d. S.-N., and fellowship SFRH/BD/18238/2004 from the FCT to J. M. I.

Edited by: W. Quax

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

 
Antelmann, H., Tjalsma, H., Voigt, B., Ohlmeier, S., Bron, S., van Dijl, J. M. & Hecker, M. (2001). A proteomic view on genome-based signal peptide predictions. Genome Res 11, 1484–1502.[Abstract/Free Full Text]

Beldman, G., Schols, H. A., Pitson, S. M., Searl-van Leeuwen, M. J. F. & Voragen, A. G. (1997). Arabinans and arabinan degrading enzymes. Adv Macromol Carbohydr Res 1, 1–64.

Beylot, M. H., Emami, K., McKie, V. A., Gilbert, H. J. & Pell, G. (2001a). Pseudomonas cellulosa expresses a single membrane-bound glycoside hydrolase family 51 arabinofuranosidase. Biochem J 358, 599–605.[CrossRef][Medline]

Beylot, M. H., McKie, V. A., Voragen, A. G., Doeswijk-Voragen, C. H. & Gilbert, H. J. (2001b). The Pseudomonas cellulosa glycoside hydrolase family 51 arabinofuranosidase exhibits wide substrate specificity. Biochem J 358, 607–614.[CrossRef][Medline]

Britton, H. T. S. & Robinson, R. A. (1931). Universal buffer solutions and dissociation constant of Veronal. J Chem Soc 456, 1456–1462.

Canakci, S., Belduz, A. O., Saha, B. C., Yazar, A., Ayaz, F. A. & Yayli, N. (2007). Purification and characterization of a highly thermostable alpha-L-arabinofuranosidase from Geobacillus caldoxylolyticus TK4. Appl Microbiol Biotechnol 75, 813–820.[CrossRef][Medline]

Débarbouillé, M., Arnaud, M., Fouet, A., Klier, A. & Rapoport, G. (1990). The sacT gene regulating the sacPA operon in Bacillus subtilis shares strong homology with transcriptional antiterminators. J Bacteriol 172, 3966–3973.[Abstract/Free Full Text]

Debeche, T., Cummings, N., Connerton, I., Debeire, P. & O'Donohue, M. J. (2000). Genetic and biochemical characterization of a highly thermostable -L-arabinofuranosidase from Thermobacillus xylanilyticus. Appl Environ Microbiol 66, 1734–1736.[Abstract/Free Full Text]

Degrassi, G., Vindigni, A. & Venturi, V. (2003). A thermostable -arabinofuranosidase from xylanolytic Bacillus pumilus: purification and characterisation. J Biotechnol 101, 69–79.[CrossRef][Medline]

Gilead, S. & Shoham, Y. (1995). Purification and characterization of -L-arabinofuranosidase from Bacillus stearothermophilus T-6. Appl Environ Microbiol 61, 170–174.[Abstract/Free Full Text]

Glaser, P., Sharpe, M. E., Raether, B., Perego, M., Ohlsen, K. & Errington, J. (1997). Dynamic, mitotic-like behavior of a bacterial protein required for accurate chromosome partitioning. Genes Dev 11, 1160–1168.[Abstract/Free Full Text]

González-Pastor, J. E., Hobbs, E. C. & Losick, R. (2003). Cannibalism by sporulating bacteria. Science 301, 510–513.[Abstract/Free Full Text]

Horton, R. M., Hunt, H. D., Ho, S. N., Pullen, J. K. & Pease, L. R. (1989). Engineering hybrid genes without the use of restriction enzymes: gene splicing by overlap extension. Gene 77, 61–68.[CrossRef][Medline]

Hovel, K., Shallom, D., Niefind, K., Belakhov, V., Shoham, G., Baasov, T., Shoham, Y. & Schomburg, D. (2003). Crystal structure and snapshots along the reaction pathway of a family 51 -L-arabinofuranosidase. EMBO J 22, 4922–4932.[CrossRef][Medline]

Inácio, J. M. & de Sá-Nogueira, I. (2007). trans-Acting and cis elements involved in glucose repression of arabinan degradation in Bacillus subtilis. J Bacteriol 189, 8371–8376.[Abstract/Free Full Text]

Inácio, J. M., Costa, C. & de Sá-Nogueira, I. (2003). Distinct molecular mechanisms involved in carbon catabolite repression of the arabinose regulon in Bacillus subtilis. Microbiology 149, 2345–2355.[Abstract/Free Full Text]

Kaji, A. & Saheki, T. (1975). Endo-arabanase from Bacillus subtilis F-11. Biochim Biophys Acta 410, 354–360.[Medline]

Kaneko, S., Sano, M. & Kusakabe, I. (1994). Purification and some properties of -L-arabinofuranosidase from Bacillus subtilis 3-6. Appl Environ Microbiol 60, 3425–3428.[Abstract/Free Full Text]

Karow, M. L. & Piggot, P. J. (1995). Construction of gusA transcriptional fusion vectors for Bacillus subtilis and their utilization for studies of spore formation. Gene 163, 69–74.[CrossRef][Medline]

Kosugi, A., Murashima, K. & Doi, R. H. (2002). Characterization of two noncellulosomal subunits, ArfA and Bga, from Clostridium cellulovorans that cooperate with the cellulosome in plant cell wall degradation. J Bacteriol 184, 6859–6865.[Abstract/Free Full Text]

Krispin, O. & Allmansberger, R. (1998). The Bacillus subtilis AraE protein displays a broad substrate specificity for several different sugars. J Bacteriol 180, 3250–3252.[Abstract/Free Full Text]

Leal, T. F. & Sá-Nogueira, I. (2004). Purification, characterization and functional analyses of an endo-arabinanase (AbnA) from Bacillus subtilis. FEMS Microbiol Lett 241, 41–48.[CrossRef][Medline]

Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Mota, L. J., Tavares, P. & Sá-Nogueira, I. (1999). Mode of action of AraR, the key regulator of L-arabinose metabolism in Bacillus subtilis. Mol Microbiol 33, 476–489.[CrossRef][Medline]

Nakai, K. & Horton, P. (1999). PSORT: a program for detecting the sorting signals of proteins and predicting their subcellular localization. Trends Biochem Sci 24, 34–35.[CrossRef][Medline]

Numan, M. T. & Bhosle, N. B. (2006). -L-Arabinofuranosidases: the potential applications in biotechnology. J Ind Microbiol Biotechnol 33, 247–260.[CrossRef][Medline]

Pascal, M., Kunst, F., Lepesant, J. A. & Dedonder, R. (1971). Characterization of two sucrase activities in Bacillus subtilis Marburg. Biochimie 53, 1059–1066.[CrossRef][Medline]

Quisel, J. D., Lin, D. C.-H. & Grossman, A. D. (1999). Control of the development by altered localization of a transcription in B. subtilis. Mol Cell 4, 665–672.[CrossRef][Medline]

Raposo, M. P., Inácio, J. M., Mota, L. J. & Sá-Nogueira, I. (2004). Transcriptional regulation of arabinan-degrading genes in Bacillus subtilis. J Bacteriol 186, 1287–1296.[Abstract/Free Full Text]

Saha, B. C. (2000). -L-Arabinofuranosidases: biochemistry, molecular biology and application in biotechnology. Biotechnol Adv 18, 403–423.[CrossRef][Medline]

Sakai, T. & Sakamoto, T. (1990). Purification and some properties of a protopectin-solubilizing enzyme that has potent activity on sugar beet protopectin. Agric Biol Chem 54, 879–889.

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Sá-Nogueira, I. & Mota, L. J. (1997). Negative regulation of L-arabinose metabolism in Bacillus subtilis: characterization of the araR (araC) gene. J Bacteriol 179, 1598–1608.[Abstract/Free Full Text]

Sá-Nogueira, I. & Ramos, S. S. (1997). Cloning, functional analysis, and transcriptional regulation of the Bacillus subtilis araE gene involved in L-arabinose utilization. J Bacteriol 179, 7705–7711.[Abstract/Free Full Text]

Sá-Nogueira, I., Nogueira, T. V., Soares, S. & de Lencastre, H. (1997). The Bacillus subtilis L-arabinose (ara) operon: nucleotide sequence, genetic organization and expression. Microbiology 143, 957–969.[Abstract/Free Full Text]

Shallom, D. & Shoham, Y. (2003). Microbial hemicellulases. Curr Opin Microbiol 6, 219–228.[CrossRef][Medline]

Shulami, S., Zaide, G., Zolotnitsky, G., Langut, Y., Feld, G., Sonenshein, A. L. & Shoham, Y. (2007). A two-component system regulates the expression of an ABC transporter for xylo-oligosaccharides in Geobacillus stearothermophilus. Appl Environ Microbiol 73, 874–884.[Abstract/Free Full Text]

Somogyi, M. (1952). Notes on sugar determination. J Biol Chem 195, 19–23.[Free Full Text]

Studier, F. W., Rosenberg, A. H., Dunn, J. J. & Dubendorff, J. W. (1990). Use of T7 RNA polymerase to direct expression of cloned genes. Methods Enzymol 185, 60–89.[Medline]

Taylor, E. J., Smith, N. L., Turkenburg, J. P., D'Souza, S., Gilbert, H. J. & Davies, G. J. (2006). Structural insight into the ligand specificity of a thermostable family 51 arabinofuranosidase, Araf51, from Clostridium thermocellum. Biochem J 395, 31–37.[CrossRef][Medline]

Weinstein, L. & Albersheim, P. (1979). Structure of plant cell walls. IX. Purification and partial characterization of a wall-degrading endo-arabanase and an arabinosidase from Bacillus subtilis. Plant Physiol 63, 425–432.[Abstract/Free Full Text]

Wipat, A., Carter, N., Brignell, S. C., Guy, B. J., Piper, K., Sanders, J., Emmerson, P. T. & Harwood, C. R. (1996). The dnaB-pheA (25 °–24 °) 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.[Abstract/Free Full Text]

Zilhão, R., Serrano, M., Isticato, R., Ricca, E., Moran, C. P., Jr & Henriques, A. O. (2004). Interactions among CotB, CotG, and CotH during assembly of the Bacillus subtilis spore coat. J Bacteriol 186, 1110–1119.[Abstract/Free Full Text]

Zverlov, V. V., Liebl, W., Bachleitner, M. & Schwarz, W. H. (1998). Nucleotide sequence of arfB of Clostridium stercorarium, and prediction of calalytic residues of alpha-L-arabinofuranosidases based on local similarity with several families of glycosyl hydrolases. FEMS Microbiol Lett 164, 337–343.[Medline]

Received 30 March 2008; revised 23 May 2008; accepted 2 June 2008.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Inácio, J. M. Articles by de Sá-Nogueira, I. Search for Related Content PubMed PubMed Citation Articles by Inácio, J. M. Articles by de Sá-Nogueira, I. Agricola Articles by Inácio, J. M. Articles by de Sá-Nogueira, I.