SGM Journals
PHYSIOLOGY AND BIOCHEMISTRY

Identification of two scyllo-inositol dehydrogenases in Bacillus subtilis

Tetsuro Morinaga, Hitoshi Ashida, Ken-ichi Yoshida

  • Department of Agrobioscience, Graduate School of Agricultural Science, Kobe University, 1-1 Rokkodai, Nada, Kobe 657-8501, Japan
  • Correspondence
    Ken-ichi Yoshida
    kenyoshi{at}kobe-u.ac.jp

    • Microbiology 2010; 156(5):1538–1546 · https://doi.org/10.1099/mic.0.037499-0

    View at publisher PubMed

    Abstract

    scyllo-Inositol (SI) is a stereoisomer of inositol whose catabolism has not been characterized in bacteria. We found that Bacillus subtilis 168 was able to grow using SI as its sole carbon source and that this growth was dependent on a functional iol operon for catabolism of myo-inositol (MI; another inositol isomer, which is abundant in nature). Previous studies elucidated the MI catabolic pathway in B. subtilis as comprising multiple stepwise reactions catalysed by a series of Iol enzymes. The first step of the pathway converts MI to scyllo-inosose (SIS) and involves the MI dehydrogenase IolG. Since IolG does not act on SI, we suspected that there could be another enzyme converting SI into SIS, namely an SI dehydrogenase. Within the whole genome, seven genes paralogous to iolG have been identified and two of these, iolX and iolW (formerly known as yisS and yvaA, respectively), were selected as candidate genes for the putative SI dehydrogenase since they were both prominently expressed when B. subtilis was grown on medium containing SI. iolX and iolW were cloned in Escherichia coli and both were shown to encode a functional enzyme, revealing the two distinct SI dehydrogenases in B. subtilis. Since inactivation of iolX impaired growth with SI as the carbon source, IolX was identified as a catabolic enzyme required for SI catabolism and it was shown to be NAD+ dependent. The physiological role of IolW remains unclear, but it may be capable of producing SI from SIS with NADPH oxidation.

    Edited by: J. M. van Dijl

    INTRODUCTION

    Inositol is the generic name for 1,2,3,4,5,6-cyclohexanehexol; epimerization of the six hydroxyl groups results in nine stereoisomers. Among the isomers, myo-inositol (MI) is the most abundant in nature and it is an essential compound in eukaryotic cell membranes as a phosphatidylinositol moiety, which plays an important role in signal transduction. scyllo-Inositol (SI), the C-2 epimer of MI, is relatively rare but is sometimes found in plants, for example in barley seeds (Kinnard et al., 1995) and chrysanthemum buds (Ichimura et al., 2000). Also, SI is sometimes present in phosphatidylinositol, since the phosphatidylinositol synthase of Tetrahymena vorax is known to act not only on MI but also on SI (Riggs et al., 2007). In mammals, SI has been shown to localize in the kidney and brain (Sherman et al., 1968), and an enzyme from bovine brain was found to interconvert MI and SI (Hipps et al., 1976). Interestingly, SI has been shown to inhibit aggregation of beta-amyloid (Abeta) peptide in the brains of Alzheimer's patients, who develop hippocampal synaptic plasticity causing memory deficits as Abeta oligomer aggregates (McLaurin et al., 2000). It was also demonstrated that oral SI administration prevents memory impairment caused by formation of Abeta oligomers (Townsend et al., 2006). In addition, recently it was reported that artificial SI derivatives inhibited Abeta aggregation and had effects on the configuration of Abeta oligomers (Sun et al., 2008). Thus, not only MI but also SI is important in eukaryotes and rarely metabolized as a mere carbon source.

    However, some micro-organisms can grow efficiently on MI. A set of enzymes involved in MI catabolism was first identified in Aerobacter aerogenes (reclassified as Enterobacter aerogenes/Klebsiella mobilis) (Berman & Magasanik, 1966), and many subsequent studies have identified MI catabolic enzymes similar to those of A. aerogenes in a number of micro-organisms, including Cryptococcus melibiosum (Vidal-Leiria & van Uden, 1973), Rhizobium leguminosarum bv. viciae (Poole et al., 1994), Bacillus subtilis (Yoshida et al., 1997), Sinorhizobium meliloti (Galbraith et al., 1998), Sinorhizobium fredii (Jiang et al., 2001), Corynebacterium glutamicum (Krings et al., 2006) and Lactobacillus casei (Yebra et al., 2007). Recently, the MI catabolic pathway in B. subtilis was fully elucidated (Fig. 1⇓). The B. subtilis iolABCDEFGHIJ operon encodes enzymes for multiple steps of the MI catabolic pathway (Yoshida et al., 1997). The first reaction is the conversion of MI (compound [1]) into scyllo-inosose (SIS; also called 2-keto-MI, compound [4]) in a reaction catalysed by IolG, an MI dehydrogenase (Fujita et al., 1991). During the second reaction, SIS is dehydrated into 3d-(3,5/4)-trihydroxycyclohexane-1,2-dione (THcHDO, compound [6]); this step is catalysed by IolE (Yoshida et al., 2004). IolD catalyses THcHDO hydrolysis to yield 5-deoxy-d-glucuronic acid (5DG, compound [7]) in the third reaction. IolB catalyses the isomerization of 5DG to 2-deoxy-5-keto-d-gluconic acid (DKG, compound [8]), which is subsequently phosphorylated by IolC kinase to DKG 6-phosphate (DKGP, compound [9]) (Yoshida et al., 2008). DKGP appeared to be the intermediate acting as inducer by antagonizing DNA binding of IolR (the repressor controlling transcription of the iol operon) (Yoshida et al., 1999a). Finally, the specific aldolase IolJ for the sixth reaction catalyses the cleavage of DKGP into dihydroxyacetone phosphate (compound [10]) and malonic semialdehyde (compound [11]) (Yoshida et al., 2008). The former is a known glycolytic intermediate and the latter has previously been shown to be converted into acetyl-CoA (compound [12]) and CO2 by an IolA-catalysed reaction (Stines-Chaumeil et al., 2006).

    Figure image not available in archive
    Fig. 1.

    Inositol catabolic pathway and the iol genes of B. subtilis. B. subtilis iol genes encoding the enzymes for the respective reactions in the inositol catabolic pathway and intermediate compounds are shown. Our previous studies have demonstrated that B. subtilis utilizes either MI or DCI as its sole carbon source via the catabolic pathway. In this study, it was revealed that B. subtilis also utilizes SI, which is dependent on the MI/DCI catabolic pathway. An additional path utilizing SI was found to merge into the branching point at SIS and involves two newly identified genes, iolX and iolW. The intermediate compounds involved in the pathway are MI [1], DCI [2], SI [3], SIS [4], 1-keto-DCI [5], THcHDO [6], 5DG [7], DKG [8], DKGP [9], dihydroxyacetone phosphate [10], malonic semialdehde [11] and acetyl-CoA [12].

    We have shown that B. subtilis utilizes not only MI but also d-chiro-inositol (DCI, compound [2]) (Yoshida et al., 2006). DCI is converted to 1-keto-DCI (compound [5]) in an IolG-catalysed reaction, and then isomerized to SIS by an IolI-catalysed reaction. Thus, DCI appears to be degraded via the MI catabolic pathway. In addition, IolG can react with pinitol (3-O-methyl-DCI), and B. subtilis has been shown to utilize pinitol as the sole carbon source via the same MI catabolic pathway (Morinaga et al., 2006).

    In this paper, we demonstrate that B. subtilis can also grow on SI as a sole carbon source and that this is also dependent on the MI catabolic pathway; we have also identified two genes encoding enzymes converting SI into SIS, i.e. SI dehydrogenases.

    METHODS

    Bacterial strains, plasmids and growth conditions.

    Bacterial strains and plasmids used in this study are listed in Table 1⇓. B. subtilis 60015 is our standard laboratory strain (Fujita et al., 1991). B. subtilis YF248 is a mutant derived from strain 60015, whose construction was described previously (Yoshida et al., 1997). Strains BFS3018 and BFS1058 were supplied by the National Institute of Genetics, Japan. The erythromycin-resistance cassette of BFS1058 was replaced with a tetracycline-resistance cassette through integration of the plasmid pEm : : Tc (Ogura et al., 2001) to obtain strain TM043. Strain TM044 was constructed by transforming BFS3018 with DNA from TM043 resistant to both erythromycin and tetracycline. B. subtilis strains were grown on tryptose blood agar base (Difco) plates supplemented with 0.18 % glucose or in S6 minimal medium (Fujita & Freese, 1981) supplemented with 0.02 % yeast extract and containing the appropriate carbon source as indicated. Escherichia coli strains DH5α (Sambrook & Russell, 2001) and BL21(DE3) (Novagen) were used as the hosts for plasmid construction and expression of the C-terminally His6-tagged fusion protein, respectively. E. coli strains were maintained in LB medium (Sambrook & Russell, 2001). Plasmid pET-30a(+) (Novagen) was used as the cloning vector for the His6-tag fusion constructs. When required, media were supplemented with antibiotics, including chloramphenicol (15 μg ml−1), erythromycin (0.5 μg ml−1), tetracycline (12.5 μg ml−1) and kanamycin (50 μg ml−1).

    Table 1.

    Bacterial strains and plasmids used in this study

    Plasmid construction.

    To express the His6-tag fusion protein in E. coli, plasmids were constructed as follows. Two types of DNA fragments were amplified by PCR using DNA from strain 60015; each of these was designed to cover either the iolX or the iolW coding regions by generating the NdeI or XhoI sites at the head or tail, respectively. Specific primer pairs used for the PCR are listed in Table 2⇓. Primer pair iolXNdeI/iolXXhoI was used to amplify the iolX fragment, and primer pair iolWNdeI/iolWXhoI was used to amplify the iolW fragment. Each of the fragments was trimmed with NdeI and XhoI and ligated with the NdeI–XhoI arm of pET-30a(+). The ligated DNA was used to transform E. coli strain DH5α to kanamycin resistance, and recombinant plasmids were extracted from the transformants. The resulting plasmids were designated pETiolX and pETiolW and carried the respective iol genes with in-frame C-terminal fusion to a His6-tag placed under the control of the pET-30a(+)-borne T7 promoter.

    Table 2.

    Oligonucleotides used in this study

    RNA techniques.

    B. subtilis 60015 was grown at 37 °C with shaking in S6 medium containing 0.5 % Casamino acids (Difco) with 10 mM MI or 10 mM SI. The cells were harvested at an OD600 of 1.0. Total RNA was extracted from the cells and purified as previously described (Yoshida et al., 1999b). The RNA samples were subjected to DNA macroarray analysis using the Panorama B. subtilis Gene Array system (Sigma-Aldrich) according to the instructions provided by the supplier. The autoradiogram of the DNA array membrane was obtained by exposing a BAS-IP SR 2040 plate (Fuji Photo Film) to the membrane and scanning it with a Typhoon 9400 phosphorimager (GE Healthcare Bio-Science). The imaging data were analysed using Array Vision software version 8.0 (Imaging Research), and the overall-spot-normalization function of the software was used for normalization and background subtraction. Each analysis was carried out twice using independently isolated RNA preparations and two different array batches. The mean of the normalized signal intensity of the spots contained in the DNA array was taken to represent the expression level of the respective genes. The RNA samples were also subjected to Northern blot analysis, performed using a process similar to that described previously (Yoshida et al., 1997); however, digoxigenin (DIG)-labelled RNA probes were used. These were prepared as follows. DNA fragments corresponding to the target genes were amplified from 60015 DNA by PCR and a T7 RNA promoter tag was generated at the downstream end. The specific primer pairs used were NiolX and NiolXDIG for the iolX probe and NiolW and NiolWDIG for the iolW probe. The PCR product was used as the template for in vitro transcription using a DIG RNA labelling kit (SP6/T7) (Roche Diagnostics). The RNA samples (15 μg) prepared from the cells were separated in an electrophoresis gel (Yoshida et al., 1997), transferred to a positively charged nylon membrane (Roche Diagnostics) and hybridized with the DIG-labelled RNA probe under the conditions recommended by the supplier. The hybridized probes on the membrane were detected as chemiluminescent signals using the DIG luminescence detection kit (Roche Diagnostics).

    Enzyme production, purification and assay.

    E. coli BL21(DE3) cells carrying pETiolW or pETiolX were inoculated into TGA medium (Kaempfer & Magasanik, 1967) containing kanamycin and shaken at 37 °C. At an OD600 of 0.35, production of the C-terminally His6-tagged fusion proteins was induced for 2 h by addition of 0.1 mM IPTG. The fusion proteins IolX-(His)6 and IolW-(His)6 were extracted from the cells and purified as described previously (Yoshida et al., 2006). Purified IolX-(His)6 and IolW-(His)6 were subjected to a spectrophotometric enzymic assay for SI dehydrogenase as follows. Five micrograms of the purified enzyme was combined in a 1 ml reaction mixture in 100 mM Tris/HCl (pH 7.5), containing the substrate (SI or SIS) and the cofactor [NAD(P)+ or NAD(P)H] at various concentrations. The mixture was incubated at 37 °C and any increase/decrease in A340 [associated with the reduction/oxidation of NAD(P)+/NAD(P)H] was measured to calculate the reaction rates as described previously (Fujita et al., 1991). Alternatively, to confirm the reaction products, a reaction mixture (100 μl) in 50 mM potassium phosphate buffer (pH 7.5) composed of 50 μg of the purified enzyme [IolX-(His)6 or IolW-(His)6 as indicated], 80 mM SIS (Yoshida et al., 2004) and 80 mM cofactor (NADH or NADPH as indicated) was used. The mixture was then incubated at 37 °C for 120 min and the reaction was terminated by the addition of AG 50W-X8 (H+ form, 200–400 mesh, Bio-Rad) and incubation at 50 °C for 10 min. After the addition of 100 μl water, the supernatant was obtained by centrifugation; an aliquot (20 μl) was subjected to HPLC analysis using a Wakosil5NH2 column (4.6×250 mm) kept at 20 °C with a flow of acetonitrile/water (80/20) at 2 ml min−1; signals for the compounds were detected by a refraction index sensor. Concentrations of the compounds were calculated from the peak area.

    RESULTS

    B. subtilis is able to utilize SI as the sole carbon source using a pathway dependent on a functional iol operon

    To investigate whether B. subtilis can utilize SI, strains of B. subtilis were inoculated into a minimal medium containing SI as the sole carbon source and allowed to grow (Fig. 2a⇓). The results clearly showed that the standard strain 60015 could grow on SI, indicating that this strain possesses an SI catabolic pathway. However, it appeared that SI was not as efficient as MI at supporting B. subtilis growth. The mutant strain YF248 (Piol : : cat), which is defective in the iol operon because of a disruption in the promoter region (Yoshida et al., 1997), did not grow on SI or MI (Fig. 2b⇓), suggesting that SI catabolism in B. subtilis requires a functional iol operon.

    Figure image not available in archive
    Fig. 2.

    Growth of B. subtilis strains using various carbon sources. Cells of the strains indicated were inoculated into S6 minimal medium containing various sole carbon sources [none (•), 25 mM glucose (○), 25 mM MI (▴) or 25 mM SI (▵)], and were then allowed to grow at 37 °C with shaking. Cell growth was monitored by the increase in OD660. Tested strains were (a) 60015 (standard), (b) YF248 (Piol : : cat), (c) BFS3018 (iolX : : pMUTIN4), (d) BFS1058 (iolW : : pMUTIN2) and (e) TM044 [iolX : : pMUTIN4 iolW : : pMUTIN2 (erm : : tet)]. Representative data are shown from three independent experiments that gave similar results.

    MI and SI differ only in their epimerization of the C-2 hydroxyl group (Fig. 1⇑). The first step in the MI catabolic pathway is dehydrogenation of MI at the C-2 position to SIS; this process is catalysed by IolG. To metabolize DCI, IolG also acts on DCI at the C-1 position to produce 1-keto-DCI, which is isomerized to SIS (Fig. 1⇑) (Yoshida et al., 2006). If SI were dehydrogenated at the C-2 position, the same product, SIS, would result; however, it has been shown previously that SI cannot be a substrate for IolG (Ramaley et al., 1979). The interconversion between MI and SI involving SIS as an intermediate was previously shown in Streptomyces griseus (Horner & Thaker, 1968). In addition, a previous study demonstrated that in bovine brain there is an inositol epimerase capable of interconverting MI and SI with formation of SIS as an intermediate (Hipps et al., 1976). Therefore, we hypothesized the existence of a pathway for the interconversion of MI and SI in B. subtilis. The pathway could involve an enzyme responsible for dehydrogenation of SI to SIS, and SIS is readily converted to MI by IolG (Ramaley et al., 1979).

    Two candidate genes for SI dehydrogenase

    As mentioned above, IolG itself does not function as an SI dehydrogenase; however, we hypothesized the existence of an unknown SI dehydrogenase sharing similarities with IolG. Within the B. subtilis genome, at least seven genes paralogous to iolG have been found: yfiI, ntdC (formerly yhjJ), iolX (formerly yisS), yrbE, yteT, yulF and iolW (formerly yvaA). All these iolG paralogues have the sequence motif of an NAD(P)+-binding domain conserved among the oxidoreductase family, including a number of inositol dehydrogenase genes (data not shown). Transcriptome analysis was performed to determine expression levels of the iolG paralogues in the presence of MI or SI (Table 3⇓). It is known that MI induces transcription of iolG. Indeed in the presence of MI iolG gave the highest signal intensity, and in the presence of SI the signal reduced by half. Among the paralogues, iolX, iolW and ntdC (Inaoka et al., 2004) were significantly expressed in the presence of SI and even more enhanced than in the presence of MI, while expression of the others was judged negligible. Since ntdC was previously identified to encode an enzyme involved in the production of the aminosugar antibiotic 3,3′-neotrehalosadiamine (Inaoka et al., 2004), this gene was omitted from further analysis. Northern blot analysis was performed to confirm the transcription of iolX (Fig. 3a⇓) and iolW (Fig. 3b⇓). The specific transcript for iolX was about 1.2 kb long, suggesting that this transcript could be monocistronic. Either MI or SI induced this iolX transcript, but the difference in intensities of the bands suggested that SI was approximately three times more efficient than MI (data not shown), which almost coincided with the ratio obtained in the transcriptome analysis (Table 3⇓). The iolW transcript was 1.3 kb long, probably also monocistronic, and was expressed constitutively even in the absence of both inositols. From these results of expression profiles, we selected iolX and iolW as candidate genes for SI dehydrogenase.

    Figure image not available in archive
    Fig. 3.

    Northern blot analysis of iolX (a) and iolW (b). RNA samples were extracted and purified from cells of B. subtilis 60015 grown in S6 medium containing 0.5 % Casamino acids and supplemented with the following inositols: none (lane 1), 10 mM MI (lane 2) or 10 mM SI (lane 3). Each lane contained 15 μg RNA sample. Preparation of the specific probes is described in the text. The positions of the size markers are indicated on the left. Methylene-blue-stained images of the membranes used to detect 23S and 16S rRNA as a control for the assessment of the quality and quantity of total RNA are shown below.

    Table 3.

    Expression levels of the iolG paralogues in the presence of MI or SI

    Both iolX and iolW encode an SI dehydrogenase

    iolX and iolW encode proteins of 342 and 358 amino acid residues, respectively. Their amino acid sequences share 21.1 % identity in an overlap of 356 residues. Each of the genes was expressed in E. coli as a C-terminally His6-tagged fusion protein, designated IolX-(His)6 and IolW-(His)6, respectively. The two recombinant proteins were purified and assayed for SI dehydrogenase activity. Under the standard conditions for the spectrophotometric enzymic assay, IolX-(His)6 and IolW-(His)6 gave SI-dehydrogenating activity converting SI to SIS of 1.36 and 0.69 μmol min−1 (mg protein)−1 in the presence of NAD+ and NADP+, respectively. No activity was detected with NADP+ for IolX and with NAD+ for IolW (data not shown). When MI was used as an alternative substrate, activities of IolX-(His)6 and IolW-(His)6 appeared as low as 0.34 and 0.39 μmol min−1 (mg protein)−1, respectively, indicating their preference for SI as the substrate. HPLC analysis of the SIS-reducing reaction mixtures revealed that both IolX-(His)6 and IolW-(His)6 reduce SIS to SI, (Fig. 4⇓). The results indicated that both IolX-(His)6 and IolW-(His)6 were SI dehydrogenases, dependent on NAD+/NADH and NADP+/NADPH as cofactors, respectively. Therefore, B. subtilis possesses two distinct types of SI dehydrogenases for interconverting SI and SIS by dehydrogenation/hydrogenation at the C-2 position (Fig. 1⇑).

    Figure image not available in archive
    Fig. 4.

    HPLC analysis of the substrates and products in reactions catalysed by IolX-(His)6 and IolW-(His)6. Reactions were carried out as described in the text. HPLC charts are shown. (a) SIS standard (substrate before the reaction), (b) after the reaction with IolX-(His)6, (c) after the reaction with IolW-(His)6, (d) SI standard (expected product of the reaction). The peaks corresponding to SI and SIS are indicated by arrows.

    Table 4⇓ summarizes the Km and Vmax values of IolX-(His)6 and IolW-(His)6. Both enzymes possessed higher affinities for SIS than for SI and larger capacities for the reduction of SIS to SI than for the dehydrogenation, suggesting similar characteristics as were observed for IolG (Ramaley et al., 1979). IolX could be a catabolic enzyme, since it is induced during growth on SI and since IolX-(His)6 uses NAD+, a known catabolic cofactor. On the other hand, IolW-(His)6 has the smallest Km and the largest Vmax values for SIS and uses NADP+, which is usually a cofactor for anabolic reactions. These results imply that IolW might function to generate SI from SIS, although we cannot explain its physiological role at present.

    Table 4.

    Km and Vmax values for purified IolX-(His)6 and IolW-(His)6

    Inactivation of iolX leads to significantly reduced growth on SI

    Mutant strains BFS3018 and BFS1058 have inactivated iolX and iolW genes, respectively. Another mutant, TM044, was generated that had simultaneous inactivation of both iolX and iolW. We tested the growth of these strains in S6 medium containing SI as the carbon source for the growth. In comparison with the standard strain 60015, BFS1058 (iolW : : pMUTIN2, Fig. 2d⇑) showed normal growth on SI as well as on glucose or MI, but neither BFS3018 (iolX : : pMUTIN4, Fig. 2c⇑) nor TM044 [iolX : : pMUTIN4 iolW : : pMUTIN2(erm : : tet), Fig. 2e⇑] grew on SI specifically. These results clearly indicated that IolX functioned as an SI dehydrogenase to enable growth on SI and that the presence of IolW alone did not support growth on SI at all.

    DISCUSSION

    Our results show that B. subtilis can utilize SI as its sole carbon source and that it has two distinct SI dehydrogenases, IolX and IolW, enabling us to draw a new branch involving SI in the inositol metabolism pathway for B. subtilis (Fig. 1⇑). SI (compound [3]) is believed to be converted into SIS (compound [4]) and then degraded further via the MI/DCI catabolic pathway characterized previously (Yoshida et al., 2008). iolX inactivation abolished growth on SI (Fig. 2c⇑), indicating that IolX functions as the intracellular catabolic enzyme in this metabolic pathway. In the absence of inositols, transcription of iolX was tightly repressed, while it was fully induced in the presence of SI (lane 3 in Fig. 3a⇑). Interestingly, we found that iolX was also partially induced in the presence of MI (lane 2 in Fig. 3a⇑), implying the possible involvement of IolR in the regulation. However, IolR is not likely to play a direct role, since the tight repression of iolX was not released upon inactivating iolR as is found for the other IolR targets (T. Morinaga, unpublished). The partial induction by MI might be due to intracellular production of SI through the possible conversion of MI to SI involving IolG and IolW as discussed below. Within the genome, two genes just upstream of iolX, yisR and degA, were predicted to encode transcription factors listed in the BSORF database (). We are currently investigating the possible involvement of these genes in the regulation of iolX induction, and the results will be reported elsewhere.

    Our in vitro enzymic analysis suggested that both purified IolX-(His)6 and IolW-(His)6 possessed higher affinities for SIS than for SI (Table 4⇑). In addition, both catalysed the reaction in the SIS to SI direction faster than in the opposite direction (Table 4⇑). IolX-(His)6 and IolW-(His)6 indeed converted SIS to SI efficiently in vitro (Fig. 4⇑). Nevertheless, IolX, when present in cells, obviously functioned as a catabolic enzyme, supporting B. subtilis growth using SI as the carbon source (Fig. 2⇑). Therefore, when the cells grow using SI as the carbon source, it is likely that intracellular concentrations of SIS are not high enough to lead to the SIS-hydrogenating reaction catalysed by IolX. However, B. subtilis grew more slowly on SI than on MI (Fig. 1⇑), indicating that SI is a less efficient carbon source than MI. In a previous report (Ramaley et al., 1979), Km and Vmax values for MI in the catabolic reaction, converting MI to SIS, catalysed by IolG were determined under conditions similar to ours (Table 4⇑) to be 18.2±2.4 mM and 21±1.8 μmol min−1 (mg protein)−1, respectively, while those for SIS in the opposite reaction were 1.61±0.12 mM and 41.2±3.1 μmol min−1 (mg protein)−1, respectively. These facts imply that IolG could produce SIS from MI more efficiently than IolX does from SI and that MI is able to support better growth than SI. Another factor limiting growth could be SI uptake into the cell. Two distinct MI transporters in B. subtilis, IolT and IolF, identified previously (Yoshida et al., 2002), are the major and minor MI transporters, respectively. Our preliminary results suggested that inactivation of iolT caused a growth defect not only with MI but also when SI was used as the sole carbon source (K. Yoshida, unpublished). However, the process of SI uptake into B. subtilis cells remains unknown and further studies are required to characterize putative transporters for SI.

    SI did not affect iolW transcription (Fig. 3b⇑), and iolW inactivation did not affect cell growth on SI medium either (Fig. 2d⇑). Our in vitro study indicated that IolW-(His)6 was capable of producing SI from SIS with NADPH oxidation, suggesting that IolW could be involved in SI generation in B. subtilis. Previous reports have demonstrated interconversion of MI and SI in the locust (Candy, 1967), in various bovine species and in S. griseus (Horner & Thaker, 1968). In addition, in the case of S. griseus (Horner & Thaker, 1968) and the bovine species (Hipps et al., 1976), SI was directly synthesized from SIS with NADPH oxidation. Some deep-sea snails contain SI as an osmolyte (Rosenberg et al., 2006), and SI inhibits aggregation of Abeta in mammalian brain (McLaurin et al., 2000). These facts suggest an unknown physiological role of SI in B. subtilis cells, and further studies are required to assess this possibility. In any case, based on our findings, an efficient bioconversion from MI to SI was performed using genetically manipulated B. subtilis coupling the IolG and IolW reactions (K. Yoshida, unpublished). This bioconversion is suggested as a possible way of enabling the inexpensive supply of SI.

    Acknowledgments

    We thank K. Asai, Saitama University, Japan, for plasmid pEm : : Tc. Strains BFS3018 and BFS1058 were supplied by the National Institute of Genetics, Japan, within the framework of the National BioResource Project: E. coli and B. subtilis. This work was supported in part by Nagase Science and Technology Foundation and by KAKENHI (Grant-in-Aid for Scientific Research) on Priority Areas ‘Applied Genomics’, as well as by grants from the Special Coordination Funds for Promoting Science and Technology and the Creation of Innovation Centres for Advanced Interdisciplinary Research Areas (Innovative Bioproduction Kobe) programs of the Ministry of Education, Science and Sports and Culture of Japan.

    References

    1. Berman, T. & Magasanik, B. (1966). The pathway of myo-inositol degradation in Aerobacter aerogenes. J Biol Chem 241, 800–806.
    2. Candy, D. J. (1967). Occurrence and metabolism of scyllo inositol in the locust. Biochem J 103, 666–671.
    3. Fujita, Y. & Freese, E. (1981). Isolation and properties of a Bacillus subtilis mutant unable to produce fructose-bisphosphatase. J Bacteriol 145, 760–767.
    4. Fujita, Y., Shindo, K., Miwa, Y. & Yoshida, K. (1991). Bacillus subtilis inositol dehydrogenase-encoding gene (idh): sequence and expression in Escherichia coli. Gene 108, 121–125.
    5. Galbraith, M. P., Feng, S. F., Borneman, J., Triplett, E. W., de Bruijn, F. J. & Rossbach, S. (1998). A functional myo-inositol catabolism pathway is essential for rhizopine utilization by Sinorhizobium meliloti. Microbiology 144, 2915–2924.
    6. Hipps, P. P., Eveland, M. R., Laird, M. H. & Sherman, W. R. (1976). The identification of myo-inositol : NAD(P)+ oxidoreductase in mammalian brain. Biochem Biophys Res Commun 68, 1133–1138.
    7. Horner, W. H. & Thaker, I. H. (1968). The metabolism of scyllo-inositol in Steptomyces griseus. Biochim Biophys Acta 165, 306–308.
    8. Ichimura, K., Kohata, K., Yamaguchi, Y., Douzono, M., Ikeda, H. & Koketsu, M. (2000). Identification of l-inositol and scyllitol and their distribution in various organs in chrysanthemum. Biosci Biotechnol Biochem 64, 865–868.
    9. Inaoka, T., Takahashi, K., Yada, H., Yoshida, M. & Ochi, K. (2004). RNA polymerase mutation activates the production of a dormant antibiotic 3,3′-neotrehalosadiamine via an autoinduction mechanism in Bacillus subtilis. J Biol Chem 279, 3885–3892.
    10. Jiang, G., Krishnan, A. H., Kim, Y. M., Wacek, T. J. & Krishnan, H. B. (2001). A functional myo-inositol dehydrogenase gene is required for efficient nitrogen fixation and competitiveness of Sinorhizobium fredii USDA191 to nodulate soybean (Glycine max [L.] Merr.). J Bacteriol 183, 2595–2604.
    11. Kaempfer, R. O. & Magasanik, B. (1967). Effect of infection with T-even phage on the inducible synthesis of beta-galactosidase in Escherichia coli. J Mol Biol 27, 453–468.
    12. Kinnard, R. L., Narasimhan, B., Pliska-Matyshak, G. & Murthy, P. P. (1995). Characterization of scyllo-inositol-containing phosphatidylinositol in plant cells. Biochem Biophys Res Commun 210, 549–555.
    13. Krings, E., Krumbach, K., Bathe, B., Kelle, R., Wendisch, V. F., Sahm, H. & Eggeling, L. (2006). Characterization of myo-inositol utilization by Corynebacterium glutamicum: the stimulon, identification of transporters, and influence on l-lysine formation. J Bacteriol 188, 8054–8061.
    14. McLaurin, J., Golomb, R., Jurewicz, A., Antel, J. P. & Fraser, P. E. (2000). Inositol stereoisomers stabilize an oligomeric aggregate of Alzheimer amyloid beta peptide and inhibit abeta-induced toxicity. J Biol Chem 275, 18495–18502.
    15. Morinaga, T., Yamaguchi, M., Makino, Y., Nanamiya, H., Takahashi, K., Yoshikawa, H., Kawamura, F., Ashida, H. & Yoshida, K. (2006). Functional myo-inositol catabolic genes of Bacillus subtilis Natto are involved in depletion of pinitol in natto (fermented soybean). Biosci Biotechnol Biochem 70, 1913–1920.
    16. Ogura, M., Yamaguchi, H., Yoshida, K., Fujita, Y. & Tanaka, T. (2001). DNA microarray analysis of Bacillus subtilis DegU, ComA and PhoP regulons: an approach to comprehensive analysis of B. subtilis two-component regulatory systems. Nucleic Acids Res 29, 3804–3813.
    17. Poole, P. S., Blyth, A., Reid, C. J. & Walters, K. (1994). myo-Inositol catabolism and catabolite regulation in Rhizobium leguminosarum bv. viciae. Microbiology 140, 2787–2795.
    18. Ramaley, R., Fujita, Y. & Freese, E. (1979). Purification and properties of Bacillus subtilis inositol dehydrogenase. J Biol Chem 254, 7684–7690.
    19. Riggs, B. M., Lansley, T. A. & Ryals, P. E. (2007). Phosphatidylinositol synthase of Tetrahymena: inositol isomers as substrates in phosphatidylinositol biosynthesis and headgroup exchange reactions. J Eukaryot Microbiol 54, 119–124.
    20. Rosenberg, N. K., Lee, R. W. & Yancey, P. H. (2006). High contents of hypotaurine and thiotaurine in hydrothermal-vent gastropods without thiotrophic endosymbionts. J Exp Zool A Comp Exp Biol 305, 655–662.
    21. Sambrook, J. & Russell, D. W. (2001). Molecular Cloning: a Laboratory Manual, 3rd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
    22. Sherman, W. R., Stewart, M. A., Simpson, P. C. & Goodwin, S. L. (1968). The identification of myo-inosose-2 and scyllo-inositol in mammalian tissues. Biochemistry 7, 819–824.
    23. Stines-Chaumeil, C., Talfournier, F. & Branlant, G. (2006). Mechanistic characterization of the MSDH (methylmalonate semialdehyde dehydrogenase) from Bacillus subtilis. Biochem J 395, 107–115.
    24. Sun, Y., Zhang, G., Hawkes, C. A., Shaw, J. E., McLaurin, J. & Nitz, M. (2008). Synthesis of scyllo-inositol derivatives and their effects on amyloid beta peptide aggregation. Bioorg Med Chem 16, 7177–7184.
    25. Townsend, M., Cleary, J. P., Mehta, T., Hofmeister, J., Lesne, S., O'Hare, E., Walsh, D. M. & Selkoe, D. J. (2006). Orally available compound prevents deficits in memory caused by the Alzheimer amyloid-beta oligomers. Ann Neurol 60, 668–676.
    26. Vidal-Leiria, M. & van Uden, N. (1973). Inositol dehydrogenase from the yeast Cryptococcus melibiosum. Biochim Biophys Acta 293, 295–303.
    27. Yebra, M. J., Zuniga, M., Beaufils, S., Perez-Martinez, G., Deutscher, J. & Monedero, V. (2007). Identification of a gene cluster enabling Lactobacillus casei BL23 to utilize myo-inositol. Appl Environ Microbiol 73, 3850–3858.
    28. Yoshida, K. I., Aoyama, D., Ishio, I., Shibayama, T. & Fujita, Y. (1997). Organization and transcription of the myo-inositol operon, iol, of Bacillus subtilis. J Bacteriol 179, 4591–4598.
    29. Yoshida, K. I., Shibayama, T., Aoyama, D. & Fujita, Y. (1999a). Interaction of a repressor and its binding sites for regulation of the Bacillus subtilis iol divergon. J Mol Biol 285, 917–929.
    30. Yoshida, K., Fujita, Y. & Ehrlich, S. D. (1999b). Three asparagine synthetase genes of Bacillus subtilis. J Bacteriol 181, 6081–6091.
    31. Yoshida, K., Yamamoto, Y., Omae, K., Yamamoto, M. & Fujita, Y. (2002). Identification of two myo-inositol transporter genes of Bacillus subtilis. J Bacteriol 184, 983–991.
    32. Yoshida, K., Yamaguchi, M., Ikeda, H., Omae, K., Tsurusaki, K. & Fujita, Y. (2004). The fifth gene of the iol operon of Bacillus subtilis, iolE, encodes 2-keto-myo-inositol dehydratase. Microbiology 150, 571–580.
    33. Yoshida, K., Yamaguchi, M., Morinaga, T., Ikeuchi, M., Kinehara, M. & Ashida, H. (2006). Genetic modification of Bacillus subtilis for production of d-chiro-inositol, an investigational drug candidate for treatment of type 2 diabetes and polycystic ovary syndrome. Appl Environ Microbiol 72, 1310–1315.
    34. Yoshida, K., Yamaguchi, M., Morinaga, T., Kinehara, M., Ikeuchi, M., Ashida, H. & Fujita, Y. (2008). myo-Inositol catabolism in Bacillus subtilis. J Biol Chem 283, 10415–10424.