INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Dextransucrases (DS) (EC 2.4.1.5.) are enzymes that transfer the glucosyl moiety from sucrose to acceptor molecules, with a concomitant fructose release. They are used in the synthesis of dextran. In the presence of sucrose and an acceptor like maltose, they synthesize gluco-oligosaccharides (25). Dextran and dextran derivatives have found several valuable applications in the production of fine chemicals such as plasma substitutes and Sephadex. Particularly, gluco-oligosaccharides are used as specialty sugars in the food and cosmetic industries (21).

Several lactic acid bacteria produce DS. Expression is constitutive in Streptococcus strains, while it is inducible in Leuconostoc strains (8). Until now sucrose has been considered to be the only inducer of DS expression in Leuconostoc spp. (32). No gratuitous inducers are known, and the mechanism of DS induction has not yet been reported.

Sugar metabolism in the genus Leuconostoc is heterofermentative. When sucrose is used as a carbon source, a specific permease is responsible for its transport into the cell, where it is transformed by sucrose-phosphorylase into fructose and glucose-1-phosphate. The latter is incorporated into the phosphoketolase pathway as glucose-6-phosphate by the action of a mutase, while fructose is excreted to the culture medium (3). Extracellular DS also uses sucrose for dextran production, with additional fructose liberation. When sucrose is depleted, the accumulated fructose is consumed (23, 32).

Leuconostoc mesenteroides NRRL B-512F produces an extracellular DS that synthesizes a soluble polymer, 95% of which is composed of -(1-6) linkages in the main chain and 5% of which is composed of -(1-3) branched linkages. Only one DS gene in this strain has been reported (33), while DS has been found in multiple forms of different molecular weights (8, 13, 18). There is insufficient genetic evidence to explain if the various proteins found result from the expression of different genes or from posttranslational modifications. There is no information concerning either the DS gene regulation mechanism or the characterization of the transcript. Although constitutive mutants have been obtained by nonspecific mutation strategies (8, 11, 12), the identification of the promoter region in L. mesenteroides DS would allow the construction of constitutive strains by site-directed mutagenesis.

In lactic acid bacteria, some metabolically related genes are organized in clusters or polycistronic operons that are regulated simultaneously (9, 17). Sucrose induces both DS and sucrose-phosphorylase genes in Leuconostoc. However, biochemical data support the fact that these enzymes are induced at different stages during fermentation (3). In this work, genetic evidence to elucidate if both genes are under the control of the same promoter is given. In addition, the production of DS from L. mesenteroides NRRL B-512F under different induction conditions is examined. Through the isolation and characterization of mRNA, molecular information on the transcript is also provided.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Strain conditions. L. mesenteroides NRRL B-512F was kindly provided by the Northern Regional Research Laboratory (NRRL), Peoria, Ill. Three successive cultures were carried out with each of the various carbon sources (see culture conditions). Cells from the exponential growth phase of the third culture were stored in 15% (wt/vol) glycerol at 20°C and used to inoculate subsequent cultures.

Culture conditions. L. mesenteroides was cultured in 100-ml flasks on a rotary shaker at 200 rpm in the standard medium reported by Dols et al. (3) at 25°C unless otherwise specified. For cultures grown with other carbon sources, sucrose was replaced by (i) D-glucose, (ii) equimolar quantities of D-fructose and D-glucose, (iii) D-fructose, and (iv) D-xylose (all purchased from Sigma Chemical Co., St. Louis, Mo.). The carbon source concentration was 50 or 117 mM, as specified below. In induction studies, 50 mM fructose cultures were grown until the mid-logarithmic phase was reached. At this point, 1.8 M sucrose was added to obtain a final concentration ranging from 1 to 102 mM.

Biomass measurements. Bacterial growth was estimated by measuring the absorbance at 600 nm. The optical density value was converted to CFU by means of a calibration graph constructed during the culture on each carbon source. CFU were determined after a 24-h cultivation in plate count agar.

DS recovery and assay. After cell removal, the pH was adjusted to 5.2 and the supernatant was filtered through a membrane with a pore size cutoff of 0.2 µm (Millipore Corp., Bedford, Mass.). Subsequently, DS was concentrated by aqueous two-phase partition with 25% (wt/vol) polyethylene glycol 1500 (24). One-half percent dextran T 70 (Sigma) was included in supernatants produced from carbon sources other than sucrose. After centrifugation (7,000 × g, 20 min, 4°C), the pellet was dispersed in 20 mM acetate buffer (pH 5.4) and DS activity was measured by monitoring the release of reducing sugars by a 3,5-dinitrosalicylic acid assay (31). One unit of DS activity is defined as the amount of enzyme that produces 1 µmol of fructose per min from a 100-g · liter¹ sucrose solution at 30°C in 50 mM sodium acetate buffer (pH 5.4) containing 0.05 g of CaCl₂ and 1 g of NaN₃ · liter¹. Specific activity is given as units per gram of total culture protein. Protein was determined after precipitation with 10% (wt/vol) trichloroacetic acid, followed by dispersion in 0.1 N NaOH. Quantification of the soluble proteins was made as described by Lowry et al. (16), with bovine serum albumin as a standard. Unless otherwise specified, all experiments were carried out in triplicate. The variation coefficients were less than 5% in all cases.

Protein electrophoresis and in situ activity analysis. Supernatants from D-glucose and D-fructose cultures were concentrated approximately 40 times by centrifugal ultrafiltration with Centricon 30 tubes (Amicon Inc., Lexington, Mass.). DS from sucrose cultures was analyzed without further concentration. Protein samples were applied in parallel to sodium dodecyl sulfate (SDS)-7% polyacrylamide gels (14). After electrophoresis at a constant current of 30 mA, the gel was cut in two and one-half was stained with Coomassie R-250. The molecular mass was estimated with the High Range SDS-polyacrylamide gel electrophoresis (PAGE) molecular weight standards (Bio-Rad Laboratories, Hercules, Calif.). The other half of the gel was washed and incubated in the presence of sucrose for the in situ DS assay as previously described (20). For a specific levansucrase assay, raffinose was used as a substrate instead of sucrose.

Oligosaccharide and dextran synthesis. Oligosaccharide synthesis was carried out at 30°C with 100 g of sucrose · liter¹ and 33.3 g of maltose · liter¹ in a solution of 20 mM sodium acetate buffer (pH 5.4) containing 0.05 g of CaCl₂ · liter¹, 1 g of NaN₃ · liter¹, and 0.25 U of DS. For dextran synthesis a reaction mixture with the same composition, but lacking maltose, was used. In all cases, DS was inactivated at 75°C.

Carbohydrate analysis. D-Glucose and D-fructose concentrations were determined by an enzymatic UV method (Boehringer Mannheim GmbH, Mannheim, Germany). Sucrose was determined by the same method after treatment with invertase (Sigma). Oligosaccharide analysis was carried out by high-pressure liquid chromatography (HPLC) in a Waters-Millipore C₁₈ column equipped with a refractive index detector as previously described (19). Dextran analysis was performed after polymer precipitation with 2 volumes of absolute ethanol; the pellet was recovered by centrifugation and washed three times with deionized water before being freeze-dried. ¹³C nuclear magnetic resonance (NMR) spectra of the polymer were obtained with an AC300 Bruker spectrometer, at 75.4768 MHz, as described by Dols et al. (4). The chemical shifts were assigned to each carbon according to the method of Seymour et al. (27).

RNA isolation and hybridization analysis. For RNA isolation, 10⁹ cells were washed twice and incubated for 30 min at 37°C with 4 × 10³ mg of lysozyme (Sigma) µl¹ and for 1 h with 1% (vol/vol) proteinase K (Boehringer Mannheim GmbH). The isolation procedure was then continued by following the guanidinium thiocyanate method (2) in combination with acidic phenol extraction and treatment with DNase I (amplification grade; Gibco BRL, Rockville, Md.). The molecular weight marker (RNA ladder; Gibco BRL) and 7 µg of total RNA of each sample were separated by electrophoresis with a denaturing formaldehyde-agarose system. Afterwards, the samples were transferred and fixed to a Hybond N nylon membrane (Amersham Corp., Arlington Heights, Ill.) by applying the standard procedure (5). RNA blotted membranes were hybridized according to the manufacturer's instructions with 10 to 20 ng of the DNA probe labeled with ³²P by using the Megaprime DNA labeling system (Amersham). The probe was obtained from the L. mesenteroides NRRL B-512F DS gene described by Wilke-Douglas et al. (33) (Calgene Inc., Davis, Calif.) after digestion with SalI and NdeI (Gibco BRL). The enzyme digestion gave one 1.13-kb fragment that includes the region encoding the catalytic domain previously reported (7, 22, 28).

mRNA 5'-end determination. RNA analysis was carried out with the system for rapid amplification of cDNA 5' ends (Gibco BRL) by following the manufacturer's procedure, which consists of cDNA synthesis and cDNA 3'-end amplification by PCR. cDNA was obtained with Superscript II Reverse Transcriptase (Gibco BRL) and the synthetic oligonucleotide 5'-GATCCGTGAATGCATACCCG-3', which is complementary to a conserved sequence in the N-terminal region of the DS gene (33). cDNA 3' ends were amplified with Taq DNA polymerase (Gibco BRL) with the gene-specific primer shown in Fig. 5. The PCR amplification conditions were one cycle of 94°C for 1 min; 30 cycles of 94°C for 1 min, 55°C for 1 min, and 72°C for 1 min; and one final cycle of 72°C for 5 min. The reaction products were purified with a High Pure PCR product purification kit (Boehringer Mannheim GmbH) before being sequenced.

Nucleotide sequence accession numbers. The following accession numbers have been assigned by the EMBL nucleotide sequence database: AJ250903 and AJ250904 (artificial oligonucleotide complementary primer used for gene sequence and oligonucleotide sequence used for the rapid amplification of cDNA 5' ends, respectively).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

L. mesenteroides NRRL B-512F DS synthesis with several carbon sources. Batch fermentation evolution under standard DS production conditions (29°C and 117 mM sucrose) is shown in Fig. 1. Fructose was released during the first four hours and was later consumed once sucrose was depleted. DS activity reached a maximum of 1.8 U · ml¹ at the end of the exponential growth phase, followed by a remarkable decrease in activity closely related to culture acidification.

View larger version (23K): [in this window] [in a new window]   FIG. 1.   Batch fermentation profile of L. mesenteroides NRRL B-512F under standard conditions at 29°C. , sucrose; , fructose; *, pH; , DS activity; , optical density (O.D.).

Protein characterization. Electrophoretic analyses were performed with supernatants, with glucose or fructose as carbon sources, and in situ activity assays were carried out to distinguish protein bands able to synthesize a polymer from sucrose. A supernatant obtained under the standard DS production conditions was used as a reference (Fig. 2, lane 1). In the stained gel, two high-molecular-mass bands can be observed (Fig. 2A): an intense band of 170 kDa and a faint one of 160 kDa. Protein profiles after Coomassie staining of proteins from glucose or fructose supernatants were similar to the one from the sucrose culture. After the in situ activity assay was performed (Fig. 2B), two bands of 170 and 116 kDa with polymer-synthesizing activity could be observed (lanes 2, 3, and 4). An additional low-activity band of 160 kDa could be observed in lane 4, and faint bands of 97 kDa were also observed (lanes 2 and 3). While the 170- and 160-kDa bands were distinguished after incubation for 24 h, longer incubation times were required to observe the 116- and 97-kDa activity bands.

View larger version (68K): [in this window] [in a new window]   FIG. 2.   SDS-PAGE analysis of L. mesenteroides NRRL B-512F DS obtained from cells grown with different carbon sources. (A) Coomassie blue-stained gel. Lane MW, molecular mass markers; lane 1, supernatant from sucrose. (B) In situ polymer production from sucrose. Lane 2, supernatants from fructose culture; lane 3, supernatant from glucose culture; lane 4, supernatant from sucrose culture.

Analysis of the DS products. The dextran ¹³C-NMR analysis and the HPLC profile of the acceptor products synthesized by the enzymes obtained from sucrose, fructose, or glucose medium are shown (Fig. 3). It may be observed that the oligosaccharide profile and the polymer structures are the same.

View larger version (18K): [in this window] [in a new window]   FIG. 3.   Analysis of products synthesized by DS obtained from cells grown in fructose (A), glucose (B), and sucrose (C). (Graphs I) HPLC chromatogram of the oligosaccharides produced. These are designated DPn, with n being the oligosaccharide degree of polymerization (DP). (Graphs II) 13C-NMR spectra of dextran synthesized. *, Carbons involved in the -(1-6) linkage. Reaction and analysis conditions are reported in Materials and Methods. mv, millivolts.

Induction experiments. In order to explore the induction effect of sucrose, L. mesenteroides was initially grown under low-level-enzyme-producing conditions with D-fructose, D-glucose, and D-xylose as carbon sources. At the mid-logarithmic stage the cells were washed and transferred to a fresh 117 mM sucrose standard medium for DS production. Appropriate cell densities were reached in order to allow the comparison of results. Before sucrose induction, the highest activity obtained was 0.011 U · mg of protein¹, corresponding to the cells grown in glucose. In all cases, the DS activity was increased after the transfer to the sucrose medium and it could be detected only after 3 h of incubation with sucrose. The highest activity was obtained from cells first grown in xylose (0.569 U · mg of protein¹), while the lowest DS expression was observed when the cells were initially grown in fructose.

DS transcription analysis. In order to evaluate the DS messenger in terms of size and level, total RNA was extracted from cells grown under different culture conditions and analyzed by Northern blotting. Well-defined rRNA bands were observed in the denaturing gel (Fig. 4A), indicating a good RNA preparation quality. The hybridization analysis (Fig. 4B) showed that the size of the DS mRNA was approximately 4.8 kb and that the highest concentration was found in the exponential growth phase of the sucrose culture. A fainter hybridization signal was observed in the sample obtained from the lag phase of the same culture. Hybridization bands were not observed for RNA samples from cells grown in alternative sugars and stationary-phase sucrose-grown cultures. They became visible after a longer time exposure, at which time the hybridization signal from the log-phase sucrose culture RNA was extremely high (results not shown).

View larger version (96K): [in this window] [in a new window]   FIG. 4.   Analysis of RNA samples extracted from cells grown in fructose (lanes 1), glucose (lanes 2), sucrose lag phase (lanes 3), sucrose log phase (lanes 4), and sucrose stationary phase (lanes 5). (A) Total RNA denaturing formaldehyde-agarose gel electrophoresis. (B) Autoradiogram obtained from the Northern blot. Lanes MW contain the Gibco RNA ladder. In all cases, 7 µg of RNA was analyzed.

View larger version (36K): [in this window] [in a new window]   FIG. 5.   Nucleotide sequence of the N-terminal DS gene and its preceding region. The 10 and 35 promoter regions are underlined, and the transcription start site and direction of transcription are indicated by arrows. SD is the possible ribosome-binding site. The boxed nucleotides correspond to the primer used for PCR amplification of the mRNA 5' end.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

DS yields obtained with L. mesenteroides NRRL B-512F grown in sucrose were similar to what has been reported previously (8). An important loss of activity occurred when the pH fell to values that were lower than 5.0 because DS are active in a pH range between 4.8 and 6.2 (18). The results presented here demonstrate that there is a substantial reduction in DS mRNA expression at this moment. Therefore, it may be concluded that at this stage there is activity loss due to enzyme inactivation, which is irreversible according to Miller et al. (18), but also due to the absence of DS gene transcription.

The experiments whose results are reported in Table 1 demonstrate the evident inducing role of sucrose. However, a low-level-induction effect of D-glucose and D-fructose was observed, since DS activities could be detected in the concentrated supernatants. The enzyme yield obtained when D-xylose was used as the carbon source represents the basal DS level. The different enzyme concentration obtained from the cultures with sucrose compared to that of the glucose-fructose mixture might indicate a selectivity difference in the regulatory mechanism. It is interesting that xylose in Leuconostoc is assimilated through D-xylulose-5P but that glucose or fructose is assimilated through the phosphoketolase pathway (3). Moreover, as mentioned before, the sucrose uptake pathway differs in its first steps from that of fructose and glucose metabolism. Therefore, differences in enzyme activity might be explained by the presence of a metabolite that plays a role as an activator of gene expression. A molecule involved in the initial sucrose uptake or initial metabolic steps may be such an activator.

It was verified that the main protein bands found in supernatants of all carbon sources studied are DS. The protein of 170 kDa corresponds to the product of the gene described by Wilke-Douglas et al. (33), and the 160-kDa protein corresponds to a DS previously reported (8, 18). We have recently found a very low proteolytic activity in this strain, which could be detected in the concentrated supernatant (26). This result suggests that the 160-kDa band may be produced from digestion of the original 170-kDa protein. The 116- and 97-kDa proteins correspond to levansucrases. This fact was verified with raffinose (specific levansucrase substrate) in an in situ assay, where only these bands were observed (result not shown). Levansucrases of the same molecular mass were also reported by Miller et al. (18). Due to the very small amount of levansucrase, the polymer production was observed only after several days of incubation.

From the analyses of the DS products (dextran and oligosaccharides), it may be concluded that the enzymes obtained in media with different carbon sources have the same specificity. That is, the ¹³C-NMR spectra of the polymers synthesized with each DS were similar to that of an -(16)-linked linear dextran (20). In all three enzymes, glucosyl is specifically transferred to maltose, producing a series of -(1-6)-linked oligosaccharides. Accordingly, we conclude that the enzymes obtained in media with different carbon sources are the same in terms of protein size and product specificity. It is interesting that although some levansucrase activity was detected in the electrophoretic assay, no levansucrase products were observed in the polymer synthesis reaction, because of the very high dextransucrase/levansucrase ratio.

A classical induction phenomenon requires contact with the cells and the inducer for only a few minutes to allow gene expression. In this case, sucrose behavior as an inducer is atypical since DS activities could be detected only after several hours of contact with sucrose and since the sucrose concentration required to stimulate enzyme production was extremely high (Table 2). These results also show that the growth of three generations in the presence of sucrose was not enough to recover the DS activity levels reached by cells that had always grown in this carbon source.

The correlation between DS mRNA amount and enzyme activity produced under different culture conditions confirms that gene regulation occurs at the transcriptional level. Northern blotting shows that even in the first hour of the sucrose culture, the amount of DS mRNA is considerably higher than the maximum obtained with any other carbon source, a fact supporting the activator hypothesis.

The low enzyme activity of cells transferred to a sucrose medium after growth in fructose may be explained by a fructose repression effect. However, when time-related gene expression was analyzed, it was found that the largest amount of mRNA was observed when 20 mM fructose and 55 mM sucrose were present in the culture, after 3.5 h of fermentation (Fig. 1). These results are consistent with the enzyme production behavior in fed-batch cultures, where in spite of the high fructose concentrations reached, an increase in DS production was obtained. According to López and Monsan, the sucrose concentration should be kept between 15 and 30 mM in order to maintain the microorganism at the maximum growth rate (15). When those results are compared to the ones obtained in this work, it may be concluded that under such culture conditions, the microorganism is also kept at the maximum stage of mRNA synthesis, despite fructose accumulation.

The size of the DS messenger corresponds to the size of the previously reported gene (33), so it is possible to conclude that the DS gene of the B-512F strain is located in a monocistronic operon. This possibility also explains the differences found by Dols et al. (3) in the expression of DS and sucrose-phosphorylase during the culture time, as they claimed that these enzymes were not coinduced by their common substrate. As the 5' ends of all the analyzed mRNAs were the same, it is concluded that only one gene is transcribed under any culture condition.

Six putative glucosyltransferase promoter sequences have been reported for Leuconostoc (19, 33) and Streptococcus (1, 6, 10, 29) species. Only one fructosyltransferase promoter sequence, from Streptococcus mutans, has been determined experimentally (30). In this work, one transcription start point was found 34 bp upstream from the ATG start codon. The DS promoter presents the sequence TATAAT in the 10 region, which is totally homologous to the conserved region in prokaryotic cells and the reported region for S. mutans. The 35 region, TTTACA, presents high homology to the hexamer consensus sequence T₈₂T₈₄G₇₈A₆₅C₅₄A₄₅, and the sequence reported for S. mutans has four additional base substitutions. The identification of the DS promoter sequence will allow further studies of the gene regulation mechanism in lactic acid bacteria and allow the rational construction of constitutive mutants by site-directed mutagenesis techniques.