Sucrose fermentation by Fusobacterium mortiferum ATCC 25557: transport, catabolism, and products

Studies of sucrose utilization by Fusobacterium mortiferum ATCC 25557 have provided the first definitive evidence for phosphoenolpyruvate-dependent sugar:phosphotransferase activity in the family Bacteroidaceae. The phosphoenolpyruvate-dependent sucrose:phosphotransferase system and the two enzymes required for the dissimilation of sucrose 6-phosphate are induced specifically by growth of F. mortiferum on the disaccharide. Monomeric sucrose 6-phosphate hydrolase (M(r), 52,000) and a dimeric ATP-dependent fructokinase (subunit M(r), 32,000) have been purified to electrophoretic homogeneity. The physicochemical and catalytic properties of these enzymes have been examined, and the N-terminal amino acid sequences for both proteins are reported. The characteristics of sucrose 6-phosphate hydrolase and fructokinase from F. mortiferum are compared with the same enzymes from both gram-positive and gram-negative species. Butyric, acetic, and D-lactic acids are the end products of sucrose fermentation by F. mortiferum. A pathway is proposed for the translocation, phosphorylation, and metabolism of sucrose by this anaerobic pathogen.

Members of the genus Fusobacterium are contributory or causative agents for a variety of infections in animals and humans (for reviews, see references 21 and 28). These gram-negative anaerobes are frequently isolated (often in mixed culture) from necrotic tissues and abscesses (6,7). The prevalence and increased numbers of certain species in patients with gingivitis and periodontitis suggest participation of these organisms in the etiology of oral disease (40).
Considerable attention has been given to the classification, methods of isolation, enumeration, ecology, and pathogenic potential of fusobacteria. However, much less is known of the biochemistry, genetics, or regulation of energy metabolism in these species. Under laboratory conditions, most species thrive in media containing Trypticase, peptone, or yeast extract, and amino acid fermentations provide the energy necessary for growth (2,14,23,32,36). In contrast to the well-documented pathways for the metabolism of amino acids (e.g., glutamic acid [8] and lysine [3]), reports of sugar dissimilation by fusobacteria are few and controversial (10,22,32,43). Some of the controversial aspects may be explained by our recent discovery that, for most species, the transport, phosphorylation, and formation of an endogenous sugar polymer is dependent (or markedly enhanced) upon provision of a fermentable amino acid as the energy source (43)(44)(45).
Bergey's Manual of Systematic Bacteriology (30) and the Virginia Polytechnic Institute Anaerobe Laboratory Manual (29) describe these organisms as asaccharolytic or weakly fermentative. A recent study in our laboratory (44) of sugar utilization by eight Fusobacterium species has largely confirmed the earlier descriptions. Interestingly, however, those strains of Fusobacterium mortiferum examined in our survey grew well on a variety of sugars, and accumulation of these sugars by resting cells was not dependent upon concomitant fermentation of an amino acid. By virtue of its capacity to metabolize carbohydrates, F. mortiferum seemed to us to be an ideal choice for delineation of the * Corresponding author. regulatory mechanisms for sugar transport and metabolism within the genus.
In this article, we describe the mode of entry and the probable pathway for sucrose fermentation by F. mortiferum ATCC 25557. The first definitive evidence is presented for operation of the phosphoenolpyruvate-dependent sugar: phosphotransferase (sugar-PEP:PTS) system (42) in the genus Fusobacterium. Finally, we show, in contrast to the homolactic fermentation of sucrose by other microorganisms, that butyric and acetic acids are the major products of sucrose catabolism by F. mortiferum.

MATERIALS AND METHODS
Organism and culture maintenance. F. mortiferum ATCC 25557 was obtained from the American Type Culture Collection (Rockville, Md.). The organism was maintained by semimonthly transfer in the modified thioglycolate medium (Fisher Scientific Co., Springfield, N.J.) described previously (43).
Growth of cells. F. mortiferum was grown anaerobically (GasPak; BBL Microbiology Systems, Cockeysville, Md.) at 37°C in modified Todd-Hewitt broth (45) supplemented with 0.25% (wt/vol) sucrose. Prereduced medium was inoculated with 10% (vol/vol) stationary-phase culture, and the cells were grown to the stationary phase (ca. 22 h). The yield was approximately 3.5 to 4.0 g (wet weight) of cells per liter.
Harvesting and preparation of cells. Cells were harvested by centrifugation (13,000 x g for 10 min at 4°C), and the cell pellets were washed by resuspension and centrifugation from 50 mM potassium phosphate buffer (pH 7.0) containing 0.7 mM MgCl2. Washed cells were stored aerobically at -20°C until required for enzyme purification.
Enzyme purification. Sucrose 6-phosphate hydrolase (S6PH) and fructokinase (FK) were purified by conventional column chromatography, and, unless otherwise stated, all procedures were conducted at 4°C. Column flow rates were maintained by a P-1 peristaltic pump interfaced with a Frac 100 collector (Pharmacia LKB Biotechnology Inc., Piscataway, N.J.). Column eluents were monitored at 280 nm by a UV-1 optical control unit connected to a single-channel chart recorder (REC-481; Pharmacia LKB).
Cell breakage and preparation of extracts. Frozen cells of F. mortiferum (about 80 g [wet weight]) were thawed in an equal weight of 50 mM potassium phosphate buffer (pH 7) containing 1 mM dithiothreitol (DTT). Cells were disrupted (at 0°C) by two 1-min periods of sonic oscillation by using a Branson sonifier operating at 75% maximum power. Five milligrams each of RNase and DNase was added to the sonicate, and, after stirring for 15 min at 4°C, the residual whole cells and cell debris were removed by centrifugation (25, Purification of S6PH. (i) Hydroxylapatite chromatography. Fractions from DEAE-Sephacel chromatography which contained S6PH activity were pooled and concentrated to 14 ml with a PM 10 ultrafiltration membrane (Amicon Corp., Lexington, Mass.). The sample was desalted and exchanged for 10 mM potassium phosphate buffer (pH 7) by passage through a PD-10 gel filtration column. The preparation was then applied (flow rate, 0.5 ml/min) to a hydroxylapatite column (2.6 by 5.5 cm) previously equilibrated with 10 mM potassium phosphate buffer (pH 7). Nonadsorbed materials were eluted with 130 ml of equilibration buffer. A 400-ml linear gradient of potassium phosphate (0 to 300 mM, pH 7) was passed through the column, and fractions of 4.5 ml were collected. Fractions containing the highest S6PH activities (31 through 34; [phosphate], =0.11 M) were pooled and concentrated to 2 ml by ultrafiltration through a membrane (PM-10, Amicon Corp.).
(ii) Ultrogel AcA54 chromatography. The 2-ml concentrate from step (i) was loaded (flow rate, 0.18 ml/min) onto a gel filtration column (2.6 by 94 cm) of Ultrogel AcA54 previously equilibrated with 50 mM potassium phosphate buffer (pH 7) containing 0.1 M NaCl. The same buffer at a flow rate of 0.3 ml/min was used for elution of S6PH, and 5-ml fractions were collected. Enzyme activity was usually highest in fractions 48 through 51. These fractions were pooled and concentrated to about 2 ml. (iii) DEAE-Sephacel chromatography (pH 6.0). After exchanging the buffer in the sample for 10 mM bis-Tris (pH 6), the preparation was applied to a second DEAE-Sephacel column (1 by 24 cm) at a flow rate of 0.5 ml/min. Elution of S6PH was achieved by passage of 300 ml of a linear NaCl gradient (0 to 0.25 M) in 10 mM bis-Tris (pH 6). Threemilliliter fractions were collected, and those containing the highest S6PH activities (fractions 61 through 75) were pooled and concentrated to 1 ml. This final step of the purification usually yields 200 to 300 ,ug of S6PH.
Purification of FK. FK from F. mortiferum was purified to homogeneity by the following procedures.
(i) Hydroxylapatite chromatography. The previously described DEAE-Sephacel (pH 7.5) concentrate (fractions 48 through 53; approximately 18 ml) was dialyzed overnight against 4 liters of 25 mM potassium phosphate buffer (pH 7) containing 1 mM D1T. The preparation was transferred at a flow rate of 0.4 ml/min to a hydroxylapatite column (2.6 by 5.5 cm) previously equilibrated with dialysis buffer. The column was washed with 2 column volumes of buffer, and then FK was eluted with 400 ml of a linear potassium phosphate (pH 7) gradient (25 to 400 mM) (flow rate, 0.4 ml/min). Fractions of 3 ml were collected, and fractions 12 through 18 (which contained the highest FK activities) were pooled and concentrated to 3 ml by ultrafiltration.
(ii) Phenyl-Sepharose CL-4B chromatography. The 3-ml concentrate from step (i) was made 1 M in NaCl, and the solution was applied (flow rate, 0.2 ml/min) to a column (1 by 28 cm) of phenyl-Sepharose CL-4B equilibrated with 10 mM potassium phosphate buffer (pH 7) containing 1 M NaCl. The column was washed with equilibration buffer, and 3-ml fractions were collected. A large protein peak was eluted (in fractions 7 through 17), but FK activity was separated from this material and the enzyme was present in fractions 25 through 42. Fractions containing the highest FK activities were pooled and concentrated to 2 ml.
(iii) Ultrogel AcA54 chromatography. The 2-ml concentrate was applied to the same gel filtration column used for the purification of S6PH. FK was eluted at a flow rate of 0.3 ml/min, and 3-ml fractions were collected. Peak fractions (79 through 82) were pooled and concentrated to 1 ml to yield about 250 ,ug of enzyme. Enzyme assays. S6PH and FK activities were determined by the glucose 6-phosphate (G6P) dehydrogenase-phosphohexose isomerase-NADP+-linked spectrophotometric assays described previously in references 57 and 58, respectively. One unit of S6PH or FK is that amount of enzyme which catalyzes the reduction (A340) of 1 jmol of NADP+ per min at room temperature (ca. 22°C).
Protein assay. Protein was routinely determined by the dye-binding method of Bradford (5) by using the Bio-Rad protein assay reagent and bovine serum albumin as the standard. Concentrations of protein in dilute solutions were estimated by the E28J/E260 ratio.
Determination of molecular weight. The relative molecular masses (Mr) of the native enzymes were determined by the gel filtration procedure of Andrews (1). A column (2.6 by 94 cm) of Ultrogel AcA54 was equilibrated with 50 mM potassium phosphate buffer (pH 7) containing 0.1 M NaCl. The column was calibrated with 5 mg each of the following standard proteins (in decreasing Me): bovine serum albumin, 67,000; ovalbumin, 43,000; carbonic anhydrase, 31,000; chymotrypsinogen A, 25,000; and RNase A, 13,700. The sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) procedure of Laemmli (35) was used to monitor enzyme purity during purification and to determine the molecular weights of denatured enzymes and subunits. Lowmolecular-weight protein standards were purchased from Bio-Rad. Conditions for electrophoresis and staining procedures have been described previously (55).
IEF. Isoelectric points of S6PH and FK were determined by analytical isoelectric focusing (IEF) (at 10°C). The pI calibration kit (pH range, 3 to 10), precast Ampholine PAG plates (pH range, 3.5 to 9.5), and the Multiphor model 2117 flat-bed electrophoresis unit were supplied by Pharmacia LKB. IEF gels were fixed, stained with Coomassie blue R-250, and destained according to the LKB instruction guide no. 1804.
In situ staining for FK and S6PH activities in gels. For the qualitative determination of FK activity in situ, cell extracts were first electrophoresed in native, anionic gels (58) by the method of Davis (12). FK activity in both native and IEF gels was localized by the G6P dehydrogenase-phosphohexose isomerase-NADPH-coupled reduction of Nitro Blue Tetrazolium as described by Gabriel (18). S6PH activity in IEF gels was localized by the method of Gabriel and Wang (19). In this procedure, the reducing sugars (glucose and fructose) formed by hydrolysis of sucrose or sucrose 6-phosphate (S6P) yield a deep red color in the presence of a hot solution of 1 N NaOH containing 0.1% (wt/vol) 2,3,5triphenyltetrazolium chloride. N-terminal amino acid sequences. N-terminal amino acid sequences were obtained by automated Edman degradation on an Applied Biosystems model 470A gas-phase sequencer as described previously (55).
Sucrose utilization by washed cells. Washed cells (equivalent to about 35 mg of cell protein) were resuspended, under anaerobic conditions, in 10 ml of a solution containing 0.1 M potassium phosphate buffer (pH 7), 10 mM sucrose, and 5 mM MgCl2. At intervals, 2.5-ml samples were removed by a syringe flushed with an anaerobic gas mixture (5% CO2, 5% H2, 90% N2), and the cells were collected by rapid (ca. 30-s) microcentrifugation at 4°C. The clarified supernant was removed and filtered through a Millex-GS (0.22-,um-poresize) filter (Millipore Corp.). Filtrates were assayed for glucose, fructose, and sucrose by the Boehringer Mannheim Biochemicals sucrose analysis kit, and fermentation products (butyric, acetic, and D-lactic acids) were determined by the high-pressure liquid chromatography procedures described in a previous communication (45).
Cell permeabilization and sucrose-PEP:PTS assay. Washed cells, previously grown on sucrose, were resuspended under anaerobic conditions in 0.1 M Tris-HCl buffer (pH 8) to a final concentration of 150 mg (wet weight) of cells per ml. To 2 ml of this suspension was added 20 ,u1 of a 1:9 (vol/vol) mixture of toluene-acetone. The suspension was agitated vigorously for 1 min on a vortex mixer, and the procedure was repeated with two further additions of toluene-acetone. The suspension of permeabilized cells was maintained at 0°C. Assays for sucrose phosphorylation were performed at room temperature (ca. Radiolabeled sugar phosphate(s) retained on the filters was determined by liquid scintillation spectrometry.
Reagents. Radiolabeled sucrose was purchased from Dupont-NEN Research Products (Boston, Mass.), and [U-14C]S6P was prepared in our laboratory (57). SDS-PAGE standards, protein calibration kits for gel filtration chromatography, IEF standards, Ampholine PAG plates, DEAE-Sephacel, and phenyl-Sepharose CL-4B were obtained from Pharmacia LKB. Ultrogel AcA54 was supplied by IBF Biotechnics Inc., Savage, Md. Enzymes and other reagents were purchased from Sigma. Sucrose-PEP:PtS activity in F. mortiferum. Several grampositive and gram-negative species phosphorylate sucrose (to S6P) simultaneously with translocation of this disaccharide via a sucrose-PEP:PTS system. Evidence for the presence of this inducible multicomponent system in F. mortifernm is presented in Fig. 2  also, reference 44). The detection and subsequent purification of both enzymes from F. mortiferum provided supportive evidence for sucrose-PEP:PTS activity in this organism.
Purification of S6PH. The five-stage purification (Table 1) yielded approximately 200 ,ug of pure enzyme from 40 g (wet weight) of sucrose-grown cells of F. mortiferum. This recovery represents an overall yield of -20% and a purification of about 2,300-fold. Hydroxylapatite chromatography was particularly effective, and this step alone resulted in a 40-fold increase in specific activity of the enzyme. Electrophoretically homogeneous S6PH was obtained from the final step of the purification (Fig. 3A). Microsequencing of S6PH allowed the identification of the first 33 amino acid residues from the NH2 terminus: Met-Asp-Lys-Lys-Glu-Tyr-Ile-Asn-Arg-Leu-Asn-Glu-Gln-Lys-(Pro)-Leu-Gln-Glu-Met-(X)-Ser-Arg-Asp-Arg-Tyr-Leu-(X)-Asn-Phe-His-Leu-Ile-Pro-.
Properties of S6PH from F. mortiferum. Activity of S6PH (optimum in 0.1 M potassium phosphate buffer [pH 7.0]) was maintained after freezing (-20°C) and thawing the sample at least four times. The purified enzyme was stable to lyophilization and reconstitution in buffer and exhibited no specific metal requirements. Although the relative molecular mass (Mr) determined by gel filtration chromatography (Ultrogel AcA54) was approximately 44,000, SDS-PAGE revealed only a single polypeptide with an Mr of 52,000 (Fig. 3A). S6PH was stable to heating for 10 min within the temperature range of 20 to 50°C. At higher temperatures, S6PH activity was progressively lost, and the enzyme was inactivated by heating at 65°C for 8 min (half-time of inactivation [t,12], 45 s). Analytical electrofocusing of purified S6PH (Fig.   3B) revealed a single polypeptide (pl, 4.7) which, by activity staining, catalyzed the in situ hydrolysis of sucrose (data not shown).
Substrate specificity and kinetic parameters of S6PH. Both sucrose and S6P were hydrolyzed by S6PH from F. mortiferum, and simple Michaelis-Menten saturation kinetics were evident from studies of initial rates versus substrate concentration (data not shown). The kinetic parameters estimated from double-reciprocal plots were as follows: Km(sucrose)q =40 mM; Vmax ucrose), 43 ,umol of sucrose hydrolyzed mg of protein-' min-; Km(S6P), 0.28 mM; and Vmax(S6P), 270 pxmol of S6P hydrolyzed mg of protein-' min-'. Stoichiometric analyses of the reaction products established that sucrose was hydrolyzed to glucose and fructose and that S6P was hydrolyzed to yield equimolar concentrations of G6P and fructose (data not shown). Lactulose (4-O-p-D-galactopyranosyl-U-D-fructose) was slowly hydrolyzed by S6PH (-1 ,umol mg of protein-' min-'), but there was no detectable hydrolysis of lactose, melibiose, trehalose, or raffinose.
FK: inducibility and purification. Previous enzymatic analyses (44) revealed ATP-dependent FK activity in sucrosegrown cells of F. mortiferum. When extracts from cells grown on various sugars (maltose, sucrose, mannose, galactose, glucose, or fructose) were electrophoresed and stained in situ for FK activity, a high level of a single ATPdependent FK was detected only in sucrose-grown cells  (data not shown). Purification of FK was achieved by the five-stage procedure summarized in Table 1. The enzyme was purified about 900-fold to a specific activity of 74 U mg of protein-' for a recovery of about 8%. Usually, 200 to 300 pg of pure protein was obtained from 80 to 90 g (wet weight) of cells.
Molecular weight of FK. The molecular weight of the native enzyme was estimated to be about 60,000 by gel filtration (Ultrogel AcA54) chromatography, but SDS-PAGE revealed a single polypeptide with an Mr of =32,000 (Fig.  3C). Automated Edman degradation provided one unambiguous sequence for the first 27 amino acid residues from the NH2 terminus: Met-Ile-Ile-Gly-Ala-Val-Glu-Ala-Gly-Gly-Thr-Lys-Phe-Val-Asp-Gly-Val-Gly-Asn-Glu-Lys-Gly-Glu-Ile-Phe-Glu-Arg-. These data attest to the purity of the preparation and show that, in the native state, FK exists as a homodimer of noncovalently linked subunits (Mr, 32,000).
Properties and enzyme stability. The ATP-dependent FK had a broad activity maximum between pH 6.5 and 7.5 in 0.1 M phosphate, bis-Tris, HEPES, and imidazole buffers. Enzyme activity was maintained during storage at -20°C for at least 1 month, and activity was retained after lyophilization and reconstitution of the enzyme in buffer. IEF (Fig. 3D) revealed a single polypeptide (pI, -5.1) which catalyzed the ATP-dependent phosphorylation of both fructose and mannose (data not shown).

DISCUSSION
The genera Fusobacterium and Bacteroides are members of the Bacteroidaceae family of gram-negative anaerobic bacteria. Previous investigations provided little or no evidence for PEP:PTS-mediated sugar transport in Bacteroides thetaiotaomicron (31), Bacteroides succinogenes S85 (17,38,39), Bacteroides ruminicola B14 (39), or Bacteroides fragilis BF-1 (51). Furthermore, although studies of fructose metabolism by Fusobactenum nucleatum 10953 (45) revealed the formation of intracellular fructose 1-phosphate (the presumptive product of fructose-PEP:PTS activity [42]), attempts to demonstrate this group translocation system in either permeabilized cells or in reconstituted cell extracts were unsuccessful (45). To our knowledge, the present communication provides the first definitive evidence for PTS function in the Bacteroidaceae family in general and in the genus Fusobacterium specifically.
S6PH has been purified from two gram-positive organisms (Bacillus subtilis [34] and Lactococcus lactis [57]  enzyme has been partially purified from strains of Strepto-through 25) is found in the N-terminal sequence of F. coccus mutans (9,27,37). Purification of S6PH from F. mortiferum. The significance of this Arg/Tyr motif is not mortiferum has allowed the first characterization of this known.
enzyme from a gram-negative bacterium. From the data Although our purification of FK from F. mortiferum is the compiled in Table 2, it is apparent that S6PHs prepared from first reported isolation of this kinase from a gram-negative the two groups of bacteria have similar physicochemical bacterium, FK has previously been purified from two gramproperties. In all cases, S6PH is a single polypeptide (Mr, positive species: Leuconostoc mesenteroides (46) and L. 50,000 to 55,000), and the affinity of the enzyme for S6P is lactis (58). The three electrophoretically pure FKs require a approximately 100to 1,000-fold greater than that for su-divalent metal ion for activation, and all catalyze the ATPcrose. The presence of the phosphate group at C-6 of the dependent phosphorylation of both fructose and mannose glucosyl moiety is clearly an important rate-determining (Table 3). From these findings, it is likely that the previously factor for enzyme-catalyzed hydrolysis of the disaccharide.
documented phosphorylation of the two hexoses by partially The NH2 terminal sequences of S6PH from B. subtilis (16), purified preparations of (manno)FK from S. mutans (41) and S. mutans (47), Vibrio alginolyticus (50), and L. lactis (57) Escherichia coli (52) may be attributable to a single enzyme have been determined either by translation of the DNA of dual substrate specificity. Ultracentrifugal analyses and sequence of the cloned scrB gene(s) or by Edman degrada-gel filtration chromatography provide an estimated Mr of tion of the purified enzyme (see Table III of reference 57). In -60,000 for the FK from both L. lactis (58) and F. mortifall cases, the sequence R(Y or R)RX(Y or L) is conserved. It erum. However, results of SDS-PAGE revealed a single is of interest that a similar sequence RDRY (residues 22 polypeptide (Mr, 32,000), which suggests that the native  activities are presented in this article. Fusobacterium species contain enzymes of the Embden-Meyerhof-Parnas pathway. Other enzymes shown in this scheme have been identified as components of amino acid fermentation pathways in these species (3,8,25). enzyme is a homodimer. It is of interest that the molecular weight calculated from the translated amino acid sequence of the cloned FK gene (scrK) from V. alginolyticus is also -33,000 (4). The FKs from L. lactis and F. mortiferum exhibit similarity with respect to (i) phosphoryl donors and acceptors, (ii) activation by Me2" ions, and (iii) the presence of an identical sequence of eight amino acids, EAGGTKFV, within the NH2 terminus. Lactic acid is the major end product of sucrose fermentation by many microorganisms, including the lactic acid bacteria. However, for F. mortiferum, the products of sucrose metabolism (as for the products of glucose fermentation by related Clostridium species [25]) are primarily butyric and acetic acids. From these considerations, and by the demonstration of PEP-dependent phosphorylation of sucrose and of S6PH and FK activities, we propose the pathway shown in Fig. 4 as a probable route for sucrose fermentation by F. mortiferum. We have not conducted a strict quantitative balance of the products of sucrose fermentation by resting cells of F. mortiferum. However, a rough approximation of the molar ratio of butyrate:acetate is -1:1 (Fig. 1). Somewhat similar yields of butyrate and acetate are generated during the fermentation of certain amino acids (e.g., lysine [3] and glutamic acid [8]) by fusobacteria. It is reasonable to assume that pyruvic acid, produced during the initial stage of sucrose catabolism via the Embden-Meyerhof-Parnas pathway, is first converted to acetyl-coenzyme A (acetyl-CoA) (Fig. 4). Subsequently, a portion of the acetyl-CoA is transformed, via acetyl-phosphate, to acetic acid. The remainder forms acetoacetyl-CoA which, after entry to the later stages of the pathway(s) employed for amino acid fermentation, ultimately yields butyric acid.