INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Interest in acetone-butanol-ethanol (ABE) fermentation by clostridia has been renewed due to advances in our understanding of the genetics and physiology of solvent production by these microorganisms, as well as for economic and environmental reasons (2, 28). Clostridium beijerinckii, a gram-positive, anaerobic, spore-forming bacterium, is a member of the solvent-producing clostridia associated with the ABE fermentation. The C. beijerinckii BA101 hyper-butanol-producing mutant was generated from C. beijerinckii NCIMB 8052 by using N-methyl-N'-nitro-N-nitrosoguanidine, together with selective enrichment on the nonmetabolizable glucose analog, 2-deoxyglucose (2-DG) (1). Pilot-scale (20-liter) fermentations in which semidefined P2 medium containing either 6% glucose or 6% STAR-DRI 5 maltodextrin was used demonstrated that C. beijerinckii BA101 produces up to 100% more butanol and acetone than the C. beijerinckii 8052 parent strain. In addition, C. beijerinckii BA101 exhibited reduced acid production and increased carbohydrate utilization compared to C. beijerinckii 8052 (9). This observation corresponds with higher butanol and total solvent production by C. beijerinckii BA101.

We have been interested in further characterization of C. beijerinckii BA101 with respect to understanding the basis for the production of elevated levels of butanol by this strain. Although several physiological and molecular changes were associated with butanol hyperproduction in C. beijerinckii BA101 (1, 4, 5), the uptake and accumulation of glucose in C. beijerinckii BA101 have not been examined.

In many bacteria, the phosphoenolpyruvate-sugar phosphotransferase system (PEP-PTS) is employed to uptake sugars, which mediate the uptake and phosphorylation of carbohydrates. The PTS is a group translocation process in which the transfer of the phosphate moiety of PEP to carbohydrates is catalyzed by the general non-sugar-specific proteins, enzyme I and HPr, in combination with the sugar-specific enzyme II proteins (17). The PTS is also recognized as a primary carbohydrate transport system in the clostridia (19).

We sought to examine glucose transport and accumulation in C. beijerinckii BA101 and NCIMB 8052. We present here findings indicating that C. beijerinckii BA101, a butanol-hyperproducing mutant, possesses defective PTS activity for glucose and other sugars despite the observation that this strain metabolizes glucose more efficiently than C. beijerinckii 8052. We also characterized the PTS in C. beijerinckii BA101 and NCIMB 8052 and suggest the possibility for the presence of an alternative glucose transport system.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains. The bacterial strains used in this study were C. beijerinckii NCIMB 8052 and BA101. C. beijerinckii 8052 is a wild-type strain, and C. beijerinckii BA101 is a hyperamyloytic, hyper-solvent-producing mutant. Stock cultures of C. beijerinckii were maintained as spores in distilled water at 4°C. The Staphylococcus aureus strains used included 797A (ptsH) and 710A (ptsI).

Growth conditions and glucose assay. Spores (5 ml) were heat shocked at 80°C for 10 min, inoculated into 10 ml of reinforced clostridial medium (RCM), and incubated anaerobically for 14 h at 35°C. RCM (5 ml) cultures were transferred to 100 ml of semidefined P2 medium containing 6% glucose (9), incubated anaerobically for 20 h at 35°C, washed, and transferred to 1 liter of 6% glucose P2 medium and incubated at 33°C. The optical density at 600 nm (OD₆₀₀) and pH were measured every 4 h for the first 16 h and then every 8 h thereafter. For the glucose assay, culture samples were withdrawn over time and assayed by using glucose (Hexokinase Kit) reagent (Sigma Chemical Co.).

Uptake of [¹⁴C]glucose by intact resting cells. C. beijerinckii NCIMB 8052 and BA101 cultures were harvested at 8 h and washed with 50 mM potassium phosphate buffer (pH 7.0). Cell density was determined from the relationship A₆₀₀ = 1.0 equivalent to 0.265 mg (dry weight). Glucose uptake was examined by adding [¹⁴C]glucose to 1 ml of cell suspension to give a final concentration of 0.1 mM and incubating the sample at 37°C. Samples (0.15 ml) were removed, filtered, and washed twice with buffer over a period of 5 min (15, 17). The radioactivity of [¹⁴C]glucose accumulated in cells bound to the filter was measured by using a Beckman LS5000TD scintillation counter (Beckman) and 20-ml scintillation vials, each containing 4 ml of scintillation cocktail.

Preparation of cell extracts. C. beijerinckii NCIMB 8052 and BA101 cultures were harvested at acidogenic (8 h) and solventogenic (28 h) stages and washed with 10 mM Tris-Cl buffer (pH 7.5) by centrifugation at 8,000 × g. Cells were disrupted by two passages through a French pressure cell at 20,000 lb/in². Cell debris was removed by centrifugation at 12,000 × g for 10 min, and the supernatant was used as a cell extract and stored at 80°C (20). The protein concentration of the sample was measured by using the Bio-Rad protein assay.

Complementary assay with S. aureus ptsHI mutants. The cell extracts of both S. aureus and C. beijerinckii were used in a colorimetric PTS assay involving o-nitrophenyl--D-galactopyranoside (ONPG) as a staphylococcal PTS substrate. The assays were carried out in a total volume of 1 ml containing 50 µl of each cell extract tested, 0.5 mM PEP, 1 mM ONPG, and 20 mM Tris-HCl buffer (pH 7.5). The production of ONPG was monitored at 420 nm.

Glucose phosphorylation assay. The glucose phosphorylation assay was performed by following the protocol described by Mitchell et al. (21). Glucose phosphorylation with either PEP or ATP as a phosphate donor was assayed by precipitation of radiolabeled glucose phosphate in ethanolic barium bromide. Samples were taken and added to 2 ml of 1% (wt/vol) barium bromide in 80% (vol/vol) ethanol. Precipitates were removed by filtration and washed with 5 ml of 80% ethanol. For the PTS assay, the rate of glucose phosphorylation by PEP was determined by measuring the amount of glucose phosphate released over a 3-min period. The PTS activity associated with cell extracts, soluble extracts, membranes, and various fractions obtained after gel filtration was examined (see below). The glucokinase assay was carried out like the PTS assay except that ATP was used as the phosphate donor.

Fractionation of the soluble extract by gel filtration. Cell extracts of C. beijerinckii NCIMB 8052 and BA101 were further fractionated into soluble extracts and membrane fractions by ultracentrifugation as previously indicated (20). Membrane fractions were washed with buffer and concentrated 10-fold with respect to the original extract. Soluble extracts were fractionated on a Sephadex G-100 column (2.5 by 75 cm) at 4°C at a flow rate of 18 ml/h.

Partial purification and phosphorylation of HPr. To purify HPr from soluble extracts derived from both C. beijerinckii BA101 and 8052, fractions 63 to 75 from gel filtration were pooled and concentrated by using CentriPlus concentrators (Amicon). Partially purified HPr was further separated by using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE; 16% Tris-Tricine gel), and a blot of the gel on a polyvinylidene difluoride membrane was used for N-terminal sequencing. N-terminal sequencing of 18 amino acids from HPr was carried out at the Protein Sciences Facility at the University of Illinois-Biotechnology Center. ATP-dependent phosphorylation of partially purified HPr (Ser-46) was carried out in the presence of [-³²P]ATP and Bacillus HPr kinase as described previously (11). Bacillus HPr was used as a positive control, and a reaction mixture without HPr was used as a negative control. Reaction mixtures were incubated for 10 min at 37°C. Reactions were stopped by addition of an equal volume of sample buffer and applied to a 16% Tris-Tricine SDS-PAGE gel. After electrophoresis, the gel was treated for 5 min in boiling 16% trichloroacetic acid before it was dried and exposed as an autoradiograph. Bacillus HPr and HPr kinase were obtained from W. J. Mitchell (Hariot-Watt University, Edinburgh, United Kingdom).

Materials. D-[U-¹⁴C]glucose, D-[U-¹⁴C]fructose, D-[U-¹⁴C]glucitol, D-[1-¹⁴C]mannitol, D-[1-³H]galactose, [-³²P]ATP, and Sephadex G-100 were purchased from Amersham Phamacia Biotech. 2-Deoxyglucose, ATP, PEP [phosphoenolpyruvate tri(cyclohexylammonium) salt], barium bromide, trichloroacetic acid, and glucose (HK) reagent were obtained from Sigma Chemical Co. CentriPlus was purchased from Amicon.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Glucose utilization and uptake. The higher rate and more complete utilization of glucose by C. beijerinckii BA101 was observed in 1 liter of semidefined P2 medium containing 6% glucose (Fig. 1). The effect of 2-DG and mannose on the glucose uptake is shown in Table 1. Glucose uptake for both C. beijerinckii 8052 and BA101 was inhibited in the presence of mannose or 2-DG, which are known PTS substrates (16, 19). The inhibition of glucose uptake in the presence of 2-DG suggests the involvement of glucose PTS in glucose transport in both C. beijerinckii BA101 and 8052. However, the decreased inhibition of glucose uptake in C. beijerinckii BA101 by 2-DG suggests that this strain may transport glucose via an alternative transport mechanism.

View larger version (19K): [in this window] [in a new window]   FIG. 1.   Growth and glucose utilization by C. beijerinckii NCIMB 8052 and BA101 in 1 liter of P2 medium containing 6% glucose. Symbols: , growth of C. beijerinckii BA101; , growth of C. beijerinckii NCIMB 8052; , glucose utilization by C. beijerinckii BA101; , glucose utilization by C. beijerinckii NCIMB 8052.

PEP-dependent sugar phosphorylation. The presence of a PTS in C. beijerinckii BA101 was confirmed by a complementary assay with cell extracts derived from pts mutants of S. aureus lacking either Enzyme I or HPr (Fig. 2). A cell extract from C. beijerinckii BA101 was able to complement cell extracts obtained from pts mutants of S. aureus, demonstrating the presence of both Enzyme I and HPr activities in the extracts derived from C. beijerinckii BA101. It was also observed that Enzyme I and HPr activity in the extracts of C. beijerinckii BA101 was lower than in those of strain 8052. 

View larger version (33K): [in this window] [in a new window]   FIG. 2.   Colorimetric PTS assay with complementation of cell extracts between S. aureus pts mutants and C. beijerinckii. Shaded bars (in each group) indicate the following, from left to right: C. beijerinckii BA101 and S. aureus, C. beijerinckii BA101 only, C. beijerinckii NCIMB 8052 and S. aureus, C. beijerinckii NCIMB 8052 only, and S. aureus only.

Properties of PTS components. The effect of different combinations of soluble extracts and membrane fractions on glucose PTS activity is shown in Fig. 3. Either soluble extracts or membrane fractions by themselves were not able to demonstrate PTS activity. The recovered PTS activity from a combination of soluble extracts and membrane fractions derived from C. beijerinckii was lower (Fig. 3) than the PTS activity observed for cell extracts shown in Table 2. This may be due to the loss of enzyme activity during the separation of soluble extracts and membrane fractions. However, a combination of soluble extracts and membrane fractions derived from C. beijerinckii BA101 demonstrated lower PTS activity (50%) than soluble and membrane fractions derived from C. beijerinckii 8052. This result is consistent with the results obtained from the glucose PTS assay with cell extracts (Table 2).

View larger version (17K): [in this window] [in a new window]   FIG. 3.   Effect of different combinations of soluble extracts and membrane fractions derived from C. beijerinckii NCIMB 8052 and BA101 on glucose PTS activity. Symbols: , soluble extract and membrane fraction of C. beijerinckii 8052 (combination 1); , soluble extract of C. beijerinckii 8052 and membrane fraction of C. beijerinckii BA101 (combination 2); , membrane fraction of C. beijerinckii 8052 and soluble extract of C. beijerinckii BA101 (combination 3); , soluble extract and membrane fraction of C. beijerinckii BA101 (combination 4).

View larger version (19K): [in this window] [in a new window]   FIG. 4.   Fractionation and PTS activity of soluble extracts of C. beijerinckii NCIMB 8052 (A) and BA101 (B) by gel filtration. PEP-dependent PTS activity in fractions from gel filtration was assayed in the presence of membrane fractions derived from either C. beijerinckii NCIMB 8052 or BA101. No PTS activity was observed in either soluble extract or the membrane fraction itself. Symbols: , OD280; , cpm.

ATP- and PEP-dependent glucose phosphorylation at different growth stages. The PEP-dependent glucose phosphorylation assay was performed by using cell extracts derived from both acidogenic and solventogenic cells (Fig. 5). PTS activity of cell extracts associated with solventogenic C. beijerinckii BA101 and 8052 was significantly lower than that associated with acidogenic-phase cells. PTS activity associated with C. beijerinckii appears to be repressed during solventogenic stage, a result which is consistent with previous findings (14). During both acidogenic and solventogenic stages, C. beijerinckii BA101 demonstrated lower (50%) PTS activity than C. beijerinckii 8052. 

View larger version (19K): [in this window] [in a new window]   FIG. 5.   Glucose phosphorylation by PEP (solid bars) and ATP (shaded bars) with cell extracts derived from cultures at the acidogenic stage (A) and at the solventogenic stage (B).

Partial purification and phosphorylation of HPr. In order to further investigate the decreased PTS activity in C. beijerinckii BA101, we partially purified HPr from both C. beijerinckii BA101 and 8052 by using size exclusion chromatography, followed by SDS-PAGE. The His-15 region from the N terminus of the HPr protein was sequenced. The results of SDS-PAGE to purify HPr are shown in Fig. 6B. In the denatured SDS-gel, HPr proteins derived from both C. beijerinckii BA101 and 8052 were the same size. A blot of the gel was used for the N-terminal sequence analysis of HPr. It was found that the His-15 region (HARP) of HPr derived from C. beijerinckii is conserved, as is the case for bacterial HPrs (Fig. 6A). The amino acid sequences of HPr protein purified from C. beijerinckii 8052 and BA101 were found to be identical. This indicates that C. beijerinckii BA101 does not have a mutation in the catalytic region (His-15) of the HPr protein so that phosphate transfer between PTS components may not be altered. In vitro phosphorylation of partially purified HPr was performed to examine an alteration of regulatory site (Ser-46) by using HPr kinase purified from Bacillus and ³²P-labeled ATP as described previously (8, 11). [-³²P]ATP phosphorylation of the HPr (Ser-46) by HPr kinase was detected quantitatively. Essentially the same levels of phosphorylation by ATP were observed for Bacillus HPr and partially purified clostridial HPrs (Fig. 6C). HPr from C. beijerinckii BA101 was phosphorylated by the ATP-dependent HPr kinase at the same level as HPr derived from C. beijerinckii 8052. HPr protein derived from C. beijerinckii BA101 did not demonstrate any difference in either amino acid sequence of the catalytic site or in activity of the regulatory site relative to HPr protein derived from C. beijerinckii 8052. These results suggest that either the His-15 or the Ser-46 residue of HPr from C. beijerinckii BA101 can be phosphorylated or dephosphorylated for PTS activity and regulation at the same rate as the HPr from C. beijerinckii 8052. It is likely that C. beijerinckii BA101 has an altered regulation of HPr activity, possibly at the transcriptional level similar to other bacterial HPrs (6).

View larger version (195K): [in this window] [in a new window]   FIG. 6.   Analysis of two conserved regions in the HPr protein. (A) Alignment of amino acid sequences around the His-15 residue in HPr of C. beijerinckii and other gram-positive bacteria. The phosphorylation of His-15 is indicated by an arrow. (B) Purification of partially purified HPr by SDS-PAGE. (C) In vitro phosphorylation of HPr (Ser-46). [-32P]ATP and Bacillus HPr kinase were incubated with semipurified HPr from C. beijerinckii BA101 and 8052. Lane 1, negative control without HPr; lanes 2 and 3, Bacillus HPr; lanes 4 and 5, partially purified HPr from C. beijerinckii 8052; lanes 6 and 7, semipurified HPr from C. beijerinckii BA101.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The PTS is a complex enzyme system that is responsible for the detection, transmembrane transport, and phosphorylation of numerous sugar substrates in both gram-negative and gram-positive prokaryotes (3, 23, 24). HPr, one of the general proteins in PTS, is conserved among bacteria and possesses two phosphorylation sites: His-15 and Ser-46. The sequences of those two phosphorylation regions (i.e., His-15 and Ser-46) are highly homologous. Phosphorylation of His-15 residue by Enzyme I is required for the PTS activity. However, the phosphorylation of Ser-46 residue by HPr kinase and ATP regulates HPr activity associated with carbohydrate metabolism, including PTS and catabolite repression in gram-positive bacteria (7, 13, 25). It has been reported that a mutation in the Ser-46 residue of the HPr protein affected catabolite repression (27, 29), non-PTS sugar transport (30) and sporulation (10) as well as sugar uptake via PTS.

HPr protein may be a key element in regulation of carbohydrate metabolism in clostridia as it is in other gram-positive bacteria (24, 30). Based on the observation of a highly homologous amino acid sequence of HPr protein catalytic site, together with in vitro phosphorylation assay results obtained with Bacillus HPr kinase, the HPr protein of C. beijerinckii appears to be functionally and structurally related to HPrs found in other bacteria.

Lower activities of Enzyme I and HPr protein in C. beijerinckii BA101 were observed in complementation assays with the combination of C. beijerinckii cell extract and S. aureus pts mutants. C. beijerinckii BA101 demonstrated lower PTS activity in both the soluble and the membrane fractions, confirming the lower activities of its Enzyme I and HPr protein. In addition, reduced PTS activity for fructose and glucitol, as well as glucose, in C. beijerinckii BA101 cell extracts implies the decreased PTS activity may be due to a defect in general PTS proteins rather than in sugar-specific permeases. Based on these observations, it is likely that PTS defective properties associated with C. beijerinckii BA101 are probably due to decreased activities of the general proteins HPr and Enzyme I. However, biochemical analysis of HPrs from both C. beijerinckii BA101 and 8052 indicates that the HPr protein in C. beijerinckii BA101 is likely to function similarly to HPr protein derived from C. beijerinckii 8052 even though decreased HPr protein activity was observed in other PTS assays. Based on these observations, we postulate that C. beijerinckii BA101 may have a mutation upstream of the pts gene or in a regulatory region for pts gene expression. If we assume that the genes for clostridial PTS are organized in operons as in other bacterial PTS systems, a decreased expression of the PTS operon in C. beijerinckii BA101 due to a mutation would affect the activity of both the HPr protein and Enzyme I, which are encoded by ptsH and ptsI genes, respectively.

Although a PTS defect in C. beijerinckii BA101 is consistent with a 2-DG-resistant phenotype and the decreased inhibition of glucose uptake in the presence of 2-DG compared to C. beijerinckii 8052, a PTS defect in C. beijerinckii BA101 does not readily explain why this strain carries out more complete glucose utilization and uptake during fermentation. In addition, a higher rate of glucose uptake by intact cells of solventogenic-phase C. beijerinckii BA101 was observed for C. beijerinckii 8052 (data not shown). If we assume that PTS is the only transport system for glucose, it is not likely that the PTS-defective C. beijerinckii BA101 would be able to achieve a similar growth rate, together with increased glucose utilization, compared to C. beijerinckii 8052 over the course of fermentation. During the solventogenic stage, growth and glucose utilization by C. beijerinckii BA101 remained at the same rate despite the observation that the glucose-PTS activity fell to 40% of the glucose uptake rate by intact cells. In other words, C. beijerinckii BA101 cells possess more glucose transport capacity than that indicated by the in vitro glucose PTS assay alone.

The simplest explanation for these observations may be the presence of an alternative glucose transport mechanism that compensates for the PTS defect in C. beijerinckii BA101. In both strains, a higher glucokinase activity was detected during solventogenic stage when PTS was repressed. The induction of glucokinase activity under PTS-repressed conditions is very similar to previous findings for Streptococcus mutans (12).

It is likely that glucokinase is primarily involved in glucose metabolism during the solventogenic stage. Acidogenic-stage C. beijerinckii BA101 may take up glucose at a lower rate than does C. beijerinckii 8052 due to the decreased PTS activity. However, unlike solventogenic-stage C. beijerinckii 8052, C. beijerinckii BA101 may be able to maintain the higher glucose uptake rate under PTS-repressed conditions because enhanced glucokinase activity compensates for the defective PTS activity observed during the solventogenic stage. It is also likely that C. beijerinckii BA101 accumulates glucose by an ATP-dependent glucokinase in preference to a PEP-dependent PTS regardless of the growth stage. If we assume that the increased glucokinase activity in C. beijerinckii BA101 solventogenic cells is accompanied by an increase in a non-PTS glucose transport system, this may explain why C. beijerinckii BA101 shows more complete glucose utilization and elevated levels of butanol production relative to C. beijerinckii 8052 in spite of this strain possessing a defective PTS.

The inverse relationship between PTS activity and glucokinase activity at various growth stages implies that PTS may be involved in the regulation of glucokinase and possibly an alternative glucose transport system. At present, the question of whether glucokinase plays a direct metabolic role in glucose utilization in C. beijerinckii cannot be answered. The presence of an alternative glucose transport mechanism that requires glucokinase is currently under investigation in our laboratory in order to examine the possible involvement of glucokinase and non-PTS transport system in glucose metabolism of C. beijerinckii. A more detailed molecular analysis of PTS and the alternative non-PTS glucose transport mechanism will provide a more direct explanation for the difference in glucose transport and utilization between these two clostridial strains.