Purification and characterization of two phosphoglucomutases from ...

Download PDF
2 downloads 0 Views 2MB Size Report
Comment
Mar 21, 1994 - on maltose, but the enzyme was absent in lactococci grown on glucose, ... of both a- and I-PGM than strain 65.1 when grown on maltose.
JOURNAL OF BACTERIOLOGY, Sept. 1994, p. 5304-5311

Vol. 176, No. 17

0021-9193/94/$04.00+0 Copyright C 1994, American Society for Microbiology

Purification and Characterization of Two Phosphoglucomutases from Lactococcus lactis subsp. lactis and Their Regulation in Maltose- and Glucose-Utilizing Cells NY QIAN, GRANT A. STANLEY, BARBEL HAHN-HAGERDAL, AND PETER RADSTROM*

Department of Applied Microbiology, Lund Institute of Technology, Lund University, S-221 00 Lund, Sweden Received 21 March 1994/Accepted 29 June 1994

Two distinct forms of phosphoglucomutase were found in Lactococcus lactis subsp. lactis, strains 19435 and 65.1, growing on maltose: 13-phosphoglucomutase (1-PGM), which catalyzes the reversible conversion of "-glucose 1-phosphate to glucose 6-phosphate in the maltose catabolism, and a-phosphoglucomutase (a-PGM). j-PGM was purified to more than 90%o homogeneity in crude cell extract from maltose-grown lactococci, and polyclonal antisera to the enzyme were prepared. The molecular mass of 1-PGM was estimated by gel filtration to be 28 kDa; its isoelectric point was 4.8. The corresponding values for a-PGM were 65 kDa and 4.4, respectively. The expression of both PGM enzymes was investigated under different growth conditions. The specific activity and amount of I-PGM per milliliter of cell extract increased with time in lactococci grown on maltose, but the enzyme was absent in lactococci grown on glucose, indicating enzyme synthesis to be induced by maltose in the growth medium. When glucose was added to maltose-grown lactococci, both the specific activity and amount of ,I-PGM per milliliter of cell extract decreased rapidly. This suggests that synthesis of 13-PGM is repressed by glucose in the medium. Although the specific activity of a-PGM did not change during growth on maltose or glucose, lactococcal strain 19435 showed a much higher specific activity of both a- and I-PGM than strain 65.1 when grown on maltose.

Lactic acid bacteria are used commercially in the dairy industry and for the production of lactate from such carbohydrates as starch obtained from potatoes or corn. The starch is usually hydrolyzed by acid or enzymes, mainly to maltose and glucose, prior to its use in lactic acid fermentation. Fermentation involving both maltose and glucose as substrates has been studied with such organisms as Salmonella typhimurium (22), Escherichia coli (7, 22), Bacillus licheniformis (25), and Streptococcus bovis (18). To our knowledge, there are no published reports on the utilization of both maltose and glucose with the bacteria involved in the industrial production of lactate. When Lactococcus lactis subsp. lactis, a member of the family of lactic acid bacteria, is grown with an excess of glucose or lactose, lactate is virtually the sole product (27). However, the end products are more complex when lactococci are grown on maltose (14) or are grown on glucose under conditions of glucose limitation (27). In both cases, this results in the formation of lesser amounts of lactate and greater amounts of formate, acetate, and ethanol. The initial events of metabolism for lactococci grown on maltose and those grown on glucose differ, however. Glucose is transported in the latter case by a phosphoenolpyruvate-dependent phosphotransferase system by which it is phosphorylated to glucose 6-phosphate, which is then metabolized to pyruvate via the Embden-Meyerhof glycolytic pathway (20, 28). The degradation of maltose, which is a disaccharide, involves a different transport mechanism, and additional steps are required after its translocation across the cytoplasmic membrane before it is transformed to glucose 6-phosphate. It has been suggested earlier that, in lactococci, a maltose-inducible permease is responsible for the transfer of maltose across the Corresponding author. Mailing address: Department of Applied Microbiology, Lund Institute of Technology, Lund University, P. 0. Box 124, S-221 00 Lund, Sweden. Phone: 46 46 103412. Fax: 46 46 104203. Electronic mail address: [email protected]. *

5304

cell membrane (5). The step following this in the degradation of maltose has been described for Neisseria meningitidis by Fitting and Doudoroff (8), who found the maltose to be phosphorolyzed by a maltose phosphorylase. Lactococci have been shown to harbor a Pi-dependent phosphorylase that catalyzes the cleavage of maltose to glucose and ,8-glucose 1-phosphate (19, 23). The further transformation of ,B-glucose 1-phosphate in Neisseria perflava has been found by Ben-Zvi and Schramm (2) to involve a 3-glucose 1-phosphate-specific phosphoglucomutase (P-PGM), which catalyzes the conversion of it to glucose 6-phosphate. Although the presence of 1-specific PGM enzymes or activity has been described for Euglena gracilis (1), Lactobacillus brevis (17), and L. lactis subsp. lactis (24), very little is known about the regulation of this enzyme. In the present article, we show that L. lactis subsp. lactis contains two distinct forms of the PGM enzyme, one specific for 3-glucose 1-phosphate (i.e.,

1-PGM)

and the other

apparently specific for a-glucose

1-phosphate (oa-PGM). In our empirical work, both enzymes were partially purified, and antisera to 1-PGM were produced. The expression of ,B-PGM in two different lactococcal strains growing on maltose and/or glucose could be monitored with the help of polyclonal antibodies and specific activity measurements. The production of lactate, formate, acetate, and ethanol was studied, as was the specific activity of a-PGM.

MATERIALS AND METHODS Bacterial strains and culture conditions. Two strains of lactococci were used to investigate the regulation of a- and 1-PGM during cell growth: L. lactis subsp. lactis 19435, obtained from the American Type Culture Collection (Rockville, Md.), and L. lactis subsp. lactis 65.1, isolated from milk and obtained from the Swedish Dairies Association, Malmo, Sweden. L. lactis subsp. lactis AS211, a mutant strain derived from strain 19435 (23), was used for the purification of ,B-PGM

VOL. 176, 1994

PURIFICATION AND REGULATION OF TWO PHOSPHOGLUCOMUTASES

because of the high activity of the enzyme in this organism. The bacteria were grown in a medium (pH 6.8) of the following composition (per liter): tryptone, 5 g; yeast extract, 5 g; Casamino Acids, 1 g; K2HPO4, 2.5 g; KH2PO4, 2.5 g; and MgSO4 7H20, 0.5 g. For the experiments in which only one carbohydrate source was used, either glucose (20 g/liter) or maltose (20 g/liter) was autoclaved separately and added to the medium. In the experiments in which two carbohydrate sources were used, the cells were grown initially on either maltose (10 g/liter) or glucose (10 g/liter), and the other carbohydrate (10 g/liter) was added during the mid-exponential phase of growth. The lactococci were grown batchwise under anaerobic conditions at 30°C in a fermentor (model FPL-3; Chemoferm AB, Hagersten, Sweden) with a working volume of 3 liters. Stirring was set at 300 rpm. The parent cultures were grown overnight, and the same medium and conditions as those of the experimental cultures were used. The inoculum, 5% (vol/vol) in size, was centrifuged and resuspended in fresh medium before being added to the experimental culture. For PGM purification, strain AS211 was grown for 10 h in four 2.5-liter standing batch cultures at pH 7.0 on M17 (26) medium containing maltose (10 g/liter). Measurement of growth, pHl, substrates, and products. Cell growth was monitored by optical density measurements at 620 nm (model UV-120-02 spectrophotometer; Shimadzu, Kyoto, Japan). The pH was recorded with a model PHM61 laboratory pH meter (Radiometer, Copenhagen, Denmark). Samples taken during culture growth were centrifuged either at 5,000 x g for 5 min (model J2-21 centrifuge; Beckman Instruments, Palo Alto, Calif.) or at 13,000 x g for 1 min (Microcentaur centrifuge; MSE, Sussex, England), and the supernatant obtained was analyzed for substrates and products. Glucose, maltose, lactate, formate, acetate, and ethanol were detected with a Waters 410 differential refractometer (Millipore Corp., Milford, Mass.) after separation of the compounds at 65°C on a prepacked Aminex HPX 87-H column (Bio-Rad Laboratories, Richmond, Calif.) by means of a Varian 9001 highperformance liquid chromatograph. The mobile phase was 0.01 N H2SO4, and the flow rate was 0.6 mVmin. Standards were injected separately before and after taking the samples. Quantification was through computer integration of the area under each chromatographic peak by using the EZChrom chromatography data system software package (Scientific Software, Inc., San Ramon, Calif.). Cell extract preparation. At various times during culture growth, samples (300 to 500 ml) were collected from the culture and centrifuged at 5,000 x g and 4°C for 5 min. The cell pellet was washed twice and resuspended in S ml of 50 mM triethanolamine (TEA) buffer containing 5 mM MgCl2 * 6H20 (pH 7.2). The cell suspensions were frozen at -80°C and pressed twice with an X-press (Biox, Goteborg, Sweden) to disrupt the cells. Cell debris was removed by centrifugation (10,000 x g, 4°C) for 10 min. The supernatants, which contained the cell extracts, were stored at -80°C until used. Enzymatic assays. The specific activities (per total cellular protein) of a- and ,B-PGM were assayed as the conversion of aor p-glucose 1-phosphate to glucose 6-phosphate in a coupled reaction with glucose 6-phosphate dehydrogenase, monitored in terms of formation of NADPH measured spectrophotometrically at 340 nm and 30°C (2, 12) by using a model U-2000 spectrophotometer (Hitachi Ltd., Tokyo, Japan). Different concentrations of cofactors were examined, and the following concentrations were found to give maximum activity in 50 mM TEA buffer (pH 7.2): 5 mM MgCl2 * 6H20, 0.4 mM NADP+ (Sigma Chemical Co., St. Louis, Mo.), 2 U of glucose 6-phosphate dehydrogenase (Sigma), 50 ,uM a-glucose 1,6-bisphos-

5305

phate (Sigma), and 1.4 mM a- or ,B-glucose 1-phosphate (Sigma). The reagents were added in the order listed, the reaction being started by the addition of either a- or ,-glucose 1-phosphate for ao- and 1-PGM activity, respectively. The activities of the two PGM enzymes were tested over a pH range of 5 to 8 in phosphate and citrate phosphate buffers by using the concentrations of cofactors described above. The protein concentration of the cell extracts was determined by the method of Bradford (3) and compared with a bovine serum albumin (BSA) standard. The protein assay reagent was from Pierce (Rockford, Ill.). Purification of a- and I-PGM. Cell extracts used for PGM purification were prepared from 10 liters of culture as described earlier, except that the TEA buffer contained 0.5 mM EDTA and 5 mM 2-mercaptoethanol (pH 7.2). All operations were carried out at 4°C unless otherwise indicated. The chromatography procedures for the purification of o- and P-PGM were performed on a fast protein liquid chromatography (FPLC) system (Pharmacia Biotech, Norden AB, Sollentuna, Sweden) containing two model P-500 high-precision pumps, a model LCC-501 plus liquid chromatography controller, two motor valves (MV-7 and MV-8), and a model REC 102 recorder. Protein elution was monitored with a UV-M II control unit (at 280 nm), and fractions were collected with a model FRAC-200 fraction collector. The cell extract was treated with 5% protamine sulfate solution to remove nucleic acids. The precipitate was removed by centrifugation at 40,000 x g for 10 min. Solid (NH4)2SO4 was then added to the protamine sulfate-treated supernatant. The precipitate, collected in the range of 45 to 85% (NH4)2SO4 saturation, was dissolved by and dialyzed against 50 mM TEA buffer containing 30 mM KCl, 5% (wtlvol) glycerol, 0.5 mM EDTA, and 5 mM 2-mercaptoethanol at pH 7.5 (buffer A). The protein preparation was concentrated by an Omegacell (Filtron Technology Corp., Northborough, Mass.) membrane PES 10k ultrafilter system. A Hiload 16/60 Superdex 200 gel filtration prepacked column (Pharmacia), which had been equilibrated with buffer A, was used to separate the proteins in the sample prepared from the ultrafiltration step. The proteins were eluted with buffer A at a flow rate of 1.0 ml/min. Fractions containing a- and 1-PGM activity were pooled and concentrated separately. Afterwards, the two protein samples were loaded separately onto a Pharmacia Mono Q HR 5/5 anion-exchange column (5 by 0.5 cm) equilibrated with 20 mM bis-Tris propane containing 45 mM KCI, 5% (wt/vol) glycerol, 0.5 mM EDTA, and 5 mM 2-mercaptoethanol at pH 6.9 (buffer B). A linear gradient of 45 to 300 mM KCI, formed by buffer B and buffer C (buffer C is the same as buffer B but contains 300 mM KCl and the pH is at 6.65), was used to elute proteins at a flow rate of 1.0 ml/min. PGM-active fractions were dialyzed against 10 mM Tris-HCl containing 4 mM MgCl2, 5% (wtlvol) glycerol, 0.5 mM EDTA, and 2 mM 2-mercaptoethanol (pH 7.3; buffer D) and concentrated. The two protein samples were subjected finally to dye affinity chromatography with Matrex-green A gel (Amicon, Lexington, Mass.) equilibrated with buffer D. Protein fractions with either a- or ,-PGM activity were eluted from the column at a flow rate of 0.4 mlmin by buffer D containing 0.05 mM a-glucose 1,6-bisphosphate and 1 mM aor ,3-glucose 1-phosphate. The purity of the two PGM enzyme preparations was determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis, and the purity of ,-PGM was calculated by scanning with a video densitometric system (Makab, Goteborg, Sweden). Molecular weight determination of the PGM enzymes. Molecular weights were determined by SDS-PAGE by using

5306

QIAN ET AL.

J. BAcrERIOL.

low-molecular-weight standards (Pharmacia) containing phosphorylase b (94.0 kDa), BSA (67.0 kDa), ovalbumin (43.0 kDa), carbonic anhydrase (30.0 kDa), trypsin inhibitor (20.1 kDa), and a-lactalbumin (14.4 kDa). For molecular weights determined by gel filtration, a low-molecular-weight calibration kit (Pharmacia) containing RNase A (13.7 kDa), chymotrypsinogen A (25.0 kDa), ovalbumin (43.0 kDa), albumin (67.0 kDa), and Blue Dextran 2000 was used. SDS-PAGE, native PAGE, and isoelectric focusing electrophoresis. All of the reagents for SDS-PAGE and ready-made native PAGE were from Bio-Rad. The high- and low-molecular-weight standard markers were from Pharmacia. SDSPAGE was performed as described by Laemmli (13). The acrylamide concentration used in SDS-PAGE was 12.5%, and that used in the native PAGE was a 4 to 20% (wt/vol) gradient. After the cell extracts and the purified PGM preparations had been heated in a buffer containing SDS and 2-mercaptoethanol and separated by SDS-PAGE, the polypeptide bands in the gels were visualized either by Coomassie brilliant blue R-250 staining or by being subjected to immunoblot analysis. For zymogram analysis, nondenatured proteins were separated by native PAGE. Gels and running buffer were kept cold on ice while the electrophoresis was performed. Isoelectric focusing electrophoresis was carried out on a PhastSystem and PhastGel IEF 3-9 (Pharmacia). After focusing at 4°C had been performed, the gel plates were subjected to Coomassie brilliant blue staining and zymogram analysis. Zymogram analysis. The staining solutions used for active PGM stainings were always freshly prepared and consisted of 50 mM TEA buffer (pH 7.2), 5 mM MgCl2, 0.4 mM NADP+, 2 U of glucose 6-phosphate dehydrogenase per ml, 50 ,uM a.-glucose 1,6-bisphosphate, 1.4 mM a- or 1-glucose 1-phosphate, 0.01% phenazine methosulfate, and 0.05% nitroblue tetrazolium chloride. The native gels were incubated in the staining solutions immediately after electrophoresis. The staining was stopped by rinsing the gels in water for a few seconds. 13-PGM antiserum production. One hundred fifty micrograms of the purified ,B-PGM preparation was emulsified with an equal volume of Freund's complete adjuvant (10) and injected subcutaneously into two 6-month-old rabbits. To increase antibody titer, two more injections (75 ,ug of the purified 13-PGM preparation in Freund's incomplete adjuvant each time) were given 4 and 6 weeks after the first injection. Blood samples were taken before and 2 weeks after the first, second, and third injections, respectively, for testing antibody titer and specificity by Western blot (immunoblot) analysis. When the antibody titer had reached an adequate level (10 weeks after the first injection), 50 ml of antiserum was collected from each of the two rabbits. Western blot and immunostaining. Denatured proteins from the cell extracts were separated by SDS-12.5% PAGE and electrotransferred in continuous buffer to 0.45-mm-poresize Immobilon polyvinylidene difluoride membranes (Millipore Corp., Bedford, Mass.) for 1 h with an LKB 2117-005 Multiphor II NovaBlot electrophoretic transfer unit (Pharmacia). Detection of 1-PGM on nitrocellulose membranes was performed by using rabbit polyclonal antisera and a goat anti-rabbit alkaline phosphatase immunoblot kit (Bio-Rad) as described in the manufacturer's instructions. RESULTS

Lactococci harbor two distinct PGM enzymes. Zymography of cell extracts from lactococci growing on maltose revealed the presence of two PGM enzymes. ot-PGM activity was detected as a band corresponding to a molecular mass of

1

2 3 4

5

6

7

8

g~~~~~~~ -. -0 669 kDa -

440 kDa 232 kDa

140 kDa -

67 kDa

FIG. 1. Zymograph analysis of a- and 13-PGM. High-molecularmass standards were stained by Coomassie R-250 (lanes 1 and 8): thyroglobulin (669 kDa), ferritin (440 kDa), catalase (232 kDa), lactate dehydrogenase (140 kDa), and BSA (67 kDa). Lanes: 2 and 3, purified 1-PGM; 4 and 5, crude cell extracts; 6 and 7, partially purified a-PGM. 1-Glucose 1-phosphate was the substrate used in lanes 2, 4, and 6, and a-glucose 1-phosphate was the substrate used in lanes 3, 5, and 7. The positions of a- and ,B-PGM are indicated by arrows.

approximately 70 kDa when a-glucose 1-phosphate was used in the staining assay (Fig. 1, lane 5). Two bands appeared when 13-glucose 1-phosphate instead of the a-anomer was used as a substrate (Fig. 1, lane 4). One of the bands was in the same position as a-PGM, and the other band was around 30 kDa, corresponding to 1-specific PGM activity. The PGM enzymes were partially purified from L. lactis subsp. lactis AS211. Gel filtration chromatography revealed two partially overlaid protein peaks containing a-PGM and 1-PGM activity, respectively. The fractions containing the highest levels of a-PGM activity and those containing the highest levels of 1-PGM activity were pooled separately. The two enzyme preparations were purified further on an anion-exchange chromatography column. The 1-PGM active protein peak was eluted at 50 mM KCl by combining a linear gradient of KCl with a pH gradient. The a-PGM active protein peak was eluted at 280 mM KCl. This purification step totally separated a-PGM activity from the pooled 1-PGM fractions. However, zymography and the activity assay still detected low 13-PGM activity in the pooled a-PGM fractions. The purification fold for 1-PGM after dye affinity chromatography was 30, and the yield was less than 1 % (Table 1). SDS-PAGE of the final purified a-PGM active preparation showed at least 12 protein bands of different color intensities (data not shown), but for the ,B-PGM active protein peak, one strong band appeared (Fig. 2, lane 2) with a calculated purity TABLE 1. Purification of 1-PGM from L. lactis subsp. lactis Purification step

Vol Total Total sp act Purification Yield (ml) protein activity (U/mg) (fold) (%)

500 3,200 Cell extract 2,000 Protamine sulfate 500 1,600 1,900 (NH4)2SO4(45 to 67 1,270 1,410 85% saturation) 12 156 336 Superdex 200 Mono Q 6 9 50 1.5 0.86 16 Matrex-green A

0.6 1.2 1.1

2 2

95 70

2.2 5.6 18

4 9 30

17 3 0.8

PURIFICATION AND REGULATION OF TWO PHOSPHOGLUCOMUTASES

VOL. 176, 1994

kDa

1

5307

2

40 -35

30-o E -25 E

94.0 ~MV 67.0 -

20 X 0

-15 D

43.0-

-1 0

IL

0

30.030

20.1 -

-25

14.4_

_

0

-20 E E 1 5 en

102 -10 0° 'aCL

FIG. 2. SDS-PAGE of purified ,B-PGM (10 R,g) (lane 2). The molecular mass standards (lane 1) are phosphorylase b (94 kDa), BSA (67 kDa), ovalbumin (43 kDa), carbonic anhydrase (30 kDa), trypsin inhibitor (20.1 kDa) and a-lactalbumin (14.4 kDa).

of at least 90% based on a video densitometric system and a molecular mass of approximately 25 kDa. The molecular masses of the two PGM enzymes were 65 and 28 kDa for o-PGM and ,B-PGM, respectively, when measured by gel filtration chromatography. Comparison of the molecular weights obtained by zymography, SDS-PAGE, and gel filtration chromatography suggests that both ax- and 1-PGM are monomeric enzymes. The size of ,B-PGM is comparable to that of the corresponding enzymes found in L. brevis (27 kDa [17]) and the alga E. gracilis (29 kDa [1]). The a-PGM is in the same size range as other a-PGM enzymes that have been characterized (4, 21). The isoelectric points of the purified a-PGM and 1-PGM were 4.8 and 4.4, respectively (data not shown). The purified preparation of ,B-PGM was used to produce rabbit antisera to the enzyme. Immunoblot analysis of cell extracts showed that the antisera detected P-PGM with a titer of 1:105 but had cross-reactions with several other denaturated proteins. However, staining of the other polypeptides on the Western blots did not interfere with the analysis of P-PGM. Zymography of the partially purified a- and ,B-PGM enzymes indicated that lactococci harbor two PGM enzymes of differing substrate specificities (Fig. 1, lanes 2, 3, 6, and 7). Both of the PGM enzymes required ot-glucose 1,6-bisphosphate and Mg2" for activity. The two PGM enzymes also exhibited different activities in the same buffer at the same temperature. In phosphate and citrate-phosphate buffers, the maximum specific activities were in the range of 6.3 to 6.7 for 3-PGM and 6.7 to 7.0 or higher for a-PGM. The specific activity of PGM both in the cell extract and in the partially purified preparations remained stable at -80°C for at least 3 months. When the enzymes were incubated overnight at room temperature, 50% of the 3-PGM specific activity and 70% of the at-PGM specific activity remained, but the specific activities of 3-PGM and a-PGM were only 4 and 3%, respectively, after 15 min at 50°C. Regulation of a- and 1-PGM and product formation by glucose or maltose. The effect of glucose and maltose on product formation and PGM specific activity was examined by

5 -

0

1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8 Time (h) Time (h)

FIG. 3. Effect of carbohydrate source on cell growth (@), pH (A), and the production of lactate (C1), formate (O), acetate (A), and

ethanol (0) in two strains of L. lactis subsp. lactis. Strains 19435 (A) and 65.1 (B) were grown in glucose medium. Strains 19435 (C) and 65.1 (D) were grown in maltose medium. OD620, optical density at 620 nm.

growing the lactococcal strains (19435 and 65.1) separately on the respective carbohydrate sources. The two strains displayed similar growth characteristics on both carbohydrate sources (Fig. 3), although the exponential growth rates of the maltosegrown cultures were slightly lower than those for the glucosegrown cells. In contrast, product formation was strongly influenced by the carbohydrate source, with some variation also occurring between the two strains. Glucose-grown cultures of both strains (Fig. 3A and B) were homofermentative in the production of lactate, although small amounts of acetate were also produced during the initial stages of growth. Lactate production was most rapid during exponential growth, and a decrease occurred once the cultures had reached the stationary phase. The high production rate of lactate in the glucose cultures is reflected by the pH decrease, which was greatest during mid- to late exponential growth, but decreased considerably once the stationary phase had been reached. A different product pattern was observed when the two strains were grown on maltose (Fig. 3C and D) since, in addition to the lactate, considerable amounts of formate, acetate, and ethanol were produced. The amount of formate produced by both strains was double the amount of acetate and ethanol formed during growth, as it was expected to be from metabolism using the mixed acid fermentation pathway (9). However, strain 19435 (Fig. 3C) produced significantly more lactate and less of the other products than did strain 65.1 (Fig. 3D) when both were grown on maltose. Equal amounts of lactate, acetate, and ethanol were produced by strain 65.1 throughout exponential growth, and it was not until the stationary phase had been reached that the production of lactate was greater than that of the other by-products. The fall in pH of the maltose-grown

5308

QIAN ET AL.

J. BACT1ERIOL.

1.2>0

E0.80.6

00.40.2-

.

0

...--..0

t-6o-

0 O

23456780 1

2

3

4

5 6780

Time (h)

Time (h)

FIG. 4. Effect of substrate on the activities of a- and ,B-PGM in strain 19435 (symbols: O,

1

2 345 6 Time (h)

7

8

1-PGM; O, a-PGM) and strain 65.1 (symbols:

A,

1-PGM; 0, ot-PGM). (A) Growth on maltose (open symbols) or glucose (closed symbols) only; (B) growth on glucose with maltose addition; (C)

growth on maltose with glucose addition. Addition of the second sugar is indicated by an arrow. The data represent single experiments.

cultures was less rapid than that of the glucose-grown cultures. This is most likely attributable to the lower rate of production of lactate by the maltose-grown cells. The specific activities of a*- and P-PGM were also dependent on the carbohydrate used to grow the cultures. a-PGM specific activity was detected in both glucose- and maltose-grown cultures, although a higher level of specific activity was recorded for the latter (Fig. 4A). During fermentation, the specific activity of ao-PGM in both strains did not vary from that present in the parent culture at the time the inoculum was prepared. No activity could be detected for 3-PGM when the

lactococci were grown on glucose (Fig. 4A), and immunoblot analysis could not detect the presence of P-PGM in cell extracts prepared from the glucose-grown cells (Fig. SA and D). The opposite occurred when the lactococci were grown on maltose. A high P-PGM specific activity was detected in both strains, but in strain 19435 the specific activity was more than double that of strain 65.1 (Fig. 4A). The specific activity of 1-PGM in both strains decreased upon inoculation but then increased throughout growth, attaining a specific activity nearly double that present when the cultures were started. The rate of increase in 3-PGM specific activity in strain 19435 was

C

B

A

P

P-PGM-i

5.2 6.5 8(h)

3 4

P-PGM-

P-PGM-

+ F

E

D

P 1 2 4.5 5.8 7.2 8.5(h)

P-PGM-

fI-PGM-

f~PGMT

T

FIG. 5. Immunoblot analysis of P-PGM expression in strain 65.1 (A, B, and C) and strain 19435 (D, E, and F). Panels A and D represent the same samples as those shown in Fig. 4A. The first four lanes are maltose-growing cells, and the last four lanes are glucose-growing cells. Panels B and E represent the same samples as those shown in Fig. 4B (maltose addition to glucose-growing cells), and panels C and F represent the same samples as those shown in Fig. 4C (glucose addition to maltose-growing cells). The addition of a second sugar is marked by an arrow. The lane numbers represent culture growth times (in hours). P, parent culture. The amount of cell extract loaded per lane was 8 ,ug of protein.

VOL. 176, 1994

PURIFICATION AND REGULATION OF TWO PHOSPHOGLUCOMUTASES 40

-35

cli

81

-3030_

-25 E

20

co

0

-15 n 0

ac

-10a.

3

0.1

5

0

0

o 1. r3 0

0

-20 0E

-a CD =

15 5 am

0

-10

°

X.

5

0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8 Time (h) Time (h)

FIG. 6. Effect of carbohydrate addition on L. lactis subsp. lactis cultures growing on a carbohydrate source different from that added. Maltose (10 g/liter) was added to cultures growing on glucose (10 g/liter) of strains 19435 (A) and 65.1 (B). Glucose (10 g/liter) was added to cultures growing on maltose (10 g/liter) of strains 19435 (C) and 65.1 (D). Cell growth (@), pH (A), and the production of lactate (O), formate (K), acetate (A), and ethanol (0) are shown. The time of substrate addition is indicated by an arrow. OD620, optical density at 620 nm.

significantly greater than that in strain 65.1. Immunoblot analysis verified the presence of P-PGM in the maltose-grown cells (Fig. SA and D). Product formation and PGM regulation in the presence of both maltose and glucose. To test for a repressive effect of glucose on the specific activity of 1-PGM, the cultures were grown initially on a single carbohydrate source until the mid- to late exponential growth phase, when an equivalent amount of the other sugar was added to the culture. Around 90% of the original substrate was still in the medium at the time the second carbohydrate was added (data not shown). In the first experiment (Fig. 6A and B), stationary-phase cells from an overnight culture grown on glucose were inoculated into a glucose (10 g/liter) medium. After approximately 4.5 h of growth, 10 g of maltose per liter was added to the culture. As observed earlier, there was no significant difference in growth between the two strains. However, after approximately 6 h, the number of cells suspended in the medium began to decline as a result of flocculation and sedimentation of the cells. This was not observed in the glucose- and maltose-only cultures (Fig. 3) or upon the addition of glucose to maltose-grown cells (Fig. 6C and D) and thus appears to be attributable to the addition of maltose. The product formation was not significantly different from that of the glucose-only cultures, being essentially lactate homofermentive throughout the entire 8 h. The addition of maltose to the cells growing on glucose did not affect product formation. The specific activity of a-PGM in both strains was almost identical to the specific activity measured in the cells

5309

grown on glucose only (Fig. 4B). The addition of maltose to cells growing on glucose did not affect the specific activity of a-PGM in these two strains. A small specific activity of ,B-PGM was detected in strain 19435 following maltose addition (Fig. 4B), but the specific activity was much less than that measured in this strain when grown on maltose only. No ,-PGM specific activity could be detected in strain 65.1 after 8 h of fermentation (Fig. 4B), suggesting that in this strain the expression of the enzyme is strongly suppressed by glucose. Immunoblot analysis confirmed the observation that ,-PGM is repressed by glucose (Fig. SB and E) and also detected the presence of ,-PGM in strain 19435 after maltose addition, verifying the small specific activity of ,3-PGM in these samples (Fig. 4B). Carbohydrate analysis revealed that only glucose was consumed during fermentation (data not shown). Although the maltose induction of ,B-PGM appeared to be stronger in strain 19435 than in strain 65.1, its repression by glucose was still considerable, and it restricted the specific activity of this enzyme to low levels. The repression of P-PGM specific activity by glucose was examined in another experiment, in which glucose was added to cultures growing on maltose (Fig. 6C and D). Stationaryphase cells from an overnight culture grown on maltose were used as the inocula. Both strains showed similar growth characteristics, although the final cell yield in strain 65.1 was slightly higher. The two strains also displayed a similar response in their product formation when glucose was added. Prior to glucose addition, both strains behaved in the same way as in the maltose-only experiment (Fig. 3C and D), i.e., in showing a mixed acid product formation. Upon the addition of glucose the product pattern changed almost immediately, showing a cessation of the production of formate, acetate, and ethanol and a rapid increase in the production of lactate. The switch from mixed acid product formation to homofermentative metabolism can be seen most clearly in the culture containing strain 65.1 (Fig. 6D). In that culture, the initial production of acetate and ethanol rivaled that of lactate, and formate was produced at twice the rate of the other metabolites. Following the addition of glucose, however, only lactate was formed. Although these cells had a history of growth on maltose and were still in the presence of maltose, they behaved as if they were growing on glucose only once glucose was added to the culture. The decrease in specific activity of the two PGMs supports this observation (Fig. 4C). Upon the addition of glucose to cells growing on maltose, the specific activity of ot-PGM appeared to fall to the slightly lower levels of specific activity observed in cells grown on glucose only (Fig. 4A), while the specific activity of P-PGM fell rapidly to the point where, after 4.5 h, the cells had a ,B-PGM specific activity which was 25% of the specific activity of ,B-PGM in cells grown on maltose only. Immunoblot analysis of the samples from strains 65.1 and 19435 showed an increase in the color intensity of the ,B-PGM protein band prior to the addition of glucose, but the color intensity of the band decreased following the addition of glucose (Fig. 5C and F). This indicates that the amount of ,B-PGM per ml of cell extract followed the same pattern as the specific activity of the enzyme before and after glucose addition. The maltose consumption rate decreased when glucose was added to the cultures, and the consumption of glucose began almost immediately after its addition (data not shown). DISCUSSION Phosphorolysis of glucose oligo- and polysaccharides usually yields ax-glucose 1-phosphate as a product. However, the phosphorolysis of the disaccharide maltose in lactococci yields

5310

QIAN ET AL.

J. BACTERIOL.

GLUCOSE

MALTOSE

Y~~~~ maltose

maltose

phosphorylase

i"

glucose

/ /z

,B-glucose-lP J3PGM

/

glucose-6P

N

CELL MASS

a-PGM > a-glucose-IP

Glycolysis

pyruvate

-

LACTATE

FORMATE ACETATE ETHANOL FIG. 7. The proposed metabolic pathways for maltose and glucose in L. lactis subsp. lactis. PEP-PTS, phosphoenolpyruvate phosphotransferase system.

the 3-anomer of glucose 1-phosphate instead, which is then converted into glucose 6-phosphate. The present study shows that lactococci harbor two PGM enzymes, one specific for B-glucose 1-phosphate and the other specific, apparently, for a-glucose 1-phosphate. Earlier reports of the presence of aand wB-PGM enzymes in L. brevis (17) and the alga E. gracilis (1) and of its probable presence in the Neisseria species (2, 8) indicate a wide distribution of these two enzymes in nature. The initial degradation of maltose in lactobacilli appears to be similar to that described for lactococci (Fig. 7), and although the roles of 3-PGM in Euglena cells and lactococci appear to be the same, Euglena cells produce I-glucose 1-phosphate from the phosphorolysis of trehalose. Although the two PGM enzymes could be expected to have many enzymatic similarities, our work has demonstrated there to be considerable differences in their biochemistries and in the regulation of their specific activities. The specific activity of a-PGM was constitutive in both strains of lactococci, both when grown on maltose and when grown on glucose. Although the specific activity of the enzyme was found to be constant during growth on a single carbohydrate source, the specific activity level depended upon the carbohydrate used for growth. Lactococci growing on maltose had a higher a-PGM specific activity than those growing on glucose. A slight decrease in the specific activity of ax-PGM was observed when glucose was added to cells growing on maltose. This suggests that the substrate used for growth determines a level of enzyme specific activity that remains constant during all phases of growth. Unlike the at-PGM specific activity, a specific activity for P-PGM could not be detected in either strain while growing on glucose. This was not due to allosteric control of 1-PGM

activity but rather was attributable to the absence of the enzyme in glucose-grown cells, indicating that glucose represses the synthesis of this enzyme. The repression of f-PGM by glucose was found to occur even in the presence of maltose (Fig. 5B and E). However, in the absence of glucose, maltose induced the expression of 1-PGM, the intracellular amount of which increased substantially in both strains during growth. Thus, the synthesis of 1-PGM requires maltose in the growth medium and the absence of glucose. The latter condition, originally called the glucose effect (15), refers specifically to the inhibitory effect of glucose on the expression of various catabolic enzymes. This phenomenon has been investigated extensively for several catabolic operons and involves mechanisms of different types, such as inducer exclusion, catabolite inhibition, and catabolite repression (15). Earlier work with lactococci indicates that both maltose permease and maltose phosphorylase are also induced by maltose (5, 11) and that the expression of maltose phosphorylase is repressed during growth on glucose (23). Therefore, it is likely that all of the enzymes involved in the initial metabolism of maltose (Fig. 7) are under the same catabolic control system as that of ,-PGM, although it is too early at this stage to suggest which regulatory mechanism is operating. It was suggested by Belocopitow and Marechal (1) that both at- and 3-PGM enzymes are controlled by the cofactors at- and 3-glucose 1,6-bisphosphate. They found that ,-glucose 1,6bisphosphate is a specific cofactor for 1-PGM in Euglena cells and functions as an inhibitor for a-PGM and, conversely, that at-glucose 1,6-bisphosphate, which is a cofactor for a-PGM, inhibits 1-PGM in this alga. However, we could not detect any inhibition of the lactococci 1-PGM when preparations of a.-glucose 1,6-bisphosphate were used as a cofactor for in vitro activity measurements. It is also unlikely that a-glucose 1,6bisphosphate served as the cofactor for 1-PGM during the activity tests, even though it was the only sugar bisphosphate preparation added to the cell extracts. The most likely explanation would seem to be the fact that commercial preparations of at-glucose 1,6-bisphosphate contain both anomers, as Marechal and Belocopitow (16) have shown, and that the ,-PGM in lactococci requires 3-glucose 1,6-bisphosphate for optimal activity, as has been shown for 3-PGM in E. gracilis. A low level of activity was also detected when 3-glucose 1-phosphate was added to cell extracts which contained only a-PGM. This activity ceased soon after the addition of the sugar phosphate (data not shown), which indicates that the commercially supplied 3-glucose 1-phosphate could also contain small amounts of the other anomer. L. lactis subsp. lactis cells have been found to contain low concentrations of fructose 1,6-bisphosphate when grown on maltose or under conditions of glucose limitation (23, 24), and yet high concentrations of fructose 1,6-bisphosphate are present when the cells are grown on an excess of glucose (23). Since lactate dehydrogenase is known to be activated by fructose 1,6-bisphosphate (6), it has been suggested (23, 24) that the mixed acid fermentation observed during growth on maltose, or under glucose limitation, is due to a lack of activation of lactate dehydrogenase by the low fructose 1,6bisphosphate concentrations under these conditions. It is conceivable that the activity of ,B-PGM, and associated maltose catabolic enzymes, may influence the level of glycolytic intermediates, such as fructose 1,6-bisphosphate, during growth on maltose, and therefore lactate formation. Also, since at-glucose 1-phosphate participates in several synthetic processes (4), it is tempting to speculate that the ,B-anomer of glucose 1-phosphate is also involved in polysaccharide formation. If the r-anomer is involved in anabolic processes, it is possible that

VOL. 176, 1994

PURIFICATION AND REGULATION OF TWO PHOSPHOGLUCOMUTASES

,-PGM activity may have a role in determining polysaccharide composition and yield. Knowledge concerning the role of 3-PGM in the maltose metabolism of lactococci will be greatly improved if a lactococcus strain that permits controlled expression of this enzyme is constructed. ACKNOWLEDGMENTS This work was supported by a grant from the Swedish Council for Forestry and Agricultural Research to Peter Radstrom and Barbel Hahn-Hagerdal. REFERENCES 1. Belocopitow, E., and L. R. Marechal. 1974. Metabolism of trehalose in Euglena gracilis. Eur. J. Biochem. 46:631-637. 2. Ben-Zvi, R., and M. Schramm. 1961. A phosphoglucomutase specific for ,B-glucose 1-phosphate. J. Biol. Chem. 236:2186-2189. 3. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254. 4. Brautaset, T., R. Standal, E. Fjeervik, and S. Valla. 1994. Nucleotide sequence and expression analysis of the Acetobacter xylinum phosphoglucomutase gene. Microbiology 140:1183-1188. 5. Citti, J. E., W. E. Sandine, and P. R Elliker. 1966. Lactose and maltose uptake by Streptococcus lactis. J. Dairy Sci. 50:485-487. 6. Crow, V. L., and G. G. Pritchard. 1977. Fructose 1,6 diphosphateactivated L-lactate dehydrogenase from Streptococcus lactis: kinetic properties and factors affecting activation. J. Bacteriol. 131:82-91. 7. Decker, K., R Peist, J. Reidl, M. Kossmann, B. Brand, and W. Boos. 1993. Maltose and maltotriose can be formed endogenously in Escherichia coli from glucose and glucose 1-phosphate independently of enzymes of the maltose system. J. Bacteriol. 175:5655-5665. 8. Fitting, C., and M. J. Doudoroff. 1952. Phosphorolysis of maltose by enzyme preparations from Neisseria meningitidis. J. Biol. Chem. 199:153-163. 9. Fordyce, A. M., V. L. Crow, and T. D. Thomas. 1984. Regulation of product formation during glucose or lactose limitation in nongrowing cells of Streptococcus lactis. Appl. Environ. Microbiol. 48:332-337. 10. Freund, J. 1947. Some aspects of active immunization. Annu. Rev. Microbiol. 1:291-308. 11. Haggstrom, M. 1981. Ph.D. thesis. University of Lund, Lund, Sweden. 12. Joshi, J. G., and P. Handler. 1964. Phosphoglucomutase. I. Purification and properties of phosphoglucomutase from Eschenichia coli. J. Biol. Chem. 239:2741-2751. 13. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685.

5311

14. Lohmeier-Vogel, E., M. Haggstrom, H. Wittgren, and B. HahnHagerdal. 1983. Levels of metabolic intermediates in Streptococcus lactis grown on different carbon sources and the effect on product formation. Acta Chem. Scand. Sect. B 37:751-753. 15. Magasanik, B., and F. C. Neidhardt. 1987. Regulation of carbon and nitrogen utilization, p. 1318-1325. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: cellular and molecular biology. American Society for Microbiology, Washington, D.C. 16. Marechal, L. R, and E. Belocopitow. 1974. Metabolism of trehalose in Euglena gracilis. Eur. J. Biochem. 42:45-50. 17. Marechal, L. R, G. Oliver, L. A. Veiga, and A. A. P. De R Holgado. 1984. Partial purification and some properties of 1-phosphoglucomutase from Lactobacillus brevis. 228:592-599. 18. Martin, S. A., and J. B. Russell. 1987. Transport and phosphorylation of disaccharides by the ruminal bacterium Streptococcus bovis. Appl. Environ. Microbiol. 53:2388-2393. 19. Moustafa, H. H., and E. B. Collins. 1968. Role of galactose or glucose 1-phosphate in preventing the lysis of Streptococcus diacetilactis. J. Bacteriol. 95:592-602. 20. Postma, P. W., J. W. Lengeler, and G. R. Jacobson. 1993. Phosphoenolpyruvate: carbohydrate phosphotransferase systems of bacteria. FEMS Microbiol. Rev. 57:543-594. 21. Ray, W. J., Jr., and E. J. Peck, Jr. 1972. Phosphomutases, p. 407-458. In P. D. Boyer (ed.), The enzymes, 3rd ed., vol. 6. Academic Press, Inc., New York. 22. Saier, M. H., Jr., H. Straud, L. S. Massman, J. J. Judice, M. J. Newman, and B. U. Feucht. 1978. Permease-specific mutations in Salmonella typhimurium and Escherichia coli that release the glycerol, maltose, melibiose, and lactose transport systems from regulation by the phosphoenolpyruvate: sugar phosphotransferase system. J. Bacteriol. 133:1358-1367. 23. Sjoberg, A. 1992. Ph.D. thesis. University of Lund, Lund, Sweden. 24. Sjoberg, A., and B. Hahn-Hagerdal. 1989. 13-Glucose 1-phosphate, a possible mediator for polysaccharide formation in maltoseassimilating Lactococcus lactis. Appl. Environ. Microbiol. 55: 1549-1554. 25. Tangney, M., P. Smith, F. G. Priest, and W. J. Mitchell. 1992. Maltose transport in Bacillus licheniformis NCIB 6346. J. Gen. Microbiol. 138:1821-1827. 26. Terzaghi, B. E., and W. E. Sandine. 1975. Improved medium for lactic Streptococci and their bacteriophages. Appl. Environ. Microbiol. 29:807-813. 27. Thomas, T. D., D. C. Ellwood, and V. M. C. Longyear. 1979. Change from homo- to heterolactic fermentation by Streptococcus lactis resulting from glucose limitation in anaerobic chemostat cultures. J. Bacteriol. 138:109-117. 28. Thompson, J. 1987. Regulation of sugar transport and metabolism in lactic acid bacteria. FEMS Microbiol. Rev. 46:221-231.