B-D-phosphogalactoside galactohydrolase activity and phosphoenolpyruvate phosphotransferase activity forboth lactose and galactose. There was also a shift.
OF BACTERIOLOGY, Feb. 1979, 0021-9193/79/02-0878/07$02.00/0
JOURNAL
Vol. 137, No. 2
p. 878-884
Influence of the Lactose Plasmid on the Metabolism of Galactose by Streptococcus lactis DONALD J. LEBLANC,* VAUGHAN L. CROW, LINDA N. LEE, AND CLAUDE F. GARONt Laboratory of Microbiology and Immunology, National Institute of Dental Research, Bethesda, Maryland 20014
Received for publication 14 August 1978
Streptococcus lactis strain DR1251 was capable of growth on lactose and galactose with generation times, at 30°C, of 42 and 52 min, respectively. Phosphoenolpyruvate-dependent phosphotransferase activity for lactose and galactose was induced during growth on either substrate. This activity had an apparent Km of 5 x 10-5 M for lactose and 2 x 10-2 M for galactose. /-D-Phosphogalactoside galactohydrolase activity was synthesized constitutively by these cells. Strain DR1251 lost the ability to grow on lactose at a high frequency when incubated at 37°C with glucose as the growth substrate. Loss of ability to metabolize lactose was accompanied by the loss of a 32-megadalton plasmid, pDR1, and Lac- isolates did not revert to a Lac' phenotype. Lac- strains were able to grow on galactose but with a longer generation time. Galactose-grown Lac- strains were deficient in B-D-phosphogalactoside galactohydrolase activity and phosphoenolpyruvate phosphotransferase activity for both lactose and galactose. There was also a shift from a predominantly homolactic to a heterolactic fermentation and a fivefold increase in galactokinase activity, relative to the Lac' parent strain grown on galactose. These results suggest that S. lactis strain DR1251 metabolizes galactose primarily via the tagatose-6-phosphate pathway, using a lactose phosphoenolpyruvate phosphotransferase activity to transport this substrate into the cell. Lacderivatives of strain DR1251, deficient in the lactose phosphoenolpyruvate phosphotransferase activity, appeared to utilize galactose via the Leloir pathway.
Although members of the group N streptococci are capable of growing at the expense of either lactose or galactose, the pathway(s) by which the latter is catabolized has not been completely described. The initial steps of lactose metabolism by Streptococcus lactis are catalyzed by two enzymes: a phosphoenolpyruvate (PEP) phosphotransferase which transports and phosphorylates lactose (24, 25, 32) and a cytoplasmic P-/8-galactoside hydrolase (P-f3-galactosidase) that cleaves lactose-phosphate, producing glucose and galactose-6-phosphate (14, 28). Galactose-6-phosphate is most likely metabolized via the tagatose-6-phosphate pathway (2), whereas glucose is presumably dissimilated via the Embden-Meyerhof pathway. Previously published reports suggest that one or both of the enzymes responsible for the initiation of lactose metabolism are borne on a plasmid (1, 10). Attempts by several laboratories to delineate the pathway of galactose metabolism and relate it to lactose utilization have led to the publication of conflicting results. For example, Bissett t Laboratory of Biology of Viruses, National Institute of
Allergy and Infectious Diseases, Bethesda, MD 20014.
878
and Anderson (2) found that galactose-grown cells of S. lactis possess enzyme activities of both the tagatose-6-phosphate and the Leloir pathways, thereby providing the organism with apparent alternative dissimilatory mechanisms. However, Lee et al. (19) showed that the uptake of free galactose by toluene-treated cells of S. lactis C2 was greater in the presence of ATP than in the presence of PEP and concluded that the Leloir pathway was the major route for galactose metabolism by this organism. McKay et al. (25) published data which contradicted the findings of Lee et al. (19). Because all Lacmutants of S. lactis C2 concomitantly lost their ability to grow at the expense of galactose, McKay and co-workers (25) suggested that enzyme IJJaC played a role in galactose metabolism. This interpretation received further support when Cords and McKay (6) subsequently found that galactose-grown S. lactis cells contained low galactokinase activity, but high levels of P,B-galactosidase, and partial Lac' revertants of Lac- mutants were able to grow on galactose and contained higher galactokinase activity than the original Lac' parent, but no lactose PEP phosphotransferase. These results were taken to
VOL. 137, 1979
GALACTOSE METABOLISM BY S. LACTIS
mean that (i) galactose is metabolized via the tagatose-6-phosphate pathway by the parent cultures and (ii) the slower growing partial revertants were utilizing the LeLoir pathway. However, Demko et al. (8) published results which conflict with the interpretations of McKay and co-workers. They isolated Lac- derivatives of S. lactis ATCC 11454 which were not Gal- but exhibited only a slightly decreased growth rate on galactose. Whereas the fermentation of galactose by the Lac' parent was essentially homolactic, the Lac- strains exhibited a heterolactic fermentation of this sugar. In this communication we confirm the results of Demko et al. (8), showing that Lac- isolates of S. lactis retain the ability to grow on galactose. Our data also suggest that Lac' S. lactis metabolizes galactose primarily via the tagatose6-phosphate pathway, using the lactose PEP phosphotransferase to transport galactose into the cell as galactose-6-phosphate. When the ability to utilize lactose is lost, the loss of both lactose PEP phosphotransferase and P-f-galactosidase activity is accompanied by the loss of a 32-megadalton (Mdal) plasmid. These stable Lac- isolates metabolize galactose by a heterolactic fermentation and possess elevated levels of galactokinase activity, the first enzyme of the Leloir pathway. MATERLALS AND METHODS Bacterial strains and culture conditions. S. lactis (ATCC 11454) was obtained from the American Type Culture Collection (Rockville, Md.) and is designated here as strain DR1251. This strain is able to grow at the expense of lactose, galactose, glucose, sucrose, and ribose and produces the small peptide antibiotic nisin. Two types of media, complex medium (CM; 26) and defined medium (35), were used. Carbohydrate growth substrates were added at a final concentration of 0.5% (wt/vol). Solid media contained 1.5% Difco agar. Stock cultures were maintained at ambient temperature in reinforced skim milk (21). Unless otherwise stated, all incubations were at 30°C. Growth on solid media was under anaerobic conditions (GasPak, Baltimore Biological Laboratory). Determination of stability of lactose utilization. The rate of loss of lactose-utilizing ability at 30 and 37°C was determined by permitting a Lac' culture to grow for approximately 12 cell doublings in CM supplemented with glucose (CM glucose) at the appropriate temperature. The culture was then subjected to a 30-s sonication at a setting of 8 on a Kontes cell disrupter to disrupt cell chains, and dilutions were spread on defined medium supplemented with glucose and incubated at 30°C. Colonies were picked successively onto defined medium supplemented with lactose and defined medium glucose agar plates with sterile toothpicks. After incubation, Lac- phenotypes were confirmed in CM broth supplemented with lactose (CM lactose).
879
The effects of curing agents on the rate of loss of lactose-metabolizing ability were examined by diluting a Lac' culture 20-fold in CM glucose containing either acridine orange (50 ,ug/ml) or ethidium bromide (2 jug/ ml). After overnight incubation at 37°C the rate of appearance of Lac- colonies was determined as described above. Labeling and extraction of DNA. Overnight cultures, in 100 ml of CM supplemented with galactose (CM galactose) were transferred to 40 ml of the same medium containing 10 mM L-threonine (3) and 4 yCi of [3H]thymidine (40 to 60 Ci/mmol; New England Nuclear) per ml. Cultures were incubated at 30°C, and the cells were harvested in midexponential growth (150 Klett units; no. 66 filter; Klett-Summerson). Cell lysis was achieved by the method of Klaenhammer et al. (15) after a 5-min lysozyme treatment. Cleared lysates were prepared according to Guerry et al. (12). Supernatant fractions were extracted twice with equal volumes of chloroform-isoamyl alcohol (24:1), and the DNA in the final aqueous phases was precipitated with 2 volumes of cold ethanol. Isolation of plasmid DNA. Ethanol precipitates were suspended in 0.01 M Tris-hydrochloride-0.01 M EDTA-0.05 M NaCl, pH 8.2 (TES buffer), and covalently closed circular plasmid DNA was separated from open circular molecules and residual chromosomal DNA in equilibrium density gradients containing ethidium bromide, as previously described (17). Plasmid-containing fractions were pooled, and ethidium bromide was extracted with CsCl-saturated npropanol. The pooled plasmids were prepared for electrophoresis and electron microscopic examination by dialysis in TES buffer. Determination of plasmid sizes. Electrophoresis in agarose gels was as previously described (18). Molecular weights were estimated according to Meyers et al. (27) and by measuring contour lengths from electron micrographs. For the latter, purified plasmid DNA was incubated with pancreatic DNase I (0.03 U of enzyme per ml for 25 min at 25°C; conditions producing approximately one nick per molecule). Samples were then mounted for microscopy by the Kleinschmidt aqueous technique as described by Davis et al. (7). The spreading solution contained 0.1 mg of cytochrome c per ml and 0.5 M ammonium acetate. Grids were rotary shadowed with platinumpalladium and examined in a Siemens Elmiskop 101 at 40-kV accelerating voltage. Electron micrographs were taken on Kodak electron image plates at x6,000 and calibrated with a grating replica (E. F. Fullam, no. 1000). Negatives were projected (5x) onto a tablet where contour lengths were measured (Neumonics Graphics calculator). Data were stored and processed in a Wang 2200 computer. Enzyme assays. Cells were harvested in the late exponential phase of growth for all enzyme assays. Protein concentrations were determined by the biuret reaction (11). For determinations of galactokinase activities cells from 100-ml cultures were centrifuged, washed twice, and suspended in 0.05 M potassium phosphate buffer, pH 7.5, containing 10 mM fl-mercaptoethanol. Cell suspensions (6 to 7 ml) were sonically treated using three 3-min treatments at 175 W with a W-350 probe
880
J. BACTERIOL.
LFBLANC ET AL.
(Branson). Kinase activities were determined by the radioactive assay of London and Chace (20). PEP phosphotransferase activities were assayed in permeabilized cells by the method of Kornberg and Reeves (16), with modifications. The cells were harvested and washed twice with 0.1 M sodium-potassium phosphate buffer, pH 7.2, containing 5 mM MgCl2 (decryptification buffer) and suspended to 10% of the original culture volume in the same buffer. After chilling in an ice-water bath the cell suspension was vigorously mixed (maximum speed; Vortex Genie), and 50 ,il of toluene-acetone (1:9) was added per ml of cell suspension. The cells were again mixed for 5 min and kept in ice water until used for assay. PEP phosphotransferase assay mixtures contained, in a total volume of 1 ml: 0.1 mM NADH; 10 U of rabbit muscle lactic acid dehydrogenase (Worthington); 5 mM PEP; 10 mM NaF; carbohydrate substrate; and toluene-acetone-treated cell suspension. The final volume was adjusted with decryptification buffer. Reaction mixtures were allowed to equilibrate at 26°C, and the reactions were started by the addition of PEP. T'he rate of NADH oxidation (16) was followed at 340 nm in a quartz cell of 1-cm light path, using a Gilford 2400 recording spectrophotometer. NADH oxidase controls consisted of assay mixtures without added PEP. Specific activities are expressed as nanomoles of PEP-dependent NADH oxidized per minute per milligram (dry weight) of cells. Cell dry weights were determined by adding I ml of toluene-acetone-treated cell suspension to a preweighed aluminum cup (Fisher) and drying in a 70°C oven overnight. The dried cells were allowed to cool in a desiccator before weighing. Cell dry weights were calculated after substracting identically treated de-
cryptification buffer-toluene-acetone controls. /,-Galactosidase and P-,B-galactosidase activities were assayed according to the method of Citti et al. (4). Analyses of fermentation products. Cultures were grown in CM glucose or CM galactose, and samples were collected approximately 30 min after growth had stopped, as determined turbidimetrically. Cells were removed by centrifugation, and the supernatant fraction was acidified by the addition of 0.01 volume of concentrated HCl. Samples were frozen at -20°C until assayed. Lactic acid concentrations were determined enzymatically as described by Lowry and lPassonneau (22). Ethanol was assayed enzymatically (9) as well as by gas chromatography (29). Acetic acid concentrations were determined by gas chromatography (29).
RESULTS Growth of S. lactis DR1251 on lactose and galactose. The optimum temperature for growth of strain DR1251 on glucose was 37°C. However, this strain exhibited more rapid growth at 30°C than at 37°C on CM lactose or CM galactose. During exponential growth at 30°C the doubling time was 42 min on lactose and 52 min on galactose (Fig. 1). Lactose metabolism by strains of S. lactis is known to be an unstable characteristic (13, 23, 36), and McKay
0 1(
z -n
2
TIME
4 3 IN HOURS
5
6
FIG. 1. Growth of S. lactis strains DR1251 and DR1251/1 on lactose and galactose. One milliliter of an overnight culture grown in CM glucose was transferred to 9 ml of CM lactose or CMgalactose. Cultures were incubated at 300C, and growth was followed using a Klett-Summerson colorimeter equipped with a no. 66filter. Symbols: (0) strain DR1251 on lactose; (a) strain DR1251/1 on lactose; (0) strain DR1251 on galactose; (0) strain DR1251/1 on galactose.
et al. (23) reported that growth of S. lactis C2 at elevated temperatures increased the rate of loss of lactose-metabolizing ability. Instability of lactose utilization at elevated growth temperatures was also observed in strain DR1251. After approximately 12 cell doublings, at 37°C in the presence of glucose, 8% (12 of 144) of the colonies examined exhibited a Lac- phenotype. After the same number of cell doublings at 30°C, 100% (144) of the colonies tested had retained the ability to utilize lactose. The frequency of loss of lactose-metabolizing ability at 37°C was further increased by treatment with either acridine orange or ethidium bromide (see Materials and Methods), to 38% (72 of 191 colonies) and 22% (43 of 192 colonies), respectively. McKay et al. (25) reported that Lac- derivatives of group N streptococci are also Gal-. In contrast, all Lac- isolates of strains DR1251 from three independent curing experiments were able to grow on galactose. A Lac- Gal' isolate obtained after growth of the parent strain at 37°C on glucose, strain DR1251/1, was chosen for most of the studies described in this communication. Strain DR1251/1 was unable to grow on CM lactose (Fig. 1). The slight increase in cell density observed was the same as normally obtained in CM with no added carbohydrate and
VOL. 137, 1979
GALACTOSE METABOLISM BY S. LACTIS
is assumed to be due to low concentrations of a fermentable growth substrate present in CM. This Lac- isolate grew well in CM galactose but differed from the Lac' parent strain on galactose in that (i) the doubling time was increased from 52 to 80 min and (ii) it exhibited a typical diauxic growth curve, suggesting a requirement for enzyme induction after depletion of the fermentable substrate from the medium (Fig. 1). These results are in agreement with the data reported by Demko et al. (8) for Lac' and Lac- isolates of this strain of S. lactis. Identification of a lactose plasmid in S. lactis strain DR1251. Plasmid DNA was isolated from strains DR1251 and DR1251/1 by dye buoyant density gradient centrifugation. Plasmid-containing fractions were pooled and further examined by electrophoresis in agarose gels. Six major bands were observed in the lane containing plasmid DNA from strain DR1251 (Fig. 2A). The slowest migrating band was missing from the plasmid preparation obtained from Lac- strain DR1251/1 (Fig. 2B). To obtain better separation of the four slower migrating bands, and to confirm the absence of the slowest migrating plasmid in strain DR1251/1, both DNA preparations were concentrated and equal amounts of each were subjected to electrophoresis for 7 h (Fig. 2C and D). The plasmid nature of the DNA in each dye buoyant density gradient pool was confirmed by electron microscopy. Each pooled fraction contained between 85 and 95% covalently closed circular molecules. The molecular weight of each plasmid harbored by strain DR1251 was determined by (i) its relative mobility in agarose gels, using the a (6 Mdal), / (17 Mdal), and y (34 Mdal) plasmids from S. faecalis strain DS5 (5) as markers, and (ii) contour length measurements after limited DNase digestion of the covalently closed circular DNA preparations (Table 1). The above data suggest that the ability of S. lactis strain DR1251 to metabolize lactose is carried on a plasmid, pDRI, with a molecular weight of approximately 32 x 106. Five independent Lac- isolates of strain DR1251 were examined, and all were found to be missing the 32-Mdal plasmid. Enzyme activities associated with the lactose plasmid. Cultures of strains DR1251 and DR1251/1 were harvested near the end of the exponential phase of growth and assayed for PEP phosphotransferase activity on lactose, galactose, and glucose, as well as for P-/3-galactosidase and 3-galactosidase activity (Table 2). Lactose PEP phosphotransferase activity was high in strain DR1251 grown on lactose and galactose and considerably lower when ribose was the growth substrate, but repressed in cells
881
FIG. 2. Agarose gel electrophoresis of plasmid pools from dye buoyant density gradients. Removal of dye and cesium chloride and buffer conditions were as described in the text. Electrophoresis uwas at 60 mA and 80 V for (A, B) 5 h or (C, D) 7 h. (A, C) Plasmid DNA from strain DR1251; (B, D) plasmid DNA from strain DR1251/1.
grown on glucose. Strain DR1251/1 was devoid of PEP phosphotransferase activity for both lactose and galactose after growth on each of the three substrates. Both strains had high glucose PEP phosphotransferase activities. Strain DR1251 has extremely high P-/3-galactosidase activity when grown on lactose, galactose, or ribose but not when grown on glucose.
882
LEBLANC ET AL.
J. BACTERIOL.
Since ribose would not be expected to be an inducer of P-,8-galactosidase, this enzyme appeared to be synthesized constitutively by S. lactis strain DR1251. No significant P-/8-galactosidase activity was observed with strain DR1251/1 after growth on any of the substrates. Neither strain had jB-galactosidase activity, as measured by o-nitrophenyl-f3-D-galactopyranoside hydrolysis. The results illustrated in Fig. 2 and Table 2 clearly show that the loss of plasmid pDR1 by S. lactis strain DR1251 is accompanied by the loss of P-,8-galactosidase activity and PEP phosphotransferase activity for both lactose and galactose. It is unlikely that the absence of P-/Igalactosidase activity in strain DR1251/1 grown on galactose is a reflection of the inability to convert galactose to the true inducer, galactose6-phosphate, as previously suggested (6), since this activity is also present in the parent strain, DR1251, grown on ribose. Thompson et al. (34) have suggested that TABLE 1. Summary ofplasmid species in S. lactis strain DR1251 Mol wt determinations Plasmid
0.7% agarose gels Relative Mol wt mobility x 106a
pDR,
Electron microscopy Contour length Mol wt x 1o6b (pm) 12.26 ± 0.66 31.8 14.86 ± 0.47 29.1
0.80 31.5 0.84 28.5 0.91 23.5 11.19 ± 0.31 21.9 1.00 20.0 9.91 ± 0.24 19.4 2.55 3.0 1.84 ± 0.14 3.6 3.51 1.5 pDR6 0.76 ± 0.04 1.49 a Based on mobility of pAMal (6 x 106), pAM/,I (17 x 106), and pAM-yi (34 x 106) from S. faecalis strain DS5 (5), with relative mobilities of 1.80, 1.15, and 0.77,
pDR2 pDR3 pDR4 pDR5
respectively.
bDerived from linear density of 1.96 x 106 daltons per pm. Based on 45 to 105 molecules counted.
transport of lactose and galactose by S. lactis is
most probably mediated by a lactose PEP phosphotransferase. The results shown in Table 3
suggested that the galactose PEP phosphotransferase activity of strain DR1251 was due to the lactose PEP phosphotransferase. Cells grown on either lactose or galactose exhibited an apparent Km of the PEP phosphotransferase system for lactose of approximately 5 x 10-5 M, whereas for galactose the apparent Km was 2 x 10-2 M. Similar results have been reported for the lactose PEP phosphotransferase of Staphylococcus aureus
(30, 31).
Fermentation products and galactokinase activities of strains DR1251 and DR1251/1 grown on galactose. Demko et al. (8) reported that Lac- isolates of S. lactis produced less lactic acid from galactose than the parent Lac' strain. We obtained similar results with strains DR1251 and DR1251/1 (Table 4). Whereas the ratio of lactic acid to either acetate or ethanol was approximately 16 to 1 when strain DR1251 was grown on galactose, strain DR1251/ 1 produced only five times more lactic acid than either acetate or ethanol from galactose as the growth substrate. Both strains appeared to metabolize glucose via a homolactic fermentation. A role of the Leloir pathway in the fermentation of galactose was suggested by the presence of galactokinase activity, a key enzyme of this pathway, in both strains. The specific activity of this enzyme increased fivefold in the Lac- strain after growth on galactose, relative to the Lac' parent strain (Table 4).
DISCUSSION S. lactis strain DR1251 produces relatively high levels of lactose PEP phosphotransferase, galactose PEP phosphotransferase, and P-/8-galactosidase when grown on either lactose or galactose at 30°C. However, growth on substrates
TABLE 2. PEP-dependent phosphotransferase, /-galactosidase, and P-fl-galactosidase activitiesa in permeabilized cells of S. lactis strains DR1251 and DR1251/1 PEP-dependent phosphotransferase sp act Strain
DR1251
Energy source for growth
5 mM lactose
with: 5 mM galactose
1 mM
Lactose 109 32 Galactose 248 31 D-Ribose 36 4 Glucose 15 5 DR1251/1 Galactose 0 0 1 D-Ribose 1 Glucose 5 0 a Activities expressed as nanomoles per minute per milligram of cell b ONPG, o-Nitrophenyl-f-D-galactopyranoside.
glu-
Hydrolysis of
ONPG-6-pb
ONPG
594 360 549 21 2 29 19
0
cose
109 195
76 137 102 55 57 dry weight.
2
32 17 14 39 13
GALACTOSE METABOLISM BY S. LACTIS
VOL. 137, 1979
other than lactose or galactose at 37°C resulted in a high frequency of loss of ability to grow at the expense of lactose. The rate of loss of this characteristic is accelerated if the organism is grown in the presence of plasmid curing agents. Whereas Lac- isolates of strain DR1251 retain the ability to grow on galactose, lactose PEP phosphotransferase, galactose PEP phosphotransferase, and P-,8-galactosidase activities are not detectable in permeabilized cell suspensions harvested from galactose-containing media. Such Lac- strains have also lost one of six plasmids normally found in the parent strain, namely, the 32-Mdal plasmid pDR1. With the exception of S. lactis strain 7962, which possesses only ,B-galactosidase activity, McKay et al. (24, 25) found that galactose was a better inducer of P-,B-galactosidase than lactose for all other strains tested. However, the observations of Demko et al. (8) disagreed with these findings and showed that the levels of PB-galactosidase activity were roughly the same when they compared extracts from lactose- and galactose-grown cells of strain ATCC 11454 (DR1251). Our results agree with those of Demko and co-workers and further suggest that this enzyme is derepressed in strain DR1251 grown on ribose. On the other hand, lactose PEP phosphotransferase activity does not appear to be derepressed under the same conditions and, although induced in the presence of either lactose or galactose, the latter substrate serves as a better inducer. These results suggest that the two enzymes may be independently regulated in TABLE 3. Relative Km for lactose and galactose of PEP-dependent phosphotransferase activities in S. lactis strain DR1251 Energy source for growth
(M)a Ko se/Ka Lactose
Galactose
lactose
4.0 x 10-5 1.5 x 10-2 375 386 7.0 x 10-5 2.7 x 10-2 a Determined from double-reciprocal plots of activity versus substrate concentration.
Lactose Galactose
883
strain DR1251. Whether the structural and/or regulatory genes for both activities are encoded on the Lac plasmid pDR, or are chromosomal genes has not been determined. The galactose PEP phosphotransferase activity found in permeabilized cells of S. lactis DR1251 after growth on either lactose or galactose is absent from galactose-grown Lac- strains. These results, coupled with the data on relative Km values of the PEP phosphotransferase system for lactose and galactose, 5 x 10-5 and 2 x 10-2 M, respectively, strongly suggest that the same transport system is operative for both substrates. Since the metabolism of galactose via the tagatose-6-phosphate pathway requires a galactose PEP phosphotransferase to phosphorylate-free galactose in the 6-carbon position (6), galactose utilization by Lac- derivatives of strain DR1251 must occur via a different biochemical pathway. Evidence supporting the utilization of the Leloir pathway as the alternative route for galactose metabolism by the Lac- isolates is suggested by the increase in galactokinase activity in these cells, relative to Lac' cells grown on galactose. Fermentation patterns provide another indirect line of evidence for the operation of the Leloir pathway. Lactose is utilized by group N streptococci via a homolactic fermentation, and the galactose moiety is metabolized via the tagatose-6-phosphate pathway (34). The metabolism of galactose by S. lactis DR1251, if occurring via the tagatose-6-phosphate pathway, should also exhibit a homolactic fermentation. The results illustrated in Table 4 show that although the proportionate amounts of ethanol and acetate produced by the parent Lac' strain grown on galactose are considerably less than the Lac- isolate on this substrate, the catabolism is to some extent a mixed-acid fermentation. However, the facts that the predominant product of galactose metabolism by strain DR1251 is lactic acid and that galactokinase activities are very low relative to the Lac- strain suggest that the primary route of galactose dissimilation by the wild-type strain is the tagatose-6-phosphate pathway.
TABLE 4. End products and galactokinase activities ofgalactose- and glucose-grown S. lactis strains DR1251 and DR1251/1 End products formed (mM)
Strain
DR1251
DR1251/1
Galactoki-
ns Lactic nase Energy sourceLatc acid/ethan- (t,mol/min for growth
Galactose Glucose Galactose Glucose
Lactic acid
Acetic acid
Ethanol
ol ratio
34.5 ± 0.1 35.3 ± 0.2 21.0 ± 0.2 36.2 ± 0.1
2.07 ± 0.05
