Kinetics, Stereospecificity, and Expression of the Malolactic Enzyme

Abstract

Mass spectrometric measurement of carbon dioxide production was used to study malolactic fermentation (MLF) in Lactobacillus collinoides isolated from cider. The kinetics and stereospecificity of the malolactic enzyme (MLE) were studied, and the stoichiometry of the reaction sequence was investigated. The optimum pH for activity of the MLE was 4.9. MLF was more rapid (in both intact cells and cell extracts) when l-malic acid was used than when d-malic acid or the racemic mixture was added. The enzyme was found to be constitutively present in L. collinoides. Addition of l-malic acid (37 mM) to the growth medium resulted in increased MLE activity; addition of the d isomer alone or the racemic mixture resulted in lower activities. Addition of the main sugars in apple juice (fructose, sucrose, and glucose) to the growth medium in the presence of malic acid repressed production of MLE to similar extents in all three cases; in the absence of malic acid, instead of inhibiting MLF, addition of sugars to the growth medium somewhat increased the residual MLE activity.

Two distinct processes are involved in cider making (9, 11). The initial step is alcoholic fermentation performed by yeast (Saccharomyces sp.) (6, 12, 19, 22). This is followed by malolactic fermentation (MLF) (18) performed by lactic acid bacteria (LAB) (3, 8, 21). The overall result of the secondary fermentation is production of the characteristic flavor and aroma of cider.

Approximately 8% of the fresh weight of an apple is accounted for by a mixture of sugars (10), and apple juice contains principally fructose, glucose, and sucrose (9). Fructose is present at the highest concentration in apples and accounts for about 70% of the total sugar (10). Sucrose and glucose are much less abundant and account for 10 to 20% of the total sugar (3). Following alcoholic fermentation all of the fermentable sugar should be converted to alcohol. Any residual sugar may influence the secondary fermentation. Hence, we decided to investigate the effects of residual sugars on the MLF, both in the presence and in the absence of malic acid. We selected concentrations of sugars which reflected the low residual levels which may occur. Sugar (fructose and glucose) catabolism is inhibited in the highly acidic environment that is characteristic of cider fermentation; the low pH values maximize the rate of malate utilization and inhibit the rate of glucose consumption by some LAB (7).

The malolactic enzyme (MLE) which is responsible for the MLF is present in many species belonging to the genera Lactobacillus, Leuconostoc, Oenococcus, and Pediococcus and has been purified (4, 20). This enzyme decarboxylates l-malic acid (the main acid found in apples [1]) to form l-lactic acid directly (to date no intermediate products have been detected), which results in an overall decline in acidity (7, 14). For each malate molecule metabolized, equimolar amounts of lactate and carbon dioxide are produced; i.e., the reaction is stoichiometrically equivalent. The level of recovery of lactic acid from malic acid is about 90% (17). The rate of the MLF can therefore be accurately determined by measuring the rate of pH change, CO₂ formation, or lactic acid formation, although it should be noted that other naturally occurring organic acids can also be metabolised by LAB (e.g., quinic acid is converted into dihydroshikimic acid) (1).

In this paper we describe the use of a membrane inlet quadrupole mass spectrometer to study (i) the kinetics, (ii) the stoichiometry, (iii) the stereospecificity, and (iv), the conditions which favor expression of the enzyme involved in MLF.

MATERIALS AND METHODS

Cell growth.

An LAB isolate obtained from cider fermentation was provided by H. P. Bulmer Ltd. (The Cider Mills, Hereford, United Kingdom). This organism was tentatively identified (by using the API system [API] bioMérieux, Vercieu, France]) as Lactobacillus collinoides. Cultures were routinely grown in 250 ml of modified Lactobacillus MRS medium (5) (Merck Ltd., Dorset, United Kingdom) containing 37 mM dl-, d, or l-malic acid in stationary 250-ml Erlenmeyer flasks without baffles sealed with rubber bungs and incubated at 25°C (standard conditions). The LAB was also grown in MRS medium supplemented with fructose, sucrose, or glucose both in the presence and in the absence of l-malic acid under conditions identical to those described above. The final pH was adjusted by using 1 M NaOH and 1 M HCl.

Harvesting.

Growth was monitored by measuring absorbance at 660 nm compared with medium blanks. Once the late exponential growth phase had been reached, the cells were harvested by centrifugation with a Sorvall model RC-5B centrifuge fitted with a type GSA rotor (6,000 × g, 6 min, 25°C). Usually, they were then washed and resuspended in 10 ml of 5 mM 2-(N-morpholino)ethanesulfonic acid (MES) (Sigma, Dorset, United Kingdom) (pH 4.5) before they were stored briefly (1 to 4 h) on ice prior to use. Preliminary experiments indicated that storage resulted in no significant changes in activity. On other occasions (as indicated below) the procedure was repeated, and the harvested cells were resuspended in 5 mM MES adjusted to pH 3.5, 4.1, 4.3, 4.5, 4.9, and 5.5.

Preparation of cell extracts.

A model Soniprep 150 sonicator (amplitude, 22 μm; five 30-s pulses over a 5-min period) was used to break the walls of cells that had been resuspended at pH 4.5. The cell extract was then obtained by centrifugation with a Sorvall model RC-5B centrifuge equipped with a type SS34 rotor (6,000 × g, 6 min, 25°C) and collection of the supernatant.

MLF. Resuspended intact cells or cell extracts were added to a specially constructed Lucite reaction vessel (total volume, 4 ml) that was surrounded by a water jacket (kept at 25°C) and was fitted with a sealed cap with a 0.5-mm-diameter injection port. A pH probe attached to an analogue pH meter (model PW9418; Philips Analytical, Cambridge, United Kingdom) and a mass spectrometer probe were inserted and cemented in place. The reaction mixture was magnetically stirred at 200 rpm. A quadrupole mass spectrometer (Hiden Analytica, Ltd., Warrington, United Kingdom) in electron ionization mode measuring at m/z 44 was equipped with a membrane inlet probe. A tubular silicone membrane (inside diameter, 0.5 mm; outside diameter, 1.0 mm; Vygon) was used with the 1.6-mm stainless steel probe (2, 13).

The MLF was initiated by adding d-, l-, or dl-malic acid in 5 mM MES (pH 4.5) to the vessel containing washed intact cells or cell extracts (pH 4.5) which had been grown in medium supplemented with 37 mM dl-malic acid. Both the amounts of CO₂ produced and pH changes were recorded.

The procedure was then repeated by adding dl-malic acid preparations with the pH adjusted to 3.5, 4.1, 4.3, 4.5, 4.9, and 5.5 to the vessel containing washed whole cells (previously grown in MRS medium [pH 4.5] containing 37 mM dl-malic acid) at pH 3.5, 4.1, 4.3, 4.9, and 5.5, respectively. The reactions were monitored by measuring CO₂ production by mass spectrometry.

dl-Malic acid (37 mM) in 5 mM MES (pH 4.5) was used to initiate the MLF in washed whole cells that had been grown in media supplemented with sugar in the presence and in the absence of different isomers of malic acid. The reaction rates were again determined by measuring CO₂ evolution by mass spectrometry.

NaHCO₃ dissociates in the presence of HCl to produce CO₂, the concentration of which depends on the amount of NaHCO₃ added. Hence, adding 10 mM NaHCO₃ to excess HCl was used to calibrate the mass spectrometer for CO₂ at m/z 44.

Protein analysis.

The protein contents of intact cell suspensions and cell extracts were determined by using a modification of the method described by Lowry et al. (15); bovine serum albumin was used as the standard. Protein was extracted by boiling preparations (for 5 min with 1 M NaOH in closed tubes) before the assay.

Reproducibility of results.

The experiments in which intact organisms and cell extracts were used were all performed in triplicate, and the standard errors of the means were found to be less than 0.2%.

RESULTS AND DISCUSSION

MLF investigated at different pH values by measuring CO₂ production.

Our results indicated that the pH at which the mass spectrometry analysis was carried out was very influential in determining the MLE activity. Figure 1 shows that L. collinoides produced very little CO₂ from malate at pH 3.5 and 5.5 but exhibited a high level of activity at pH 4.9 (K_m, 100.2 mM; V_max 1,410 nmol/min/mg of protein). The optimum pH for l-malate transport in Lactobacillus plantarum has previously been found to be 4.5 (16).

FIG. 1.

MLF rates in samples from L. collinoides cultures growing under standard conditions and analyzed at different pH values.

Although malate-driven changes in pH were monitored by using samples of washed whole-cell suspensions taken at intervals during the MLF, the results obtained were not as reliable as the results obtained when carbon dioxide production was measured by mass spectrometry. This may be attributed to the less sensitive method for measuring pH in the presence of a buffer (5 mM MES) which counteracted pH changes.

MLF in intact cells in which different isomers of malic acid were used as substrates for the MLE.

The MLF rate when a racemic mixture of malic acid was used as the substrate was greater than the MLF rate when equimolar amount of either isomer alone was used. The results obtained for L. collinoides are shown in Table 1. Although the racemic mixture was inferior as a substrate compared with the l isomer, this mixture was more effective than the d isomer. The l isomer induced l-malate transport in L. plantarum to a greater degree (100%) than either the racemic mixture (82%) or the d isomer (66%) (16).

TABLE 1.

Substrate affinities, reaction velocities, and sustrate specificities for the MLF in L. collinoides^a

Acid	Whole cells		Cell extracts
	Km (mM)	Vmax (nmol/min/mg of protein)	Km (mM)	Vmax (nmol/min/mg of protein)
d-Malic acid	33.1	20	8.3	1,470
l-Malic acid	83.6	50	3.7	3,600
dl-Malic acid	17.3	42	8.0	1,490

^aK_m and V_max values were obtained from best-fit lines on Lineweaver-Burk plots obtained for whole-cell suspensions or cell extracts obtained from cultures grown under standard conditions. 

The V_max values obtained reflect either the transport of malic acid into the cell or, alternatively, the metabolism of malic acid. There are a number of possible explanations for our findings.

The results which we obtained may be attributed to a concerted transport mechanism, in which one isomer reduces the uptake and, therefore, the metabolism of the other. This proposal supports the hypothesis that malic acid cannot readily diffuse across the cytoplasmic membrane (16). On the basis of the V_max values obtained (Table 1), we deduced that it is uptake and metabolism of the l isomer that are reduced in the presence of d-malate. It has been reported previously that the l-malate carrier is stereospecific and is not significantly inhibited by d-malate (16).

Alternatively, racemization (intracellular or extracellular) may be the rate-limiting factor. If this is so and intracellular racemization occurs, both isomers of malate would be transported into the cell at similar rates. Once inside the cell, the d isomer would then have to be transformed into the l isomer before it was accepted as a substrate for the MLE in order to undergo the decarboxylation reaction that results in production of lactic acid.

A third possibility is that extracellular racemization is the rate-limiting step. If this were the case, the l isomer would be expected to exhibit the highest V_max value, followed by the racemic mixture and finally the d isomer. The results obtained in this study clearly show that this was the case (Table 1).

The final possibility is that MLF rates (as calculated in this study) are determined primarily by the ability of the MLE to convert l-malic acid to lactic acid. In this process the d isomer might interact with the enzyme to alter its affinity for l-malate.

In order to distinguish between the hypotheses described above and thus determine whether the measured V_max values are the V_max values for the malic acid transporter or are V_max values for the MLE itself, MLF was studied in cell extracts. With a cell extract if identical V_max were obtained with all three substrates (the d isomer, the l isomer, and the racemic mixture), then the substrate affinity-determining step would be assumed to be the transport mechanism. However if different V_max values were observed, the results could be explained by either (i) the enzyme having a different affinity for each of the isomers or (ii) the d isomer requiring racemization before it could be utilized as a substrate by the MLE.

MLF in cell extracts in which different isomers of malic acid were used as substrates for the MLE.

The MLF was investigated with cell extracts in order to determine whether the V_max values obtained with intact cells represented the rate of transport of malic acid into the cell or were a measure of the rate of conversion of malic acid into lactic acid (Table 1).

The cell extract V_max values were lower than the V_max values for intact cells, indicating that the transport system for malic acid favors the l isomer over the d isomer (Table 1); i.e., l-malic acid is transported into the cell by a higher-affinity system than the system used for d-malic acid. We concluded, therefore, that the presence of d-malic acid significantly reduces the uptake of l-malic acid into the cell.

The K_m values described in this paper (Table 1) correlate well with previously reported values; e.g., the purified preparation of MLE obtained from L. plantarum was found to have a K_m for malate of 9.5 mM by Caspritz (4).

The V_max values obtained for cell extracts indicate that the MLE has the greatest affinity for l-malic acid (V_max 3,600 nmol/min/mg of protein). When cell extracts are used, it is the rate of conversion of l-malic acid to l-lactic acid which is the rate-limiting step.

Interestingly, the racemic mixture and the d isomer (when used individually as substrates for the MLE) had very similar V_max values (1,490 and 1,470 nmol/min/mg of protein, respectively). Therefore, we propose that the d isomer reduces the affinity of the MLE for l-malic acid and also must be converted into l-malic acid extracellularly before it is accepted as a suitable substrate by the MLE.

In summary, we propose that the d isomer has two effects on MLF in LAB; not only does it reduce the uptake of the l isomer, but it also reduces the affinity of the enzyme for l-malic acid. In addition, d-malic acid requires extracellular racemization in order to form l-malic acid before it can be used as a substrate for the MLE. This proposal adequately explains the results obtained for both cell extracts and intact cells described in this paper.

Expression of the MLE with different isomers of malic acid.

LAB were grown in the presence and absence of 37 mM d-, l-, or dm-malic acid, and the activity of the MLE was determined. When grown in the absence of malic acid, L. collinoides was able to perform the MLF, which indicated that the enzyme was constitutively present. l-malate has previously been reported to be responsible for inducing elevated levels of the enzyme (14). Thus, as expected, addition of malic acid to the growth medium also increased the level of expression of the enzyme in L. collinoides. l-malic acid induced the MLE to a greater extent than either d-malic acid or the racemic mixture induced this enzyme (Fig. 2). Both the d and l isomers of malic acid (when used individually or in combination) enhanced expression of the MLE, and it was clear that l-malic acid resulted in greater expression than the d isomer (Fig. 2).

FIG. 2.

MLF rates in samples from L. collinoides cultures growing with various concentrations of d-, l-, or dl-malic acid. The standard medium was supplemented with d-malic acid (▴), l-malic acid (■), or dl-malic acid (⧫).

Effect that sugars in the growth medium have on induction of the MLE.

When MRS medium containing 37 mM dl-malic acid (with no added sugars) was used as the growth medium for L. collinoides, the highest level of MLF activity was observed. Addition of 15 mM sucrose, addition of 28 mM glucose, or addition of 28 mM fructose to this medium inhibited expression of the MLE to similar extents (Fig. 3). The lowest level of MLF activity occurred when L. collinoides was grown in MRS medium (pH 4.5) which had not been supplemented with either malic acid or carbohydrates.

FIG. 3.

MLF rates in samples from L. collinoides cultures growing with sugar supplements. (a) Standard growth medium containing 37 mM malic acid. (b through d) Medium supplemented with 28 mM fructose (⧫ and ◊). (b) 28 mM glucose (■ and □) (c), and 15 mM sucrose (▴ and ▵) (d) in the presence of 37 mM malic acid (solid symbols) and in the absence of malic acid (open symbols). The rates in the absence of both malic acid and sugars are also indicated (○).

When L. collinoides was grown in the absence of malic acid, additions of sugars to the growth medium increased MLE activity instead of inhibiting MLF. The enhancement of MLF activity was greater with either glucose or sucrose than with fructose (Fig. 3). Maximum expression of the MLE was observed when the growth medium was supplemented with 37 mM l-malic acid alone.