Selection of Strain, G r o w t h C o n d i t i o n s , and E x t r a c t i o n Procedures f o r O p t i m u m P r o d u c t i o n of Lactase f r o m K l u y v e r o m y c e s fragilis R. R. MAHONEY, T. A. NICKERSON, and J. R. WHITAKER Department of Food Science and Technology University of California Davis 95616 ABSTRACT Forty-one strains of Kluyveromyces fragilis (J/Srgensen) van der Wait 1909 varied 60-fold in ability to produce lactase (~-galactosidase). The four best strains were UCD #55-31 (Northern Regional Research Center NRRL Y-1196), UCD # C 2 1 ( ) , UCD #72-297(--), and UCD #55-61 (NRRL Y-1109). Biosynthesis of lactase during the growth of K. fragilis strain UCD #55-61 was followed on both lactose and sweet whey media. Maximum enzyme yield was obtained at the beginning of the stationary phase of growth. Best lactase yields from K. fragilis UCD (455-61 were obtained with 15% lactose and an aeration rate of at least .2 mmol oxygen/liter per min. Supplementary growth factors were unnecessary for good lactase yields when yeast was grown on whey media. Best extraction of lactase from fresh yeast cells was obtained by toluene autolysis (2% vol/vol) at 37 C in .1 M potassium phosphate buffer, pH 7.0, containing .1 mM manganese chloride and .5 mM magnesium sulfate. The enzyme was concentrated and purified partially by acetone precipitation. At least 95% of the enzyme activity of the concentrated solution was retained after storage for 7 days at 22 C, for 3 wk at 4 C, and for 6 wk at - 2 0 C. INTRODUCTION Lactose, in milk at about 4.8%, is a rather insoluble, not sweet sugar which may have a mild laxative effect when consumed in large amounts because many adults and some infants have a low tolerance for it (15, 18). These lactose-sensitive individuals suffer from bloat-

Received May 6, 1974.

ing, cramps, diarrhea, and malabsorption when they consume sufficient quantities of milk or some milk products. As a result, a valuable source of food is unacceptable to them. The low solubility of lactose contributes to its crystallization which is often responsible for a gritty texture in concentrated dairy products. Low tolerance for and insolubility of lactose have limited its use in foods such as ice cream, candy, and animal feeds. These factors also have limited the utilization of whey because of its high lactose content. Consequently, whey constitutes a severe problem of disposal for the dairy industry. There is widespread interest, both from the nutritional and commercial viewpoints, in reducing lactose in certain dairy products. One of the most promising methods is hydrolysis of lactose into glucose and galactose by lactase (13-galactosidase, EC 3.2.1.23). Lactase has been isolated from several microbial sources including Escbericbia coli (14), fungi (3), and lactose fermenting yeasts (2). Kluyveromyces j?agilis (formerly Saccbaromyces fragilis) is one of the best sources of the enzyme. The yeast has the added advantage of being approved for use in foods. Methods of preparing crude Iactase from K. fragilis have been described by Caputto et al. (4), Van Dam et al. (21), Stimpson (19), and Wendorff (22). Some factors affecting biosynthesis of lactase by K. fragilis were reported by Davies (9). Nutrient requirements and growth conditions for biosynthesis of lactase have been described by Wendorff et al. (23) and Young and Healey (25). The partially purified enzyme has been characterized by Chan (5) and Wendorff and Amundson (24). Several methods have been reported for extraction o f lactase from K. fragilis. The method used most frequently in recent years is buffered extraction of dried yeast (5, 17, 24). Other methods include autolysis with chloroform (21) and with toluene (11) and mechani-

1620

LACTASE FROM K. FRAGILIS cal breakage of the ceils with glass beads (10). The purpose of this study was to investigate factors which influence biosynthesis of lactase by K. fragilis and to select strains and conditions of yeast growth and enzyme extraction which maximize lactase yield. MATERIALS AND METHODS Reagents and Organisms

Lactose (USP grade) and dried sweet whey (Teklac, containing 65% lactose) were obtained from Foremost Foods, San Francisco, CA. Yeast nitrogen base and yeast extract were from Difco Laboratories, Detroit, MI. Spraydried K. fragilis cells, used in part of the work, were obtained from Kraftco Company, Chicago, IL. Corn steep liquor samples were obtained from: American Maize-Products Company, Hammond, IN; Clinton Corn Processing Company, Clinton, IA; Penick and Ford Limited, Cedar Rapids, IA, and Staley Manufacturing Company, Decatur, IL. o-Nitrophenyl /3-Dgalactopyranoside (ONPG) and isopropylthio fi-D-galactopyranoside (IPTG) were from Calbiochem, La Jolla, CA. Glucostat Reagent was from Worthington Biochemical Corporation, Freehold, NJ. All other compounds were reagent grade. Distilled-deionized water was used. Strains of K. fragilis were from the Yeast Culture Collection, Department of Food Science and Technology, University of California, Davis (see Table 1 for listing). These strains had been maintained on slants of 10% malt-extract agar. The yeasts were transferred to 1.5% agar slants containing 2% (wt/vol) lactose and .5% (wt/vol) yeast extract and, after suitable growth at 30 C, were stored at 4 C. Transfers were made to new lactose-containing slants, to achieve fresh growth, just prior to use. Propagation

Yeast was transferred from the slants to a sterile solution of 2% (wt/vol) lactose containing .7% (wt/vol) yeast nitrogen base and the solution gently agitated for 18 h at 30 C. Suspensions of .1% and 1% (vol/vol) yeast were added to the propagation medium in shake flasks and fermentor jars, respectively. The propagation medium was a modification of that used by Wendorff et al. (23) and contained 10% lactose in water together with "supplementary

1621

nutrients" [.3 % (wt/vol) K2 HPO4, . 5% (wt/vol) yeast extract, and .3% (vol/vol) 5 N NH4 OH]. The medium was adjusted to pH 4.5 with HzSO4, autoclaved for 15 rain at 121 C, cooled to 30 C, and inoculated as described above. In 9ther experiments, this medium was varied either by substituting dried whey for lactose, varying the amount of lactose, or altering the source of growth factors. Whey-containing media were prepared by dissolving sufficient dried sweet whey in water to give 15% (wt/vol) lactose. The solution was adjusted to pH 4.5 with concentrated HC1, heated at 90 C for 5 rain, cooled, and filtered through a layer of neutral diatomaceous earth to remove coagulated protein. "Supplementary nutrients" (above) were added, the pH adjusted to 4.5 with H2SO4, and the medium pasteurized at 70 C for 30 s. Shake flask cultures were propagated by shaking 1 liter of inoculated medium in a 2.8 liter Fernbach flask at 30 C on a rotary shaker at 170 -+ 20 rotations per min. The oxygen absorption rate over the first 4 h was .10 mmol O2/liter per rain. Propagations also were in either 4- (3 liters medium) or 12- (9 liters medium) liter fermentor jars (New Brunswick Scientific Co., New Brunswick, NJ), equipped with spargers, at 30 C under controlled rates of aeration and agitation. Oxygen absorption rates were measured by a sulfite oxidation method (8). Increase in cell mass was followed by an increase in absorbance as measured in a KlettSummerson colorimeter (red filter). Lactase Extraction

Spray-dried K. fragilis yeast from Kraftco was used to establish basic extraction parameters in the initial studies because of the large amount of cells needed. Potassium phosphate buffer, containing .1 mM MnCI2 and .5 mM MgSO4 in all cases, was used in the extraction experiments. Unless otherwise indicated, after extraction the suspension of cells and enzyme was centrifuged for 30 rain at 5000 x g, and lactase activity was determined on the supernatant liquid. The effect of pH on extraction of lactase was determined by suspending spraydried cells (100 mg/ml) in .1 M phosphate buffers at various pH values for 40 h at 22 C. Other extractions were under similar conditions except that temperature (4 to 37 C) and concentrations of phosphate (1 mad to 1 M), Journal of Dairy Science Vol. 58, No. 11

1622

MAHONEY ET AL.

TABLE 1. Lactase biosynthesis by strains of Kluyveromyces fragilis (Jorgensen) van der Walt 1909. a

Yeast strain

Yeast per liter of medium

Activity per mg dry weight

Total yield per liter of medium

(UCD #)b

(NRRL #)c

(g)

(lactose units)

(lactose units X 10-3)

55-31 C21 72-297 55-61 55 -25 55-6 3 55-59 55-35 55-3 55-38 55-62 55-29 55-52 55-22 55-57 55-2 55-6 55-13 72-5 5 5-9 55-32 55-16 50-16 50-18 55-27 55-36 55-20 55-55 C-106 55-66 55-56 55-1 55-19 55-4 55-30 50-84 50-29 55-12 55-7 55-5 55-26

Y-1196 ...... ...... Y-1109 Y-1159 Y-1156 Y-1207 Y-1100 Y-108 Y-1208 Y-1137 Y-1183 Y-1163 Y-1151 Y-1195 Y 88 Y-610 Y-1012 ...... Y-872 Y-1204 Y-1122 ...... ...... Y-1177 Y-1200 Y-1149 Y-1176 ...... Y-1198 Y-1182 ...... Y-1131 Y-176 Y-1189 ...... ...... Y-908 Y-653 Y-208 Y-1172

5.121 5.313 4.984 4.325 3.870 5.004 3.669 4.328 3.742 3.819 4.933 4.424 4.476 3.501 4.067

2.45 2.30 2.37 2.46 2.45 1.79 2.36 1.87 2.08 2.03 1.55 1.40 1.22 1.51 1.20 1.23 1.12 1.17 .786 .973 .945 .805 1.21 .814 .946 1.07 .853 .786 .781 .728 .464 .645 .705 .914 1.00 .736 .202 .340 .099 .241 .039

12.5 12.2 11.8 10.7 9.48 8.96 8.66 8.09 7.78 7.75 7.65 6.19 5.46 5.29 4.88 4.87 4.42 4.39 4.26 4.20 4.13 4.02 3.70 3.41 3.38 3.37 3.06 3.00 2.95 2.85 2.85 2.84 2.79 2.71 2.70 2.68 1.21 1.18 .597 .449 .179

3.963 3.947 3.753 5.421 4.318 4.368 4.992 3.058 4.189 3.576 3.150 3.584 3.817 3.774 3.918 6.140 4.405 3.953 2.964 2.697 3.645 6.004 3.462 6.031 1.864 4.578

ayeast cells grown to beginning of stationary phase on 10% lactose plus supplementary nutrients in shake flasks. bDesignation o f the Yeast Culture Collection, Department of Food Science and Technology, University of California, Davis. C

.

.

DeslgnaUon of the Northern Regional Research Center (NRRL), USDA, Agricultural Research Service, Peoria, IL.

t o l u e n e (0 t o 5%), 2 - m e r c a p t o e t h a n o l (0 t o .1 M), a n d d i t h i o t h r e i t o l (0 t o .1 34) w e r e v a r i e d in separate experiments. W h e n f r e s h y e a s t w a s u s e d , cell walls w e r e Journal of Dairy Science Vol. 58, No. 11

broken by organic solvent (toluene or chlorof o r m ) autolysis or b y h o m o g e n i z a t i o n with glass b e a d s p r i o r t o l a c t a s e e x t r a c t i o n . Cells, a t a c o n c e n t r a t i o n o f 1 0 0 m g / m l , in .1 34 p h o s -

LACTASE FROM K. FRAGILIS phate buffer, pH 7.0, and various concentrations of organic solvent (0 to 10%) were autolyzed for various times at 4, 22, 30, and 37 C, the suspension centrifuged, and lactase activity determined. Cells (175 mg/ml) in .1 M phosphate buffer, pH 7.0, also were completely broken (verified by microscopic examination) by shaking with an equal volume of glass beads (.45 to .50 mm diameter) for 1 min in a Braun Homogenizer (Bronwill Scientific Co., Rochester, NY) at 3 + 2 C. The homogenate was extracted with an equal volume of the same phosphate buffer for 4 h at 4 C and then centrifuged to remove cell debris prior to determination of activity. Enzyme Purification K. fragilis UCD #55-61 was grown in shake flasks on whey medium, harvested, and autolyzed with toluene at 37 C for 20 h under the conditions described above. The autolysate was centrifuged at 20,000 x g for 30 min. The supernatant liquid contained about 300 ONPG units of lactase activity per ml (specific activity 12 ONPG units per mg protein). The enzyme was concentrated and partially purified by slowly adding one volume of cold acetone (4 C) to each volume of solution also at 4 C. The precipitated protein was collected by centrifugation for 10 min at 6000 × g. Residual acetone was evaporated in an air stream, and the precipitate was dissolved in one volume or less of .1 M potassium phosphate buffer, pH 7.0. All buffers used in purification or analysis of the enzyme contained .1 mM MnC12 and .5 mM MgSO4. After centrifugation for 30 min at 30,000 x g, aliquots of the solution were stored at %22, 4, and - 2 0 C to test stability of the lactase.

1623

and weighed. During growth experiments, enzyme analyses were on duplicate samples that were removed from propagation vessels, centrifuged for 10 rain at 5000 × g, and washed twice with cold water. After dilution with cold water to give a concentration of about 100 mg/ml (dry weight), 1 ml aliquots in small glass bottles were quick-frozen in acetone-dry ice, lyophilized, and stored at - 2 0 C in a desiccator. For the comparative enzyme extraction studies, some samples were dried in a forced-flow, hot-air oven for 4 h at 60 C (19) while others were dried with acetone (13). All dried samples were stored at - 2 0 C in a desiccator. Enzyme activity was determined on the dried cells (2 to 20 mg per sample) by incubating them in 100 ml of .1 M potassium phosphate buffer solution, at pH 6.6, containing 15% (wt/vol) lactose at 37 C for periods of 1 to 3 h. Aliquots were removed from the reaction vessel, heated for 10 min in boiling water to stop hydrolysis, and centrifuged. The supernatant liquid was assayed for glucose with the Glucostat Reagent. Following the recommendation of the International Union of Biochemistry, one "lactose unit" is defined as the amount of enzyme which liberates 1 /zmol of glucose per min under these conditions. There was a linear relationship between glucose production and time, and glucose production and weight of cells (Fig. 1). Separate studies in which yeast cells were incubated with glucose [1% (wt/vol) glucose in .1 M potassium phosphate buffer a t p H 6.6] for up to 5 h a t 3 7 C

I

I

16f

B

I

I

A

3O

Analyses

Protein was determined by a modified Folin-Ciocalteau method (20). Lactose was determined polarimetrically (12). Determinations of cell dry weight were made in duplicate on 25 ml samples removed from propagation vessels at various times and centrifuged for 10 rain at 5000 × g. Cells were washed twice with cold water, suspended in sufficient water to transfer to pre-dried, cooled, and tared aluminum pans, dried in a hot-air oven at 100 C for 12 h, cooled in a desiccator,

1

20

I

I

20 60 WEIGHT OF YEAST (rag)

I00

2 REACTION TIME

4 (hours)

6

FIG. 1. Relationships between weight of yeast (A), reaction time (B), and percentage hydrolysis of lactose. Journal of Dairy Science Vol. 58, No. 11

1624

MAHONEY ET AL.

showed that zymase activity was absent. Precision of the method was demonstrated by analysis of 40 identical samples. Hot-air dried samples and lyophilized samples had 2.72 +- .23 and 2.40 + .06 lactose units per mg dry weight. Cell-free extracts of yeast were assayed on lactose as described or on the chromogenic substrate o-nitrophenyl /3-D-galactopyranoside (ONPG). Activity on ONPG was determined b y incubating .1 ml of suitably diluted enzyme with 4 ml of 1.25 mM ONPG containing .1 M potassium phosphate buffer at pH 6.6 and 37 C. After 5 min, the reaction was stopped b y adding 1 ml of .5 M Na2CO3 and the absorbance of the nitrophenolate ion in alkaline solution measured at 420 nm. Under the conditions of the assay, e M of o-nitrophenol was 4.50 × 103 M-~ cm -1, based on absorbance measurements of known concentrations of onitrophenol and on absorbance changes when known concentrations of ONPG were hydro]yzed completely with the enzyme. One "ONPG unit" of enzyme activity is defined as the amount of enzyme which liberates 1 /amol of o-nitrophenol per min under the conditions described. RESULTS A N D D I S C U S S I O N Lactase Formation During Yeast Growth

The best conditions for comparing the biosynthesis of lactase by several strains of K. fragilis were determined by measuring lactase activity of K. fragitis strain UCD #55-61 during growth under different conditions. The pH, lactose concentration, cell mass, and lactase

,

,

,

,

.

.

.

.

.

.

.

.

.

.

.

~12

45

44

a~

m

so

90

3o

so

so

TIME OF GROWTH (hours)

FIG. 3. Growth of K. fragilis strain #55-61 o n whey (15% lactose) plus supplementary nutrients in shake flasks at 30 C.

activity were determined at various times during propagation (Fig. 2 to 4). The cell mass (Fig. 2 to 4) rose rapidly in a logarithmic fashion to the "stationary phase." The actual cell mass, determined by dry weight, paralleled the absorbance readings with two exceptions: (a) there was a decline in cell mass after 60 h when yeast cells were grown in shake flasks, a change not reflected by absorbance (Fig. 2 and 3); and (b) in the fermentor, there was a secondary phase of growth after the initial logarithmic phase whose magnitude was underestimated by chnages in absorbance (Fig. 4). Lactase activity per mg dry weight of cells and total lactase yield rose sharply during the logarithmic phase of growth and reached a maximum at the beginning of the stationary phase. Lactase activity and total lactase yield

z

8R~ 3

4

e 5

~g f ~ 0L

0

' ~ ' ' ~ ";~o

30

6o

9°

30

,

, so

,

,

90

3C

60

'

so

3

12

o

o

~,

TIME OF GROWTH (hours}

TIME OF GRCWTH (hours)

FIG. 2. Growth of K. fragitis strain #55-61 on 10% lactose plus supplementary nutrients in shake flasks at 30C. Journal of Dairy Science Vol. 58, No. 11

FIG. 4. Growth of K. fragilis strain #55-61 o n whey (15% lactose) plus supplementary nutrients in a fermentor at 30 C with an oxygen absorption rate of 2.75 retool/liter per minute.

LACTASE FROM K. FRAGILIS decreased shortly after the beginning of the stationary phase. In shake flasks, activity and yield declined almost to zero after 70 h (Fig. 2 and 3), but in the fermentor the decline was less marked (Fig. 4). There were marked differences in both cell mass and lactase activity per mg dry weight of cells under the three experimental conditions. As a result, there were large differences in the overall yield of lactase. Yeast grown on lactose media in shake flasks produced only 9 lactose units/ml (Fig. 2) while on whey the yield was 15 units/ml in shake flasks (Fig. 3) and 42 units/ml in the fermentor (Fig. 4). Commercial production of the enzyme would require propagation in fermentors, but the rapid decrease in both activity and total yield immediately after reaching the stationary phase would make control of harvest time imperative. Lactose utilization was most rapid during the logarithmic phase of growth. In shake flasks, the lactose concentration continued to decrease after the cell mass had stabilized whereas in the fermentors lactose was exhausted by the end of the first phase of growth. The decrease in pH during the logarithmic phase of growth was due to uptake of NH4 + and metabolite production. The decrease of pH was much smaller in the whey- than in lactosecontaining medium because of greater buffer capacity of the whey. The rise in pH of the medium in the fermentor after 25 h was concurrent with the secondary phase of growth.

Effect of Lactose Concentration on Lactase Yield

Effect of lactose concentration on biosynthesis of lactase by K. fragilis strain #55-61 is in Fig. 5A. Lactase activity per mg yeast increased slowly as lactose concentration increased. Total yield of lactase increased sharply as the lactose concentration was raised from low levels because of a sharp increase in cell mass. Thereafter, total lactase yield was approximately proportional to lactase activity per mg yeast. Optimum lactose concentration for biosynthesis of lactase under these conditions was about 15%, in agreement with results of Wendorff et al. (23). Effect of Aeration Rate on Lactase Yield

Effect of aeration rate on production of lactase is in Fig. 5B. Low aeration rates (

0.6

>, 0.4 "o E

0.2

(/) LU .~-

m

g __.o