Enzyme preparations yielding a clean taste

Abstract

The present invention describes a intracellular produced lactase, which comprises less than 40 units arylsulfatase activity per NLU of lactase activity. The invention also provides a process comprising treating a substrate with an enzyme preparation, wherein the enzyme preparation is substantially free from arylsulfatase.

Claims

1. A process for producing a lactase comprising (a) producing an arylsulfatase deficient yeast or fungal host cell using mutagenesis or recombinant genetic engineering to decrease the expression of a native arylsulfatase expression in the yeast or fungal host cell, wherein the arylsulfatase deficient host cell expresses less arylsulfatase than a parent yeast or fungal host cell not subjected to the mutagenesis or recombinant genetic engineering when cultured under the same conditions; (b) isolating the arylsulfatase deficient yeast or fungal host cell: (c) cultivating the arylsulfatase deficient yeast or fungal host cell in a nutrient medium under conditions resulting in expression of a lactase polypeptide: and (d) recovering the lactase polypeptide from the nutrient medium or from the yeast or fungal host cell.

2. The process according to claim 1, wherein the mutagenesis is random mutagenesis.

3. The process according to claim 2, wherein the random mutagenesis is physical or chemical mutagenesis.

4. The process according to claim 1, wherein the recombinant genetic engineering is one-step gene disruption, marker insertion, site-directed mutagenesis or deletion.

5. The process according to claim 1, wherein the yeast or fungal host cell is transformed with a nucleotide encoding a lactase following step (a) or step (b).

6. The process according to claim 1, wherein the yeast host cell is Kluyveromyces lactis.

7. The process according to claim 1, wherein the fungal host cell is Aspergillus niger or Aspergillus oryzae.

Description

LEGEND TO THE FIGURES

(1) FIG. 1: Cloning of the 5′-flank of the K. lactis arylsulfatase gene in TOPO vector

(2) FIG. 2: Cloning of the 3′-flank of the K. lactis arylsulfatase gene in TOPO vector

(3) FIG. 3: Cloning of the 3′-flank of the K. lactis arylsulfatase gene, lacking the SacII site, in TOPO vector

(4) FIG. 4: Combining the 5′-flank and the amdS selection cassette in one plasmid

(5) FIG. 5: Combining the 5′-flank, 3′-flank and the amdS selection cassette in one plasmid

(6) FIG. 6: Final construction of the arylsulfatase knockout construct

(7) FIG. 7: shows the endoprotease profile using Dabcyl-Edans as substrate.

MATERIALS & METHODS

(8) Activity assay arylsulfatase: Arylsulfatase activity was determined using p-nitrophenylsulfate (obtained from Sigma) as a substrate. For activity measurements, 0.5 ml of substrate solution (20 mM p-nitrophenylsulfate in 100 mM NaP.sub.i buffer pH6.5) was mixed with 0.5 ml sample solution containing the arylsulfatase activity. The solution was incubate at 37° C. for 3 hours. Than the reaction was stopped by addition of 1.5 ml 0.5M NaOH. The OD at 410 nm was determined (1 cm pathlength) against a blank experiment in which water was added instead of sample solution. As reference, a solution was prepared in which the enzyme was added after the reaction was stopped with NaOH. The OD.sub.410 of this reference solution was subtracted from the OD.sub.410 determined for the solution in which the enzyme was active for three hours. An aryl sulfatase unit (ASU) is expressed as the change in OD.sub.410*10E6/hr. For liquid products, the aryl sulfatase activity can expressed as the change in OD.sub.410*10E6/hr per ml of product. For solid products, the aryl sulfatase activity can expresses as the change in OD.sub.410*10E6/hr per g of product. When the activity of the enzyme of interest is known, the arylsuffatase activity can also be expressed as the as the change in OD.sub.410*10E6/hr per unit of activity of enzyme of interest. Activity assay acid lactase: Acid lactase is incubated during 15 minutes with o-nitrophenyl-beta-D-galactopyranoside (Fluka 73660) at pH 4.5 and 37 degrees C. to generate o-nitrophenol. The incubation is stopped by adding 10% sodium carbonate. The extinction of the o-nitrophenol generated is measured at a wave length of 420 nm and quantifies acid lactase activity. One acid lactase unit (ALU) is the amount of enzyme that under the test conditions generates 1 micromol of o-nitrophenol per minute.

(9) Activity assay proline-specific endoproteases: Overproduction and chromatographic purification of the proline specific endoprotease from Aspergillus niger was accomplished as described in WO 02/45524. The A. niger proline specific endoprotease activity was tested using CBZ-Gly-Pro-pNA (Bachem, Bubendorf, Switzerland) as a substrate at 37° C. in a citrate/disodium phosphate buffer pH 4.6. The reaction products were monitored spectrophotometrically at 405 nM. The increase in absorbance at 405 nm in time is a measure for enzyme activity.

(10) The activity of proline-specific endoproteases with near neutral pH optima is established under exactly the same conditions but in this case the enzyme reaction is carried out at pH 7.0.

(11) The activity of proline-specific dipeptidyl peptidases such as DPP IV is established under conditions specified for proline-specific endoproteases with near neutral pH optima, but in this case Gly-Pro-pNA is used as the substrate.

(12) A Proline Protease Unit (PPU) is defined as the quantity of enzyme that releases 1 μmol of p-nitroanilide per minute under the conditions specified and at a substrate concentration of 0.37 mM.

(13) Activity assay carboxypeptidases: The activity of the A. niger derived carboxypeptidase PepG (“CPG”; Dal Degan et al., Appl. Env. Microbiol. 58 (1992)2144-2152) was established using the synthetic substrate FA-Phe-Ala (Bachem, Bubendorf, Switzerland) as a substrate. Enzymatic hydrolysis of this substrate (1.5 mM FA-Phe-Ala at pH 4.5 and 37 degrees C.) results in a decrease of absorbance which is monitored at a wavelength of 340 nm. One unit (CPGU) is the amount of enzyme needed to decrease the optical density at 340 nm by one absorbency unit per minute under the test conditions.

(14) Activity Assay Amino Peptidases.

(15) The activity of aminopeptidases is established using the synthetic substrate X-pNA in which pNA represents p-nitroanilide and “X” an amino acid residue. Because different aminopeptidases can have different selectivities, the nature of amino acid residue “X” depends on the cleavage preference of the aminopeptidase activity tested. Thus, “X” represents the residue for which the specific aminopeptidase has the highest preference. Because many aminopeptidases show the highest reactivity towards Phe, Phe-pNA represents a preferred substrate. Various X-pNA substrates can be obtained from Bachem (Bubendorf, Switzerland). Enzymatic hydrolysis of this substrate (1.5 mM at pH 6.5 and 37 degrees C.) results in a color development which is monitored at a wavelength of 410 nm. One unit (APU) is the amount of enzyme needed to increase the optical density at 410 nm by one absorbency units per minute under the test conditions.

(16) Activity assay esterases/lipases: Esterases and lipases catalyse the release of free fatty acids from triglycerols. In the present assay glycerol tributyrate is used as the substrate. To establish the esterase/lipase activity, the butyric acid released from tributyrate is titrated with sodium hydroxide to a constant pH of 7.5. Therefore, the amount of sodium hydroxide dosed per time unit in order to keep the pH constant, is directly proportional to the esterase activity of the enzyme sample

(17) The measurement is carried out using a Radiometer pH-stat unit and the following reagents.

(18) Arabic Gum solution:Consecutively dissolve, while gently stirring, 100 g Arabic gum (Sigma) and 500 mg Thymol (ICN) in approximately 800 mL demineralised water in a 1 L volumetric flask. Make up to one litre with water and mix. Centrifuge the solution for 15 minutes at 4000 rpm. The resulting arabic gum solution may be kept in the refrigerator for 2 months but should be prepared at least one day before use.

(19) Sodium hydroxide 0.02 mol/l: quantitatively transfer the contents of an ampoule containing 0.01 mol/L NaOH into a 500 mL volumetric flask with water. Make up to volume with water and mix.

(20) SDS/BSA solution: Dissolve, while gently stirring, 1 g SDS (Merck) and 1 g BSA (fraction V, Sigma) in approximately 40 mL water. Prevent the formation of foam. Make up the volume to 1 litre with water after complete dissolving of the SDS and BSA. Only use a freshly prepared solution.

(21) Substrate emulsion: Weigh 50 g glycerol tributyrate in a 600 mL glass beaker and add 300 mL Arabic gum solution. Prepare an emulsion by stirring 5 minutes at maximum speed with the Ultra Turrax. Adjust the pH to 7.5 with NaOH 0.5 mol/L.

(22) To test esterase/lipase activity of a particular enzyme sample, weigh in approximately 1 g of enzyme sample and dissolve in SDS/BSA solution. This sample solution should have a final enzyme content equivalent to approx. 0.2 to 0.8 NBGE/ml (see further). Keep the sample solution on ice until the start of the measurement.

(23) Carry out the measurement by subsequently transferring the following solutions into the heated reaction vessels: 20 mL substrate emulsion, 5.0 mL water (pre-heated at 40° C.) and allow to pre-heat for 15 minutes, then start the measurement by adding 5.0 mL of control sample or the sample solution and start the VIT 90 esterase program of the Radiometer pH-stat unit.

(24) The esterase/lipase unit (NBGE) is defined as the amount of enzyme that releases 1 pmol free fatty acid from glycerol tributyrate per minute at a temperature of 40° C. and pH 7.5 in the following procedure.

EXAMPLES

Example 1

Identification of Off-Flavour Compounds in UHT-Milk

(25) Maxilact LG5000 (DSM, Netherlands) was added under sterile conditions to semi-skimmed UHT milk (Friesche Vlag, Netherlands) to levels of 10,000 and 40,000 NLU per liter and incubated for 4 days at room temperature. In the reference experiment, no Maxilact was added. Prior to assessment of the samples by a taste-panel, a fresh lactase-hydrolyzed milk sample was prepared by adding 40,000 NLU per litre semi-skimmed milk and incubate for 18 hours at room temperature. Sample analysis was performed at NIZO Food Research (The Netherlands) using the SOIR procedure which is a common procedure at NIZO Food Research and which includes a sensory and chemical analysis. Sensory analysis was performed directly on the prepared samples and aliquots of each milk sample were frozen at −25° C. in small portions for further chemical analysis.

(26) Sensory analysis was performed by a 9-membered trained panel. The reference sample was described as cooked, the other samples were classified as not standard UHT milks. The main attributes that described the off-flavour were chemical, medicinal, urine/unclean and stable/manure.

(27) Volatile compounds were isolated with a simultaneous high vacuum distillation near room temperature, creating a watery extract of the sample. The volatile compounds were subsequently isolated from the watery extract using a dynamic headspace and collected at an absorbant. The isolated compounds were injected into a Gas Chromatograph making use of a thermal desorption and separated on a GC-colomn. The GC-effluent was evaluated by two trained assessors (GC-sniff) and described in odour terms (olfactometry). Duplicated high and low concentrated GC-Sniff analyses were carried out by using two different purge times (30 minutes and 24 hours) during dynamic head space sampling. Subsequently the peaks (compounds) indicated during the olfactometric analysis as corresponding with the off-flavour characteristics of the lactase-treated UHT-samples were identified by mass spectrometry. The compounds of interest that may explain the cause of the off-flavour were identified as 1) esters (ethyl butanoate); 2) sulphur compounds (dimethyl sulfide, dimethyl trisulfide and benzothiazole); 3) sulfur esters (methyl thioacetate, methylthiobutyrate); 4) 1-octen-3-ol; 5) 2-nonenal; 6) 3-damascenone; 7) borneol and 8) p-cresol. The p-cresol could originate from conjugates in milk. The only compound that was associated with the most offensive sensory attribute ‘medicinal’ was p-cresol. The concentration of p-cresol in the samples was determined using GC-analysis by addition of standard quantities of p-cresol to the samples. The concentration of p-cresol in the UHT-milk sample 4 days incubation) was estimated at 12 μg per litre. This is clearly above the flavour treshold of 1 ppb and 2 ppb for air and water respectively (Ha et al, (1991) J Dairy Sci 74, 3267-3274). It also is in the range of p-cresol-concentrations commonly found in cows milk. The results were confirmed by recombination experiments in milk, confirming that p-cresol is responsible for the medicinal off-flavour in lactase-treated UHT-milk.

Example 2

Determination of aryl-sulfatase and β-galactosidase Activity

(28) Arylsulfatase activity was determined using p-nitrophenylsulfate (obtained from Sigma) as a substrate. For activity measurements, 0.5 ml of substrate solution (20 mM p-nitrophenylsulfate in 100 mM NaP.sub.i buffer pH6.5) was mixed with 0.5 ml sample solution containing the arylsulfatase activity. The solution was incubated at 37° C. for 3 hours. Than the reaction was stopped by addition of 1.5 ml 0.5M NaOH. The OD at 410 nm was determined (1 cm pathlength) against a blank experiment in which water was added instead of sample solution. As reference, a solution was prepared in which the enzyme was added after the reaction was stopped with NaOH. The OD.sub.410 of this reference solution was subtracted from the OD.sub.410 determined for the solution in which the enzyme was active for three hours. The sulfatase activity is expressed as the change in OD.sub.410*10E6/hr and per NLU. The lactase activity (NLU) for the sample solution was determined as given below.

(29) Lactase activity was determined as Neutral Lactase Units (NLU) using o-nitrophenyl-β-D-galactopyranoside (ONPG) as the substrate, according to the procedure described in FCC (fourth ed, July 1996, p 801-802: Lactase (neutral) β-galactosidase activity).

Example 3

Addition of Aryl-Sulfatase to UHT-Milk

(30) The off-flavour test in milk was performed with commercially available arylsulfatase (Sigma, Aerobacter aerogenes, type VI; 4.9 mg protein/ml; 3.9 arylsulfatase units as defined by Sigma/mg protein). In the experiment, 50 ml of UHT milk (Campina, The Netherlands) was incubated with 1 ml enzyme solution at 30° C. The development of off-flavour was followed by sniffing the sample. The typical off-flavour smell that was also described in example 1 for the UHT-milk incubated with lactase was clearly noticeable after 2 hours of incubation. The smell was more intense after 17 hours of incubation. Apparently, the aryl-sulfatase generated a similar off-flavour as lactase. Based on the findings, described in example 1, this can be explained by the release of p-cresol from the conjugate p-cresylsulphate in milk. Experiments were also performed in which acid phosphatase (wheat germ, Sigma, 6 phosphatase units as defined by Sigma in 40 ml milk) or glucuronidase (from E. coli, Sigma, 6350 glucuronidase units as defined by Sigma per 40 ml milk) were added instead of arylsulfatase. In these incubations the typical off-flavour did not develop. This suggests that the sulphate conjugates are the most important conjugates for the formation of off-flavour in cows milk, This is consistent with literature findings (Lopez et al (1993) J Agric Food Chem. 41, 446-454). The results do not completely exclude the presence of other off-flavour compounds, which could be generated by glucuronidase or acid phosphatase but apparently these compounds do not reach levels that are higher than the flavour thresholds.

Example 4

Off-flavor Test UHT-Milk: Procedure

(31) Semi skimmed UHT milk (Campina, The Netherlands) was incubated with 20,000 NLU/L milk during 48 hours at 30° C. The lactase was added via a sterile filter under sterile conditions to prevent bacterial infection. The milk was tasted after 48 hours by a trained taste panel and compared with a milk solution that was incubated under identical conditions but without addition of lactase. A reference solution was prepared briefly before tasting by adding 5000 NLU/L milk and incubation for 2 hours at 30° C. This sweet milk was used as the reference solution by the taste panel. The off-flavour is scored by the panel as follows: the blank milk was set as ‘−’. Low off-flavour products containing a noticeable but light off-flavour are given ‘+’, whereas products containing higher amounts of off-flavour are expressed as ‘++’ or ‘+++’. The indication ‘+++’ indicates a high level of off-flavour, perceived as very unpleasant. Terms used to characterize the off-flavour were the same as those described in example 1.

Example 5

Purification of K. lactis Lactase: Removal of Aryl-Sulfatase Activity

(32) Maxilact LX5000 (DSM, Netherlands), a commercially available K. lactis lactase, was diluted 10 times with water and applied to a Q-Sepharose column (Amersham Biosciences), equilibrated in 55 mM KP.sub.i (pH7.0). Loading was continued until lactase activity was detected in the run-through of the column. The column was subsequently washed with 4 column volumes of 55 mM KP.sub.i (pH7.0), followed by elution of lactase with 65 mM KP.sub.i (pH7.0) containing 0.16M NaCl. Fractions were collected and assayed for lactase activity. The lactase containing fractions were pooled, and loaded on a butyl Sepharose column (Amersham Biosciences) equilibrated in 55 mM KP.sub.i (pH7.) containing 1 M NaSO.sub.4. The lactase was applied to the column in presence of 1M NaSO.sub.4 (pH7.0) until lactase was detected in the run-through of the column. The column was washed with 4 column volumes of 55 mM KP.sub.i (pH7.) containing 1 M NaSO.sub.4 Lactase was eluted using a 15 column volumes near gradient from 55 mM KP.sub.i (pH7.0) containing 1 M NaSO.sub.4 to 55 mM KP.sub.i (pH7.0). The elution profile was monitored by UV-detection (280 nm). Fractions were collected and assayed for lactase activity. Lactase containing fractions were pooled, with omission of those fractions that were collected after the lactase peak (OD 280 nm) had decreased to 50% of the maximum peak value. Omission of these fractions is critical to prevent contamination of the lactase preparation with aryl-sulfatase. The elution of lactase partly overlaps with the elution of arylsulfatase. The product was concentrated and desalted by ultrafiltration on a 10 kdalton filter and preserved by addition of glycerol to 50% w/w.

Example 6

Protease Levels in Purified Lactase

(33) Protease activity was determined using a series of substrates with the general formula Glu(EDANS)-Ala-Ala-Xxx-Ala-Ala-Lys(DABCYL). (Xxx: any of the 20 natural amino acids). The substrates were obtained from PEPSCAN (Lelystad, The Netherlands), and are internally quenched fluorescent substrates. When such peptide substrates are cleaved, this results in a fluorescent signal. The appearance of fluorescence therefore signals the presence of endo-protease activity. Endo-protease activity was determined in 96-wells microtiter plates by adding 50 μl enzyme solution to 200 μl solution containing 50 μM of the substrate in 100 mM Tris-Bis (pH 6.7). The reaction mixture was incubated for 10 minutes at 40° C. in a TECAN Genius microtiter plate reader using Magellan4 software. Development of fluorescence was followed in time (excitation filter: 340 nm, emission filter: 492 nm). Protease activity was quantified as the slope of the fluorescence line, expressed as RFU/minute/NLU. (RFU: relative fluorescent units, as given by the Genius equipment). NLU-units of the enzyme sample are determined as given in example 2. FIG. 7 shows the enormous reduction in protease activity when LX5000 is purified over the Q-Sepharose column. The pooled fractions after Q-sepharose (example 4) have a factor of at least 5-10 lower protease activity compared to the starting material (LX5000). Lactase samples in which the RFU/min/NLU is <0.5 for each of the substrates used (see FIG. 1) are defined as preparations that have low levels of protease activity.

Example 7

Comparison of Non-Purified and Purified Lactase

(34) Several lactase preparations were submitted to the off-flavour test which is described in example 4. The lactase preparations differed in aryl-sulfatase content. For each preparation, at least two samples were used; individual samples varied in aryl-sulfatase activity, and activity ranges are indicated in the right column of table 1. The results of the off-flavour test are given in table 1. Clearly, levels of arylsulfatase activity are correlated with off-flavour formation. Low levels of aryl-sulfatase (19 or less, see table 1) do not cause off-flavour formation whereas increasing levels lead to increased off-flavour formation. It is also clear that the lactase preparation after Q-Sepharose still shows off-flavour development, even though the protease levels are low (see example 5).

(35) TABLE-US-00001 TABLE 1 Off flavour development for various lactase preparations. Level of Aryl sulfatase activity off- in preparation flavour delta OD * 10E6/hr Lactase preparation formation per NLU Milk without addition −  0 Lactase after Q-Sepharose + to ++ 100-300 (pooled fractions; example 4) Lactase GODO YNL-2 +  40-120 (GODO, Japan) Lactase after butyl −  <8-19.sup.2 sepharose (pooled fractions; example 4) Lactase containing high aryl- +++ 723 sulfatase.sup.1 .sup.1Fraction with high aryl-sulfatase activity, selected from the Q-Sepharose elution fractions described in example 4. .sup.2the level of 8 arylsulfatase units (as defined in example 2) is the detection limit of the assay. <8 means no arylsulfatase activity was observed.

Example 8

Different Commercial Enzyme Preparations Contain Arylsulfatase Activity

(36) Various enzyme products produced from different sources and recovered by different processing routes were collected and were analysed for arylsulfatase activity using the assay specified in the Materials & Methods section. From the results obtained (see Table 2), it is clear that enzyme preparations obtained from various microorganisms such as Aspergillus oryzae, Kluyveromyces lactis, Rhizomucor miehei, Talaromyces emersonii and Trichoderma harzianum can be seriously contaminated with arylsulfatase activity.

(37) These enzyme preparations can advantageously be purified by the process according to the innvention.

(38) TABLE-US-00002 TABLE 2 Aryl sulfatase activity in various commercial enzyme preparations arylsulfatase activity (in delta OD * 10E6/hr per g or ml of enzyme Enzyme product Supplier batch code Prod. organism preparation) Sumizyme FP Shin- U-E529 A. oryzae 39300*10E3 U/g (microbial proteases) Nihon (chem syst.) Sumizyme LP Shin- S-9906-02 A. oryzae 14950*10E3 U/g (microbial proteases) Nihon Maxilact LG2000 DSM AE0050 K. lactis  283*10E3 U/ml Acid lactase Amano LAFD1050508 A. oryzae 12550*10E3 U/g (20 mg)* 3.3 h reaction Lipase F-AP15 Amano LFB A. oryzae  1230*10E3 U/g (lipases) 1251507 Piccantase A DSM F5583 R. miehei  250*10E3 U/g (microbial (20 mg) esterase/lipase) Filtrase BR-X DSM AF0392 T. emmersonii  513*10E3 U/ml β-glucanase (microbial hemicellulases) Oenozyme Elevage DSM KM616001 T. harzianum  1985*10E3 U/g β-glucanase (microbial hemicellulases)

Example 9

Chromatographic Removal of Arylsulfatase Activity from the Proline-Specific Protease from Aspergillus niger Using Ion Exchange Chromatography

(39) In order to remove the arylsulfatase side activities from the proline-specific endoprotease secreted by A. niger (WO 02/046381), a number of chromatographic resins were screened. Because the isoelectric points of the protease and the main secreted arylsulfatase activities secreted by A. niger were found to be approximately 0.5 pH units apart, the identification of a chromatographic separation that allows an acceptable separation of the two activities, even under large scale, industrial conditions, is quite demanding.

(40) Finally, the cation exchanger SP Sepharose 6FF and the hydrophobic interaction (HIC) resin butyl Sepharose 6FF (Amersham Biosciences Europe) were selected for further tests. Both resins were tested in Tricorn 5/100 columns (CV=2,2 ml) using an ÄKTA Explorer 100 controlled by UNICORN 3.20 and an AKTA Purifier controlled by UNICORN 3.21 in combination with a FRAC-950 fraction collector. After elution all fractions generated were tested for proline-specific endoprotease activity and arylsulfatase activitiy using methods specified in the Materials and Methods section.

(41) TABLE-US-00003 TABLE 3 Conditions under which the SP-Sepharose-6FF chromatography was conducted: Buffer A 20 mM Citrat, 0.085M NaCl, pH3.0 Buffer B 20 mM Citrat, 1.0M NaCl, pH3.0 Start conc. B (%)/Start cond. 0/10.7 (mS/cm) Flow rate (ml/min) 0.48 Sample volume (ml) 0.40 Wash volume (CV) 6.1 Flow through and wash 1.0 and 11.0 fraction sizes (ml) Gradient 0-40% B in 10 CV; 100% for 3 CV Eluate fraction size (ml) 1.0

(42) After pooling of the fractions showing proline-specific activity towards the chromogenic peptide Z-Gly-Pro-pNA (Bachem, Bubendorf, Switzerland), arylsulfatase activities of the crude and chromatographically purified enzyme preparations were compared. It turned out that in preparations showing exactly the same proline-specific activity (9 PPU/ml), the arylsulfatase activity was lowered from 3800*10E3 units/ml in the crude preparation to less than 30*10E3 units/ml in the chromatographically purified preparation.

Example 10

Chromatographic Removal of Arylsulfatase Activity from the Proline-Specific Protease from Aspergillus niger Using Hydrophobic Interaction Chromatography

(43) The HIC chromatography was conducted under the following conditions. A diafiltrate of the A. niger derived proline-specific endoprotease having an activity of 10 PPU/ml was used as the starting material. This diafiltrate was diluted two times with 20 mM citrate buffer containing 2 M Na.sub.2SO.sub.4 (pH 4.2, G=121 mS/cm) and was subsequently sterilized by filtration (0.2 μm) before loading on the column.

(44) TABLE-US-00004 TABLE 4 Conditions under which purification of example 10 is performed. Resin Butyl Sepharose 6 FF Column type XK26 Column volume (ml) 107 Buffer A 20 mM citrate + 1M Na.sub.2SO.sub.4 (pH4.2; G = 94 mS/cm) Buffer B 20 mM citrate + 0.02M Na.sub.2SO.sub.4 (pH4.2; G = 6 mS/cm) Flow rate (ml/min) 15 (or 170 cm/h) Equilibration 0 or 20% buffer B (94 or 82 mS/cm) Sample volume (ml) 76-77 ml (with 1M Na.sub.2SO.sub.4 as end concentration) Wash 20% buffer B (83 mS/cm) for 24 CV Flow through and 38.5 ml and collection of total wash volume or wash fraction sizes (ml) total selection of flow through and wash Elution (step) 100% buffer B for 12 or 15 CV Eluate fraction size (ml) 10 or 50 ml

(45) As the result of a considerable tailing after loading of the enzyme on the column, a long washing procedure was required to obtain baseline separation. Finally, the proline-specific endoprotease could be eluted from the column with buffer B. The fractions containing the proline-specific proteolytic activity were pooled. Though diluted, this purified material showed significantly lowered arylsulfatase activity if calculated back to the original proteolytic activity demonstrating that the proline-specific endoproteolytic activity and the arylsulfatase activity were effectively separated using this hydrophobic interaction chromatography protocol. Also here it turned out that in preparations showing exactly the same proline-specific activity (9 PPU/ml), the arylsulfatase activity was lowered from 3800*10E3 units/ml in the crude preparation to less than 30*10E3 units/ml in the chromatographically purified preparation.

Example 11

The Purified Proline-Specific Endoprotease from A. niger Generates Casein Hydrolysates without Off Odors

(46) To test the performance of the chromatographically purified proline-specific endoprotease, two casein hydrolysates were prepared using a chromatographically purified and a non-chromatographically purified proline-specific endoprotease in exactly the same protocols.

(47) To a solution containing 100 g/L of sodium caseinate (Murray Goldbern, New Zealand) and water, subtilisin (Protex{circumflex over ( )} L; 25 milliliter/gram protein was added and incubated for 4 hours at 60° C. and a pH as is. The precipitate formed slowly dissolved while stirring. At the end, a clarified solution was obtained with a minor precipitate. Then pH of the solution was adjusted to pH 4.5 and the liquid was split into equal volumes. To one of these volumes, 1 PPU of a crude A. niger prolyl endopeptidase per gram of casein hydrolysate was added; to the other volume 1 PPU of a chromatographically purified A. niger prolyl endopeptidase. Incubation was continued for 9 hours at 55° C. followed by a 10 kDa ultrafiltration for both solutions. After a further heat inactivation step (5 seconden 120 degrees C.) and a cooling down period, the taste and the odor of the two liquids were evaluated by a panel of 5 people trained in detecting and ranking off flavours and off odors in milk hydrolysates. The panel was unanimous in their conclusion that the hydrolysate prepared with the crude proline-specific endoprotease had a characteristic, “barn-like” odour and flavor which was missing in the preparation prepared with the chromatographically purified proline-specific endoprotease.

(48) TABLE-US-00005 TABLE 5 overview of results of examples 11 and 13 In substrate In enzyme (during application) preparation AS-activity in AS-activity AS-activity substrate (In Delta (In Delta (in Delta OD*10.sup.6 OD*10.sup.6/hr OD*10.sup.6/hr per liter of per ml) per PPU) substrate) Crude preparation  3.8*10.sup.6  422*10.sup.3   42*10.sup.6 Barn-like “proline-specific” flavour (9 PPU/ml) Purified   30*10.sup.3  3.3*10.sup.3  330*10.sup.3 No preparation barn-like “proline-specific” flavour (9 PPU/ml) Crude preparation 31.2*10.sup.6 15.1*10.sup.3  3.2*10.sup.6 off-flavour “carboxy- peptidase” (2060 CPG/ml) Crude preparation   10*10.sup.3  4.8 960 No off- “carboxy- flavour peptidase” (2060 CPG/ml)

Example 12

Chromatographic Removal of Arylsulfatase Activity from a Carboxypeptidase from Aspergillus niger

(49) Because the isoelectric points of the carboxypeptidase (i.e.p. 4.5) and the main secreted arylsulfatases from A. niger (i.e.p.'s of 5.0 and 5.4) are close together, chromatographic separation of the two enzymesproved to be quite difficult. However, the following procedure allowed us to obtain a pure carboxypeptidase, free from arylsulfatase activity. Most importantly, the method is relatively simple so that it can be carried out on an industrial scale.

(50) Again a SP-Sepharose FF resin was used. The chromatography was conducted under the following conditions:

(51) TABLE-US-00006 TABLE 6 conditions under which chromatography of example 12 is performed Buffer A: 20 mM NaCitrate + 40 mM NaCl pH 3.1 ± 0.1; conductivity 5.5 ± 0.3 mS/cm Buffer B: 20 mM NaCitrate pH 5 ± 0.2; conductivity 3 mS/cm Equilibration: PH 3.1 ± 0.1; cond 5.5 ± 0.3 mS/cm  5 cv Load 3100 U/ml resin  7 cv Washing Buffer A 10 cv Elution buffer Buffer B  4 cv Collected pepG Buffer B 1.3-1.5 cv Caustic cleaning 1M NaOH  2 cv

(52) After the crude enzyme was applied to the column, the column was washed with buffer A to remove unbound/slightly bound contaminations. Finally, the carboxypeptidase is eluted with buffer B. The peak containing activity towards the hydrolysis of the chromogenic peptide FA-Phe-Ala-OH (Bachem, Bubendorf, Switserland) was collected. In carboxypeptidase preparations containing comparable carboxypeptidase activities (2060 CPG/g; see Materials and Methods for the activity determination), the arylsulfatase activity decreases from an initial 31200*10E3 units/ml in the crude enzyme to 10*10E3 units/ml in the purified preparation.

Example 13

The Purified Carboxypeptidase from A. niger Accelerates the Aging of Gouda Cheeses without Generating Off-Flavours

(53) Milk was unstandardised and collected from the NIZO. Gouda cheese was manufactured using the NIZO method. Briefly: After starter addition, stir for 15-20 minutes. Then add rennet, stir for 3 minutes and set (approx. 45-50 minutes). Cut coagulum using the gradual increase dial. This will take 10 minutes and have a final speed of 8.5. Turn the blades and stir for another 10 minutes at speed 11. Drain till 120 L remains in vat. Add 36 L (30% of remaining volume) of water at 55° C. to achieve an end temperature in the vat of 35.5-35.7° C. while stirring at speed 16. Stir for 60 minutes at speed 16. Collect the curd and rest for 15 minutes. Divide the curd over the moulds (weight) and rest the filled moulds for 30 minutes. Press for 30 minutes at 0.7 bar (add the cheese code after first pressing), 30 minutes at 1.2 bar and then another 30 minutes at 1.7 bar. Turn the curd after each. After pressing, the pressure is removed and the cheeses are rested in the moulds (overnight in the brining room at 13° C.).The cheeses are removed from the moulds and entered into the brine for 30 hours and turned twice to ensure uniform brining. Ripening at 13° C., 88% humidity. The method standardizes at 1.05 fat to protein ratio (equivalent to about 0.85 fat to casein Following pasteurisation, the milk was pumped into the 200 L cheese vats. Delvo®-TEC UX21A (1.5U; DSM Food Specialities, Delft, The Netherlands) was used as starter culture and Maxiren® 600 (55 IMCU/L milk; DSM Food Specialities, Delft, The Netherlands) as rennet. The cheeses were brined for 26 hours and ripened at 13° C., 88% RH. Purified and non-purified PepG was added with the rennet at a level of 200 CPGU/liter milk.

(54) Following 6 and 24 weeks of ripening, representative samples of a cheese from each cheese vat were graded using an internal panel. These sessions have taken place in a round-the-place manner which means that the graders are informed of the trial details and afterwards discuss their findings which are then summarised.

(55) TABLE-US-00007 TABLE 7 results of example 13: 6 weeks (n = 7) Control Young Gouda cheese, no off flavours, little bit acid and buttery odour. Purified More mature Gouda cheese, farmhouse type flavours, PepG a strong odour and a fuller flavour. Unpurified Like the purified PepG cheese only bitter notes in the PepG flavour and after taste.

(56) TABLE-US-00008 TABLE 8 results of example 13: 24 weeks (n = 6) Control Mature Gouda cheese, bit salty Purified PepG Cheese with an intense flavour, touch of sweetness Unpurified Very piquant flavoured cheese, farmhouse cheese, PepG not in balance, off flavour

(57) A clear effect was found as a result of the addition of PepG, both purified and unpurified. The bitter notes and the disbalances recorded for the cheeses in which the non-purified was used lead to the conclusion that PepG should be purified.

Examples 14-15

(58) In the examples described hereinbelow, standard molecular cloning techniques such as isolation and purification of nucleic acids, electrophoresis of nucleic acids, enzymatic modification, cleavage and/or amplification of nucleic acids, transformation of E. coli, etc., were performed as described in the literature (Sambrook et al. (2000) “Molecular Cloning: a laboratory manual”, third edition, Cold Spring Harbor Laboratories, Cold Spring Harbor, N.Y., and Innis et al. (eds.) (1990) “PCR protocols, a guide to methods and applications” Academic Press, San Diego).

Example 14

Construction of an Arylsulfatase Knock-Out Strain of Kluyveromyces lactis

(59) Isolation of Kluyveromyces lactis chromosomal DNA:

(60) A 100 ml YEPD (1% Yeast-extract; 1% Bacto-peptone; 2% glucose) shake flask was inoculated with a single colony of K. lactis CBS 2359 and cultivated for 24 hours at 30° C. shaking at 280 rpm. The amount of cells was counted using a counting chamber and an amount of culture corresponding to 4.1*10.sup.8 cells was used. Extraction of chromosomal DNA was performed using the Fast DNA Spin Kit supplied by Q-BIOgene (Cat #6540-600). The yeast protocol was used: one homogenizing step using the Fastprep FP120 homogenizer (B10101 Savant) of 40 seconds at speed setting 6.0 was used.

(61) Subsequently the sample was cooled on ice and subsequently homogenized again using the same conditions.

(62) The purity and yield of the extracted genomic DNA was determined using the Nanodrop ND1000 spectrofotometer. It was found that the concentration of the extract was 114 nanogram/microliter. The A260/280 and A260/230 ratio was found to be respectively 1.57 and 0.77.

(63) PCR amplification of 5′ and 3′ arylsulfatase flanks:

(64) TABLE-US-00009 5′ flank arylsulphatase primers: DFS-15289 (5′.fwdarw.3′): TCG CCG CGG TTG TCA ACT ATA TTA ACT ATG DFS-15290 (5′.fwdarw.3′): GAT AGA TCA TAG AGT AAC AAT TGG 3′ flank arylsulphatase: DFS-15291 (5′.fwdarw.3′): GCA ACT GAA GGT GGT ATC AAT TG DFS-15292 (5′.fwdarw.3′): CAC CCG CGG CAC CAG ATA ATG GAG GTA G 3′ flank SacII arylsulphatase: DFS-15291 (5′.fwdarw.3′): GCA ACT GAA GGT GGT ATC AAT TG DFS-15340 (5′.fwdarw.3′): CGG CAC CAG ATA ATG GAG GT

(65) The arylsulfatase flanks were amplified using Phusion High-Fidelity DNA Polymerase, (Finnzymes, Espoo Finland). The K. lactis CBS 2359 genomic DNA was diluted 100 times with Milli-Q water and 5 μl was used as a template in a 50 μl PCR mix, according to suppliers' instructions. A Hybaid MBS 0.2G PCR block using the following programs:

(66) PCR Program 5′ flank arylsulfatase:

(67) TABLE-US-00010 Stage 1 (1 cycle) 98° C. 30 s Stage 2 (30 cycles) 98° C. 10 s 60° C 30 s 72° C 30 s Stage 3 (1 cycle) 72° C. 10 min  4° C. Hold

(68) PCR Program 3′ flank arylsulphatase:

(69) TABLE-US-00011 Stage 1 (1 cycle) 98° C. 30 s Stage 2 (30 cycles) 98° C. 10 s 72° C. 30 s 72° C. 30 s Stage 3 (1 cycle) 72° C. 10 min  4° C. Hold

(70) PCR Program 3′ flank SacII arylsulphatase:

(71) TABLE-US-00012 Stage 1 (1 cycle) 98° C. 30 s Stage 2 (30 cycles) 98° C. 10 s 65° C. 30 s 72° C. 30 s Stage 3 (1 cycle) 72° C. 10 min  4° C. Hold

(72) Construction of an arylsulphatase knock-out vector:

(73) The obtained 5′-, 3′- and 3′ SacII arylsulfatase flank PCR fragments were cloned into the pCR-Blunt II-TOPO vector using the Zero Blunt TOPO PCR Cloning Kit (Invitrogen; Part. no. 45-0245), according to suppliers' instructions. The TOPO cloning reactions were transformed to One Shot TOP10 Chemically Competent E. coli (Invitrogen; Part. no. 44-0301) according to suppliers' instructions. Correct clones were selected based on restriction pattern analysis using MunI, SacII, XcmI, and DraI; MunI, SacII, EcoRI and EcoRV; MunI, EcoRI, EcoRV and SacII for respectievely, 5′ TOPO, 3′ TOPO and 3′ SacII.sup.− TOPO.

(74) The amdS cassette was isolated from the pKLAC1 vector (New England Biolabs). The pKLAC1 plasmid was transformed to chemically competent dam-/dcm-E. coli cells (New England Biolabs; Cat. No C2925H) and the un-methylated plasmid was isolated. Large plasmid DNA batches of 5′ TOPO, 3′ TOPO, 3′ SacII.sup.− TOPO and pKLAC1 vector were isolated from overnight LBC cultures containing 50 μg/ml Kanamycin using the

(75) GeneElute Plasmid MidiPrep Kit (Sigma; Cat. No. NA0200).

(76) The pKLAC1 vector was digested Sail and XbaI and the 5′ TOPO vector was digested XbaI and Xhol. Digests were purified using the Nucleospin ExtractII Kit (Machery Nagel) according to suppliers' instructions.

(77) The SalI/XbaI digested amdS cassette was ligated into the XbaI/XhoI digested 5′ TOPO vector using the Quick ligation Kit (New England Biolabs; Cat. No. M2200S) according to suppliers' instructions. The ligation mix was transformed to One Shot TOP10 Chemically Competent E. coli (Invitrogen; Part. no. 44-0301) according to suppliers' instructions. A correct clone was selected based on restriction pattern analysis using MunI, EcoRI and SacI. This resulted in the following vector: 5′amdS TOPO vector (FIG. 1) A large batch of the 5′amdS TOPO plasmid was isolated from overnight LBC cultures containing 50 μg/ml Kanamycin using the GeneElute Plasmid MidiPrep Kit (Sigma; Cat. No. NA0200) according to suppliers' instructions. The 5′amdS TOPO vector was digested MunII and AscI and the 3′ SacII.sup.− TOPO vector was digested MunI and EcoRI. The MunI/EcoRI 3′ SacII.sup.− TOPO fragment was isolated and purified by means of gel extraction. An electrophoresis was performed on 1% agarose in TAE buffer containing SYBR Safe DNA Stain (Invitrogen; Cat. No. S33102), according to suppliers' instructions. The fragment was visualized using the dark reader transilluminator (Clare Chemical Research; Cat. No. DR-45M), excised from the gel and extracted from the agarose using the Nucleospin ExtractII kit (Machery Nagel; Cat. No. 740 609.250) according to suppliers' gel extraction protocol.

(78) The 5′amdS TOPO vector was digested MunI and AscI and purified using the Nucleospin ExtractII Kit (Machery Nagel) according to suppliers' PCR purification protocol. Subsequently the MunII/AscI digested 5′amdS TOPO vector was dephosphorylated using Shrimp Alkaline Phosphatase (Roche; Cat. No. 1 758 250) according to suppliers' instructions.

(79) The MunI//EcoRI 3′ SacII.sup.− TOPO fragment was ligated into the dephosphorylated MunI/Ascl digested 5′ amdS TOPO vector using the Quick ligation Kit (New England Biolabs; Cat. No. M2200S) according to suppliers' instructions. The ligation mix was transformed to chemically competent dam-/dcm-E. coli cells, (New England Biolabs; Cat. No C2925H) according to suppliers' instructions. A correct clone was selected based on restriction pattern analysis using EcoRI and EcoRV. This resulted in the following vector: 5′ amdS 3′ SacII-TOPO vector.

(80) A large batch of 5′amdS 3′ SacII TOPO vector was isolated from overnight LBC cultures containing 50 μg/ml Kanamycin using the GeneElute Plasmid MidiPrep Kit (Sigma; Cat. No. NA0200) according to suppliers' instructions.

(81) The 5′amdS 3′SacII vector was digested with XbaI. The 3′ TOPO vector was digested with XbaI and SpeI. Digests were purified using the Nucleospin Extractll Kit (Machery Nagel) according to suppliers' instructions. The XbaI/SpeI 3′ TOPO fragment was isolated by means of gel extraction, as described above. The XbaI digested 5′amdS 3′SacII vector was dephosphorylated using Shrimp Alkaline Phosphatase, Roche (Cat. No. 1 758 250) according to suppliers' instructions. The XbaI/SpeI 3′ fragment was ligated in the dephoshorylated XbaI digested 5′ amdS 3′ SacII vector using the Quick ligation Kit (New England Biolabs; Cat. No. M2200S) according to suppliers' instructions. The ligation mix was transformed to One Shot TOP10 Chemically Competent E. coli (Invitrogen; Part. no. 44-0301). A correct clone was selected based on restriction pattern analysis using MfeI, KpnI, EcoRI, SacII, ScaI. This resulted in the final K. lactis arylsulphatase knock-out vector (FIG. 3).

(82) A large batch of the arylsulphatase knock-out vector was isolated from overnight LBC cultures containing 50 μg/ml Kanamycin using the GeneElute Plasmid MidiPrep Kit (Sigma; Cat. No. NA0200) according to suppliers' instructions. The K. lactis arylsulphatase knock-out vector was digested with SacII so the linear knock-out cassette would be obtained, lacking the TOPO vector part. The digest was purified using the Nucleospin Extractll Kit (Machery Nagel) according to suppliers' instructions.

(83) Transformation of K. lactis CBS 2359 with arylsulphatase knock-out vector A 100 ml YEPD culture of K. lactis CBS2359 was incubated at 30° C., shaking at 280 rpm for 24 hours. This culture was used to inoculate a 100 ml YEPD culture which was grown under the same conditions until an OD610 between 0.5 and 0.8 was reached. Cells were harvested by means of centrifugation for 5 minutes at 1559 g and 4° C. The cell pellet is washed with 50 ml sterile electroporation buffer (EB): 10 mM Tris pH 7.5, 9.2% (w/v) Sucrose, 1 mM MgCl.sub.2 at 4° C. The cell pellet was resuspended in 50 ml YEPD containing 25 mM DTT and 20 mM HEPES buffer pH 8.0 at room temperature. The cells were incubated 30 minutes at 30° c. without shaking. The cells were harvested by means of centrifugation for 5 minutes at 1559 g and 4° C. and washed with 10 ml ice cold EB. The cells were again pelletted by means of centrifugation for 5 minutes at 1559 g and 4° C. and resuspended in 0.1 ml ice cold EB. The cell suspension was distributed in 40 microliter aliquots in 1.5 ml eppendorf tubes. To one aliquot of cells 0.2-1.0 microgram (1-5 microliter) of the linear knock out construct was added, mixed by pipetting and incubated on ice for 15 minutes. The cell-DNA mix was added to a chilled electroporation cuvette with a 2 mm gap size (BTX; Part. No. 45-0125). Electroporation was performed on a BioRad electroporator composed of a Gene Pulser (BioRad, Model No. 1652077) and a Pulse Controler (BioRad, Model No. 1652098) using the following settings: 1000 V, 400 Ohm and 25 μF. Immediately after electroporation 1 ml YEPD was added and the cells were transferred to a sterile 12 ml tube and incubated during 2 hours in an shaking incubator at 30° C. The cells were pelletted for 5 minutes at 1559 g and washed in fysiologic saline solution (0.85% (w/v) sodium chloride). The cells were again pelletted and resuspended in 1 ml fysiologic saline solution. Several aliquots of 25 μl, 50 μl and 100 μl were plated on selective amdS agar plates: 1.25% (w/v) agar, 1.17% (w/v) Yeast Carbon Base, 30 mM phosphate buffer pH 6.8 and 5 mM acetamide. Plates were incubated for 2 days at 30° C. followed by 2 days incubation at room temperature. Colonies were selected and purified by streaking them on YEPD agar plates so single colonies would appear and incubated at 30° C. for 24 hours. These single colonies were tested for targeted integration of the knockout construct using a colony PCR with oligonucleotides targeted against the amdS cassette and downstream of the integrated knock out construct. Colony material was suspended in 20 mM NaOH, 0.2% (w/v) and incubated for 5′ at 98° C. The cell suspension was diluted 2 times with water and 2.5 microliter was used directly as template in a 25 microliter PCR reaction using Phusion High-Fidelity DNA Polymerase (Finnzymes; Espoo Finland; Product code F-5305) according to suppliers' instructions.

(84) TABLE-US-00013 Fw 3′ amdS: GAC AAT TGA TAC CAC CTT CAG TTG Rv downstream: CTG GGA AAT GTG GTG ACT CCA TA

(85) Program Targeting PCR:

(86) TABLE-US-00014 Stage 1 (1 cycle) 98° C. 30 s Stage 2 (30 cycles) 98° C. 10 s 68° C. 30 s 72° C. 30 s Stage 3 (1 cycle) 72° C. 10 min  4° C. Hold

(87) PCR was analysed on 1% agarose gels and targeted transformants that show a clear amplified band were selected. The arylsulfatase knockout strains were named 2359ΔARY1-10, and stored until further use.

Example 15

Detection of Arylsulfatase Activity in 2359ΔARY

(88) Motherstrain CBS 2359 and strain 2359ΔARY were all cultivated in shakeflask in 100 ml YEP+2% galactose for 3 days at 30° C. Biomass was collected by centrifugation for 5 minutes at 1559 g and 4° C. Biomass was washed twice with ice-cold water to remove medium components. Yeast biomass was treated with Yeast Protein Extraction reagent (Y-PER) according the instructions of the manufacturer (Pierce), to extract intracellular enzymes like arylsulfatase. It will be clear to those skilled in the art that other yeast lysis protocols can be used to extract arylsulfatase activity, like mechanical sheering with glasbeads, or enzymatic treatment to dissolve the cell wall with e.g. Zymolyase (see i.e. Glover and Hames, DNA cloning 2—a practical approach, IRL Press 1995). Arylsulfatase was measured in the extract using the method described in Example 2. From this experiment it became clear that while the motherstrain contained an appreciable amount of arylsulfatase activity, no such activity could be detected in the 2359ΔARY strain. When the β-galactosidase (lactase) activity was measured in this extract according to Example 2, no difference in lactase activity could be detected between the wild type strain CBS 2359 and the mutant strain 2359ΔARY, showing that the mutant strain is specifically disturbed in arylsulfatase activity.

(89) The mutant strain can be used to make a lactase preparation at industrial scale, virtually devoid of arylsulfatase activity.