Cheese ripening is the term used to describe the process whereby changes occur in the cheese, resulting in the development of flavour, texture and aroma in the finished product. In commercial cheese production, ripening is initiated by the addition of a bacterial starter culture and rennet to milk. The starter culture bacteria convert lactose into lactic acid, producing an acidic environment in which biochemical reactions occur that are critical for cheese ripening (Scott, R.,'Cheesemaking practice', Chapter 11, pages 145-146,1986, Publ. Elsevier). Furthermore, enzymes released from the bacteria are also involved in the degradation of proteins into peptides and amino acids, and the breakdown of fatty acids into keto acids, ketones and esters by lipolysis. These breakdown products are important in the development of flavour, aroma and texture (Aston et al., Aust. Journal Dairy Technology, 38: 55,1983). In all, the ripening process can take up to a year, or more.

Modem cheese-making techniques employ a number of different methods in order to speed up the ripening process, so that the resulting cheese requires less storage time before maturation is complete and is available for sale more quickly. These methods include the use of elevated temperatures, modified cells, adjunct cultures and the use of cheese slurries.

One further method involves the addition of exogenous enzymes, such as purified proteases, to the developing cheese. Protease enzymes break down proteins to shorter chain peptides, which are further broken down to form compounds which impart flavour to the cheese. However, the addition of such enzymes can be problematical.

Proteases act upon casein, the main milk protein, to produce a number of different peptides. Some of these peptides. especially those derived from the breakdown of parts of-casein, can impart a bitter flavour to the cheese. Many proteases can produce these bitter tasting peptides, and this effect is enhanced if the protease enzymes are present at high levels. In addition, problems can arise from the use of rennet. which is added to milk as a coagulant. Rennet is a protease mixture, and residual rennet in the curd is thought to catalyse initial proteolysis ('Cheese, Chemistry. Physics and Biology'.

Second Edition, 1993, Publ. Chapman & Hall. Chapter 7, page 262).

The problem with protease action is not confined to exogenously added enzymes. Starter cultures bacteria possess their own intracellular protease enzymes. which can also lead to the development of a bitter cheese flavour. even in the absence of purified proteases.

Some attempts have been made to tackle the problem of bitterness per se. by using aminopeptidase enzymes. These enzymes counteract the action of proteases. by acting on the small peptide fragments which cause the bitter flavours. Aminopeptidases may be added as purified enzymes, but these are expensive. Alternatively, it has been suggested that starter strains with high inherent levels of aminopeptidase activity can be used to minimise the adverse effects of protease activity in rennet (Powell, I.. Australian Dairy Foods. 1997. October, p. 42-43). Current cheese-making practice. therefore, is focused on reducing the overall effect of proteases, such as those found in coagulants.

However, this practice takes no consideration of the need for accelerated ripening.

In addition to the problems with bitterness, the use of exogenous proteases in accelerated ripening has not been commercially successful. as it appears that primary proteolysis is not the rate limiting reaction in flavour development. Uniform incorporation of exogenous enzymes into the curd is also problematic (Fox et al., Antonie van Leeuwenhoek 70: 271-297,1996).

There still remains, therefore. a need to identify a method for accelerated ripening of cheese using proteases which overcomes the above problems.

We have now, surprisingly, discovered a method for accelerating cheese ripening using an exogenous protease, without any undesirable flavour impairment. using an exogenous protease in combination with a starter culture that has a high level of intracellular, aminopeptidase activity.

Thus, in a first aspect, the present invention provides a method for accelerating cheese ripening, the method comprising the steps of : a selecting a starter culture strain having a significant amount of aminopeptidase activity; b treating milk with the selected starter culture: and c adding an exogenous protease enzyme to the milk and/or curd.

It will be appreciated that steps b and c may be performed in either order or together.

The term'starter culture'relates to any bacterial culture that is suitable for use in cheese ripening, such as Bifidobacteria. Brevibacteria. Lactobacilli. Lactococci.

Leuconostocs, Micrococci and Pediococci. We prefer that the culture is a member of the lactic acid bacteria. It will be appreciated that the term starter culture may encompass a culture containing a single strain of bacterium, or more than one bacterial strain. The term'starter culture'may also include genetically modified organisms (GMO's). In any event, the term,'starter culture'is well known in the art. and the invention extends equally to all known starter cultures.

The present invention relates to starter culture strains that have a high aminopeptidase activity. The term. 'aminopeptidase', as used herein. refers to any enzyme capable of breaking down peptide fragments into shorter amino acid chains or single amino acids. We prefer that the aminopeptidase enzyme is capable of producing a de-bittering effect. We further prefer that the aminopeptidase enzymes with high activity are PepN and/or PepXP, with high PepN activity particularly preferred.

However, it will be appreciated that the present invention applies equally to any aminopeptidase enzyme and an suitable culture producing such. Other suitable enzymes will be readily apparent to the person skilled in the art.

The term'significant', in relation to activity of the aminopeptidase enzyme. indicates that the enzyme must be capable of catalysing the breakdown of the exogenous protease. Accordingly, a high activity is preferred. specific for each aminopeptidase enzyme tested. The term'high activity', with respect to PepN. preferably refers to levels of PepN activity greater than 3 units/g dry weight of starter culture strain. The definition of one unit of aminopeptidase activity is the amount of enzyme required to liberate p-nitroaniline at a rate of 1 umole per minute from a 0.7mM solution of L-leucine p-nitroanilide (Sigma) at pH 7 and 30°C. the reaction being carried out in 100 mM phosphate buffer. We prefer that PepN activity is greater than 5 U/g dry weight of starter culture strain, with levels of 6 U/g dry weight of starter culture strain particularly preferred.

In addition to. or separate from, high levels of PepN activity. we prefer that levels of PepXP activity are greater than 20U/g dry weight of starter culture strain, more preferably greater than 24U/g dry weight of starter culture strain, with activity greater than 35U/g dry weight of starter culture strain most preferred. Starter culture bacteria which have high levels of PepXP activity in conjunction with high PepN activity allows the production of cheese in the process of the present invention without associated bitterness.

Many strains of starter culture contain both PepN and PepXP enzymes.

However, the present invention may be carried out equally successfully with strains that contain only one aminopeptidase enzyme. Such strains may occur naturally or may be genetically engineered, where appropriate.

Starter culture strains having high aminopeptidase activity may be selected from known starters or may be produced by genetic manipulation. The selection of strains from currently available cultures involves the testing of levels of. for example. PepN activity, involving standard enzyme assays which are well known to the person skilled in the art, such as those detailed in Example 1.

The production of a modified bacterial strain which expresses high levels of aminopeptidase activity is also a standard procedure. The aminopeptidase gene may be cloned into a plasmid vector. transformed into the bacterial cell and expressed from an inducible promoter within the cell, for example. Such techniques of genetic manipulation are standard in the art [see, for example. Sambrook et al..'Molecular cloning, A laboratory Manual', second edition. Cold Spring Harbour Press 1989].

It will be appreciated that the identification and use of a starter culture with high inherent aminopeptidase activity allows the use of an increased level of external protease enzyme without the need for exogenous aminopeptidase addition. This has advantages with respect to cost and the ease of cheese production.

The starter culture may be used with any suitable protease enzyme. We prefer that the protease used is a neutral protease. It will be appreciated that the exact selection of protease is not an essential feature of the invention, and it is the combination of protease with a high level of aminopeptidase enzyme that provides the accelerated ripening without bitterness.

The term'exogenous', as used herein, indicates that the protease is provided separately from those proteases found within milk. such as those found within starter culture bacteria, for example. Exogenous proteases may be provided as purified enzymes, or as partially purified mixtures, and may be derived from any suitable source.

Suitable sources of protease will be readily apparent to the person skilled in the art.

The phrase,'treatment of milk', in its simplest form, refers to the addition of the starter culture bacteria to milk. The starter culture is preferably added to the milk prior to the addition of coagulant, such as rennet. In this respect, the production of cheese using the method of the invention may be performed in a conventional manner. The chosen protease for use in the invention may be added directly to the milk. or at the time of salting. As such, it will be appreciated that the exact details of the cheese-making process are not essential to the present invention. The general cheese making processes used to make the cheese are normal, and well known to the person skilled in the art.

The present invention will now be illustrated by way of the following Examples. which are illustrative of the present invention, but not limiting upon it.

EXAMPLE 1 Screening of Starter Culture Strains A range of lactic acid bacteria and cheese starters of the 0-type (Lactococcus lactis subsp. cremoris and Lactococcus lactis subsp. lactis) were screened to identify the levels of PepN activity (one unit of aminopeptidase activity is defined as the amount of enzyme required to liberate p-nitroaniline at a rate of 1 pmole per minute from a 0.7mM solution of L-leucine p-nitroanilide (Sigma) at pH 7 and 30°C. the reaction being carried out in 100 mM phosphate buffer) and PepXP activity (one unit of dipeptidyl hydrolase activity is defined as the amount of enzyme required to liberate p- nitroaniline at a rate of 1 umole per minute from a 1.2 mM solution of glycine-proline- p-nitroanilide (Sigma) at pH 7 and 30°Ct the reaction being carried out in 50 mM phosphate buffer).

The rate of liberation of p-nitroaniline may be determined from the rate of increase in absorbance of the solution at 410 nm, and the extinction coefficient of p-nitroaniline (Aldrich Chemical Co.) dissolved in the same buffer under the same conditions.

The results are given in Table 1.

TABLE 1 Identity PepN PepXP U/gdrywtU/gdrywt L. lactis subsp. cr/McK335 NCDO 924 Z../ocMsubsp./ac5 NCDO 712 Chr. Hansen604363 AJ059 2160 3763 4322 B1675 B2841 jl624 C2449 C449 C4690 D1367 D2321 Lb. helveticus CNRZ32 38 51 EXAMPLE 2 Screening of selected strains and cultures in a model cheese system A number of the strains tested in Example 1 were used in the production of cheese. Cheese was made in 10 L cheese vats using a standard cheddar cheese make.

Resultant curd was treated with salt and then used in aliquots in a curd slurry system (moisture approximately 60%). Samples were incubated at 30°C for 6 days in the presence or absence of neutral protease (0.018 Anson Units per Kg of curd).

The extent of protein breakdown in cheese samples was assessed by measuring the amount of Amino Nitrogen soluble in a given organic acid, in this case 5- SulphoSalicylic Acid. Free Nitrogen was detected colourimetrically using 2,4,6 trinitrobenzenesulphonic Acid (TNBS). Samples were also assessed by a trained sensory panel for flavour evaluation. The panel scored the slurries for cheese flavour intensity, cheddar flavour intensity, savour flavour intensity and bitterness. Protein breakdown levels are used as an indicator of age advancement and when taken in context with sensory analysis provides key findings regarding the cheese. The results are presented in Table 2.

TABLE 2 Strain PepN PepXP SSA'Cheese"CheddarBittedSavoury I NCD09243353"2"2"21 +Neutral Protease 9 2 3 3 1 CH604 3 63 6 2 2 +Neutral Protease 13 3 332 +CRNZ32 38 51 7 3 1 1 1 +NeutralProtease13I1 +CRNZ32 Al9593"2'"2*11 +Neutral Protease 10 3 3 1 1 A3 7 63 3 2 2 1 1 +NeutralProtease93312 B2 8 41 2 3 2 1 1 +Neutral Protease 14 3 3 1 1 Cl24"3"22'21 +Neutral Protease 11 2 2 1 1 C4690'42"211 +NeutralProtease10 3 311 D13"67 2 2 2 2 1 +Neutral Protease 13 2 2 3 1 (1) SSA. soluble Nitrogen: mg g'Cheese Solids (2) Cheese/Cheddar/Savoury flavour intensity Scale 1 = low, 4 = high (3) Bitterness intensity 1 = absent, 4 = very bitter The results indicate that strains with low PepN activities (less than 6 U/g) give low flavour intensity slurries, with a tendency to give bitter off-flavours. Strains having high Pep XP (>24 U/g) and PepN activities equal or greater than 6 U/g gave better flavour profiles and showed no bitterness.

EXAMPLE 3 Cheese production with selected starter cultures Cheese production was scaled up to 500L, using a standard cheddar cheese make procedure. Two of the cultures (CH604 and C4) were selected. CH604 has low PepN and PepXP activities (3 and 21U/g respectively.) C4 has a PepN activity of 6 U/g and a Pep XP activity of 90 U/g. Cheese curd was salted with and without the addition of neutral protease (0.018 Anson U/KG) and the cheese stored at 10°C and periodically analysed for protein breakdown and evaluated by a trained sensory panel. The results of the 12 week evaluations are presented in Table 3.

TABLE 3 Strain PepN PepXP SSA Cheese Cheddar Maturity Bitter Savoury Culture CH604 3 63 5 2.5 2.1 1. 7 1.5 1.1 +NP8 2.2.2.2. 4 2. 8 2.5 1 C4 6 90 6 2.3 2.1 2. 5 1 1 +NP9 3.1 2.8 4. 0 1 1.2 (1) SSA. soluble Nitrogen: mg g 1 Cheese Solids: (2) Cheese/Cheddar/Savoury flavour intensity: Scale 1-4 low-high (3) Maturity index: 1 (mild) 2 (mild/medium) 3 (medium) 4 (medium/mature) 5 (mature) 6 (extra mature) (4) Bitter intensity: 1 absent to 4 very bitter The results in Table 3 show similar trends to those observed in the smaller scale model system; low PepN (low PepXP) starters are prone to bitter off-flavours, high PepN and PepXP cultures used in conjunction with neutral protease accelerated the ripening without producing bitter off-flavours.