Infl uence of the Probiotic Lactobacillus acidophilus NCFM and Lactobacillus rhamnosus HN 001 on Proteolysis Patterns of Edam Cheese

Probiotics are selected cultures of bacteria and yeast, mostly lactic acid bacteria of the genus Lactobacillus, which benefi t the health of the consumer. Probiotics deliver a health-enhancing eff ect by limiting lactose intolerance, inhibiting the growth of pathogenic microfl ora, producing vitamins, reducing cholesterol levels and stimulating intestinal immunity (1). In line with FAO/WHO guidelines (2), before a bacterial strain is classifi ed as a probiotic, its health benefi ts have to be demonstrated in a series of clinical trials. It is recommended that a product with a high probiotic count (approx. 6–7 log CFU/g) be consumed in quantities higher than 100 g/mL daily (3). However, daily consumption of 100 g of cheese cannot be expected due to its high fat content, which is why producers are encouraged to increase the probiotic content of cheese to approx. 8–9 log CFU/g. Daily consumption of 20–25 g of cheese high in probiotics would deliver positive health eff ects (4). ISSN 1330-9862 original scientifi c paper


Introduction
Probiotics are selected cultures of bacteria and yeast, mostly lactic acid bacteria of the genus Lactobacillus, which benefi t the health of the consumer.Probiotics deliver a health-enhancing eff ect by limiting lactose intolerance, inhibiting the growth of pathogenic microfl ora, producing vitamins, reducing cholesterol levels and stimulating intestinal immunity (1).In line with FAO/WHO guidelines (2), before a bacterial strain is classifi ed as a probiotic, its health benefi ts have to be demonstrated in a series of clinical trials.It is recommended that a product with a high probiotic count (approx.6-7 log CFU/g) be consumed in quantities higher than 100 g/mL daily (3).However, daily consumption of 100 g of cheese cannot be expected due to its high fat content, which is why producers are encouraged to increase the probiotic content of cheese to approx.8-9 log CFU/g.Daily consumption of 20-25 g of cheese high in probiotics would deliver positive health eff ects (4).
Probiotic eff ects of a product are determined by the biochemical and physiological properties of probiotic bacteria as well as the applied delivery medium.To maximize their viability, probiotic bacteria should be supplied with products whose chemical composition protects and increases the abundance of benefi cial cultures.Fermented milk products are a traditional carrier of probiotic strains which are also found in other dairy products, including fermented and non-fermented frozen dairy desserts (5), powdered milk (6), fresh cheese (7) and ripened cheese (8,9).
Ripened cheese is increasingly regarded as a bio-matrix for probiotics, which remain viable over prolonged periods of time.In comparison with fermented milk, cheese is characterized by a higher fat content, a higher buff ering capacity, and a lower water content and activity.Cheese ripening and storage processes off er a more supportive environment for probiotics than fermented dairy products (4,10).
A very important criterion, apart from the high probiotic viability, is also the sensory quality of the product.A new product on the market is required to have a high, constant sensory quality to be accepted by potential consumers.Ripened cheese sensory quality is dependent on the biochemical processes, especially during ripening.One of the most important parameters indicative of a proper technological process is lactic acid fermentation.The acidity of the cheese immediately a er production depends on the growth and activity of mesophilic citrate-fermenting streptococci.Diacetyl, a by-product of fermentation, determines the typical fl avour of Edam cheese, while its taste depends on the extent and depth of proteolysis, peptidolysis and fat content.Carbon dioxide is responsible for the formation of the eye.Although proteolysis range depends on the coagulant activity of the enzyme and to a lesser extent on plasmin and perhaps somatic cell proteinases, which results in the formation of large (water-insoluble) and medium-sized (water-soluble) peptides.Subsequently, the depth of proteolysis activity (the increase in the trichloroacetic acid-soluble nitrogen (TCA-SN) and phosphotungstic acid-soluble nitrogen (PTA-SN) fractions) is determined by microbial enzymes from the starter and non-starter bacteria of the cheese (8,11,12).Starter cultures do not aff ect paracasein degradation signifi cantly during cheese ripening.Mesophilic Lactococcus strains produce intracellular enzymes which are activated only a er bacterial autolysis.Due to enzyme-substrate specifi city, secon d ary microfl ora, mostly Lactobacillus bacteria, are characterized by much higher proteolytic and peptidolytic activities (11)(12)(13).
Autolysis of starter cultures, probiotic cultures and the survival of secondary microfl ora (including technically harmful microfl ora) and bacterial enzymes signifi cantly depend on the water activity (14).Water activity (a w ) is an important parameter aff ecting the rate of microbiological and biochemical processes and is determined mostly by water content (free water) and concentrations of soluble compounds.Similar to sodium chloride, low-molecular--mass substances produced by casein and milk fat hydrolysis also reduce water activity (15).The addition of salt and its penetration into the cheese gradually lowers water activity in the ripening process.A drop in a w is also induced by the presence of nitrogen compounds, the products of α S1 -and β-casein proteolysis and short-chain fa y acids (12,16).The results of previous research indicate that proteolysis and peptidolysis can be controlled by limiting the growth of non-starter Lactobacillus microorganisms and replacing them with probiotic bacteria.This approach results in the production of cheese with various sensory properties (11).
Most publications concerning incorporation of bacteria into cheese focus on their survival during manufacture and viability during ripening or storage, but few studies have considered the eff ect of their incorporation on cheese composition and biochemical changes.Besides, most studies were conducted at laboratory scale, therefore comparing or repeating the results of diff erent studies is sometimes impossible.Production in an automatic cheese forming system (Casomatic, Obram, Olsztyn, Poland) provides more repeatable results.Furthermore, the production at industrial scale provides be er microbiological quality (including viability of probiotic culture) and biochemical parameters.
For this reason, this study was performed at industrial plant (10 000 L).The objective of this study is to determine the viability rates of Lactobacillus acidophilus NCFM and Lactobacillus rhamnosus HN001 in Edam cheese during ten weeks of ripening and the eff ect of probiotics on paracasein decomposition and changes in water activity levels during ripening.

Edam cheese manufacture
The experimental material comprised ripened cheese produced in an industrial plant in Giżycko, Poland.Dutch-type cheese (control and experimental) was produced from 10 000 L (each) of premium class milk, which was thermized at 65 °C for 15 s and then cooled to 4 °C.It was then bactofugated, pasteurized at 72.5 °C for 15 s and standardized to 3.0 % fat content.A erwards, the milk was heated to 32 °C before inoculation with cheese starter culture and probiotic bacteria.The milk was combined with 3 kg of calcium chloride (Ciech, Warsaw, Poland), 110 mL of Anna o colouring agent (Ch.Hansen, Cząstków Mazowiecki, Poland), 500 mL of AFILACT ® (Ch.Hansen), frozen cheese starter culture (CHOOZIT™ Classic 111 (0.06 %) DuPont, Poznan, Poland).Apart from Lactobacillus rhamnosus HN001 (0.028 % by volume) or Lactobacillus acidophilus NCFM (0.028 % by volume) (DuPont), 430 mL of rennet (CHY-MAX ® , Ch. Hansen) were added directly to the batch of the experimental cheese.In experimental cheese, the probiotic was added in an amount not less than 8 log CFU/g a er the brining.The cheese sampels were brined for 24 h at 4 °C.A er brining, the cheese samples were wrapped in Cryovac ® (Sealed Air, Duchnice, Poland) oxygen barrier bags and stored in a cold room under controlled conditions.The cheese was ripened for 10 weeks at 12 °C and relative humidity of 85 %.Each phase of the production process was in agreement with industrial standards and according to Dutch-type cheese technology.

Chemical analysis
Grated cheese samples were analyzed in triplicate to determine their salt content using the Volhard method (17), fat content using the Van Gulik method (18) and moisture content by oven drying at 102 °C (17).The pH of the cheese slurry, prepared by blending 10 g of grated cheese with 10 mL of H 2 O, was measured with a pH meter (Elmetron CP 501, Zabrze, Poland, electrode: Inode, Zabrze, Poland) a er calibration with pH=4.0 and 7.0 buff ers (Merck, Darmstadt, Germany).
Water activity values were determined at 25 °C using a water activity meter (Novasina, Pfäffi kon, Switzerland) a er reaching equilibrium at 20 °C, according to the manufacturer's instructions.The samples were brought to room temperature prior to evaluation.A er 4, 6 and 10 weeks of ripening, cheese samples were analyzed to determine the pH and water activity.The degree of paracasein hydrolysis was investigated by determining the concentrations of total nitrogen (TN) and water-soluble nitrogen (WSN), and the fractions of trichloroactic acid-soluble nitrogen (TCA-SN/TN) and phosphotungstic acid-soluble nitrogen (PTA-SN/TN) in total nitrogen (19).

Enumeration of lactic acid bacteria in Edam cheese
Microbiological analyses were performed a er brining and a er 2, 4, 6, 8 and 10 weeks of ripening.A cheese sample of 10 g was added to 90-mL solution (40 °C) of sodium citrate (20 g/L, POCh, Gliwice, Poland).Samples were homogenized in a stomacher (BagMixer 400W, Interscience, Saint Nom, France) to produce a homogenous emulsion.A er sample homogenization, serial tenfold dilutions were made using the M66 solution (15 g/L, Merck, Darmstadt, Germany) as the solvent.
Total starter culture count was determined in the control and experimental cheese samples on M17 agar (Merck).Samples were incubated aerobically at 30 °C for 48 h.Total non-starter lactic acid bacteria from genus Lactobacillus (NSLAB) were determined on Rogosa agar (Merck).The samples were incubated anaerobic ally at 37 °C for 72 h in the AnaeroGen system (Oxoid, Poznan, Poland).
The isolated strains that met the initial classifi cation criteria for the genus Lactobacillus were subjected to a PCR analysis with the aim of identifying the genus and the spe-cies of L. acidophilus NCFM and L. rhamnosus HN001 bacteria.

Isolation and identifi cation of genomic DNA
In order to confi rm that the bacteri a belong to the genus Lactobacillus and species L. rhamnosus and L. acidophilus, fi ve randomly selected bacterial colonies were collected.Then, the material from the collected colonies was grown in MRS broth at 37 °C for 24 h under anaerobic conditions.Bacterial cells were centrifuged and suspended in 100 µL of 10 mM Tris-HCl buff er (pH=8.0).The cells were treated with lysozyme (10 mg/mL) at 37 °C for 1 h.Next, total genomic DNA was extracted using the Genomic Mini DNA purifi cation kit (A&A Biotechnology, Gdansk, Poland) according to the manufacturer's instructions.The isolated DNA was stored at -80 °C (Kaltis, New Taipei, Taiwan) for further analyses.

Statistical analysis
The results of physicochemical and microbiological analyses were evaluated for normal distribution and homogeneity of variance.The signifi cance of diff erences between means or medians was analyzed by Student's t-test (normal distribution) or the Mann-Whitney U test (abnormal distribution).Interactions between two factors were determined by ANOVA.Selected dependencies were evaluated using correlation analysis.
The results were processed with the STATISTICA v. 10 so ware (StatSo , Krakow, Poland), p<0.05,N=3 (physicochemical and biochemical parameters); p<0.05,N=3 (microbiological analysis, done in duplicate).Production of Edam cheese was repeated twice.All data are presented as mean values±standard error of the mean.

Chemical composition of cheese
The control and experimental cheese samples were characterized by similar water content, from 42.9 to 43.6 % (Table 1).The average protein content of control cheese was 25.83 % and of experimental cheese 26.97 %.The analyzed cheese samples diff ered in their fat and sodium chloride content, but the observed diff erences were not signifi cant (Table 1).The addition of the probiotic culture did not lead to signifi cant (p>0.05)diff erences in the chemical composition of the cheese.

Cheese acidity during ripening
Immediately a er brining, the average pH of control cheese samples was 5.22, and it was comparable to experimental cheese samples (containing L. rhamnosus HN001 and L. acidophilus NCFM), which had a pH=5.20 (Fig. 1).The highest pH growth rate was observed a er two weeks of ripening in all samples.Between the 2nd and the 6th week of ripening the pH gradually increased, and it was always higher in the control samples than in the experimental ones.A minor increase in the pH (∆pH=0.01)was observed in control cheese and the cheese with L. acidophilus NCFM between the 6th and 10th week of ripening.Variations in pH values were signifi cantly (p<0.05)correlated with changes in Lactobacillus counts (R= -0.807) and the content of PTA-SN/TN (R=0.775)fraction produced during proteolysis.

Change of the water activity in Edam cheese
Immediately a er brining, the average water activity (a w ) in all cheese samples was 0.976 (Fig. 2).A slow decrease in a w was reported in control and experimental cheese samples during ten weeks of ripening.The water activity decreased at a faster rate in the cheese with L. acidophilus NCFM (∆a w =0.014) and a slightly slower rate in the cheese with L. rhamnosus HN001 (∆a w =0.011), whereas the slowest decrease was noted in the control cheese (∆a w =0.006).The statistical analysis revealed a strong negative correlation (R=-0.728;p<0.05) between viability of probiotic bacteria and changes of a w .A negative correlation was noted between the changes in a w and PTA-SN/TN (R=-0.824;p<0.05) and TCA-SN/TN (R=-0.785;p<0.05).

Proteolysis and peptidolysis in Edam cheese during ripening
The average WSN/TN fraction in control cheese samples was lower (1.25 %) than in experimental cheese samples with probiotic culture (Fig. 3).The fraction of WSN/ TN in all cheese samples increased progressively a er the  2nd week of ripening, but the increase was more intense a er the 4th week of ripening period.The increase in WSN/TN fraction was higher in probiotic than in the control cheese samples.A er ten weeks of ripening, the highest WSN/TN fraction was in the cheese samples with Lactobacillus rhamnosus HN001 (28.22 %) and L. acidophilus NCFM (26.80 %) and the lowest was observed in the control (19.50 %).
Immediately a er brining, the average TCA-SN/TN fraction in the experimental cheese samples was comparable, which was approx.1.64 %, and was lower than in the control (2.08 %).A er two weeks of ripening, the increase in TCA-SN/TN was low (approx.0.50 %) in all analyzed cheese samples (Fig. 4).A gradual increase in TCA--SN/TN fraction was observed between the 2nd and 10th week of ripening and the TCA-SN/TN fraction of the experimental cheese samples increased at a faster rate.A er ten weeks of ripening, the fastest increase in TCA-SN/TN fraction was found in the cheese with L. rhamnosus HN001 (655 %) and a slightly lower increase (520 %) in the cheese with L. acidophilus NCFM.The lowest increase dynamics in TCA-SN/TN was found in the control cheese samples (300 %).The changes in the TCA-SN/TN values were signifi cantly correlated with the viability of Lactobacillus sp.bacteria (R=-0.717;p<0.05).
The PTA-SN/TN fraction of the analyzed cheese samples directly a er brining ranged from 0.63 to 0.75 %.A slow increase in the PTA-SN/TN fraction was reported in the control and experimental cheese samples during ten weeks of ripening (Fig. 5).The lowest increase of PTA--SN/TN fraction was observed a er two weeks of ripening (control: 0.36 %, L. rhamnosus HN001: 0.23 %, L. acidophilus NCFM: 0.07 %).From the 2nd to the 10th week of ripening, the increase in PTA-SN/TN fraction was faster in the cheese with strains NCFM and HN001, whereas the slowest increase was reported in the control cheese.A er ten weeks of ripening, the highest PTA-SN/TN fraction was observed in the cheese samples with the added strains NCFM (3.48 %) and HN001 (3.13 %), but the lowest was in the control cheese (2.24 %).The changes in PTA-SN/TN values were signifi cantly correlated with the viability of pro biotic counts (R=-0.813;p<0.05).

Viability of lactic acid bacteria in Edam cheese
A er brining, the average Lactococcus sp.counts in experimental cheese samples were determined at 8.64 log CFU/g and were signifi cantly (p<0.05)lower than in control cheese (8.80 log CFU/g) (Fig. 6).A gradual increase of Lactococcus sp. was noted between the 2nd and the 4th week of ripening (control: 0.54 log CFU/g, strain HN001: 0.58 log CFU/g, and strain NCFM: 0.51 log CFU/g).A gradual drop of Lactococcus sp.count was noted between the 6th and the 10th week of ripening.Changes in the pH value during ripening occurred due to the production of lactate and the proliferation of non-starter lactic acid bac- teria (NSLAB).For this reason, changes in Lactobacillus sp.counts were analyzed in this research.The average Lactobacillus sp.counts in the control cheese samples were determined at 3.16 log CFU/g and were signifi cantly (p<0.05)lower than in the experimental cheese samples (approx.8.3 log CFU/g) (Fig. 7).A signifi cant increase (approx.4 log CFU/g) in Lactobacillus sp.count was observed in the control cheese samples a er four weeks of ripening.A signifi cant (approx. 1 log CFU/g) reduction of probiotic bacteria population was observed in the analyzed experimental cheese samples between the 2nd and the 4th week of ripening.The viability of probiotic bacteria remained stable in all experimental cheese samples from the 4th week.A signifi cant (p<0.05)decrease (approx.0.78 log CFU/g) in Lactobacillus sp.count was observed in the control cheese samples a er ten weeks of ripening.

Discussion
The chemical composition of all ripened cheese samples was infl uenced mainly by the production process.The production of cheese with an identical chemical composition is practically impossible.Cheeses produced with the use of traditional methods using the same ingredients diff er in their water content by approx. 1 %, on average.The dry ma er content of cheese is also infl uenced by the production process: acidifi cation, coagulation, dehydration (cu ing the coagulum, cooking, stirring, pressing, salting and other operations that promote gel syneresis), shaping (moulding and pressing) and brining.According to Van den Berg et al. (21), the chemical composition of cheese can be infl uenced by its microbiological quality.The results of the current study confi rm the above observation.The use of the probiotic culture in the production of Edam cheese changes its chemical composition.Cheese samples were characterized by a higher protein and fat content than the control cheese samples, but the diff erences were not signifi cant.In contrast to our research, other authors reported that Lactobacillus sp. had no significant eff ect on the chemical composition of Cheddar (22), semi-hard (23) or so cheese (24).The slightly higher protein content in the experimental cheese samples could be caused by the proteolysis of whey proteins (particularly β-lactoglobulin by lactobacilli) and the incorporation of hydrolyzed forms in the curd structure.
The high survival rates of Lactobacillus sp. were largely determined by the water activity.In the fi rst stage of the   production process, a w reached around 0.99 and it did not aff ect signifi cantly the bacterial populations.At successive production stages, including cheese pressing and salting, water activity decreased, reducing the populations of starter cultures and Lactobacillus.This was refl ected in the correlation between the changes in L. acidophilus NCFM and L. rhamnosus HN001 counts (R= -0.728; p<0.05), depending on the water activity.The analyzed cheese samples were characterized by signifi cantly higher water activity (a w =0.976) than semi-hard cheeses investigated by other authors, who reported a w of 0.932-0.935,while the values noted for Gouda cheese varied in the range of 0.913-0.960(25).Due to a radical improvement in raw material quality and the use of bactofugation, most of the referenced cheese products had higher water content.A comparison of various types of ripened cheese suggests that the moisture is a result of higher levels of water activity.An exception to the above rule was Cheddar cheese due to its high acidity and the application of the dry salting process (26).High levels of a w were a consequence of cheese forming, pressing in large moulds and cu ing.This process shortened the salting stage and enhanced production as well as cost eff ectiveness.Due to the absence of a natural cheese rind, the resulting surface absorbed much more brine, thus increasing the water content of the product, water activity levels and intensifying microbiological and biochemical processes during ripening (8).The growth of culture populations increased the rate of proteolysis and raised the concentrations of low-molecular-mass peptides, which reduce water activity.
Acidity, similarly to the chemical composition of fresh cheese, determined its microbiological quality, i.e. an increase in the amount of secondary bacteria and technologically harmful microorganisms (i.e.Escherichia coli, Clostridium sp. and Enterococcus sp.).Cheese ripening involves various biochemical processes which are determined by the presence of lactic acid bacteria (LAB) and non-starter lactic acid bacteria (NSLAB) cultures.At the initial stage of ripening, an increase of LAB counts was stimulated by substrates, i.e. organic acids (lactic acid, citric acid) and sugar residues (27,28).A er their depletion, lactates became a source of carbon.Most LAB are capable of drawing energy from lactate oxidation, but only in the presence of oxygen (29).The increase in the mesophilic populations of Lactobacillus bacteria was closely related to the presence of lactose and probably glycomacropeptide residues from casein degradation (30).The ripening process is accompanied by the gradual autolysis of starter cultures, which leads to the isolation of cell wall components and nucleic acids -the source of substrates supporting NSLAB growth (11).
Directly a er brining, cheese was characterized by desirable pH values (approx.5.20).Ripening led to a gradual increase in pH values due to secondary fermentation whose metabolism involves lactic acid and calcium lactate, as well as the production of ammonia from amino acid catabolism, and by increasing the water-binding capacity of the curd through the fabrication of new α-carboxylic and α-amino groups produced during hydrolysis of peptide bonds (31).
Variations in pH values were signifi cantly correlated with changes in Lactobacillus counts and similar results were reported by other authors (32).The increase in Lactobacillus sp.populations was observed during ten weeks of ripening.NSLAB cultures consist predominantly of L. plantarum, L. casei and L. brevis (33).The presence of NSLAB in cheese directly a er production is a consequence of the protective properties of milk fat, which counteracts the destructive eff ects of pasteurization (34).The presence of Lactobacillus sp. is also associated with the protective properties of specifi c proteins (DnaJ and GrpE) which are responsible for the tolerance of high pasteurization temperatures (35).Various authors (11,36,37) have determined the presence of NSLAB in ripened cheese as its natural microfl ora and they emphasized the key contribution of NSLAB to the desirable a ributes of the end product.
The L. acidophilus NCFM and L. rhamnosus HN001 probiotic cultures are characterized by high viability during the production stage and ripening.A minor drop in the probiotic population was observed in all experimental Edam cheese samples.According to other authors (9,23,37,38), various Lactobacillus sp.strains, including L. acidophilus, L. paracasei and Bifi dobacterium bifi dus, are characterized by good viability in semi-hard cheese.The presence and diversity of the available substrates (peptides, amino acids, sugars, organic acids and their salts) stimulates the viability of bacterial cultures in cheese.Moreover, probiotic bacteria counts exceeded 7 log CFU/g at the end of ripening period, thus supporting the classifi cation of the produced Edam cheese as potentially probiotic.
The growth of lactic acid bacteria (LAB) and nonstarter lactic acid bacteria (NSLAB) cultures signifi cantly contributed to the degradation of β-casein.High-molecular-mass, hydrophobic peptides were produced during the hydrolysis of β-casein by plasmin and bacterial enzymes (38).The water-soluble nitrogen (WSN) fraction is a very heterogeneous complex, which includes whey proteins, high-molecular-, medium-molecular-and low-molecular-mass peptides and free amino acids.High (approx.5 %) content of WSN/TN in cheese a er brining and slightly increased content during ripening are a consequence of the action of residual coagulant, milk proteinases, somatic cell proteinases and cell envelope proteases from the cheese microfl ora.Additionally, the higher WSN/ TN fraction could be a result of higher plasmin activity in cheese a er the removal of plasminogen inhibitors of plasmin activators from curd during the whey removal (39).Moreover, in the milk used for the production of all experimental cheese samples the somatic cell count was high: approx.650 000 (Aljewicz, unpublished data).Generally, plasmin dominates proteolysis in milk with a low amount of somatic cells, with a minor contribution from other enzymes.However, when the amount of somatic cells increases in milk, the relative activity of other enzymes is higher than the plasmin activity (40).Marino et al. (41) showed that adding somatic cells from mastitic milk to cow's milk resulted in much faster hydrolysis of α S1 -casein in experimental cheese compared to control cheese, indicating accelerated proteolysis of this casein.The higher WSN/TN fraction could be a consequence of high acidity and moisture values, and these conditions are ideal for the activity of chymosin on α S1 -casein (42).The increase of WSN/TN fraction during the later stage of ripening is connected with bacterial proteinases, which are extensively released via lysis of their source microfl ora.
In this study, we have found that the inclusion of L. acidophilus NCFM and L. rhamnosus HN001 changed the peptidolysis pa ern during Edam cheese ripening.TCA--SN/TN (peptides from 2 to 22 amino acids, and amino acid residues <3000 Da) fraction in control and probiotic cheese samples during ripening was signifi cantly aff ected by factors such as time of ripening and population of non--starter lactobacilli in control cheese and adjunct bacteria in experimental cheese samples.TCA-SN/TN fraction increased (p<0.05)progressively during ripening, especially in the cheese samples with L. rhamnosus HN001.The coagulation (i.e.rennet) and bacterial enzymes (proteinases and peptidases) are responsible for the formation of some of the non-protein compounds which were precipitated with 12 % TCA (43).
The fraction of PTA-SN/TN (di-and tripeptides and amino acids) was considerably higher in the experimental cheese than in the control cheese samples throughout ripening, suggesting that the peptidases of probiotic culture origin contribute signifi cantly to the release of amino acids during ripening.The above can be a ributed to high activity of cell wall proteinases and peptidases (aminopeptidases and dipeptidases) synthesized by L. acidophilus (9,11).In addition to aminopeptidases synthesized by mesophilic LAB, carboxypeptidases and tripeptidases were produced by various Lactobacillus strains (11,44).Starter cultures had a marginal infl uence on the proteolysis and peptidolysis during ripening because intracellular aminopeptidases were released only a er cell autolysis.This confi rms the stronger negative correlations between the viability of Lactobacillus sp. and changes in the TCA-SN/ TN (R= -0.813; p<0.05) and PTA-SN/TN (R= -0.717; p<0.05) fractions than the correlations between the viability of Lactococcus sp. and changes in TCA-SN/TN (R= -0.225; p>0.05) and PTA-SN/TN (R= -0.111; p<0.05) fractions.The stimulating eff ect of probiotic bacteria on the abundance of TCA-SN/TN and PTA-SN/TN has been demonstrated by various authors (9,12,45).In contrast to our fi ndings regarding Edam cheese, other authors (33,46) reported that Lactobacillus sp. had no signifi cant eff ect on the increase in the level of water-soluble nitrogen in Cheddar cheese.

Conclusions
The use of Lactobacillus rhamnosus HN001 and Lactobacillus acidophilus NCFM in ripened cheese production led to minor changes in its chemical composition, but the changes were not signifi cant.The cheese samples containing L. rhamnosus HN001 or L. acidophilus NCFM were characterized by lower acidity than the control cheese samples (without probiotic cultures).The populations of probiotic cultures used in the production of ripened Edam cheese decreased by around 1 log cycle.A er ten weeks of ripening, probiotic culture counts reached approx.7 log CFU/g.The consumption of 80-100 g of the obtained cheese per day could deliver health benefi ts.A signifi cant correlation was determined between the increase in probiotic culture counts and the drop in water activity levels.The use of L. acidophilus NCFM and L. rhamnosus HN001 probiotic cultures in the production of Edam cheese changed the proteolysis pa ern by intensifying proteolysis and peptidolysis.

Table 1 .
Composition of control and experimental Edam cheese samples aAnalyses were conducted a er brining of Edam cheese samples.Results are expressed as mean value±standard error of the mean.The values with the same le er in superscript do not diff er significantly (p<0.05) ; FDM=fat content in dry ma er, SDM=salt content in dry ma er