| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Instituto de Lactología Industrial Facultad de Ingeniería Química, Universidad Nacional del Litoral–CONICET Santiago del Estero 2829, S3000AOM Santa Fe, Argentina
1 Corresponding author: ehynes{at}fiqus.unl.edu.ar
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: nonstarter lactobacilli • acidifying activity • proteolytic activity • cheese making
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Nonstarter lactic acid bacteria constitute the only main factor that remains uncontrolled in today’s industrial cheese making, and consequently, they may be the main source of quality inconsistencies and defects in cheese products (Crow et al., 2001). The NSLAB counts can increase from a very low level (10 to 104 cfu/g in 1-d-old Cheddar cheese manufactured with pasteurized milk) up to 107 to 108 cfu/g during the first weeks of ripening, becoming dominant in cheese microbiota after starter decrease (Peterson and Marshall, 1990; Fox et al., 1998). The NSLAB are made up mainly of lactobacilli in cheeses manufactured with pasteurized milk, but may also include other lactic acid bacteria such as pediococci and enterococci, Streptoccocus thermophilus, and Leuconostoc spp. (Corsetti et al., 1998; Swearingen et al., 2001; Sánchez et al., 2006). To date, no suitable method to obtain NSLAB-free cheeses is available: aseptic cheese making is only appropriate for laboratory conditions and is not completely effective in keeping low levels of NSLAB during ripening (Kleter, 1977; McSweeney et al., 1994; Shakeel-Ur-Rehman et al., 2000c). On the other hand, the decrease in ripening temperature has been a successful approach to impair NSLAB growing, but with a concomitant slow down of the biochemical transformations involved in ripening (Shakeel-Ur-Rehman et al., 2000a,b). The alternative of indirectly controlling secondary microflora in cheese by means of the addition of an adjunct culture has been suggested (Crow et al., 2001; Di Cagno et al., 2003). For that purpose, a relatively important amount of research work has been committed to isolate and characterize strains of NSLAB from good quality cheeses. Most available strains have been obtained from Cheddar cheese (Crow et al., 2001; Swearingen et al., 2001; Banks and Williams, 2004), although recent studies are more diverse (De Angelis et al., 2001; Bude-Ugarte et al., 2006).
Overall, NSLAB grow poorly in milk, but they are able to grow in young cheeses (Stanley, 1998). They contribute to proteolysis in cheese mainly via their peptidolytic potential, increasing the amount of small peptides and free AA (Lynch et al., 1999). Some NSLAB have been also reported to possess key enzymes for cheese flavor formation by AA catabolism (Kieronczyk et al., 2003; Thage et al., 2005). The criteria that a nonstarter Lactobacillus strain should meet to become an appropriate adjunct culture include: to reach and maintain high levels of cell density during ripening, cause no defect in the product, and if possible, impact positively on cheese overall quality (Crow et al., 2001; Di Cagno et al., 2003). The influence of a given adjunct culture in different cheese models may differ because its growth and biochemical expression rely on the technology applied and the starter used (Thomas, 1987; Lane et al., 1997; Hynes et al., 2001b). In addition, some biochemical activities may be considered defective in one cheese but desirable in others [e.g., CO2 production, L-lactate isomerization, aldehyde production (Di Cagno et al., 2003)].
The objective of the present study was to assess the technological properties of nonstarter Lactobacillus strains isolated from Argentinean cheeses by means of in vitro and in situ studies. For that purpose, the strains were examined for proteolytic and acidifying activities, phage-resistance and tolerance to NaCl and KCl. After that, 4 lactobacilli were selected and tested as adjunct cultures in soft and semihard cheeses.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
In Vitro Characterization
Milk acidification kinetics, as well as proteolytic (o-phthaldialdehyde spectrophotometric assay, OPA Test, Church et al., 1983) and acidifying activities were determined by inoculation (2% vol/vol) of the strains in sterile, reconstituted (10% wt/vol), commercial, dry skim milk followed by incubation for 24 h at 34°C. The pH values were measured with a model SA 720 pH meter (Orion, Beverly, MA) and plotted against time. Proteolytic activities were expressed as the differences in absorbance at 340 nm (A340) between strain cultures and a control of uninoculated milk. The acidity developed, measured by titration with 0.1 N NaOH, was expressed as percentage of lactic acid.
The fast or slow character of the strains was determined by inoculation (2% vol/vol) of strain cultures in milk, milk supplemented with 1% (wt/vol) glucose or with 0.25% (wt/vol) casein hydrolysate, and milk supplemented with both 1% glucose and 0.25% casein hydrolysate (McKay and Baldwin, 1974; Efstathio and McKay, 1976), followed by the assessment of pH after incubation by 24 h at 34°C.
The fraction of milk cultures soluble in trichloroacetic acid 8% wt/vol obtained for the OPA test was also analyzed by reverse phase HPLC. The HPLC equipment consisted of a quaternary pump, an inline degasser and UV/VIS detector, all Series 200, purchased from Perkin Elmer (Norwalk, CT). An interface module connected to a computer was used for acquisition of chromatographic data with the software Turbochrom (Perkin Elmer). A 220 mm x 4.6 mm Aquapore OD-300 C18, 7 µm –300 Å analytical column was used (Perkin Elmer). Samples were filtered through 0.45-µm membranes (Millex, Millipore, Sao Paulo, Brazil), and 60 µL was injected into the HPLC chromatograph. Detection was performed at 214 nm, and column temperature was maintained at 40°C. The gradient starting from 100% of solvent A (H2O:trifluoroacetic acid (TFA) 1,000:1.1, vol/vol) and 0% of solvent B (acetonitrile: H2O: TFA 600:400:1, vol/vol) was generated 10 min after injection. The proportion of solvent B was increased by 1%/min (80 min), 20%/min (1 min), 0%/min (4 min), and then returned to starting conditions, which took 1 min. These last setting conditions were maintained for 10 min (Hynes et al., 2003).
Lactobacillus casei strains were tested for cross sensitivity to bacteriophages using the Spot and Turbidity Tests, as described by Svensson and Christiansson (1991). The phages were two collection phages: lytic phage 
g J-1 (sensitive strain L. casei ssp. casei ATCC 27139), and lytic phage gPL-1 (sensitive strain Lactobacillus paracasei ssp. paracasei ATCC 27092, and 4 phages isolated in INLAIN from faulty batches of fermented milks in dairy industry: gMLC-A (Capra et al., 2006), gMLC-B, gMLC-C, and gMLC-D.
Salt resistance was studied by inoculating strains in de Man, Rogosa, Sharpe (MRS) broth, which contained 1, 2, or 3% (wt/vol) NaCl; KCl was also assayed in the same conditions, as it has been proposed to replace part of the NaCl in cheeses (Reddy and Marth, 1993). Cultures were incubated at 34°C, and after 24-h cell growth (A560) was measured and compared with a control without salt addition. The results were expressed as the percentage of growth in the presence of NaCl or KCl compared with the control.
In Situ Characterization: Cheese Making Experiments
Four strains were selected according to the results of in vitro tests and assayed individually as adjunct cultures in cheese making experiments at pilot plant scale. Cheese making experiments consisted of 2 trials; in the first one a soft cheese model was chosen (Cremoso Argentino cheese) and in the second trial a semi-hard cheese variety was assayed (Pategrás Argentino). Each trial was single block completely randomized, where cheese making days were subblocks. In both trials, control cheeses did not contain adjunct cultures but only a commercial starter composed of several strains of Strep. thermophilus. Experimental cheeses were made with the same starter plus a culture of the tested strain of Lactobacillus.
For both types of cheese, raw milk obtained from a nearby dairy factory (Milkaut S.A., Franck, Santa Fe, Argentina), was batch pasteurized at 65°C for 20 min, and cooled to 37°C. Calcium chloride (Merck, Darmstadt, Germany) was added to a final concentration of 0.02% (wt/vol) in the milk. After that, milk was divided in 4 aliquots of 40 L each and sent to cheese vats; 2 vats were used for Cremoso cheeses and the other 2 for Pategrás cheeses. A lyophilized culture of Strep. thermophilus (Chr. Hansen Argentina, Quilmes, Argentina) was used as primary starter; it was dispersed in approximately 100 mL of pasteurized milk and maintained for 5 to 10 min at 37°C before addition to cheese milk. After that, adjunct cultures were added in experimental cheeses. All the lactic bacteria (starter and adjunct) were added in a dose high enough to achieve 106 cfu/mL in cheese milk. After 15 min, 1 g of chymosin produced by fermentation of genetically modified Kluyveromyces lactis (Maxiren 150, Gist Brocades, France) was dispersed in 25 mL of distilled water and added into each vat.
In Cremoso cheeses, coagulation took place at 37°C. When the coagulum strengthened to the typical consistency of this enzymatic curd, it was cut to gross cubes of approximately 2 cm3. The curd grains were then gently stirred in the whey to allow light whey drainage, but without ulterior cutting of curd particles. This step was finished after 10 to 15 min, and then the curd was allowed to set in the bottom of the vat. Whey was pumped out and the curd was put in the mold. Before salting, the curd was placed in a hot chamber (40°C) until a pH value within the range of 5.2 to 5.3 was reached. Salting was performed by immersion in brine (20% wt/vol, pH 5.4) at 5°C for 4 h, to stop the starter activity. A 
4-kg cheese was obtained from each vat. The cheeses were vacuum packed in plastic bags and ripened at 5°C (± 0.5°C) for 60 d (Hynes et al., 2001a).
As for Pategrás cheese, the coagulation step was performed at a somewhat lower temperature: 35°C. When the curd reached the appropriated strength, it was cut in successive steps (with manual stirring between steps) until it reached the size of a corn grain, maintaining temperature about 35°C. The cutting-stirring step took 20 min approximately. The curd grains and the whey were then gently stirred and heated at the rate of 0.5°C/min until 45°C then maintained for 10 to 15 min approximately to reduce the moisture content of curd grains. After that, the curd was separated from whey and molded. The two molds obtained from the 2 vats were piled and pressed during 24 h (0.2 to 0.3 kg cm2). Young cheeses were brined in 20% (wt/vol) pH 5.4 brine for 24 h and ripened for 60 d at 12°C (± 0.5°C) and 80% relative humidity (Bergamini et al., 2006).
Gross composition, pH, microbiology, and overall quality of the cheeses were assessed. Gross composition was determined in 3-d-old cheeses, whereas pH was monitored during Cremoso cheese making and during ripening of all cheeses. Curves of pH were not obtained for Pategrás cheese making as the procedure for this cheese variety does not imply a curd acidification (cheddaring) step. Dry matter was analyzed by drying the sample at 105 ± 1°C until constant weight according to IDF standards (IDF, 1982). Fat matter was quantified by a butyrometric method (IDF, 1997) and pH was assessed according to the American Public Health Association standard (Bradley et al., 1993). Protein content was determined by the Kjeldahl method according to IDF standards (IDF, 1993). For microbial counts, five to ten 10-g cylinders were taken from the cheeses with a sterile sampler and ground aseptically. Then, a 20-g sample was homogenized for 3 min in a stomacher lab blender (PBI International, Milan, Italy) with sterile sodium citrate solution (2% wt/vol). From cheese homogenates, decimal dilutions were made in 0.1% (wt/vol) sterile peptone water. For lactobacilli adjunct cultures and NSLAB, counts were performed on MRS-Agar after evaluation of a set of different culture media, which demonstrated that recuperation on MRS was the best and that the starter strain colonies did not interfere (Bude-Ugarte et al., 2006). Tested media were MRS agar (Biokar, Beauvais, France), bile-MRS agar (Vinderola and Reinheimer, 2000) acid-MRS agar (pH 5.5) (IDF, 1988), and Elliker agar NaCl (6.5% NaCl wt/vol; Biokar) for lactobacilli. Surface plating was made, and the plates were incubated for 48 h at 34°C. Lactic acid starter bacteria (Strep. thermophilus) were enumerated on skim milk agar (48 h at 37°C).
Overall quality of the cheeses was evaluated by 3 independent experienced graders. The overall quality of cheese samples involved the evaluation of flavor acceptability and texture acceptability and was scored using a 0 to 8 point scale (0–1: unacceptable, 2–3: substandard, 4–5: standard, 6–7: good, and 8: excellent; Lynch et al., 1996).
Statistics
The results of compositional analyses of the cheeses were compared by 1-way ANOVA. Posthoc LSD test was applied to detect groups of mean values (Statistix 8, Analytical Software, Tallahassee, FL).
Principal components analysis (PCA) was applied to the peptide maps of soluble fraction in 8% trichloroacetic acid, to reduce dimensionality, compare chromatograms objectively, and detect subjacent structures in the data ensemble. The areas of peaks expressed on arbitrary units were considered as independent variables for PCA, with standardization to a mean of zero and their original variances (covariance matrix; Bergamini et al., 2006). Multivariate analysis was performed with the software Unscrambler 7.6 (CAMO ASA, Oslo, Norway).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
. The other strains were slow in which case cultures only reached pH 5.78 ± 0.19. Fast strains were L. rhamnosus (3 strains), L. curvatus (1 strain), L. plantarum (2 strains), and L. delbrueckii ssp. lactis (1 strain). Average acidifying and proteolytic activities of fast strains were 1.33 ± 0.03% lactic acid and A340 0.19 ± 0.03, respectively.
|
= 0.05), similarly to fast strains, when cultured in milk supplemented with casein hydrolysate, glucose, or both. This group represented 44% of the slow strains (7 strains) and was mainly composed of L. casei strains (4 strains), but also by strains of L. perolens (1), L. plantarum (1), and L. rhamnosus (1). The rest of the strains did not grow in milk when it was enriched with casein hydrolysate or glucose, not even when it was inoculated with both. This group of strains consisted of L. casei (3), L. curvatus (2), L. fermentum (1), and L. plantarum (2) strains. The proteolytic activity of the strains assayed was low or moderate, taking into account the comparative activities of the other lactic acid bacteria assayed in the same conditions. The average value was A340 0.18 ± 0.02 for NSLAB strains, whereas A340 ranged from 0.02 for Strep. thermophilus 5-C to 0.83 for L. helveticus SF209.
As for PCA results (Figure 2), the peptide profiles of NSLAB cultures were grouped together on the score plot of the first vs. the second principal components (PC1 and PC2), occupying the quarter of the chart defined by positive values of PC1 and PC2, as well as other moderately proteolytic lactic acid bacteria assayed in the same test. The peptide maps corresponding to the most proteolytic strains, which were tested in the same conditions, differed from NSLAB profiles, both in the amount and area of the peaks, and were plotted separately, in the quarter of the graph defined by positive values of PC1 and negative values of PC2. The Strep. thermophilus strains, which were weakly proteolytic, were also grouped apart, in the opposite quarter of the chart (Figure 2a). The impact of different peaks from the peptide profile on PC1 and PC2 is shown in Figure 2b: only 4 peaks showed high influence, whereas the rest had a low impact. The former eluted early during chromatography: they probably represent hydrophilic oligopeptides.
|
). When NaCl or KCl concentration was increased to 2 or 3%, the proportion of the strains that grew as much as in the control diminished; however, all the strains assayed showed significant growth in the tested conditions (Figure 3b and 3c). No sensitive strains were found among the nonstarter L. casei strains, for the 6 bacteriophages tested.
|
In Situ Characterization: Cheese Making Experiments
Primary starter was about 108 cfu/g in curd samples at molding and 109 cfu/g in cheeses after brining. This last level remained more or less constant during ripening of Cremoso cheeses. In Pategrás cheeses, the starter population decreased slightly during ripening. Primary starter counts were similar in control and experimental cheeses in Cremoso and Pategrás cheese varieties (Figures 4 and 5).
|
|
). The NSLAB level in Cremoso cheeses without added lactobacilli (control cheeses) was always under 105 cfu/g (results not shown).
As for Pategrás cheeses, adjunct lactobacilli rapidly increased from 107 cfu/g in curd up to 108–109 in 7-d-old cheeses. The L. casei I90 population reached 109 cfu/g in 7 d and then remained constant at a level that equaled the starter culture population. The L. plantarum I91 and L. rhamnosus I77 and I73 also attempted populations similar to that of the starter, but after the first month of ripening. The NSLAB in Pategrás cheeses without added lactobacilli augmented up to 106 cfu/g in 7 d of ripening and reached 108 in 15-d-old cheeses (Figure 5).
The evolution of pH during Cremoso cheese makings was similar in control and experimental cheeses with adjunct cultures of L. casei I90 and L. plantarum I91. On the contrary, experimental cheeses that included L. rhamnosus strains showed an increased rate of acidification during the incubation step in the hot chamber, before salting (Figure 6).
|
).
|
).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The fact that most of the tested strains were not able to coagulate milk is not a drawback for their use as adjunct cultures in cheese making because secondary lactic cultures need not significantly contribute to acidification during cheese making. In fact, excessively high acidifying activity can cause curd over acidification and change the moisture content of the cheese (Crow et al., 2001). Highly acidifying nonstarter lactobacilli strains may be evaluated for mixed primary starter cultures; L. plantarum and L. casei isolates have been assayed as a part of mixed starters in several cheese varieties (Mendia et al., 2000; Macedo et al., 2004; Herreros et al., 2007).
Proteolytic activity, on the contrary, is an interesting property for all, adjunct or primary cheese starters as proteolysis and peptidolysis during cheese ripening can influence the background flavor of the product, and provide most of the precursors of cheese aroma (Yvon, 2006). Among the tested strains, fast lactobacilli showed moderate proteolytic and peptidolytic activities compared with other lactic acid bacteria, such as L. helveticus SF209, which is highly proteolytic and peptydolitic in vitro and in cheese (Quiberoni et al., 1998; Hynes et al., 2003), and Strep. thermophilus 5-C, which scarcely shows any proteolytic activity (Binetti et al., 2007). Several peptidases have been identified in nonstarter lactobacilli of the species L. plantarum and the L. casei group, including aminopeptidases, dipeptidases, and peptidases for proline-containing peptides (Williams and Banks, 1997; Martinez-Cuesta et al., 2001). Even if no carboxypeptidasé has been purified from lactic acid bacteria yet, carboxy-peptidase activity was detected in nonstarter lactobacilli (Macedo et al., 2000). Peptidolytic activities in NSLAB are species- and strain- dependent and rely on the growing conditions of the culture and the physiological state of the cells (Williams et al., 2002). Because fast strains showed similar proteolytic activities and produced equivalent peptide profiles that other lactic acid bacteria already tested in cheese with positive results (Hynes et al., 2003), it can be expected that they will have a favorable influence in peptide breakdown during cheese ripening.
Rotation of starter cultures is a common strategy in dairy industry to control phage proliferation and prevent fermentation failures (Moineau and Lévesque, 2005). A relatively large choice of primary starters is available for this purpose, but secondary cultures such as adjunct or probiotic lactobacilli do not offer as many possibilities because available strains are still few (Capra et al., 2006). In this work, we have found that all the L. casei strains were resistant to phages specific of L. casei and L. paracasei, both from collections or isolated from Argentinean dairy plants during infection episodes. This characteristic constitutes a promising feature of the tested bacteria, especially taking into account that several strains in the same set of lactobacilli have shown probiotic potential (Bude-Ugarte et al., 2006).
Not surprisingly, all the assayed strains were resistant to NaCl and KCl in concentrations compatible with those usually found in cheeses, which is consistent with the fact that these strains have been isolated from cheeses and are probably adapted to such environmental conditions. The NSLAB have been reported as more salt tolerant than starter lactic acid bacteria (Fox et al., 1993), but this characteristic depends on the species and the strain of the starter (Lane et al., 1997). Resistance to salt is an evident technological requisite for cheese cultures.
All the lactobacilli cultures tested in cheese making experiments reached high levels in the cheeses during the first 24 h after manufacture and remained more or less constant during 60 d of ripening. This result indicates that the lactobacilli studied were able to proliferate in the cheese, although most of them had not been able to grow in milk. This is in agreement with previous reports on the subject (Stanley, 1998; Crow et al., 2001). In cheese, energy source for nonstarter lactobacilli remains unclear (Thomas, 1987; Williams et al., 2000; Beresford et al., 2001), whereas their nitrogen nutritional requirements are probably fulfilled by peptides and free amino acids derived from the proteolytic activity of the primary starter (Lane et al., 1997; Hynes et al., 2001b; Di Cagno et al., 2003).
In control cheeses without added lactobacilli, NSLAB attained high numbers after 15 d in Pategrás cheese, but remained below 105 cfu/g until the end of ripening in Cremoso cheese. Similar lactobacilli counts were found in semihard Argentinean cheeses, whereas soft cheeses from industrial environment showed a somewhat higher level of NSLAB (Bude-Ugarte et al., 2006). The difference in NSLAB development in both cheese models is probably due to the different ripening temperature (Shakeel-Ur-Rehman et al., 2000a).
The 2 L. rhamnosus strains did not perform as well as adjunct cultures than L. casei I90 and L. plantarum I91, especially in the soft cheese model. These cultures modified the pH curve during curd cheddaring in Cremoso cheese making, which resulted in lower pH and increased DM in the product. These changes were verified for the 2 tested L. rhamnosus, although one of the strains had been classified as fast (L. rhamnosus I73) and the other, slow (L rhamnosus I77). In Pategrás cheese, only L. rhamnosus I77 caused significant decrease of pH toward the end of the ripening, and gross composition was not altered by the acidifying activity of the adjunct cultures. The L. rhamnosus 77 was not expected to be an over-acidifying adjunct culture because it was classified as a slow strain according to its ability to grow in milk. In cheeses, the existence of other lactic cultures in the food ecosystem, and the higher availability of oligopeptides and free amino acids, probably favored the growing of the adjunct culture, although information about NSLAB growing in cheese, especially concerning energy source, is still lacking.
Adjunct lactobacilli added to cheese both to improve the control over NSLAB adventitious flora or as probiotic cultures are required not to impact negatively on quality and, if possible, to improve sensory characteristics of the products (Crow et al., 2001; Bergamini et al., 2006). Some cultures of lactobacilli have been reported to produce cheeses more intensely flavored than controls, although aroma and flavor were not always improved (McSweeney et al., 1994; Lynch et al., 1999). In this study, the sensory quality of cheeses was generally considered good, except for Cremoso cheeses with L. rhamnosus that received lower scores, probably because low pH impacts significantly in soft cheese texture (Hynes et al., 1999). Cremoso cheeses with L. plantarum I91 and L. casei I90 were the best scored; similar results were found for one of these strains in miniature Cremoso cheeses (Mercanti et al., 2006).
In the present work, a group of nonstarter lactobacilli strains has been studied by focusing on their technological characteristics. The majority of the strains have shown promising technological properties, such as resistance to phage infections and tolerance to salts. In addition, several strains were able to coagulate milk and evidenced diverse proteolytic and peptidolytic activities. In cheeses, all the assayed lactobacilli reached high levels and survived during ripening. The L. casei I90 and L. plantarum I91 were appropriate for their use as adjunct cultures in both soft and semihard cheeses because they did not alter the composition of the products and improved their sensory attributes, whereas the L. rhamnosus strains tested were only suitable for the semihard cheese model. The impact of these strains in proteolysis and sensory profile of the cheeses will be the object of further study.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Received for publication March 8, 2007. Accepted for publication June 13, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
ez, P. Torre, and Y. Barcina. 2000. Effect of pasteurization and use of a native starter culture on proteolysis in a ewes’ milk cheese. Food Contr. 11:195–200.[CrossRef]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, •: cheeses with Lactobacillus rhamnosus I73; 
, 
: L. rhamnosus I77; 
, 
: Lactobacillus casei 90, 
, 
: Lactobacillus plantarum I91. Nonstarter lactobacilli in control cheeses remained always below 105 UFC/g and are not shown.
