HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Interpretive Summary Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ferreira, I. M. P. L. V. O. Articles by Mendonça, A. Search for Related Content PubMed PubMed Citation Articles by Ferreira, I. M. P. L. V. O. Articles by Mendonça, A.

Casein Breakdown in Terrincho Ovine Cheese: Comparison with Bovine Cheese and with Bovine/Ovine Cheeses

I. M. P. L. V. O. Ferreira^*,1, C. Veiros^*, O. Pinho^*,, A. C. A. Veloso, A. M. Peres and A. Mendonça

^* REQUIMTE—Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto, R. Aníbal Cunha 164, 4050-047 Porto, Portugal
Faculdade de Ciências da Nutrição e Alimentação, Universidade do Porto, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal
Escola Superior Agrária de Bragança—Instituto Politécnico de Bragança, Quinta de Santa Apolónia, Apartado 1172, 5301-855 Bragança, Portugal

¹ Corresponding author: isabel.ferreira{at}ff.up.pt

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Terrincho cheese is an uncooked, pressed cheese made from raw whole ovine milk from the "Churra da Terra Quente" breed. It requires a minimum ripening time of 30 d. A detailed evaluation of the effect of ripening time on the breakdown of the casein fractions, along with the formation of major breakdown products of casein hydrolysis, was monitored by HPLC to contribute to a more complete characterization of this product. In 30-d-old cheeses, only 20% of _S1-casein remained intact; the ß-casein fraction was more resistant to hydrolysis. The ripening time of Terrincho cheese can be predicted using 2 variables of normalized peak areas of _S1-casein and _S1-I peptide, and a constant; the estimation error is 2.5 d. The pH 4.3-insoluble fraction of Terrincho and cheeses manufactured with bovine milk and with ovine milk combined with 2 levels of bovine milk (10 and 20%) revealed different chromatographic and electrophoretic profiles, especially the _S1-casein fraction. Similar proteolysis progress was observed, particularly in the percentage of casein fraction degradation. However, using both analytical methods, the detection of 10% bovine milk at 30 d of ripening was no longer possible as result of _S1-casein hydrolysis. The discriminate analysis applied to HPLC data indicated that at 30 d of ripening, differences between the casein fractions of Terrincho cheese and mixture cheeses were mainly from ß₁-casein content. The function thus obtained was able to correctly classify all the samples according to cheese type. Using the descriptive sensory profile, Terrincho cheese at 30 d of ripening could be distinguished from bovine and mixture cheeses owing to its higher fracturability and adhesiveness and lower elasticity and hardness, which correlated with its lower total casein content.

Key Words: cheese • proteolysis • caseins • high performance liquid chromatography

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Terrincho cheese is an uncooked, pressed cheese made from raw whole ovine milk from "Churra da Terra Quente" breed in the regions of northern Portugal. The Protected Denomination of Origin (PDO) for this cheese falls within a very specific geographic region, and its manufacture is based on traditional methods. The making of Terrincho cheese is performed according to the regulations issued by the Denomination of Origin Regulatory Board. Terrincho cheese is produced by slowly draining the curds obtained upon coagulation via the action of animal rennet. No starter cultures are added. Terrincho is characterized as a cylinder-shaped (dish-type) cheese with a flat rind; its weight ranges between 0.8 and 1.2 kg, and it has a straw-yellow color, a semi-soft consistency, a slightly oily texture containing some holes, and a soft, very characteristic flavor. Because the cheese is made from raw milk, and for safety reasons, it requires a minimum ripening time of 30 d before cheese can be sold.

The average composition of Terrincho cheese is 45% total solids, 50 g of fat/100 g of DM, 40 g of protein/100 g of DM, and 2.5% salt on a DM basis (Pinho et al., 2004a). A number of variable factors are known to affect the composition of ovine cheeses: breed, climate, variation in the physicochemical and microbiological composition of the milk, rennet type, and ripening conditions, such as ripening time, temperature, and relative humidity. All these factors affect cheese lipolysis and pro-teolysis and strongly contribute to flavor and texture development (Fox, 1989; Ordonez et al., 1997; Gaiaschi et al., 2000, 2001; Irigoyen et al., 2000; Marino et al., 2000; Sousa et al., 2001; Bustamante et al., 2003; Dave et al., 2003).

Previous studies on the lipolysis and proteolysis of Terrincho cheese have mostly focused on the characterization of its composition relating to microbiology, texture, and quantification of volatile free fatty acids and other volatile compounds, and have examined low primary proteolysis and production of free biogenic amines (Pinho et al., 2003, 2004a,Pinho et al., b,c). Therefore, a detailed evaluation of the effect of ripening time on the breakdown of the different caseins during ripening, along with the formation of peptides and other breakdown products of CN hydrolysis is necessary for a more complete characterization of the product (Veloso et al., 2004).

Analysis of proteolysis by nonspecific methods, for example, by quantification of nitrogen in different peptide and amino acid fractions, provides information about the degree of proteolysis in cheese ripening (Pinho et al., 2004c). Nevertheless, the results of these methods do not express well the complexity and speci-ficity of proteolytic development during cheese ripening; namely, the evolution of the 4 major CN, _S1-, _S2-, ß-, and para -CN, and their proteolytic products. Electrophoretic and chromatographic techniques, however, resolve proteins or groups of peptides and thereby provide proteolytic profiles that can give useful information about proteolysis extension, ripening time, and product authenticity (Restani et al., 1996a; Trujillo et al., 2000; Veloso et al., 2002a,b; Borková and Snáselová, 2005).

Adulteration of traditional ovine cheeses with bovine milk presents a problem for food monitoring and one solution for this problem is the analysis of milk proteins (Borková and Snáselová, 2005). However, the proteoly-sis that occurs during cheese ripening incurs a risk of complex formation and the formation of insoluble new compounds and smaller peptides that can interfere with evaluation of authenticity. Examples of these compounds are -CN and proteoso-peptone fragments of ß-CN, originating from the action of plasmin. The -CN, presumably fragments of _S1-CN, glycomacropeptides, and para -CN, are the result of chymosin action (Borková and Snáselová, 2005).

To date, there have been reports on the use of chemo-metric analysis for evaluation of the proteolytic process in different types of cheese during ripening, as well as for predicting cheese ripening time (Pham and Nakai, 1984; Santamaría et al., 1986; García-Ruiz et al., 1998; Herrero-Martínez et al., 2000; Pripp et al., 2000). However, limited data are available on the application of multivariate statistical analysis to predict ripening time and authenticity of Terrincho cheeses. Previous studies carried out by our research group using small cheeses (70 g) indicated that HPLC profiles of the pH 4.3-insoluble fraction could be used not only to follow proteolysis but also to evaluate the authenticity of ewe cheeses (Veloso et al., 2004). However, differences may occur in ripening relating to cheese size variations, native microflora, and others. For that reason, further studies with Terrincho cheeses of real dimensions were performed for quantitative measurements of CN degradation in these cheeses.

The objective of the present study was to evaluate the proteolytic process during 30 d of ripening of Terrincho cheese based on the protein patterns obtained by HPLC of the pH 4.3- insoluble fractions. Chemometric analysis of the HPLC data was also used to predict the ripening time. Additionally, the influence of milk origin on prote-olysis and cheese sensory characteristics was studied through evaluation of HPLC and urea-PAGE CN pro-files during ripening of cheeses manufactured using the same methodology to guarantee Terrincho PDO authenticity.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Cheese Making
A batch of Terrincho PDO cheeses was manufactured during the winter at a certified dairy plant, according to the traditional method. Briefly, prefiltered pure raw ewe’s milk of the "Churra da Terra Quente" breed was heated to 35°C, and coagulated with animal rennet derived from artisanal production; the curds were slightly drained and placed in molds, where they were subjected to pressure to help remove the remaining whey. The cheeses were salted upon unmolding and placed in ripening chambers held at 10 to 12°C and 88 to 89% relative humidity.

Simultaneously in the same dairy plant, 3 other batches of cheeses were manufactured according to Ter-rincho technology, one batch using raw bovine milk and 2 batches using mixtures of bovine and ovine raw milks [10 to 90% mixture (M1) and 20 to 80% mixture (M2) bovine to ovine, respectively]. These percentages were selected because use of less than 10% of bovine milk is not profitable economically, whereas use of more than 20% is easily detected by consumers.

From these 4 batches, groups of 2 cheeses were randomly taken at 0, 7, 14, 21, and 30 d of ripening and the pH 4.3-insoluble fraction was assayed by HPLC and urea-PAGE. Cheeses from each batch at 30 d of ripening were used for sensory analysis; 40 cheeses were selected.

All cheeses were ripened locally under the aforementioned conditions. At each sampling time, cheeses were duly conditioned in refrigerated boxes and sent promptly to our laboratory. Two samples of each cheese were cut, labeled, and frozen at –40°C pending analysis of the study parameters. The analyses were carried out in duplicate, so that 8 analyses were determined for each type of cheese on each of the collection dates.

Milk used for preparation of each batch of cheese was also analyzed for the chromatographic and electrophoretic profile of pH 4.3-insoluble fraction.

Reagents and Protein Standards
All reagents used were of analytical grade purity. Eluents for HPLC were filtered through 0.22-µm NL 17 filters (Whatman, Brentford, UK) and degassed under vacuum for at least 15 min before use.

Bovine casein, with a minimum purity of 75%, determined by the method of Bradford (1976), was supplied by Sigma Chemical Co. (St. Louis, MO). Purified bovine -, ß-, and -CN were obtained from Sigma Chemical Co., and had a minimum purity of 85, 90, and 80% (according to Sigma), respectively. Casein standards were dissolved in a mixture of 70% water and 30% acetonitrile (vol/vol) for HPLC analysis, and in deionized water at pH 10 for electrophoretic assays.

Chemical Analysis
Analyses were performed using cheese samples that were thawed at 3 to 4°C for 12 h. The pH was measured by inserting a combined Ag-AgCl electrode into a comminuted sample according to the method of Berdague and Grappin (1987). Dry matter was evaluated using equipment from Scaltec Instruments GmbH (Goet-tingen, Germany).

Caseins Extraction
Isoelectric caseins were obtained by precipitation from 2.5 g of homogenized cheese in 15 mL of water by adding 1 M HCl to pH 4.3, followed by centrifugation at 4°C for 15 min at 3,000 x g to recover the precipitated caseins. To isolate the casein fraction completely from whey, the precipitate was washed once with 1 mM ammonia acetate buffer (pH 4.3) and centrifuged at 3,000 x g for 10 min, at 4°C. This procedure was repeated twice. To remove the remaining fat, the sample was washed with acetone, and left to dry in a fume hood at room temperature. Finally, the dried, powdered casein was weighed, and stored in a desiccator at 8°C until analyzed. The dried powder was dissolved in a mixture of 70% water and 30% acetonitrile (vol/vol) and in deionized water at pH 10 for the HPLC and electrophoretic analysis, respectively. Total casein content at each ripening time (expressed as g of casein/100 g of cheese) was measured from the weight of extracted caseins.

HPLC Separations
The chromatographic analysis was carried out in an analytical HPLC unit (Jasco, Tokyo, Japan) equipped with 2 type PU-980 pumps, a type UV-970 detector, and a type 7125 Rheodyne injector with a 20-µL loop. Borwin PDA Controller software (JMBS Developpe-ments, Grenoble, France) was also used. The column was a reversed-phase Chrompack P 300 RP column (Varian, Harbor City, CA) that contained polystyrene-divinylbenzene copolymer-based packing (8 µm, 300 Å, 150 x 4.6 mm i.d.); a Chrompack P RP (24 x 4.6 mm i.d.) precolumn was used.

Gradient elution was carried out with a mixture of 2 solvents. Solvent A consisted of 0.1% trifluoroacetic acid (TFA) in water and solvent B consisted of 0.1% TFA in 80% aqueous acetonitrile (vol/vol). Proteins were eluted with a series of linear gradients: 29% solvent B in solvent A during 5 min, from 29 to 37% B over 5 min, 37 to 54% solvent B over 15 min, holding for 2 min, finishing with 54 to 100% of solvent B in 3 min, followed by 5 min for column reequilibration. The flow rate was 1.0 mL/min, the column temperature was 46 ± 0.1°C, and the detection wavelength was 280 nm.

Peaks were characterized using caseins from bovine and ovine milk previously identified in our laboratory (Veloso et al., 2002b) according to previous identifica-tion (Visser et al., 1986). The areas under the 5 most representative peaks on the chromatogram, designated -CN (para -CN), _S1-I (_S1-I peptide), _S2-CN, ß₂-CN, _S1-CN, and ß₁-CN in order of elution (Figure 1), were used to study the evolution of proteolytic process.

View larger version (10K): [in this window] [in a new window]   Figure 1. Chromatographic patterns (HPLC) of Terrincho cheese from pH 4.3-insoluble fraction at 1 d of ripening. = para--casein; S2 = S2-casein; S1 = S1-casein; ß2 = ß2-casein; ß1 = ß1-casein.

Urea-PAGE Analyses
Polyacrylamide gel electrophoresis of casein samples was performed according to the method of Andrews (1983) with some modifications. The assays were carried out in a vertical vat (SE 250/260 Amersham Phar-macia Biotec, model Mighty, Amersham, Freiburg, Ger-many), using an EPS 301 power supply (Amersham Pharmacia Biotec).

The slab gels consisted of a 4% stacking gel and a 10% running gel. The stacking gel buffer was 0.06 M Tris and 4.5 M urea at pH 7.6, and the resolving gel buffer was 0.76 M Tris and 9 M urea at pH 8.9. The electrophoresis buffer was a solution of 0.02 M Tris and 0.19 M glycine. The run was performed at 4°C at 20 mA until the end of the stacking gel, followed by a current of 30 mA. The gels were stained with Coomassie Brilliant Blue R250.

Statistical Analyses
Data were autoscaled before statistical analysis. This normalization involved dividing each value of a given variable by the standard deviation of all the values for this variable over the entire sample collection period (Garrido Frenich et al., 1995). After normalization, all variables had the same weight because they had a mean of zero and unitary variance. Exploration of data, descriptive statistics, t-test, and ANOVA with pairwise comparisons of mean values using Tukey’s test, discriminant analysis, and multiple linear regressions were performed with SPSS for Windows version 13 (SPSS, Chicago, IL).

Sensory Analysis
Sensory analyses were performed by a jury of 20 panelists of both sexes who had been previously selected (ISO 8586-1, 1993). The taste panel evaluated texture, odor, and flavor attributes using a 7-point scale.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Proteolysis of Terrincho Cheese CN Fraction by Reverse-Phase HPLC
Figure 1 indicates the chromatographic patterns obtained from Terrincho cheese CN at 1 d of ripening. The HPLC profiles of the Terrincho cheese pH 4.3-insoluble fraction at the beginning of ripening were similar to the profiles of CN from ovine milk (chromatograms not shown).

Two peaks observed in all the chromatograms near the 30-min retention time are the result of elution of small amounts of TFA, which adsorbs to the stationary phase and is eluted when eluent returns to initial composition. Similar behavior was observed by Elgar et al. (2000).

Only slight modifications were noted in the chromatograms at 7 d of ripening. However, important changes in chromatographic areas occurred between 7 and 30 d of ripening as shown in Table 1.

One-way ANOVA was used to verify whether the average values obtained for normalized peak areas for the major peaks, para -CN, _S1-I peptide, _S2-CN, _S1-CN, ß₂-CN, and ß₁-CN, could be considered different or not during ripening. This was possible because data obtained for each ripening time were presented in general normal distribution (P > 0.05; Shapiro-Wilk test) and homoscedasticity of variances (P > 0.05; Levene test). Results obtained for ANOVA and posthoc tests (Tukey HSD) are shown in Table 1. It was concluded, with 95% confidence, that there were significant differences in the effect of ripening time of all quantified fractions (for para -CN, _S1-I peptide, _S2-CN, _S1-CN, ß₁-CN, and ß₂-CN). A significant reduction in the para-CN, _S1-CN, and _S2-CN contents depending on the ripening time was observed, together with an increase in _S1-I peptide. A reduction of 80% of _S1-CN was observed during the 30 d of ripening. This is due to the action of residual rennet, together with the action of the hydrolytic enzymes of the microorganisms present in cheese. The primary site of chymosin action on _S1-CN is the Phe₂₃-Val₂₄ bond, with the appearance of _S1-I peptide breakdown product; this peptide suffers further hydrolysis (Michaelidou et al., 1998; Irigoyen et al., 2000; Prieto et al., 2004). A reduction of 75% was observed for _S2-CN fraction, owing to chymosin action on these fractions (Singh et al., 1994). Some oscillations in ß₁-CN and ß₂-CN were observed, but without a significant decreasing trend. Thus, the fractions that corresponded to the ß-CN region were less degraded than those that corresponded to the _S1-CN and _S2-CN, which were hydrolyzed more quickly during the first stages of ripening. The greater resistance of ß-CN to enzyme hydrolysis was already pointed out by several authors (Izco et al., 1999; Irigoyen et al., 2000). This fraction is mainly hydrolyzed by plasmin from milk (Gaiaschi et al., 2001) and it most likely does not sig-nificantly contribute to proteolysis of Terrincho cheese, owing to its pH between 5 and 6 (Lawrence et al., 1987).

A reduction of 40% was observed for fragments of para -CN; this fraction is thought to be very resistant to further proteolysis after milk coagulation. However, peptides from -CN were isolated in Feta cheese, and its formation could be the result of the action of lactococcal proteinase at Met₉₅-Ala₉₆, which exhibits the characteristics of a susceptible cleavage site for such an enzyme (Reid et al., 1991; Michaelidou et al., 1998).

Prediction of Terrincho Cheese Ripening Time by Multiple Linear Regression Analysis
Terrincho cheese ripening time was estimated with stepwise variable selection involving the 6 variables by multiple linear regression analysis. The general formula of the estimation equation is as follows:

where y is cheese ripening time, x₁, x₂, ..., x_n are the normalized areas of the major casein fractions, and represents the errors or residuals of the model associated with the y values.

The correlation between the measured and estimated values is shown in Figure 2. The product ripening time can be estimated with 2 variables: normalized peak area of _S1-I peptide and _S1-CN, as well as a constant (correlation coefficient = 0.917). The estimation error is 2.5 d. This error is lower than that obtained for prediction of Terrincho cheese ripening time based on texture and color parameters measured instrumentally (Pinho et al., 2004a). Our results are in agreement with other authors who identified the pattern of _S-CN and related fragments as useful tools for the control of cheese ripening and quality (Gaiaschi et al., 2000).

View larger version (12K): [in this window] [in a new window]   Figure 2. Prediction of ripening time for Terrincho cheeses; linear regression with 95.0% individual prediction interval; y = ripening time; x = major casein fractions normalized areas. y = 13.56 + 10.7x S1-I peptide –11.8x S1-casein. SE = 2.5, R = 0.92, n = 32, P < 0.01.

The total CN content also decreased during ripening. At the beginning of ripening, total CN content was 19.5 ± 2.3, 21.5 ± 2.1, 20.1 ± 2.0, and 20.5 ± 1.8 g/100 g of cheese, respectively, for Terrincho, bovine, and M1 and M2 cheeses. No significant differences were found between CN levels of the 4 types of cheese. In 30-d-old cheeses, Terrincho had significantly lower CN content than other cheeses, 14.1 ± 0.6 g/100 g of product, with no significant differences (P < 0.05) found among levels in M1, M2, and bovine cheeses (17.5 ± 1.1, 16.1 ± 0.9, and 17.3 ± 0.8 g/100 g of product, respectively).

The HPLC profiles of the pH 4.3 insoluble fractions of Terrincho, bovine, M1, and M2 cheeses with 7 d of ripening are shown in Figure 3. The chromatographic profile of Terrincho cheese CN was significantly different from that of bovine cheese CN concerning proportions of -CN, _S-CN, and ß-CN fractions (24:47:29 and 10:56:34, respectively). Differences were also observed in retention times of _S-CN from ovine and bovine milk. As a result of variations in retention times of ovine and bovine _S1-CN, the chromatographic patterns of M1 and M2 cheeses were different compared with the Terrincho cheese pH 4.3-insoluble fraction, resulting in an overlap of bovine _S1-CN with ovine ß₂-CN and _S1-CN. This was more prominent in mixed cheeses with 20% bovine milk.

View larger version (14K): [in this window] [in a new window]   Figure 3. Chromatograms for cheese whole casein fraction after ripening for 7 d: a) Terrincho cheese; b) bovine; c) mixture 1 (10% bovine milk and 90% ovine); d) mixture 2 (20% bovine milk and 80% ovine). = para--casein; S2 = S2-casein; S1 = S1-casein; ß2 = ß2-casein; ß1 = ß1-casein.

View larger version (13K): [in this window] [in a new window]   Figure 4. Chromatograms for cheese whole casein fraction after ripening for 30 d: a) Terrincho cheese; b) bovine; c) mixture 1 (10% bovine milk and 90% ovine); d) mixture 2 (20% bovine milk and 80% ovine). = para--casein; S2 = S2-casein; S1 = S1-casein; ß2 = ß2-casein; ß1 = ß1-casein.

View larger version (48K): [in this window] [in a new window]   Figure 5. Urea-PAGE of the pH 4.3-insoluble fraction of ovine and bovine milks (lanes 1 and 6) and Terrincho cheese (lane 2), mixture 1 (10% bovine milk and 90% ovine; lane 3), and mixture 2 (20% bovine milk and 80% ovine; lane 4) cheeses, and bovine cheese (lane 5) after manufacture.

View larger version (59K): [in this window] [in a new window]   Figure 6. Urea-PAGE of the pH 4.3-insoluble fraction of ovine and bovine milks (lanes 1 and 6) and Terrincho cheese (lane 2), mixture 1 (10% bovine milk and 90% ovine; lane 3) and mixture 2 (20% bovine milk and 80% ovine; lane 4) cheeses, and bovine cheese (lane 5) after 30 d of ripening.

In a second stage of the study, discriminant analysis was applied to data obtained by HPLC from Terrincho, mixture, and bovine cheeses with 30 d of ripening, selecting only peaks corresponding to -CN, _S1-I peptide, _S2-CN, and ß₁-CN; the peaks that overlapped in mixture cheeses were excluded. The discriminant analysis indicated that 3 variables contributed significantly to explaining the variability among the 4 types of cheese: -CN and _S1-I peptide were the major contributors to function 1 (loadings of –1,223 and 1,083)—the first was higher in Terrincho and mixture cheeses and the second in bovine cheeses (loading of 1,083)—and ß₁-CN was positively correlated with function 2 (loading of 0,703). Plots of the samples in Figure 7 of the 2 main discriminant functions explained 98.0% of the total variance. Differences between Terrincho and mixture cheeses result from higher ß₁-CN content in mixture cheeses, because this fraction is higher in bovine milk and more resistant to enzyme hydrolysis until 30 d of ripening. The function thus obtained was able to correctly classify all the samples according to cheese type.

View larger version (8K): [in this window] [in a new window]   Figure 7. Canonical discriminant functions showing 4 groups of different types of cheeses classified by the 2 functions of the proposed model.

View larger version (31K): [in this window] [in a new window]   Figure 8. Mean scores obtained by 20 panelists for the 15 attributes assessed in Terrincho (T), mixture 2 (M), and bovine (B) cheeses.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Data presented in this and our previous paper (Veloso et al., 2004) has contributed to deepening the knowledge on the casein degradation of Terrincho cheese throughout ripening. In particular, the relationship between CN fraction composition, ripening time, milk origin, and cheese sensory characteristics were approached in this paper in a statistically validated manner.

High performance liquid chromatography is well suited to analyze the pH 4.3-insoluble fraction during cheese ripening. The method was able to quantify CN degradation and to study the effect of the ripening time on the proteolytic process in Terrincho cheese.

During the 30 d of Terrincho cheese ripening, a relatively high resistance of ß-CN to hydrolysis by residual chymosin and microorganisms present in cheese was verified, as observed in most cheeses (Fox, 1989). In contrast, the hydrolysis of _S1-CN and _S2-CN was significant. In 30-d-old cheeses, only 20% of _S1-CN remained intact. The ripening time of Terrincho cheese can be predicted using the 2 variables of normalized peak areas of _S1-CN and _S1-I peptide, and a constant. The estimation error was 2.5 d. In spite of the different proportions of CN fractions in bovine and mixture cheeses, the evolution of cheese proteolysis followed a similar trend. Differences between Terrincho and mixture cheeses result from higher ß₁-CN content in mixture cheeses. The _S1-CN fraction related to Terrincho cheese ripening time, and ß₁-CN fraction was a marker for Terrincho cheese authenticity.

Terrincho cheese descriptive sensory profile at 30 d of ripening was distinguished from bovine and mixture cheeses owing to its higher fracturability and adhesiveness and lower elasticity and hardness; these parameters were correlated with its lower total CN content.

In conclusion, chemometric analysis of HPLC proteo-lytic profiles has been shown to be a powerful method to examine the biochemical process of proteolysis and to predict the ripening time and authenticity of Terrin-cho cheese.

Received for publication October 26, 2005. Accepted for publication January 11, 2006.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 


Andrews, A. T. 1983. Proteinases in normal bovine milk and their action on caseins. J. Dairy Res. 50:45–55.[Medline]

Berdague, J. L., and R. Grappin. 1987. Affinage et cualité du Gruyère de Comté. II. Influence de l’affinage sur l’ évolution des caractéris-tique physico-chimimiques des fromages. Lait 67:237–241.

Borková, M., and J. Snáselová. 2005. Possibilities of different animal milk detection in milk detection in milk and dairy products—A review. Czech. J. Food Sci. 23:41–50.

Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye-binding. Anal. Biochem. 72:248–254.[Medline]

Bustamante, M. A., M. Virto, M. Aramburu, L. J. R. Barron, F. J. Pérez-Elortondo, M. Albisu, and M. de Renobales. 2003. Lamb rennet paste in ovine cheese (Idiazabal) manufacture. Proteolysis and relationship between analytical and sensory parameters. Int. Dairy J. 13:547–557.

Dave, R. I., D. J. McMahon, C. J. Oberg, and J. R. Broadbent. 2003. Influence of coagulant level on proteolysis and functionality of Mozzarella cheeses made using direct acidification. J. Dairy Sci. 86:114–126.[Abstract/Free Full Text]

Elgar, D. F., C. S. Norris, J. S. Ayers, M. Pritchard, D. E. Otter, and K. P. Palmano. 2000. Simultaneous separation and quantification of the major bovine whey proteins including proteose peptone and caseinomacropeptide by reversed-phase high performance liquid chromatography. J. Chromatogr. A 878:183–196.[Medline]

Fox, P. F., and P. L. H. McSweeney. 1998. Milk proteins. Pages 146–238 in Dairy Chemistry and Biochemistry. 1st ed. Blackie Academic & Professional, London, UK.

Fox, P. F. 1989. Proteolysis during cheese manufacture and ripening. J. Dairy Sci. 72:1379–1400.[Abstract/Free Full Text]

García-Ruiz, A., L. Cabezas, P. J. Martín-Alvárez, and D. Cabezudo. 1998. Prediction of the ripening times of Manchego cheese using multivariate statistical analysis: A preliminary study. Z. Lebensm. Unters. Forsch. 206:382–386.

Garrido Frenich, A., D. Jouan-Rimbaud, D. L. Massart, S. Kutta-tharmmakul, M. Martinez Galera, and J. L. Martinez Vidal. 1995. Wavelength selection method for multicomponent spectroscopic determinations using partial least squares. Analyst 120:2787–2792.

Gaiaschi, A., B. Beretta, C. Poiesi, A. Conti, M. G. Giuffrida, C. L. Galli, and P. Restani. 2000. Proteolysis of _S-casein as a marker of Grana Padano cheese ripening. J. Dairy Sci. 83:2733–2739.[Abstract]

Gaiaschi, A., B. Beretta, C. Poiesi, A. Conti, M. G. Giuffrida, C. L. Galli, and P. Restani. 2001. Proteolysis of ß-casein as a marker of Grana Padano cheese ripening. J. Dairy Sci. 84:60–65.[Abstract]

Herrero-Martínez, J. M., E. F. Simó-Alfonso, G. Ramis-Ramos, C. Gelfi, and P. G. Righetti. 2000. Determination of cow’s milk and ripening time in nonbovine cheeses by capillary electrophoresis of the ethanol-water protein fraction. Electrophoresis 21:633–640.[Medline]

Irigoyen, A., J. M. Izco, F. C. Ibáñez, and P. Torre. 2000. Evaluation of the effect of rennet type on casein proteolysis in an ovine milk cheese by means of capillary electrophoresis. J. Chromatogr. A. 881:59–67.[Medline]

ISO. 1993. 8586-1. Analyse sensorielle, Guide général pour la sélection, l’entraînement et le contrôle des sujets. Partie 1: Sujets qualifiés. In Contrôle de la qualité des produits alimentaraires—Analyse sensorielle, 5e édition (1995). ISO, Geneva, Switzerland.

Izco, J. M., P. Torre, and Y. Barcina. 1999. Capillary electrophoresis: Evaluation of the effect of added enzymes on casein proteolysis during the ripening of a ewe’s milk cheese. Adv. Food Sci. 21:110–116.

Lawrence, R. C., L. K. Creamer, and J. Gilles. 1987. Texture development during cheese ripening. J. Dairy Sci. 70:1748–1760.[Abstract/Free Full Text]

Marino, M., M. Maifreni, S. Moret, and G. Rondinini. 2000. The capacity of Enterobacteriaceae species to produce biogenic amines in cheese. Lett. Appl. Microbiol. 31:169–173.[Medline]

Michaelidou, A., E. Alichanidis, H. Urlaub, A. Polychroniadou, and G. K. Zerfiridis. 1998. Isolation and identification of some major water-soluble peptides in Feta cheese. J. Dairy Sci. 81:3109–3116.[Abstract]

Ordonez, A. I., C. I. Francisco, P. Torres, and Y. Barcina. 1997. Characterization of the casein hydrolysis of Idiazábal cheese manufactured from ovine milk. J. Dairy Sci. 81:2089–2095.

Pham, A., and S. Nakai. 1984. Application of stepwise discriminant analysis to high-pressure liquid chromatography profiles of water extract for judging ripening of Cheddar cheese. J. Dairy Sci. 67:1390–1396.[Abstract/Free Full Text]

Pinho, O., I. M. P. L. O. Ferreira, and M. Ferreira. 2003. Quantifica-tion of short-chain free fatty acids in "Terrincho" ewe cheese: Intravarietal comparison. J. Dairy Sci. 86:3102–3109.[Abstract/Free Full Text]

Pinho, O., I. M. P. L. O. Ferreira, and M. Ferreira. 2004b. Discriminate analysis of the volatile fraction from "Terrincho" ewe cheese: Correlation with flavour characteristics. Int. Dairy J. 14:455–464.

Pinho, O., E. Mendes, M. M. Alves, and I. M. P. L. O. Ferreira. 2004a. Chemical, physical and sensorial characteristics of "Terrincho" ewe cheese. Changes during ripening and intravarietal comparison. J. Dairy Sci. 87:249–257.[Abstract/Free Full Text]

Pinho, O., A. I. E. Pintado, A. M. P. Gomes, M. M. E. Pintado, F. X. Malcata, and I.M.P.L.V.O. Ferreira. 2004c. Interrelationships among microbiological, physicochemical, and biochemical properties of terrincho cheese, with emphasis on biogenic amines. J. Food Prot. 67:2779–2785.[Medline]

Prieto, B., I. Franco, J. M. Fresno, J. G. Prieto, A. Bernardo, and J. Carballo. 2004. Effect of ripening time and type of rennet (farmhouse rennet from kid or commercial calf) on proteolysis during ripening of León cow milk cheese. Food Chem. 85:389–398.

Pripp, A. H., L. Stepaniak, and T. Sorhaug. 2000. Chemometrical analysis of proteolytic profiles during cheeses ripening. Int. Dairy J. 10:249–253.

Reid, J. R., C. H. Moore, G. G. Midwinter, and G. G. Pritchard. 1991. Action of a cell wall proteinase from Lactococcus lactis subsp. cremoris SK11 on bovine _S1-casein Appl. Microbiol. Biotechnol. 35:222–227.[Medline]

Restani, P., T. Velonà, A. Carpen, M. Duranti, and C. L. Galli. 1996a. -Casein as a marker of ripening and/or quality of Grana Padano cheese. J. Agric. Food Chem. 44:2026–2029.

Santamaría, G., M. Ramos, and J. A. Ordonez. 1986. Application of linear-discriminant analysis to different proteolysis parameters for assessing the ripening of Manchego cheese. Food Chem. 19:225–234.

Singh, T. K., P. F. Fox, P. Hojrup, and A. Healy. 1994. A scheme for the fractionation of cheese nitrogen and identification of principal peptides. Int. Dairy J. 4:111–122.

Sousa, M. J., Y. Ardo, and P. L. H. McSweeney. 2001. Advances in the study of proteolysis during cheese ripening. Int. Dairy J. 11:327–345.

Trujillo, A. J., C. Isidre, and G. Buenaventura. 2000. Analysis of major ovine milk proteins by reversed-phase high-performance liquid chromatography and flow injection analysis with electrosp-ray ionization mass spectrometry. J. Chromatogr. A 870:371–380.[Medline]

Veloso, A. C. A., N. Teixeira, I. M. P. L. V. O. Ferreira, and M. A. Ferreira. 2002a. Detecção de adulterações em produtos ali-mentares contendo leite e/ou proteínas lácteas. Quim. Nova 25:609–615.

Veloso, A. C. A., N. Teixeira, and I. M. P. L. V. O. Ferreira. 2002b. Separation and quantification of the major caseins fractions by reverse-phase high performance liquid chromatography and urea-polyacrylamide gel electrophoresis—Detection of milk adulterations. J. Chromatogr. A. 967:209–218.[Medline]

Veloso, A. C. A., N. Teixeira, A. M. Peres, A. Mendonça, and I. M. P. L. V. O. Ferreira. 2004. Evaluation of cheese authenticity and proteolysis by HPLC and urea-polyacrylamide gel electrophoresis. Food Chem. 87:289–295.

Visser, S., K. J. Slangen, and H. S. Rollema. 1986. High-performance liquid chromatography of bovine caseins with the application of various stationary phases. Milchwissenschaft 41:559–562.

This Article Abstract Full Text (PDF) Interpretive Summary Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ferreira, I. M. P. L. V. O. Articles by Mendonça, A. Search for Related Content PubMed PubMed Citation Articles by Ferreira, I. M. P. L. V. O. Articles by Mendonça, A.