Free Amino Acids and Water-Soluble Nitrogen As Ripening Indices in Montasio Cheese N Innocente

Free amino acids and water-soluble nitrogen as ripening indices in Montasio cheese N Innocente

To cite this version:

N Innocente. Free amino acids and water-soluble nitrogen as ripening indices in Montasio cheese. Le Lait, INRA Editions, 1997, 77 (3), pp.359-369. hal-00929531

HAL Id: hal-00929531 https://hal.archives-ouvertes.fr/hal-00929531 Submitted on 1 Jan 1997

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Lait (1997) 77, 359-369 359 © ElsevierlINRA

Original article

Free amino acids and water-soluble nitrogen as ripening indices in Montasio cheese*

N Innocente

Dipartimento di Scienze degli Alimenti, Università di Udine, 33100 Udine, Italy

Summary - Traditionally, in semi-hard and hard cheeses the development of proteolysis is fol- lowed by the increase of water-soluble nitrogen (SN) as a percentage of total nitrogen (TN), but the characterisation of proteolysis for each particular cheese must con sider ail the processes that lead to the formation of peptides, free amino acids and products of their catabolism. In Montasio cheese, a considerable variability in the SN values and the amino acid profiles of cheeses at the same stage of ripening was recorded. This variability was probably due to the different traditional technologies adopted. Therefore, no chemometric model could be set up to evaluate the typical nature of Monta- sio cheese. On the other han d, a relationship between the total free amino acid content and the SNrrN ratio was demonstrated and it was found that certain amino acids cou Id be used as an index for determining the level of proteolysis reached by the Montasio chee se. proteolysis / free amino acid / total nitrogen / water-soluble nitrogen / Montasio cheese

Résumé - Les acides aminés libres et l'azote soluble dans l'eau pour la caractérisation du fromage Montasio. Traditionnellement, dans les fromages à pâte dure et semi-dure, le développe- ment de la protéolyse est suivie d'une augmentation du rapport en pour cent entre l'azote soluble dans l'eau et l'azote total. Toutefois, pour caractériser la protéolyse dans chaque fromage particulier, il faut considérer l'ensemble des processus qui mène à la formation de peptides, acides aminés libres et produits de leur catabolisme. Dans le fromage Montasio, on a remarqué au même stade de maturation une forte variabilité des valeurs d'azote soluble dans l'eau, ainsi que des profils d'acides aminés. Cette variabilité est probablement due aux différentes technologies traditionnelles. Ainsi, aucun modèle chémométrique ne peut être créé pour déterminer la nature caractéristique du fromage Mon- tasio. D'un autre côté, on peut établir une relation entre le contenu total d'acides aminés et le rapport azote soluble dans l'eau/azote total. Certains acides aminés peuvent être aussi utilisés comme indice pour déterminer le niveau de protéolyse dans le fromage Montasio. protéolyse / acide aminé libre / nitrogène total/azote soluble dans l'eau / fromage Montasio

* Oral communication at the lOF Symposium 'Ripening and Quality of Cheeses', Besançon, France, February 26-28, 1996. 360 N Innocente

INTRODUCTION teolysis is the most commonly used index for maturity. Proteolysis occurs in ail cheese Many Italian cheeses have particular fea- varieties and is considered to be a prereq- tures related to the milk employed, the par- uisite for good f1avour development, so that ticular technology applied and the geo- attention has been focussed on the parame- graphical environment. These typical ters that quantify and characterise proteolysis characteristics allow such chee ses to be (McSweeney and Fox, 1993). Traditionally, legally designated as 'DOC' (controlled proteolysis is monitored by the increase in denomination of origin) if they conform water-soluble nitrogen (SN) as a percent- with the local milk production regulations, age of total nitrogen (TN), but the charac- the traditional cheese manufacturing tech- terisation of proteolysis for each particular niques and the cheese ripening process. Nev- chee se must consider ail the processes that ertheless, to ensure that these regulations lead to the formation of peptides, free amino and transformations are effective in prac- acids and the products of their catabolism tice, it is necessary to define methods for (Creamer, 1970). In fact, the first level of evaluating the qualitative characteristics of proteolysis results from enzymatic action typical cheeses (Corradini and Innocente, on curdled milk and the consequent release 1995). The organoleptic qualities of chee se of amino acids, but later the se amino acids are generally determined by the physical are further broken down by enzymes and che mi cal changes that occur during involved in deamination, decarboxylation, ripening. In fact, the basic composition and transamination and/or demethiolation structure of cheese are determined by the (Hemme et al, 1982; Bech, 1992). The curd-producing operations, but it is during amino acids that are subjected to a more ripening that the individuality and unique rapid degradation are methionine (Strecker characteristics of each chee se variety degradation) and tyrosine (degraded to tyra- develop, and these are influenced by the mine by tyrosine decarboxylase, produced composition of the curd and other factors, by Streptococcus faecalis and only inacti- such as the microflora established during vated by the pasteurisation of the milk), manufacture (Fox et al, 1993). The bio- while catabolic products are œ-aminobu- chemical events involved in cheese ripen- tyric acid and y-aminobutyric acid, formed ing are traditionally assessed from a sen- by decarboxylation of glutamic acid, and sory evaluation of f1avour, body, texture and ornithine, which is derived from the enzy- typical eyes, finish and overall quality by matie degradation of arginine (Hemme et experienced judges or trained consumer pan- al, 1982). These reactions may be introduc- els (Lavanchy et al, 1994). However, sen- tory to the development of important and sory analyses are subjective, and although at specifie f1avours (Griffith and Hammond, present the best index of consumer accept- 1989). Thus, the typical nature of cheese ability, they provide data that are difficult can be evaluated by means of the relative to evaluate scientifically and compare content of certain amino acids, and their between laboratories or studies (Boume et al, increase and/or degradation or formation as 1975). Therefore, chemical and physical products of amino acid catabolism (Reiter et analyses of cheese are used to objectively al, 1969; Weaver et al, 1978; Aston et al, monitor ripening and to assess quality, usu- 1983; Amantea et al, 1986). This is partie- ally to complement sensory evaluation. ularly important in cheeses with character- Although it may never be possible to assess istic eyes, such as Montasio. Montasio is a cheese quality by chemical criteria alone, semi-hard cheese cooked at 44-46 "C, pro- they can provide valuable indices of quality. duced from unpasteurised (raw or thermised) For most hard and semi-hard cheeses, pro- milk in Friuli Venezia-Giulia and the Veneto Ripening indices in Montasio chee se 361

(north-east Italy) and is consumed as a table Reagents cheese after at least 2 months of ripening, whereas 8-9-month-old cheese is used for The amino acid standards used were from Sigma grating. One of the characteristics of Mon- (L-amino acid kit). The calibration standard solu- tasio cheese is the typical eyes, which tion contained alanine, a-aminobutyric acid (Aaba) and y-aminobutyric acid (Gaba), alanine depend on the use of natural milk culture in (Ala), glycine (Gly), valine (Val), threonine (Thr), which the heterofermentative microorgan- leucine (Leu), isoleucine (Ile), serine (Ser), pro- isms are not excessively dominant over the line (Pro), methionine (Met), aspartic acid + homofermentative microbes. However, at asparagine (Asx), phenylalanine (Phe), glutamic present, to limit the development of defects acid + glutamine (Glx), tyrosine (Tyr), omithine commercial starters can also be used sorne- (Om), lysine (Lys) and homoserine (Hom) (inter- naI standard) in 7% ammonia. The final amino times combined with thermal treatments of acid concentration was 1 mg/mL. Other reagents the milk. Consequently, it is possible that of analytical grade were from Carlo Erba the characteristic eyes and the typical taste (ethanol, 28% ammonia solution, isobutanol, sul- and f1avour could be excessively changed. phuric acid, sodium hydroxide pellets), Fluka (heptafluorobutyryl anhydride, > 99% purity, a To evaluate the typical nature of Monta- gas chromatography reagent) Aldrich-Chemie sio cheese by chemical parameters, prote- (Dowex 50W x 8 100-200 mesh) and Janssen olysis during ripening and the free amino (methylene chloride). acid composition in ripened cheese were determined. A high correlation between the intensity of f1avour and the concentration Methods of free amino acids had already been demonstrated in sorne Italian cheeses (Resmini et Determination of free amino acid al, 1993; Pellegrino et al, 1995). The quantitative analysis of amino acids, as their N-heptafluorobutyryl iso-but yi ester derivatives, MATERIALS AND METHODS was achieved by high resolution gas chromatography using a Varian Model 3700 instru- ment equipped with a HD detector, a Hewlett- Materials Packard 3396 A integrator and a 25 m x 0.32 mm id fused silica capillary column coated with a Cheese samples 0.2-llm thick OV 170 1 film. The carrier gas was helium with a column flow rate of 1.6 mUmin at 80 "C. Both inje ctor and detector temperature The first phase of the experimental work involved were 280 oc. The column temperature ranged the analysis of cheeses from four different cheese- from 80 to 280 -c, with an 8 OC/min rate. The makers located in the typical production zone. cheese samples were prepared, purified and At the time of sampling, a sensory evaluation derivatised as described by Bertacco et al (1992). forrn was filled out, which included a graphie representation of the arrangement of the typical holes. Later, the cheeses from a single producer Determination of nitrogen content were considered. These were manufactured by the genuine traditional method, the only variable The water-soluble extract was prepared by grind- being the different storage temperature for mat- ing 4 g grated cheese in a mortar with water at uration, thereby demonstrating the influence of 50 "C until a homogeneous sample was obtained, this processing parameter on the proteolytic trend and the paste then diluted to 100 mL (Stadhouders, during maturation. In particular, several cheeses 1960). The homogenate was filtered and anal- were stored in the traditional manner in ware- ysed for nitrogen by the Kjeldahl method hou ses at 11°C, while others were stored for 15, (AOAC, 1980). The TN was deterrnined by plac- 30 or 60 days in warehouses at a controlled tem- ing 0.5 g grated cheese directly in a Kjeldahl perature of 5 "C and only later taken into tradi- digestion flask. The protein content was obtained tional maturation cells. by multiplying the percentage ofTN by 6.38 and 362 N Innocente

I:-Nool:- NI:-lO 000l<'11.01.OC'j OIo"foiC'ÏooC'Ïr...:C'Ïr...:v:3"fC'ÏoiC'Ïoo<'1 00 00000-OIN- N ool.OooO~ N 'T-<'1 <'1NI:-N <'1

~00 o N

N 0000~1:-1:-00'T C'ÏoC'Ïvioioicioi<'"i ("f')- --C"1tn- 1.O~001:- 'TOIN<'1oo~'T~ ~~~ ~ ~~~oogv:3~r...:~~~~0'1~~- 'T 0'1 ]' <'")~ 0;C'jV:<'1~0;~C'j00~1:-'T -O~-O("f')("f')("f')N~~V)-~~~O 1.0 V) V)~NNV)~~ 1:- ~"" --oo~~_~ o ~ - -- "" 00 'T 1.0 ~ <'1 ~ ~ 'OllON 1:- -: 00 c:; <'1 è'C'l ~O~'T~'T~~~C'Ï"fS~N~~~ ~ N N-NN N <'1 ~ 00<'1<'1 ~~I.O~~'T 'T~ oio~cioir...:<'1oo--cil:-l:-"foi 1.0 OI~-'TOO<'1I.ON<'1-OINN ~~ - - N - ~ 1.0 <'10;00 1:-'T<'1<'1oo ooN~ NOIO oi C'Ï~~~~I:-~~v:3~~~~~g - ]' ~~<'1~<'1ooOlC'j~ OII.O~I.ON~<'1 ~~~~S~~~~~~~8~~~~ ~"" N --""Ï"-..q- ('f')C'f')V)C""l-V) ~ C'l è'C'l ~ <'") <'10;~<'")0000-:~'TV:~'TV:~'T ~O~M~~~~~~~~~~~~~- -

00 1:- 0'11:- 1.0<'100<'") 0I01~ ~o~Sg~~~~<'"i~~~~~~~ .- N---V1-C'f') MNOO--tn Ripening indices in Montasio cheese 363 the ratio of SNrrN was used as an index of pro- Table II. Normalityof distributionof free amino teolysis. acids: values assessed by the Kolmogorov- Smimov test. Estimation de la normalité de la répartition des Statistics valeurs en acides aminés libres avec le test de Kolmogorov-Smirnov. Calculation of mean and median values, interquartile range, Kolmogorov-Smimov nor- Amino Kolmogorov-Smirnov mality test, standard deviation (SO), coefficient P-level of variation (CV), linear and second-order poly- acids test nomial curve fitting, coefficientof regression (r) and coefficient of determination (r2) was per- Ala 0.5137 0.9545 formed using the statistical package SPSS for Aaba 1.5510 0.0160 Windows, release 6.0. Gly 0.7899 0.5606 Val 0.6820 0.7410 Thr 0.7756 0.5843 RESULTS Ile 0.9120 0.3764 Leu 0.6439 0.8014 Fifteen samples of 60-day-old Montasio Ser 0.5898 0.8775 cheese from four different cheese makers Pro 0.7156 0.6851 were analysed first. The working conditions Gaba 1.3418 0.0546 adopted (Bertacco et al, 1992) allowed the Met 0.0899 0.3933 identification of the following amino acids: Asx 0.6964 0.7173 Aaba, Ala, G1y, Val, Thr, Leu, Ile, Ser, Pro, Phe 0.6137 0.8457 Gaba, Met, Asx, Phe, G1x, Tyr, Orn and 0.9634 0.3113 Lys, while the internai standard was G1x 0.9933 0.2773 homoserine (Hom). Table 1 reports the free Tyr amino acid composition of the samples am 1.1249 0.1591 under examination expressed in mg/l 00 g Lys 1.0275 0.2416 cheese dry matter. It can be seen that the Total 0.5592 0.9133 free amino acid profile was very variable. Before the statistical analysis, the verifica- tion of the distribution of the data for the individual amino acids was performed by very high. The considerable variability the Kolmogorov-Smirnov normality test: recorded in the amino acid profiles of the the comparison between the experimental 60-day-old chee ses prevented the setting up and theoretical distribution based on the of a chemometric model for Montasio, as assumption of normality was highly signif- has been done using the same techniques icant, with a high P level (table II). In for other typical Italian chee ses (Resmini et table IIIthe statistical parameters of the indi- al, 1988). vidual amino acids determined in cheese of the same age but from four different pro- The amino acid profile expressed in terms ducers are reported. It can clearly be seen of relative percentage (g amino acid/g total that, even without considering the amino free amino acids) showed less variability. acids that are easily degraded during the ln this case, the coefficient of variation was course of ripening (methionine and tyro- on average lower (see table IV) if those sine) or that are released following catabolic amino acids subject to degradation and those processes at the expense of other amino of catabolic origin were excluded. The rel- acids (ornithine, œ-aminobutyric acid and ative content of the individual amino acids y-arninobutyric acid), the variation remained does not reflect the intensity of proteolysis 364 N Innocente

Table III. Statistical results of the free amino acid content (mgllOO g dry matter), determined in cheeses of the same age but produced in four different dairies (n = 15). Résultats statistiques de la teneur en acides aminés libres (mg/IOO g de matière sèche) de fromages du même âge mais provenant de quatre laiteries différentes.

Amino acids Min Max Median Mean IQR SD CV

Ala 9.35 223.1 69.50 76.68 93.65 56.45 73.62 Aaba 0.00 11.18 0.00 1.34 1.40 3.01 223.61 Gly 12.78 393.79 95.77 145.09 159.00 107.90 74.36 Val 15.82 110.62 50.34 62.38 58.43 31.79 50.97 Thr 12.75 110.33 43.12 52.64 63.30 34.34 65.24 Ile 16.39 176.74 47.13 78.80 110.61 57.66 73.17 Leu 102.31 579.01 240.52 300.59 359.12 170.90 56.85 Ser 7.32 173.21 73.48 77.28 73.59 47.86 61.93 Pro 39.84 426.45 161.59 196.39 241.61 127.33 64.83 Gaba 187.11 38.93 19.41 39.47 38.93 54.21 137.34 Met 4.55 98.96 29.87 32.00 26.57 24.99 78.08 Asx 351.28 191.29 159.10 192.25 191.29 93.78 48.78 Phe 57.52 303.37 140.40 159.71 188.61 87.93 55.05 G1x 123.44 839.89 237.60 369.56 337.86 223.94 60.60 Tyr 19.74 159.52 50.04 64.37 78.65 41.62 64.66 Om 29.13 282.60 86.58 127.30 198.08 91.45 71.83 Lys 70.49 520.04 124.63 225.24 248.70 157.08 69.74 Total 723.20 4210 1976.50 2204.89 2604.35 1238.70 56.18

IQR: interquartile range; SD: standard deviation; CV: coefficient of variation.

Table IV. Statistical results of the free amino acid content, expressed as the relative percentage of total amino acid content determined in cheeses of the same age but produced in four different dairies. Résultats statistiques exprimés en pourcentage relatif de teneur totale en acides aminés, de la teneur en acides aminés libres defromages de même âge mais provenant de quatre laiteries différentes.

Amino acids Mill Max Median Mean IQR SD CV

Ala 1.26 6.30 3.17 3.40 2.18 1.690 49.936 Aaba 0.00 0.34 0.00 0.06 0.17 0.109 178.039 Gly 1.77 11.12 5.42 5.99 1.94 2.383 39.784 Val 2.16 4.56 2.98 2.97 1.25 0.751 25.295 Thr 1.42 2.73 2.47 2.25 0.74 0.375 16.633 Ile 2.19 4.99 3.02 3.19 1.60 0.906 28.411 Leu 11.73 16.09 13.63 13.59 2.24 1.419 10.441 Ser 1.01 5.3 3.44 3.36 1.15 0.944 28.096 Pro 5.51 11.93 8.01 8.34 2.86 1.757 21.068 Gaba 0.00 5.96 1.55 1.90 4.02 1.914 100.557 Met 0.24 7.86 1.45 1.79 1.30 1.759 98.016 Asx 7.53 12.34 8.75 9.26 2.59 1.416 15.299 Phe 6.22 8.92 7.27 7.36 0.83 0.653 8.878 G1x Il.67 21.16 16.25 17.06 4.88 3.641 21.341 Tyr 1.51 4.14 2.92 2.98 1.60 0.820 27.503 Om 1.39 14.39 5.60 6.30 5.93 3.453 54.801 Lys 5.05 12.89 Il.41 9.99 3.96 2.527 25.285

For abbreviations see table III. Ripening indices in Montasio cheese 365

Table V. Statistical results conceming the SNrrN % ratio, calculated from the analyses performed on 98 samples of Montasio cheese from four different chee se producers. Résultats statistiques du rapport SNI7N calculé à partir d'analyses effectuées sur 98 échantillons de Montasio provenant de quatre producteurs différents.

Dairy n Min Max Mean Median IQR SD CV

1 26 18.07 43.28 26.55 24.73 11.23 7.32 27.57 2 21 14.49 40.91 25.07 23.38 9.01 7.28 29.06 3 23 13.93 34.71 21.36 17.99 10.44 5.96 27.90 4 28 14.07 24.82 18.65 18.00 3.29 2.53 13.56

Total L 13.93 43.28 22.73 20.81 8.54 6.68 29.41

For abbreviations see table III

as much as it indicates several qualitative On the basis of the se considerations, we aspects and can thus suppl y very useful decided to evaluate the relationship between information on the characterisation of the the total free amino acid contentffN ratio cheese, particularly if the amino acids used to the SNrrN ratio, and to extend analysis to are those with the lowest interquartile ranges cheeses at different stages of ripening. In such as Phe, Leu, Thr and Asx. particular, it is known that storage at low temperatures for the first few months of In contrast, the ratio of SNrrN was cho- ripening, although an ideal means of obtain- sen as a parameter to indicate the progress of ing good control over swelling, generally ripening. Table V reports the maximum and slows down the biochemical ripening phe- minimum values, means, medians, interquar- nomena, which become more marked the tile ranges, standard deviations and coeffi- longer the cheese is stored at low tempera- cients of variation of the SNrrN ratio cal- ture (Aston et al, 1985). This was exploited culated from the analyses of cheeses (a total to obtain chee ses with different degrees of of 98 samples) produced by the four cheese ripening, in which the SNrrN ratio and the makers considered in this study, represen- ami no acid profile were determined after tative of about 25% of the entire annual Montasio production. Table VI presents an assessment of the normal distribution of the above-mentioned ratio. A high coefficient Table VI. Normality of distribution of the SNfTN ratio assessed by the Kolmogorov-Smimov test. of variation can be seen both within and Estimation de la normalité de répartition du rap- between cheese producers. The variability in port SNITN avec le test de Kolmogorov-Smir- the SN value and the amino acid profile of nov. cheeses at the same stage of ripening thus appeared to be related to a different evolu- Dairy n Kolmogorov-Smirnov tion of the proteolytic processes, probably test P-level due to the different processing conditions adopted by the various chee se producers 1 26 0.8565 0.4455 (temperature of ripening cells, type of starter 2 21 0.8688 0.5965 culture, thermal treatment of the milk) or 3 23 1.2844 0.7338 due to unexpectedly abnormal conditions 4 28 0.9850 0.2864 in the biological ripening agents that are the Total L 1.4550 0.0290 basis of the proteolytic process. 366 N Innocente

Table VII. Regression coefficients between the individual amino acids and the SN/TN ratio. (y = individual amino acids; x = SNrrN). Coefficients de régression entre les acides ami- nés individuels et le rapport SNIlN (y = acides aminés individuels; x = SNIlN).

2 1 -+-~-+---+-I -+--< Amino acids y=ax+b y = ax + bx + c 20 40 60 80 100 120 140 (r) (r) Maturation Ume (day) Ala 0.7205 0.8459 Fig 1. Trend regarding the SNrrN ratio for the Gly 0.8862 0.9352 chee ses matured at three different temperatures Val 0.8914 0.9370 during the early stages of maturation; -0- normal Thr 0.8913 0.9400 maturation; .-6-. conditioned at 5 "C for 30 days; Ile 0.8691 0.9289 -0- conditioned at 5 "C for 60 days. Leu 0.9418 0.9686 Évolution du rapport SNIlN pour des fromages Ser 0.8359 0.8444 affinés à trois températures différentes dans les Pro 0.8922 0.9659 premiers stades de la maturation. Asx 0.9321 0.9642 Phe 0.9288 0.9676 Glx 0.8996 0.9408 different ripening times. In particular, sev- Tyr 0.8213 0.8746 era! cheeses were stored at 5 "C for 15,30 or Lys 0.8938 0.9559 60 days during the early stages of ripening, Met 0.8769 0.9340 while others were maintained at the tradi- Om 0.8658 0.9224 Aaba 0.7305 0.5392 tional ripening temperature (11°C) for the Gaba 0.5549 0.5601 entire ripening period. The evolution of the SNrrN ratio as a function of the maturation time for this cheese is shown in figure 1: it is c1ear that there was a net slowing down of ing. Figure 2 describes the re!ationship the proteolytic process involved in the mat- between the SN/TN ratio and the total free uration of the cheeses conditioned at low amino acid content, expressed as a percent- temperature during the early stages of ripen- age of total free amino acids in total pro-

y = 0.02 x2 - 0.4056x + 3.4836 ,2 = 0.9723 ~ 18 ê 16 '" 14 f Ë y = 0.5138 x - 5.9866 Co 12 ... ,2 = 0.8932 1$ 10 t '".".. 8 c.. 6 ïs 4 J..t .. 1 r. ~ .. S 2 T g 1 0 -'-~-~--+--or=------,--+-~---"-'4~~~~1 1 10 20 30 40 (watcr-solublc Nltotal N)%

Fig 2. Graphie representation of the relationship between total free amino acid content (expressed as a percentage of total protein content) and the SNrrN % ratio. Représentation graphique de la relation entre la teneur totale en acides aminés libres (exprimée en pourcentage de la teneur totale en protéine) et le rapport SNIlN. Ripening indices in Montasio cheese 367 tein. This relationship is weIl described by a with a good degree of certainty as an index second-order polynomial equation, which for determining the level of proteolysis had the best coefficient of determination (stage of ripening) reached by Montasio (,-2) of the various functions evaluated. The cheese independently of its age (in days lack of a linear relationship was due to the from the date of production). The amino fact that release of amino acids was initiaIly acids which, in contrast, did not appear to slower than the increase in the SNffN ratio, have any type of relationship whith the as during the early stages of seasoning the degree of proteolysis (% SNffN) were Aaba proteins are initiaIly hydrolysed to peptides and Gaba (fig 3); these amino acids are Gly of differing molecular weights and only later catabolic products. However, one of the are these peptides further degraded to amino principal characteristics of Montasio cheese, acids. Table VII reports the coefficients of as previously mentioned, is the presence of regression (r) calculated for the individual holes that should have a weIl-defined shape, ami no acids; in this case also, the r values distribution and size. The preliminary data were higher wh en it was a second-order show that the characteristic openings in this equation. The coefficient of regression was cheese, rather than the extent of proteoly- > 0.95 for Leu, Pro, Phe, Asx, Lys and Leu. sis, can be related to the presence of Aaba These amino acids can therefore be used and Gaba (fig 4). Furthermore, the presence

lL~ l":'" 0.45"" !. 0.4 'il= 0.35 0.3 ëe, .. 0.25 j o~;~ ,.z = 0.3137

~O'I~~~~- 0.05 o o 10 15 20 30 40 (Water soluble N!total N)%

0.06

~ 0.05

OB=- 0.04 ,.z = 0.2907 E "- li 0.03 <:> ~ 0.02 ,.z = 0.5336 ...., $. 0.01

10 ~ W ~ ~ ~ ~ (Water soluble N!total N)%

Fig 3. Correlation between the Aaba and Gaba content, expressed as a percentage of the total protein content, and the SNrrN % ratio. Corrélation entre la teneur en Aaha et Gaha (exprimée en pourcentage de la teneur totale en protéine) et le rapport SNII'N. 368 N Innocente

Cheeses Gaba nt Aaba nt Tot FAA (nt) SN nt) ---(-;0) ---(-10) --- -10-(-10 ,.------, TotFAA TotFAA Totprot TN

1 o o 8.67 26.29

0 0 p 0 0 0 0 ~ " " °v 4 ,,0 CV 0 2 0' 0 ~ cPo 0.11 0 2.30 17.05 0 , 0 0; 0 o 0 D 50 o ù ~ 3 Où 0

00 00 l'v f1'0 ~ o G Eh 0° 0 0°00, (}ôO~ClC~OOC' '10 D 'i)"" DO OC"Oc O'b°,p'V 3 o 0 C> 0 0 '" t 0 q:, fi ooco' 0.94 0 4.95 19.90 o cP"" 0 00 tf ,}" 0 0 0 ".C;: 0 ~o o 0 '" °ODl) L Cl e 0 ~ c 0 1> 1) ,/"'". 0 C' o~ 0

Cl c> " ,,0 :-> vo ~ 01 é) e:;, 1.84 1.20 2.44 13.86 4 j C' 0 <:::> cP 0 Ç)' «) a. 00 0 Cl",

Fig 4. Holes and proteolytic parameters in Montasio chee se after 60 days of ripening. FAA: free amino acids; prot: protein; SN: water-soluble nitrogen; TN: total nitrogen. Ouvertures et paramètres de protéolyse de fromages Montasio après 60 j d'affinage. of aminobutyric acids (small amounts of age of a Montasio cheese from its free amino Aaba and a moderate amount of Gaba) has acid content. On the other hand, considerable been reported for Montasio chee se with differences in the degree of proteolysis good sensorial characteristics (Bertacco et al, (expressed as the SNffN ratio) in cheeses of 1994). the same age were noted owing to the variability in several processing parameters, of which the temperature of the ripening cells DISCUSSION AND CONCLUSION appeared to have the greatest influence. In contrast, the usefulness of an analysis The results obtained from the analysis of of free amino acids was confirmed as an data on free amino acid composition showed indicator of the degree of proteolysis in that, owing to the wide range of values cheese, independently of age. In fact, a rela- recorded for samples with the same ripening tionship between the total free amino acid time, it was not possible to determine the content and the SNITN ratio was demon- Ripening indices in Montasio cheese 369

strated. This relationship was not valid for 1 June, 1995, Holland College, Delft, the Nether- the aminobutyric acids, derived from sec- lands ondary catabolic processes: instead, the Creamer LK (1970) Protein breakdown in Gouda cheese. NZ J Dairy Sci Technol S, 152-154 higher or lower concentration of these amino Fox PF, Law J, Mc Sweeney PLH, Wallace J (1993) acids appeared to be related to the forma- Biochemistry of cheese ripening. In: Cheese: Chem- tion of the characteristic eyes in Montasio istry, Physics and Microbiology. Vol /: General cheese. Aspects (Fox PF, ed) Chapman & Hall, London Griffith R, Hammond EG ( 1989) Generation of Swiss cheese t1avour components by the reaction of amino acids with carbon yi compounds. J Dairy Sei 72, ACKNOWLEDGMENT 604-613 Hemme D, Bouillame C, Métro F, Desmazeand MJ The author would Iike to thank C Corradini for (1982) Microbial catabolism of amino acids dur- his valuable suggestions and for critically read- ing chee se ripening. Sei Aliments 2, 113-123 ing the manuscript. Lavanchy P, Bérodier F, Zannoni M, Noël Y, Adamo C, Squella J, Herrero L (1994) Guida per la valu- tazione sensoriale della struttura dei formaggi a pasta dura 0 semidura. INRA, Paris REFERENCES Mc Sweeney PLH, Fox PF (1993) Methods of chemical analyses. In: Cheese: Chemistry, Physics and Amantea GF. Shiva BJ, Nakai S (1986) Culture effect Microbiology. Vol/: General Aspects (Fox PF, ed) on ripening characteristics and rheological Chapman & Hall, London behaviour of Cheddar cheese. J Food Sei 51, 912- Pellegrino L, Hogenboom JA, Pazzaglia C, Tedesco 918 R (1995) Caratterizzazione analitica deI formag- AOAC (1980) Official Methods of Analysis. Assoc gio Fontina sulla base della sua composizione in Official Analytical Chemists, Washington, DC amminoacidi liberi. Latte 10, 1086-1093 Aston JW, Durward IG, Dulley JR (1983) Proteolysis Reiter B, Sorokin Y, Pickering A, Hall AJ (1969) and t1avour development in Cheddar cheese. Aust Hydrolysis of fat and protein in small cheeses made under aseptic conditions. J Dairy Res 36, 75-80 J Dairy Technol38, 55-59 Resmini P, Pellegrino L, Pazzaglia C, Hogenboom JA Aston JW, Giles JE, Durward 1G, Dulley JR (1985) (1985) Gli amminoacidi liberi nella tipizzazione Effect of elevated ripening temperatures on prote- deI formaggio Parmigiano Reggiano, ed in parti- olysis and t1avour development in Cheddar cheese. colare nel prodotto grattugiato. Sei Tecn Latt- J Dairy Res 52, 565-572 Casearia 36, 557-592 Bech AM (1992) Enzymes for the acceleration of Resmini P, Pellegrino L, Hogenboom JA, Bertuccioli cheese ripening. ln: Fe rmentation-produced M (1988) Gli amminoacidi liberi nel formaggio enzymes and accelerated ripening in cheese-mak- Parmigiano Reggiano. In: Ricerca triennale sulla ing. Bull/nt Dairy Fed 269,24-28 composizione e su alcune peculiari caratteristiche Bertacco G, Boschelle 0, Lercker G (1992) Gas chro- dei formaggio Parmigiano Reggiano. Consorzio matographie determination of free amino acids in dei Formaggio Parmigiano Reggiano, Reggio cheese. Mllchwissenschaft 47,348-350 Emilia, 113-121 Bertacco G, Giorno A, Boschelle 0, Corradini C (1994) Resmini P, Hogenboom JA, Pazzaglia C, Pellegrino Carallerizzazione deI formaggio Montasio in L (1993) Gli amminoacidi liberi nella caralleriz- relazione al contenuto in amminoacidi liberi. Nota zazione analitica del formaggio Grana Padano. Sei . II. Sei Tecn Latt-Casearia 45, 421-441 Tecn Latt-Casearia 1,7-19 Boume MC, Sandoval AMR, Villalobus MC, Buckle Stadhouders J (1960) De eiwithjtrolise tijdens de kaas- TS (1975) Training a sensory texture profile panel rijping de enzymen die het eiwit inkaashjdroljseren. and development of standards scales in Columbia. Neth Milk Dairy J 14, 83-90 J Texture Stud 4,204-223 Weaver JC, Kroger M, Thompson MP (1978) Free Corradini C, Innocente N (1995) Evaluation by char- amino acid and rheological measurements on acteristic parameters of the ripening in few typical hydrolysed lactose Cheddar chee se during ripen- cheeses. In: Proc Congr on Cheese Ripening, ing. J Food Sei 43,579-583