Journal of Dairy Science
Volume 89, Issue 11 , Pages 4126-4143, November 2006

Comparison of the Compositional, Microbiological, Biochemical, and Volatile Profile Characteristics of Nine Italian Ewes’ Milk Cheeses

  • R. Coda

      Affiliations

    • Dipartimento di Protezione delle Piante e Microbiologia Applicata, Università degli Studi di Bari, Bari 70126, Italy
    • ,
  • E. Brechany

      Affiliations

    • Hannah Research Institute, Ayr, KA6 5HL Scotland
    • ,
  • M. De Angelis

      Affiliations

    • Dipartimento di Protezione delle Piante e Microbiologia Applicata, Università degli Studi di Bari, Bari 70126, Italy
    • Corresponding Author InformationCorresponding author.
    • ,
  • S. De Candia

      Affiliations

    • Dipartimento di Protezione delle Piante e Microbiologia Applicata, Università degli Studi di Bari, Bari 70126, Italy
    • ,
  • R. Di Cagno

      Affiliations

    • Dipartimento di Protezione delle Piante e Microbiologia Applicata, Università degli Studi di Bari, Bari 70126, Italy
    • ,
  • M. Gobbetti

      Affiliations

    • Dipartimento di Protezione delle Piante e Microbiologia Applicata, Università degli Studi di Bari, Bari 70126, Italy

Received 1 March 2006; accepted 25 May 2006.

Article Outline

Abstract 

Nine Italian ewes’ milk cheeses were compared for compositional, microbiological, biochemical, and volatile profile characteristics. Mean values for the gross composition were rather similar among cheeses. The lowest pH values were found for cheeses that used primary starters. At the end of ripening, cheeses made from raw milk contained >6.0 log10 cfu/g of nonstarter lactic acid bacteria. Several species of lactobacilli were identified, but Lactobacillus plantarum and Lactobacillus paracasei were dominant. Random amplified polymorphic DNA-PCR analysis showed the biodiversity among the strains, and in several cases a relationship with the cheese of provenance. Cheeses differed mainly for secondary proteolysis, as shown by the principal component analysis applied to reversed-phase fast protein liquid chromatography data of the pH 4.6-soluble fractions and by determination of the free AA. A total of 113volatile components were identified in the Italian Pecorino cheeses by solid-phase microextraction coupled with gas chromatography–mass spectrometry analysis. The volatile profiles of the 9 cheeses differed significantly. Quantitatively, alcohols were the most abundant chemical class for some cheeses, whereas ketones were the most abundant for other cheeses. Esters and carboxylic acids were largely found. Specific volatile components seemed to distinguish specific cheeses.

Key words: Pecorino cheese, nonstarter lactic acid bacteria, proteolysis, volatile component

 

Back to Article Outline

Introduction 

Cheese is considered the most diverse group of dairy products. Italy is one of the countries in the world with the largest and most diverse production of cheeses made from cows’, ewes’, goats’, and buffalos’ milks. “Pecorino” is the trivial name given to Italian cheeses made from ewes’ milk. In 2004, the Italian production of cheeses made from ewes’ milk was approximately 89,976 t. Beyond the production of well-known cheeses such as Pecorino Romano, Fiore Sardo, and Canestrato Pugliese (approximately 36,249 t), there is a large variety of Pecorino cheeses (approximately 53,727 t), mostly without a “designation of origin,” which have typical characters and originate from a delimited geographical area (Largo Consumo, 2004; ISTAT, 2001). Pecorino del Tarantino, Pecorino Leccese (Apulia region, southern Italy), Pecorino di Filiano (Basilicata region, southern Italy), Pecorino del Reatino (Lazio region, central Italy), Pecorino Sardo (the only one with a “protected designation of origin,” or PDO status; Sardinia region, southern Italy), Pecorino Umbro (Umbria region, central Italy), Pecorino di Pienza (Tuscany region, central Italy), Pecorino Marchigiano (Marche region, central Italy), and Pecorino Piemontese (Piedmont region, northern Italy) may be considered typical examples of Pecorino cheeses that are produced throughout the Italian territory. They are manufactured in industrial or mostly semi-industrial plants according to local or regional traditions, they have different national and international markets, and they are obtained by technologies that differ in part. Although particular cheeses may be consumed after different periods of ripening, the above-mentioned cheeses belong to the semihard category, are manufactured by using raw milk (except for Pecorino Umbro and Pecorino di Pienza) and rennet paste (except for Pecorino di Pienza), and do not use primary natural or commercial starter cultures (except for Pecorino Sardo and Pecorino Umbro, respectively). As a consequence, adventitious microorganisms, represented mainly by nonstarter lactic acid bacteria (NSLAB) that derive from raw milk (Berthier et al., 2001) or from the dairy environment and surfaces of equipment used in cheese manufacture (Somers et al., 2001), play the most important role in cheese during ripening.

The most famous Italian PDO Pecorino cheeses, for example, Pecorino Romano (Battistotti and Corradini, 1993; Pirisi et al., 2000; Di Cagno et al., 2003), Fiore Sardo (Mannu et al., 2000; Larráyoz et al., 2001), and Canestrato Pugliese (Albenzio et al., 2001; Corbo et al., 2001), have been subjected to studies regarding proteolysis, lipolysis, microbiology, technology, volatile profile analysis, and sensory analysis. The same has been done for the most famous Spanish PDO cheeses made from ewes’ milk such as Manchego (Martinez-Castro et al., 1991; Villasenor et al., 2000), Roncal (Izco and Torre, 2000; Ortigosa et al., 2001), Idiazabal (Pérez Elortondo et al., 1998), and Zamorano (Barron et al., 2004), and for Portuguese PDO cheeses, namely, Serra (Macedo et al., 2004) and Terrincho (Pinho et al., 2003a, 2004a,b). Nevertheless, no studies have been carried out on the Italian Pecorino cheeses mentioned, which deserve interest in terms of either market popularity or their typical features. Comparative studies based on microbiological, compositional, biochemical, and volatile profile characteristics of several cheeses belonging to the same variety may be helpful for 1) differentiating cheeses, 2) establishing the effect of selected technological parameters on specific differences in the microbial flora and related biochemical activities, and, in general, 3) finding the most appropriate characteristics suitable for obtaining a legal “designation of origin,” which may increase the market popularity of individual cheeses.

Consequently, 2 major hypotheses should be investigated: 1) Pecorino cheeses produced in Italy could present differences in microbiological, compositional, biochemical, and volatile profile characteristics, and 2) an integrated characterization could define the major distinguishing traits of Italian Pecorino cheeses. In this study, the microbiological, compositional, biochemical, and volatile profile characteristics of the 9 Italian Pecorino cheeses mentioned were compared.

Back to Article Outline

Materials and Methods 

Cheese Samples 

Nine Italian Pecorino cheeses were considered in this study: Pecorino del Tarantino, Pecorino Leccese, Pecorino di Filiano, Pecorino del Reatino, Pecorino Sardo, Pecorino Umbro, Pecorino di Pienza, Pecorino Marchigiano, and Pecorino Piemontese. The manufacturing protocols are shown in Figure 1. Cheeses were supplied in triplicate (different batches) by local cheese markets and were stored at 4°C for a few hours before analyses. All the analyses were carried out at least in duplicate for each batch of cheese (a total of 6 analyses).

  • View full-size image.
  • Figure 1. 

    Protocols for the manufacture of the 9 Italian Pecorino cheeses. T, Pecorino del Tarantino; L, Pecorino Leccese; F, Pecorino di Filiano; R, Pecorino del Reatino; S, Pecorino Sardo; U, Pecorino Umbro; P, Pecorino di Pienza; M, Pecorino Marchigiano; PI, Pecorino Piemontese.

Enumeration and Isolation of NSLAB 

Samples (20g) of cheeses were diluted in 180mL of a sodium citrate (2% wt/vol) solution and homogenized with a Stomacher Lab-Blender 400 (PBI International, Milan, Italy). Serial dilution were made in quarter-strength Ringer's solution and plated on de Man, Rogosa, and Sharpe (MRS) agar (Oxoid Ltd., Basingstoke, Hampshire, UK) for viable counts. Mesophilic lactobacilli were enumerated after incubation at 30°C for 48 to 72h.

At least 10 colonies, possibly with different morphologies, were isolated from the last plate dilution. Gram-positive, catalase-negative, nonmotile rod isolates were cultivated in MRS broth (Oxoid Ltd.) at 30°C for 24h, and restreaked onto MRS agar. All the isolates considered for further analyses showed the capacity of acidifying the culture medium and grew at 15°C but not at 45°C. Mesophilic microbial cultures were stored at −20°C in 10% (vol/vol) glycerol.

Random Amplified Polymorphic DNA–PCR Analysis 

Genomic DNA were extracted as reported by De Los Reyes-Gavilán et al. (1992), from 2mL of overnight cultures grown in MRS at 30°C. Three primers (Invitrogen Life Technologies, Milan, Italy), with arbitrarily chosen sequences (P4, 5′-CCGCAGCGTT-3′; P7, 5′-AGCAGCGTGG-3′; and M13, 5′-GAGGGTGGCGGT TCT-3′; De Angelis et al., 2001; Rossetti and Giraffa, 2005) were used singly in 3 series of amplification. The reaction mixture contained 200μM of each 2′-deoxynucleoside 5′-triphosphate, 1 to 2μM primer, 1.5 to 3μM MgCl2, 1.25 U of Taq DNA polymerase (Invitrogen), 2.5μL of PCR buffer, 25ng of DNA, and sterile double-distilled water to 25μL. For amplifications with primers P4 and P7, the PCR program comprised 45 cycles of denaturation for 1min at 94°C, annealing for 1min at 35°C, and elongation for 2min at 72°C; the cycles were preceded by denaturation at 94°C for 4min and followed by elongation at 72°C for 5min. For primer M13, amplification reactions were performed according to the protocol described by Giraffa et al. (2000): one cycle at 94°C for 60s (denaturing), 42°C for 20s (annealing), and 72°C for 2min (elongation). Polymerase chain reaction products were separated by electrophoresis (2h at 130 V) on 1.5% (wt/vol) agarose gel (Invitrogen), and the DNA was detected by UV transillumination after staining with ethidium bromide (0.5μg/mL). The molecular weight of the amplified DNA fragments was estimated by comparison with a 1 Kb Plus DNA Ladder (Invitrogen) ranging from 100 to 12,000 bp. For random amplified polymorphic DNA (RAPD) markers, the presence or absence of fragments was recorded as 1 or 0, respectively. Only reproducible well-marked amplified fragments were scored, with faint bands being ignored. Two series of RAPD-PCR profiles were combined to obtain a unique dendrogram. Pairwise comparison of banding patterns was evaluated, with an index of genetic similarity calculated using the simple matching coefficient (Sokal and Michener, 1958).

Genotypic Identification by 16S rRNA Gene Sequence Analysis 

Genomic DNA from each strain was extracted as reported above. Two primer pairs (Invitrogen), LacbF/ LacbR and LpCoF/LpCoR (De Angelis et al., 2006), were used to amplify the 16S rRNA gene fragment of lactobacilli. Fifty microliters of each PCR mixture contained 200μM of each 2′-deoxynucleoside 5′-triphosphate, 1μM of both forward and reverse primer, 2mM MgCl2, 2 U of Taq DNA polymerase (Invitrogen) in the supplied buffer, and approximately 50ng of DNA. The expected amplicons of about 1,400 and 1,000 bp (after amplification with the primers pairs LacbF/LacbR and LpCoF/ LpCoR, respectively) were eluted from gel and purified by the GFX PCR DNA and Gel Band Purification Kit (Amersham Biosciences, Piscataway, NJ). Taxonomic strain identification was performed by comparing the sequences of each isolate with those reported in the Basic BLAST database (Altschul et al., 1997). Strains showing homology of at least 97% were considered to belong to the same species (Goebel and Stackebrandt, 1994).

Compositional Analysis 

Moisture, NaCl, and pH were determined as reported by the International Dairy Federation (IDF 1970, 1988, 1989). Protein and fat were determined by the micro-Kjeldahl method (IDF, 1964) and Soxhlet method using diethyl ether (Gobbetti et al., 1999), respectively. All determinations were carried out on 3 different sections of each cheese and values were averaged.

Assessment of Proteolysis 

The pH 4.6-insoluble and -soluble fractions of the cheeses were obtained as described by Kuchroo and Fox (1982). All fractions of the cheeses were analyzed by urea-PAGE, using an SE 600 electrophoresis unit (Hoefer, Amersham Biosciences) and the stacking gel system described by Andrews (1983). The gels were stained directly by the method of Blakesley and Boezi (1977) with Coomassie brilliant blue G250.

The peptide profiles of the pH 4.6-soluble fractions were determined by reversed-phase fast-protein liquid chromatography (RP-FPLC) using a Resource RPC column and ÄKTA FPLC equipment with a UV detector operating at 214nm (Amersham Biosciences). For each cheese, aliquots (1mL) of the pH 4.6-soluble extracts, containing 1.5 to 3mg of peptides as determined by the o-phthaldialdehyde method (Church et al., 1983), were added to 0.05% (vol/vol) trifluoroacetic acid, and centrifuged at 10,000×g for 10min. The supernatant was filtered through a Millex-HA 0.22-μm pore size filter (Millex-HA, Millipore S.A., Saint Quentin, France) and loaded onto the column. Gradient elution was performed at a flow rate of 1mL/min using a mobile phase composed of water and acetonitrile containing 0.05% trifluoroacetic acid. The CH3CN content was increased linearly from 5 to 46% between 16 and 62min, and from 46 to 100% between 62 and 72min. Total and individual free AA of the pH 4.6-soluble fraction were analyzed by a Biochrom 30 series Amino Acid Analyzer (Biochrom Ltd., Cambridge Science Park, UK) with a Na- cation-exchange column (20×0.46cm i.d.). A mixture of AA of known concentration (Sigma Chemical Co., St. Louis, MO) was added to cysteic acid, methionine sulfoxide, methionine sulfone, Trp, and Orn, and used as the standard. Proteins and peptides in the samples were precipitated by addition of 5% (vol/vol) cold solid sulfosalicylic acid, holding at 4°C for 1h, and centrifuging at 15,000×g for 15min. The supernatant was filtered through a 0.22-μm pore size filter and diluted, when necessary, with sodium citrate (0.2 M, pH 2.2) loading buffer. Amino acids were postcolumn derivatized with ninhydrin reagent and detected by absorbance at 440 (Pro and hydroxyproline) or 570nm (all other AA).

Determination of Volatile Components 

The determination of volatile components was performed by solid-phase microextraction coupled with gas chromatography–mass spectrometry (SPME–GC-MS). Prior to analysis, cheese samples were sliced, frozen in liquid nitrogen, and then pulverized into small granules and stored at −20°C. Three grams of each cheese was then placed in a 15-mL vial and allowed to equilibrate at 40°C for 30min. Extraction of the volatiles was carried out by injecting a carboxen-polydimethylsiloxane fiber into the vial and exposing it to the headspace for 30min at 40°C. Samples were desorbed onto an Agilent FFAP column, 50 m×0.2 mm×0.33μm (Agilent Technologies, Inc., Milan, Italy). The oven was held at 40°C for 2min, then increased at 5°C per minute to 70°C, where it was held for 2min. The temperature was then increased at 10°C per minute to 240°C and held to give a run time of 35min. The mass spectrometer was set to record 33 to 450amu (threshold 1,000) at a sampling rate of 1.11 scans per second. The components were identified and a database was set up to quantify relative amounts of each. The database was constructed using selected ion monitoring as the parameter to determine the amount of each component. The data from the custom report were transferred into Excel (Microsoft Excel, 2003). The data are in the form (area×e5) and are normalized to a weight of 1g of sample. An ANOVA of the area transformed data was carried out using Mini-tab (Minitab Ltd., Brandon Court, UK) on the basis of cheese type and group.

Statistical Analysis 

Data from microbiological and physicochemical analyses were subjected to one-way ANOVA (SAS Institute, 1985 and pairwise comparison of treatment means was achieved by Tukey's procedure at P<0.05, using the statistical software Statistica for Windows (Statistica 6.0 per Windows 1998; StatSoft Italia srl, Padova, Italy). Cluster analysis was conducted on similarity estimates using the unweighted pair group method with arithmetic average, from which a dendrogram representing the relationship between isolates was obtained. Analysis was performed using the statistical software Statistica for Windows (Statistica 6.0 per Windows 1998; StatSoft Italia srl). Peptide profiles of the pH 4.6-soluble fractions of the cheeses were analyzed by using multivariate statistical techniques. Data for the factor reduction analysis were obtained by visually recognizing the peaks and taking peak heights as variables. Factor reduction analysis was performed on the data by the covariance matrix for the determination of principal components (PC; Pripp et al., 1999) using the statistical software Statistica for Windows (Statistica 6.0 per Windows 1998; StatSoft Italia srl).

Back to Article Outline

Results 

Compositional Analysis 

All Pecorino cheeses had moisture values ranging from 35.0 to 38.2% (wt/wt) and were considered as semi-hard cheeses (Table 1). Range values for fat and protein were 27.5 to 29.7, and 26.4 to 27.7% (wt/wt), respectively. These values were in agreement with those recommended by the manufacturers or indicated in the guidelines for cheese manufacture. Large differences were found regarding the pH values 4.68 to 5.80. In particular, Pecorino Sardo and Pecorino Umbro cheeses were characterized by the lowest values, pH 4.68 and 5.05, respectively. The NaCl content was rather uniform in all the cheeses and ranged from 1.5 to 2.3% (wt/wt). No particular differences were found for dry or brine-salted cheeses.

Table 1. Mean values1 for the gross composition parameters of the 9 Italian Pecorino cheeses at the end of ripening
ParameterPecorino del TarantinoPecorino LeccesePecorino di FilianoPecorino del ReatinoPecorino SardoPecorino UmbroPecorino di PienzaPecorino MarchigianoPecorino Piemontese
Moisture, % (wt/wt)37.5±1.038.2±1.436.5±0.835.8±0.535.4±0.636.5±1.135.5±0.936.5±1.035.0±1.3
Fat, % (wt/wt)27.5±0.328.5±0.529.2±0.429.7±0.628.5±0.828.5±0.429.5±0.329.3±0.728.0±0.7
NaCl, % (wt/wt)1.5±0.12.1±0.22.3±0.31.7±0.11.7±0.12.2±0.31.8±0.32.0±0.22.3±0.2
Protein, % (wt/wt)26.8±0.827.3±0.626.4±0.327.2±0.727.7±0.426.5±0.526.5±0.627.5±0.226.8±0.4
pH5.80±0.25.70±0.25.61±0.35.45±0.25.05±0.34.68±0.25.44±0.35.55±0.35.50±0.1
pH 4.6-soluble N:% total N22.1±1.122.4±0.822.5±0.824.2±1.024.0±0.519.0±1.018.4±0.820.0±0.425.0±0.6

1Mean values±standard deviations for 3 batches of each type of cheese, analyzed in duplicate (n=6).

Enumeration and Identification of NSLAB 

Cell numbers of presumptive mesophilic lactobacilli at the end of ripening varied from 3.2 to 8.27 log10 cfu/g (Figure 2). The lowest values were found for Pecorino Umbro and Pecorino di Pienza (3.2 and 5.28 log10 cfu/g, respectively), whereas the other cheeses always had values higher than 6 log10 cfu/g.

  • View full-size image.
  • Figure 2. 

    Cell numbers (log10 cfu/g) of presumptive mesophilic lactobacilli found in the 9 Italian Pecorino cheeses at the end of ripening. Data are the mean of 3 batches of each type of cheese, analyzed in duplicate, and standard deviations are reported.

Gram-positive, catalase-negative, nonmotile rods, growing at 15°C and acidifying, were isolated from the last plate dilution and subjected to preliminary RAPD-PCR analysis by using single primers P4, P7, or M13. Primers M13 and P7 generated zero or a low number of bands for most of the isolates. For this reason, cluster analyses were performed by using RAPD-PCR profiles obtained with P4 and P7, and P4 and M13 (Figure 3, panels A and B, respectively). All isolates were further identified by partial sequencing of the 16S rRNA. The following species were identified for each Pecorino cheese: Pecorino del Tarantino, Lactobacillus plantarum, Lactobacillus casei, and Lactobacillus brevis; Pecorino Leccese, Lb. plantarum and Lactobacillus paracasei; Pecorino di Filiano, Lb. plantarum and Lb. paracasei; Pecorino del Reatino, Lb. plantarum and Lb. paracasei; Pecorino Sardo, Lb. plantarum and Lb. paracasei; Pecorino Umbro, Lb. plantarum and Lb. paracasei; Pecorino di Pienza, Lb. brevis; Pecorino Marchigiano, Lb. plantarum and Lb. brevis; and Pecorino Piemontese, Lb. plantarum. Overall, Lb. plantarum and Lb. paracasei were the species found most frequently, with 56 and 29 isolates over the total of 99 isolates. In most cases, strains of the same species did not group in a unique cluster but were separated into different clusters, depending mainly on the cheese. For instance, Lb. plantarum strains analyzed by primers P4 and P7 (Figure 3, panel A) were separated into 4 clusters at a similarity level of 80%, in which cluster C isolates were from Pecorino di Filiano only, and cluster G and H isolates were from Pecorino del Tarantino only. The same was found (Figure 3, panel B) for clusters C and D, with Lb. plantarum strains isolated only from Pecorino del Reatino and Pecorino Sardo, respectively.

  • View full-size image.
  • Figure 3. 

    Dendrogram obtained by combined random amplification of polymorphic DNA patterns for the isolates from Pecorino cheeses using primer P4 and P7 (A), and P4 and M13 (B). Isolates were numbered based on cheese variety: T, Pecorino del Tarantino; L, Pecorino Leccese; F, Pecorino di Filiano; U, Pecorino Umbro; P, Pecorino di Pienza; M, Pecorino Marchigiano (A); R, Pecorino del Reatino; S, Pecorino Sardo; PI, Pecorino Piemontese (B). Cluster analysis was based on the simple matching coefficient and unweighted pair grouped method with arithmetic average.

Proteolysis 

The level of pH 4.6-soluble nitrogen, expressed as the percentage of total nitrogen in cheese, was in the range of 18.4 to 25.0% (Table 1). The highest values (>22.5%) were found in the order of Pecorino Piemontese>Pecorino del Reatino>Pecorino Sardo>Pecorino Umbro>Pecorino di Filiano. Urea-PAGE of the pH 4.6-insoluble fractions (Figure 4) showed that αs1-CN was completely degraded in almost all the cheeses. β-Casein persisted at the end of ripening and formation of γ-CN was evident, indicating plasmin activity. The urea-PAGE electrophoretograph of the pH 4.6-soluble fractions showed some differences among the cheeses. In particular, Pecorino del Tarantino, Pecorino di Filiano, Pecorino Sardo, and Pecorino Piemontese had characteristic protein bands (data not shown). The pH 4.6-soluble fractions were also analyzed by RP-FPLC (Figure 5). For each of the 9 different RP-FPLC chromatograms, 35 peaks were recognized and matched visually with the Unicorn program (Amersham Biosciences). Quantitative, and especially qualitative, differences were evident for several cheeses. In particular, Pecorino del Tarantino, Pecorino di Filiano, Pecorino di Pienza, and Pecorino Marchigiano were characterized by various peptide peaks distributed throughout the acetonitrile gradient (Figure 5). Principal component analysis (PCA) was applied to RP-FPLC data. The score plot and loading plot of the first and second PC after PCA, based on 35 variables (peaks) of 9 Italian Pecorino cheeses, are shown in Figure 6 (panels A and B). The 2 PC explained 64.17% of the total variance. Pecorino del Reatino, Pecorino Sardo, and Pecorino Piemontese, characterized by the lowest peptide concentrations, were grouped together, whereas the other 6 cheeses occupied separate zones of the plane. The highest concentration of free AA was found for Pecorino Piemontese (51.12mg/g), followed by Pecorino del Reatino (37.15mg/g), Pecorino Sardo (31.00mg/g), and Pecorino di Filiano (26.43mg/g; Table 2). Overall, the AA found at the highest concentrations in all the cheeses were Asp, Glu, Pro, Ile, Leu, Phe, and Lys. In particular, Pecorino del Reatino and Pecorino di Pienza showed, respectively, Thr and His at the highest concentrations, and Pecorino di Filiano and Pecorino Piemontese were characterized by very elevated concentrations of Asp and Glu, respectively.

  • View full-size image.
  • Figure 4. 

    Urea-PAGE of the pH 4.6-insoluble fractions of the 9 Italian Pecorino cheeses at the end of ripening. Lanes: T, Pecorino del Tarantino; L, Pecorino Leccese; F, Pecorino di Filiano; R, Pecorino del Reatino; S, Pecorino Sardo; U, Pecorino Umbro; P, Pecorino di Pienza; M, Pecorino Marchigiano; PI, Pecorino Piemontese; St, ewes’ milk CN.

  • View full-size image.
  • Figure 5. 

    Reversed-phase fast-protein liquid chromatography of the pH 4.6-soluble fractions of 9 Italian Pecorino cheeses at the end of ripening. Chromatograms: T, Pecorino del Tarantino; L, Pecorino Leccese; F, Pecorino di Filiano; R, Pecorino del Reatino; S, Pecorino Sardo; U, Pecorino Umbro; P, Pecorino di Pienza; M, Pecorino Marchigiano; PI, Pecorino Piemontese. mAU = milli-arbitrary units.

  • View full-size image.
  • Figure 6. 

    Score plot (A) and loading plot (B) of the first and second principal components (PC) after PC analysis based on individual peaks obtained from reversed-phase fast-protein liquid chromatograms of the pH 4.6-soluble fractions of 9 Italian Pecorino cheeses at the end of ripening. T, Pecorino del Tarantino; L, Pecorino Leccese; F, Pecorino di Filiano; R, Pecorino del Reatino; S, Pecorino Sardo; U, Pecorino Umbro; P, Pecorino di Pienza; M, Pecorino Marchigiano; PI, Pecorino Piemontese.

Table 2. Concentration (mg/g)1 of individual and total free AA of the 9 Italian Pecorino cheeses at the end of ripening
CheeseAspThrSerGluProGlyAlaCysValMetIleLeuTyrPheHisLysArgTotal
Pecorino del Tarantino4.740.230.263.850.990.180.520.221.100.521.941.790.331.140.261.380.0121.83±0.50
Pecorino Leccese4.140.260.434.620.710.220.420.131.110.531.831.801.971.350.291.23020.96±0.65
Pecorino di Filiano1.550.551.522.762.500.501.360.081.381.081.652.231.441.940.591.992.1226.43±0.87
Pecorino del Reatino2.511.453.963.633.860.801.572.451.591.462.162.521.732.441.252.510.1137.15±1.33
Pecorino Sardo2.980.66005.211.261.680.261.691.572.422.601.812.601.512.760.0431.00±1.12
Pecorino Umbro0.180.110.431.330.330.110.620.141.170.380.421.790.331.280.230.610.1810.58±0.13
Pecorino di Pienza0.200.100.561.040.440.060.210.400.130.250.311.170.450.932.040.580.029.13±0.21
Pecorino Marchigiano10.450.601.411.401.550.390.750.101.180.711.371.790.581.360.321.710.0417.20±0.85
Pecorino Piemontese1.361.032.8821.23.020.871.110.101.751.122.332.950.842.871.632.831.3651.12±1.21

1Mean values±standard deviations for 3 batches of each type of cheese, analyzed in duplicate (n=6).

Volatile Components 

A total of 113volatile components were identified in the Italian Pecorino cheeses and grouped according to chemical classes (Table 3). In most cases, the variability of the volatile components among the 3 batches of each type was not significant (P>0.05). In contrast, except for 18 components, all the volatiles identified differentiated the cheeses with various degrees of statistical significance.

Table 3. Volatile components1 (area×e5/1g of sample) found in the 9 Italian Pecorino cheeses at the end of ripening
Chemical classPecorino del TarantinoPecorino LeccesePecorino di FilianoPecorino del ReatinoPecorino SardoPecorino UmbroPecorino di PienzaPecorino MarchigianoPecorino PiemonteseSignificance
Cheese2Sample3
FPP
Esters
Methyl butanoate0.961.100.970.822.451.380.721.351.66ΔΔ***NS
Methyl octanoate0.840.820.580.832.610.900.600.890.68ΔΔ***NS
Ethyl acetate7.9417.794.683.186.893.372.9625.4217.42ΔΔΔΔ******
Ethyl propanoate2.062.200.000.001.290.210.002.674.09ΔΔΔΔ***NS
Ethyl-2-methyl-propanoate1.511.700.000.000.000.000.000.000.00ΔΔΔ****
Ethyl butanoate11.0514.1011.394.9713.19.083.4616.9722.46ΔΔ****
Ethyl-2-butanoate0.002.070.000.000.000.000.000.000.47ΔΔΔΔ******
Ethyl-2-methyl-butanoate0.580.520.000.000.000.000.000.000.00ΔΔΔΔ******
Ethyl-3-methyl-butanoate0.980.980.000.000.000.000.000.000.00ΔΔΔΔ******
Ethyl pentanoate1.011.090.760.541.090.680.001.442.22ΔΔΔ****
Ethyl hexanoate10.819.549.013.7010.829.542.1612.5614.28ΔΔ****
Ethyl octanoate3.683.703.651.755.413.631.136.765.77Δ***NS
Ethyl decanoate2.882.602.671.632.873.311.174.222.86Δ***NS
Ethyl-9-decenoate0.000.000.000.000.200.530.000.720.65ΔΔ******
Propyl acetate2.480.590.870.003.850.000.001.940.77ΔΔΔ****
Propyl butanoate2.780.961.880.666.250.840.051.431.55ΔΔΔΔ***NS
Propyl hexanoate1.980.580.980.563.670.690.390.600.32ΔΔ***NS
Propyl octanoate0.370.000.220.281.190.000.000.070.00ΔΔ***NS
Butyl acetate1.541.115.741.063.530.620.750.591.40ΔΔ***NS
Butyl butanoate1.560.646.140.773.381.260.310.001.06ΔΔΔ***NS
Butyl hexanoate1.111.032.860.591.180.730.090.000.70ΔΔ****
Butyl octanoate0.281.750.750.000.543.910.000.000.00ΔΔΔ****
3-Methylbutyl acetate3.24.021.740.005.870.000.002.210.00ΔΔΔ******
3-Methylbutyl butanoate1.992.600.460.470.860.580.041.201.53ΔΔ***NS
Diethyl succinate0.000.000.000.000.000.000.001.100.00ΔΔΔΔ******
Total esters50.7871.4955.3518.6375.0541.2613.8382.1479.89
Ketones
2-Propanone8.385.608.3610.138.367.5917.043.338.81ΔΔΔ****
2-Butanone30.565.465.364.6911.648.556.0732.428.68Δ***NS
2-Pentanone11.7221.2713.4524.8520.2310.4837.610.6220.22ΔΔ****
4-Methyl-2-pentanone1.330.000.001.080.910.001.090.000.91Δ******
3-Methyl-2-pentanone1.750.870.001.110.630.000.570.000.90ΔΔ***NS
2-Hexanone1.743.462.624.923.051.556.242.362.96ΔΔ****
2-Heptanone8.8420.5817.4327.0920.0510.6933.1614.1514.54ΔΔ***NS
6-Methyl-5-hepten-2-one0.000.000.000.000.860.000.000.000.00Δ***NS
2-Octanone0.932.343.323.562.741.334.132.201.77Δ***NS
2-Nonanone4.648.6216.9113.6413.435.4516.8011.516.89/***NS
2-Undecanone1.201.412.251.342.040.702.292.101.42/***NS
3-Octanone0.550.480.370.430.510.420.601.040.53/***NS
1-Hydroxy-2-propanone1.340.001.421.660.361.682.060.000.00ΔΔ***NS
3-Hydroxy-2-butanone12.695.055.924.175.806.659.222.214.24ΔΔ******
3-Hydroxy-2-pentanone1.951.071.421.131.401.151.050.620.90ΔΔ***NS
2-Hydroxy-3-pentanone1.810.950.931.071.801.160.980.740.96Δ***NS
Diacetyl9.227.778.787.776.446.9011.270.006.00/***NS
2,3-Pentanedione1.871.130.941.672.131.052.681.280.95/***NS
Acetophenone0.330.370.430.580.460.550.470.480.48/***NS
Phenylacetone0.730.000.000.000.000.000.000.000.00ΔΔ****
Total ketones101.5886.4389.91110.89102.8465.90153.3285.0681.16
Alcohols
Ethanol15.9026.2212.996.8915.7110.7511.6635.2933.14ΔΔΔΔ***NS
2-Propen-1-ol0.000.000.000.000.000.000.004.620.00ΔΔΔΔ******
2-Methyl-1-propanol1.002.650.000.000.910.001.411.871.94ΔΔ***NS
1-Butanol4.463.2210.292.047.411.811.252.894.60Δ***NS
2-Buten-1-ol0.000.000.000.001.750.000.000.660.00ΔΔΔΔ******
2-Methyl-1-butanol2.334.610.520.601.120.851.112.993.06ΔΔΔ****
3-Methyl-1-butanol4.768.790.951.261.750.922.695.115.59ΔΔΔ***NS
3-Methyl-3-buten-1-ol0.931.090.590.781.310.501.100.730.96/***NS
3-Methyl-2-buten-1-ol0.811.050.640.811.210.451.220.731.00Δ***NS
1-Pentanol1.053.001.712.561.641.094.872.371.78ΔΔΔΔ******
1-Hexanol3.122.296.491.683.771.981.513.754.05Δ***NS
2-Ethyl-1-hexanol1.490.670.941.241.390.761.391.341.35/****
1-Octanol0.900.560.700.530.710.490.500.910.79/***NS
2-Propanol2.624.743.681.434.601.681.321.414.35ΔΔ****
2-Butanol10.434.583.711.3710.992.731.1619.612.81ΔΔΔ******
2-Pentanol4.6917.915.807.5010.692.664.226.1613.57ΔΔ***NS
2-Hexanol0.682.700.001.251.590.000.841.261.93ΔΔΔΔ***NS
2-Heptanol3.799.934.295.298.152.104.007.698.30ΔΔ***NS
2-Nonanol1.232.002.312.544.050.682.133.553.09/***NS
3-Ethoxy-1-propanol0.000.000.000.000.000.000.001.360.00ΔΔΔΔ***NS
2-Butoxy-ethanol0.003.300.260.840.930.971.041.001.49Δ******
1-Butoxy-2-propanol0.000.000.001.040.000.000.000.552.07ΔΔ******
Benzyl0.220.390.110.800.450.940.300.761.01Δ***NS
Phenethyl1.351.860.950.750.450.860.852.632.45Δ****
Total alcohols61.76101.5656.9341.2072.5042.9744.57109.2499.33
Aldehydes
2-Propenal0.000.000.000.000.000.000.002.360.00ΔΔΔΔ******
2-Methyl-propanal0.430.400.451.251.170.650.620.000.54ΔΔ*****
Butanal0.150.070.600.930.970.780.530.000.00ΔΔ***NS
2-Butenal0.000.000.001.8212.070.000.390.000.00ΔΔΔΔ***NS
2-Methyl-2-butenal0.000.000.000.602.420.320.000.770.00ΔΔΔ***NS
2-Ethyl-2-butenal0.000.000.000.002.700.000.000.330.00ΔΔΔΔ******
2-Methyl-butanal0.850.650.832.171.790.840.680.241.01ΔΔΔΔ******
3-Methyl-butanal1.751.231.503.071.951.341.890.481.64Δ***NS
Hexanal1.291.281.371.661.701.472.322.701.67/******
Benzaldehyde0.800.801.092.132.061.871.521.070.98/***NS
Phenylacetaldehyde0.900.730.621.250.951.061.010.621.23/***NS
Methional0.000.000.171.350.800.630.140.000.00ΔΔ*****
Total aldhehydes6.175.166.6316.2328.588.969.108.577.07
Lactones
γ-Butyronolactone0.000.000.670.000.000.000.001.571.14ΔΔ***NS
γ-Hexanolactone0.650.981.392.181.141.101.441.001.15Δ Δ***NS
3-Methyl-2(5H)-furanone0.000.000.000.001.390.630.000.000.00Δ Δ***NS
2-Methylfuran0.770.000.000.000.000.000.000.000.00Δ Δ Δ******
Total lactones1.420.982.062.182.531.731.442.572.29
Miscellaneous
Phenol1.500.530.500.420.450.610.390.480.68/******
3-Methyl-phenol1.240.500.560.230.430.000.070.060.07Δ Δ Δ Δ******
Carbon dioxide4.595.535.014.514.775.315.024.834.43/***NS
Methanethiol0.250.290.030.290.180.070.040.000.09/***NS
Carbon disulfide4.582.223.573.072.493.322.230.30.68Δ***NS
Dimethyl sulfide0.760.550.370.710.930.342.380.350.31Δ Δ***NS
Dimethyl disulfide3.031.460.380.710.960.260.810.180.55/******
Dimethyl trisulfide1.451.560.240.470.680.040.360.030.81Δ Δ******
S-Methyl-ethanethioate0.410.500.070.000.000.000.000.000.00/*****
3-Ethyl-thyophene0.000.000.000.001.450.000.000.000.00Δ Δ Δ Δ******
2-Methyltetra-hydrothyophen-3-one0.711.260.000.620.000.280.000.000.00Δ Δ Δ Δ***NS
Dimethyl pyrazine0.570.791.200.830.770.370.560.060.50Δ Δ***NS
Trimethyl pyrazine0.700.921.140.891.881.380.421.090.22Δ******
Tetramethyl pyrazine0.330.000.730.331.122.420.120.000.00Δ Δ***NS
Limonene1.180.611.201.538.151.701.450.531.58Δ Δ****
α-Pinene0.450.570.670.360.810.000.390.001.29Δ Δ***NS
β-Pinene0.000.000.250.001.490.000.800.000.93Δ Δ***NS
Styrene0.800.920.951.471.100.801.620.861.81Δ*****
Cymene0.330.510.670.712.050.610.720.170.75Δ Δ***NS
Benzene0.750.570.400.480.510.470.560.421.06Δ****
Toluene4.523.743.083.154.432.173.332.123.22Δ Δ******
Acetonitrile3.163.866.344.464.344.495.991.502.06Δ Δ******
Chloroform2.933.902.412.692.952.353.561.512.02Δ Δ***NS
1,3-Pentadiene0.350.440.761.136.011.0319.800.000.50Δ Δ Δ Δ*****
Pentyl-nitrate0.000.000.000.000.000.001.170.070.00Δ Δ***NS
Heptane3.033.741.602.742.651.073.451.832.55Δ Δ***NS
Octane2.432.862.291.221.340.661.692.002.06Δ Δ******
2,2,4,6,6-Pentamethyl-heptane4.944.921.270.801.620.891.002.241.42Δ Δ Δ Δ******
Total miscellaneous44.9942.7532.6233.8253.5630.6458.2220.6329.59

1Average values for 3 batches of each type of cheese, analyzed in duplicate (n=6).

2Statistical significance among the 9 Pecorino cheeses.

3Statistical significance among the 3 batches of each type of cheese. F-values are coded as follows: range 50 /; 50 to 99 Δ; 100 to 499 Δ Δ; 500 to 999 Δ Δ Δ; >1,000 Δ Δ Δ Δ. P-values are coded as follows

***P<0.001

**P<0.01

*P<0.05NS (P>0.05).

The largest amounts of total esters were found in Pecorino Marchigiano, Pecorino Piemontese, Pecorino Sardo, and Pecorino Leccese (Table 3). Pecorino del Reatino and Pecorino di Pienza contained one-fourth the amount or less of the esters found in the other cheeses. Butyl, propyl, and especially ethyl esters (ethyl acetate, ethyl butanoate, and ethyl hexanoate) were those characterizing the cheeses. The 3 ethyl-methyl esters identified were characteristic of Pecorino del Tarantino and Pecorino Leccese, which are manufactured in the same region (the Apulia region of southern Italy).

Except in Pecorino Leccese, Pecorino Marchigiano, and Pecorino Piemontese, ketones were the volatile components found at the highest levels (Table 3). 2-Alkanones with odd numbers of carbon atoms were found at the highest levels, showing quantitative differences among the cheeses. In particular, 2-propanone in Pecorino di Pienza; 2-pentanone in Pecorino di Pienza, Pecorino del Reatino, and Pecorino Leccese; and 2-heptanone in Pecorino di Pienza, Pecorino del Reatino, Pecorino Leccese, and Pecorino Sardo were those characterizing the cheeses. In contrast, 2-butanone was found mainly in Pecorino Marchigiano and Pecorino del Tarantino. 3-Hydroxy-2-butanone was found at the highest levels in cheeses (Pecorino del Tarantino and Pecorino di Pienza) in which diacetyl (2,3-butanedione) was also determined at the highest concentration.

Quantitatively, alcohols were the most abundant chemical class of volatiles in Pecorino Marchigiano, Pecorino Leccese, and Pecorino Piemontese (Table 3). Large amounts of ethanol were contained in all the cheeses, especially in these 3. Primary alcohols such as 1-butanol (especially in Pecorino di Filiano and Pecorino Sardo); secondary alcohols such as 2-butanol (especially in Pecorino Marchigiano, Pecorino Sardo, and Pecorino del Tarantino), 2-pentanol (especially in Pecorino Leccese, Pecorino Piemontese, and Pecorino Sardo), and both 2-heptanol and 2-nonanol (in Pecorino Leccese, Pecorino Sardo, and Pecorino Piemontese); and branched-chain alcohols such as both 2-methyl-1-butanol and 3-methyl-1-butanol (in Pecorino Leccese, Pecorino Piemontese, Pecorino Marchigiano, and Pecorino del Tarantino) were identified as having the highest levels of these alcohol components.

Aldehydes were found at low levels in all the Pecorino cheeses (Table 3). Some aldehydes were found only in one cheese (e.g., 2-propenal in Pecorino Marchigiano) or at levels markedly higher in one cheese (e.g., 2-butenal, 2-methyl-2-butenal, and 2-ethyl-2-butenal in Pecorino Sardo) than in the others. 2-Methyl-butanal, 3-methyl-butanal, hexanal, benzaldehyde, and phenyl-acetaldehyde were identified in all cheeses. Lactones seemed to be a very minor fraction of the volatile components of the 9 Italian Pecorino cheeses (Table 3).

Among the miscellaneous components identified were methanethiol, dimethyl sulfide, dimethyl disulfide, and dimethyl trisulfide (the last 2 compounds especially in Pecorino del Tarantino and Pecorino Leccese), and various terpenes, without major differences among the cheeses with the exception of limonene, which was found at a very high level in Pecorino Sardo (Table 3). Pecorino di Pienza showed a level of 1,3-pentadiene approximately 20-fold higher than the other cheeses (except for Pecorino Sardo).

Volatile FFA 

Table 4 shows the profile of carboxylic acids found in the 9 Italian Pecorino cheeses. Nineteen volatile FFA were identified. The highest levels were found in Pecorino Umbro, Pecorino del Tarantino, and Pecorino di Filiano cheeses. Pecorino di Pienza was characterized by the lowest level. Although statistically significant differences were found among the cheeses, the FFA profiles were qualitatively similar. Ethanoic, butanoic, hexanoic, octanoic, and decanoic acids represented the largest proportions of the total of carboxylic acids, which varied from 65.2% (Pecorino del Tarantino) to 82.6% (Pecorino Sardo).

Table 4. Carboxylic acids1 (area×e5/1g of sample) found in the 9 Italian Pecorino cheeses at the end of ripening
Chemical classPecorino del TarantinoPecorino LeccesePecorino di FilianoPecorino del ReatinoPecorino SardoPecorino UmbroPecorino di PienzaPecorino MarchigianoPecorino PiemonteseSignificance
Cheese2Sample3
FPP
Carboxylic acids
Benzoic acid1.091.131.041.040.851.091.160.970.75/*****
Formic acid1.481.761.850.991.553.442.642.521.09Δ Δ***NS
Ethanoic acid24.9025.3422.1914.3320.2526.7711.9320.2116.80Δ Δ***NS
Propanoic acid9.723.142.792.342.782.971.626.493.75Δ Δ Δ***NS
2-Methyl propanoic acid11.215.804.443.481.692.902.781.564.13Δ Δ Δ Δ******
Butanoic acid31.8226.2640.5736.4932.0444.2423.6627.8629.14Δ Δ***NS
2-Methyl-butanoic acid9.143.853.222.661.031.321.090.943.36Δ Δ Δ******
3-Methyl-butanoic acid10.754.835.824.741.243.032.151.957.62Δ Δ*****
3-Methyl-2-butenoic acid0.980.000.000.000.000.000.000.000.00Δ Δ Δ Δ***NS
Pentanoic acid4.312.905.625.813.957.982.493.483.57Δ Δ***NS
Hexanoic acid27.9319.7431.4928.7726.4134.6714.5721.1118.88Δ Δ***NS
5-Hexenoic acid1.280.000.000.001.380.000.000.140.00Δ Δ***NS
2,4-Hexadienoic acid0.000.000.000.000.310.001.120.000.00Δ Δ Δ******
4-Methyl-hexanoic acid0.970.000.000.000.310.000.000.000.00Δ Δ******
Heptanoic acid3.062.133.423.743.374.821.532.191.86Δ Δ***NS
Octanoic acid14.469.5116.9314.2214.8619.737.518.667.22Δ***NS
Nonanoic acid1.851.461.961.651.571.480.981.190.98/***NS
Decanoic acid7.304.456.756.405.739.153.963.883.23/***NS
9-Decenoic acid1.050.531.100.950.831.490.580.600.60/***NS
Total carboxylic acids163.30112.83149.19127.61120.15165.0879.77103.75102.98

1Average value for 3 batches of each type of cheese, analyzed in duplicate (n=6).

2Statistical significance among the 9 Pecorino cheeses.

3Statistical significance among the 3 batches of each type of cheese. F-values are coded as follows: range 50 /; 50 to 99 Δ; 100 to 499 Δ Δ; 500 to 999 Δ Δ Δ; >1,000 Δ Δ Δ Δ. P-values are coded as follows

***P<0.001

**P<0.01; *P<0.05; NS (P>0.05).

Back to Article Outline

Discussion 

Mean values for the gross composition were rather similar for cheeses, except for pH. Cheeses manufactured by using primary starters showed the lowest pH values. The major role of primary starter cultures is to start the production of lactic acid from lactose, which occurs early in the manufacturing phase of cheese. Overall, in the early phase of manufacture NSLAB are present at very low numbers, significantly contributing to the decrease of pH (Berthier et al., 2001).

During cheese ripening, microbiology is frequently characterized by successions of communities. Whereas a large part (≥8.0 log10 cfu/g) of primary starter biomass declines throughout ripening, NSLAB increases from approximately 2.0 log10 cfu/g in hygienically produced raw milk cheeses to ≥6.0 log10 cfu/g in ripened cheese (Berthier et al., 2001). They grow at low temperature; are acid-tolerant; and tolerate the lack of fermentable carbohydrates, low pH, and aw and the presence of bacteriocins, which make the environmental conditions very hostile during ripening. Nonstarter lactic acid bacteria find some components for growth in ripening cheese (e.g., lactate, citrate, glycerol, sugars, AA, and other metabolites; Peterson and Marshall, 1990; Wouters et al., 2002). The major part of the Italian Pecorino cheeses had >6.0 log10 cfu/g of NSLAB at the end of ripening. The only exception concerned Pecorino Umbro and Pecorino di Pienza cheeses, which were manufactured from pasteurized ewes’ milk. Although some lactobacilli strains may survive the heat treatment (Jordan and Cogan, 1999), the majority are inactivated by pasteurization (Turner et al., 1986). However, recent studies have shown that the main source of NSLAB is milk (Buffa et al., 2001). Except for Pecorino di Pienza, which contained low numbers of Lb. brevis, all the other cheeses contained Lb. plantarum, Lb. paracasei, or both as the dominant species of NSLAB. This finding is in agreement with several other studies aimed at characterizing the microbial composition of cheeses made from ewes’ milk (Mannu et al., 2000; De Angelis et al., 2001; Di Cagno et al., 2003; Oneca et al., 2003; Macedo et al., 2004). In spite of the rather low differences at species level, the RAPD-PCR analyses highlighted a large bio-diversity at the strain level. Although the role of NSLAB in the overall cheese quality is still debated because of the unpredictable and dynamic nature of nonstarter lactobacilli influenced by compositional and environmental factors (Lane et al., 1997), the positive contributions to flavor development by peptidase and AA catabolism activities of NSLAB have been reported (Poveda et al., 2002; Wouters et al., 2002).

Primary proteolysis of Italian Pecorino cheeses was characterized by the complete hydrolysis of αs1-CN, which indicated considerable chymosin activity. Unlike other Italian ewes’ milk cheeses (e.g., Pecorino Romano), the manufacture of these Pecorino cheeses did not include cooking of the curd. Under this condition, chymosin activity toward αs1-CN may proceed intensively during ripening. A considerable amount of β-CN persisted at the end of ripening. Overall, chymosin activity on β-CN is lower than that toward αs1-CN, mainly because of hydrophobic interactions between salt and proteins (Fox, 1989). The same primary proteolysis was described during ripening of Canestrato Pugliese ewes’ milk cheese (Albenzio et al., 2001). The PCA applied to chromatogram data of the pH 4.6-soluble fractions (Poveda et al., 2002; Pripp et al., 2000) and the concentrations of free AA showed that secondary proteolysis varied among the Pecorino cheeses. The concentrations of free AA of some cheeses approached those found in other Italian ewes’ milk cheeses ripened for longer times (Di Cagno et al., 2003). Pecorino Umbro and Pecorino di Pienza cheeses, which contained the lowest number of NSLAB, also had the lowest concentrations of free AA. Although the release of some AA seemed to be the highest in specific cheeses, Asp, Glu, Pro, Ile, Leu, Phe, and Lys were found at the highest concentrations in almost all the Pecorino cheeses. These AA are typically released during ripening of several Italian semihard and extra-hard cheese varieties (Resmini et al., 1988; Gobbetti et al., 1999; Albenzio et al., 2001; Di Cagno et al., 2003).

The SPME–GC-MS technique was used to characterize the volatile components of Italian Pecorino cheeses. Compared with other techniques, SPME–GC-MS was preferable because of its reduced sample preparation time, high sensitivity, and limited risk of artifacts caused by the use of solvents (Kataoka et al., 2000; Lord and Pawliszyn, 2000; Pinho et al., 2002). The SPME–GC-MS technique was also used previously to characterize the volatile profiles of Terrincho, Roncal, Pecorino Sardo, and Fiore Sardo cheses (Pinho et al., 2003b; Larráyoz et al., 2001). Quantitatively, alcohols were the most abundant chemical class for Pecorino Marchigiano, Pecorino Leccese, and Pecorino Piemontese, whereas ketones were the most abundant for the other cheeses. Overall, esters were the main volatile components of other Italian and Spanish cheeses also made from ewes’ milk but ripened for longer times (e.g., Canestrato Pugliese, Pecorino Romano, Manchego, Roncal, Castellano; Martinez-Castro et al., 1991; Izco and Torre, 2000; Villasenor et al., 2000; Larráyoz et al., 2001; Di Cagno et al., 2003; Fernàndez-Garcìa et al., 2003). Nevertheless, esters, especially ethyl esters, were also found largely in the Pecorino cheeses. Esters contribute in a synergistic way to the fruity aroma of the cheese because they have a low perception threshold concentration that is 10-fold lower than their alcohol precursors (Preininger and Grosch, 1994). Ethyl hexanoate, which has a distinct odor of unripe apple, and ethyl butanoate, which mainly characterizes cheeses containing the largest amounts of esters, were typically found to increase during the ripening of ewes’ milk cheeses (Mariaca et al., 2001; Di Cagno et al., 2003). Ethyl-2-methyl-butanoate and -propanoate, and ethyl-3-methyl-butanoate were found only in cheeses produced in the Apulia region. 2-Alkanones with odd numbers of carbon atoms were found, with differences in the amounts of individual compounds depending on the Pecorino cheese. Free fatty acids liberated through lipolysis might be catabolized to methyl ketones by microbial activity (Izco and Torre, 2000). Ketones also characterized the profiles of Fiore Sardo and Manchego cheeses when manufactured with raw ewes’ milk (Villasenor et al., 2000; Di Cagno et al., 2003). 2-Pentanone was the most abundant methyl ketone in aged Manchego cheese (Villasenor et al., 2000) and may impart an orange-peel aroma to cheese because of its much lower perception threshold than the other ketones (Arora et al., 1995). 3-Hydroxy-2-butanone (acetoin) was detected at the highest concentrations in Fiore Sardo, Pecorino Romano, Canestrato Pugliese, and Roncal ewes’ milk cheeses (Izco and Torre, 2000; Di Cagno et al., 2003). Because of its low perception threshold (0.12mg/kg), the effect of 3-hydroxy-2-butanone on the aroma of Roncal cheese was considered very important. Acetoin is produced by the reduction of diacetyl (2,3-butanedione) or it may be synthesized from pyruvate, lactose, or citrate by lactic acid bacteria (Crow, 1990). Overall, the strong reducing conditions in cheese may favor the rapid reduction of aldehydes and ketones to primary and secondary alcohols (Molimard and Spinnler, 1996). Although primary alcohols were found mainly in Pecorino di Filiano cheese, secondary and branched-chain alcohols mainly distinguished Pecorino Leccese, Pecorino Piemontese, and Pecorino Sardo cheeses. Manchego cheese showed the largest percentage of primary alcohols, whereas Zamorano cheese showed the largest percentage of branched-chain alcohols (Barron et al., 2004). Alcohols were quantitatively the main chemical family found in the volatile fraction of La Serena and Castellano cheeses (Carbonell et al., 2002; Fernàndez-Garcìa et al., 2003). Secondary alcohols such as 2-pentanol and 2-heptanol may be derived by the reduction of methyl ketones by microbial reductases as a defense mechanism against toxicity (Molimard and Spinnler, 1996). 3-Methyl-1-butanol has a fruity (Karahadian et al., 1985), fusel oil, or whisky odor (Moio and Addeo, 1998) and is responsible for the pleasant aroma of fresh cheese. Methyl-branched alcohols may be derived through the reduction of aldehydes formed via Strecker degradation from AA (Jollivet et al., 1994). Compared with the other volatile components, aldehydes were found at low levels in all the Pecorino cheeses. This was in agreement with the volatile profile of other Italian cheeses such as Pecorino Romano, Fiore Sardo, and Canestrato Pugliese (Di Cagno et al., 2003). The low level of aldehydes indicated an optimal maturation because a higher concentration of aldehydes may cause off-flavors (Moio and Addeo, 1998). Aldehydes are unstable compounds that are reduced to alcohols or oxidized to acids during cheese ripening (Carbonell et al., 2002). Limonene was found at a level approximately 8-fold higher in Pecorino Sardo, and the level of 1,3-pentadiene was markedly higher in Pecorino Sardo and especially in Pecorino di Pienza compared with the other cheeses. Their presence in cheese is probably not related to the ripening process but to the ewes’ diet (Carbonell et al., 2002). Methional, dimethyl disulfide, and dimethyl trisulfide were probably related to the breakdown of the sulfur-containing AA during ripening by microbial enzymes, which produced hydrogen sulfide and methanethiol, which after oxidation may yield the above components. Those compounds are considered indispensable for the characteristic aroma of cheeses such as Cheddar and Emmenthal, but some authors speculate (Izco and Torre, 2000) that they are not particularly important for the aroma of Spanish ewes’ milk cheeses.

The strong, balanced piquant flavor that characterizes Pecorino cheeses is primarily due to the relatively high levels of short-chain FFA. Except for Pecorino di Pienza, which showed the lowest level of FFA, all the other cheeses were manufactured by using rennet paste. It contains the pregastric esterase, which preferentially hydrolyzes fatty acids esterified at the sn-3 position of glycerol (Woo and Lindsay, 1984), where the major portion of the short-chain fatty acids are located. Because most of the cheeses were manufactured without pasteurization, one should not exclude a minor role played by the milk endogenous lipoprotein lipase. Nevertheless, NSLAB, especially when found at high cell numbers, may contribute to lipolysis (Gobbetti et al., 1996, 1997). The same profile of fatty acids was also found in Italian PDO ewes’ milk cheeses (Di Cagno et al., 2003). High levels of ethanoic and butanoic acids were also found in hard and semihard Greek (Kondyli and Katsiari, 2001) and Spanish (Izco and Torre, 2000; Villasenor et al., 2000) ewes’ milk cheeses.

Back to Article Outline

Supplementary data 

Download file

download text

Interpretive summary.

Back to Article Outline

References 

  1. Albenzio M, Corbo MR, Rehman SU, Fox PF, De Angelis M, Corsetti A, et al. Microbiological and biochemical characteristics of Canestrato Pugliese cheese made from raw milk, pasteurized milk or by heating the curd in hot whey. Int. J. Food Microbiol. 2001;67:35–48
  2. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, et al. Gapped BLAST and PSI-BLAST: A new generation of protein database search programs. Nucl. Acids Res. 1997;25:3389–3402
  3. Andrews AT. Proteinases in normal bovine milk and their action on caseins. J. Dairy Res. 1983;50:45–55
  4. Arora G, Cormier F, Lee B. Analysis of odour-active volatiles in Cheddar cheese headspace by multidimensional GC/ MS/sniffing. J. Agric. Food Chem. 1995;43:748–752
  5. Barron LJR, Redeondo Y, Flanagan CE, Pérez Elortondo FJ, Albisu M, Nàjera AI, et al. Comparison of the volatile composition and sensory characteristics of Spanish PDO cheeses manufactured from ewes’ raw milk and animal rennet. Int. Dairy J. 2004;15:371–382
  6. Battistotti B, Corradini C. Cheese: Chemistry, physics and microbiology. In:  Fox PF editors. Italian Cheese. Vol. 2:London, UK: Chapman and Hall; 1993;p. 221–243
  7. Berthier F, Beuvier E, Dasen A, Grappin R. Origin and diversity of mesophilic lactobacilli in Comtè cheese, as revealed by PCR with repetitive and species-specific primers. Int. Dairy J. 2001;11:293–305
  8. Blakesley RW, Boezi JA. A new staining technique for proteins in polyacrylamide gels using Comassie Brilliant Blue G250. Anal. Biochem. 1977;82:580–582
  9. Buffa M, Guamis B, Royo C, Trujillo AJ. Microbiological changes throughout ripening of goat cheese made from raw, pasteurized and high-pressure-treated milk. Food Microbiol. 2001;18:45–51
  10. Carbonell M, Nunez M, Fernández-Garcìa E. Evolution of the volatile components of ewe milk La Serena cheese during ripening. Correlation with flavour characteristics. Lait. 2002;82:683–698
  11. Church FC, Swaisgood HE, Porter DH, Catignani GL. Spectrophotometric assay using o-phthaldialdehyde for determination of proteolysis in milk and isolated milk proteins. J. Dairy Sci. 1983;66:1219–1227
  12. Corbo MR, Albenzio M, De Angelis M, Sevi A, Gobbetti M. Microbiological and biochemical properties of Canestrato Pugliese hard cheese supplemented with bifidobacteria. J. Dairy Sci. 2001;84:551–561
  13. Crow VL. Properties of the 2,3-butanediol dehydrogenase from Lactococcus lactis subsp. lactis in relation to citrate fermentation. Appl. Environ. Microbiol. 1990;56:1656–1662
  14. De Angelis M, Corsetti A, Tosti N, Rossi J, Corbo MR, Gobbetti M. Characterization of non-starter lactic acid bacteria from Italian ewe cheeses based on phenotypic, genotypic and cell wall protein analyses. Appl. Environ. Microbiol. 2001;67:2011–2020
  15. De Angelis M, Siragusa S, Berloco M, Caputo L, Settanni L, Alfonsi G, et al. Selection of potential probiotic lactobacilli from pig feces to be used as additives in pelleted feeding. Res. Microbiol. 2006;doi:10.1016/ j.resmic.2006.05.003
  16. De Los Reyes-Gavilán CGG, Limsowtin KY, Tailliez P, Séchaud L, Accolas JP. A Lactobacillus helveticus-specific DNA probe detects restriction fragment length polymorphisms in this species. Appl. Environ. Microbiol. 1992;58:3429–3432
  17. Di Cagno R, Banks J, Sheehan L, Fox PF, Brechany EY, Corsetti A, et al. Comparison of the microbiological, compositional, biochemical, volatile profile and sensory characteristics of three Italian PDO ewes’ milk cheeses. Int. Dairy J. 2003;13:961–972
  18. Fernàndez-Garcìa E, Gaya P, Medina M, Nuñez M. Evolution of the volatile components of raw ewes’ milk Castellano cheese: Seasonal variation. Int. Dairy J. 2003;14:39–46
  19. Fox PF. Proteolysis during cheese manufacture and ripening. J. Dairy Sci. 1989;72:1379–1383
  20. Giraffa G, Rossetti L, Neviani E. An evaluation of SELEX-based DNA purification protocols for the typing of lactic acid bacteria. J. Microbiol. Meth. 2000;42:175–184
  21. Gobbetti M, Folkertsma B, Fox PF, Corsetti A, Smacchi E, De Angelis M, et al. Microbiology and biochemistry of Fossa (pit) cheese. Int. Dairy J. 1999;9:763–773
  22. Gobbetti M, Fox PF, Smacchi E, Stepaniak L, Damiani P. Purification and characterization of a lipase from Lactobacillus plantarum 2739. J. Food Biochem. 1996;20:227–246
  23. Gobbetti M, Fox PF, Stepaniak L. Isolation and characterization of a tributyrin esterase from Lactobacillus plantarum 2739. J. Dairy Sci. 1997;80:3099–3106
  24. Goebel BM, Stackebrandt E. Cultural and phylogenetic analysis of mixed microbial populations found in natural and commercial bioleaching environments. Appl. Environ. Microbiol. 1994;60:1614–1621
  25. IDF (International Dairy Federation). Determination of the protein content of processed cheeses products. Standard 25. Brussels, Belgium: International Dairy Federation; 1964;
  26. IDF (International Dairy Federation). Determination of dry matter content in whey cheese. Standard 58. Brussels, Belgium: International Dairy Federation; 1970;
  27. IDF (International Dairy Federation). Determination of salt content. Standard 12B. Brussels, Belgium: International Dairy Federation; 1988;
  28. IDF (International Dairy Federation). Determination of pH. Standard 115A. Brussels, Belgium: International Dairy Federation; 1989;
  29. ISTAT (Istituto Nazionale di Statistica). 2001. Raccolta di latte e produzione lattiero-casearia italiana. www.istat.it/agricoltura/agricoltura Accessed Jan. 12, 2006.
  30. Izco JM, Torre P. Characterization of volatile flavour compounds in Roncal cheese extracted by the “purge and trap” method and analyzed by GC-MS. Food Chem. 2000;70:409–417
  31. Jollivet N, Chateaud J, Vayssier Y, Bensoussan M, Belin J. Production of volatile compounds in model milk and cheese media by eight strains of Geotrichum candidum Link. J. Dairy Res. 1994;61:241–248
  32. Jordan KN, Cogan TM. Heat resistance of Lactobacillus spp. isolated from Cheddar cheese. Lett. Appl. Microbiol. 1999;29:136–140
  33. Karahadian CD, Josephson BJ, Lindsay RC. Contribution of Penicillium spp. to the flavours of Brie and Camembert cheese. J. Dairy Sci. 1985;68:1865–1877
  34. Kataoka H, Lord H, Pawliszyn J. Applications of solid-phase microextraction in food analysis. J. Chromatogr. A. 2000;880:35–62
  35. Kondyli E, Katsiari MC. Differences in lipolysis of Greek hard cheeses made from sheep's, goat's, or cow's milk. Milchwissenschaft. 2001;56:444–446
  36. Kuchroo CN, Fox PF. Soluble nitrogen in Cheddar cheese: Comparison of extraction procedures. Milchwissenschaft. 1982;37:331–335
  37. Lane CN, Fox PF, Walsh EM, Folkertsma B, McSweeney PLH. Effect of compositional and environmental factors on the growth of indigenous non-starter lactic acid bacteria in Cheddar cheese. Lait. 1997;77:561–573
  38. Largo Consumo. 2004. Produzione di formaggi DOP in Italia. www.largoconsumo.info Accessed Jan. 1, 2006.
  39. Larráyoz P, Addis M, Gauch R, Bosset JO. Comparison of dynamic headspace and simultaneous distillation extraction techniques used for the analysis of the volatile components in three European PDO ewes’ milk cheeses. Int. Dairy J. 2001;11:911–926
  40. Lord H, Pawliszyn J. Evolution of solid-phase microextraction technology. J. Chromatogr. A. 2000;885:153–193
  41. Macedo AC, Tavares TG, Malcata FX. Influence of native lactic acid bacteria on the microbiological and sensory profiles of Serra da Estrela cheese. Food Microbiol. 2004;21:233–240
  42. Mannu L, Comunian R, Scintu MF. Mesophilic lactobacilli in Fiore Sardo cheese: PCR-identification and evolution during cheese ripening. Int. Dairy J. 2000;10:383–389
  43. Mariaca RG, Fernanàndez-Garcìa E, Mohedano AF, Nunez M. Volatile fraction of ewe's milk semi-hard cheese manufactured with and without the addition of a cysteine proteinase. Food Sci. Technol. 2001;212:239–246
  44. Martinez-Castro I, Sanz J, Amigo L, Ramos M, Martin Alvarez P. Volatile components of Manchego cheese. J. Dairy Res. 1991;58:239–246
  45. Moio L, Addeo F. Grana Padano cheese aroma. J. Dairy Res. 1998;65:317–333
  46. Molimard P, Spinnler HE. Review: Compounds involved in the flavor of surface mold-ripened cheeses: Origins and properties. J. Dairy Sci. 1996;79:169–184
  47. Oneca M, Irigoyen A, Ortigosa M, Torre P. PCR and RAPD identification of Lactobacillus plantarum strains isolated from ovine milk and cheese. Geographical distribution of strains. FEMS Microbiol. Lett. 2003;227:271–277
  48. Ortigosa M, Torre P, Izco M. Effect of pasteurization of ewe's milk and use of a native starter culture on the volatile components and sensory characteristics of Roncal Cheese. J. Dairy Sci. 2001;84:1320–1330
  49. Pérez Elortondo FJ, Aldàmiz Echobarria P, Albisu M, Barcina Y. Indigenous lactic acid bacteria in Idiazàbal ewes’ milk cheese. Int. Dairy J. 1998;8:725–732
  50. Peterson SD, Marshall RT. Non-starter lactobacilli in Cheddar cheese: A review. J. Dairy Sci. 1990;73:1395–1410
  51. Pinho O, Ferriera IMPLVO, Ferreira MA. Solid-phase microextraction in combination with GC/MS for quantification of the major volatile free fatty acids in ewe cheese. Anal. Chem. 2002;74:5199–5204
  52. Pinho O, Ferriera IMPLVO, Ferreira MA. Quantification of short-chain free fatty acids in “Terrincho” ewe cheese: Intravarietal comparison. J. Dairy Sci. 2003;86:3102–3109
  53. Pinho O, Pérés C, Ferriera IMPLVO. Solid-phase microextraction of volatile compounds in “Terrincho” ewe cheese. Comparison of different fibers. J. Chromatogr. A. 2003;1011:1–9
  54. Pinho O, Ferriera IMPLVO, Ferreira M. Discriminate analysis of the volatile fraction from “Terrincho” ewe cheese: Correlation with flavour characteristics. Int. Dairy J. 2004;14:455–464
  55. Pinho O, Pintado AIE, Gomes AMP, Pintado MME, Malacata FX, Ferriera IMPLVO. Interrelationship among microbiological, physicochemical and biochemical properties of Terrincho cheese, with emphasis on biogenic amines. J. Food Prot. 2004;67:2779–2785
  56. Pirisi A, Achilleos C, Jaros D, Noel Y, Rohm H. Rheological characterisation of Protected Denomination of Origin (PDO) ewe's milk cheeses. Milchwissenschaft. 2000;55:257–259
  57. Poveda JM, Sousa MJ, Cabezas L, McSweeney PLH. Preliminary observations on proteolysis in Manchego cheese made with defined-strain starter culture and adjunct starter (Lactobacillus plantarum) or a commercial starter. Int. Dairy J. 2002;13:169–178
  58. Preininger M, Grosch W. Evaluation of key odorants of the neutral volatiles of Emmentaler cheese by the calculation of odour activity values. Lebensm. Wiss. Technol. 1994;27:237–244
  59. Pripp A, Shakeel-Ur-Rehman H, McSweeney PLH, Fox PF. Multivariate statistical analysis of peptide profiles and free amino acids to evaluate effects of single-strain starters on proteolysis in miniature Cheddar-type cheeses. Int. Dairy J. 1999;9:473–479
  60. Pripp A, Shakeel-Ur-Rehman H, McSweeney PLH, Sørhaug T, Fox PF. Comparative study by multivariate statistical analysis of proteolysis in a sodium caseinate solution under cheese-like conditions caused by strains of Lactococcus. Int. Dairy J. 2000;10:25–31
  61. Resmini P, Pellegrino L, Hogenboom J, Bertuccioli M. Atti giornata di studio. Consorzio del Formaggio Parmigiano Reggiano. Italy: Reggio Emilia; 1988;
  62. Rossetti L, Giraffa G. Rapid identification of dairy lactic acid bacteria by M13-generated, RAPD-PCR fingerprint databases. J. Microbiol Meth. 2005;63:135–144
  63. SAS Institute. User's Guide: Statistics, Version. 5 Edition.. Cary, NC: SAS Inst., Inc; 1985;
  64. Sokal RR, Michener CD. A statistical method for evaluating systematic relationships. Univ. Kans. Sci. Bull. 1958;38:1409–1438
  65. Somers EB, Johnson ME, Wong ACL. Development of amino acids and organic acids in Norvegia, influence of milk treatment and adjunct Lactobacillus. J. Dairy Sci. 2001;84:1926–1936
  66. Turner KW, Lawrence RC, Lelievre J. A microbiological specification for milk for aseptic cheese-making. N. Z. J. Dairy Sci. Technol. 1986;15:249–254
  67. Villasenor MJ, Valero E, Sanz J, Martìnez-Castro I. Analysis of volatile components of Manchego cheese by dynamic headspace followed by automatic thermal desorption-GC-MS. Milchwissenschaft. 2000;55:378–382
  68. Woo A, Lindsay RC. Concentration of major free fatty acids and flavor development in Italian cheese varieties. J. Dairy Sci. 1984;67:960–968
  69. Wouters TM, Ayad EHE, Hugenholtz J, Smith G. Microbes from raw milk for fermented dairy products. Int. Dairy J. 2002;12:91–109

PII: S0022-0302(06)72458-4

doi:10.3168/jds.S0022-0302(06)72458-4

Journal of Dairy Science
Volume 89, Issue 11 , Pages 4126-4143, November 2006

Article Tools

* Image Usage & Resolution