Chemical characterization of ‘Pecorino Di Farindola’ cheese during ripening

Serena Niro, Alessandra Fratianni*, Annacristina D’Agostino, Ivan Notardonato, Gianfranco Panfili

Dipartimento di Agricoltura, Ambiente e Alimenti, Università degli Studi del Molise, via F. De Sanctis, 86100 Campobasso, Italy


This study evaluated the nutritional and sensorial characteristics of Pecorino di Farindola cheese at different commercial ripening stages. Moreover, in order to assess effectively the peculiar features of this product, the evolution of proteolysis and lipolysis, together with that of free amino acids (FAAs), was studied throughout ripening. A marked proteolysis of Pecorino di Farindola was found. At the end of ripening, FAAs with the highest content were glutamic acid, valine, leucine and lysine. Long-ripened cheeses had a light spicy feature that distinguishes them from other Italian Pecorino cheeses.

Key words: free amino acids, lipolysis, Pecorino di Farindola, pig rennet

*Corresponding Author: Allesandra Fratianni, Dipartimento di Agricoltura, Ambiente e Alimenti, Università degli Studi del Molise, via F. De Sanctis, 86100 Campobasso, Italy. Email:

Submitted: 28 September 2020; Accepted 13 January 2021; Published 01 February 2021

DOI: 10.15586/ijfs.v33i1.1964

© 2021 Codon Publications

This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0). License (


The term ‘Pecorino’ refers to a cheese obtained from ewe milk, in most cases of protected origin. In Italy, however, several cheeses, even not of protected origin, have typical features and are prepared in limited geographical areas (Coda et al., 2006). Pecorino cheeses are often produced traditionally in central and south Italy and are characterised by different ripening stages (Di Cagno and Gobbetti, 2011; Schirone et al., 2011), and the use of different milk, rennet and technology of production (Gobbetti, 2016). ‘Pecorino di Farindola’ is a traditional food product (Prodotto Agroalimentare Tradizionale – PAT, published in the Official Gazette of the Italian Republic on 20th February 2020, General series n. 42 , Ordinary supplement n. 9) and has a limited production in the eastern part of the Gran Sasso area, Abruzzo, Italy. This is an original Pecorino cheese, since it is made from pig rennet that gives the cheese particular flavours and taste. The milk comes from sheep of Pagliarola Appenninica breed, which are bred in the wild, with a limited milk production. The cheese can be found at a short ripening time (3 months), demonstrating a soft texture and a yellow crust, or at a long ripening time (over 12 months), having a harder texture, a more intense and spicy flavour and a darker crust (Schirone et al., 2011). The cheese-making process is reported by Schirone et al. (2011). It starts from raw milk without the addition of natural cultures or selected starters, added with porcine rennet. The microbiota, derived mostly from mesophilic lactobacilli coming from the raw milk and the cheese-making environment, plays an important role during ripening, contributing to the development of typical aromas of this cheese (Aquilanti et al., 2007; De Angelis et al., 2001; Tofalo et al., 2015). The chemical and microbiological features of 10 Pecorino di Farindola cheeses coming from different dairies of the Consortium, at 90 days of ripening, are reported by Schirone et al. (2011). Information about the chemical and microbiological characteristics of Pecorino di Farindola, together with that of proteolytic and lipolytic phenomena during its ripening process is still limited (Di Giacomo et al., 2013; Suzzi et al., 2014; Tofalo et al., 2015). The aim of this work is to have a deeper insight into the chemical and nutritional characterization of the Pecorino di Farindola cheese and to evaluate its proteolysis and lipolysis at different ripening stages.

Materials and Methods

Sample collection and preparation

Three batches of Pecorino di Farindola cheeses were analysed. The three batches were different according to ripening times: three samples were ripened for 3 months (3R), three for 5 months (5R) and three for 12 months (12R). Samples came from a dairy in the area of Pecorino of Farindola, which includes nine towns located within the provinces of Pescara and Teramo (reported in the ‘Production Disciplinary of Pecorino di Farindola’- All cheeses were produced following the cheese-making phases stated in the production disciplinary. The 3- and 5-month cheeses were produced during spring and summer, respectively, while the 12-month ripened samples in the autumn season. Cheeses were grounded and carefully mixed. Three bulk samples were prepared by combining the samples of each ripening month and stored at –20°C until analysis.

Chemical–physical analysis

For each ripening stage, each sample was analysed in triplicate. Cheese samples were analysed, following the international methods of Association of Official Analytical Chemists (AOAC, 2000), for fat (method: 933.05), protein (method: 920.123), moisture (method: 948.12) and ash (method: 935.42). Proteolysis was assessed by determining the content of water-soluble nitrogen (SN) and non-protein nitrogen (NPN), as done in Niro et al. (2014). The amount of SN and NPN, expressed as a percentage of total nitrogen (TN) (SN/TN% and NPN/TN%) indicates the extent of proteolysis. The nitrogen content was determined by the Kjeldahl method (AOAC, 2000; method: 920.123). Free amino acids (FAAs) were analysed by a Biochrom 30 series Amino Acid Analyzer (Biochrom Ltd., Cambridge Science Park, UK), with a Li-cation-exchange column (20 × 0.46 cm). A mixture of basic, acid and neutral amino acid (AA) of a known concentration (Sigma Chemical Co., St. Louis, MO) was used as standards. The FAA extraction procedure is reported in Niro et al. (2017a). Lipolysis was expressed as Acid Degree Value (ADV) (Deeth and Fitz-Gerald, 1976).

Sensorial analysis

Each sample was evaluated for three times. Samples were evaluated by a panel comprising 10 trained components. Sensory evaluation was conducted according to the Etana method described by Bozzetti et al. (2004) modified by Chiavari et al. (2006). The evaluated attributes were flavour and aroma: odour intensity, aroma intensity, hardness, solubility, sweet, salty, bitter, spicy and acidic. The definition of the descriptive attributes is reported in Niro et al. (2014). Samples were served at room temperature. The intensity of each attribute was rated on an increasing scale from 1 to 10 (from absence to maximum).

Statistical analysis

An ANOVA was applied to the data. Least significant differences were obtained using the least significant difference test (P < 0.05).

Results and Discussion

The chemical composition of different Pecorino cheeses (g/100 g fresh weight), at different ripening stages, is reported in Table 1. The moisture value ranged from 29.6% in 3R samples to 28.5% in 5R and 23.2% in 12R samples. Significant differences (P < 0.05) were found in proteins and fats, depending on different raw milk, which, in 3R and 5R samples, came from grazing sheep, while in 12R samples from sheep fed with forage. Different ash values could depend on the variability of salting process. The cheese composition is in accordance with the values reported in the ‘Production Disciplinary’ and the ranging values are from literature for Pecorino di Farindola (Bellocci et al., 2018). Schirone et al. (2011) and Tofalo et al. (2015) have reported lower fat values in several Pecorino di Farindola cheeses. Different Pecorino cheeses were found to have similar or higher fat contents (Coda et al., 2006; Di Cagno et al., 2007).

Table 1. Proximal composition of Pecorino di Farindola cheeses at different ripening times (g/100-g fresh weight) (mean ± S.D).

Ripening time Moisture Proteins Fats Ash
3 months 29.6 ± 0.04a 25.7 ± 0.29a 39.0 ± 0.09a 4.1 ± 0.03a
5 months 28.5 ± 0.03b 27.7 ± 0.16b 36.3 ± 0.12b 5.4 ± 0.02b
12 months 23.2 ± 0.08c 31.6 ± 0.27c 40.1 ± 0.27c 5.0 ± 0.00c

Different letters within the same column indicate a significant difference (P < 0.05).

The proteolysis and lipolysis indices are shown in Table 2. The value of SN/TN% increased significantly during ripening, ranging from 32.5% at 3 ripening months (3R) to 37.6% at 12 ripening months (12R). This value is an indicator of hydrolysis of casein caused by the action of rennet and milk proteases present at the beginning of the ripening process. The SN is very variable for composition, including high-, medium-, low-molecular weight peptides and AAs. Moreover, a significant part of SN is produced during curd acidification and, consequently, it is partly lost into the water or brine (Alichanidis and Polychroniadou, 2008). The breaking off of casein and high- and medium-molecular mass peptides by microorganisms, endogenous enzymes and rennet into low-molecular mass peptides and AAs, which are soluble in 12% trichloroacetic acid (TCA), is expressed by the NPN/TN% (Corradini, 1995). This value increased significantly from 12.1% in 3R samples to 19.6 % in 12R samples. These results demonstrated a marked proteolysis of Pecorino di Farindola cheese, similar of that of common Italian Pecorino cheeses (Di Cagno and Gobetti, 2011). The principal proteolytic agents in the curd are coagulant, microbial proteinases and peptidases and indigenous milk proteinase (plasmin) (Di Cagno and Gobetti, 2011). The proteinases and peptidases of rennet are among some of the proteolytic enzymes acting during the cheese-making process and the ripening phase (Fox and Stepaniak, 1993). Tofalo et al. (2015) found faster casein breaking off in Pecorino di Farindola cheeses, attributed to higher proteolytic activity of the enzymes of pig rennet. On the contrary, a lower proteolytic activity of pig rennet than that of calf rennet was found by Di Giacomo et al. (2013) in Pecorino di Farindola cheese samples.

Table 2. Proteolysis and lipolysis indices of Pecorino di Farindola cheeses at different ripening times (mean ± SD).

Ripening time Proteolysis indices Lipolysis index
SN/TN% NPN/TN% ADV (meq KOH/100-g fat)
3 months 32.5 ± 0.75a 12.1 ± 0.61a 2.9 ± 0.34a
5 months 33.7 ± 1.91a 13.8 ± 1.68b 4.0 ± 0.23b
12 months 37.6 ± 0.12b 19.6 ± 0.24c 7.0 ± 0.05c

Different letters within the same column indicate a significant difference (P < 0.05).

A measure of lipolysis is represented by the determination of ADV. This value is a measure of the content of free fatty acids (FFA) dissolved in a certain amount of fat by lipases that can be correlated to the sensorial quality of finished products (Deeth and Fitz-Gerald, 1976; Mcsweeney and Sousa, 2000). The 3-, 5- and 12-month ripened cheeses demonstrated an ADV value of 2.9, 4.0 and 7.0 meq KOH/100-g fat, respectively (Table 2). Similar values were reported for ewe’s cheeses (Georgala et al., 2005), while higher indices were found in mixed cow/ewe Caciocavallo cheeses by Niro et al. (2014).

The evolution of FAAs of cheeses during ripening is shown in Table 3. The concentration of total free amino acids (TFAA) increased significantly (P < 0.05) to 936.48 mg/100 g at 12 months of ripening. At different ripening stages, the average TFAA content is in accordance with literature data for other Italian Pecorino cheeses (Coda et al., 2006). No proline and arginine were found in all analysed samples. The latter evidence could be due to its consumption by bacteria; in fact, many species of LAB are able to convert arginine to citrulline and ornithine (Diana et al., 2014; Niro et al., 2017a). During ripening, the content of single FAAs increased, with the exception of ornithine. As also reported by Di Giacomo et al. (2013), at 12-month ripening, the highest FAAs found were glutamic acid (Glu), valine (Val), leucine (Leu) and lysine (Lys) in Pecorino di Farindola cheese at 6-month ripening and in long-ripened Pecorino cheese (Coda et al., 2006; Mangia et al., 2008). Glutamic acid is one of the most abundant amino acids in milk protein and, as almost free, in mature cheeses (Redruello et al., 2020) and is strictly correlated with the ‘umami’ taste (McSweeney and Sousa, 2000).

Table 3. Free amino acid (FAA) content of Pecorino di Farindola cheeses at different ripening times (mg/100-g fresh weight).

FAA 3 months* 5 months 12 months
Aspartic acid 10.74a 32.20b 45.26c
Threonine 3.24a 10.06b 17.49c
Serine 0.00a 5.68b 11.01c
Asparagine 10.15a 13.27b 54.20c
Glutamic acid 15.36a 99.14b 146.21c
Glutamine 4.56a 4.01a 9.64b
Glycine 4.34a 6.45a 20.89b
Alanine 11.35a 14.13b 43.82c
Valine 30.71a 42.08b 88.03c
Methionine 7.95a 14.62b 33.68c
Isoleucine 12.03a 25.50b 60.10
Leucine 42.97a 57.66b 115.54c
Tyrosine 0.00a 2.27b 12.41c
Phenylalanine 26.75a 37.07b 63.56c
γ-Aminobutyric acid 21.55a 30.02b 56.98c
Ornithine 14.33a 15.08a 10.64a
Lysine 20.34a 38.92b 115.87c
Histidine 0.00a 9.13b 31.15c
Total 236.37a 457.28b 936.48c

*Different letters within the same raw indicate a significant difference (P < 0.05).

Leu, phenylalanine (Phe) and Lys e Val can be found in other long-ripened cheeses such as Caciocavallo cheeses (Corsetti et al., 2001; Niro et al., 2017b; Succi et al., 2016), Idiazabal (Barcina et al., 1995), Picante (Freitas et al., 1998), Serra da Estrela (Tavaria et al., 2003), Teleme (Pappa and Sotirakoglou, 2008), goat cheeses (Poveda et al., 2016) and Italian hard cheeses (Niro et al., 2017b). Leu, with the other branched chain AA, isoleucine (Ile) and Val, the aromatic AA, Phe and tyrosine (Tyr), and metionine (Met) are the main precursors of key aroma compounds (Yvon and Rijnen, 2001). Some authors have attributed the bitter flavour in cheese to the concentration of arginine (Arg) (Barcina et al., 1995; Pappa and Sotirakoglou, 2008). In all samples, ornithine (Orn) and γ-aminobutyric acid (GABA) were found; they do not originate from casein but are the products of microbial metabolism and are established functional AAs (Diana et al., 2014; Tofalo et al., 2019). As for GABA, its anti-hypertensive and anti-diabetic properties and its ability to reduce stress and anxiety are widely recognised (Redruello et al., 2020; Tofalo et al., 2019). GABA is synthesised by glutamate decarboxylase through the decarboxylation of L-glutamate; milk origin, milk treatment, proteolytic activity, fat content, texture, ripening time and climate are reported to be the key factors governing its accumulation (Redruello et al., 2020). Tofalo et al. (2019), in Pecorino di Farindola cheeses made from pig rennet, reported similar amounts of GABA than those found in this study, higher than the values of different commercial cheeses.

Regarding sensorial analysis, all samples were characterised by low acid and bitter attributes and a high salty score (Figure 1). As reported by Suzzi et al. (2014), cheeses made from pig rennet demonstrated the lowest elasticity, bitter taste and fruity and hay flavour intensities, compared with the cheeses made from calf and kid rennet. These data were also confirmed by Di Giacomo et al. (2013). Low-ripened cheeses had a higher solubility, and tasted sweeter, less hard and less spicy than the corresponding long-ripened cheeses (P < 0.05). Flavour and aroma intensity increased during ripening, reaching the highest score in Pecorino at 12-month ripening (P < 0.05). The 12-month ripened cheeses had a light spicy feature that distinguishes them from other Italian Pecorino cheeses, probably because of the use of pig rennet instead of the commonly used lamb rennet. As reported by different authors (Di Giacomo et al., 2013; Kindstedt et al., 2004), cheeses with a more evident lipolysis differ for strong flavours. The effects of FAAs on taste and flavour are reported by McSweeney and Sousa (2000) and Yvon and Rijnen (2001). In raw milk cheeses, similar to Pecorino di Farindola, the native microbiota may have played an important role and contributed to the distinct sensorial characteristics (McSweeney and Sousa, 2000; Niro et al., 2014).

Figure 1. Sensorial profile of Pecorino di Farindola cheeses at different ripening times. 3R: three-month ripening; 5R: five-month ripening; 12R: 12-month ripening.


Different Pecorino di Farindola cheeses were characterised by certain variability because of different composition of the used raw milk, the low standardization of the cheese making process and different salting and ripening conditions. The use of raw ewe’s milk and pig rennet contributed to the peculiar features of lipolysis and proteolysis and the sensorial attributes of the Pecorino of Farindola cheese. Sensorial analysis confirmed Pecorino of Farindola as sweet, lightly spicy and never bitter cheese, even when the ripening is extended a distinctive feature that is appreciated by the consumer.

Data emerging from this work could add new knowledge to the investigations of this cheese, giving an effective characterisation of the final nutritional and sensorial quality of the product. An additional investigation could be conducted to have a deeper insight into the fatty acid evolution and the influence of lipolysis and proteolysis on the profiles of volatiles during ripening.


Alichanidis E. and Polychroniadou A. 2008. Characteristics of major traditional regional cheese varieties of East-Mediterranean countries: a review. Dairy Sci. Technol. 88:495–510. 10.1051/dst:2008023

Aquilanti L., Silvestri G., Zannini A., Osimani S., Santarelli S. and Clementi F. 2007. Phenotypic, genotypic and technological characterization of predominant lactic acid bacteria in Pecorino cheese from central Italy. J. Appl. Microbiol. 103:948. 10.1111/j.1365-2672.2007.03513.x

Association of Official Analytical Chemists (AOAC). 2000. Official Methods of Analysis. Vol. 2, 17th Ed. AOAC, Washington, DC. ISBN-13 978-093558467-7

Barcina Y., Ibanez F.C. and Ordonez A.I. 1995. Evolution of free amino acids during Idiazabal cheese ripening. Food Contr. 6:161. 10.1016/0956-7135(94)00007-Z

Bellocci M., Stramenga A., Marchegiani F., Cianciavicchia S., D’Aloise A., Melai V. and Migliorati G. 2018. Caratteristiche chimiche e valore nutrizionale del Pecorino di Farindola. In: “Il Pecorino di Farindola. Indagine fisica, chimica e microbiologica. Riferimenti di territorialità”, p 85-97. IZSAM-Istituto Zooprofilattico Sperimentale dell’Abruzzo e del Molise “G. Caporale”. Teramo, Italia.

Bozzetti V., Morara B. and Zannoni M. 2004. ETANA: un modello per definire il profilo organolettico di tutti i formaggi. Il Latte 11:66. 10.3168/jds.2013-7550

Chiavari C., Nanni M., Ferri G., Morara B. and Qualizza G. 2006. Formare assaggiatori per la valutazione sensoriale della Mozzarella di bufala. Il Latte 11:66.

Coda R., Brechany E., De Angelis M., De Candia S., Di Cango R. and Gobetti M. 2006. Comparison of the compositional, microbiological, biochemical, and volatile profile characteristics of nine Italian ewes’ milk cheese. J Dairy Sci. 89:4126. 10.3168/jds.S0022-0302(06)72458-4

Corradini C. 1995. Chimica e tecnologia del latte. Tecniche Nuove. ISBN 13: 978884810005

Corsetti A. M. R. Corbo, M. Albenzio, R. Di Cagno, M. Gobetti and P. F. Fox. 2001. Microbiology and biochemistry of Caciocavallo silano cheese. Ital. J. Food Sci. 3:297. ISSN : 1120-1770

De Angelis M., Corsetti A., Tosti N., Rossi J., Corbo M.R. and Gobbetti M. 2001. Characterization of non-starter lactic acid bacteria from Italian ewe cheeses based on phenotypic, genotypic, and cell wall protein analysis. Appl Environ Microbiol. 67:2011. 10.1128/FAEM.67.5.2011-2020.2001

Deeth H.C. and Fitz-Gerald C.H. 1976. Lipolysis in dairy products: a review. Aust J Dairy Technol. 31:53.

Di Cagno R., Buchin S., de Candia S., De Angelis M., Fox P.F. and Gobetti M. 2007. Characterization of Italian cheeses ripened under nonconventional conditions. J Dairy Sci. 90:2689. 10.3168/jds.2006-654

Di Cagno R. and Gobetti M. 2011. Hard Italian cheese. In: “Encyclopedia of Dairy Science.” Roginski H., Fox P.F. and Fuquay J.W. (Eds.). Elsevier Cambridge, MA. 10.1016/B978-0-12-374407-4.00087-X

Diana M., Rafecas M., Arco C. and Quílez J. 2014. Free amino acid profile of Spanish artisanal cheeses: importance of gamma-aminobutyric acid (GABA) and ornithine content. J Food Compos Anal. 35:94–100. 10.3168/jds.2017-13308

Di Giacomo F., Del Signore A. and Giaccio M. 2013. Pig rennet in making Farindola ewe cheese. Prog. Nut. 15:226.

Freitas A.C., Fresno J.M., Prietro B., Franco I., Xavier Malcata F. and Carballo J. 1998. Influence of milk source and ripening time on free amino acid profile of Picante cheese. Food Contr. 9:187. 10.1016/S0956-7135(97)00073-X

Fox P.F. and Stepaniak L. 1993. Enzymes in cheese technology. Int. Dairy J. 3:509. 10.1016/0958-6946(93)90029-Y

Georgala A., Moschopoulou E., Aktypis A., Massouras T., Zoidou E., Kandarakis I. and Anifantakis E. 2005. Evolution of lipolysis during the ripening of traditional Feta cheese. Food Chem. 93:73. 10.1016/j.foodchem.2004.09.007

Gobbetti M. 2016. “Hard Italian Cheese. Reference Module in Food Science.” Elsevier, New York, NY. 10.1016/B978-0-08-100596-5.00671-5

Kindstedt P., Caric M. and Milanovic S. 2004. Pasta-filata cheeses. In: ”Cheese: Chemistry, Physics and Microbiology”. Vol. 2, 3rd Edition. P. Fox F., McSweeney P. L. H., Cogan T. M. and Guinee T. P. (Eds), p 251–277. Elsevier Academic Press, London, UK. ISBN 13: 9780122636530

Mangia N.P., Murgia M.A., Garau G., Sanna M.G. and Deiana P. 2008. Influence of selected lab cultures on the evolution of free amino acids, free fatty acids and Fiore Sardo cheese microflora during the ripening. Food Microbiol. 25:366. 10.1016/

McSweeney P.L.H. and Sousa M.J. 2000. Biochemical pathways for the production of flavor compounds in cheese during ripening: a review. Lait 80:293–324. 10.1051/lait:2000127

Niro S., Succi M., Tremonte P., Sorrentino E., Coppola R., Panfili G. and Fratianni A. 2017a. Evolution of free amino acids during ripening of Caciocavallo cheeses made with different milks. J Dairy Sci. 100:9521. 10.3168/jds.2017-13308

Niro S., Fratianni A., Colavita G., Galassi L., Zanazzi M. and Salimei E. 2017b. Technological use of donkey milk in cheesemaking. Int J Dairy Technol. 70:1. 10.1111/1471-0307.12342

Niro S., Fratianni A., Tremonte P., Sorrentino E., Tipaldi L., Panfili G., et al. 2014. Innovative caciocavallo cheeses made from a mixture of cow milk with ewe or goat milk. J Dairy Sci. 100:1. 10.3168/jds.2013-7550

Pappa E.C. and Sotirakoglou K. 2008. Changes of free amino acid content of Teleme cheese made with different types of milk and culture. Food Chem. 111:606. 10.1016/j.foodchem.2008.04.027

Redruello B., Zwengiel A., Madero V., del Rio B. and Alvarez M.A. 2020. Identification of technological/metabolic/environmental profiles of cheeses with high GABA contents. Food Sci Technol (LWT) 130:1. 10.1016/j.lwt.2020.109603

Schirone M., Tofalo R., Mazzone G., Corsetti A. and Suzzi G. 2011. Biogenic amine content and microbiological profile of Pecorino di Farindola cheese. Food Microbiol. 28:128. 10.1016/

Succi M., Aponte M., Tremonte P., Niro S., Sorrentino E., Iorizzo M., et al. 2016. Variability in chemical and microbiological profiles of long ripened Caciocavallo cheeses. J Dairy Sci. 99:9521. 10.3168/jds.2016-11585

Suzzi G., Sacchetti G., Patrignani F., Corsetti A., Tofalo R., Schirone M., et al. 2014. Influence of pig rennet on fatty acid composition, volatile molecule profile, texture and sensory properties of Pecorino di Farindola cheese. J Sci Food Agr. 95(11):2252. 10.1002/jsfa.6944

Tavaria F., Franco K.I., Carballo F.J. and Malcata F.X. 2003. Amino acid and soluble nitrogen evolution throughout ripening of Serra da Estrela cheese. Int Dairy J. 13:537. 10.1016/S0958-6946(03)00060-8

Tofalo R., Perpetuini G., Battistelli N., Pepe A., Ianni A., Martino G., et al. 2019. Accumulation γ-aminobutyric acid and biogenic amines in a traditional raw milk ewe’s cheese. Foods. 8:401. 10.3390/foods809040.1

Tofalo R., Schirone M., Fasoli G., Perpetuini G., Patrignani F., Manetta A.C., et al. 2015. Influence of pig rennet on proteolysis, organic acids content and microbiota of Pecorino di Farindola, a traditional Italian ewe’s raw milk cheese. Food Chem. 175:121. 10.1016/j.foodchem.2014.11.088

Yvon M. and Rijnen L. 2001. Cheese flavour formation by amino acid catabolism. Int Dairy J. 11:185. 10.1016/s0958-6946(01)00049-8