International Dairy Journal 13 (2003) 961–972 Comparison of the microbiological, compositional, biochemical, volatile profile and sensory characteristics of three Italian PDO ewes’ milk cheeses Raffaella Di Cagno a , Jean Banks b , Liz Sheehan c , Patrick F. Fox c , E.Y. Brechany b , Aldo Corsetti d , Marco Gobbetti a, * a Dipartimento di Protezione delle Piante e Microbiologia Applicata, Facolt " a di Agraria, Universit " a degli Studi di Bari, Via Amendola 165/a, Bari 70125, Italy b Hannah Research Institute, Ayr, KA6 5HL, UK c Department of Food and Nutritional Sciences, University College Cork, Ireland d Dipartimento di Scienze degli Alimenti, Universit " a degli Studi di Perugia, Italy Received 4 February 2003; accepted 23 June 2003 Abstract Three batches of three Italian ewes’ milk cheeses (Canestrato Pugliese, Fiore Sardo and Pecorino Romano), which were manufactured under the Protected Denomination of Origin (PDO), were compared for microbiological, compositional, biochemical, volatile profile and sensory characteristics. After ca. 1 year of ripening, the three cheeses had a complex microbial population, composed mainly of non-starter lactic acid bacteria (NSLAB); several species of mesophilic lactobacilli and enterococci were identified. Pecorino Romano had a very high level of NaCl, 8.770.1% w/w. The level of pH 4.6-soluble nitrogen increased in the order: Canestrato Pugliese>Fiore Sardo>Pecorino Romano. Urea-PAGE electrophoresis showed that the hydrolysis of a s1 -casein was more consistent than b-casein. RP-HPLC profiles of the ethanol-soluble and ethanol-insoluble fractions of the pH 4.6-soluble nitrogen showed differences between the cheeses which agreed with the content of free amino acids. Glutamic acid, histidine, valine, isoleucine, leucine and phenylalanine were the free amino acids present at the highest levels in all the cheeses. Sixty-two volatile components were identified by Gas Chromatography-Mass Spectrometry analysis of steam distillates of the cheeses. The volatile profile of the three cheeses differed significantly. Esters were the main volatiles in the Canestrato Pugliese cheese but were the lowest in Fiore Sardo cheese. Ketones and alcohols were the principal class of volatile components in Fiore Sardo and Pecorino Romano cheeses, respectively. Lactones and products of the breakdown of the sulphur-containing amino acids were found at considerable levels in all cheeses, but aldehydes were present at low levels. Only 30 carboxylic acids were identified; Canestrato Pugliese had the highest content of total acids. Butanoic, hexanoic, octanoic and decanoic acids represented the 30–33% of the total carboxylic acids in the cheeses. The cheeses were subjected to descriptive sensory analysis; 19 discriminating and 2 descriptive attributes were analysed by Principal Component Analysis (PCA). The sensory characteristics of the three ewes’ milk cheeses were distinct. r 2003 Elsevier Ltd. All rights reserved. Keywords: Pecorino cheeses; Proteolysis; Volatile compounds; Sensory evaluation 1. Introduction Pecorino Romano, Fiore Sardo and Canestrato Pugliese are three of the more important Italian ewes’ milk cheeses which have the Protected Denomination of Origin (PDO) status. All three belong to the category of Italian Pecorino cheeses; Pecorino is the trivial name given to the Italian cheeses made from ewes’ milk. In 2001, the production of these cheeses was 35 311, 700 and 180t for Pecorino Romano, Fiore Sardo and Canestrato Pugliese, respectively (Industria Italiana Lattiero-Casearia, 2002). These ewes’ milk cheeses are manufactured traditionally as extra-hard varieties. Pecorino Romano is manufactured in the regions around Rome and in Sardinia. It is usually made from raw or thermised milk which is inoculated with a natural culture, ‘‘scotta fermento’’, which is produced by acidifying the ‘‘scotta’’, the whey obtained from the ARTICLE IN PRESS *Corresponding author. Tel.: +39-080-544-2949; fax: +39-80-544- 2911. E-mail address: gobbetti@agr.uniba.it (M. Gobbetti). 0958-6946/$-see front matter r 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0958-6946(03)00145-6