Introduction Brač cheese is produced from raw sheep milk and its quality is mainly affected by the composition of milk as in most of Croatian traditional cheeses. If good hygiene is practised during milking of healthy sheep, wholeness and intact levels of milk constituents (especially casein) are retained. In the coagulation process of milk, casein forms the continuous and hard para-casein matrix which in relation to the other milk constituents mostly occludes the milk fat. Hard para-casein matrix also keeps its functional properties during syneresis process and represents basic substrate for proteolytic enzyme activity during process of cheese ripening. Proteolytic changes during cheese ripening are fundamental process which plays a vital role in development of characteristics of cheese texture. Therefore, the motivation for this study is to evaluate effect of raw sheep milk composition on textural curd characteristics and intensity of proteolytic and textural changes during the ripening of Brač cheese. Hypothesis and aims of the research The hypothesis of the research is that raw sheep milk composition has effect on curd syneresis and texture. The present research also aimed to assess the hypothesis that proteolytic and textural changes during ripening of brač cheese depends on raw sheep milk composition. Because casein network in cheese influence its mechanical behaviour, it can be hypothesized that intensity of proteolysis in brač cheese changes its textural characteristics during ripening. The objective of this research is to determine: • physicochemical composition and hygienic quality of raw sheep milk, • texture and syneresis characteristics of curd as indicators of sheep milk coagulation ability, • effect of sheep milk composition on texture, syneresis ability and whey expulsion of curd at four different times of coagulation, • chemical, textural and proteolytic changes in Brač cheese during ripening, • relationship between sheep milk composition and textural, proteolytic and composition changes of Brač cheese during ripening, • relationship between proteolytic pattern and texture of Brač cheese. Materials and methods The research was conducted on a family farms in the area of Pražnice and Supetar on the Brač island. Milk samples were collected twice a month throughout the different lactation periods (Supetar – 125 and Pražnice – 75 days in milking). Raw milk samples were collected into a laboratory bottles, transported in a portable refrigerator to a laboratory and stored at 4⁰ C until its coagulation within the next 24 hours. The milk samples intended for physicochemical analysis were filled in 100 ml capacity polyethylene laboratory bottles, preserved with azidiol and sent to the Reference Laboratory for Milk and Dairy Products at the Faculty of Agriculture in Zagreb. Raw sheep milk was analysed for fat, protein, casein, lactose, dry matter, solids-non-fat, urea, total and ionic calcium content as well as somatic cell counts, total number of bacteria and acidity. Milk was poured in a laboratory beaker and warmed in water bath at 32⁰ C. After rennet addition milk was immediately transferred into eight glass beakers and conical bottom centrifuge tubes. A hundred milliliters of milk for curd preparation was added in each of eight glass beakers. Furthermore a thirty grams of milk for syneresis test was weighed in each of sixteen centrifuge tubes. Each analysis was done in duplicate. Milk samples were placed in water bath at 25°C and allowed to coagulate for 30, 45, 60 and 75 minutes in laboratory under climatized conditions at same temperature. Strength of obtained curd in glass beakers were measured by texture analyser (Lloyd instruments, model TA Plus) equipped with a 500 N load cell (model XLC -0500-A1) and cylindrical probe (model FG/CY3, 12,5 mm in diameter). Syneresis was measured by centrifuging the curd at 5000 RPM for 15 min. After centrifugation the free water was separated, weighed and expressed as percentage of the total weight of milk (30 g). Fifteen batches of Brač cheese were produced with four cheeses in each batch. The cheeses were periodically sampled after 0, 30, 60, 90 and 120 days. Using a drill, cheese samples (about 20 g) were taken from the lateral side of each cheese in order to determine chemical composition and proteolytic changes of cheese during ripening period of four months. Cheese samples for texture measurement were taken by cork borer parallel to the direction of pressing of cheese between the edge of the top and bottom surface. Cheese cores were cut into cylindrical samples (diameter=17 mm; height=25 mm). One part of the cheese, with core, was cut at a depth of 1 cm which was nearest the surface, and was then replaced in the hole in the cheese. The hole was coated with cheese wax, covered with plastic mash and finally sealed with cheese wax before the cheese was replaced in the ripening room. Cheese were analysed for content of fat, protein, dry matter and salt as well as pH value. Proteolysis in cheese was monitored by polyacrylamide gel electrophoresis and determining content of water soluble nitrogen (WSN) as a percentage of total nitrogen (% TN), trichloroacetic acid-soluble nitrogen (TCA-SN) as a percentage of total nitrogen as well as free amino acid (as g of leucine/100 g dry matter). The texture profile analysis (TPA) was carried out using the texture analyser (Lloyd instruments, model TA Plus). The textural characteristics of Brač cheese were analysed for hardness 1, strain at hardnes 1, fracturability, strain at fracture, hardness 2, adhesiveness, cohesiveness and springiness. Descriptive statistics, analysis of variance (ANOVA) using a GLM procedure, pairwise comparisons of mean values with Fisher’s LSD test, Pearson's correlation coefficients and linear regression analysis were performed with SPSS (version 21). Results and conclusions The results showed that composition of sheep milk had a greater effect on curd syneresis ability in comparision to its strength. Casein and fat content, casein to fat ratio as well as urea concentration in sheep milk had no significant effect on curd strength. Increasing of casein and fat content in sheep milk significantly (P<0.01; P<0.05) reduced syneresis ability of curd while increasing casein to fat ratio significantly (P<0.01) enhanced its syneresis ability. Enhanced curd strength and its syneresis ability were significantly affected by increase of total (P<0.01; P<0.05) and ionic calcium (P<0.01) in sheep milk. The increase of urea concentration in sheep milk had no significant effect on curd strength but obtained curd showed significant reduction (P<0.01) in syneresis ability. Somatic cell count in sheep milk significantly (P<0.05) affected curd strength but did not affect its syneresis ability. The increase of casein content in sheep milk significantly (P<0.01; P<0.05) reduced the extent of secondary proteolysis in Brač cheese. The level of β-casein and its ripening index were significantly (P<0.01; P<0.05) affected by increase of casein content in sheep milk. The levels of the rest of caseins in Brač cheese did not show statistically significant differences due to the increase of casein content in sheep milk. The research showed that increase of casein content in sheep milk had significant effect (P<0.01; P<0.05) on hardness 1 and hardness 2, fracturability and springiness of Brač cheese while changes in adhesiveness, cohesiveness as well as strain at fracture and hardness 1 did not show statistically significant differences. The content of TCA-SN (%TN) in Brač cheese was significantly (P<0.01) reduced as a result of an increase of milk fat in sheep milk. The increase of the aforementioned parameter in sheep milk also reduced the content of free amino acid and WSN (% TN) in Brač cheese but without statistical significances. The increase of milk fat in sheep milk significantly (P<0.05) reduced level of γ-caseins in Brač cheese while the levels of the rest of caseins in Brač cheese did not show statistically significant differences. The fracturability, hardness 1 and hardness 2 of Brač cheese significantly (P<0.01) decreased with increase of milk fat in sheep milk. On the other hand, adhesiveness, cohesiveness as well as strain at fracture and hardness 1 of Brač cheese were not affected by fat content in sheep milk. The extend of secondary proteolysis in Brač cheese was not changed regardless of casein to fat ratio in sheep milk. The increase of casein to fat ratio in sheep milk caused a significant increase in levels of β and γ-caseins in Brač cheese. Although level of αs1casein in Brač cheese did not undergo statistically significant changes, level of αs1-I- casein was significantly modified with increase of casein to fat ratio in sheep milk. The fracturability, hardness 1 and hardness 2 of Brač cheese significantly (P<0.01) increased as a result of an increase of casein to fat ratio in sheep milk. Increase of casein to fat ratio significantly (P<0.05) affected strain at fracture of Brač cheese but did not affect its adhesiveness, cohesiveness, springiness and strain at hardness 1. The increase of urea concentration in sheep milk significantly (P<0.01) modified levels of β-caseins but did not affect content of the rest of caseins in Brač cheese. The extent of secondary proteolysis in Brač cheese was significantly (P<0.05) affected by the increase of urea concentration in sheep milk. The fracturability, hardness 1 and hardness 2 of Brač cheese were significantly (P<0.01) modified with increase of urea concentration in sheep milk. However, no significant differences were found between urea concentration in sheep milk and adhesiveness, cohesiveness, springiness as well as strain at fracture and hardness 1. Extend of secondary proteolysis in Brač cheese were significantly (P<0.01; P<0.05) affected by increase of somatic cell count in sheep milk. The levels of α and βcasein significantly (P<0.05) increased with increasing somatic cell count in sheep milk while changes in γ and αs1-I caseins did not show statistically significant differences. The increase of somatic cell count content in sheep milk significantly increased fracturability (P<0.05) and hardness 2 (P<0.01) as well as significantly reduced adhesiveness (P<0.01) in Brač cheese. Changes in strain at fracture and hardness 1 as well as springiness were also significantly (P<0.01; P<0.05) affected by increase of somatic cell count in sheep milk. There were no significant differences in hardness 1 and cohesiveness in Brač cheese due to increase of somatic cell count in sheep milk. The content of TCA-SN (%TN) in Brač cheese was significantly (P<0.01) reduced as a result of advancement of lactation. The stage of lactation had no significant effect both on the content of free amino acid and WSN (% TN) in Brač cheese. The level of βcasein in Brač cheese was significantly (P<0.05) reduced in Brač cheese with advancement of lactation. The levels of the rest of caseins in Brač cheese did not show significant differences due to stage of lactation. With advancement of lactation, the cheeses became significantly (P<0.01) softer. The major consequences were a significant (P<0.01) increase of fracturability, hardness 1 and hardness 2 with stage of lactation. The other texture parameters were not affected by stage of lactation. The levels of αs1 and β-casein were significantly (P<0.01; P<0.05) reduced and the levels of its degradation products were significantly (P<0.01) increased during the 120-day maturation period of Brač cheese. The extend of secondary proteolysis in Brač cheese significantly (P<0.01) increased by increasing of its ripening time. On the other hand, Brač cheese became significantly (P<0.01) harder, less springy and less cohesive.