**3. Results**

#### *3.1. Chemical Composition of Milk and Cheese*

Taking into account the milk production over the entire duration of the dietary zinc enrichment, no significant di fferences were evidenced between the two groups, reflecting the fact that such parameter was not a ffected by diet. Regarding the chemical quality of milk (Table 1), all the analyzed parameters did not undergo variations during the experimental period. Similarly, no significant di fferences were observed as regards the ureic content and pH, whereas the EG samples showed a lower SCC with respect to the CG (*p* < 0.01). In regards to the amount of Zn, in the experimental group were found higher average values (4.82 ± 0.23 vs 5.42 ± 0.34 mg/kg, in CG and EG respectively; *p* < 0.05).


**Table 1.** Milk yield and chemical composition of milk obtained from the control group (CG) and the experimental group (EG).

Data are expressed as mean ± S.D. 1 Somatic cell count (SCC) is reported in linear score (LS): LS = log2[(cells/μL)/100] + 3. \**p* < 0.05; \*\* *p* < 0.01; n.s. = not significant.

As evidenced for milk, the dietary supplementation did not influence cheese yield (*p* > 0.05). Regardless of the feeding strategy, no significant differences in composition of cheeses were evidenced (Table 2). Regarding the ripening time, as expected a significant reduction in moisture was found in T90 samples (*p* < 0.05); protein and lipids were not influenced by ripening, as well as the zinc amount which maintained similar values between the two groups. Furthermore, in T90 samples obtained from EG, the nitrogen fractions were significantly higher (0.94% vs. 0.63%, *p* < 0.05, for WSN; 0.56% vs. 0.35%, *p* < 0.05, for TCA-SN).

**Table 2.** Chemical composition of Pecorino cheese obtained from the control group (CG) and the experimental group (EG), analyzed after 1 (T1) and 90 (T90) days after the cheese-making.


Data are expressed as mean percentage ± S.D. 1 Data are reported on a dry matter (DM) basis. a,b Means with different superscripts are significantly different by diet (*p* < 0.05). A,B Means with different superscripts are significantly different by ripening time (*p* < 0.01).

#### *3.2. Fatty acid Profile of Milk and Cheese*

The fatty acid composition of individual milk samples collected at the beginning and at the end of the trial by both the experimental groups is reported in Table 3. At T0 not significant variations between CG and EG were evidenced testifying to the homogeneity of the animals selected for the study. At the end of the experimental period, samples of milk obtained from EG evidenced an increase in the content of vaccenic acid (C18:1 *trans*11; *p* < 0.05), rumenic acid (RA; *p* < 0.01) and total polyunsaturated fatty acids (PUFA, *p* < 0.05).


**Table 3.** Fatty acid profile of milk and fresh cheese obtained from the control group (CG) and the experimental group (EG).

Analysis on milk have been performed at the beginning (T0) and at the end of the trial (T30). SFA = saturated fatty acid; MUFA = monounsaturated fatty acid; PUFA = polyunsaturated fatty acid; RA = rumenic acid; AI = atherogenic index; TI = thrombogenic index; DI = desaturation index. Data are expressed as mean (%) ± S.D. *\* p* < 0.05; \*\* *p* < 0.01; n.s. = not significant. a,b Means with different superscripts (in milk) are significantly different by time (*p* < 0.05).

Similarly, the evaluation of the total fatty acids profile in cheese evidenced modifications already evident in milk (Table 3), with an increase in concentration, in EG, of vaccenic acid (C18:1 trans11; *p* < 0.01), rumenic acid (RA; *p* < 0.05) and total PUFA (*p* < 0.05). Based on the obtained FA profile, calculations of desaturation, thrombogenic and atherogenic indices were performed. Dietary supplementation with zinc did not induce significant modifications of desaturation index (*p* > 0.05) both in milk and cheese; thrombogenic index was significantly lower only in cheese samples obtained from EG (*p* < 0.05), whereas atherogenic index decreased both in milk and cheese obtained from the EG (*p* < 0.05).

#### *3.3. Analysis of the Oxidative Stability in Pecorino Cheese*

Diet enrichment with zinc did not induce alterations of the oxidative stability in samples of fresh cheese (Figure 1). Very interesting is instead the result obtained after 90 days from the cheese making, as expected lipid peroxidation increased in all the analyzed samples, but in ewes' milk cheese obtained from EG the value of malondialdehyde was stood at significantly lower values if compared to samples of the control group (0.092 vs 0.066 μg MDA/g of cheese, in CG and EG respectively; *p* < 0.05).

**Figure 1.** Lipid peroxidation in fresh and ripened Pecorino cheese samples obtained from control group (CG) and experimental group (EG).

#### *3.4. Volatile Profile of Cheese*

The analysis of the volatile profile allowed to identify 24 volatile compounds (VOC) in samples of T1 and T90 cheese obtained from CG and EG: 6 carboxylic acids, 5 ethyl esters, 3 aldehydes, 2 alcohols, 2 lactones, 2 ketones and 4 classified as aromatic hydrocarbons (Table 4).


**Table 4.** Volatile compounds (VOCs) detected in cheese samples obtained from control group (CG) and experimental group (EG).

Data are expressed as mean (%) ± S.D. \* *p* < 0.05; \*\* *p* < 0.01; n.s.: not significant; n.d.: not detectable.

Regarding the carboxylic acids, is interesting the increase in concentration of the hexanoic acid in the EG samples after 90 days of ripening (6.75% vs. 2.06% in EG and CG samples respectively; *p* < 0.01); in T90 samples should also be underlined the significant reduction of the longer chain acids: decanoic acid (9.67% vs. 5.09% in CG and EG respectively; *p* < 0.01) and dodecanoic acid (9.57% vs. 6.67% in CG and EG respectively; *p* < 0.01).

In the case of esters, the dietary zinc supplementation induced the increase of butanoic acid ethyl ester (*p* < 0.01) and of hexanoic acid ethyl ester (*p* < 0.05) at the end of the ripening period. Different resulted the behavior of octanoic acid ethyl ester, which is higher in the T90 samples obtained from CG (11.85% vs. 7.31%, *p* < 0.01).

In the most ripened samples was also found the hexanal increase in EG samples (9.86% vs. 7.46%, *p* < 0.05), and variations of both nonanal (1.68% vs. 1.12% in CG and EG respectively; *p* < 0.05) and 3-methylbunanol (5.98% vs. 3.27% in CG and EG respectively; *p* < 0.05).

Additionally, in T1 samples have been evidenced some significant differences, specifically in EG cheeses have been observed lower concentrations of nonanoic acid (*p* < 0.05), nonanal (*p* < 0.05) and 2-heptanone (*p* < 0.01).
