The average proportions of seeds and pericarps in the whole EMB after drying at 40 °C were 58.60 ± 4.8 and 41.40 ± 4.8 (mean ± SD), respectively.
3.1. Chemical Composition
The results of the chemical analyses performed on whole EMB and the separated seeds and pericarps are reported in Table 1
. The seeds presented slightly higher values for DM and organic matter than for pericarps (p
< 0.01), whereas the ash content was higher for the pericarps than for seeds (p
< 0.01). Regarding the fiber content, NDF was higher in seeds than in pericarps (p
< 0.01), whereas ADF and lignin contents were both higher in pericarps than in seeds (p
< 0.01). The differences in fiber content and composition for the two EMB fractions was further highlighted by the higher hemicellulose (27.75 vs. 21.24) and cellulose (25.21 vs. 8.62) contents in seeds than in pericarps (p
< 0.01). Non-fiber carbohydrates (NFC) were more abundant in pericarps than in seeds (p
< 0.01). The crude protein and fat contents were higher in seeds than in pericarps (p
< 0.01); in particular, the values for crude protein and fat were about 2-fold and almost 10-fold higher in seeds compared with pericarps, respectively. These differences result in a significantly higher energy value for seeds compared with pericarps (445 vs. 384 kcal/100 g DM, p
< 0.01). Overall, the chemical composition of whole EMB showed interesting value from a nutritional point of view, suggesting a possible use as feedstuff. This is evidenced by the value of gross energy (425 kcal/100 g of DM), which is comparable to that of typical feeds used in ruminant nutrition, as soybean meal (350–450 kcal/100 g of DM). Recently, the EMB was used as supplement in two nutritional trials in sheep [31
], evidencing contrasting results in term of milk production (no effect or reduction of milk yield) and milk composition (no effect or reduction of protein and fat content, and reduction or no effect of milk urea content), but both studies agreed on the suitability of this by product as feed in sheep.
3.2. Fatty Acid Composition
The fatty acid (FA) profiles of pericarps, seeds and whole EMB are presented in Table 2
. The FA profiles for EMB and seeds were very similar due to the low contribution of pericarps to the lipid content of whole EMB (see Table 1
). The FA profile of pericarps showed a composition similar to that of seeds, but with a different proportion of each FA. Linoleic acid (C18:2 n-6, LA) was the most abundant FA in seeds and in whole EMB, accounting for 75% and 71% of total FA, respectively. The other most representative FAs in seeds and whole EMB were oleic acid (C18:1 cis-9, OA; 9.25% and 9.41%, respectively), palmitic acid (C16:0, PA; 8.30% and 9.34%, respectively), and stearic acid (C18:0, SA; 3.99% and 4.26%, respectively). In pericarps, the most abundant FA was PA, accounting for about 25%, followed by LA, OA, SA, arachidic acid (C20:0, AA) and LNA (17.31%, 11.69%, 8.12%, 5.20% and 4.24%, respectively). Interestingly, pericarps showed a higher proportion of LNA and saturated and unsaturated long chain FAs when compared with seeds (p
< 0.01). In general, these results are in line with the FA composition of seeds and pericarps of fresh myrtle berries as reported in previous studies [33
]. However, a different FA profile was reported by Cakir [35
], who found the OA content of seeds and mesocarps to be 64% and 72%, respectively, with LA accounting for only 12.7% and 1.7%, respectively. This discordance could be ascribed to a difference in the maturation stage of the berries. In fact, when the variations in FA composition of myrtle berries were studied at different time points during fruit maturation [36
], PA and OA were shown to be the most abundant FA in the first stage of ripening (37.03% and 21.89%, respectively), whereas their proportions decreased progressively throughout all stages of ripening (until 13.58 and 6.49%, respectively). On the other hand, the proportion of LA only accounted for 12.21% at 30 days after flowering and increased progressively to 71.34% in fully ripe fruit (180 days post flowering), thus reaching comparable values to those observed in our study.
The high amount of PUFA makes the lipid content of EMB potentially useful because of the beneficial biological and nutritional properties of these compounds. Indeed, LA could be included in cosmetic formulations since it exhibits important skin protection properties [37
]. Moreover, the inclusion of lipid sources (with a high proportion of PUFA) in ruminant diets represents a useful strategy to increase the proportion of beneficial FA in meat and milk and their derived products [38
]. Values of LA that exceed 50% of total FA are typical of plant oils, such as soybean, sunflower and grape seed oils [39
]. In particular, the FA profile reported in our study for the myrtle seeds is very close to that of the grape seed byproduct [40
], which was found to enhance the concentration of beneficial FA in sheep milk when added to the animals’ diet [14
3.3. Polyphenolic Compounds
A preliminary screening of polyphenol total content was performed using the Folin–Ciocalteau method and data were expressed as µg GAE/mg of dry extract; the results are in line with those reported by Wannes and Marzouk [41
] relating to fresh berry parts. As evidenced by the results (Table 3
), the total polyphenol content was higher in seed extracts than pericarps (p
< 0.01). In two recent trials on sheep nutrition, the presence of polyphenols in EMB has been associated to the reduction in blood and milk urea concentration [31
] and in ammonia accumulation in rumen [42
]. It seems correlated to the ability of polyphenols to bind dietary proteins and to reduce their ruminal degradation. In addition, EMB was found to be effective in reducing the proteolytic bacteria in rumen [42
]. These findings also point out that Myrtus
byproduct could be used to increase feed efficiency in animals, in terms of better protein utilization.
All secondary metabolites detected in EMB samples were identified by comparing their chromatographic behaviors and their MS and MS/MS spectra with those of standard reference compounds, when available.
The MS conditions were optimized using reference standards to achieve optimal MS sensitivity for detection and to obtain abundant fragment ions for structural elucidation. Molecules that were identified in negative ion mode belonged to the flavonoid and phenolic acid compound classes. On the other hand, due to the presence of a positive charge in the chemical structure of anthocyanin, good signal sensitivity could also be obtained in positive ion mode.
All compounds were finally confirmed by monitoring their characteristic transitions in MRM mode and comparing their retention times with those of the corresponding authentic standards.
The analytes listed in Table S1
were monitored for their occurrence and 31 compounds were identified in the investigated samples (Table 4
The precursor/product transitions selected to develop the MRM method are described in Table S1
. Quantitative results are reported in Table 4
. Each of the three samples was analyzed in triplicate, and the results obtained are expressed as average values of the three analyses.
As shown, ellagic acid was found as the most representative compound in all samples with the highest content in seeds (345 mg/100 g FW), followed by whole EMB (281 mg/100 g FW) and pericarps (244 mg/100 g FW). The other most abundant acids were gallic and quinic acids, ranging 63–123 mg/100 g FW and 77–121 mg/100 g FW, respectively.
With regard to flavonoids, quercetin and quercetin 3-O-rhamnoside were the most abundant (the greatest levels being found in seeds [21 mg/100 g FW and 24 mg/100 g FW, respectively]) followed by isorhamnetin, with values in the range 8–15 mg/100 g FW. Myricetin 3-O-galactoside content was higher in pericarps (10 mg/100 g FW) than in seeds or whole EMB. Overall, the seeds contained the highest level of total polyphenols, at 566 mg/100 g FW. No anthocyanin compounds were found in our samples; this is probably because these compounds are exhaustively extracted during the hydroalcoholic infusion of the myrtle berries in liqueur production. In addition, the low stability of these compounds, which are easily degraded by light, high temperature and air, is widely reported in the literature [43
Only few studies have assessed and quantified the polyphenolic composition of myrtle berries: three were focused on whole fresh berries [5
]; one on pericarps [46
]; and one specifically looked at the various myrtle berry parts [41
]. Thus, a real comparison of our data with other published results is difficult. Nevertheless, the majority of secondary metabolites identified in our samples have previously been reported as present in fresh myrtle fruit; with the exception of caffeic acid, p-coumaric acid, ferulic acid, sinapic acid, quinic acid, syringic acid, chlorogenic acid, isorhamnetin, robinin, isorhamnetin 3-O-rutinoside, neohesperidin, phloridzin, apigenin, luteolin and epicatechin, which were not investigated in the cited papers.
The liquor preparation by hydroalcoholic infusion of berries, extract some of the polyphenolic compounds. Consequently, as expected, the detected levels of the main bulk of polar compounds in EMB were lower than those reported in the literature for fresh myrtle fruit, apart from ellagic acid that was more abundant in our samples. Ellagic acid is a naturally occurring phenolic compound found at high concentrations in many berries; in plants, it forms structural components in the plant cell wall and cell membrane in the form of hydrolysable tannins (ellagitannins), where it is esterified with glucose. Several papers have investigated the biological properties of ellagic acid, which include antioxidant, antimicrobial, anti-inflammatory and antimutagenic activities, as reviewed in [47
3.4. Antioxidant Activity
The free radical-scavenging properties of the exhausted myrtle berry byproduct are presented in Figure 1
, where a lower IC50
value (µg/mL) implicates higher antioxidant activity. The ability of DPPH radical scavenging was significantly higher in seeds (p
< 0.01) than in pericarps, with a three-fold higher antioxidant activity at both time points investigated (0 and 30 min). Our results are in line with those reported by Wannes and Marzouk [41
] for the separate myrtle fruit parts, where seeds showed the highest antioxidant activity. This result could be explained by considering the higher content of phenolic acids and flavonols in seeds than in pericarps, as the antioxidant activity of fruit is mainly obtained from phenolic compounds [41
assay showed that antioxidant activity was also significantly higher in seeds (p
< 0.01) than in pericarps, with values two-fold higher at both time points (Figure 1
). A highly significant positive correlation was found by comparing the results obtained using the Folin–Ciocalteau method with the DPPH and ABTS results, respectively (Table 3
), confirming the well documented [48
] role of phenols in antioxidant activity.