Results and discussion

Table 1 shows the DM intake, 6% fat-corrected milk, the daily yield of milk, fat, and protein, as well as the composition of milk. DDGS inclusion in the ewe diets did not affect intake, the yield of milk, or the fat-corrected milk. There is only one study (Alshdaifat and Obeidat, Reference Alshdaifat and Obeidat2019) testing corn DDGS in dairy sheep production. This reported increased milk yields with DDGS inclusion in the diets (daily milk production of 0.95, 1.17 and 1.19 kg/d for inclusion rates of 0, 20 and 30% DM, respectively). Small increases (not always significant) have also been reported in some studies with lactating cows (Janicek et al., Reference Janicek, Kononoff, Gehman and Doane2008; Chibisa et al., Reference Chibisa, Christensen and Mutsvangwa2012; Benchaar et al., Reference Benchaar, Hassanat, Gervais, Chouinard, Julien, Petit H and Massé2013) and it has been attributed to either the slightly increased energy density, due to higher fat content in DDGS diets, or to greater DM intake and, therefore, higher energy acquisition by the animals offered DDGS. Nevertheless, most cow studies with isoenergetic and isonitrogenous experimental diets, and similar intakes, as those applied in the present experiment, reported no effect on milk yield, when the applied inclusion level of DDGS was up to 30% DM (Abdelqader et al., Reference Abdelqader, Hippen, Kalscheur, Schingoethe and Garcia2009; Gaillard et al., Reference Gaillard, Sørensen, Vestergaard, Weisbjerg, Basar, Larsen, Martinussen, Kidmose and Sehested2017a, Reference Gaillard, Sørensen, Vestergaard, Weisbjerg, Basar, Larsen, Martinussen, Kidmose and Sehested2017b; Testroet et al., Reference Testroet, Beitz, O'Neil, Mueller, Ramirez-Ramirez and Clark2018)

Table 1. Milk production and chemical composition of milk from ewes fed diets contained 0 (control, DG0), 6 (DG6) or 12 (DG12) g of wheat – based dried distillers' grains with solubles per 100 g DM

^a–cMeans within a row not sharing a common superscript differ due to the different diet examined (P < 0.05), ¹Probability of significant effects due to diet (D), time (T), and their interaction (D × T); *P < 0.05; **P < 0.01; ***P < 0.001; NS, Non-significant; †P < 0.1: tendency, ²6% FCM, milk yield × (0.453 + 0.0912 × fat), ³SNF, Solids non-fat.

DDGS inclusion did not affect milk protein levels in the present study (Table 1), a fact that was also reported in a previous related study in sheep (Alshdaifat and Obeidat, Reference Alshdaifat and Obeidat2019). Nonetheless, the results in previous studies with cows on the effect of DDGS on milk protein content are controversial, reporting either an increase (Testroet et al., Reference Testroet, Li, Beitz and Clark2015), or a decrease (Kleinschmit et al., Reference Kleinschmit, Schingoethe, Kalscheur and Hippen2006; Benchaar et al., Reference Benchaar, Hassanat, Gervais, Chouinard, Julien, Petit H and Massé2013; Kurokawa et al., Reference Kurokawa, Shibata, Tateno, Kanda, Takaura, Ishida and Itabashi2013), or no significant effect (Anderson et al., Reference Anderson, Schingoethe, Kalscheur and Hippen2006; Janicek et al., Reference Janicek, Kononoff, Gehman and Doane2008; Sasikala-Appukuttan et al., Reference Sasikala-Appukuttan, Schingoethe, Hippen, Kalscheur, Karges and Gibson2008; Chibisa et al., Reference Chibisa, Christensen and Mutsvangwa2012). Reviewing the studies in cows, Kalscheur et al. (Reference Kalscheur, Garcia, Schingoethe and Diaz-Royon2012) concluded that milk protein content is rarely reduced by feeding DDGS, unless protein is limited in the diet or DDGS levels are higher than 30% DM of the diet reflecting the high rumen undegradable protein of the diet, which may result in a reduction of provided lysine for milk protein synthesis in the mammary gland. However, none of these prerequisites took place in our study where adequate feed protein levels and lower DDGS inclusion rates were used.

Milk fat percentage in the present study was reduced with the inclusion rate of 12%, but not with 6% of DDGS in the diet (on DM basis) compared to control, without however affecting the fat yield (Table 1). Reduction of milk fat was not observed in a previous study in sheep with either 20 or 30% DM inclusion rates of corn DDGS (Alshdaifat and Obeidat, Reference Alshdaifat and Obeidat2019). Nevertheless, it is well established in lactating cows that the addition of high levels of DDGS in diets could result in a decrease of milk fat percentages when NDF content and forage participation in the diets are low, a fact that has also been observed in practice (Schingoethe et al., Reference Schingoethe, Kalscheur, Hippen and Garcia2009). A meta-analysis by Kalscheur et al. (Reference Kalscheur, Garcia, Schingoethe and Diaz-Royon2012) demonstrated that milk fat depression in cows fed with DDGS occurred only when diets contained less than 22% forage NDF and 50% total forage. This finding led these authors to suggest that the key of maintaining milk fat percentage is to provide not only adequate NDF and specific percentage of forages in the diet but also sufficient amounts of effective fiber (e.g. hay as opposed to corn silage). Although our study was designed to provide adequate dietary NDF content (34 and 35% DM in DG6 and DG12, respectively) and the barley straw component was at 40% of the diet DM (usual practice in Cyprus), it is not known if these practices were adequate to prevent milk fat reduction in sheep with the 12% DDGS inclusion in the diet.

All individual FA and their aggregated groups determined in milk of DG0, DG6, and DG12 ewes are presented in Table 2. The DDGS inclusion had a marked effect on milk FA composition by reducing SFA and increasing unsaturated FAs in both DG6 and DG12 treatments. The content of saturated FAs with less than 16 carbons were linearly reduced with increased DDGS proportion in the ewe diets, while the only saturated FA that came close to increase was C18:0 and this difference was not significant. These results are in line with those of a previous study in ewes (Cais-Sokolińska et al., Reference Cais-Sokolińska, Wójtowski, Pikul, Danków, Majcher, Teichert and Bagnicka2015) and with many studies conducted in cows (Leonardi et al., Reference Leonardi, Bertics and Armentano2005; Anderson et al., Reference Anderson, Schingoethe, Kalscheur and Hippen2006; Sasikala-Appukuttan et al., Reference Sasikala-Appukuttan, Schingoethe, Hippen, Kalscheur, Karges and Gibson2008; Abdelqader et al., Reference Abdelqader, Hippen, Kalscheur, Schingoethe and Garcia2009; Hippen et al., Reference Hippen, Schingoethe, Kalscheur, Linke, Rennich, Abdelqader and Yoon2010; Kurokawa et al., Reference Kurokawa, Shibata, Tateno, Kanda, Takaura, Ishida and Itabashi2013; Gaillard et al., Reference Gaillard, Sørensen, Vestergaard, Weisbjerg, Basar, Larsen, Martinussen, Kidmose and Sehested2017a, Reference Gaillard, Sørensen, Vestergaard, Weisbjerg, Basar, Larsen, Martinussen, Kidmose and Sehested2017b) showing diminished de novo FA percentages and increased levels of long-chain FA (LCFA) as a result of DDGS dietary inclusion. The reduction of SFA in the present study was accompanied by a linear increase in total PUFA and MUFA with the increasing concentrations of DDGS in the sheep diets. Individual FA with a special interest for human health such as oleic, vaccenic, linoleic, and conjugated linoleic acid (CLA cis-9, trans-11), were linearly elevated with the addition of 6% and 12% DDGS in diet DM (Table 2). Similar effects have been reported in studies that examined diets from 5 up to 20% DM inclusion levels of DDGS in cows (Kalscheur et al., Reference Kalscheur, Garcia, Schingoethe and Diaz-Royon2012; Kurokawa et al., Reference Kurokawa, Shibata, Tateno, Kanda, Takaura, Ishida and Itabashi2013; Gaillard et al., Reference Gaillard, Sørensen, Vestergaard, Weisbjerg, Basar, Larsen, Martinussen, Kidmose and Sehested2017a, Reference Gaillard, Sørensen, Vestergaard, Weisbjerg, Basar, Larsen, Martinussen, Kidmose and Sehested2017b) and also observed in ewes with low DDGS inclusion (3.5% in diet DM, Cais-Sokolińska et al. Reference Cais-Sokolińska, Wójtowski, Pikul, Danków, Majcher, Teichert and Bagnicka2015).

Table 2. Fatty acid composition (expressed as a percentage of total fatty acid methyl esters) of milk from ewes fed diets contained 0 (control, DG0), 6 (DG6) or 12 (DG12) g of wheat – based dried distillers' grains with solubles per 100 g DM

^a–cMeans within a row not sharing a common superscript differ due to the different diet examined (P < 0.05), ¹Probability of significant effects due to diet (D), time (T), and their interaction (D × T); *P < 0.05; **P < 0.01; ***P < 0.001, NS, Non-significant; †P < 0.1: tendency, ²Other C18:1 cis = C18:1 cis-11; C18:1 cis-12; C18:1 cis-13; C18:1 cis-14; C18:1 cis-15, ³SCFA = short-chain fatty acids (C4:0 to C8:0), ⁴MCFA = medium-chain fatty acids (C10:0 to C16:1), ⁵LCFA = long-chain fatty acids (C17:0 and above), ⁶Atherogenic index = (C12:0 + 4 × C14:0 + C16:0)/(ΣMUFA + ΣPUFA), ⁷Desaturation index = (C14:1 cis-9/C14:0 + C14:1 cis-9) × 100, Unidentified picks were less than 2% in all groups.

The reduction of FA with carbon chain less than 16 is due to either a higher secretion of LCFA from the blood and/or a lower de novo synthesis of FAs in the mammary gland (Chilliard et al., Reference Chilliard, Glasser, Ferlay, Bernard, Rouel and Doreau2007; Dorea and Armentano, Reference Dorea and Armentano2017). DDGS diets contained higher levels of ether extract than the control diet, indicating that ewes fed with DDGS consumed greater amounts of LCFA, including PUFA such as linoleic acid. Furthermore, de novo FA secretion can be inhibited by bioactive PUFA, produced during incomplete ruminal biohydrogenation of dietary LCFA, particularly CLA trans-10, cis-12 and trans monoenes, such as C18:1 trans-10 (Dorea and Armentano, Reference Dorea and Armentano2017) with the latter found elevated in the milk of ewes fed DDGS in our study (Table 2). With regards to the predominant CLA isomer, cis-9, trans-11, (rumenic acid), it is well known that the majority of rumenic acid in the milk fat is synthesized endogenously, in the mammary gland through the action of mammary Δ9-desaturase with the substrate being vaccenic acid, while a small amount of this specific CLA originates from ruminal biohydrogenation of unsaturated FA by rumen bacteria (Dorea and Armentano, Reference Dorea and Armentano2017). Thus, the elevated proportions of rumenic acid observed in the milk of DDGS groups are either due to the higher concentrations of vaccenic acid observed in those groups (Table 2) or due to the incomplete biohydrogenation of longer FA that took place in the rumen of ewes fed with the DDGS diets.

In conclusion, the observed changes in the FA profile as a consequence of feeding dried distillers' grains with solubles resulted in an enrichment with beneficial FA and a reduction of atherogenic index of sheep milk by 20% (6% by DM inclusion) and 35% (12% inclusion), suggesting an improvement of health-related attributes of milk and possibly related dairy products. These results are useful for exploiting a cost-effective bioethanol byproduct to improve the FA profile of ewe milk and possibly other related dairy products without adverse effects on milk production.