In general, untreated whole flaxseed is readily accepted by dairy cows and feeding up to 15% of the total dry matter (DM) as flaxseed has no effect on DM intake of mid- (Secchiari et al. Reference Secchiari, Antongiovanni, Mele, Serra, Buccioni, Ferruzzi, Paoletti and Petacchi2003) and early-lactating (Petit, Reference Petit2002) dairy cows, which makes it a good source of energy and protein for milk production. Flaxseed is also one of the richest sources of the plant lignan precursor, secoisolariciresinol diglucoside (SDG; Axelson et al. Reference Axelson, Sjovall, Gustafsson and Setchell1982). Epidemiological and experimental studies strongly suggest that lignans have a potential role in the prevention of menopausal symptoms, hormone-dependent cancers (e.g. breast and prostate cancer), cardiovascular disease, and possibly osteoporosis (Murkies et al. Reference Murkies, Wilcox and Davis1998). Under the action of intestinal glycosidases, SDG is transformed into secoisolariciresinol (SECO; Saarinen et al. Reference Saarinen, Smeds, Mäkelä, Ämmälä, Hakala, Pihlava, Ryhänen, Sjöholm and Santti2002). In non-ruminant animals, colonic microbiota convert SECO to mammalian lignans, mainly enterodiol (ED) and enterolactone (EL; Setchell et al. Reference Setchell, Lawson, Mitchell, Adlercreutz, Kirk and Axelson1980). Plant lignan matairesinol (MAT) is another precursor of the mammalian lignan EL but there is no in-vitro conversion of MAT into ED by faecal microbiota (Heinonen et al. Reference Heinonen, Nurmi, Kiukkonen, Poutanen, Wähälä, Deyama, Nishibe and Adlercreutz2001). Mammalian lignans are subsequently absorbed and undergo enterohepatic circulation (Borriello et al. Reference Borriello, Setchell, Axelson and Lawson1985). Secoisolariciresinol diglucoside and mammalian lignan metabolites ED and EL have greater antioxidant activity than vitamin E (Prasad, Reference Prasad2000), which could then be a tool to decrease milk oxidation and produce value-added milk with health benefits. In fact, people with higher blood concentrations of EL have a lower incidence of cardiovascular diseases (Vanharanta et al. Reference Vanharanta, Voutilainen, Lakka, Van der Lee, Adlercreutz and Salonen1999).
There is little information on flaxseed and flaxseed meal as a source of dietary antioxidants and their concentrations in milk. One study has looked at the transfer of lignans in milk of rodents and it has been demonstrated that 3H-SDG is transferred to offspring through milk when rat dams are fed 3H-SDG (Tou et al. Reference Tou, Chen and Thompson1998). There is detectable concentration of EL in the milk of dairy cows (Antignac et al. Reference Antignac, Cariou, Le Bizec and André2004) which may result from the conversion of plant lignan SDG into mammalian lignan EL (Setchell et al. Reference Setchell, Lawson, Mitchell, Adlercreutz, Kirk and Axelson1980) but there is no information on the potential for increasing the concentration of mammalian lignans in milk by feeding flaxseed products. Lignans in grain are concentrated in the outer fibre-containing layers (Adlercreutz & Mazur, Reference Adlercreutz and Mazur1997) which leads to higher concentration of SDG in flax products without than with oil (Côrtes et al. Reference Côrtes, Gagnon, Benchaar, da Silva, Santos and Petit2008). Therefore, the objective of the experiment was to determine the effects of feeding the same dietary level (10% of DM) of two sources of flax products with different concentrations of SDG, whole flaxseed and flaxseed meal, on concentrations of the mammalian lignans ED and EL in milk. Feed intake, digestion, milk production, milk composition and milk fatty acid profile were also determined to compare the use of whole flaxseed and flaxseed meal for milk production because information on this comparison was not found. Although ether extract concentration is lower in meal than seed, there are some lipids left in the meal, which might contribute to modifying milk fatty acid profile.
Materials and Methods
Cows and their diets
The experiment was conducted at the Atlantic Dairy and Forage Institute (ADFI) of Fredericton Junction, NB, Canada, from June to December 2003 using 24 multiparous Holstein cows averaging 620 (se=29) kg of body weight. Cows were cared for in accordance with the guidelines of the Canadian Council on Animal Care (CCAC, Reference Olfert, Cross and McWilliam1993). Cows were blocked for similar days in milk. The experiment was conducted from week 17 to week 21 of lactation. Cows were housed in tie stalls, fed individually and milked twice daily at 6.15 and 15.30. Milk production was recorded at every milking. Cows within blocks were assigned randomly to one of three dietary treatments based on a mixture of silages (Table 1). The three dietary treatments (Table 1) consisted of supplements based on no flaxseed product (CO), 10% flaxseed meal (FM) or 10% whole flaxseed (FS) in the DM. All diets were formulated to meet requirements for cows that averaged 620 kg of body weight and produced 40 kg/d of milk with 3·90% fat (NRC, 2001). Yield of 4% fat-corrected milk was calculated according to the equation of Tyrrell & Reid (Reference Tyrrell and Reid1965). Diets were fed twice daily at 6.00 and 14.00 for 10% orts. Feed consumption was recorded daily. Total mixed diets were sampled daily in the last week of the experiment, frozen, and composited on a 7-d basis. Composited samples were mixed thoroughly and subsampled for chemical analyses.
Table 1. Composition of total mixed diets of Holstein cows fed no flaxseed product (CO), 10% flaxseed meal (FM) or 10% whole flaxseed (FS) in the dry matter (DM). Values are least squares means with pooled se
a,b,c Means within rows with different superscripts differ (P<0·05)
† Protein supplement contained (as fed basis) 40·0% CP, 1·5% fat, 9·0% crude fibre, 45 000 i.u. of vitamin A/kg, 10 000 i.u. of vitamin D/kg, 175 i.u. of vitamin E/kg, 3·6% Ca, 1·5% P, 0·87% Mg, 1·3% Na, 0·3% S, 0·7; 12 mg/kg of Se, 667 mg/kg of Zn, 10 mg/kg of I, 100 mg/kg of Co, 350 mg/kg of Mn, 64 mg/kg of Co, 100 mg/kg of Cu, 360 mg/kg of F and 3000 mg/kg of Fe
‡ Mean of 8 weekly samples prepared by compositing daily samples collected on the digestion week
§ Calculated using published values (NRC, 2001)
¶ Mean of one pool made from seven daily samples collected from each cow during the digestibility trial
Measurements
Milk samples were obtained weekly from each cow for two consecutive milkings and were analysed separately to determine milk composition. Milk samples were kept at room temperature with a preservative (bronopol-B2) for determination of protein, fat and lactose concentrations. One sample without preservative was kept frozen to determine milk fatty acid profile and lignan-metabolite concentrations on week 5 of the experiment. Body weight of cows was determined at the beginning and the end of the experiment. Total faeces were collected from all cows during week 5 of the experiment for 7 d. Faeces were collected from a rubber mat placed behind the animals and were stored in plastic containers. Daily faeces were weighed and mixed thoroughly. A 10% subsample was taken daily and stored at −15°C for subsequent drying at 55°C.
Chemical analysis
DM content of the total mixed diets was determined by drying at 100°C for 48 h. Nitrogen determination was done by the Kjeldahl method (AOAC, 1990). Neutral and acid detergent fibre components were measured according to the nonsequential procedures of Van Soest et al. (Reference Van Soest, Robertson and Lewis1991). Nitrogen, fat and lactose in milk were determined by infrared spectroscopy (Bentley model 2000; Bentley Instrument Inc., Chaska MN, USA) except for fat in milk samples collected during the digestion trial, which were analysed by the method of Roese-Goettlib (AOAC, 1990). Somatic cells were counted using an optical somatic cell counter (Fossomatic 90; Foss Electric, Hellerød, Denmark).
Milk fat was extracted for determination of fatty acid profile using a method adapted from Jiang et al. (Reference Jiang, Bjoerck, Fonden and Emanuelson1996). Extracted lipid samples were methylated (Lock & Garnsworthy, Reference Lock and Garnsworthy2002) and fatty acid methyl esters were separated on a HP 5890 GLC fitted with a HP autosampler (model 18596C, HP 7673 injector, Hewlett-Packard Ltd., Mississauga ON, Canada) and a flame ionization detector. Hewlett Packard 3365 Chemstation software was used for chromatogram integration and analysis. Samples were introduced onto a 100-m Supelco (Oakville ON, Canada) SP-2560 column (part number 24056) via 1-μl splitless injections. The temperature programme was as follows: level one, 120°C held for 15 min; level two, 120°C to 160°C at 5 deg C/min, then held for 15 min; level three, 160°C to 240°C at 4 deg C/min, then held for 30 min. Injector temperature was set at 220°C and the detector was set at 275°C. Column head pressure was set at 30 psi. A 2-mm I.D. splitless injection sleeve (Chromatographic Specialties Inc., Brockville ON, Canada) was used for all injections. Gas flow rates were: helium (carrier) 1·7 ml/min, helium (make up) 29 ml/min, compressed air 320 ml/min, and hydrogen 34 ml/min.
Extraction and analysis of plant SDG in diets was performed as described in the patent of Pizzey (Reference Pizzey2006). Lignans were hydrolysed and extracted from milk according to the method of Frank & Custer (Reference Frank and Custer1996) originally developed for daidzein and genistein determination in human milk, with some modifications. Briefly, frozen milk was thawed and warmed at 40°C for 30 min. Each sample was analysed in triplicate. For hydrolysis, 500 μl of milk were mixed with 5 μl of β-glucuronidase/arylsulphatase from Helix pomatia (Roche-Diagnostics, Laval QC, Canada) and then shaken in a water bath at 37°C for 1·5 h. Hydrolysed samples were washed with 3 ml of hexane. The upper phase was discarded and extraction was done twice with 2 ml of diethyl ether. Extracts were combined and evaporated by vacuum (Speed-Vac; Thermo Savant, Holbrook NY, USA) at room temperature. The dry extract was warmed at 37°C and redissolved in 150 μl of methanol and 50 μl of 0·2 m-sodium acetate buffer (pH 4) by vortexing. After centrifugation at 1800 g for 1·5 min, 150 μl of clear supernatant was transferred into a HPLC vial and a 50-μl aliquot was injected into the HPLC system. Standards of ED and EL were obtained from Sigma-Aldrich (Oakville ON, Canada). The standard used for quantification was a mixture of ED and EL. Average recoveries of ED and EL, which were calculated by addition of different doses of standards to a pooled sample, were 98% and 110%, respectively. The intra- and inter-assay CV was 4% and 8%, respectively.
ED and EL in milk samples were quantified by a HPLC (model 210, Varian, Mississauga ON, Canada) solvent delivery system equipped with a 100-μl injection loop autosampler (model 410, Varian) and an electrochemical detector (Coulochem II; ESA, Concord ON, Canada) set as follows: guard cell (model 5020; E=+1400 mV) and analytical cell (model 5010, E1=0 mV; E2=+1000 mV). Sensitivity of the detector was set to 1 μA. Analyses were carried out on a C18 reversed-phase column (4·6 mm×250 mm Microsorb-mv; R0086200C5; Varian). The mobile phase consisted of a mixture of two eluents (50/50): A, 50 mm-sodium acetate buffer, pH 4·8/methanol, 80/20; and B, 50 mm-sodium acetate buffer, pH 4·8/methanol/CAN, 40/40/20. The mobile phase was filtered on a 0·22-μm nylon membrane filter (Gelman-Sciences, Ann Arbor MI, USA) and then pumped through the column at a flow rate of 1 ml/min. All the solvents were HPLC-grade from Fisher Scientific (Nepean ON, Canada).
Statistical analysis
All results were analysed using the MIXED procedure of SAS (2000). Data recorded during the digestibility trial were analysed as a randomized block design and block and treatment were the main sources of variation. Data on EL concentrations were transformed (log) as previously performed by Côrtes et al. (Reference Côrtes, Gagnon, Benchaar, da Silva, Santos and Petit2008). Data on feed intake, milk production and milk composition were analysed as a randomized block design with repeated measurements using PROC MIXED of SAS and as mean values for the 5-week experiment when there was no interaction between week and treatment (P>0·10). When a significant F test was detected (P<0·05) treatment means were separated using Tukey's multiple-range test.
Results
Whole flaxseed and flaxseed meal were fed at 11·1% and 9·4% of DM (Table 1), respectively, which are close to the 10% planned at initiation of the experiment. Concentration of SDG in the DM of the total mixed diets increased significantly from CO (0·002%), FS (0·072%) to FM (0·138%).
There was no interaction between week and treatment for any measurement. Intake of DM averaged 20·7 kg/d and was similar among treatments (Table 2). Diet had no effect (P>0·10) on initial and final body weight, body weight change, average body weight, milk yield, milk composition, yield of milk components and 4% fat-corrected milk yield (Table 2). Daily intake of SDG was significantly higher for cows fed FM than for those fed the other diets. Cows fed FS had higher intake of SDG than those fed CO but lower intake of SDG than those fed FM. Cows fed CO had the lowest (P=0·001) intake of SDG.
Table 2. Intake, body weight, milk production and milk composition of Holstein cows fed no flaxseed product (CO), 10% flaxseed meal (FM) or 10% whole flaxseed (FS) in the dry matter (DM). Values are least squares means with pooled se
†a,b,c Means within rows with different superscripts differ (P<0·05)
‡ Somatic cell score=log10 somatic cell count
The mammalian lignan ED was not detected in the milk of cows. Therefore, no data are reported on ED concentration. On the other hand, concentration of EL in milk was above detection level and differed (P=0·01) among treatments (Table 2). Cows fed CO had a lower concentration of EL in milk than those fed FM and FS (P=0·02 and P=0·03, respectively). Yield of EL in milk differed (P=0·05) among treatments (Table 2). Cows fed FM had higher (P=0·02) yield of EL in milk than those fed CO and there was no difference between cows fed FS and CO (P=0·13) and between those fed FM and FS (P=0·95). Daily data on milk EL yields and SECO intakes were used to calculate the apparent recoveries from feed to milk, which averaged 1·224, 0·048 and 0·063 μg of EL per mg of SECO intake for treatments CO, FM, and FS, respectively. Digestibilities of DM, N and acid detergent fibre were similar among diets (Table 3). Digestibility of neutral detergent fibre tended (P=0·08) to be higher for cows fed FS compared with those fed CO or FM.
Table 3. Digestibility of dietary fractions in Holstein cows fed no flaxseed product (CO), 10% flaxseed meal (FM) or 10% whole flaxseed (FS) in the dry matter (DM). Values are least squares means with pooled se
Concentration of 16:1 in milk fat was lower for cows fed FS than for those fed CO or FM (Table 4). Concentration of cis11-18:1 in milk fat tended (P=0·06) to be lower for cows fed FS compared with those fed FM. Concentration of trans11-18:1 in milk fat was significantly higher for cows fed FS than for those fed FM or CO. Concentration of cis9,trans11-18:2 in milk fat tended (P=0·07) to be higher for cows fed FM than for those fed CO and cows fed FS had intermediate values. Concentration of cis3-18:3 in milk fat was significantly higher in milk fat of cows fed FS than in that of cows fed CO, and cows fed FM had concentrations similar to those of cows fed FS or CO. Concentration of medium-chain fatty acids in milk fat tended (P=0·10) to be higher for cows fed CO than for those fed FS (50·8 and 42·7% of total fatty acids, respectively). On average, concentration of omega 3 fatty acids in milk fat was 184% higher (P=0·001) for cows fed FM and FS than for those fed CO. The highest (P<0·001) omega 6 to omega 3 fatty acids ratio in milk fat was observed for cows fed CO (4·84) and the lowest ratio was obtained for cows fed FS (2·36) and cows fed FM had an intermediate value (2·98). Concentration of total conjugated linoleic acids tended (P=0·07) to be higher for cows fed FM than for those fed CO.
Table 4. Average milk fatty acid concentration (percentage of total fatty acids) of Holstein cows fed no flaxseed product (CO), 10% flaxseed meal (FM) or 10% whole flaxseed (FS) in the dry matter. Values are least squares means with pooled se
†a,b,c Means within a row without a common subscript differ (P<0·05)
‡ MUFA=monounsaturated fatty acids
§ PUFA=polyunsaturated fatty acids
¶ SCFA=short-chain fatty acids
†† MCFA=medium-chain fatty acids
‡‡ LCFA=long-chain fatty acids
§§ Omega 3=cis3-18:3+cis3-20:5+cis3-22:5
¶¶ Omega 6=cis6-18:2+trans6-18:2+cis6-20:4
Discussion
Feeding the two flaxseed products resulted in similar intakes of DM but different intakes of SDG. It is known that lignans in grain are concentrated in the outer fibre-containing layers (Adlercreutz & Mazur, Reference Adlercreutz and Mazur1997) which would lead to higher concentration of SDG in flax products with lower concentrations of oil. Indeed, Côrtes et al. (Reference Côrtes, Gagnon, Benchaar, da Silva, Santos and Petit2008) reported that concentration of SDG is 3·4-times greater in flaxseed hulls than whole flaxseeds (32·0 and 9·2 nmol/mg, respectively).
Concentration of ED in milk was below detection level and only EL was present in significant amounts. Côrtes et al. (Reference Côrtes, Gagnon, Benchaar, da Silva, Santos and Petit2008) have shown that the main mammalian lignan-metabolite produced by ruminal microbiota from flaxseed hulls and whole flaxseeds was EL while faecal microbiota led mainly to the net production of ED. Therefore, these results may suggest that the enhanced EL concentration in the milk of cows fed FM and FS compared with those fed CO is a result of the absorption of EL from the rumen following the conversion of SDG from flax products into EL by rumen microbiota. Other phyto-oestrogens such as isoflavones are known to be metabolized by rumen microbes (Dickinson et al. Reference Dickinson, Smith, Rander and Pemberton1988) and absorbed mainly in the rumen (Lundh et al. Reference Lundh, Pettersson and Martinsson1990). Moreover, different studies have confirmed the presence of polyphenolic compounds such as equol, daidzein and genistein (Bannwart et al. Reference Bannwart, Adlercreutz, Wähälä, Kotiaho, Hesso, Brunow and Hase1988; King et al. Reference King, Mano and Head1998) and mammalian lignan EL (Antignac et al. Reference Antignac, Cariou, Le Bizec and André2004) in the milk of dairy cows, further corroborating the hypothesis that EL may be absorbed from the rumen and transferred into milk. Although the concentration of EL in the milk of cows fed CO was lower, the concentration was above detection level. Plant lignans such as MAT, pinoresinol and lariciresinol, also are converted in mammalian lignans (Heinonen et al. Reference Heinonen, Nurmi, Kiukkonen, Poutanen, Wähälä, Deyama, Nishibe and Adlercreutz2001) and Milder et al. (Reference Milder, Arts, van de Putte, Venema and Hollman2005) reported that pinoresinol and lariciresinol are present in higher concentrations than SDG and MAT in many plants. Therefore, sources other than flax products may contribute to the presence of EL in the milk of cows fed CO.
Concentrations of EL in the milk of French cows produced from organic and conventional agriculture averaged, respectively, 39·1 and 42·1 μg/l while those of full creamed and skimmed milk averaged 47·3 and 36·4 μg/l, respectively, with a sd of 19 μg/l (Antignac et al. Reference Antignac, Cariou, Le Bizec and André2004). Moreover, the range of EL concentrations varied from 22 to 39 μg/l in milk of cows fed different clover species and concentrate supplementation (Steinshamn et al. Reference Steinshamn, Purup, Thuen and Hansen-Møller2008). Lower EL concentrations were observed in the present experiment, suggesting that differences in EL concentrations among milk samples may be influenced by the type of ingredient present in the diet and factors such as animal species. Indeed, higher milk EL concentrations were observed for cows fed white clover-grass silage compared to those fed red clover-grass silage, which was related to the higher dietary content and intake of their precursors (Steinshamn et al. Reference Steinshamn, Purup, Thuen and Hansen-Møller2008). The richest source of plant lignans is flaxseed although other feed ingredients also contain some lignans (Kurzer & Xu, Reference Kurzer and Xu1997), which may also contribute to the formation of mammalian lignans.
Although SDG intake of cows fed FM was 1·9-times higher (P<0·001) than that of cows fed FS, output of EL in milk was not different for both diets, which could partly be a result of the 8·6 unit difference in diet digestibility of neutral detergent fibre. Feeding whole flaxseed compared with flaxseed meal tended (P=0·08) to increase digestibility of neutral detergent fibre and plant lignans are known to be mainly in the fibre layer of plants (Adlercreutz & Mazur, Reference Adlercreutz and Mazur1997). Therefore, an improvement in fibre digestion could result in enhanced availability of SDG for conversion into EL by rumen microbes for cows fed FS compared with those fed FM, thus increasing the amount of EL produced in the rumen per unit of SDG consumed and thereafter transferred in milk. Similarly, increased availability of flaxseed lignans in the human digestive system following crushing and milling resulted in enhanced concentrations of EL and ED in blood (Kuijsten et al. Reference Kuijsten, Arts, van't Veer and Hollman2005).
Lower SECO intake for cows fed CO resulted in better apparent recovery rate of EL in milk than for cows fed flaxseed products. Steinshamn et al. (Reference Steinshamn, Purup, Thuen and Hansen-Møller2008) suggested that the metabolism (conjugation) in the gastrointestinal tract and the hepatic reconjugation before blood transport and secretion into the udder could be rate-limited, resulting in a higher transfer rate with low intake of phyto-oestrogens such as isoflavones than with high intake. Moreover, in the present experiment, only SDG was analysed in the diets and it is known that other precursors lead to the formation of EL. For example, Smeds et al. (Reference Smeds, Eklund, Sjoholm, Willfor, Nishibe, Deyama and Holmbom2007) recently quantified a new precursor of EL in cereals, 7-hydroxymatairesinol and CO, FS, and FM diets contained different proportions of barley and corn. However, further studies are required to better understand the mechanisms involved in the transfer of EL into milk.
Until recently, only SDG and MAT were seen as mammalian lignan precursors, but new precursors have recently been identified, of which lariciresinol and pinoresinol have a high degree of conversion (Heinonen et al. Reference Heinonen, Nurmi, Kiukkonen, Poutanen, Wähälä, Deyama, Nishibe and Adlercreutz2001). However, plant lignans such as matairesinol, pinoresinol and lariciresinol represent only around 2·4% of the total lignans found in flaxseed (Liu et al. Reference Liu, Saarinen and Thompson2006). Therefore, in the present experiment, the main source of plant lignans in cow diets would most likely be flaxseed with SDG being the main precursor for the synthesis of EL and ED, although precursors of mammalian lignans other than SDG may have been produced by the ruminal microbiota as shown by unidentified peaks in chromatographic profiles (Côrtes et al. Reference Côrtes, Gagnon, Benchaar, da Silva, Santos and Petit2008). Milder et al. (Reference Milder, Arts, van de Putte, Venema and Hollman2005) reported that pinoresinol and lariciresinol are present in higher concentrations than SDG and MAT in many plant foods consumed by man. Moreover, many cereal species may contain lignans (Adlercreutz & Mazur, Reference Adlercreutz and Mazur1997; Smeds et al. Reference Smeds, Eklund, Sjoholm, Willfor, Nishibe, Deyama and Holmbom2007). Smeds et al. (Reference Smeds, Eklund, Sjoholm, Willfor, Nishibe, Deyama and Holmbom2007) reported both MAT (0·21 and 0·42 mg/kg of DM) and SECO (1·25 and 0·42 mg/kg of DM) in the bran of corn and barley. The concentrates used in the present study were based on a mixture of corn and barley grains, which may contribute to the formation of mammalian lignans.
Similar feed intake, milk yield and milk composition among diets may indicate that flaxseed meal and whole flaxseed are two feed ingredients adequate for milk production of mid-lactating dairy cows. In general, untreated whole flaxseed is readily accepted by dairy cows and feeding up to 15% of the total DM as flaxseed has no effect on DM intake of mid- (Secchiari et al. Reference Secchiari, Antongiovanni, Mele, Serra, Buccioni, Ferruzzi, Paoletti and Petacchi2003) and early-lactating (Petit, Reference Petit2002) dairy cows. On the other hand, feeding whole flaxseed has controversial effects on milk production with no effect (Petit & Benchaar, Reference Petit and Benchaar2007) or increases (Petit et al. Reference Petit, Germiquet and LeBel2004) having been reported.
Milk fatty acid profile was slightly improved by feeding flaxseed products as shown by higher concentrations of fatty acids (e.g. omega 3) recognized as being beneficial for human health (Wright et al. Reference Wright, McBride and Holub1998). Similarly, greater trans-18:1 relative percentage in milk fat has been reported previously when flaxseed was fed to dairy cows compared with when raw cracked soybeans were fed (Dhiman et al. Reference Dhiman, Satter, Pariza, Galli, Albright and Tolosa2000) and higher omega 3 fatty acid concentrations in milk fat of cows fed flaxseed (Petit et al. Reference Petit, Germiquet and LeBel2004). Greater proportions of trans-18:1 in milk fat of cows fed FS could also result from differences in the proportion of forage in the DM. Total trans-18:1 proportion in milk fat increases more markedly with a high- than with a low-forage diet with linseed oil supplementation (Loor et al. Reference Loor, Ferlay, Oillier, Doreau and Chilliard2005) which agrees with earlier findings (Chilliard et al. Reference Chilliard, Ferlay, Mansbridge and Doreau2000) and with the greater duodenal flow of trans11-18:1 (Loor et al. Reference Loor, Ueda, Ferlay, Chilliard and Doreau2004). On the other hand, concentrations of cis9, trans11-18:2 (CLA) in milk fat were similar among treatments.
In a previous study (Gagnon et al. unpublished) infusing flax oil in the rumen or the abomasum of dairy cows supplemented with flaxseed hulls, which are rich in SDG, resulted in similar EL concentration in the rumen, suggesting that rumen microbes involved in SDG metabolism are not affected by the presence of readily available linolenic acid. On the other hand, Henderson (Reference Henderson1973) showed that the growth of some strains of important rumen bacteria such as Butyrivibrio, Ruminococcus and Methanobacterium is strongly inhibited by the presence of long-chain fatty acids and the Butyrivibrio group has been shown to have a dominant role in fatty acid biohydrogenation (Maia et al. Reference Maia, Chaudhary, Figueres and Wallace2007). The activity of β-glucuronidase, which plays important roles in the metabolism and absorption of lignans (Jenab & Thompson, Reference Jenab and Thompson1996) has been attributed to bacteria belonging to the dominant human intestinal microbiota, such as Ruminococcus, Bacteroides, Bifidobacterium and Eubacterium (Beaud et al. Reference Beaud, Tailliez and Anba-Mondoloni2005) and ruminal fluid activity of microbial β-glucuronidase was influenced by the presence of flax oil in the rumen (Gagnon et al. unpublished). Therefore, more research is required to better understand the relationships between supplementation with polyunsaturated fatty acids and SDG metabolism in the rumen.
In conclusion, the results provide new information on the conversion of plant SDG from two flaxseed products into mammalian lignans in dairy cows. The mammalian lignan enterolactone concentration in milk was increased by feeding flaxseed products but there was no difference between whole flaxseed and flaxseed meal fed at 11·1 and 9·4% of the DM, respectively. Of the two metabolites studied, enterolactone was the mammalian lignan that was detected in milk of cows. Feed intake, milk production, and milk composition were also similar for all diets, indicating that both flaxseed meal and whole flaxseed were adequate feed ingredients for milk production of mid-lactating dairy cows. However, milk fatty acid profile was slightly improved by feeding flaxseed products as shown by higher concentrations of fatty acids recognized as being beneficial for human health.
The authors thank N Clark, L Veilleux and S Dallaire for technical assistance and the Atlantic Dairy and Forage Institute, Fredericton Junction, NB E5L 1R1 for their financial help. This project was sponsored in part by the Matching Investment Initiative of Agriculture and Agri-Food Canada.
