Voi 6 (19971: 161-172. Effects of gradual replacement of rapeseed cake with linseed cake in a grass silage-based diet for dairy cows MarkettaRinne, Seija Jaakkola Agricultural Research Centre ofFinland, Institute ofAnimal Production, FIN-3 1600 Jokioinen, Finland, e-mail: marketta.rinne@mtt.fi Matti Järvi Koivikko Agricultural College, FIN-9 1500Muhos, Finland Pekka Huhtanen Agricultural Research Centre of Finland, Institute ofAnimal Production, FIN-3 1600 Jokioinen, Finland Eight Finncattle cows were used in two replicated 4x4 Latin squares with 21 -day periods to study the effects of replacing rapeseed cake with linseed cake in proportions of 0, 1/3, 2/3 and 1 (on air dry basis), the total amount of supplement being 1.5 kg/day (on air dry basis). The basal diet consisted of silage fed ad libitum and a 4.5 kg/day (on air dry basis) barley:oats (1:1) mixture. The experimental diets had no effect on feed intake. Effective protein degrability (EPD) deter- mined by the nylon bag method was higher for linseed cake than for rapeseed cake. Milk production decreased linearly (P<0.01), from 18.5 to 17.1 kg/day, when the proportion of linseed cake was in- creased. Milk fat content increased (P<0.05) by 3.4 g/kg but milk protein content tended to decrease (P<(). 10) with an increase in linseed cake feeding. Despite clear differences in the fatty acid compo- sition of linseed and rapeseed oils, experimental treatments had only minor effects on milk fatty acid composition. Several factors, including the slightly higher ether extract content, higher EPD and/or lower amino acid content of linseed cake than rapeseed cake, the different fatty acid composition of the two sup- plements and the presence of antinutritional compounds in linseed cake, may be responsible for the impaired milk production with linseed cake feeding. Key words: flaxseed, Linum usitatissimum L., milk production, milk fatty acid composition, oilseed cake Introduction Linseed or flaxseed (Linum usitatissimum L.j is one of the world’s oldest and most versatile crops but its cultivation in Finland had almost ceased by the 19605. A new interest in cultivating lin- seed, especially the oil linseed varieties, has emerged in the 19905. Linseed oil and meal are recognised to be highly nutritious to humans whereas linseed oil can be used in paints and varnishes, and the fibre of oil linseed in techni- © Agricultural and Food Science in Finland Manuscript received February 1997 161 AGRICULTURAL AND FOOD SCIENCE IN FINLAND https://www.c-info.fi/en/info/?token=FE7LwnwtE1HCL98e.IouLG6MPAHl2WO-ukOQwUg.nM2nYMeR27XfXB7zYELhVhGsoyzwPKRP3o7mBs0qTRqUizAEeyNyYtcB0s_3ALVjyUkeznE_LWrZk4_6kouhWDciD_zkLl3D-Pk6dsvsfcOxsoe4P_0KkYf0zICc4WKsL_l-_qohk9cRf_bphJBr0d1r6GaYRe3mka63379Y9Gsr7EqceN8MJ5_HRxvtLBYcteJ89lKmqSz8174H2ngaj4IesFYgMjhk5ju_ZgR_p1Y6KwjE80KD7qISGUnRBC6B0qVXPac_zKEDFwowxyp8QueIIXtRW8NEeKsyhSSx5UbROY-il6qGkEM74ksHGhgr8TS-ppFjTA cally innovative developments in, for example, hard board production (van Kempen and Jans- man 1994). Once the oil has been removed, lin- seed meal or cake can be used as animal feed. Rapeseed products are the most important pro- tein supplements used in dairy cow feeding in Finland, and their effects on milk production have been extensively studied (e.g. Tesfa 1992, Tuori 1992). Knowledge of the value of linseed products is, however, more scarce. In a histori- cal perspective, linseed-based feeds have been highly appreciated, but in more recent times a range of negative factors such as antinutritional compounds and the poor amino acid composi- tion oflinseed products have been reported (Ols- son et al. 1988, van Kempen and Jansman 1994). Apart from the interest in utilizing by-prod- ucts of linseed oil production for feeding dairy cows, consumers are becoming increasingly aware of the health aspects of food and foods promoting health, i.e. functional foods. If milk fatty acid composition could be manipulated by feeding cows linseed oil, it might be possible to create special milk products with health-promot- ing effects. Rumen microbes are known to bio- hydrogenate dietary unsaturated fatty acids. Milk fatty acid composition can, however, be changed by feeding linseed oil to dairy cows (McDonald and Scott 1977, Kennelly 1996). Studies of typ- ical Finnish dairy cow diets based on grass si- lage supplemented with locally produced linseed cake are, however, lacking. The objective of this study, then, was to com- pare the production potential of linseed cake with that of rapeseed cake in milk production. Spe- cial attention was paid to the feasibility of alter- ing milk fatty acid composition by feeding cows linseed cake. Material and methods Animals and basal care Eight autumn-calving Finncattle cows, of which five were in their first lactation and three in their second lactation, were used. The cows had calved 70 (s.e. 8) days before the start of the experi- ment and their average live weight during the experiment was 468 kg (s.e. 17). They were fed and housed in individual stalls in the dairy barn at Koivikko Agricultural College. Grass silage was given twice daily ad libitum and concentrate feeds were offered as two equal meals at 06.30 and 14.30. The cows had free access to water, and a commercial mineral mixture was included according to their requirements. Experimental design The experiment was conducted as two replicated 4x4 Latin squares withperiods of 21 days. The Latin squares were balanced for carry-over ef- fects, within a square each treatment being once first and once preceeded by all other treatments. Experimental diets were formed by gradually replacing rapeseed cake (RC) with linseed cake (LC) as the protein supplement. The proportion of LC was 0, 1/3, 2/3 and 1 in the four treat- ments. The total amount of protein supplemen- tation was 1.5 kg of air dry feeds per animal/ day. Feeds The linseed cake was produced by Elixi Oil Ltd. and had been cold-pressed from oil linseed, va- riety Noriin. The rapeseed cake was a commer- cial heat-moisture -treated (Öpex®) product. The cows were also given a mixture of 4.5 kg (on air dry basis) of rolled barley and oats (1:1). Grass silage was produced from timothy-meadow fes- cue (4:1). It was ensiled in clamp silos with a formic acid-based additive (AIV 2) at a rate of4 litres of formic acid per tonne of fresh grass. Measurements and sampling Data from the last week ofeach period were used to calculate concentrate intake. Silage intake was 162 Rinne, M. et al. Linseed cake in a grass silage-based dietfor cows AGRICULTURAL AND FOOD SCIENCE IN FINLAND Vol. 6(1997): 161-172. measured over the last 3 days only, a relatively short period due to practical reasons. Represent- ative feed samples were collected over each pe- riod of intake measurements. Fresh silage sam- ples were stored at -20°C before analysis. Sam- ples of concentrate feeds were bulked over all periods, but silage samples were analysed sepa- rately for each period. Milk production was re- corded during the last week ofeach period. Milk samples, in proportion to yield, were taken on the last four consecutive milkings of each peri- od, and analysed for fat, protein and lactose us- ing an infra-red milk analyser and for urea with an enzymatic UV test (Valio Ltd.). For milk fat- ty acid (FA) analyses, samples of two replicate animals were bulked. Chemical analyses The dry matter (DM) content of feed samples was determined by oven drying at 103°C. Silage DM was corrected for loss of volatile substances [lactate, volatile fatty acids (VFA), ethanol and ammonia] according to Huida et al. (1986). The ash content of samples was obtainedby ignition in a muffle furnace at 550°C for 6 h. Concen- trate feeds were analysed for nitrogen (N) by the method of Sweeney (1989) on a Leco FP 428 nitrogen analyser, and fresh silage samples by the Kjeldahl method (EEC 1993). The crude pro- tein (CP) content was obtained by multiplying the N content by 6.25. Silages were analysed for water soluble carbohydrates (Somogyi 1945), lactate (Barker and Summerson 1941), VFA (Hu- ida 1973), ethanol (Huida 1982) and ammonia N (McCullough 1967). Crude fibre (EEC 1992) and ether extract (EEC 1984; method A) analy- ses were conducted according to the official pro- cedure for feed analysis. For concentrate feeds, the samples were treated with HCI before ether extraction (EEC 1984; method B with ethyl- ether). Neutral detergent fibre (NDF) was deter- mined according to Van Soest et al. (1991). Si- lage D-value was estimated in vitro with cellu- lase-based digestion (Friedel 1990). To quantify individualmilk FAs, milk fat was separated using the International Dairy Federa- tion standard method and methylated as given in Antila and Kankare (1983). The methylated FAs were separated on an HP-Innowax (cross- linked PEG) column (30 m x 0.32 mm, film thickness 0.5 (im) in a gas chromatograph Hewlett Packard 5890 fitted with a flame-ioni- zation detector and an automatic injector (Hewlett Packard 7673). The carrier gas used was helium and the split ratio was 50:1. The injector and detector temperatures were 250 and 275°C, respectively. The column temperature, which was 90°C at the start, was first raised to 105°C at a rate of 5°C/min, then to 225°C at a rate of40°C/ min, and finally to 250°C at a rate of 2°C/min. Individual FAs and their mass fractions were identified using the milk fat standard CRM 164 (Pocklington et al. 1993). Feed values The metabolizable energy (ME) content of con- centrate feeds was calculated from the chemical composition and the reported digestibility val- ues (Tuori et al. 1996). Effective protein degrad- ability (EPD) in the rumen was calculated for both protein supplements from nylon bag incu- bations of 0,2, 4,8, 16, 24, 48 and 72 h in the rumen of three fistulated cows. Rapeseed cake was incubated as fed and LC was ground through a 6 mm sieve before incubation. After incuba- tion the bags were washed in a household wash- ing machine, dried at 60°C and analysed for N. The values of amino acids absorbed from the small intestine (AAT) and the protein balance in the rumen (PBV) for the feeds were calculated according to Tuori et al. (1996) but the meas- ured EPD values were used for RC and LC. Statistical analyses Data were analysed with the GLM procedure of 163 AGRICULTURAL AND FOOD SCIENCE IN FINLAND i Table 1. Chemical composition and feed values of the feeds. Barley and Linseed Rapeseed Grass oats mixture cake cake silage Dry matter (DM), g/kg 854 913 908 278 In DM, g/kg Ash 25 59 64 77 Crude protein 137 359 342 164 Ether extract 51" 160" 117" 43 21 Neutral detergent fibre 250 213 279 520 Feed values, g/kg DM Metabolisable energy, MJ/kg DM 13.0 14.1 12.8 10.9 AAT31 100 95 149 84 PBV> -25 207 115 20 " With HCI hydrolysis 2) Without HCI hydrolysis 31 Amino acids absorbed from the small intestine 4) Protein balance in the rumen the Statistical Analyses System (SAS Institute Inc. 1989) using the following model: y ijk , =H+ S, + 8(C), +p k +T, + (SxP) jk + (SxT)tl + ejjkl’ where S is a square, S(C) is a cow within a square, P is the period, T is the dietary treatment and SxP and SxT the subsequent interactions. Degradability of protein The sums of squares for treatment effects were further separated, using orthogonal contrasts, into single degree of freedom comparisons of linear, quadratic and cubic effects of dietary LC increases. Quadratic and cubic effects were non- significant in virtually all cases. Consequently only linear effects are included in the tables. Results The composition of the feeds used in the present experiment is given in Table 1. The higher ether extract content of LC explains why its ME con- tent was higher than that of RC. LC contained slightly more CP and clearly less NDF than RC. There was a marked difference in the protein values of the two supplements despite the small differences in CP content. This was due to the clearly higher EPD of LC than of RC (0.88 vs 0.61). Protein degradation curves for both sup- plements are presented in Figure I. The silage clamp was changed in the middle of the experi- ment. The values describing the fermentation Fig. 1.Degradability oflinseed cake andrapeseed cake pro- tein incubated in nylon bags in the ramen of three replicate cows as a function of time. 164 Rinne, M. et al. Linseed cake in a grass silage-based dietfor cows AGRICULTURAL AND FOOD SCIENCE IN FINLAND Vol. 6 (1997): 161-172. Table 2. Effects of gradual replacement of rapeseed cake with linseed cake on feed and nutrient intake. Proportion of linseed cake Statistical 0 1/3 2/3 1 SEM significance Feed intake, kg DM/d Silage 9.06 9.33 8.71 9.17 0.288 ns Barley and oats mixture 3.83 3.83 3.76 3.62 Rapeseedcake 1.37 0.91 0.45 0.00 Linseed cake 0.00 0.46 0.89 1.29 Total 14.26 14.53 13.81 14.08 0.343 ns Nutrient intake, g/d Organic matter, kg/d 13.38 13.63 12.97 13.21 0.318 ns Crude protein 2461 2530 2418 2445 55.2 ns Ether extract Total 743 769 759 786 17.7 ns From rapeseed cake 160 107 52 0 From linseed cake 0 73 143 207 AAT" 1349 1343 1260 1258 30.2 * PBV2 » 223 292 316 335 8.9 *** Metabolisable energy (ME), MJ/d 166.4 168.9 162.0 165.9 4.02 ns SEM = Standard error of the mean; Statistical significance: *** (P<0.001), * (P<0.05) "Amino acids absorbed from the small intestine : ' Protein balance in the rumen quality of silage in the first and second halves of the experiment were: DM 229 and 328 g/kg, pH 4.78 and 4.14, water soluble carbohydrates 9 and 93 g/kg DM, lactate 44 and 29 g/kg DM, acetic acid 33 and 11 g/kg DM, butyric acid 2 and 0 g/kg DM, ammonium N 79 and 29 g/kg N and soluble N 525 and 645 g/kg N. Values indi- cate that the first batch of silage was extensively fermented, with a rather high pH, whereas the sec- ond batch was fermented to a restricted extent. Dietary treatments had no effect on daily feed intake, but there were slightly more concentrate refusals when cows were fed high proportions ofLC (Table 2). Organic matter. ME and CP in- takes remained the same for all diets. Ether ex- tract intake showed a slight numerical increase when more LC was fed, owing to an increase of 3.6 g/kg (P<0.001) in the ether extract content of the total diet. As a result of the difference in protein degradability characteristics between supplements, AAT intake was smaller (P<0.05) and PBV intake greater (P<0.001) on high LC diets. Milk production (P<0.01) and energy correct- ed milk (ECM) production (P<0.05) declined with an increasing proportion of LC in the diet (Table 3). The decrease was smaller for ECM than for milk (5.6% v.v 8.2%) because LC in- creased the milk fat content (P<0.05). The milk protein content tended to decrease (P<0.10) and the milk urea content increased (P<0.01) in high LC diets. The daily output of milk protein (P<0.001) and lactose (P<0.01) was clearly de- creased by the gradual inclusion of LC but milk fat output was not significantly affected. The efficiency of dietary CP utilisation for milk pro- tein was impaired (P<0.01) by a gradual increase ofLC in the dietbut feed conversion in terms of AAT, DM or ME into ECM was not affected by diet. Gradual replacement of RC by LC had little effect on the milk short and medium chain (C4:O 165 AGRICULTURAL AND FOOD SCIENCE IN FINLAND Table 3. Effects of gradual replacement of rapeseed cake with linseed cake on milk production and milk composition and feed conversion. Proportion of linseed cake Statistical Ö 1/3 2/3 1 SEM significance Milk produktion, kg/d Milk 18.5 18.2 17.2 17.1 0.38 ** ECM" 20.7 20.6 19.7 19.6 0.43 * Milk composition, g/kg Protein 33.5 33.8 33.1 32.5 0.43 o Fat 43.6 44.8 46.4 47.0 0.85 * Lactose 49.1 48.5 49.0 49.0 0.20 ns Milk urea, mmol/1 6.08 6.32 6.83 6.70 0.153 ** Output in milk, g/d Protein 617 617 558 549 15.9 *** Fat 808 824 787 798 25.3 ns Lactose 906 888 827 829 25.7 * Efficiency of feed conversion Protein output/CP intake 0.252 0.245 0.232 0.226 0.0063 ** g AAT2)/kg ECM 49.2 50.0 48.0 47.6 1.62 ns ECM/DM intake 1.46 1.42 1.44 1.41 0.034 ns ME, MJ/kgECM 5.55 5.81 5.73 5.89 0.212 ns SEM = Standard error of the mean; Statistical significance: *** (P<0.001), ** (P<0.01), * (P<0.05), o(P<0.10) " Energy corrected milk calculated with the formula of Sjaunja et al. (1990) 21 Amino acids absorbed from the small intestine to C 16:0) FA composition other than a tendency (P<0.10) for the C14:l proportion to decrease (Table 4). In contrast, C 16:1 and C20:0 decreased (P