Effect of amount and source of vegetable oils in a high fibrous cattle diet on in vitro rumen fermentation, nutrient degradability and rumen cis-9, trans-11 CLA concentration

Abstract

Four diets with two vegetable oils at two different levels were evaluated for their in vitro rumen fermentation pattern, nutrient degradability and cis-9, trans-11 CLA production using a rumen simulation technique (RUSITEC). Control diet (C) was having roughage to concentrate in the ratio of 65:35. Treatment diets were formed by supplementing with sunflower oil (SF) and soya bean oil (SB) at 4.5 (SF4.5, SB4.5) and 6.0% (SF6, SB6) levels and incubated in RUSITEC fermenters. After seven days of adaptation period, fermentation parameters, nutrient degradability and rumen cis-9, trans-11 concentration were assessed for the next 3 days. Cis-9, trans-11 conjugated linoleic acid (CLA) content was higher in effluent from diets supplemented with 6% oil compared to that of control and 4.5% oil. SF6 and SF4.5 were more effective in increasing rumen cis-9, trans-11 CLA compared with corresponding levels of soya bean oil. However, the degradability of dry matter, acid detergent fibre, neutral detergent fibre and hemicellulose was significantly (P<0.05) reduced in SF6 and SB6 diets. Significantly (P<0.05) higher pH, lower proportion of acetate and butyrate and higher proportion of propionate were indicative of the aforementioned changes in the corresponding fermenters. Rumen fermentation pattern and nutrient degradability were not affected in SF4.5 and SB4.5 diets. The study indicated that vegetable oil supplementation at 6% level was deleterious for rumen fermentation and nutrient degradability despite an increase in cis-9, trans-11 CLA.

Keywords:

• conjugated linoleic acid
• degradability
• fermentation
• RUSITEC
• vegetable oils

Introduction

Animal-derived food products are known to contain some micro-components that have positive effects on human health and disease prevention beyond those associated with traditional nutritive value. The cis-9, trans-11 isomer of conjugated linoleic acid (CLA) represents one of these micro-components in animal products because of numerous potential health benefits (Belury 1995). CLA in ruminant products originates from the incomplete biohydrogenation of dietary linoleic acid (LA) and by endogenous synthesis from trans-11 C_18:1 (TVA), another intermediate of biohydrogenation, in mammalian tissues (Harfoot and Hazlewood 1988). Amounts of biohydrogenation intermediates produced in the rumen influence their concentrations in tissues or milk (Demeyer and Doreau 1999; Chilliard et al. 2000). The supplementation of plant oils rich in polyunsaturated fatty acids (PUFA) such as sunflower oil and soya bean oil is an effective strategy to increase rumen CLA and TVA and finally the milk CLA (Qiu et al. 2004; Shingfield et al. 2008).

However, the amount of PUFA that can be added in the diets of dairy cows is limited due to adverse effects of these PUFA on the metabolism of rumen bacteria, thereby impairing rumen fermentation and animal performance (Jenkins 1993). Type and source of oil and forage concentrate ratio of the basal diet are the major factors that determine the effect of lipid supplementation on ruminal fermentation (Toral et al. 2009). In most of the previous studies, which involved vegetable oil supplementation to produce CLA-enriched livestock products, simultaneous effect on nutrient digestibility had not been investigated. Therefore, it is crucial to optimise the level of supplemental fat that increases CLA without impairing rumen fermentation.

Two vegetable oils which are varying in their fatty acid composition were selected at two levels to assess the variation due to amount and source of fat on cis-9, trans-11 CLA and TVA production and rumen fermentation. Sunflower oil contains more LA compared with soya bean oil, whereas the proportion of oleic acid and LA in the latter was more. Hence the objective of the study was to evaluate the effect of supplementation of two different levels of vegetable oils varying in their fatty acid composition on rumen fermentation pattern, nutrient degradability and cis-9, trans-11 CLA and TVA production.

Materials and methods

The study was carried out using RUSITEC apparatus (Czerkawski and Breckenridge 1977) consisting of five fermenters with an effective volume of 800 ml each, immersed in a water bath maintained at 39 °C. The fermenter inoculum was obtained from three crossbred (Jersey×Red Sindhi) cows (mean body weight 350 kg) immediately after slaughter. The animals were maintained on rice straw and crushed maize in 70:30 ratio until 8 hour prior to slaughter. The fermentation inocula (solid and liquid) were withdrawn from the rumen within 1 hour of slaughter and transferred to laboratory at 39°C under CO₂ flushing. Ruminal contents were strained through four layers of cheese cloth. The fermenters were filled with 650 ml strained rumen fluid and 150 ml McDougall buffer (McDougall 1948). Eighty grams of solid inoculum was weighted into a nylon bag (100 µm pore size, 75×125 mm), which was then placed inside the feed container kept in each vessel together with a bag containing 10 g of experimental diet. The experimental feed consisted of hybrid napier grass and concentrate (65:35, DM basis; Table 1) which was dried at 60°C in a forced-air oven for 48 hour, ground to pass a 2-mm screen and supplemented with either sunflower oil or soya bean oil. The fermenters were randomly assigned to the C (control basal diet without oil), SF4.5, SB4.5 (sunflower oil and soya bean oil, respectively, at 4.5% level), SF6 and SB6 (sunflower oil and soya bean oil, respectively, at 6%). The fermenters were mixed constantly at 10 rpm via a stirrer and flushed with CO₂ gas. McDougall buffer was supplied to each vessel at a flow rate of 0.56 ml/min using a peristaltic pump. Displaced effluents and fermentation gases were collected in effluent collection vessels (cooled with ice) and gas collection bags, respectively. After 24 hours, the vessels were opened and the solid inoculum bags were replaced by new feed bags, the removed bags were squeezed, washed in artificial saliva and the washing was returned to the vessel. On subsequent days, the nylon bags that had spent 48 hours were replaced, thus achieving an incubation period of 48 hours for each diet. The first seven days were utilised for the adaptation of rumen microbes to the respective diet followed by measurement period of three days. The experiment was repeated thrice and in each run three measurements were taken for each diet.

Table 1. Ingredient and chemical composition (%) of experimental diets containing vegetable oils at two different levels.

Composition	C	SF4.5	SB4.5	SF6	SB6
Ingredient (%)
Hybrid napier grass	65	65	65	65	65
Maize	11	10.5	10.5	10	10
Pearl millet	4.2	3.5	3.5	3	3
Deoiled rice bran	10.25	8.45	8.45	7.95	7.95
Sunflower cake	7	7	7	7	7
Mineral mixture^a	0.7	0.7	0.7	0.7	0.7
Salt (NaCl)	0.35	0.35	0.35	0.35	0.35
Sunflower oil^b		4.5		6
Soya bean oil^c			4.5		6
Chemical (% DM basis)
Dry matter	91.4	91.38	91.35	91.41	91.36
Crude protein	10.38	10.07	10.01	9.68	9.58
Ether extract	1.1	5.48	5.53	7.02	7.05
Total ash	8.01	8.18	8.6	8.15	8.13
NDF	64.22	64.82	64.37	64.37	64.52
ADF	31.43	31.95	31.81	31.88	35.91
Hemicellulose	32.79	32.87	32.56	32.49	32.61
^aContains Ca 28 g/kg, P 12 g/kg, I 0.026 g/kg, Cu 0.077 g/kg, Mn 0.01 g/kg, Co 0.013 g/kg and Zn 0.16 g/kg. C, Control diet; SF, C + Sunflower oil; SB, C + Soya bean oil.
^bSunflower oil: C18:1, 15.31; C18:2, 65.23; C18:3, 0.21.
^cSoya bean oil: C18:1, 21.53; C18:2, 51.65; C18:3, 8.83.

Chemical analysis

The five experimental diets were analysed for dry matter (DM), crude protein (CP), ether extract (EE) and ash as per the method of AOAC (2000). The neutral detergent fibre (NDF) and acid detergent fibre (ADF) were determined as per the method of Van Soest et al. (1991). NDF was assayed without amylase and sodium sulphite. NDF and ADF were expressed inclusive of residual ash.

Five millilitre of effluent was collected prior to daily replacement of feed bags, stored frozen (–20°C), freeze dried and ground to pass through a 0.50-mm screen. The fatty acids (FA) from vegetable oils and effluent were extracted according to the extraction-transesterification procedure of Sukhija and Palmquist (1988). Separation of the individual FA in the samples was achieved using a gas chromatograph (model 5890, Hewlett Packard Co.) fitted with a FID and a fused-silica capillary column (SP-2560; 100 m, 0.25 mm i.d., 0.20 µm film; Supelco Inc., Bellefonte, PA). A sample containing fatty acid methyl esters (FAME) in hexane (1 µl) was injected by automatic sampling with split injection. Helium was used as carrier gas, with split ratio of 50:1 and column flow of 2 ml/min. The oven was initially kept at 140°C for 5 min, then heated at 4°C/min to 240°C and held for 35 min. The injector and detector temperature were maintained at 250°C; heptadecanoic acid (C17:0) was used as a quantitative internal standard. Each peak was identified using FAME for CLA and TVA (Nu-Chek Prep, Elysian, MN; Matreya, Pleasant Gap, PA) and 37 Component FAME mix for other FAME (Supelco Inc., Bellefonte, PA). The percentage of each FA was calculated by dividing the area under the FA peak (minus the area under the peak for heptadecanoic acid) by the sum of the areas under all of the reported peaks. All results were reported as g/100 g of FAME.

Five millilitre of culture fluid was collected from each fermenter with help of an injection syringe to study pH, ammonia nitrogen (NH₃-N) and volatile fatty acids (VFA). pH was immediately recorded and samples were then transferred into two micro-centrifuge tubes for VFA (1.1 ml of effluent plus two drops of concentrated orthophosphoric acid) and NH₃-N (1 ml of effluent plus 0.1 ml of 2 M HCl) analyses. The NH₃-N concentration was estimated colorimetrically as per the method of Weatherburn (1967). The samples for VFA were centrifuged, frozen and analysed by GLC (model 5890, Hewlett Packard Co., Avondale, PA) as per the procedure of Chase (1990). The nylon bags containing weighed amount of experimental diets were incubated for 48 hours and on removal, the bags were washed in a washing machine for 10 min and oven dried at 60°C for 48 hours. After weighing the fermentation residues, it was analysed for CP (AOAC 2000) and fibre fractions (Van Soest et al. 1991) and degradability of DM (IVDMD) and nutrients was found out.

The data on in vitro rumen fermentation parameters, nutrient degradability and fatty acid concentration in different treatment groups were subjected to ANOVA for completely randomised design using SPSS software (10.0). All the values were shown as mean and standard error of the mean.

Results

The effect of addition of vegetable oil on rumen fermentation parameter is presented in Table 2. The rumen fluid pH due to the addition of 6% vegetable oil was significantly (P<0.05) higher than that of control as well as 4.5% oil added groups. Regarding NH₃-N concentration, there was no significant difference among the control and oil-added feeds. Total VFA production was unaffected due to source and level of oil and varied from 38.33 to 41.66 mmol/day. Oil addition at 6% level significantly (P<0.05) reduced the molar proportion of acetate compared with control. The molar proportion of propionate was increased in SF6 and SB6. The molar proportion of butyrate showed a significant (P<0.05) reduction due to oil addition at both levels. Consequently, A/P ratio in the oil-added fermenters differs significantly (P<0.05) from the control and change was observable between two levels also.

Table 2. In vitro rumen fermentation pattern as affected by supplementation of vegetable oils at two levels to a complete diet.

Parameter	C	SF4.5	SB4.5	SF6	SB6
PH^A	6.97^ab±0.01	6.95^ab±0.01	6.93^a±0.02	7.06^c±0.00	7.01^bc±0.01
NH3-N (mg/100 ml)	19.46±0.56	18.89±1.12	19.93±1.50	16.56±2.02	16.80±1.93
Total VFA (mmol/day)	41.00±1.20	39.00±2.72	41.66±3.76	38.33±2.02	38.66±6.06
Acetate^A	54.79^b±1.38	54.87^b±0.48	55.21^b±2.39	51.29^a±2.61	50.69^a±1.23
Propionate^A	20.14^a±0.93	23.24^ab±2.86	24.55^ab±1.81	26.34^b±1.77	26.70^b±0.27
Butyrate^A	16.39^b±0.55	12.57^ab±2.82	14.12^ab±1.83	12.92^ab±0.88	10.62^a±0.68
A/P ratio^A	2.54^c±0.33	2.24^b±0.20	2.12^b±0.32	1.97^a±0.17	1.92^a±0.08
^AMean bearing different superscript letters in the row differ significantly (P<0.05).
C, control; SF, sunflower oil; SB, soya bean oil; A/P, acetate/propionate; VFA, volatile fatty acid.

The in vitro nutrient degradability of the experimental diets after 48 hours is presented in Table 3. The IVDMD of experimental diets ranged from 40.62 to 46.96%. The IVDMD was significantly (P<0.05) affected due to oil addition at 6% level. Moreover, the oil supplementation at 6% level significantly (P<0.05) reduced the degradability of NDF, ADF and hemicellulose. However, CP degradability was unaffected at this level. The apparent degradability of DM, CP, NDF, ADF and hemicellulose of SF4.5 and SB4.5 diets did not differ from control.

Table 3. In vitro nutrient degradability as affected by supplementation of vegetable oils at two levels to a complete diet.

Nutrient degradability (%)	C	SF4.5	SB4.5	SF6	SB6
DM^A	44.12^b±2.24	45.35^b±0.53	46.29^b±1.77	40.62^a±2.01	41.11^a±1.98
CP	41.81±0.56	41.88±0.29	42.42±0.53	41.99±0.33	42.16±0.33
NDF^A	67.04^b±2.20	68.22^b±0.03	70.59^b±2.27	63.66^a±0.54	62.15^a±0.42
ADF^A	34.98^b±0.36	36.65^b±0.01	35.99^b±0.42	31.69^a±1.10	30.50^a±2.64
Hemicellulose^A	34.25^b±0.35	36.43^b±0.01	35.05^b±1.43	31.89^a±0.48	30.59^a±0.30
^AMean bearing different superscript letters in the row differ significantly (P<0.05). C, control; SF, sunflower oil; SB, soya bean oil.

The cumulative fatty acid concentration after 24 hours (g/100g of FAME) in effluent is presented in Table 4. The proportion of C_16:0 was comparable among the treatments. C_18:0 content in effluent was significantly (P<0.01) lower in SF6 and SB6 fermenters, whereas it was significantly (P<0.01) higher due to addition of oils at 4.5% level and highest in control diet. Cis-9 C_18:1 content in different fermenters was not significantly different from each other. TVA concentration was significantly (P<0.05) lower in C fermenter and significant (P<0.05) increase was visible due to oil addition at both levels. In the effluent from oil-added diets, per cent of C_18:2 was found to be significantly (P<0.05) higher than that of control. The SF6 diet produced significantly (P<0.05) highest content of CLA (0.68 g/100 g of FAME) followed by SB6 (0.52 g/100 g of FAME). SF6 and SF4 were more effective in increasing rumen cis-9, trans-11 CLA compared with corresponding levels of soya bean oil. CLA was not even detectable in the control. C_18:3 content from different fermenters was not significantly different.

Table 4. Rumen fatty acid profile (g/100 g FAME) as affected by supplementation of vegetable oils at two levels to a complete diet.

Fatty acids	C	SF4.5	SB4.5	SF6	SB6
C16:0	23.02±0.24	22.66±0.21	23.55±0.36	23.64±0.30	23.83±0.47
C^B 18:0	34.54^c±0.33	22.91^b±0.49	23.99^b±0.57	20.35^a±0.33	20.82^a±0.27
Cis-9 C18:1	8.14±0.18	7.96±0.17	8.39±0.17	8.56±0.20	8.28±0.29
Trans-11 C18:1 (TVA)^A	8.26^a±0.27	16.8^b ±0.42	17.6^b ±0.21	21.47^c±0.59	22.40^c±0.37
C18:2	1.66^a±0.13	2.61^b±0.13	2.55^b ±0.11	2.65^b ±0.08	2.57^b ±0.10
Cis-9, trans-11 CLA^A	ND	0.47^c±0.02	0.37^b±0.01	0.68^e±0.25	0.52^d±0.21
C18:3	2.11±0.09	2.08±0.06	1.97±0.15	2.04±0.04	2.09±0.08
Note: C, control; SF, sunflower oil, SB, soya bean oil; ND, non detectable.
^AMean bearing different superscript letters in the row differ significantly (P<0.05).
^BMean bearing different superscript letters in the row differ significantly (P<0.01).

Discussion

The response of rumen pH to higher level of inclusion of vegetable oil was similar to the findings of Jalc and Ceresnakova (2001). They observed significantly higher pH values for linseed oil, sunflower oil and rapeseed oil (10%) supplemented diets compared with unsupplemented control diet. The higher pH values observed with 6% level of oil addition was an indication that total VFA production might be lowered at this level, which could be correlated with the reduced fibre digestibility.

NH₃-N concentration in all the culture fluid was found to be above 5.0 mg/100 ml of rumen fluid as suggested by Satter and Slyter (1974) for maximal microbial growth and was not affected by the supplementation of fat. Similarly, Hervas et al. (2008) observed that lipid supplementation usually results in no variation in rumen ammonia concentration. In contrast to this, ammonia nitrogen concentration was higher in RUSITEC fermenters due to the supplementation of oil blends (Jalc et al. 2006).

The pattern of total VFA (mmol/day) production observed in this study was consistent with the findings of Bateman and Jenkins (1998), who reported a non-linear reduction in total VFA concentration from diet supplemented with 0, 4 and 6% of soya bean oil. Analysis of molar proportion of individual VFA showed that oil addition at high level reduced acetate, increased propionate and reduced butyrate compared with basal diet. Acetate:propionate ratio showed a significant (P<0.05) reduction at both levels of oil addition. Similarly, higher level of inclusion of sunflower oil shifted the VFA towards propionate at the expense of acetate (Shingfield et al. 2008).

The degradability of DM, NDF, ADF and hemicellulose was significantly affected by oil supplementation at 6% level. Effect of high level of oil supplementation to bring about depression in DM degradability was also observed by Jalc and Ceresnakova (2001). In contrast to this, apparent ruminal digestibility of DM, OM, N or NDF was not affected due to addition of soya bean oil up to 8% level in a high fibrous cattle diet (Bateman and Jenkins 1998). They proposed that high fibre content of the diet promoted rapid lipolysis and hydrogenation of soya bean oil and these changes to fat sources prevented any deleterious effect on ruminal digestibility. Absence of any negative effect due to 4.5% oil supplementation can be explained based on the earlier theory. On the other hand, impairment of rumen fermentation and nutrient degradability at 6% level indicates that either the microbial population was modified or microbial activity was inhibited at this level (Devendra and Lewis 1974). CP degradability remained unaffected due to amount and source of fat and finding was in agreement to that of Shingfield et al. (2008) where OM and N digestibility did not differ significantly due to incremental level of oil supplementation in cattle diet.

In this study, ruminal nutrient degradability was not affected due to difference in source of oil. Similarly, Kalscheur et al. (1997) observed that apparent rumen digestibility of DM, OM and NDF was not affected by different fat source varying in their fatty acid composition. In contrast to this, different oil sources affected IVDMD in different ways and may be due to the corresponding changes in microbial population (Jalc and Ceresnakova 2001).

Inclusion of sunflower and soya bean oil in the diet as a source of LA enhanced the proportion of TVA and cis-9, trans-11 CLA in fermenters and reduced the proportion of C_18:0. The vegetable oil supplementation at 6% level resulted in greater cis-9, trans-11 CLA and TVA concentration compared with 4.5% oil. Considering the source, sunflower oil was more effective in increasing cis-9, trans-11 CLA at both the levels. The higher concentration of CLA resulting from the supplementation of sunflower oil at both the levels could be correlated to higher content of CLA precursor, LA (C_18:2) in it. The only source of ruminal cis-9, trans-11 CLA is the dietary LA. The higher level of cis-9, trans-11 CLA from 6% oil-added diets was similar to the findings of Shingfield et al. (2008) where incremental level of sunflower oil in cattle diet increased the flow of TVA and cis-9, trans-11 CLA at the omassal level. TVA is important because in mammalian tissues (i.e. the mammary gland) can be also desaturated to CLA. In contrast to the present findings, the changes in cis-9, trans-11 CLA were not significant due to different level of sunflower oil (Sackmann et al. 2003) or soya bean oil (Beaulieu et al. 2002) in the diet of cattle.

The TVA concentration was not affected by the source of oil. TVA is a common intermediate in the BH of C_18:1, C_18:2 and C1_8:3. C1_8:1 and C_18:2 of sunflower oil and C_18:1, C_18:2 and C_18:3 of soya bean oil contributed to TVA and hence its concentration remained unaffected due to oil source. Variation in C_18:0 was apparent due to difference in source of oil with significantly (P<0.01) lowest value for SF6 diet. In turn, the C_18:0 content in SB6 fermenter was significantly (P<0.01) lower than that of control as well as 4.5% oil-added diets.

All the earlier changes could be accounted to the inhibition of biohydrogenation pathway due to oil supplementation and oils at 6% level inhibited the pathway to greater extent than 4.5% level. Concentration of LA exceeding 1 mg/ml of culture contents was reported to interfere with the final step of biohydrogenation leading to accumulation of TVA (Harfoot and Hazlewood 1988).

It was concluded that even though CLA and TVA in fermenter fluid were increased by 6% oil supplementation, rumen fermentation and nutrient degradability were hampered at this level. However, oil supplementation at 4.5% level was safe for rumen fermentation and nutrient degradability, simultaneously increasing the content of rumen CLA and TVA. In addition, the study recommended the addition of dietary oil at 4.5% level especially sunflower oil in terms of CLA production.