In vitro ruminal biohydrogenation of C18 fatty acids in mixtures of Indigofera zollingeriana and Brachiaria decumbens

This research was aimed at studying the in vitro ruminal biohydrogenation (BH) of C18 fatty acids (FA) in mixtures of Indigofera zollingeriana and Brachiaria decumbens. Four combinations of experimental rations of I. zollingeriana : B. decumbens were tested i.e., IZ 1 (45%:45%), IZ 2 (60%:30%), IZ 3 (75%:15%), and IZ 4 (90%:0%). The remaining 10% in in each ration was rice bran. The experimental design was based on a completely randomized design with five replicates. Results revealed that there was a statistically significant difference (P<0.01) in the composition of C18 unsaturated FA (UFA) and saturated FA (SFA) for each in vitro incubation period of 1, 2, 4, 8 and 24 h. The highest accumulation of C18 UFA at 24 h was observed in the incubation of IZ 4 (19.87%). The BH of C18:3, C18:2, and C18:1 showed no differences (P>0.05). Composition of C18:0 after incubation showed a significant difference (P<0.01) with the lowest composition was observed in IZ 2 (22%). In conclusion, combination of I. zollingeriana and B. decumbens at different ratio has minor inhibition 124 J.Indonesian Trop.Anim.Agric. 45(2):124-135, June 2020 effect on BH of C18 UFA.

The BH process that takes place in the rumen is strongly influenced by the species and composition of plants that are consumed by cattle. The phenol component and fraction of secondary metabolite compounds from feed plants are believed to be able to modify and reduce the ruminal BH C18 UFA (Lourenco et al., 2008, Khiaosa-ard et al., 2009. Previous studies have indicated that phenol components from various plant sources were able to inhibit isomerisation and reduction phase of the stages of the metabolic process of BH C18 UFA so as to increase the composition of beneficial bypass FA C18:3, C18:2 and, C18:1. Further, the secondary compounds enhance the synthesis of bioactive conjugated FA isomers such as cis-9, trans-11 C18:2 and trans-11 C18:1 as well as reduce the proportion of C18:0 (Jayanegara et al., 2011;Buccioni et al., 2012;De Neve et al., 2018).
The strategy of reducing BH ruminal activity in vitro and in vivo using polyphenol sources at various levels with subcategories, namely total phenol, phenolic acids, total tannin, hydrolyzable tannin, condensed tannin and, flavonoids are reported to have roles as anti-bacterial and protozoa inhibitors of rumen microbial extracellular enzymes, and are more specifically reported to be able to influence the morphological shape and cell wall degradation of microbes that are directly associated with BH activity and methanogenesis (Jafari et al., 2016a;Jafari et al., 2017;Toral et al., 2018;Vasta et al., 2019). Tree legume plant group of the genus Indigofera has various phytochemical compounds (65 components) and high total phenol content so that it has the of bioactivity as an anti-microbial and natural antioxidant (Bakasso et al., 2008, Rahman et al., 2018. Therefore it can be hypothesized that the relatively abundant phenol component in the legume of Indigofera sp. can be used as a source of BH reduction agents and is expected to contribute to the increase in the accumulation of C18 UFA at simultaneously suppressing BH terminal products in the form of C18 saturated FA (SFA).
The present study adopted a predictive model approach by Makmur et al. (2019) who reported Indigofera zollingeriana and Brachiaria decumbens as a promising tropical species that have a high composition of C18 UFA as well as abundant presence of total phenolic compounds. However, effect of these tropical forage plants on ruminal biohydrogenation of C18 FA could not be found. The aim of the research was to study the effect in mixtures of I. zollingeriana and B. decumbens on ruminal biohydrogenation (BH) of C18 FA during in vitro conditions.

Experimental Diets and Chemical Analysis
As much as 3 kg of fresh edible parts (soft stems, leaves, and flowers) of I. zollingeriana and B. decumbens were harvested from tropical forages at UPT Teaching Farm, Andalas University, Padang, Indonesia. These were used to formulate experimental diets. Meanwhile, rice bran was obtained from poultry shops in the local area. I. zollingeriana and B. decumbens were selected as base materials for ration formulation because of their high C18 UFA compositions (C18:3, C18:2, C18:1) and their abundant secondary metabolite components that play vital role in inhibiting BH of C18 FA when compared to other tropical plant species used as feeds (Makmur et al., 2019;Makmur et al., 2020). At the time of collection, it was ensured that the B. decumbens samples collected were at the vegetative phase while I. zollingeriana samples were at the generative phase. These plant species have been identified as the basis of tropical forage rations without any negative effect on production performance for small ruminant farmers in Indonesia. After indoor storage overnight at room temperate, the samples were dried in an oven at 60°C for 3 h, ground and filtered through 1-mm mesh size. The samples were then constituted into the following rations: IZ 1 45% I. zollingeriana + 45% B. decumbens + 10% rice bran, IZ 2 60% I. zollingeriana + 30% B. decumbens + 10% rice bran, IZ 3 75% I. zollingeriana + 15% B. decumbens + 10% rice bran, and IZ 4 90% I. zollingeriana + 0% B. decumbens + 10% rice bran. Each ration was then analyzed by proximate analysis to determine the quantity of crude protein (CP), ether extract (EE), and ash according to AOAC (2005) standards. While the Van Soest analysis (Goering and Van Soest, 1970) was performed to determine the neutral detergent fiber (NDF). The non-fiber carbohydrate (NFC) content was based on the calculation of Sarvvl et al. (2018). Quantification of bioactive components was done by using a spectrophotometer, in which total phenols and tannin analyses were performed based on Makkar (2003) method, while the quantity of saponins was determined following Hiai et al. (1976) method.

In vitro Rumen Incubation
Incubation media in the form of rumen fluid from Kacang goat breed with ± 20 kg of live weight were obtained from the goat rearing cages in Padang region. Before collecting rumen fluid, the goats were adapted in ad libitum herbaceous forage (Passiflora foetida L.) and 250 gram rice bran per day for 14 days. After the slaughtering process, fresh rumen fluid was filtered using nylon (of 100 µm sieve size) and filled into prewarmed (39 o C) thermos flasks. The in vitro rumen process was carried out for 24 h according to Tilley and Terry (1963) method. Each treatment was repeated five times. About 2.5 g of the diet substrate was weighed and poured into a fermentation tube (250 ml). Next, 200 ml of McDougall buffer solution and 50 ml of rumen liquid were mixed at ratio 4:1 in the fermentation tube. The CO 2 gas was then pumped into the mixture for 30 s, after which the tube was sealed with a rubber cap and plastic wrap. Preparation of 0 h rumen fluid was carried out by separating out three tubes of Erlenmeyer containing rumen fluid before incubation and then storing them in a frozen state. Each fermentor was placed in a shaker water bath at a temperature of 39°C with a rotational speed of 100 rpm for 24 h. Rumen fluid collection was carried out over a period of 1, 2, 4, 8, and 24 h. After the fermentation period, each fermentor was then immersed in ice water to stop microbial fermentation activities. The seal on each fermentation tube was then opened and 40 ml of rumen fluid was centrifuged with a rotation of 3.000 rpm for 5 minutes. Supernatants were then collected for FA analysis and determination of molar concentrations of volatile fatty acids (VFA) and iso-VFA using gas chromatography based on Abdurachman and Surayah (2000) procedures. Estimation of in vitro methane (CH 4 ) emissions was done based on the concentration of partial VFA components using Moss et al. (2000) equation.

Determination of C18 FA
Fatty acid composition of experimental rations and rumen fluid were determined using an external standard solution (Sigma-Aldrich, Inc., USA) FA methyl ester (FAME). Preparation of standard FAME solution was done by adopting AOCS (1993) procedure. Extraction of fat was done using 2-3 g of each sample from the ration, Soxhlet tube was used for determination of fat content in the form of solid particles. While the Weibull extraction method was used for rumen fluid samples. About 0.03 -0.04 g of fat from the extraction results was weighed and put in a threaded tube; and was allowed to undergo methylation process until a analyte layer of hexane-methyl ester fatty acid was formed. The threaded tube was then adequately inserted into the auto-sampler vial for injection into a gas chromatography (GC) system. Analysis of fatty acid samples was done using GC Agilent 7890B (Agilent Technologies, Inc., USA) equipped with Supelco SPTM 2560 capillary columns with specifications of 100 m x 0.25 mm x 0.2 μm and flame ionized detector (FID). The injector temperature was set at 225°C and the detector temperature at 240°C. Nitrogen gas acted as a carrier gas with a flow of 18.0 cm/sec and a split ratio of 1: 100. The FA analysis method was based on AOAC (2000) method. Identification of C18 FA samples was determined by the peak retention time of the standard FAME solution (Ratnayake et al., 2006). The compositions of C18:0, C18:1, C18:2, and C18:3 were interpreted as percentage (%) FA component and the composition of total C18 FA in % ether extract.

Statistical Analysis
The experimental design for in vitro treatment of BH of C18 FA was based on a completely randomized design. The data collected were further analyzed for analysis of variance (ANOVA) using JASP software version 0.9.2. (Goss-Sampson, 2018). Data groups that showed statistical significance (P < 0.05) or higher (P < 0.01) were further analyzed using least significant difference.

Biohydrogenation of C18 FA
In general, the composition of C18 SFA in rumen fluid (Table 2) at legume-based rations of I. zollingeriana and B. decumbens increased dramatically during the incubation period 4 h. There was a significantly difference (P < 0.01) in the composition of C18 FA for each incubation period in vitro 1 -24 h. C18 UFA decreased dramatically in the incubation period in vitro. Conversely, the composition of C18 SFA increased. The highest proportion of C18 UFA (19.87%) after 24 h of incubation was recorded in IZ 4 while the lowest proportion (11.54%) was recorded in IZ 1. The highest proportion of C18 SFA after 24 h of incubation was recorded in IZ 1 (88.46%), followed by IZ 3 (88.33%) and the lowest was recorded in IZ 4 (80.13%).

Rumen Fluid Composition
The combined rations of I. zollingeriana and B. decumbens (Table 3) formed VFA in the range of 43.65-40.94%. The total VFA decreased significantly (P < 0.01) from IZ 1 (43.65 mM) to IZ 4 (40.94 mM). There was a statistically significant decrease (P < 0.01) in the iso-VFA content as the the ratio of I. zollingeriana increase in the experimental rations. The acetic acid proportion tend to increase significantly (P < 0.01) when compared to IZ 1 with the highest proportion in IZ 3 (45.84% total VFA). Propionate acid proportion increased significantly (P < 0.01) in IZ 3 in contrast to butyric acid. Methane proportion increased significantly (P < 0.01) in IZ 3 (35% total VFA). There was no significant difference (P > 0.05) between treatments for the parameters of valeric acid, acetate: butyrate ratio, and propionate: acetate + butyrate ratio.

DISCUSSION
Differences in the contents of C18 FA and C18 UFA in the experimental rations are closely related to the compositions of the basic ingredients of the rations. An increase in the ratio of legumes I. zollingeriana from 45% to 90% and a decrease in the ratio of B. decumbens from 45% to 0% indicates an increase in the composition of the C18 FA ration. This shows that the majority (> 80%) of C18 FA feed content was supplied from I. zollingeriana. This indicates that the content of galactolipid (high composition of C18:3) in the specific form of monogalactosyldiacylglicerol (MGDG) and digalactosyldiacylglycerol (DGDG) present in legume plants is higher than that of  grass. which is in line with the report of Buccioni et al. (2012). The study conducted by Makmur et al. (2019) showed that I. zollingeriana had a high total PUFA and a high proportion of C18:3n-3 (% total FA), i.e., 63.3 and 47.9%, respectively. Similiar findings was reported by Cabiddu et al. (2009) who conducted a comparative study of four Mediterranean legume species namely Vicia sativa, Vicia villosa, Trifolium incarnatum, and Trifolium alexandrinum which had average FA concentrations (% FAME) in the vegetative phase for C18:3 (33.83-58.28%), C18:2 (10.59% -15.28%), C18:1 (0.76% -1.62%), and PUFA (40-60%). An interesting study conducted by Saarvl et al. (2018), who combined wild hay ryegrass (Leymus chinensis) with alfalfa hay in a composition of 134:128 g/kg -1 DM, as a total mixed ration (TMR), showed that the composition significantly increased C18 UFA intake when compared to TMR that was constituted using corn stover only.
The presence of secondary metabolite plants, namely tannin, total phenol, and saponin is expected to reduce the activity of rumen BH through the inhibitory action of BH of rumen microbes in isomeration and saturation of unsaturated fatty acids. However, in the experimental diet, the highest content (g/100g DM) for tannin (1.02), total phenol (2.21), and saponin (2.25) were not able to change the BH pattern. Concentrations of plant secondary compounds in the materials are related to their biological activity (Kondo et al., 2014). So far, tannin doses and the total phenols reported are capable of altering BH of microbial populations, increasing the escape rate of C18 UFA (C18:3, C18:2, C18:1 and trans-11 C18:1) after reducing the accumulation of C18:0 in the rumen digesta, at doses of 2 and 7% DM (Jayanegara et al., 2011;Carreño et al., 2015). Whereas the saponin component, reported at level 40 g/head of dairy cows is able to provide a significant UFA transfer to milk products (Wang et al., 2017). On the other hand, Toral et al. (2018) reviewed the effect of various forms of plant secondary metabolite constituents, e.-g. condensed tannin extract, hydrolyzable tannin, and tannin; these were able to reduce the concentration of C18:0 (<15%) in the digesta. Likewise, the use of papaya leaves as a source of polyphenols (30.31 gallic acid equivalent/g) up to a composition of 25% of the substrate, increased the population of Butyrivibrio fibrisolvens, using methanogens, besides having no effect on the total protozoa (Jafari et al., 2016b).
The decrease in the composition of C18 UFA during in vitro incubation shows that BH activity massively occurs within a 4 h period and slowly decreases the BH level to a duration of 24 h. However, the trend is inversely proportional to the level of appearance of the C18 SFA digesta. This is related to the amount of substrate availability in  the form of abundant UFA for BH microbes in the initial incubation period (0-4 hours); which over time undergoes saturation thereby reducing the composition of C18 UFA and increasing the percentage of C18 SFA as the final product of the rumen FA metabolic process. Beam et al. (2000) stated that there is a positive correlation between the FA content of the substrate and the BH level. In vitro rumen lipid metabolism studies showed the same characteristics of BH levels for C18 PUFA using a ration of subtropical plant species namely a mixture of red clover (Trifolium repens): perennial ryegrass (Lolium perenne) with a ratio of 100: 0, 75:25, 50: 50, 25: 75, and 0: 100 (Van Ranst et al., 2013). The use of a single legume red clover substrate leads to gradual increase of BH of C18:3 and C18:2 as well as biosynthesis of trans-11 cis-9 conjugated linoleic acid (CLA), and conjugated linolenic acid (CLNA) incubation period of 0-5 h; and C18:0 concentration increased significantly in 5-10 h (Van Ranst et al., 2010). Lejonklev et al. (2013) stated that the rate of BH of C18:3 and C18:2 was slower in red clover legume silage than ryegrass silage.
The profile of 24 h in vitro biohydrogenation activity in tropical forage-based I. zollingeriana and B. decumbens showed massive activity of C18:3, C18:2, and C18:1 (> 95%). This indicates that the bioactive components of plants contained in the experimental rations have not been able to inhibit the BH and increase the accumulation of C18 UFA bypassing the rumen system. A comparative study by Toral et al. (2016) concluded that the accumulation of beneficial FA using in vitro batch culture in alfalfa species (5 g/kg DM tannin) and sainfoin (35 g/kg DM tannin) has not given satisfactory results in modulating BH. Gudla et al. (2012), using alfalfa hay as a forage source with a high proportion of ration (700 g/kg DM), showed the highest synthesis yield at trans-10 C18:1 and trans-10, cis-12 CLA and concentration deoxyribonucleic acid (DNA) of Anaerovibrio lipolytica, Butyrivibrio fibrisolvens, Butyrivibrio proteoclasticum, Ruminococcus albus, and Ruminococcus flavafaciens when compared with low forage ration (300 g/kg DM). In vivo studies have shown that different tannin sources, namely quebracho and chestnut with intake levels of 5.3 and 16% DM, increased the population of rumen In vitro Biohydrogenation of Indigofera zollingeriana and Brachiaria decumbens (M. Makmur et al.) 131 BH microbial group A, B. fibrisolvens but decreased the composition of C18:0 (Buccioni et al., 2015;Buccioni et al., 2017). This has the effect of using tannin content as a reference in selecting plant species that are able to inhibit BH activities, to be less precise. Delgadillo-Puga et al. (2019), supplementing goat ration using rich-phenols legume species, Acacia farnesiana pods (10-30% diets), did not record a significant effect on increasing the total concentration of monounsaturated FA, polyunsaturated FA and decreasing n-6/n-3 in milk products. In contrast, in vitro evaluations of phenol plants from subtropical and tropical plant species have been reported to have lipolysis reduction effects, increased C18 UFA preservation capabilities and CLA formation (Cabiddu et al., 2010;Jayanegara et al., 2011). The C18:0 composition of the rumen at the end of the 24 h incubation period gave surprising results, where IZ 2 and IZ 4 showed percentages of C18:0 to be 22 and 37%, respectively. This is also possible when a higher proportion of the legume in the ration contains the C18 UFA class, namely C18:2 and C18:1 during the in vitro incubation period, thus affecting the composition of FA in the rumen digesta. In vivo studies of C18 UFA protection at Hereford x Friesian steers, using fish oil as a supplement (10-30 g/kg DM diet) in grass-based feeds and red clover-based silage showed the role of fish oil as an inhibitor of BH of C18:3 and C18:2 feed. Red clover based diets (90% dietary intake) can increase the flow of C18 UFA from the rumen to the duodenal when compared to grass silage based diets, besides giving no reduction effect of BH of C20:5 n-3 and C22:6 n-3 on both treatment (Lee et al., 2008). The composition of C18:0 in this research is not much different from a research using UFA protection technology in the form of microencapsulation (50% arabic gum: 50% maltodextrin) canola and sesame oil combined with Sapindus rarak extract at a dose of 1 mg.ml -1 which produces a range of stearic acid 22.48 -37.71% (Suharti et al., 2019).
The profile of in vitro volatile fatty acids (VFA) displayed in each experimental ration indicates an abundant production of VFA. An increase in the ratio of I. zollingeriana to B. decumbens tends to decrease acetate production and tends to increase propionate. However, the VFA production have slight difference beetwen IZ 1 (45% I. zollingeriana: 45% B.decumbens) and IZ 4 (90% I. zollingeriana). Nevertheless the VFA profile in IZ 4 is still superior when compared to VFA characteristics of 100% I. zollingeriana . This fermentation pattern is related to the decreased dietary fiber content of the experiment with an increase in the composition of I. zollingeriana, thereby reducing the formation of acetic acid and the acetate/propionate ratio as the main product of the degradation of crude fiber fraction and other organic materials by rumen microbes. This profile contrasts with the results of the study of Tarigan et al. (2018) where the use of I. zollingeriana (90% DM) as a base for concentrate showed superiority in the acetate ratio and total VFA (127.08 mM) when compared to the controls, using palm kernel meal (35% DM) and the combination of I. zollingeriana and Calliandra calothyrsus (45%: 45% DM). Suharlina et al. (2016), using concentrated feed with a lower composition of I. zollingeriana, 40% DM, showed molar production of acetate, propionate, and butyrate at the respective levels of 14.4, 4.40, and 1.44. The reduced degradation of feed protein indicates a decrease in the total production of VFA and iso-VFA. These results showed that secondary plant metabolites such as tannin, phenol, and saponin contents acted as antinutritional components and the proportions of secondary plant metabolites continued to increase linearly with increase in I. zollingeriana in the experimental ration. The inclusion of I. zollingeriana in a ratio of 75 -100% in a diet based on palm oil fronds also showed a decrease in production (g/kg organic matter) of microbial biomass and microbial nitrogen (Fakhri et al., 2017). Estimated rumen methane emissions also show a slight upward trend with increase in legume proportions. The methane profile showed in this experiment is almost close to in vitro methane gas production using ammoniated rice straw-based rations with legume (Leucaena leucocephala leaf meal) supplementation at 10 and 20% DM yields of 14.41-12.88 mM (Ningrat et al., 2019).

CONCLUSION
Increased composition of C18 UFA and the content of secondary metabolite (tannin, total phenol, and saponin) were accompanied by an increase in the ratio of I. zollingeriana to B. decumbens in the experimental ration. However, combination of I. zollingeriana and B. decumbens at different ratio had a minor effect on the inhibition of BH of C18 UFA which resulted in an extensive BH activity of C18:3 n-3, C18:2 n-6, and C18:1 n-9 during the in vitro. Interestingly, the influence of I. zollingeriana showed a lower formation of C18:0 and improved in vitro rumen volatile fatty acid profiles. I. zollingeriana might be a promising tropical plant species as part of a feeding strategy for modulation of rumen BH.