The rise in atmospheric greenhouse gas (GHG) concentrations (1) has accelerated climate change processes (2). The main GHG are carbon dioxide (CO₂), methane (CH₄) and nitrous oxide (N₂O), which have increased by 40, 150 and 20%, respectively, since 1990 (3).

With regard to CH₄, it was estimated that total emissions from solely ruminant livestock in 2014 were around 97.1 million tons (1) which constituted 18 % of the total CH₄ released into the environment (4).

During the digestive process, ruminants generate CH₄ made up of methanogenic bacteria, (5) which is expelled through burping (6,7), which has an environmental impact and constitutes an energy loss for the animal in the order of 5 to 12% (6). However, CH₄ emissions could be reduced through sanitary improvement, herd management, diet or with the inclusion of plant secondary metabolites (PSM) (5).

There are different PSM, such as condensed tannins (CT), essential oils (EO) and cyanogenic glucosides (CG), with different mechanisms of action that enable them to reduce rumen CH₄ generation (8,9). These PSM are chemical compounds synthesized by the plant itself (5) and the primary mechanism related to the reduction of methanogenesis is the modification of antimicrobial activity (7).

The reducing action of the CH₄ production from CT and EO has been widely proven with favorable results. In this regard, when 20% of Amaranthus spinosus (plant containing CT) was replaced in an experimental diet in an in vitro fermentation trial, a 26 % reduction of CH₄ was obtained (10). Moreover, when clove (Syzygium aromaticum), white thyme (Thymus mastichina) and anise (Pimpinella anisum) EO were used, CH₄ was reduced by 37, 76 and 32%, respectively (8). Likewise, when clove EO 200 mg was used, CH₄ was reduced by 84.21, 69.49 and 80.34%, respectively (9), when the substrate was a mixed diet, concentrated and hay, respectively, demonstrating the effect of PSM as CH₄ reducers.

Linamarin, a kind of CG, is found mainly in cassava (Manihot esculenta Cranz), with a higher concentration in bitter varieties. It has been observed that using cassava under in vitro conditions has an anti-methanogenic effect (11). When analyzing the concentrations of Linamarin in cassava cortex, it was found 28.40 ± 3.38 g/kg of dry weight and 7.71 ± 0.97 g/kg in fresh weight, while in cassava parenchyma it was found 14.71 ± 1.91 and 5.77 ± 0.74 g/kg in dry and fresh form respectively (12). The effect of cassava on the CH₄ reduction in vitro fermentation, was demonstrated whit the inclusion of 12 mg/DM of root and leaf cassava. The CH₄ production was 70.50 ± 1.32 and 65.70 ± 1.32 mL/g of digested cassava root or leaf DM respectively, while when cassava was not included, the CH₄ production was 74.2 ± 1.32 mL/g of digested DM (13).

Linamarin, has a potential to reduce CH₄ production in ruminants, when suitable doses are used. However, the effect of the inclusion of different doses of pure Linamarin in vitro is little known. The objective of this study was to quantify the effects of including rising doses of Linamarin on the CH₄ concentration in a period of time during in vitro ruminal fermentation.

Animal handling. Two four-years-old rumen fistulated Merino Precoz sheep were used, which were fistulated in 2014 in accordance with the protocol of the Bioethics and Animal Welfare Committee of the Faculty of Veterinary Sciences of the University of Chile. They were handled at the Dryland Young and Grazing Ruminants section of Germán Greve Silva Experiment Station, which belongs to the Faculty of Agricultural Sciences of the University of Chile, located in Maipú Commune, Metropolitan Region, Chile (Lat. 33º 28’ S and Long. 70º 51’ W; 470 m.a.s.l.). The experiment was conducted at the Laboratory of the Department of Animal Production of the Faculty of Agricultural Sciences, University of Chile.

Diet. During the experimental period, the sheep were fed 1.2 kg of alfalfa hay and 300 g of oat grain per animal per day, dispensed in two portions, one given in the morning and the other in the afternoon. Table 1 shows the nutritional composition of the diet offered and the supplies that were used as fermentation substrates (FS) for incubation.

Table 1 Table 1 Table 1. Bromatological composition of the diet supplied and components

	DM (%)	NDF (%)	ADF (%)	CP (%)	Energy (Mcal/kg DM-)
Alfalfa Hay	92.84	49.33	38.09	16.66	2.21
Oat Grain	89.89	28.69	13.79	11.06	2.72
Diet	91.37	39.01	25.94	13.86	2.50
MD: Dry Matter, NDF: Neutral detergent fiber, ADF: Acid detergent fiber, CP: Crude Protein, (Mcal/kg DM) Mega Calories per kilogram of DM.

Ruminal fluid collection (RF). On each incubation day, the collection of RF was carried out in the morning, in preprandial conditions. Two liters were collected during the three incubation days. The RF was filtered with a double-layer cotton cloth and kept in thermoses, which were kept in a container with water at 39ºC, with the purpose of maintaining a stable temperature in order to be taken to the laboratory and to carry out the inoculation.

Experimental design. This study was repeated measures with a single factor having five levels, where Linamarin (purity ≥98%; Sigma-Aldrich Chemical, Darmstadt, Germany, Cat. No. 68264-50 mg) was tested in five different dosages (0, 6.0, 13.0, 20.0, and 26.0 mg/L). Each dose represented a treatment. In each treatment, a RF sample was supplemented with the respective Linamarin dose and in vitro incubated in 5 test tubes (repetitions), measuring CH₄ concentration hourly, during 8 hours of incubation. This protocol was repeated in three consecutive days.

Inoculation. The in vitro technique described by Theodorou et al (14) was used. Twenty-five 100 mL test tubes with a rubber stopper and Bunsen valve were used. The FS, consisting of 0.5 g of alfalfa hay and 0.5 g of 1 mm-ground oat grain, was put into each tube in a Wiley mill. Each test tube was added with RF 30 mL and buffer solution 40 mL. The buffer solution composition was: 238 mL macromineral solution (5.7 g Na₂HPO₄ + 6.2 KH₂PO₄ + 0.6g MgSO*7H₂O + distilled water 1 L) plus 238 mL phosphate buffer solution (35 g NaHCO₃ + 4g NH₄HCO₃ + distilled water 1 L, pH 7.0) in 1 L distilled water (15).

Incubation. The tubes were placed on test tube racks and put into a thermoregulated chamber (Memmert 854, Germany) at 39ºC for a period of eight consecutive hours. The temperature was constantly monitored, and the tubes underwent rotary motion.

CH₄ measurements. This procedure was carried out every hour during a period of eight consecutive hours, using an RKI Eagle 2 gas monitor (RKI instruments, California, USA). This monitor is a high-precision instrument, using a thermal conductivity (infrared) sensor, capable of detecting CH₄ concentrations within a range of 0 to 50.000 ppm.

Statistical analysis. Data were analyzed by means analysis of variance, using Statgraphics 5.0 software. These were normally distributed, no differences were obtained when compared data between days of incubation, then data from different incubation days were analyzed together. The data were analyzed by time, treatment and the interaction time vs treatment when it was relevant. The differences among groups were analyzed using the Tukey test. Differences were considered when p≤0.05. Data are presented as group average±SEM. Additionality polynomial contrasts were performed to determine linear or quadratic effects.

Reduction of CH₄ concentration. Table 2 shows the general averages of CH₄ concentrations, according to each Linamarin dose, the linear and quadratic effect, as well as the significance of the treatments, time of fermentation and the interaction time vs treatment.

Table 2 Table 2 Table 2. Mean concentration values of CH4 at increasing doses of Linamarin in ruminal fermentation in vitro.

	Treatment (LIN concentration mg/L)
	0	6	13	20	26
CH4 (ppm)	2738a	2472a	2485a	2243ab	1932b
SEM1	P2			P3
	T	H	T*H	L	Q
154.2	<0.001	<0.001	<0.001	0.002	0.015
a, b Means with different letters are different (p≤0.05). 1SEM, Standard error of the mean; 2Probability of differences between treatments (T), sampling schedule (H) and interaction (T*H);3Probability linear (L) or quadratic (Q) to concentration of LIN.

The CH₄ reductions in relation to the doses (6, 13, 20, 26 mg/L) were 9.7, 9.2, 18.1 and 29.4%, respectively. When the CH4 concentration values were regressed in accordance with the treatments, a significant linear (p=0.002) and quadratic (p=0.015) effect was obtained in the reduction of CH₄ with increasing doses of Linamarin.

The analysis of the data by incubation time shows a reduction (p≤0.05) in CH₄ concentration (Figure 1), where hours 1, 2 and 3 were the higher (p≤0.05), but similar among them.

Figure 1 Figure 1 Figure 1. Reduction of methane concentration during eight hours of fermentation whit Linamarin.

A significant decrease begun at hour 4 of incubation, with a CH₄ concentration of 3.480 ± 20 ppm; this value was similar to that of hour 5. Low concentrations of CH₄ were obtained at hours 6 and 7 of incubation, with values of 1.218 ± 20 and 1.494 ± 20 ppm, respectively. Hour 8 was the lowest in CH₄ concentration with 840 ± 20 ppm. When the CH₄ concentration values were regressed as a function of incubation time, a significant decreasing linear ratio was obtained (R²=0.843; p≤0.001).

An interaction between time of incubation and treatments was obtained (p≤0.005). The less value of CH₄ concentration was obtained with the higher dose of Linamarin at hour 1. At this time, CH₄ concentration increased while Linamarin dose decreased (Figure 2), showing a dose/response effect. However, after 2 hours of fermentation, a similar CH₄ reducing pattern throughout the rest of the fermentation time was observed.

Figure 2 Figure 2 Figure 2. Effect of five doses of Linamarin on the production of CH4, during eight hours of in vitro fermentation.

No studies regarding the use and dosage of pure Linamarin in the reduction of CH₄ were found in literature. As was previously mentioned, Linamarin is a PSM present in cassava. Concentrations of Linamarin in cassava fluctuate according to the variety, place, age of the plant, harvest time and how it is supplied in diets. In one reported analysis, 538.4 ± 4.91 mg of Linamarin were found in 100g DM cassava, while 102.3 ± 0.93 mg of Linamarin was obtained from 100 mL of aqueous mixture, therefore, it is important to consider the way in which cassava is offered, when a specific dose of Linamarin must be supplied (16).

The use of bitter varieties of cassava as a food additive has been demonstrated to reduce CH₄ concentrations. In this regard, Phuong et al. (11) obtained a 50 % CH₄ reduction in a 24 hours in vitro fermentation assay when included 2.16 g/DM of bitter cassava leaves. We have estimated the amount of Linamarin included in the previously described experiment, in accordance to Maherawati et al (16), although these authors did not report the used cassava variety for the analysis. Thus, the inclusion of bitter cassava had an equivalent to 11.62 mg Linamarin. Then, the results of Phoung et al (11) are similar to the results obtained in the present study, with the dose of 13 mg/L of pure Linamarin. Other results reported by Inthapaya and Preston (17), when they supplemented with 12 g/DM of cassava leaf flour, with an equivalent of 64 mg/Linamarin, obtaining the lowest CH₄ values, with a reduction of 35.5% CH₄ at 24 hours of in vitro fermentation. These percentage of reduction is higher than the maximal reduction obtained in our study. However, the inclusion of Linamarin used by Inthapaya and Preston (17) was about 2 folds than the maximal dose used in our experiment, confirming the dose/response effect of the Linamarin on CH₄ reduction in in vitro ruminal fermentation.

The findings of this research show that the four Linamarin doses had a linear response in the reduction of CH₄ concentrations. This result is primarily related to the way in which Linamarin was used, since having ≥98% purity makes it highly available for rumen microorganisms. In this regard, when Do et al (18) used cassava leaves as substrate, plus cassava root flour as an additive, equivalent to a 10.76 mg/Linamarin, they obtained CH₄ 32.2 mL/g of substrate, while when they did not include the additive, CH₄ 43.5 mL/g substrate was obtained. Furthermore, the addition of cassava root flour to the substrate lead to reduce total gas production from 237 mL to 197 mL.

The regression of the incubation time vs CH₄ concentration, show that the addition of Linamarin into the incubation with sheep RF, presents a trend to reduce CH₄ concentration. PSM have a period of action that could vary depending on the doses given, the diet of the animal (primarily) and the antimicrobial capacity, that leads to the reduction of protozoa and methanogenic bacteria, which as a result leads to a constant reduction of CH₄ in a certain period (19,20).

Previous studies have demonstrated that fermentation is influenced by the type of process in the substrates used as Linamarin source. In this regard, Inthapaya et al (21) assessed the effects using fresh, ensiled, sun- or oven-dried cassava leaves on in vitro CH₄ production, observing that the CH₄ percentage was lower for fresh and ensiled leaves compared to sun- and oven-dried leaves, with values of 12, 11, 14 and 15% of CH₄, respectively. The fresh leaves of cassava contain a greater amount of Linamarin, therefore the reduction of CH₄ is greater in comparison to dry leaves of cassava, which due to drying process, decreases the concentration of Linamarin (21). These results are in agreement with those of the present study, in which higher doses of Linamarin results in less CH₄ concentration. However, Inthapaya et al (21), also observed that as the fermentation time elapses, the total gas production increased, as well as the percentage of CH₄ from 11 to 16% from 12 to 24 hours of fermentation. It has been observed that in long fermentation times, the CH₄ reducing activity tends to decrease, probably due to a habituation of the microorganisms to the additive inclusion and/or to a reduction of the additive effectiveness (22,23).

In conclusions this study demonstrates that the addition of Linamarin to the rumen fermentation substrate reduces CH₄ production in a dose-dependent manner. Also, the reducing action of the methanogenesis of Linamarin is dependent on the fermentation time. This study constitutes a basis for any future tests including vegetable sources of Linamarin as well as other rumen variables and longer in vitro fermentation times.

Conflict of interests

There is no conflict of interest.