CHEMICAL COMPOSITION AND ITS RELATIONSHIP WITH IN VITRO METHANE MITIGATION POTENTIAL OF SELECTED WILD LEGUME SEEDS

Background . Increasing atmospheric concentrations of methane (CH 4 ) have led scientists to examine its sources of origin. Mitigation of enteric CH 4 production by ruminants has been recognized as an important goal because it reduces greenhouse gas emission and improves feed efficiency. Objective . The study evaluated the chemical composition and its relationship with in vitro total gas (GP 24 ) and in vitro methane (CH 4 ) production parameters of five tropical wild legume seeds [( Luffa cylindrica (LC), Piliostigma thonningii (PT) , Detarium microcarpum (DM) , Daniellia oliveri (DO) and Afzelia africana (AA)]. Methodology . Chemical compositions were analysed, while total GP volume was measured and CH 4 estimated after 24 h incubation. Results. Total GP 24 produced by the seeds steadily increased and was most pronounced (P<0.05) in AA seed meal (64.71 mL/200 mg DM), and the least in LC (37.83 mL/200 mg DM). CH 4 concentration (MC) varied ( P <0.05) from 9.90 in AA to 23.93 in LC. Methane reduction potential (MRP) was higher ( P <0.05) for AA seed meal and lowest for LC. There were positive correlations ( r = 0.685**, r = 0.763* respectively) between crude protein and non-fibre carbohydrates (NFC) contents of the seeds and total gas production at 24 h incubation. Fibre fractions (NDF and ADF) were positively, ( r = 0 . 978 and r = 0.874 respectively) correlated with MC, and negatively ( r = - 0.927 and r = - 0.870, respectively) associated with total GP 24 and MRP . CP, EE and NFC had a more pronounced positive correlation ( r = 0.948**, r = 0.851** and r = 0.852** respectively) with MRP. Implication . Results suggest that all the selected seeds have the potential to reduce methane production and positively impact rumen fermentation. Conclusion . Seed containing more nutrients (CP, EE, and NFC) reduced enteric methane production more than any other of the chemical components in the study.


INTRODUCTION
Climate change is transforming the planet's ecosystems and threatening the well-being of current and future generations. One of the main greenhouse gases (GHG) is methane (CH4), which has a heat-trapping potential 23 times that of CO2 (IPCC 2001). Increasing atmospheric concentrations of methane have led scientists to examine its sources of origin. On a global scale, enteric CH4 produced by domesticated ruminants contributes 8% of total GHG emissions (Black et al., 2021)). Methane production by ruminants also results in a loss of energy intake of up to 2% to 14% of gross energy (Johnson and Johnsons, 1995). Therefore, reduced CH4 production by ruminants has been recognized as an important goal because it reduces GHG emission and improves feed efficiency. A number of dietary and management mitigation options and policies have been advocated for lowering methane production from livestock production systems (Hristov et al., 2013). Methane reduction strategies have included the introduction of methane inhibitors, both biological and chemical, in the animal feed, either to kill or at least reduce the activity of the methanogenic microbiome in the lower gut. Such mitigation options include the use of plant secondary compounds (Soliva et al., 2004). Many studies reported that tannins and saponins containing plants appeared to be useful in suppressing methane release by reducing the activity of rumen ciliate protozoa and methanogens (Anantasook et al., 2014). Also, the nutritive value of feed in terms of crude protein and types and concentration of carbohydrate and fibre fraction as well as their intake can influence the quantity of CH4 generated during enteric fermentation (Singh et al., 2016). Therefore, the objective of this study was to determine the in-vitro gas production (GP), enteric methane reduction property as well as correlation between chemical composition and methane reduction potential of different tropical wild legume seeds :Afzelia africana, Daniellia oliveri, Luffa cylindrica, Piliostigma thonningii and Detarium microcarpum (Keay, 1989) that can be used as alternative plant protein supplements for ruminants.

Description of the experimental site
This study was carried out at the Federal College of Wildlife Management, New Bussa, Niger State, Nigeria. New Bussa is located at longitude 9° 81̍ 95̋ N and 9° 49̍ 10̋ N and latitude 4° 58̍ 05̋ N and 4° 34̍̍̍̍ ̍ ̍ 49̋ N in the Savanna Areas of Niger Basin, North Central Zone of Nigeria.

Chemical analysis
Five tropical wild legume seeds (Luffa cylindrical, Afzelia africana, Daniellia oliveri, Piliostigma thonningii and Detarium microcarpum) were collected from plant stands naturally established within the rangeland of Federal College of Wildlife Management, New Bussa, Niger state Nigeria. The seeds were dried at 60 o C for 48 h and the ground using 1 mm sieve with electrically operated Wiley mill. The milled samples were analysed for dry matter (DM), crude protein (CP), ether extract (EE), crude fibre (CF) and ash according to AOAC (2005) and neutral detergent fibre (NDF) and acid detergent fibre according to Van Soest et al. (1994). Condensed tannins and saponins were determined by the methods of Babayemi et al. (2004). Nonfibre carbohydrates (NFC) were calculated as (100−CP−NDF−EE−ash).

Source of inoculums
In vitro GP (total gas and CH4) determination from the incubation of selected wild legume seeds was carried out by using rumen liquor from three Red Sokoto goats fed a mixed diet of Panicum maximum (60% DM) and concentrates (40% DM). The animals had free access to water and mineral. Rumen fluid was collected from the goats with the use of suction tube prior to morning feeding in a pre-warmed steel Thermos flask and immediately brought to the laboratory for analysis. The collected rumen liquor was strained through four layers of cheesecloth and kept at 39°C. All laboratory handling of rumen fluid was carried out under a continuous flow of carbon dioxide.

In vitro gas and methane determination
The in vitro GP was determined according to Menke and Steingass (1988). Samples (200 mg) of the oven-dried and milled seed were accurately weighed into 100 ml glass syringes fitted with plungers. In vitro incubation of the samples was conducted in triplicates. Syringes were filled with 30 ml of medium consisting of 10 ml of rumen fluid and 20 ml of buffer solution (g/liter of 1.985 (Na2)HPO4 + 1.30 2KH2PO4 + 0.105 MgCl2.6H2O + 1.407 NH2HCO3 + 5.418 NaHCO3 + 0.390 Cysteine HCl + 0.100 NaOH) and three blank samples containing 30 ml of medium (inoculums and buffer) only were incubated at the same time. The syringes were placed in a rotor inside the incubator (39°C) with about one rotation per min. The gas production was recorded at 3, 6, 9, 12, 18, 24, 36 and 48 h. At post-incubation period, 4 mL of 10 M (NaOH) was dispensed into the each incubated sample. Sodium hydroxide was added to absorb carbon dioxide that was produced during the process of fermentation and the remaining volume of gas was recorded as methane (Fievez et al., 2005).
Methane production reduction potential (MRP) was calculated by taking the highest % net methane values for the control (Luffa cylindrica) as 100 %. MRP was calculated with respect to methane concentration for hay since all the samples studied can be regarded as forage: MRP = %Net methane in control − %Net methane in the test / %Net methane × 100

Statistical analysis
Data were subjected to one-way ANOVA in a completely randomized design using version 9.1 of SAS software (SAS Institute, 2012). The significant difference between individual means was separated by the tukey test of the same software. Mean differences were considered significant at P < 0.05. Correlation and regression options of the SAS were used to test the relationships between the seed meals nutrients and GP at 24 h incubation, MC and MRP.

Chemical compositions of selected wild legume seed meals
The chemical composition of the legume seeds is shown in Table 1. There were variations (P < 0.05) in the chemical composition of the wild legume seeds. Crude protein (CP) was highest (P < 0.05) in AA and least in LC. Highest and lowest (P<0.05) levels of EE were observed for PT and LC respectively. Non-fibre carbohydrate (NFC) was highest in AA and DO and lowest in LC (P < 0.05). Concentrations of NDF and ADF were the greatest and the lowest in AA and LC, respectively (P < 0.05). Higher concentrations of tannins and saponins were recorded for LC, while DM seed had the lowest values of both (P < 0.05).

In vitro gas and methane production of selected legume seed meals
The net gas volume and CH4 produced increased steadily and significantly (p<0.05) at all stages of incubation (Table. 2). The highest total gas volume (NGV) and lowest methane production after 24 h of incubation were produced by AA seed meal, while the lowest NGV and highest methane were produced by LC seed meal.

Methane concentration and methane reduction potential
The amounts of methane concentration (MC) and MRP produced by the different tropical wild legume seed meals are shown in the table 3. Although, methane production from feed fermentation implies an energy loss to ruminant animals, the actual CH4 concentration production of any feed is obtained when CH4 production is expressed as the percentage of total gas volume produced. The percentages of MC produced and MRP after 24 hours of fermentation varies between 13.86 and 23.94 mL/200 mg DM, and 75.85 and 82.89 mL/200 mg DM respectively.

Correlations coefficients
Relationship between chemical composition and in vitro GP at 24 h, MC and MRP are presented in Table 4. The CP content of the seed meals was negative correlated (r = -0.915; p = 0. 001) with MC and strongly positively correlated (r = 0.685; p = 0.005, r = 0.948; p = 0.001) with GP and MRP respectively. The EE is negatively correlated (r = -0.562; p < 0.029, r = -0.722; p < 0.001) with NGV and MC at 24 hr incubation periods and strongly (r = 0.851; p < 0.001) correlated with MPR. Positive correlation was observed between NFC content, NGV and MRP at 24 hr incubation periods (r = 0.763, r = 0.852; p < 0.001). NFC had strong negative relationship (r = -0.830; p < 0.001) with MC. Fibre fractions (NDF and ADF) were strongly negatively correlated (r = -0.927; r = -0.870; p < 0.001) with in vitro GP24 and MRP (r = -0.947; r = -0.885; p < 0.001) respectively at 24 hr incubation periods, while there are strong positive correlations (r = 0.874; r = 885; p < 0.001) between MC and NDF or ADF contents of wild legume seed species. No relationship was observed between either condensed tannins (CT) or saponins and in vitro GP24 and methane production parameters.

Chemical compositions of selected wild legume seed meals
The chemical compositions of all the legume seed meals were comparable with earlier reports (Adubiaro et al., 2011;Fasoyiro et al., 2012). 4 However, lower or higher values might have been reported in previous literatures, chemical composition for forages is subjected to wide fluctuations depending largely on soil, parts of plant analyzed and climate characteristics. Chemical composition of the legumes seed meals indicates their nutritive potential as high quality feedstuffs in the diets of livestock. In this study, the least concentration of CP we recorded exceed 10 g/100 g DM for ruminant maintenance requirements and about 19 g/100 g DM for high-producing dairy cows or young growing stock (Waghorn and Clark 2004). This justifies the use of all the legume seeds studied as supplements to poor quality natural pastures and crop residues that are deficient in CP. The quantity and rate of fermentability of NFC affect ruminal pH, volatile fatty acids production and incorporation of the ruminal NH3-N into microbial protein. The NFC contents of the selected seeds are adequate to stimulate NH3-N utilization in the rumen (Tylutki et al., 2008). The optimal concentration of NFC is important in ruminant diets to avoid acidosis and other metabolic problems. The moderate NFC of the legumes seed meals is of nutritional benefits because the quantity and rate of fermentability, ruminal pH, volatile fatty acids production and incorporation of the ruminal NH3-N into microbial protein (Olafadehan et al., 2016).
The fibre fraction (NDF and ADF) contents of the wild legume seeds were generally moderate and within the limits established by Harper and McNeill (2015) for ensuring proper digestion and rumination in ruminants animals. The fibre contents of the browse seeds indicate their high nutritive value since fibre plays a significant role in voluntary intake, ruminal fermentation and digestibility (Okunade et al., 2014).
The generally moderate concentration of tannins and saponins of the seed meals is indicative of the possibility of utilising the seeds meals as a feedstuff without jeopardizing the performance and health of animals consuming such feed, particularly ruminants. Sheep and cattle can tolerate 20-50 g/kg DM condensed tannins (CT), unlike goats which can tolerate up to 100 g/kg DM (Adissu 2016). Saponins levels in all the samples were lower than the tolerable level of 15-20 g/kg DM reported for goats (Onwuka 1983), which suggests that the levels reported herein are not likely to affect nutritional potentials of the seed when fed to ruminants.

In vitro gas and methane production of selected legume seed meals
The values of GP24 and CH4 production in the current study are relatively higher than those reported by Seifdavati and Taghizadeh (2012) for some legume seeds. Differences in the substrate chemical composition and management of the inoculums donors may cause the disparity between results. Higher GP24 and CH4 production of AA relative to other seed species indicates superior ruminal fermentability, in response to the higher CP, lower fibre fractions and higher NFC. Soluble carbohydrates (NFC) do not bind with tannins and are, therefore, readily fermented in the rumen to release gas. The low gas produced by LC seed meal after 24 hours may be related with its relatively high fibre fractions, CT and saponins and low level of NFC. Plants' structural carbohydrates and secondary metabolites have been reported to inhibit fermentability which affects gas and methane production (Delgado et al., 2007;Okunade et al., 2014). Phenolic compound such as tannins or saponins may impair microbial fermentation and gas production due to their bactericidal and bacteriostatic effects on the rumen microbes (Ebert et al., 2017) and inactivation of microbial enzymes. This may be responsible for low NGV and methane generation observed in LC with higher level of tannins and saponins compared with other seed meals In the current study, it appears the NFC and fibre fraction influenced in vitro gas and CH4 production more than any other chemical constituent. The result collaborates the reports of (Fluck, et al., 2013;Kulivand and Kafilzadeh, 2015). However, in the present study we opine that the amount of GP24 and CH4 produced generally could not have been as a result of phenolic compounds in the legume seed meals because the source of inoculums is from goat. Goats in the tropics can tolerate as high as 10 g/100g DM tannins content which is higher than the highest level recorded. Goats in the tropical environment have been reported to have evolutionary adaptation for high tannins rich browses (Yisehak et al., 2016). Likewise, saponins concentration recorded in this study was lower than the tolerable level of 1.2 -2.0 g/100g DM for ruminant animals (Onwuka 1983).

Methane concentration and methane reduction potential
Methane concentration production after 24 h of anaerobic fermentation can be assessed to rank the feedstuffs in terms of anti-methanogenic potential (Uslu, et al., 2018). The lowest MC and consequently the highest percentage MRP observed for AA and other seed meals may be as a result of its lower fibre fractions and higher NFC levels compared to other legume seed meals. Studies have shown that low fibre fraction (ADF and NDF) and high NFC (easily fermentable carbohydrates) produce low MC and consequently high MRP (Uslu, et al., 2018). The results suggest that among the legume seed meals, Afzelia seed meal was the best potential protein source that could be used in ruminant diets to reduce loss of dietary energy through methane production and consequently mitigate enteric methane production contribution from livestock to global warming.

Correlations between chemical composition and response variables
Reports have shown that the volume of gas produced during anaerobic fermentation is related with differences in chemical compositions in the feeds and the availability of those components for rumen microorganisms (Elghandour et al., 2015;Pilajun and Wanapat 2018). The strong positive correlation between CP content of the seed meals and total GP may be attributed to the generally high CP of all the legume seed meals. Normal rumen microbial activities get hampered when dietary CP is below the threshold of 8 % DM which is the minimum level required for optimal ruminal microbial function (Norton 2003). The far higher CP of the seed meals than this threshold possibly increased the microbial multiplication activities and fermentation, resulting in enhanced in vitro GP. Parallell results were obtained in previous studies (Karabulut et al., 2007;Njidda and Nasiru 2010;Elghandour et al., 2017). The significant negative relationship between CP and MC and strong positive correlation of CP with MRP are desirable as is indicative of decreases MC and increased MRP. Several reports on correlations between CP and GP, MP and MRP are not consistent. Kalivand and Kafizadeh (2015) obtained a positive but insignificant correlation between CP and MC and a negative insignificant relationship between CP and MRP. Inconsistence between results may be due to variations in substrate, chemical composition and the diets of the animals from which inoculums were collected. The negative significant correlation between EE and in vitro GP24 and methane concentration and its consequent positive correlation with MRP in our current study confirm earlier findings that dietary fat inhibits in vitro GP24 and methane production in the rumen due to suppression of some rumen microbiota, particularly protozoa (Yusuf et al., 2009), although the extent of the reduction varies depending on types of fatty acid, ruminant species used and the inclusion level, diet, (Patra 2013;Patra, 2014;Dey et al., 2018).
In vitro GP24 and MRP increased as NFC increased. Generally, the types of carbohydrate present in the diet to a larger extent dictate the volume of in vitro GP24 and methane production. Readily fermentable carbohydrates (NFC) diets are fermented very rapidly by rumen microbiome with concomitant increase in lactic acid and volatile fatty acids production, particularly propionate (Olafadehan et al., 2016), resulting in increased gas production and reduced pH which inhibits protozoal and methanogen activity and hence CH4 production. Similarly, more propionate production reduces hydrogen availability for methanogensis (CH4 production) by the protozoa and methanogens. The results agree with reports of Dong and Zhao (2013) and Kulivand and Kafilzadeh (2015) who reported that the carbohydrate fractions were closely correlated to the in vitro rumen total gas and methane production.
The negative correlation between fibre fractions (NDF and ADF) and GP may be a result of bulkiness of structural carbohydrates (Okunade et al., 2014;Olafadehan et al., 2014) which affects ruminal fermentation due to reduced microbial activity (Isah et al., 2015) and hence GP (Kamalak et al., 2005). The negative relationship between fibre fraction and MRP indicates that a diet with a higher fibre level would reduce methane mitigating potential, in agreement with earlier finding (Kulivand and Kafilzadeh, 2015). Therefore, fibrous feeds would contribute more to anthropogenic methane emissions from livestock. The result is further buttressed by the positive correlation between fibre fraction and MC, implying that fibrous feeds could increase methane production. Ruminal fermentation of structural carbohydrates favours the synthesis of acetic acid and production of H2 which is used to reduce CO2 to CH4 (Kennedy and Charmley, 2012). This result is consistent with earlier reports by Heidary and Kafilzadeh (2012).
Condensed tannins and saponins have been reported to decrease the population of protozoa, which have symbiotic relationship with methanogen (Ningrat et al., 2017), resulting in reduced GP, methanogenesis and methane production. In addition, beneficial effect of moderate CT in methane abatement has been reported Olafadehan, 2019, Olafadehan et al., 2020). Though insignificant, the negative relationships between CT or saponins and NGV24, MC and weak positive relationship with MRP confirms CT and saponins to some extent reduced MC and consequently increased MRP in this study. Although, some previous reports contradict positive correlation between CT and MC and MRP (Oliveira et al., 2007;Beauchemin et al., 2007), the earlier reports of Jayanegara et al. (2009) and Piñeiro-Vázquez et al. (2015) agree with present study. The discrepancies in the effect of tannins on methane reduction potential, may be as result of doses, types, molecular weight, sources of tannins and quality of diets (Belete and Abubeker, 2018;Agrawal et al., 2014;Jayanegara et al., 2011). Likewise, previous studies (Goel and Makkar 2012;Liu et al., 2019) suggest that saponin supplementation in the diet of ruminant animals could reduce methane emission by inhibiting the growth of ruminal methanogens and protozoa, and may have different effects on cellulolytic bacteria. Weak relationship between saponins and NGV24, MC and MRP we observed in this study could be as a result of low concentration and source of saponins present in the seed meals. The result is in agreement with earlier reports (Patra and Yu, 2015;Ramírez-Restrepo 2016). Overall, anti-methanogenic property of any phenolic rich browse plant depends not only on the presence of one of phenolic compound, but the presence of other phenolic compounds.

CONCLUSION
All the studied tropical legume seed meals have good nutrient profile, moderate and safe levels of condensed tannins and saponins, relatively high in vitro gas production and methane mitigation potential, which qualify them as suitable feed supplements to low quality basal diets or as alternative protein sources in the diets of ruminants. However, Afzelia africana demonstrated superior feeding potential over other tropical wild legume seed meals based on its nutrient profile, gas production, as well as enteric methane mitigation potential. It may, therefore, be used as alternative protein source for sustainable and environmentally friendly with livestock production. Further in vivo trials are, however, required to confirm its nutritive value and methane mitigation potential as a protein feed ingredient.