In Vitro Screening of East Asian Plant Extracts for Potential Use in Reducing Ruminal Methane Production

Simple Summary Methane from ruminants is a major contributor to total greenhouse gases. Therefore, ruminant nutritionists have proposed strategies to mitigate methane emissions, such as chemical inhibitors and ionophores. However, dietary manipulation including natural feed additives is more practical, considering consumer preferences. Therefore, the current experiment screened 137 plant species, indigenous to East Asian countries, to select novel anti-methanogenic candidates as natural feed additives. Among these species, an extract from the seeds of Pharbitis nil exhibited a maximum 37% reduction of methane in a conformation assay. Identification of active compounds present in the seeds of Pharbitis nil revealed enrichment of polyunsaturated fatty acids, which were dominated by linoleic acid (18:2). The extract had negative effects on the populations of ciliated protozoa and H2-producing Ruminococcus flavefaciens, thereby increasing the proportion of propionate, similar to the effect of monensin. This is the first report to suggest that the seeds of P. nil could be a promising anti-methanogenic alternative to ionophores or oil seeds. Abstract Indiscriminate use of antibiotics can result in antibiotic residues in animal products; thus, plant compounds may be better alternative sources for mitigating methane (CH4) production. An in vitro screening experiment was conducted to evaluate the potential application of 152 dry methanolic or ethanolic extracts from 137 plant species distributed in East Asian countries as anti-methanogenic additives in ruminant feed. The experimental material consisted of 200 mg total mixed ration, 20 mg plant extract, and 30 mL diluted ruminal fluid-buffer mixture in 60 mL serum bottles that were sealed with rubber stoppers and incubated at 39 °C for 24 h. Among the tested extracts, eight extracts decreased CH4 production by >20%, compared to the corresponding controls: stems of Vitex negundo var. incisa, stems of Amelanchier asiatica, fruit of Reynoutria sachalinensis, seeds of Tribulus terrestris, seeds of Pharbitis nil, leaves of Alnus japonica, stem and bark of Carpinus tschonoskii, and stems of Acer truncatum. A confirmation assay of the eight plant extracts at a dosage of 10 mg with four replications repeated on 3 different days revealed that the extracts decreased CH4 concentration in the total gas (7–15%) and total CH4 production (17–37%), compared to the control. This is the first report to identify the anti-methanogenic activities of eight potential plant extracts. All extracts decreased ammonia (NH3-N) concentrations. Negative effects on total gas and volatile fatty acid (VFA) production were also noted for all extracts that were rich in hydrolysable tannins and total saponins or fatty acids. The underlying modes of action differed among plants: extracts from P. nil, V. negundo var. incisa, A. asiatica, and R. sachalinensis resulted in a decrease in total methanogen or the protozoan population (p < 0.05) but extracts from other plants did not. Furthermore, extracts from P. nil decreased the population of total protozoa and increased the proportion of propionate among VFAs (p < 0.05). Identifying bioactive compounds in seeds of P. nil by gas chromatography-mass spectrometry analysis revealed enrichment of linoleic acid (18:2). Overall, seeds of P. nil could be a possible alternative to ionophores or oil seeds to mitigate ruminal CH4 production.


Introduction
Ruminal methane (CH 4 ) production is regarded as the cause of a loss of 3-10% of the gross energy intake of the animal and leads to the unproductive use of dietary energy [1]. Concerns regarding feed energy loss and climate change have led to many scientific studies aimed at lowering enteric CH 4 production by ruminants through various mitigation options [2,3]. Notably, feed additives (e.g., CH 4 analogues, hydroxymethylglutaryl-CoA reductase inhibitors, and nitrate and organic nitro compounds that are capable of decreasing rumen methanogenesis) have been extensively studied over the past two decades (reviewed in [4][5][6][7][8]). However, the use of certain chemically modified/synthesised compounds has adverse effects on fermentation at effective concentrations [4][5][6][7]. Intriguingly, 3-nitrooxypropanol is widely regarded as a promising candidate for enteric CH 4 mitigation [9,10]. In addition to its potential to mitigate CH 4 , consumer preference may factor into the acceptance of such a synthetic compound if commercially available. Furthermore, concerns have been raised regarding the potential use of antibiotics because of their residues in final products, which have led to bans in the Republic of Korea since 2011 [11,12]. Therefore, natural plant feed additives that might be environmentally friendly and have a high level of acceptance among consumers are desired to improve livestock productivity.
Several studies have suggested that adding plant essential oils or plant extracts rich in plant secondary metabolites (PSM; e.g., tannins, saponins, and flavonoids) to ruminant diets may have beneficial effects on ruminal fermentation and CH 4 production (reviewed in [13][14][15]). A comprehensive review by Patra et al. [4] also elaborated the direct and indirect roles of such PSMs against the growth and activity of rumen methanogens and the protozoan population. Similarly, numerous studies have shown that increasing fatty acid concentrations in the diet decreases CH 4 production to a greater extent, but often exerts detrimental effects on digestibility and fermentation of feeds, as well as animal performance [16][17][18][19]. Therefore, it would be desirable to discover plant-based fatty acidrich feed additives that decrease CH 4 production, with additional effects of improved digestibility and animal performance.
Screening natural sources at a large scale is an initial step in the discovery and development of new compounds and feed additives. Few studies have performed screening experiments; these include the European Union project "Rumen-up" that evaluated 450 plants and plant extracts [20], 58 plants, herbs, and spices in Spain [21], 156 plants from natural grasslands in France [22], and 93 plant extracts in India [23]. Nevertheless, there is a persistent need to identify potential anti-methanogenic plants for the development of new compounds as natural feed additives, because many in vivo studies have shown adaptation of the additives by rumen microbes [24,25]. Furthermore, no study has focused on large-scale screening of plant species that are widely distributed in East Asian countries for their CH 4 reduction potential. Hence, our objectives in this study were to screen 152 plant extracts from 137 plant species of East Asian origin for their potential to reduce CH 4 production, in vitro; to study the effect on volatile fatty acids (VFAs) production, to quantify the bioactive compounds of the selected candidates; and to uncover their actions on methanogens, protozoa, and several other rumen cellulolytic bacteria using real-time polymerase chain reaction (PCR) analysis.

In Vitro Rumen Fermentation Assay
Two cannulated Holstein steers (mean body weight 680 ± 30 kg), cared for in accordance with the guidelines of the Animal Ethical Committee, Seoul National University, Republic of Korea (approval number SNU-160105-1), were used as rumen fluid donors. The animals were fed twice daily with 3.5 kg rice straw containing (k −1 dry matter [DM]) 857 g organic matter; 48 g crude protein; 26 g ether extract; 768 g neutral detergent fibre; 417 g acid detergent fibre; and 2.0 kg of commercial concentrate with (k −1 DM) 896 g organic matter, 156 g crude protein, 53 g ether extract, 310 g neutral detergent fibre, and 122 g acid detergent fibre. Ruminal digesta of approximately 800 mL was collected from each steer before the morning feeding and strained through four layers of muslin into a pre-warmed flask flushed with O 2 -free CO 2 . The fluid was diluted with O 2 -free buffer (adjusted to pH 7.0) [26] at a ratio of 1:2 (v/v) and placed in a water bath pre-heated to 39 • C with continuous CO 2 flushing. Briefly, an in vitro screening assay was performed by incubating 20 mg of the extracts (dissolved in 1 mL of 10% dimethyl sulphoxide) with 30 mL of mixed rumen microorganisms in 60-mL serum bottles containing 200 mg DM of total mixed ration as the substrate. The ingredient and nutrient compositions of the substrate are provided in Table 2. The in vitro screening trial of all 152 plant extracts were tested in 2 different cycles with approximately 9-10 extracts per fermentation run with a total of 8 runs per cycle. Each run contained a control (i.e., with substrate and without plant extract), treatment (i.e., with substrate and 20 mg of plant extract), positive control (i.e., with substrate and 30 ppm of monensin; CAS No. 22373-78-0, Sigma-Aldrich, St. Louis, MO, USA), and three replicates. The bottles were sealed with rubber stoppers, covered with aluminium, and incubated at 39 • C for 24 h. After the completion of eight fermentation run (one cycle), potential candidates were chosen based on their abilities to decrease CH 4 production by more than 20%, compared to their respective controls [21]. The same experimental procedure was followed for the screening assay in the second cycle. In vitro confirmation incubations using the selected potential candidates from each cycle of the screening test were performed to validate the results. In this assay, there were four replications of the control, monensin, and each candidate at a lower dosage of 10 mg. The fermentation run was repeated on three different days to check consistency.

Measurements and Chemical Analysis
After 24 h of incubation, the total gas volume in the headspace of the bottle was measured using a water displacement apparatus [27]. A gas sample was transferred to a 10-mL vacuum tube (ref 364979, BD Vacutainer, Becton Dickinson, Franklin Lakes, NJ, USA) for CH 4 analysis. Then, the bottles were placed on ice to stop fermentation, the incubation medium was transferred to a 50-mL beaker, and the pH was measured using a pH meter (model AG 8603; Seven Easy pH, Mettler-Toledo, Schwerzenbach, Switzerland). For the microbial analysis, a 10-mL sample of incubation medium was stored at −80 • C until DNA was extracted. The remaining medium was centrifuged at 12,000× g for 10 min (Centrifuge Smart 15, Hanil Science Industrial, Seoul, Korea), and the supernatant was stored at −20 • C to determine the ammonia nitrogen (NH 3 -N) and volatile fatty acid (VFA) concentrations. CH 4 concentration in the headspace gas was determined using the Agilent 7890B GC system (Agilent Technologies, Santa Clara, CA, USA) with a flame ionization detector. The inlet and detector temperature were maintained at 200 • C and 250 • C, respectively. A 10-mL sample was injected through the back inlet using a 10-mL graduated syringe connected to a two-way stopcock (KOVAX, Seoul, Korea) with a split ratio of 10:1 into a 30 m × 0.53 mm × 20 µm HayeSep Q-ValcoPLOT fused-silica capillary column (CFS-PQ3053-200, VICI Metronics, Danvers, MA, USA). The carrier gas helium (99.99%; Air Korea) was set to a flow rate of 10 mL/min and the oven temperature of 80 • C was held constant for 2.5 min. CH 4 content was calculated by external calibration, using a certified gas mixture (8% mol/mol balance N 2 ; Air Korea). The NH 3 -N concentration was determined using a modified colorimetric method [28]. For VFA analysis, 5.0-mL aliquot of sample was mixed with 1.0 mL 25% HPO 3 and 0.2 mL 2% pivalic acid [29], then analysed using gas chromatography as described previously to identify the VFAs [30]. The feed and substrate samples were dried in a forced-air oven at 65 • C for 72 h to estimate DM content and then ground to pass through a 1-mm screen (Model 4, Thomas Scientific, Swedesboro, NJ, USA). Nutrient compositions were determined using methods described previously [30].

Analysis of Plant Secondary Metabolites
Total phenols, total tannins, and condensed tannins were determined in the selected crude extracts based on the method described by Makkar [31]. For extraction, 60 mg of crude methanol or ethanol extract was mixed with 3.5 mL of aqueous acetone (70:30 v/v), vortexed, and incubated at room temperature for 1 h. Subsequently, the mixture was centrifuged at 3000× g (Hanil Science Industrial, Gimpo, Korea) for 10 min, and the supernatant was collected and used for assays. Total phenols and total tannins were expressed as catechin (CAS No. 225937-10-0, Sigma-Aldrich) equivalents and condensed tannins were expressed as cyanidin (CAS No.528-58-5, Sigma-Aldrich) equivalents. Total tannic acids or hydrolysable tannins (HTs) were estimated as the difference between total tannins and condensed tannins [32]. Total saponin (TS) content was determined [33], and expressed as escin (CAS No. 6805-41-0, Sigma-Aldrich) equivalents. PSMs were expressed as units per milligram of extract, because the DM contents of the plant parts and extraction yield were unknown.

Gas Chromatography-Mass Spectrometry (GC-MS) Analysis
Seeds of Pharbitis nil (100 g) were ground and extracted with 1000 mL of ethanol (98%) for 24 h at room temperature in an orbital shaker. The extract was filtered through Whatman No. 2 filter paper and concentrated using a rotary vacuum evaporator (Heidolph Instruments, Schawabatch, Germany). The resulting extract (without derivatization) was diluted 10-fold, and the GC-MS analysis was performed using a TSQ 8000 triple quadrupole MS interfaced with a TRACE 1310 GC (Thermo Scientific, Waltham, MA, USA) equipped with a TG-5MS (30 × 0.25 mm× 0.25 µm; Agilent Technologies) 5%-phenyl-methylpolysiloxane fused capillary column. Pure helium gas (99.99%; Air Korea) was used as the carrier gas at a constant flow rate of 1.2 mL/min and a splitless injection volume of 1 µL. The injector temperature was maintained at 280 • C and oven temperature was programmed from 80 • C (isothermal for 2 min), with an increase of 15 • C/min to 250 • C (isothermal for 5 min), then 15 • C/min to 300 • C, ending with a 4-min isothermal incubation at 300 • C. Mass spectra were collected at 70 eV with a scan-interval of 1.0 s and fragments ranging from 50 to 550 m/z. The solvent delay was 0 to 2 min, and total run time was 25 min. Phytochemicals present in the extracts were identified based on a comparison of their mass spectral patterns with the spectral database at the library of the National Institute of Standards and Technology (NIST, Gaithersburg, MD, USA).

DNA Extraction and Real-Time PCR
Genomic DNA from the incubation medium was extracted using the NucleoSpin soil kit (Macherey-Nagel, DuÈren, Germany), and nucleic acid concentrations were measured as described previously [30]. The integrity of the gDNA was confirmed by visualising the bands using eco dye-stained (Biofact, Seoul, Korea) agarose gel electrophoresis. Real-time PCR assays to determine the relative abundances of major cellulolytic bacteria, such as Ruminococcus albus, Ruminococcus flavefaciens, Fibrobacter succinogens, total methanogens, and ciliated protozoa were performed using the SYBR Green real-time-PCR Master Mix (Bioneer, Daejeon, Korea) and the CFX96 Touch™ Real-Time PCR Detection System (Bio-Rad Laboratories Inc., Hercules, CA, USA). Thermal cycling was performed based on the annealing temperature that showed high product band intensity and determined by multiple gradient PCR for each primer set as shown in Table 3. The primers targeted the 16 s or 18 s variable region for relative quantification. Briefly, the PCR was carried out in 20-µL total reaction volumes containing 20 ng gDNA, 10 µL SYBR Green RT-PCR Master Mix, and 1.0 µL of each 10-µM primer. Thermal cycling consisted of initial denaturation at 95 • C for 10 min, followed by 40 cycles at 94 • C for 15 s and annealing for 30 s followed by extension at 72 • C for 30 s [34]. The annealing was carried out at specific temperatures corresponded for each primer sets as mentioned in Table 3. After an amplification cycle, a melting curve analysis was performed starting at 65 • C with an increase of 0.5 • C to 95 • C, followed by a plate read. The 2 −∆∆CT method was used to determine the relative fold-changes [35], and all data were normalised to the abundance of total bacteria.

Statistical Analysis
In screening assay, Student's t-test was used to compare the total gas and CH 4 production levels in the control bottles with those levels in bottles containing a given plant additive from the same incubation run. The effects were expressed as relative change to the value of the control for the specific incubation run. The confirmation assay results were analysed using one-way analysis of variance, followed by Newman-Keuls multiple comparison tests. All statistical analyses were performed using GraphPad Prism, version 5.0 (GraphPad Software Inc., La Jolla, CA, USA), and a p-value < 0.05 was considered statistically significant. To identify bacterial lineages and other parameters that differentiated the control and treatment groups, we performed principal component analysis using the fviz_pca_biplot function in the FactoMineR [39] package of R-software, version 4.0.3 (The R Foundation for Statistical Computing, Vienna, Austria). The non-parametric Kendall rank-correlation coefficient was calculated to identify correlations among CH 4 production, fermentation characteristics, bacterial communities, and PSMs using the PROC CORR function in SAS software, version 9.4 (SAS Institute, Cary, NC, USA).

Results and Discussion
While many strategies have been proposed to mitigate enteric CH 4 [2,3], most (e.g., defaunation, direct-fed microbials, ionophores, and bacteriocins) are difficult to implement at the farm level due to practical difficulties. Therefore, dietary manipulations, such as plant-based anti-methanogenic feed additives, offer highly effective CH 4 mitigation approaches [4,[13][14][15][16]19,[40][41][42]. In vitro experimental models are very useful for the preliminary screening of a large number of plant additives to select a few potent additives with desired characteristics. Plants are either directly used in the reaction mixture [20][21][22] or used as dry extracts during the screening process [23]. Therefore, we initially obtained 152 dry methanolic or ethanolic extracts of 137 plant species that are widely distributed in Korea and could be readily available as potential feed additives.
The relative effects of each plant extract on total gas and CH 4 production (mmol per g of DM) during a screening assay conducted during the two different cycles are shown in Figure 1. CH 4 production decreased by more than 10% in 20% of the extracts tested. Although the extracts from stems of Acer tegmentosum Maxim., leaves of Carpinus laxiflora (Siebold & Zucc.) Blume, leaves of Cleyera japonica Thunb., aerial parts of Erigeron annuus Pers., stems of Taxus cuspidate Siebold & Zucc., and stems of Ginkgo biloba L. exhibited a reduction of CH 4 close to 20%, they were not included as candidates for the confirmation assay. Only eight extracts (5% of the extracts tested) reduced (p < 0.1) CH 4 production by more than 20% (Supplementary Table S1) and were considered promising candidates for subsequent confirmation assays. These included stems , and stems of Acer truncatum Bunge (AC) from cycle 2. Among these, PA exhibited the maximum reduction of CH 4 by 63%, compared to the control. Most potential plant extracts decreased (p < 0.1) total gas production by 12-35%, except VI and TR, which had a negligible effect (Supplementary Table S1). These results could be attributed to the dosage (20 mg) of the plant extracts, which may have had a detrimental effect on ruminal microbes. Thus, the plant extracts were tested at a relatively lower dosage (10 mg) in subsequent confirmation assays, compared to the dosage in screening assays. , and stems of Acer truncatum Bunge (AC) from cycle 2. Among these, PA exhibited the maximum reduction of CH4 by 63%, compared to the control. Most potential plant extracts decreased (p < 0.1) total gas production by 12-35%, except VI and TR, which had a negligible effect (Supplementary Table S1). These results could be attributed to the dosage (20 mg) of the plant extracts, which may have had a detrimental effect on ruminal microbes. Thus, the plant extracts were tested at a relatively lower dosage (10 mg) in subsequent confirmation assays, compared to the dosage in screening assays.
The effects of the selected candidates on CH4, gas production, fermentation characteristics, and microbial abundance were confirmed in an in vitro assay (Tables 4 and 5). Significant decreases (p < 0.05) in CH4 production (mmol per g of DM incubated) in response to adding VI (17%), AM (17%), RE (19%), TR (22%), PA (37%), AL (27%), AC (23%), and CA (23%) were observed at half extract concentrations, compared to the screening assay. This also corresponded to reductions of CH4 concentration in total gas of 7%, 11%, 9%, 9%, 15%, 11%, 10%, and 10% (p < 0.05), respectively, compared to the control. Principal component analysis also discriminated the treatments from their respective controls, explaining 57.6% and 47.8% of variation during cycles 1 and 2, respectively (Figure 2). Furthermore, this is the first study to report the anti-methanogenic activities of these extracts, although reports regarding such activities are available for leaves of VI [43], and gross saponins from TR [44]. However, the extents of CH4 mitigation in previous studies might not be comparable with the extent in the current study because of the different plant parts and dosages used. In addition, despite the lower dose of supplemented extracts compared to the screening assay, an increase (p < 0.05) in pH and decreases (p < 0.05) in total gas production, total VFA, and NH3-N were detected in the confirmation The effects of the selected candidates on CH 4 , gas production, fermentation characteristics, and microbial abundance were confirmed in an in vitro assay (Tables 4 and 5). Significant decreases (p < 0.05) in CH 4 production (mmol per g of DM incubated) in response to adding VI (17%), AM (17%), RE (19%), TR (22%), PA (37%), AL (27%), AC (23%), and CA (23%) were observed at half extract concentrations, compared to the screening assay. This also corresponded to reductions of CH 4 concentration in total gas of 7%, 11%, 9%, 9%, 15%, 11%, 10%, and 10% (p < 0.05), respectively, compared to the control. Principal component analysis also discriminated the treatments from their respective controls, explaining 57.6% and 47.8% of variation during cycles 1 and 2, respectively ( Figure 2). Furthermore, this is the first study to report the anti-methanogenic activities of these extracts, although reports regarding such activities are available for leaves of VI [43], and gross saponins from TR [44]. However, the extents of CH 4 mitigation in previous studies might not be comparable with the extent in the current study because of the different plant parts and dosages used. In addition, despite the lower dose of supplemented extracts compared to the screening assay, an increase (p < 0.05) in pH and decreases (p < 0.05) in total gas production, total VFA, and NH 3 -N were detected in the confirmation assay. A higher pH and reduced VFA concentrations are indications of overall inhibition of rumen microbial fermentation, which would not be nutritionally beneficial to the host animal, since VFAs are major energy source for the ruminants [45]. However, this effect is comparable with the effect of monensin, suggesting that the extracts have similar properties to those of monensin. This could be attributed to the greater concentrations of PSMs in the tested extracts, which are known for their anti-microbial activities [46]. Most of the plant extracts tested in this study (except seeds of TR and PA) were rich in total phenols, total tannins, HTs, and TSs (Table 6). This is consistent with previous studies reporting greater concentrations of polyphenols, flavonoids, and saponins in tested plant species with anti-microbial properties [47][48][49][50][51][52]. It has also been reported that HTs reduce the production of total VFAs through actions on ruminal microbes [53,54]. This is further supported by the significant decrease (p < 0.001) in the F. succinogens population in this experiment (Table 5), which is an efficient producer of succinate and the major precursor for propionate synthesis [55].  Similarly, the decrease in NH3-N might be related to proteolysis inhibition through the formation of insoluble tannin-protein complexes [56,57]. Getachew et al. [58] reported  Similarly, the decrease in NH 3 -N might be related to proteolysis inhibition through the formation of insoluble tannin-protein complexes [56,57]. Getachew et al. [58] reported a decrease in protein degradation and NH 3 -N after supplementation with tannic acids. This finding suggests that the addition of a tannin-rich extract might minimise the degradability of protein in the rumen and exert beneficial effects similar to those that occur when ruminants are supplemented with rumen undegradable protein (reviewed in [59]). Hydrolysable tannins with low molecular weight and less structural variability than condensed tannins result in more consistent reduction of CH 4 due to gallic acid subunit binding to methanogens [60]. In the current study, the HT concentration provided by the extracts (1.15-1.35 g/100 g DM) was comparable with the level (1.43 g/100 g DM) supplemented in the study by Aboagye et al. [60], who observed a 9% decrease in CH 4 yield. In addition, Jayanegara et al. [61] showed that HTs decrease the methanogen population and microbes, which provide H 2 to a greater extent, compared to condensed tannins. Pure saponins and saponin-containing plants or extracts have inhibitory effects on protozoans (reviewed in [62]), which contribute to CH 4 production via interspecific H 2 transfer to methanogens [63]. In the current study, the abundances of total methanogens in VI, AM, and RE decreased (p < 0.001), as did ciliated protozoa in VI and RE (p < 0.001), compared to the control (Table 4). These findings clearly showed the effects of HTs and TSs on H 2 and CH 4 production, which thereby affect total gas production. These findings were supported by stronger negative (τ = −0.51, p = 0.070) and positive (τ = 0.64, p < 0.05) correlations between TS content and CH 4 production, and protozoan abundance and gas production, respectively (Table 7). However, AL, AC, and CA reduced CH 4 without any negative effects on methanogens or the protozoan population, compared to the control (Table 5). Expression analysis of methyl-co reductase (MCR) gene can provide a better understanding of complex methanogenesis processes than methanogen abundance analyses based on 16s rDNA [64]. Other studies have also demonstrated that CH 4 production is not correlated with methanogens abundance, but with its composition (reviewed in [65]). Furthermore, saponins may decrease the activities of CH 4 producing genes or the rate of CH 4 production in methanogenic cells [66], suggesting that PSMs from different sources have different effects on microbes and methanogenesis [67]. However, directly or indirectly inhibiting CH 4 production entails a change in the VFA profile, mostly favouring greater propionate production [68]. Gram-positive ruminal bacteria generally produce acetate and butyrate, while Gram-negative bacteria produce propionate [69]. The decrease in CH 4 production caused by most of the tested extracts in this experiment, without any changes in the proportions of individual VFAs (except PA), suggests broad spectral antibacterial activities of PSMs targeting Gram-positive and negative bacteria. However, no negative effects were observed on selected microbes, such as R. flavefaciens and R. albus, in this experiment. Some studies have reported that PSMs target other ruminal microbes with minimal effects on Ruminococcus spp. (reviewed in [46]). Despite the ban on the non-therapeutic use of monensin in the Republic of Korea, it remains one of the most commonly used ionophores in ruminants in other countries. Monensin supplementation has been associated with decreased methanogenesis accompanied by improved feed digestibility, increased propionate synthesis, and decreased NH 3 -N production [70]. A recent study [71] also showed a decrease in CH 4 production coupled with a decrease in H 2 -producing microorganisms (e.g., protozoa, fungi, and Gram-positive Firmicutes) after supplementation with monensin. Intriguingly, in the current study, the decrease in CH 4 production caused by PA alone was accompanied by decreases in protozoan abundance and NH 3 -N concentration, as well as an increase in the proportion of propionate, similar to the effect of monensin (Table 5).
Principal component analysis grouped PA and monensin, explaining 47.8% of the variation from their respective controls (Figure 2b). The PCA analysis also exhibited a strong correlation of propionate towards PA and monensin, further supporting our statement. The observed effect of PA with a very low concentration of TSs and near absence of HTs suggests the presence of other potentially bioactive compounds in PA. GC-MS analysis revealed the presence of a heterogeneous mixture, dominated by polyunsaturated fatty acids ( Table 8). Seeds of P. nil had greater concentrations of 9,12-octadecadienoic acid (Z,Z)-(23%), commonly known as linoleic acid (18:2), followed by 9,12-octadecadienoic acid (Z,Z)-,2,3-dihydroxypropyl ester (18%) commonly known as alpha-glyceryl linoleate. Overall, 60% of the compounds identified were classified either as fatty acids or fatty acid amides. A meta-analysis by Patra et al. [19] established negative associations between total dietary polyunsaturated fatty acid concentrations and CH 4 , VFAs, and NH 3 -N production in the rumen. The effects of polyunsaturated fatty acids on CH 4 production were attributed to the change in H 2 thermodynamics in the rumen, caused by inhibition of protozoa, biohydrogenation of unsaturated fatty acids, and increased production of propionic acid, which compete with methanogenesis for metabolic H 2 [72,73]. A strong negative association (τ = −0.51, p = 0.070) was noted between protozoan abundance and propionate proportion in the current study. A meta-analysis by Guyader [74] reported a decrease in protozoan abundances in experiments supplemented lipids on ruminants' diet, which was due to changes in membrane permeability, resulting in cell lysis [75]. In addition, Dohme et al. [76] reported a detrimental effect of linoleic acid (18:2) on the protozoan and total bacterial populations. This is consistent with the decreased (p < 0.001) abundance of the ciliated protozoa, R. flavefaciens and F. succinogens in PA, in the current study. However, complete metabolite profile of PA using chromatographic techniques with proper derivatization procedures would give deeper understanding of the compound responsible for the action. Moreover, enrichment of dietary linoleic acid (18:2), a precursor of bioactive conjugated linoleic acids [77], suggested that PA seeds might be a promising feed additive for ruminants. In addition, PA seeds have been widely used in Korean and Chinese traditional medicine for their roles in improving digestibility and intestinal motility (reviewed in [78]). Therefore, PA seeds could act as a source of fatty acids, probably replacing oil seeds that have been reported to decrease DM and neutral detergent fibre digestibility [19]. However, future in vitro or in vivo trials are needed to confirm their effects on rumen nutrient digestibility and animal performance, since the protozoal defaunation was associated with decrease in rumen organic matter digestibility and specifically NDF and ADF digestibility [79].

Conclusions
The extracts rich in phenolic compounds from stems of A. asiatica, fruit of R. sachalinensis, seeds of T. terrestris, leaves of A. japonica, stems and bark of C. tschonoskii, and stems of A. truncatum reduced CH 4 production and fermentation rates in vitro. The negative effects on total gas and VFA production suggest the need to standardise the doses of plant extracts that are effective for inhibiting CH 4 emissions with minimum adverse effects on fermentation. These supplemental plant extracts seem to decrease the output of ammonia from protein degradation, although the post ruminal nitrogen use efficiency is still remained to be elucidated in ruminants. Notably, the maximum reduction in CH 4 production by the extracts from the seeds of P. nil, which are rich in linoleic acid (18:2) and other fatty acid amides, is a promising alternative to ionophores and oilseeds to mitigate CH 4 emissions. In vivo trials must be conducted to elucidate the adaptation of rumen microbes to the seeds of P. nil over a prolonged feeding period.