In vitro screening of selected essential oils from medicinal plants acclimated to Benin for their effects on methane production from rumen microbial fermentation

Enteric methane production lowers the efficiency of feed utilization in ruminants and contributes to greenhouse gas emissions which are responsible for global climate change. This study examined the effects of nine essential oils (EO) from Citrus aurantifolia, Cymbopogon citratus, Eucalyptus citriodora, Laurus nobilis, Lippia multiflora, Mentha piperita, Ocimum basilicum, Ocimum gratissimum and Zingiber officinalis on enteric methane production in in vitro batch cultures screening experiments using Andropogon gayanus grass. Two in vitro batch culture incubation runs were conducted independently on separate days at two different ranges of dosages: 0 (control), 150, 300, 600 and 1200 mg/L inoculum and 0 (control), 25, 50, 100 and 150 mg/L inoculum. The effects of EO on in vitro gas production, methane production and apparent dry matter disappearance (DMD) were assessed relative to the control containing no additive. O. basilicum, E. citriodora, O. gratissimum and C. aurantifolia, significantly inhibited (Z’ ˃ 0 and relative decrease ≥ 15%) enteric methane production (g DM incubated) relative to control at dosages of 300-1200 mg/L and L. nobilis, C. citratus and M. piperita significantly decreased it at 600 and 1200 mg/L. A substantial decrease (Z’ ˃ 0 and relative decrease ≥ 15%) in methane production per g DM incubated was apparent for Z. officinalis and L. multiflora at dosage of 1200 mg/L. Most EO had globally negligible effects on methane production (Z’ ≤ 0 and relative decrease < 15%) at dosages of 25 to 150 mg/L. Substantial decrease in apparent DMD together with gas production (g DM) incubated was observed relatively to the control with Z. officinalis and L. multiflora at 1200 mg/L and with the remaining EO at 600 and 1200 mg/L. Overall, this screening investigation demonstrated that addition of assayed EO (except Z. officinalis and L. multiflora) at dosages close to 300 mg/L seem to potentially decrease enteric methane production with limited negative effects on dry matter digestibility of forage grass in vitro.


INTRODUCTION
Methane is known as a potent greenhouse gas and its accumulation in the atmosphere is thought to be a key factor in global anthropogenic warming besides carbon dioxide and nitrous oxide (Intergovernmental Panel on Climate Change, 2013).One of the significant contributions to the increase of methane in the atmospheric  (Lassey, 2007).Therefore, the interest in modulating the rumen microbial fermentation occurring in ruminant has been increasing with reduction in enteric methane production being the ultimate target.To this end, a number of strategies were recently explored using products with antimicrobial properties including plant secondary metabolites of enteric fermentation which is mediated by microorganism activity in the rumen (Ribeiro et al., 2015;Satyanagalakshmi et al., 2015).Most plant extracts are considered as safe in animal production owing to their natural source unlike chemical additives and antibiotics.
As a result, it has been demonstrated that essential oils (EO), among other researched natural products, can favorably alter rumen microbial fermentation and reduce enteric methane production owing to their antimicrobial activity (Bodas et al., 2012).
Antimicrobial properties of EO depend on their chemical composition, which is a function of plant species, harvesting seasons, geographical origin, analytical methodology and the plant organ used (Burt, 2004).Many EO have been shown to inhibit methane production in in vitro incubations at high doses together with a decrease in total volatile fatty acid concentrations and feed digestion (Benchaar and Greathead, 2011).There is a challenge in identifying EO that could potentially benefit ruminal fermentation and lower methane production.Therefore, this study aimed to screen the effects of EO from medicinal plants acclimated to Benin on methane production from rumen microbial fermentation of Andropogon gayanus grass using in vitro batch cultures.The plants were selected because their EO are edible and commonly used in medicinal pharmacopoeia in Benin.

In vitro experimental design and treatment
The EO were evaluated for their effects on methane production, gas production and apparent dry matter disappearance (DMD) relative to the controls (inoculums plus substrate without EO) in in vitro batch cultures using A. gayanus grass.Essential oils were tested in an initial screening in a single run as suggested by Secundino et al. (2010) at 4 dosages of 150, 300, 600 and 1200 mg/L inoculum (run 1) as used in a previous study (Benchaar and Greathead, 2011), which observed that most EO inhibited methane production at dosages above 300 mg/L.Based on the results from this experiment where apparent DMD and gas production decreased with most EO at dosages between 300-1200 mg/L, all EO were evaluated in a second incubation run at lower dosages of 25, 50, 100 and 150 mg/L (run 2) to limit their adverse effect on apparent DMD.
Incubation sets containing only inoculum served as blanks and were used to correct fermentation residues, gas and methane production resulting from the inoculum.In the preliminary and second screening assays, a single incubation run was carried out as suggested by Secundino et al. (2010) and each treatment as well as control and blank samples were tested in triplicate.

Essential oils
Six EO were prepared by processing various parts (leaves, rhizomes or fruit peel) of six medicinal plants (Table 1) as previously described by Baba-Moussa et al. (2012).The plants were collected by the departments of Ouémé (Porto Novo, Sèmè, Djérègbé), Plateau (Kétou) and Zou (Setto) in Benin and were botanically identified by the National Herbarium of Benin where their voucher specimens were deposited.For each EO, appropriate plant fractions to obtain the desired volume were collected from many plants with combination of all collected samples.The EO were extracted from the collected plant materials after 72 h of airdrying by steam-distillation for 180-240 min using a Clevenger-type apparatus (Clevenger, 1928) and EO composition was analyzed by gas chromatography coupled to mass spectrometry (GC-MS).In steam distillation, vapours of the volatile components were carried from a plant material by steam, which was produced from water contained in a heated round-botton flask.The resultant mixture of steam flow and oil vapour was condensed in a refrigeration tube, and then collected in a Florentine flask where water and oil were separated.A gas chromatograph (DELSI GC 121 G, Type 300 N°464, DELSI instruments: 92, Surenes, France) equipped with a flame ionization detector and a capillary column (CP WAX 52 CB made by Chrompack, Middelburg, Netherland; length 25 m × 0.25 mm interior diameter, 0.25 µm film thickness) was used.GC-MS was performed (model 5970, Hewlett-Packard, Palo Alto, CA, USA) *Corresponding author.E-mail: gjdjim@yahoo.fr.Tel: (229) 97 53 35 51.
Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution License 4.0 International License using a DB-5 non-polar capillary column (length 25 m × 0.25 mm interior diameter) with ionization energy of 70 eV.The column temperature was kept at 50°C for 5 min and programmed to subsequently increase at a rate of 5°C/min to 300°C.Essential oils constituents were identified on the basis of their Kovats retention indices and mass spectral fragmentation, using standards, literature data (Adams, 2001) and an established laboratory data bank (Laboratory of Pharmacognosy and Essential Oils at the University of Abomey-Calavi, Benin).

Fermentation substrate
The aerial part of A. gayanus grass was collected by cutting plants 10 cm above the soil surface at the flowering stage during July and August 2012 at the pilot farm of the Faculty of Agronomic Science, UAC, Benin (longitudes 1° and 30°40' East).The grass was sundried over a period of 3 days followed by oven drying at 60°C for 48 h.Grass was then ground through a 1 mm screen with combination of all the collected samples before use as the fermentation substrate.

Ruminal inoculum and in vitro batch incubations
Two ruminally fistulated non-lactating cows fed barley grain and barley silage in a DM ratio of 1:3 were used as donors of rumen fluid for the entire study.Cows were cared for in accordance with standards set by the Canadian Council on Animal Care (CCAC, 1993).Rumen fluid was collected 2 h after the morning feeding by straining rumen contents from four sites within the rumen through four layers of sterilized cheesecloth into pre-warmed insulated containers.Collected rumen fluid was transported immediately to the laboratory under anaerobic conditions, pooled in equal portions and mixed with a mineral buffer (Menke et al., 1979) in 1:2 ratio to prepare the rumen microbial inoculum.
The in vitro batch incubations were conducted as described by Wang et al. (2000).Incubations were performed in 125-mL serum vials pre-loaded with 500 mg of substrate and warmed to 39°C.Inoculum (40 mL) was dispensed under a stream of O2-free CO2 and an appropriate amount of each EO was added into the inoculum by using a pipette to obtain the final desired concentration.Vials were immediately sealed and affixed to a rotary shaking incubator (120 revolutions per minute) at 39°C for 48 h.

Sampling and analysis
Gas production from each of the three culture vials used for each treatment as well as control and blank was measured at 6, 12, 24 and 48 h of incubation using a water displacement technique (Fedorak and Hrudey, 1983).Prior to measuring of gas production at each time point, 10 mL of headspace gas was sampled for methane analysis (Chaves et al., 2006;Wang et al., 2008).
After 48 h of incubation, each vial was removed from the incubator and its content was transferred into a pre-weighed 50-mL centrifuge tube and centrifuged at 500 ×g (4°C, 10 min) to obtain a solid fraction (fermentation residue) containing undegraded substrate and residual feed particle-associated microbial biomass (Narvaez et al., 2013).The fermentation residue from each vial was washed with dH2O and centrifuged (500 ×g, 4°C, 10 min) two times, then dried at 50°C and weighed to determine apparent DMD (Narvaez et al., 2013).
The substrate was analyzed for dry matter (DM) and organic matter (OM) content as described by Association of Official Analytical Chemists (AOAC, 2003), neutral detergent fiber (NDF) and acid detergent fiber (ADF) as described by Van Soest (1991).Heat-stable α-amylase and sodium sulphite were used in NDF procedure and expressed inclusive of residual ash.Combustion analysis (NA2100, Carlo Erba Instruments, Rodano, Milan, Italy) was used to determine N.

Calculations and statistical analysis
At each incubation time point, cumulative methane produced was calculated (López et al., 2007) and total net gas production per g of DM incubated (considering the 10 mL sampled for methane analysis) or methane production per g of DM incubated was estimated by subtracting the mean values of blanks from that of the control and test vials.Net cumulative methane and gas production per g of DM incubated were estimated after 48 h of incubation.The apparent DMD was calculated as the difference between incubated weight of the substrate and the dry weight of the fermentation residue corrected for residue weight in the blank (Narvaez et al., 2013).
Because the incubation run was not repeated in each screening trial, the statistical evaluation of results from this study was limited.Therefore, the statistical Z' (Z-prime) factor was calculated as 1-3 (σtest + σcontrol)/|μtest -μcontrol|, where σ indicates standard deviation and μ indicates mean, and it was used to assess the separation between test (EO treatment) and control values regarding net cumulative methane production, net cumulative gas production and apparent DMD after 48 h of incubation (Secundino et al., 2010).Test and control values were declared overlapping at Z' ≤ 0, whereas both values are different when Z' ˃ 0 (Secundino et al., 2010).As the acceptance criterion required for Z' to be confirmed as a sample with a substantial decrease in methane production was not established as observed by Secundino et al. (2010), the effect of each treatment on methane production relative to the control was expressed as the percentage of induced change (increase or decrease) in methane production as compared to the control using the equation R = (μtest − μcontrol) × 100/μcontrol where R (%) = relative effect of EO on methane production, μtest = net cumulative methane production after 48 h of incubation with a given treatment; μcontrol = net cumulative methane production after 48 h of incubation from the control (Secundino et al., 2010).Relative decrease (R) in methane production (per g of DM incubated) with addition of EO was considered as significant when Z' ˃ 0 and R value was 15% or higher as Secundino et al. (2010) observed from previous laboratory studies which shows that a plant causing a relative decrease of 15% or higher can be considered a promising plant additive deserving further investigation.

Chemical composition of fermentation substrate and essential oils
The substrate used in this study had nutrient composition (g /kg in DM) of OM 920, NDF 679, ADF 347 and CP 105.
The main components in EO analyzed by GC-MS in the

Incubation run 1
Off all the EO, O. basilicum, E. citriodora, O. gratissimum and C. aurantifolia significantly inhibited (Z' ˃ 0 and relative decrease ≥ 15 %) methane production per g dry matter (DM) incubated at dosages of 300-1200 mg/L, whereas apparent DMD and gas production per g DM incubated were decreased (Z' ˃ 0) mainly at 600 and 1200 mg/L relative to the control (Table 3).A substantial decrease (Z' ˃ 0 and relative decrease ≥ 15%) in methane production per g DM incubated was apparent together with a reduction (Z' ˃ 0) in apparent DMD and gas production per g DM incubated for L. nobilis, C. citratus and M. piperita mainly at dosages of 600-1200 mg/L.Z. officinalis and L. multiflora significantly inhibited (Z' ˃ 0 and relative decrease ≥ 15 %) methane production per g DM incubated at dosage of 1200 mg/L together with a reduction (Z' ˃ 0) in apparent DMD and gas production per g DM incubated at 1200 mg/L.

Incubation run 2
Most EO had globally negligible effects on methane production (Z' ≤ 0 and relative decrease < 15 %) at dosages of 25 to 150 mg/L (Table 4).At such dosages, EO treatments and control were overlapping (Z' ≤ 0) regarding apparent DMD and gas production per g DM incubated.

Chemical composition of essential oils
Results from the present study reveal the specificity in qualitative and quantitative composition of the EO from each plant species.Essential oil constituents of a given plant species may vary depending on harvesting seasons, geographical origin, analytical methodology and the part of the plant that they are extracted from (Burt, 2004).This may explain the divergence between the chemical composition of EO analyzed in the present study and that previously reported for their chemotypes (Benchaar et al., 2008;Cimanga et al., 2002;Marzouki et al., 2009).For example, a previous report (Marzouki et al., 2009) from Tunisia observed 1,8-cineole, α-terpinyl acetate and methyl eugenol as main components in the L. nobilis chemotype.Benchaar et al. (2008) reported camphene (14.1%), ar-curcumene (14.5%) and βbisadolene (22.1%) as the major compounds in Z. officinalis.Similarly to this study, O. gratissimum from Democratic Republic of Congo was a thymol type, but it contained more thymol (53.2%) and less p-cymene (7.3%) (Cimanga et al., 2002).Cimanga et al. (2002) also identified eugenol and γ-terpinene as main components in O. gratissimum chemotype, whereas γ-terpinolene was absent.Iscan et al. (2002) identified menthol and menthone as major components in four chemotypes of M. piperita, but their concentrations were less than those found for M. piperita in the present study.A concentration of 53.53% was reported by Javari et al. (2011) for limonene in C. aurantifolia from Iran, which agrees with the results of the present study.Alpha terpineol and γterpinene are among other major components thought to be responsible for the antimicrobial properties of C. aurantifolia, but their concentrations were higher in the previous study than in the present study.The chemotype of E. citriodora from Democratic Republic of Congo was found to contain 72.7% citronellal (Cimanga et al., 2002),  The decrease in apparent DMD together with gas production observed in this study suggests that EO may exert a general inhibition on rumen microbial fermentation.The extent of the observed effects of EO on apparent DMD and gas production depended on EO type and EO dosages.This is likely a reflection of the differences in EO composition within plant types (Benchaar et al., 2008;Burt, 2004).The reduction in apparent DMD with Z. officinalis and L. multiflora at 1200 mg/L and with the remaining EO at 600-1200 mg/L suggests that beneficial inhibition in methane production is likely offset by adverse effects on rumen microbial fermentation.Effects of L. nobilis, O. gratissimum, O. basilicum, M. piperita and C. aurantifolia on apparent DMD and gas production as observed in this work are consistent with previous cited studies (Kamalak et al., 2011;Jahani-Azizabadi et al., 2011;Patra and Yu, 2012) that worked on a range of dietary substrates.To the best knowledge of the authors, there is no study that examined the effects of Z. officinalis (α-zingiberene and ar-curcumene type), L. multiflora (pcymene type), C. citratus (neral and geranial type) and E. citriodora (citronellal type) on apparent DMD and gas production.
In conclusion, reductions in enteric methane production with limited negative effects on apparent DMD seem to potentially occur in vitro with all EO (except Z. officinalis and L. multiflora) at dosages close to or less than 300 mg/L.

Table 1 .
List of medicinal plants (scientific name, common name and registration number) and plant parts from which the essential oils were extracted.

Table 3 .
Effects of essential oils (150-1200 mg/L) on gas production, methane production and apparent dry matter disappearance after 48 of in vitro incubation.

Table 4 .
Effects of essential oils (25-150 mg/L) on gas production, methane production and apparent dry matter disappearance after 48 h of in vitro incubation.Citrus sinensis (limonene type), at 200-1200 mg/L, reduced methane production (per g DM).Results obtained with addition of O. basilicum are in agreement with Jahani-Azizabadi et al. (2011) which showed that O. basilicum (estragole type) reduced methane production (per g DM) at 1000 mg/L.Information relating to effects of Z. officinalis (α-zingiberene and ar-curcumene type), L. multiflora (p-cymene type), C. citratus (neral and geranial type) and E. citriodora (citronellal type) on methane production is scarce.