Pithecellobium dulce , Tagetes erecta and Cosmos bipinnatus on reducing enteric methane emission by dairy cows

The aim of the present research was to evaluate the effect of Pithecellobium dulce, Tagetes erecta and Cosmos bipinnatus on methane emission, milk yield and dry matter intake in dairy cattle. A 4×4 Latin square experimental design was employed, using four multiparous Holstein cows of 553±72.4kg body weight, at mid lactation and average milk yield of 17.3±3kg/day. The experiment lasted 92 days, divided into four experimental periods of 23 days each. All cows had free access to maize and alfalfa silage in a 50:50 proportion, 4kg of concentrate/day and ad libitum access to water. Treatments consisted in supplementation of 0.5kg/day of the experimental plants, with one control treatment without supplementation. Each cow received one of each treatment in turn during one of the four periods. The C. bipinnatus reduced methane production by 16% (P<0,05) in comparison with the control diet. Milk production, milk composition and dry matter intake were not affected (p>0 0.05) by the use of C. bipinnatus or any other plant species. Supplementation at low doses of C. bipinnatus showed a reduction in ruminal methane production in dairy cows.


INTRODUCTION
The main sources of anthropogenic methane (CH 4 ) are the production of ruminants for meat and milk production, the extraction and use of fossil fuels, and rice production.Agriculture contributes to approximately 13% of the total global emission of greenhouse gases (GHG) (IPCC, 2014), of which cattle are responsible for 53% (CHARMLEY et al., 2016).Thus, several studies, mostly in vitro experiments, have been conducted on the effect of tanniferous plants on reducing rumen CH 4 production, in order to find natural alternatives to mitigate the environmental impact generated by the emissions of this GHG by the cattle industry (CUARTAS et al., 2014).However, results in literature are contradictory, because in some cases a small methane reduction has been observed whereas in others a large reduction potential was identified.For example, NAUMANN et al. (2013) observed a small CH 4 reduction of 3.2% and 0.5% in an experiment conducted to evaluate the in vitro effect of two types of Acacia angustissima -the South Texas ecotype and Cross timbers ecotype -with a condensed tannin content of 8.4% and 8.9%, respectively.Similarly, in a study conducted by BHATTA et al. (2009) in order RESUMO: O objetivo deste estudo é avaliar o efeito do Pithecellobium dulce, Tagetes erecta e o Cosmos bipinnatus na emissão do metano, a produção de leite e de ingestão diária de matéria seca pelo gado.Um desenho experimental do quadrado latino 4x4 foi usado, com quatro vacas da raça Holandesa, multíparas, de 553±72,4kg, no seu segundo terço da lactação e rendimento médio de leitedo de 17,3±3kg/dia.O experimento durou 92 dias e foi dividido em quatro períodos de 23 dias cada um.Das vacas tinham acesso livre a milho e silagem de alfafa, numa proporção de 50:50, com 4kg de concentrado por dia, com acesso a água ad libitum.Os tratamentos consistiram na ingestão de 0,5kg por dia, com as plantas experimentais; as vacas no tratamento de controle não receberam nenhuma planta, porém, elas receberamum tratamento, em cada um dos quatro períodos.O C. bipinnatus reduziu significativamente a produção de metano em 16% (P<0,05), em comparação com a dieta controle.A produção de leite, sua composição e o consumo de matéria seca não foram afetados (P>0,05) pelo uso de C.bipinnatus ou outras espécies de plantas.A ingestão com doses baixas de C. bipinnatus, que é uma planta arbustiva com um centéudo moderado de taninos, mostrou ter potencial para reduzir a produção de metano ruminal em vacas leiteiras.Palavras-chave: metano, gado, mudança climática, mitigação, taninos.

Pineda et al.
to evaluate the effect of different tannins (in their pure form) on in vitro CH 4 production, it was observed that hydrolysable tannins reduced CH 4 production by only 0.6%, while a mixture of hydrolysable and condensed tannins reduced it by 5.5%.Unfortunately, there are few studies where the effects of entire tanniferous plants on rumen methano genesis in cattle and animal performance have been evaluated in vivo (PATRA & SAXENA 2010).For example, BEAUCHEMIN et al. (2007) conducted an experiment to determine if 1% and 2% of dietary DM as condensed quebracho tannin extract could be used to reduce enteric CH 4 emissions in cattle.They used six spayed Angus heifers and 6 Angus steers and concluded that feeding with up to 2% of dietary DM as quebracho tannin extract failed to reduce enteric methane emissions from growing cattle.No information was reported in the literature on the in vivo anti-methanogenic properties of Tagetes erecta, Cosmos bipinnatus and Pithecellobium dulce.However, our group conducted several in vitro studies to evaluate the potential of Tagetes erecta for reducing CH 4 production; for example, ANDRADE et al. (2012) observed a reduction of 39% in comparison with a control diet.Similarly, MAHMOUD et al. (2016) reported a 35% reduction (p<0.05) in methane production in comparison with a control diet when only 10% dietary dry matter (DM) of T. erecta was included in the diet.Likewise, a reduction of 11% (P<0,001) in methane production was also reported by MAHMOUD et al. (2017) with the inclusion of only 10% of dietary DM of C. bipinnatus in the experimental diet.These studies also suggested that the anti-methanogenic capacity of T. erecta and C. bipinnatus is not linear but quadratic, and that the best effect was observed at low inclusion levels.Therefore, it was hypothesized that T. erecta, C. bipinnatus and P. dulce could reduce in vivo methane production and that this effect could be detected at low inclusion levels without affecting animal productive performance.The objective of the present study was to evaluate the effect of supplementation of Tagetes erecta, Cosmos bipinnatus and Pithecellobium dulce on reducing enteric methane emission, and on milk production and dry matter intake in dairy cattle.

Area of study
The experiment was conducted at the Laboratory for Research on Livestock, Environment and Renewable Energies of the Faculty of Veterinary Medicine and Animal Science of the Universidad Autónoma del Estado de México, located in Toluca, State of Mexico, Mexico at 19°27'N and 98° 38'W and 2600m.a.s.l.

Experimental design:
A 4×4 Latin square experimental design was employed, using four multiparous Holstein cows of 553±72.4kg of body weight, at mid lactation and average milk yield of 18±3kg/day.The experiment lasted 92 days, which were divided into four experimental periods of 23 days each.All cows had free access to maize-alfalfa silage in a 50:50 proportion, 4kg of concentrate/day and ad libitum access to water.The concentrate was composed of 48.7% corn, 20% soya bean, 14.8% canola bean, 14.7% wheat bran and 1.8% of mineral premix.The four treatments consisted of the supplementation of 0.5kg DM/day/cow for each of the three experimental plants plus the control, which was the diet alone without the addition of experimental plants.Each cow received each treatment in turn, once in each of the four periods.Before the beginning of the trial, the experimental plants were dried over a period of 8 weeks away from sunlight to prevent denaturalization of phenolic compounds as indicated in MAKKAR et al. (1993).Once dried they were ground and incorporated into the concentrate to ensure that the cows would eat the entire daily ration provided, as the aim of the experiment was to evaluate the effect of the entire plants and not only the effect of the phenolic compounds within them.Samples of the experimental plants were collected for later chemical analysis in the laboratory.For each 23day experimental period, 15 days were used for diet adaptation and the remaining to measure methane production, dry matter intake and digestibility of the diet.The cows were confined in individual pens of 4×4m, equipped with drinker and trough, throughout the experiment.Milk yield was weighed daily during the entire 23-day period and milk composition was evaluated daily during the measuring period of eight days.A Lactichek ™ -01 (Rapi Read, Page & Pedersen International Ltd.Hopkinton, Massachusetts) analyzer was used to determine milk composition: fat, lactose, protein, non-fat milk solids.Body weight was measured once at the beginning and once at the end of each experimental period.We used adult cows and the body weight was recorded only.Dry matter intake (DMI) was determined by weighing the silage and concentrate offered in the morning and collecting and weighing the rejected feed the next morning.The difference between the offered feed minus the rejected feed was the daily DMI.Digestibility of the dry matter (DMD) was calculated as shown in equation (1).Samples of silage were taken every week for later chemical analyses in the laboratory.

Methane production measurement:
The CH 4 production was measured with a respiration chamber of the head-box type as described previously in PEDRAZA- BELTRÁN et al. (2016).The CH 4 analyzer that was used measures emissions highly accurately over a wide range by combining noise amplification and signal processing with a computerized controller and digital filtering to provide a maximum resolution of 0.0001% to 0.01%.Before each assay two calibrations of this instrument were performed: a zero calibration using high-purity nitrogen (N 2 ) (Praxair Inc., Mexico), and a calibration against a reference gas, which is also known as a span gas.The N 2 in the zero calibration was first passed through a drying unit to remove moisture and then through the analyzer at a flow rate of 0.3L/min to obtain a reading close to zero.The span calibration was performed using a known CH 4 concentration gas mixture (1000ppm of CH 4 in high-purity N 2 ).The span gas passed through the analyzer (0.3L/min) to obtain a reading corresponding to the concentration of CH 4 in the span gas.Every assay started at 10:00h, the mass flow generator was set at 480L/min, the analyzer was set to measure CH 4 concentration every second and then the chamber was closed.The CH 4 emissions were measured for 24-h period.The cows were removed from the chamber for milking at 6:00h and 15:00h, and each milking lasted 1.5h, subsequently they were returned to the chamber.The diet was weighed before the beginning of the assay, and all animals were given the same amount at the same time (9:00h and 16:00h).The next morning, the orts were removed and weighed to calculate DMI.Diet samples were collected and kept in a freezer until laboratory analysis.Feces were collected and weighed at the end of each measurement.A sample of approximately 1kg of feces was obtained and kept frozen until laboratory analysis.Four assays were completed in each experimental period.

Chemical analysis of the feed
Silage, concentrate feed and stool samples were dried in a forced air oven at 60°C for 72h until constant weight was obtained, then later ground and passed through a 1-mm sieve.The DM and organic matter (OM) contents were determined according to the procedures of the Official Methods of Analysis (AOAC, 1997).The nitrogen content in the silage and concentrate was determined by the Kjeldhal method (AOAC, 1997), and subsequently multiplied by a factor of 6.25 to obtain the protein content.The neutral detergent fiber (NDF), acid detergent fiber (ADF) and lignin contents were determined by the method of VAN SOEST et al. (1991) using an ANKOM A200 fiber analyzer (Technology Corporation, Fairport, NY, USA).The DM content in the silage was corrected as in HAIGH (1995) to include the volatile solids in the DM.The concentration of total phenols in the experimental plants was determined by the Folin-Ciocalteu method and the tannin content by the polyvinylpolypyrrolidone method as described in MAKKAR (1993).Table 1 shows the chemical composition of silage and concentrates, and the phenol and tannin content of the experimental plants.

Analysis of results
The experimental variables were analyzed with an analysis of variance for a Latin square experimental design as shown in equation 2. In order to eliminate Type II error (KAPS & LAMBERSON 2009) the Fisher's Least Significant Difference (LSD) test between control and treatment means was also carried out.The analytical procedures were carried out using the lmer function of the lme4 package (BATES et al., 2015) in R software (R core team, 2016).Post hoc pairwise comparison was carried out using the Tukey HSD test using the lsmeans function in the lsmeans package (LENTH, 2016).

Y ijk =µ+TX i +Per j +Cow k +εijk
Eq. ( 2) where Y ijk is the individual observation, µ is the overall mean, TX i is the fixed effect of treatment (i =1, 2, 3, and 4), Per j is the effect of period (j=1, 2, 3, and 4; treated as a random effect), Cow k is the effect of cow (k=1, 2, 3, and 4; treated as a random effect) and ε ijk is the residual error term.A multiple correlation analysis between all variables was also performed in order to find associations that help in explaining CH 4 production.The corr plot routine from the corr plot v.0.77package (WEI & VILIAM 2016) of the R software v.3.2.2. was used (R, CORE TEAM, 2011).

RESULTS
Table 2 shows the effect of the experimental plants on DMI, DMD, milk yield and CH 4 production.Results show that no differences (P>0,05) were observed between the control and the treatments for all variables except for CH 4 production.The treatment with C. bipinnatus reduced CH 4 production Pineda et al. by 98.2L/day (<16%) relative to the control (P<0,05), whereas P. dulce and T. erecta showed no effect on CH 4 production (P>0,05).The LSD between the control and P. dulce was 86.9L/day (P<0,05).All the productive parameters, such as body weight, milk yield, DMI and DMD were not affected by any of the plants tested at the level they were supplemented.Percentages of each plant in relation to the total dry matter intake for the treatments of control, P. dulce, C. bipinnatus and T. erecta were 3.5, 2.6, 3.6 and 3.1%, respectively.Milk composition was also unaffected (P>0,05) by the experimental plants, as shown in Table 3.The multiple correlation analysis showed positive associations between CH 4 L/day and DMI (r=0,5, P<0,05), and between DMD and CH 4 L/day (r=0.6,P<0,01).In contrast, a negative association (r=-0.64,P<0,001) was observed between milk yield and CH 4 in L/Kg of milk.No other association was observed among the studied variables.A large numerical difference of 22.4L was observed between the control and P. dulce treatments for the variable CH 4 , L/kg of DMD; however, this was not significant since the LSD is 53 L (P>0,05).

DISCUSSION
The CH 4 production for individual cows in the control treatment was within the range reported for animals of similar body live weight and intake (NIU et al., 2016).On the other hand, the percentage OM organic matter, CP crude protein, NDF neutral detergent fiber, ADF acid detergent fiber.
Table 2 -Effect of supplementation with P. dulce, C. bipinnatus and T. erecta on methane production, dry matter intake, dry matter digestibility, live weight and milk yield.
- of CH 4 reduction was achieved due to the inclusion of C. bipinnatus and was comparable to the values reported in similar studies, even though the percentage of plants in the diet and the tannin content, 2.6% and 69.7g/kg of plant's DM respectively, were lower than in most published studies.This confirmed our hypothesis that the supplementation of small quantities of this experimental plant can reduce in vivo methane emission by adult cows.The inclusion rate of experimental plants in our research ranged from 2.6 to 3.1% of the average total DMI, whereas in similar studies the supplementation rate is significantly higher.For example, PIÑEIRO et al. (2017) evaluated the effect of increasing levels of Leucaena leucocephala (0, 20, 40, 60 and 80%) on methane production in heifers fed with a low-quality diet.The DM intake in their work ranged from 7 to 7.15kg of DM/day, meaning that the intake of L. leucocephala reached up to 5.72kg of DM/day at an 80% inclusion level.In other studies, the tanniferous plant constituted 100% of the diet, as in in WOODWARD et al. (2002), where dairy cows were fed only with Hedysarum coronarium with no effect on daily CH 4 production.The reduction in methanogenesis observed for C. bipinnatus may be attributed to the higher content of total tannins than P. dulce, although T. erecta has more tannins than C. bipinnatus.This suggested that it may be the type of tannins, particularly condensed tannins that may be responsible for the reduction in methane emissions.-------------------------------------------------------------------------------Treatment-------------------------------------------------------------------------- that the response is not linear, but quadratic, and that there is an interaction (P<0,01) between the plants species and the level of inclusion in the diet for CH 4 production.The positive associations between CH 4 L/ day and DMI (r=0.5, P<0,05) and between DMD and CH 4 L/day have been reported before, because intake is the most important variable that determines CH 4 production; similarly, more CH 4 will be produced in response to more degraded substrate (HALES et al., 2013).In contrast, the negative association observed between milk yield and CH 4 in L/Kg of milk has been reported before by GARG et al. (2013) where increased milk yield per cow was associated with lower methane production per kilogram of milk produced.In fact, intensification of milk production is considered an option for mitigating CH 4 production in dairy systems (GERBER et al., 2011).

CONCLUSION
It can be concluded that supplementation at low doses with C. bipinnatus has the potential to reduce ruminal methane production in dairy cows without affecting animal production parameters.More research is needed in order to determine whether higher levels of C. bipinnatus may lead to further reduction in rumen methanogenesis.Also more research is needed for higher supplementation doses of T. erecta because it showed a trend towards reducing methanogenesis.

Table 1 -
Chemical composition of the silage and concentrate; and chemical composition, total phenols and total tannin concentration of the experimental plants (all in g/Kg of DM, except DM, which is g/Kg).