USING SAPONINS TO REDUCE GASEOUS EMISSIONS FROM STEERS By Wenting Li A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILSOPHY Animal Science 2012 ABSTRACT USING SAPONINS TO REDUCE GASEOUS EMISSIONS FROM STEERS By Wenting Li Enteric methane (CH4) production from beef cattle accounts for more than 71% of the total enteric CH4 fermentation from ruminants. Many nutritional strategies have been investigated in vitro to mitigate CH4 production from ruminants. Saponin is a plant extract that has been demonstrated to be effective in vitro. In this thesis, a series of studies were conducted to investigate the effects of dietary inclusion of steroid (Yucca schidigera) and triterpenoid (Quillaja saponaria and Camellia sinensis) saponins on animal and manure-derived CH4 and other gaseous emissions. In addition, the effects of adding saponin extracts to manure on manure-derived CH4 and other gaseous emissions were also investigated. Dietary inclusion of up to 1.5% of quillaja and yucca saponins or 0.25% of tea saponin did not change animal-derived CH4 emissions, while CH4 emissions were significantly reduced when steers were fed 0.5% tea saponin. The reductions of CH4 production can be possibly attributed to reduced DMI in 0.5% tea saponin treatment. Manure-derived CH4 emissions were reduced in steers fed 0.64% yucca saponins, increased in steers fed 1.5% quillaja saponin treatment and not affected in steers fed 0.25% tea saponin treatment compared to the control treatment. However, direct saponin addition to manure showed no effects on CH4 emissions. Feeding steers up         to 1.5% yucca saponin or 0.5% of tea saponin did not affect animal-derived NH3 emissions. Manure-derived NH3 emissions were reduced in 0.64% yucca saponin treatment. Increased animal-derived NH3 daily emissions were observed in 1.5% quillaja treatment in one of the studies, whereas in another study, 1.5% dietary quillaja saponin supplementation did not change NH3 emissions. The differences may be explained by variation among animals. Animal-derived H2S, NMTHC and N2O emissions were not influenced by dietary saponin inclusion or direct addition. Dietary inclusion of 1.5% of quillaja saponin reduced manure-derived H2S emissions, increased NMTHC emissions but did not affect N2O emissions. Both NMTHC and H2S emissions from manure were reduced as a result of dietary inclusion of 0.64% yucca. Dietary inclusion of 0.25% tea saponin treatments reduced NMTHC, H2S and N2O emissions. Overall, dietary inclusion of all saponin sourced failed to change animal-derived CH4 emissions without affecting growth the performance. Effects of dietary saponin supplementation on manure-derived air emissions were varied by saponin type. Direct application of saponins to manure had no effects on manure-derived air emissions.         ACKNOWLEDGMENTS Though the following dissertation is an individual work, it could not have been written without the help, support and friendship from a lot of people. First of all, I would like to gratefully and sincerely thank my major professor Dr. Wendy Powers, for her guidance, understanding, patience and most importantly her support during my graduate study at Michigan State University. I am not sure many graduate students are given the opportunity to be involved in different projects and develop their own individuality by being allowed to work with such independence. For everything you’ve done for me, Dr. Powers, I thank you. I would also like to gratefully and sincerely thank Dr. Rozeboom, the chair of my guidance committee. He’s been motivating, encouraging, and enlightening especially towards the end of my Ph.D. study. Thirdly, I am very grateful to the remaining members of my guidance committee, Dr. Allen, Dr. Trottier, Dr. Liao and Dr. Yokoyama. Their academic support and input and personal cheering are greatly appreciated. Thank you. I would like to take this opportunity to thank Jolene Roth for her friendship and help during these years. She’s been as tough on me as she’s been supportive. I want to thank Andrew Fogiel and Mark Schilling, without their help and support, my projects could not be completed. Thank you Jolene, Andy and Mark for the many precious memories along the way. I also want to extend my appreciation to the beef farm staffs Ken Metz, Fred Openlander, Phil Summer, technicians Dewey Longuski, Dave Main, Lei Zhang and Kristine Van Winkle and undergraduate students for their assistance throughout my projects.       iv  A very special thank you to my friends Rui Li, Yangmu Liu, Ruoning Wu, Jiesi Luo, Ting Li and Jing Wang for the support they’ve lent me over all these years. Thanks a lot for everything. Last but not least, I would like to thank my dearest friend Danni Ye, my parents Xiaohong Fu, Shengping Li and my family members. Through their love, support and unwavering belief in me, I’ve been able to complete this journey. Finally, I would like to dedicate this work to my past grandmothers. I hope my work makes you proud. v      TABLE OF CONTENTS LIST OF TABLES ......................................................................................................... viii LIST OF ABBREVIATIONS .......................................................................................... x CHAPTER 1 REVIEW OF LITERATURE .................................................................. 1 CLIMATE CHANGE AND GREEN HOUSE GAS EMISSIONS ................................ 2 METHANE EMISSIONS FROM RUMINANTS .......................................................... 3 USING PLANT SAPONIN EXTRACTS TO REDUCE ENTERIC METHANE EMISSIONS .................................................................................................................. 11 USING DYNAMIC ROOMS TO MEASURE GASEOUS EMISSIONS FROM RUMINANTS ............................................................................................................... 22 OBJECTIVES AND RATIONALE STATEMENT ..................................................... 25 LITERATURE CITED ................................................................................................. 27 CHAPTER 2 USING DIETARY SAPONIN EXTRACTS TO REDUCE METHANE EMISSIONS FROM HOSTEIN STEERS ................................................................... 38 INTRODUCTION ......................................................................................................... 41 MATERIALS AND METHODS .................................................................................. 42 RESULTS...................................................................................................................... 48 DISCUSSION ............................................................................................................... 50 CONCLUSION ............................................................................................................. 57 LITERATURE CITED ................................................................................................. 70 CHAPTER 3 EFFECTS OF SAPONIN EXTRACTS ON GASEOUS EMISSIONS FROM STEERS .............................................................................................................. 73 INTRODUCTION ......................................................................................................... 76 MATERIALS AND METHODS .................................................................................. 78 RESULTS...................................................................................................................... 83 DISCUSSION ............................................................................................................... 87 CONCLUSION ............................................................................................................. 94 LITERATURE CITED ............................................................................................... 106 CHAPTER 4 EFFECT OF SAPONIN EXTRACTS, IN THE DIET OF HOLSTEIN STEERS OR ADDED DIRECTLY TO THEIR MANURE, ON GASEOUS EMISSIONS FROM THE MANURE......................................................................... 113 INTRODUCTION ....................................................................................................... 116 vi      MATERIAL AND METHODS .................................................................................. 118 RESULTS.................................................................................................................... 122 DISCUSSION ............................................................................................................. 125 CONCLUSION ........................................................................................................... 130 LITERATURE CITED ............................................................................................... 145 CHAPTER 5 SUMMARY............................................................................................ 148 EFFICACY OF SAPONINS IN REDUCING CH4 EMISSIONS .............................. 149 EFFICACY OF SAPONINS IN REDUCING NH3 AND OTHER GASEOUR EMISSIONS ................................................................................................................ 150 IMPLICATIONS AND RECOMMENDATIONS ..................................................... 151 vii      LIST OF TABLES Table 2.1. Diet and nutrient composition from experimental diets (DM basis) ............... 59 Table 2.2. Effects of Yucca schidigera (YS), Quillaja saponaria (QS) and Camellia sinensis (TS) on gas production, methane concentration at 0 and 4 hr incubation, in vitro ........................................................................................................................................... 61 Table 2.3. Effects of Yucca schidigera (YS), Quillaja saponaria (QS) and Camellia sinensis (TS) on gas production, methane concentration at different time points during 24hr incubation period, in vitro ........................................................................................ 63 Table 2.4. Effects of Yucca schidigera (YS), Quillaja saponaria (QS) and Camellia sinensis (TS) on VFA concentration at 4 hr incubation period, in vitro ........................... 64 Table 2.5. Effects of Yucca schidigera (YS), Quillaja saponaria (QS) and Camellia sinensis (TS) on gas production, methane concentration at 8 and 12 hr incubation period, in vitro ............................................................................................................................... 65 Table 2.6. Effects of Yucca schidigera (YS), Quillaja saponaria (QS) and Camellia sinensis (TS) on gas production, methane concentration at 24 hr incubation period, in vitro ................................................................................................................................... 66 Table 2.7. Growth performance from Holstein steers fed corn-corn silage based diets with different saponin sources .................................................................................................. 67 Table 2.8. Daily manure excretion from Holstein steers fed corn-corn silage based diets with different saponin sources .......................................................................................... 68 Table 2.9. Least squares means from CH4 emissions from Holstein steers fed corn-corn silage based diets with different saponin sources ............................................................. 69 Table 3.1. Diet and nutrient composition from experimental diets (DM basis) ............... 95 Table 3.2. Growth performance from Holstein steers fed corn-corn silage based diets with and without saponin addition ............................................................................................ 97 Table 3.3. Daily manure excretion from Holstein steers fed corn-corn silage based diets with and without saponin addition .................................................................................... 98 Table 3.4. Least squares means from NH3 emissions from Holstein steers fed corn-corn silage based diets with and without saponin addition ....................................................... 99 Table 3.5. Least squares means from N2O emissions from Holstein steers fed corn-corn silage based diets with and without saponin addition ..................................................... 100 ix      Table 3.6. Nitrogen balance from Holstein steers fed corn-corn silage based diets with and without saponin addition .......................................................................................... 101 Table 3.7. Least squares means from CH4 emissions from Holstein steers fed corn-corn silage based diets with and without saponin addition ..................................................... 103 Table 3.8. Least squares means from non-methane total hydrocarbon (NMTHC) emissions from Holstein steers fed corn-corn silage based diets with and without saponin addition ........................................................................................................................... 105 Table 4.1. Ingredient and nutrient composition of diets fed to steers (DM basis) .......... 132 Table 4.2. Least squares means of saponin’s effect on manure CH4 emissions............. 134 Table 4. 3. Least squares means of saponin’s effect on manure NH3 emissions ........... 136 Table 4. 4. Least squares means of saponin’s effect on manure H2S emissions ............ 138 Table 4.5. Least squares means of saponin’s effect on manure N2O emissions ............ 140 Table 4.6. Least squares means of saponin’s effect on manure NMTHC emissions...... 141 Table 4. 7. Initial N content of manure collected from steers fed with or without saponin supplementation (27.5 kg wet manure) ........................................................................... 143 ix      LIST OF ABBREVIATIONS AAQRF …………………….……….. Animal Air Quality Research Facility ADG …………………….…………... average daily gain A:P …………………….……………. acetate to propionate BCTRC ……………………….…….. Beef Cattle Teaching Research Facility BW ………………………………….. body weight CH4 ………………………………...... Methane CO2 Eq. .…………………………….. CO2 equivalent CP …………………………………… crude protein DE …………………………………… digestible energy DM ……………………………....….. dry matter DMI ………………………………..... dry matter intake F:C …………………………………... forage to concentrate ratio GHG …………………….…………... greenhouse gas GI…………………………...………….. gastro-intestinal GWP ……………………….……....... global warming potential EDMI ……………………………….. estimated dry matter intake H2S ………………………………….. hydrogen sulfide MAU ………………………………… Make-up Air Unit MPS …………………………………. microbial protein synthesis N2O ………………………………...... nitrous oxide x      NH3 ………………………………...... ammonia NMTHC ……………………………... none-methane total hydrocarbon RFI …………………………………... residual feed intake TMR ………………………………… total mixed ration Tg ………………………………….... tetra gram VFA ………………………………..... volatile fatty acid RFI ………………………………....... Residual feed intake xi      CHAPTER 1 REVIEW OF LITERATURE 1      CLIMATE CHANGE AND GREEN HOUSE GAS EMISSIONS Greenhouse gases (GHG), natural and anthropogenic originated, absorb thermal infrared radiation from the atmosphere and the Earth’s surface as a mechanism to maintain the Earth’s surface temperature. For those long-lived GHG such as methane (CH4), carbon dioxide (CO2) and nitrous oxide (N2O), the ability to absorb radiation is described by the global warning potential (GWP) index as a comparison to the relative effectiveness of CO2 based on certain duration of time. The 100 yr GWP for CH4 and N2O are 21 and 298 times greater than CO2 (IPCC, 2007). Non-anthropogenic and anthropogenic activities produce GHG. Nevertheless, human related GHG emissions, for example agricultural activities, energy use, land use and industrial processes are considered to be the principal contributors to the drastic increase in atmospheric GHG concentrations (US EPA, 2011). Since pre-industrial times, CO2 concentration has increased by 38%, from 280 parts per million (ppm) to 382 ppm (NOAA, 2008), while CH4 and N2O concentration in the atmosphere has increased 148% and 18%, respectively, compared with the pre-industrial levels (IPCC, 2007). Because the complex and interactive global climate system is very susceptible to the atmospheric concentrations of gases, increased atmospheric GHG concentrations can consequently change global climate (IPCC, 2007). Increased atmospheric temperature is the most direct consequence of increased GHG concentration due to the heat-trapping properties of GHG. It is reported that global surface temperature on average was 0.54 °C higher in 2005 than in 1988 (Hansen et al., 2006). Prediction of temperature by 2040 is suggested to be increased by 1 to 2.7 °C based on different scenarios (Allen et al., 2000). The 2      warmed planet changes climatic patterns, resulting in many severe consequences such as increased sea level and oceanic pH and more frequent extreme weather, which in combination, may greatly affect people’s livelihood (Allen et al., 2010). Because human activities have overwhelmed natural processes of changing climate, reducing GHG emissions from anthropogenic activity could have important implications in the future climate. METHANE EMISSIONS FROM RUMINANTS In the U. S. enteric fermentation from ruminants produces 139.8 Tg CO2 Eq. annually, accounting for more than 20% of total CH4 emissions from human activities (US EPA, 2011). Beef (71%) and dairy (23%) cattle are mainly responsible for the enteric CH4 emissions, which, together represent over 95% of entire enteric CH4 production, with sheep, swine and goat accounting for the rest 5% (US EPA, 2011). In addition, enteric CH4 fermentation from ruminants is also responsible for 2-12% of energy loss (Nelson et al., 1960; Czerkawski, 1978). Methanogenesis in the rumen Methane is produced by methanogens in the rumen, which are obligate anaerobes belonging to the domain of Archaea. Different from bacteria, the cell wall peptidoglycan of methanogens is replaced by pseudomurein (Methanobrevibacter and Methanobacteruim), heteropolysaccharide (Methanosarcina) and protein (Methanomicrobium) (Balch et al., 1979). Bacteria, protozoa and fungi hydrolyze feed nutrients into amino acids and simple carbohydrates such as sugars, which can be further fermented into volatile fatty acids 3      (VFA), principally acetic acid, propionic acid and butyric acid, and utilized by animals as the energy sources (McAllister et al., 1996). Meanwhile, reducing equivalents, mainly + NADH and H , are produced as the electron carriers and need to be oxidized timely to facilitate the process of fiber digestion (Wolin et al., 1997). Although there are several pathways to uptake the reducing equivalents, such as lactic acids, ethanol and H2S formation, CH4 production by methanogens is considered as the more effective electron sink in the rumen (Sharp et al., 1998). Bauchop and Dauglas (1981) demonstrated that in the mono-culture of ruminal fungi, concentration of acetate, ethanol, lactate and hydrogen in final products was 73, 37, 67 and 35 mol/100 mol of hexose units, respectively, without CH4 production. In contrast, when methanogens were co-cultured with fungi, considerable amounts of CH4 were detected (59 mol/100 mol of hexose units) and no accumulation of hydrogen was observed. Meanwhile, improved rate and extent of cellulose degradation were found in co-culture compared to the mono-culture. Formation of acetate increased to 135 mol/100 mol of hexose units; besides the yields of lactate and ethanol decreased to 3 and 19 mol/100 mol of hexose units, respectively. Approximately, 82% of the CH4 formed in the rumen is produced from CO2 follows the reduction of CO2 to formyl, formaldehyde and methyl groups and the conversion of methyl group to methane (Ferry, 1992). To yield one mole CH4, 1 mole of CO2 and 4 moles of H2 are involved, generating 103.4 kJ: CO2 + 4H2 → CH4 +2H2O 4      Although, this reaction is major pathway for most methanogens (Jones et al., 1987), other substrates such as formate and acetate can also be utilized by some methanogens (Garcia et al., 2000). Profiles of methanogens can be highly affected by dietary factors. Grazing sheep were found to have more diversified methanogen populations and strain Methanobrevibacter M6 was more prevalent compared to sheep fed on oaten or lucerne hay diets (Wright et al., 2004). Zhou et al. (2010) observed a shift of dominant methanogen community from Methanobrevibacter ruminantium NT7 when fed a lowenergy diet to Methanobrevibacter smithii and/or Methanobrevibacter sp. AbM4 when fed a high energy diet in beef cattle. Methanogens from feedlot beef cattle fed on cornbased diets with potato by-product based diets, contained only 67% of total clones were found to exist in both herds (Wright et al., 2007). King et al. (2011) reported that under the same feeding regime, management and environmental conditions, there remained a 15% discrepancy in the combined genome library between Holstein and Jersey cows, suggesting that internal factor from the host breeding genetics also has an effect on methanogen community. Role of protozoa in ruminal methanogenesis Protozoa constitute a considerable part of the rumen biomass and are responsible for the extensive production of ruminal ammonia (NH3) by metabolizing rumen bacteria and proteins. The majority of protozoa present in wild and domestic ruminants belong to the order Entodiniomorpid and Holotrich (Williams and Coleman, 1992). The population of protozoa is dynamic and subject to many factors, such as host specificity, geographical distribution, feed composition and young ruminant’s infection (Williams and Coleman, 5      1992). Protozoa population is more diversified in ruminants fed high forage diets rather than high concentrate diets (Dehority, 1978), while they are found to be less diversified (Hristov et al., 2001) or even absent (Lyle et al., 1981) when ruminants are under subclinical acidosis or fed with high grain diets. Butyric and acetic acids are the principal end products of carbohydrate fermentation by protozoa (Howard, 1959; Hansen et al., 2006). Reducing equivalents are usually accompanied by butyric and acetic acids production can later be converted to H2 and used by methanogens to produce CH4. The symbiosis of methanogens and protozoa provides the advantage for fast removal of the H2, which is also recognized as interspecies hydrogen transfer. In the rumen, 20% of the methanogen population is associated with protozoa of which on an average, 43% and 20% are Methanobrevibacter gottschalkii and Methanomicrobium, respectively (Janssen and Kirs, 2008), contributing 9 to 37% of total CH4 production in the rumen (Newbold et al., 1995). High methanogenic activity was observed in fractions containing greater density of protozoa in vitro (Krumholz et al., 1983) and variations of CH4 production in calves were in accordance with composition of ciliate population (Itabashi et al., 1994). Because the presence of protozoa is important to methanogen populations and methanogenic activity, defaunation is suggested to reduce methanogenesis (Newbold et al., 1995). Hu et al. (2005) reported that defaunation resulted in 60% reduction in methanogenesis in rumen fluid. Hess et al. (2003) found 40% to 50% reduction of CH4 6      yield from a defaunation treatment compared to a faunated control treatment, in vitro. Similar results were also reported from in vivo research. Sheep kept under a protozoa-free environment produced 17% to 20% less CH4 compared to faunated sheep during both short-term (10 wk) and long-term (2 yr) studies (Mosoni et al., 2011). When using molecular techniques to examine the changes of methanogen population, studies have found that reduced CH4 production is not always associated with the abundance of methanogens (Guo et al., 2008; Mosoni et al., 2011). Those results indicated that the elimination of protozoa was likely to reduce the amount of hydrogen that is available for methanogens and cause a possible shift from dominant methanogens to less active stains (Denman et al., 2007), as such reducing the ruminal CH4 production, rather than reducing the biomass of methanogen, directly (Hess et al., 2003). Dietary factors affecting methane emissions Decreased forage level or increased forage quality reduces energy losses as CH4 (Johnson and Johnson, 1995; Boadi et al., 2004). Reduced CH4 production is the result of a shift of rumen fermentation patterns from acetate to propionate, which favors an in increased rumen fermentation rate (Demeyer and Van Nevel 1975). In addition, easilyfermented carbohydrates often lead to lower rumen pH, which may reduce the activity of rumen methanogens, resulting in reduced CH4 production (Hristov et al., 2001). However, the reduced CH4 that results from feeding steers a high concentrate diet only occurs when dietary concentrate levels exceed 70% and CH4 production is not linearly associated with concentrate levels. Lovett et al. (2003) compared CH4 emissions from finishing beef 7      heifers fed diets with different forage to concentrate (F:C) ratios (65:35, 40:60 and 10:90). Quadratic responses of CH4 output to reduced F:C ratio were found as emissions were expressed by daily mass, per kg DMI as well as percentage of gross energy intake. The greatest CH4 production was observed as a result of feeding the 40:60 (F:C) ratio treatment. Similar results were also reported by Moss et al. (1995). Possible associative effects could have occurred when a grass hay diet was supplemented with small amount of maize (Blaxter and Vainman, 1964). The effects of dietary crude protein (CP) on CH4 emissions have also been investigated. Kurihara et al. (1997) reported that when increasing the CP content from 4% to 9% in goats fed at maintenance levels, an 18% increase in CH4 production per kg DMI was observed in 9% CP treatment without affecting CH4 production on energy basis. However, when animals were fed above the maintenance level, a negative relationship occurred between dietary CP intake and CH4 emissions (Sekine et al., 1986a and b). Legumes usually contain lower fiber but higher crude protein compared to grasses. In addition, the presence of tannins in legume forages has been demonstrated to reduce methanogenesis (Ahn et al. 1989; Puchala et al., 2005). As a result, lower CH4 yield is generally found when ruminants are fed legume based diets compared with grass based diets (McCaughney et al., 1999). However, the extent of reduction on enteric CH4 emissions is largely determined by the maturity stage of the legume. Advanced maturity of alfalfa for grazing cattle could result in greater CH4 production than grass of less maturity (Chaves et al., 2006). 8      In addition, pelleting and particle size changes CH4 production from ruminants. Hironaka et al. (1996) reported that pelleted alfalfa hay diet reduced daily CH4 emissions on a DMI basis without changing the CH4 emissions on a digestible energy (DE) basis when steers were fed at maintenance level compared to chopped alfalfa hay diet. When increasing the feeding level to 1.6 times maintenance, CH4 production (both DMI and DE basis) was reduced as a result of feeding the pelleted alfalfa hay diet. Intake levels The amount of feed consumed is another factor that is important to determine the daily CH4 emission from ruminants. The relationship between feed intake and CH4 emissions has been investigated intensively. Models used to predict CH4 emissions suggested that DMI accounts for 64% of the daily CH4 production variation (Boadi and Wittenberg; 2002). Generally, increasing DMI levels results in higher CH4 production, whereas the CH4 emissions per unit of DMI decreases or is unaffected (Mills et al., 2001; Mills et al., 2003; Ellis et al., 2007; Ellis et al. 2009). However, poor predictions have been reported when animals are fed low quality diets such as tropical forage and straw (Kurihara et al., 1997; Kurihara et al., 1999). Herd et al. (2002) found that beef cattle of low residual feed intake (RFI) produced approximately 5% less CH4 than high RFI beef cattle. Other studies have shown that differences in CH4 production between low RFI and high RFI beef was 25- 9      28%, corresponding to approximately 16100 L/yr less CH4 in low RFI beef cattle (Nkrumah et al., 2006; Hegarty et al., 2007). Genetic variations Genetic differences may lead to ecological changes of microbial communities in the rumen, translating into different CH4 productions (Hackstein et al., 1996; Nkrumah et al., 2006). Persistent difference of CH4 emissions exists among sheep managed under the same grazing conditions, where 37% more CH4 on a gross energy intake basis was produced in high emitters compared to low CH4 emitters (Pinares-Patino et al., 2003). Robertson and Waghorn (2002) compared cows in New Zealand originating from overseas with domestic cows and found no CH4 production per kg DMI differences at 240 days of lactation, while at 60 and 150 days of lactation, overseas originated cows emitted 15% less CH4 on a DMI basis compared to domestic cows. Machmüller and Clark (2006) reviewed 32 CH4 emission trials with grazing animals and concluded that CH4 emission mass from female cattle was twice more than male cattle. However, because DMI in female cattle was greater than DMI in male cattle, the production of CH4 was reversed when adjusting the daily mass by estimated dry matter intake (EDMI). In sheep species, although females tended to emit 52% more CH4 per day than males, neither the daily mass nor adjusted emissions on EDMI basis showed differences between the two genders (Machmüller and Clark, 2006). Variations may exist in terms of 10      management and diets between male and female animals, but the generic differences of gender should no doubt be taken into account when estimating CH4 emissions. The genetic differences between animals provide the opportunity for selection of low CH4 producers in terms of mitigating CH4 emissions. However, it also reveals the complexity and difficulty in accurate prediction of CH4 emissions. Future strategies, for a better achievement of CH4 reduction, need to take these variations into account. USING PLANT SAPONIN EXTRACTS TO REDUCE ENTERIC METHANE EMISSIONS Saponins are natural glycosides that occur widely in various parts of plants, including the fruits, roots, stems, leaves and seeds (Vincken et al., 2007). Saponins are characterized by several properties, which, most significantly, are the foaming, haemolytic and emulsifying properties (Oda et al., 2000; Price et al., 1987). Chemically, saponins comprise a large family of structurally related compounds containing a steroid or triterpenoid aglycone (sapogenin) linking to one or more oligosaccharide moieties by glycosidic linkage (Makkar et al., 2007). Usually, the aglycone of a steroid saponin is derived from spirostanol or furostanol (Hostettmann and Marston, 1995a). Triterpenoid saponins, however, are more diversified than steroid saponins. Depending on whether amyrin (α- or β- type) or lupeol group is presented in the sapogenin, the triterpenoid saponins can be classified into three classes (Hostettmann and Marston, 1995a). Although it is suggested that the distribution of sapogenins are not subclass-specific (Vincken et al., 2007), triterpenoid saponins are found to be more prevalent in plants compared to steroid saponins (Hostettmann and Marston, 1995b). 11      Saponins are not evenly distributed in plant parts. They are found more abundant in tissues vulnerable to fungi or bacterial infections. For example, in Bacopa monnieri, shoots and leaves are responsible for over half of the total saponin content of the plant (Phrompittayarat et al., 2011). Saponin concentration can be determined by the growth stage of plant as well. Generally, concentration increases as the maturity of plants proceeds (Singh et al., 1986; Phrompittayarat et al., 2011). In addition, sowing dates, growing locations and organs of plants can also affect the saponin composition and concentration considerably (Tsukamoto et al., 1995). Different extract method also can be a factor in terms of determining the saponin concentration in plants (Adebayo et al., 2009). Effects on rumen protozoa population Saponins are toxic to rumen protozoa by forming an irreversible complex with the steroid in protozoal cell wall (Francis et al., 2002). However, the degree of this effect is dose-dependent and subject to saponin types. When 1.2 mg/ml saponin-rich fraction from Quillaja saponaria was added to substrates containing only hay or a 50:50 hay and concentrate mixture, in vitro, 38% to 54% reduction of protozoal population was observed and accordingly, ruminal NH3-N concentration was 12% to 15% lower in Quillaja saponaria treatments compared to the control treatment (Makkar et al., 1998). In another study, 8% less protozoa were observed, in vitro, in saponin treatment where extract from Quillaja saponaria was added at concentrations from 0.1 to 0.4% of DM compared to control treatment (Hristov et al., 2003). The anti-protozoal effects of Yucca schidigera and Camellia sinensis have also been confirmed. When 1 or 10 mg/ml yucca saponin was added to rumen fluid with no substrate, a 22% reduction of protozoal counts 12      was observed in the 1 mg/ml treatment and protozoa were totally eliminated in the 10 mg/ml treatment (Wallace et al., 1994). Linearly dose-dependent effects on protozoa population reduction of yucca saponin was reported by Lovett et al. (2006) and Hristov et al. (1999), in vitro. Protozoal population, measured by real-time PCR was 50% (in vitro; sheep) and 40% (in vivo; sheep and lamb) lower in Camellia sinensis saponin treatments compared to no saponin treatments (Guo et al., 2008; Mao et al., 2010; Zhou et al., 2011). Similar results were also reported from in vivo studies (Hristov et al., 1999; Hess et al., 2004; Lovett et al., 2006). Nasri et al. (2011) reported that administrating saponin extract from Quillaja saponaria to male lambs fed a hay diet with daily supplementation of 400 mg concentration reduced protozoal population by 30 to 47% compared with the control diet. Inclusion of saponins does not always reduce protozoa populations in vivo. Holtshausen et al. (2009) reported that when diets containing 10 g/kg DM saponin extract from Yucca schidigera or Quillaja saponaria were fed to dairy cows, protozoa population was not affected. Both saponin extracts in their study were added at lower concentration in vivo than the lowest concentration tested in vitro, hence the lack of effect can possibly be attributed to the low doses offered. Protozoal communities are significantly influenced by dietary composition and animal species (Williams and Coleman, 1992). This may help to explain the discrepancies between studies in saponin’s efficacy in reducing protozoal population. The inconsistency of saponin’s effects may also arise from different extraction methods. A more pronounced effect of saponin of Acacia concinna against protozoa was observed by methanol extracts compared with water or ethanol extracts (Patra et al., 2006). In 13      addition, it is suggested that saponin’s anti-protozoal effects may be selective towards individual protozoa species. Saponin-rich fruit of Sapindus saponaria was found to reduce Entodiniomorphs numbers without affecting Holotrichs (Abreu et al., 2004). However, this finding is not confirmed in other plant species such as Enterolobium cyclocarpum (Ivan et al., 2004) or Yucca schidigera (Benchaar et al. 2008), suggesting that variations may exist among plant species. Effects on N metabolism Protozoa account for approximately half of the total microbial biomass in the ruminant. One of the major activities of protozoa is the proteolytic effect towards rumen bacteria with NH3 produced as the end product. Inhibition of protozoal populations can therefore prevent the degradation of bacteria, reducing ruminal NH3 concentration and increase the net biosynthesis of microbial crude protein (MCP) in the rumen (Mao et al., 2010; Zhou et al., 2011). The positive relationship between ruminal protozoa population and NH3 concentration has been well established both in vitro and in vivo (Hart et al., 2005, review; Wina et al., 2005, review). Reduction of protozoa is not always accompanied by decreased ruminal NH3 concentrations. In studies with sheep species, dietary inclusion of Enterolobium cyclocarpum saponin at 200 g/d (Ivan et al., 2004) or intra-ruminal (8 g/kg BW0.75) or oral (5 g/kg BW0.75) addition of saponin-rich fruit from Sapindus saponaria (Abreu et al., 2004; Hess et al., 2004) suppressed protozoa population but failed to reduce NH3 concentration in the rumen. This can be attributed to low concentrations of saponin. Hristov et al., (1999) reported that, in heifers, association of decreased NH3 with reduced 14      protozoa population only occurred when treating with higher Yucca schidigera saponin concentrations (5.83 g/kg DM), whereas in lower concentration (1.96 g/kg DM) treatment, protozoa population was reduced by Yucca schidigera saponin without affecting ruminal NH3 concentrations. The inhibition of protozoa may also reduce the protozoal predation activity of bacteria, leading to increased efficiency of microbial protein synthesis (MPS). Saponin extracts from Camellia sinensis (Mao et al., 2010; Zhou et al., 2011), Biophytum petersianum (Santoso et al., 2007), Sapindus saponaria (Abreu et al., 2004) and Yucca schidigera (Santoso et al., 2006) have been identified to improve ruminal MPS. Zhou et al. (2011) noted that inclusion of Camellia sinensis saponin into both faunated and defaunated sheep increased rumen MPS by 16% and 36%, respectively, indicating the lack of interaction between saponin and defaunation. The lack of saponin’s effects on MPS suggests that improved efficiency of MPS in saponin treatments could be due to a greater amount of digested substrate partitioned in to microbial mass synthesis (Makkar et al., 1998). In addition, diet composition can also play a role in MPS (Lu and Jorgensen; 1987). However, the benefits of MPS as a result of saponin inclusion may be compromised at high saponin concentration by suppression of bacteria and fungi population (Wang et al., 2000, Guo et al., 2008). When examined on the whole animal basis, saponin usually did not affect N metabolism. Quillaja saponaria administrated at 30 to 90 mg/kg DM to lambs showed no effect on N intake or losses, suggesting N retention was not affected by saponin treatment (Nasri et al., 2011). Similar results were reported by Hristov et al. (1999) when offered 20 or 60 g/d Yucca schidigera to heifers fed barley grain and alfalfa hay based diet 15      (61:39). Urinary N was not affected by supplementing 120 mg/kg DM of Yucca schidigera saponin (Santoso et al., 2004), while a 15% reduction of urinary N was reported when the dose increased to 240 mg/kg DM in sheep species (Santoso et al., 2006), indicating N metabolism may be changed by saponin at higher concentrations. Effects on rumen VFA fermentation and animal production Regardless of the sources, inclusion of saponins usually does not change total VFA production in vitro, whereas the acetate to propionate ratio (A:P) almost always declines in saponin treatments albeit sometimes propionate concentration does not increase (Hess et al., 2004; Goel et al., 2008; Guo et al., 2008; Pen et al., 2008; Holtshausen et al., 2009; Wang et al., 2011; Zhou et al., 2011). The shift of VFA products from acetate to propionate can probably be explained by the reduction of protozoa population. In some cases, propionate concentration was found to be decreased in rumen fluid containing 100 g/kg Enterolobium cyclocarpum or Pithecellobium saman (Hess et al., 2003). However, it should be noted that the decline of propionate production was due to increased protozoa population. Unlike in vitro studies, rumen VFA production from in vivo experiments showed great variation. Yucca schidigera extract had no influence on total VFA production when fed at 0.075, 10 and 5.83 g/kg DM to steers (Hussain and Cheeke, 1995), dairy cows (Holtshausen et al., 2009) and heifers (Hristov et al., 1999), respectively. Biophytum petersianum (Santoso et al., 2007) reduced total rumen VFA production at the dose of 19.5 mg/kg BW, whereas at a higher dose (26 mg/kg BW) it failed to affect ruminal VFA production. Lu and Jorgensen (1987) reported that 20 g/kg DM lucerne saponin reduced total rumen VFA concentration when diets containing 40% forage and 60% concentrate 16      were fed, but not when diets containing 60% forage and 40% concentrate. Interaction between feed type and saponin concentration has also been reported by Singer et al. (2008), suggesting that response to saponin in terms of rumen VFA production is dietdependent, therefore conclusions should be made diet and saponin concentrations specifically. Reduced rumen A:P ratios, in most cases, were observed in sheep and goat species (Lu and Jorgensen, 1987; Abreu et al., 2004; Hess et al., 2004; Santoso et al., 2006; Santoso et al., 2007). In large ruminants, such as cows, heifers and steers, rumen A:P ratios were usually not changed by saponin treatments (Hussain and Cheeke, 1995; Lovett et al., 2006; Holtshausen et al., 2009), indicating responses to saponin inclusion maybe species-dependent. Fewer studies have investigated effects of saponin containing-diets on animal performance (growth, meat quality, wool and milk production). Saponin extracts from Yucca schidigera, Quillaja saponaria and Camellia sinensis are the few commercialized saponin-rich products and have therefore been used as the main sources of saponin in performance studies. Dairy cows (BW = 586 kg; 69 d post calving) offered 25 g/d Yucca schidigera saponin had greater BW but the milk yield and composition did not differ from that of non-saponin treatment cows (Lovett et al., 2006). Similar results were reported by Singer et al. (2008), where dairy cows (BW = 810 kg; late lactation) fed up to 150 g/d of Yucca schidigera showed no difference on milk production. Milk efficiency of dairy cows (BW = 627; early lactation), when expressed as per kg DMI, was improved when 10 g/kg DM Yucca schidigera (equivalent 230 g/d) and Quillaja saponaria (equivalent 225 g/d) 17      treatments were fed, although milk production was not affected in these treatments (Holtshausen et al., 2009). It needs to be noted that these comparisons are based on crude saponin extracts rather than the actual saponin content, which is affected by the plant source and product. Because animals used in every experiment are not at the same BW and lactation stages, conclusions should be made carefully regarding the effect of saponin-containing diets on milk production. In small ruminants, differences in growth performance and meat quality in lambs were not observed among Quillaja saponaria and control treatments at up to 90 mg/kg DM inclusion (Nasri et al., 2011). Lambs fed 3 g/d Camellia sinensis showed no difference in growth performance (Mao et al. 2011). However, a dose-dependent response of increased BW change was reported when goats were offered 90 to 160 mg/kg DM Yucca schidigera (Aregheore, 2005), suggesting that the lack of effect observed in Mao et al. (2011) and Nasri et al. (2011) could be due to the low doses. Greater DMI and enhanced nutrient digestibility in Yucca schidigera treatment might explain the better growth performance (Aregheore, 2005). In sheep fed a forage-only diet, both 100 g/d and 300 g/d Enterolobium cyclocarpun dietary inclusion improved ADG and wool growth, although in the 300 g/d treatment, DM digestibility was reduced (Navas-Camacho et al., 1993). Effects on methane production A number of studies have demonstrated that ruminal methanogenesis is reduced by dietary inclusion of saponin in vitro. Saponins from Yucca schidigera and Quillaja saponaria have been the most extensively studied and both saponins are suggested to reduce ruminal methanogenesis (Makkar et al., 1998; Wang et al. 1998; Pen et al., 2008). 18      Two studies compared the effects of Yucca schidigera and Quillaja saponaria on methanogenesis and found Yucca schidigera was more effective than Quillaja saponaria. In the first study (Pen et al., 2006), Yucca schidigera extract and Quillaja saponaria extract, at 0, 2, 4, and 6 mL/L, were administered to culture media containing oat hay and concentrate (50:50). Methane was reduced in a dose-dependent manner by up to 42% in Yucca schidigera treatment, while no effect was observed in Quillaja saponaria. In the second study (Holtshausen et al., 2009), saponin extract of Yucca schidigera (6% saponin) and Quillaja saponaria (3% saponin) was added to rumen fluid at 1.5%, 3.0% and 4.5% of substrate DM. Methanogenesis was reduced by 8%, 12% and 26%, respectively in Yucca schidigera and 6%, 11% and 12%, respectively, in Quillaja saponaria. When the reduction of CH4 production was corrected by actual saponin content in both extracts, Yucca schidigera still had a stronger effect on CH4 production over Quillaja saponaria. Guo et al. (2008) found that 5.3 g/kg DM crude Camellia sinensis saponin extract significantly reduced methane production by 8% from sheep. In another study, Hu et al. (2005) observed that methanogenesis was suppressed up to 26% when Camellia sinensis saponin concentration was increased to 40 g/kg DM. However, the suppressive effect of Camellia sinensis on ruminal methanogenesis only occurred in faunated rumen fluid rather than defaunated rumen fluid. Guo et al. (2008) used mcrA gene to monitor the methanogen population and found that in saponin treatment the abundance of methanogens were not affected. This evidence indicates that rather than direct targeting methanogens, declined CH4 production as a result of saponin inclusion could because of 19      their toxicity towards protozoa population, which in turn reduces the availability of hydrogen available for CH4 formation. In other in vitro studies, saponin-rich tropical fruit Sapindus saponaria was found to inhibit methanogenesis by 14% when supplemented at 100 g/kg DM to a forage-based diet containing low quality meadow grass hay (Hess et al., 2003). Methanogenesis was increased in the Enterolobium cyclocarpum treatment but was not affected when Pithecellobium saman was added. The authors suggested that this was because the crude saponin in Sapindus saponaria (120mg/g) is higher than in Enterolobium cyclocarpum (19 mg/g) and Pithecellobium saman (17 mg/g), therefore the lack of effect could be attributed to a dose dependent response (Hess et al., 2003). Given the amount of work conducted in vitro, unfortunately, information relating to the effect of saponins on CH4 production from in vivo studies is relatively sparse. Sheep fed roughage based diet (roughage:concentrate, 70:30) with 120 mg/kg DM supplementation of Yucca schidigera were found to have lower CH4 emissions when expressed as metabolic body weight (per kg BW0.75) and g/kg DMI but not as g per digestible organic matter intake (Santoso et al., 2004). In other studies, neither supplementation of the Yucca schidigera nor Quillaja saponaria reduced CH4 emissions in sheep or dairy cows (Pen et al., 2006, Holtshausen et al., 2009), although in both of those studies, numerical reductions were found in saponin treatments. Considerable reduction of CH4 emissions was found in lambs receiving diets containing 3 g/d Camellia sinensis saponin, where CH4 production mass was 27.2% 20      lower in saponin treatment compared to control treatment (Mao et al., 2010). Similar results (8.71% reduction) were observed in sheep fed 5 g/d Camellia sinensis saponins for 21 days (Yuan et al., 2007). The variability in response to saponin inclusion from in vivo studies could be partially attributed to the sources which may contribute to the variation of saponin concentration in the diets. On the other hand, low dietary saponin concentration offered in some studies in order to avoid negative effects of saponins have on animal performance may also explain the variation. Metabolism and adaptation of saponins in the rumen One of the challenges that impedes application of saponin is the microbial degradation of saponins in rumen. Gestetner et al. (1968) found microorganisms in the cecum and colon of mice were able to deglycosylate soybean saponins. Similar results were also observed in in vitro cultures of both steroid and triterpenoid saponins with rumen fluid (Wang et al., 1998, Makkar et al., 1998). Bacteria that were capable of attacking soluble Medicago sativa (alfalfa) saponin were isolated from steers fed fresh cut alfalfa diet (Gutierrez et al., 1959), suggesting some rumen bacteria might be able to use the sugar moiety of saponin leaving the intact sapogenin part in the rumen. The structure of sapogenin (steroid or triterpenoid) matters more than sugar moieties in the aspect of saponin’s haemolytic activity on bacteria cell membranes (Segal et al., 1966), suggesting that deglycosylation of saponin can enhance the biological activities of saponins in the rumen. Hydrolyzed sapogenin moieties of Narthecium ossifragum underwent oxidative and reductive reactions into epismilagenin, smilagenone, smilagenin and tigogenin in the rumen (Flaoyen and Wilkins, 1997). However, the microbial 21      degradation of sapogenin was found to be limited in Yucca schidigera saponins, even when microbes were pre-exposed to saponin for 20 days (Wang et al, 2008). Another challenge lies in the transient characteristic of the antiprotozoal effect of saponins. Ivan et al. (2004) found that when Enterolobium cyclocarpum was fed to sheep, protozoa population was significantly reduced, but only during the first 11 days, whereas the population increased to almost the same level compared to non-saponin treatment after 14 d. Newbold et al. (1997) noted that protozoa counts in Sesbania sesban saponin treatment recovered after 10 d of feeding. The microbial degradation of saponin was suggested to be one of the explanations and the increased glycosidase activity was considered as one of the adaptation processes (Newbold et al., 1997). In addition, Wang et al. (2000) proposed that rumen protozoa may also adapt to saponin by developing their cell morphology for a better resistance of saponins’ toxicity. Nevertheless, Wina et al. (2006) found the protozoa population did not recover over the 3-months study when Sapindus rarak was fed to sheep, suggesting no adaptation occurred over the long term. To explain the differences in the aspect of microbial adaptation, Teferedegne et al. (1999) compared sheep bred in Scotland and Ethiopia, suggesting that species differences and environmental factors might both contributed to animal’s tolerance to the presence of saponin in the diet and the adaptation might not happen in all animal species. USING DYNAMIC ROOMS TO MEASURE GASEOUS EMISSIONS FROM RUMINANTS A number of techniques have been developed to measure CH4 emissions from ruminants including meteorological techniques, ventilated hood techniques, static or dynamic room techniques and tracer gas technique. The tracer gas and ventilated hood 22      techniques are designed more specifically for enteric CH4 measurement from ruminants, whereas other techniques are also capable of measuring other gaseous emissions, for example ammonia, nitrous oxide, hydrogen sulfide from various sources. Two types of rooms have been used in measuring CH4 emissions from ruminants, static rooms and dynamic rooms. Static rooms are more often applied in measuring CH4 emissions from soil, crop, landfill or manure (Raich et al., 1990). In order to monitor the CH4 emissions, an area needs to be enclosed by a room for a duration of time to allow the accumulation of gaseous concentration in this area, therefore any leakage can contributed to the error of measurement. Commonly, gas samples are collected by vacuumed gas bulbs or syringes and analyzed by GC later. Cheap and easy to use are the two most significant advantages of static rooms. However, when restricting animals in an enclosed area without ventilation, they are very likely to suffer from stress, resulting in inaccuracy of measurement. In addition, temperatures in static rooms usually increase during the enclosed period, uncontrollably (Livingston and Hutchinson, 1995); the lack of temperature regulation is prone to affect the volume of gas samples, causing changes in animal’s metabolism and welfare issues. In dynamic rooms, animals are confined in sealed rooms with inlets to supply fresh air and outlets for exhausted gases. Rooms, depending on the size, are usually capable of housing one or more ruminants, with fixed or regulated temperatures (McGinn et al., 2004; Li et al., 2011). Concentrations of targeting gases are calculated by the difference of exhausted and background ambient air. To account for the accuracy of 23      emitted gaseous concentrations from animals, air flow rates and sophisticated instruments calibrated on a regular basis are required. The extent of control allows dynamic rooms to be less susceptible to be affected by ambient changes compared to other techniques, providing the benefit of using them to compare treatment effects on gaseous emissions from ruminants. Nevertheless, the spatial limitation may change animal’s behaviors and activities thus affecting their emissions. The Animal Air Quality Research Facility (AAQRF) at Michigan State University facilitates 12 dynamic rooms (height = 2.14 m, width = 3.97 m, length = 2.59 m), capable of occupying different animal species, such as swine (6 head/room), laying hens (7 bird/cage, 8 cage/room), boilers (50 bird/room), turkeys (10 bird/room), heifers (1 head/room) and steers (1 head/room). Since 2007, more than 10 research projects investigating the air emissions from different animal species and manure management have been conducted at AAQRF. Each room is constructed with an individual inlet for incoming fresh air and an outlet for exhausting air. Incoming air ducts are designed at the same length and size with one end attached to the room and the other end linked to a main duct which is responsible for the overall supply of ambient air. Through the ceiling on one side of the room enters the ambient air through a tri-directional vent and an exhaust duct, 12 cm above floor level, is located at the corner of the room on the diagonal side to expel the room air. This design allows room air to be well-mixed, providing homogenous gas samples for analysis. Temperatures are controlled through a Make-up Air Unit (MAU) system. Within the desired range, temperatures can be adjusted automatically by air flow rate through a dedicated fan in the main duct. 24      A standard sampling cycle is 195 min. During every cycle, each of the 12 rooms is sampled for a 15-min period. Within the 15-min sampling period, the line is purged for the first 10 min and then data are saved for the remaining 5 min of the sampling period (concentration readings every half second and averaged over the 10 readings). After each of the 12 rooms is sampled, a background sample is collected to obtain baseline readings. Through software control (LabVIEW Version 8.2; National Instruments Corp., Austin, TX), gaseous concentration monitoring of each room occurs in a sequential manner. Daily mass of emitted gas is calculated by summing the mass emitted during each sampling period for that day (7 to 8 daily observations per room (Powers et al., 2007). All emission factors are calculated from emission mass, which is calculated based on the emission rate (the product of concentration and airflow). Gas emission rates are calculated as the product of ventilation rates and concentration differences between exhaust and incoming air using the following equation: ER Q C C 10 where ER is emission rate, g/min; Q is ventilation rate at room temperature and pressure, L/min; T is air temperature in room exhaust, in Kelvin; Co is gas concentration in room exhaust, ppm; Ci is gas concentration in the incoming air, ppm; MW is molecular weight of the gas, g/mol; Vm is molar volume of gas at standard condition (22.414 L/mol). OBJECTIVES AND RATIONALE STATEMENT Due to facility limitations, available data from research efforts regarding the nutritional impacts on enteric fermentation from ruminants are primarily from in vitro studies. Significant knowledge gaps of the relationship between in vitro and in vivo 25      studies and nutritional impacts in vivo still exist. In fact, product and diet-specific impacts on methane emissions in ruminants in vivo are still unknown. Regardless of the efficiencies of saponins in reducing methane emission in vitro, in vivo evidence is required before products can be applied to animal industries. The overarching aim of this project is to investigate potential application of dietary saponin inclusion to reduce CH4 emissions from steers. Because the detrimental effect of saponin against protozoa population has been well established in vitro, we hypothesize that CH4 emissions will be reduced as a result of dietary saponins inclusion, in vivo. This study will combine in vitro approaches with utilization of environmentalcontrolled facilities for air sampling/monitoring at MSU’s AAQRF. The specific objectives of the study were to:  Establish the dose-dependent response of ruminal fermentation and methanogenesis to saponin inclusion, in vitro.  Determine the effects of dietary saponin inclusion on animal-derived methane and other air emissions, in vivo;  Determine the effects of dietary supplementation of saponin and direct application to manure on manure-derived methane and other emissions. 26      LITERATURE CITED 27      LITERATURE CITED Abreu, A., J. E. Carulla, C. E. Lascano, T. E. Diaz, M. Kreuzer, and H. D. Hess. 2004. Effects of Sapindus saponaria fruits on ruminal fermentation and duodenal nitrogen flow of sheep fed a tropical grass diet with and without legume. J. Anim. Sci. 82:1392–1400. Adebayo, E. A., O. R. Ishola, O. S. Taiwo, O. N. Majolagbe, and B. T. Adekeye. 2009. Evaluations of the methanol extract of Ficus exasperata stem bark, leaf and root for phytochemical analysis and antimicrobial activities. Afr. J Plant Sci. 3:283–287. Allen, M. R., P. A. Stott, J. F. B. Mitchell, R. Schnur, and T. Delworth. 2000. Quantifying the uncertainty in forecasts of anthropogenic climate change. Nature. 407:617–620. Allen, C. D., A. K. Macalady, H. Chenchouni, D.Bachelet, N. McDowell,M. Vennetier, T. Kitzberger, A. Rigling, D. D. Breshears, E. H. Hogg, P. Gonzalez, R. Fensham, Z. Zhang, J. Castro, N. Demidova, J. Lim, G. Allard, S. W. Running, A. Semerci, and N. Cobb. 2010. A global overview of drought and head–induced tree mortality reveals emerging climate change risks for forests. Forest Ecol. Manag. 259:660–684. Ahn, J. H., B. M. Robertson, R. Elliott, R. C. Gutteridge, and C. W. Ford. 1989. Quality assessment of tropical browse legumes: tannin content and protein degradation. Anim. Feed Sci. Technol. 27:147–156. Aregheore, E. M. 2005. Effect of Yucca schidigera saponin on the nutritive value of urea–ammoniated maize stover and its feeding value when supplemented with forage legumes (Calliandra calothyrsus) for goats. Small Rum. Res. 56:95–102. Balch, W. E., G. E. Fox, and L. J. Magrum. 1979. Methanogens: reevaluation of a unique biological group. Microbiol. Rev. 43:260–296. Bauchop, T., and D. O. Mountfort. 1981. Cellulose fermentation by a rumen anaerobic fungus in both the absence and the presence of rumen methanogens. Appl. Environ. Microbiol. 42:1103–1110. Benchaar, C., T. A. McAllister, and P. Y. Chouinard. 2008. Digestion, ruminal fermentation, ciliate protozoal populations, and milk production from dairy cows fed cinnamaldehyde, quebracho condensed tannin, or Yucca schidigera saponin extracts. J. Dairy Sci. 91:4765–4777. Blaxter, K. L. and F. W. Wainman. 1964. The utilisation of the energy of different rations by sheep and cattle for maintenance and for fattening. J. Agric. Sci. 63:113–128. Boadi, D. A., and K. M. Wittenberg. 2002. Methane production from dairy and beef heifers fed forages differing in nutrient density using the sulfur hexafluoride (SF6) tracer gas technique. Can. J. Anim. Sci. 82:201–206. 28      Boadi, D. A., K. M. Wittenberg, S. L. Scott, D. Burton, K. Buckley, J. A. Small, and K. H. Ominski. 2004. Effect of low and high forage diet on enteric and manure pack greenhouse gas emissions from a feedlot. Can. J. Anim. Sci. 84:445–453. Czerkawski, J. W. 1978. Reassessment of efficiency of synthesis of microbial matter in the rumen. 61(9):1261–1273. Dehority, B. A. 1978. Specificity of rumen ciliate protozoa in cattle and sheep. J. Protozool. 25:509–512. Demeyer, D. I. and C. J. Van Nevel. 1975. Methanogenesis: an integrated part of carbohydrate fermentation and its control. In W. MacDonald and A. C. Wamer (eds.). Digestion and metabolism in the ruminant. The University of New England. Publishing Unit, Armidale, Australia. pp. 366–382. Denman, S. E., N. Tomkins, and C. S. McSweeney. 2007. Quantitation and diversity analysis of ruminal methanogenic populations in response to the antimethanogenic compound bromochloromethane. FEMS Microbiol. Ecol. 62:313–322. Ellis, J. L., E. Kebreab, N. E. Odongo, B. W. McBride, E. K. Okine, and J. France. 2007. Prediction of methane production from dairy and beef cattle. J. Dairy Sci. 90:3456– 3467. Ellis, J. L., E. Kebreab, N. E. Odongo, K. Beauchemin, S. McGinn, J. D. Nkrumah, S. S. Moore, R. Christopherson, G. K. Murdoch|, B. W. McBride, E. K. Okine, and J. France. 2009. Modeling methane production from beef cattle using linear and nonlinear approaches. J. Anim. Sci. 87: 1334–1345. Ferry, J. G. 1992. Biochemistry of methanogenesis. Crit. Rev. Biochem. Mol. Biol. 27:473–503. Flaoyen, A., and A. L. Wilkins. 1997. Metabolism of saponins from Narthecium ossifragum – a plant implicated in the aetiology of alveld, a hepatogenous photosensitization of sheep. Vet. Res. Commun. 21:335–345. Francis, G., Z. Kerem, H. P. S. Makkar, and K. Becker. 2002. The biological action of saponins in animal systems: A review. Br. J. Nutr. 88:587–605. Garcia, J. L., B. K. C. Patel, and B. Ollivier. 2000. Taxonomic, phylogenetic, and ecological diversity of methanogenic archaea. Anaerobe. 6:205–226. Gestetner, B., Y. Birk, and Y. Tencer. 1968. Soybean saponins. Fate of ingested saponins and the physiological aspect of their haemolytic activity. J. Agric. Food Chem. 16:1031–1035. Chaves, A. V., L. C. Thompson, A. D. Iwaasa, S. L. Scott, M. E. Olson, C. Benchaar, D. M. Veira, and T. A. McAllister. 2006. Effect of pasture type (alfalfa vs. grass) on 29      methane and carbon dioxide production by yearling beef heifer. Can. J. Anim. Sci. 86:409–418. Gutierrez, J., R. E. Daviss, and I. L. Lindahl. 1959. Characteristics of saponin–utilizing bacteria from the rumen of cattle. Appl. Microbiol. 7:304–308. Guo, Y., J. Liu, Y. Lu, W. Zhu, S. E. Denman and C. S. McSweeney. 2008. Effect of tea saponin on methanogenesis, microbial community structure and expression of mcrA gene, in cultures of rumen micro–organisms. Lett. Appl. Microbiol. 47:421–426. Hackstein, J. H. P., P. Langer, and J. Rosenberg. 1996. Genetic and evolutionary constraints for the symbiosis between animals and methanogenic bacteria. Environ. Monit. Assess. 42:39–56. Hansen, J., M. Sato, R. Ruedy, K. Lo, D. W. Lea, and M. Medina–Elizade. 2006. Global Temperature Change. Proc. Natl. Acad. Sci. 101:16109–16114. Hart, K. J., D. R. Yáñez–Ruiz, S. M. Duval, N. R. McEwan, and C. J. Newbold. 2008. Plant extracts to manipulate rumen fermentation. Anim. Feed Sci. Technol. 147:8–35. Hegarty, R. S., J. P. Goopy, R. M. Herd, and B. McCorkell. 2007. Cattle selected for lower residual feed intake have reduced daily methane production. J. Anim. Sci. 85:1479–1486. Herd, R. M., P. F. Arthur, R. S. Hegarty, and J. A. Archer. 2002. Potential to reduce greenhouse gas emissions from beef production by selection for reduced residual feed intake. Proc. 7th World Congr. Genet. Appl. Livest. Prod. 31:281–284. Hess, H. D., M. Kreuzer, T. E. Díaz, C. E. Lascano, J. E. Carulla, C. R. Soliva, and A. Machmüller. 2003. Saponin rich tropical fruits affect fermentation and methanogenesis in faunated and defaunated rumen fluid. Anim. Feed Sci. Technol. 109:79–94. Hess, H. D., R. A. Beuret, M. Lötscher, I. K. Hindrichsen, A. Machmüller, J. E. Carulla, C. E. Lascano, and M. Kreuzer 2004. Ruminal fermentation, methanogenesis and nitrogen utilization of sheep receiving tropical grass hay–concentrate diets offered with Sapindus saponaria fruits and Cratylia argentea foliage. Anim. Sci. 79:177–189. Holtshausen, L., A. V. Chaves, K. A. Beauchemin, S. M. McGinn, T. A. McAllister, N.E. Odongo, P. R. Cheeke, and C. Benchaar 2009. Feeding saponin-containing yucca schidigera and quillaja saponaria to decrease enteric methane production in dairy cows. J. Dairy Sci. 92:2809-2821. Hostettmann, K. and A. Marston. 1995a. In: K. Hostettmann and A. Marston (eds.): Saponins. Cambridge University Press Cambridge, UK, pp.12 Hostettmann, K. and A. Marston. 1995b. In: K. Hostettmann and A. Marston (eds.): Saponins. Cambridge University Press Cambridge, UK, pp.18. 30      Howard, B. H. 1959. The biochemistry of rumen protozoa. 1. Carbohydrate fermentation by Dasytricha and Isotricha. Biochem. J. 71:671–675. Hironaka, R., G. W. Mathison, B. K. Kerrigan, and I. Vlach. 1996. The effect of pelleting of alfalfa hay on methane production and digestibility by steers. Sci. Total Envir. 180:221–227. Hristov, A. N., T. A. McAllister, F. H. Van Herk, K. J. Cheng, C. J. Newbold, and P. R. Cheeke. 1999. Effect of Yucca schidigera on ruminal fermentation and nutrient digestion in heifers. J. Anim. Sci. 77:2554–2563. Hristov, A. N., M. Ivan, L. M. Rode, and T. A. McAllister. 2001. Fermentation characteristics and ruminal ciliate protozoal populations in cattle fed medium– or high–concentrate barley–based diets. J. Anim. Sci. 79:515–524. Hristov, A. N., M. Ivan, L. Neill, and T. A. McAllister. 2003. Evaluation of several potential bioactive agents for reducing protozoal activity in vitro. Anim. Feed Sci. Technol. 105:163–184. Hu, W, Y. Wu, J. Liu, Y. Guo, and J. Ye. 2005. Tea saponins affect in vitro fermentation and methanogenesis in faunated and defaunated rumen fluid. J Zhejiang Univ. Sci. B. 6(8):787–792. Hussain, I., and P. R. Cheeke. 1995. Effect of dietary Yucca schidigera extract on rumen and blood profiles of steers fed concentrate– or roughage–based diets. Anim. Feed Sci. Technol. 51:231–242. IPCC. 2007. In: S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.): Contribution of working group I to the fourth assessment report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Itabashi, H., Y. Washio, A. Takenaka, S. Oda, and T. Ishibashi. 1994. Effects of a controlled fauna on methanogenesis, fiber digestion and rate of growth in calves. Proceedings of the Society of Nutrition Physiology. 3:179. Ivan, M., K. M. Koenig, B. Teferedegne, C. J. Newbold, T. Entz, L. M. Rode, and M. Ibrahim. 2004. Effect of the dietary Enterolobium cyclocarpum foliage on the population dynamics of rumen ciliate protozoa in sheep. Small Rumin. Res. 52:81–91. Janssen, P. H., and M. Kirs. 2008. Structure of the archaeal community of the rumen. Appl. Environ. Microbiol. 74:3619–3625. Johnson, K. A. and D. E. Johnson. 1995. Methane emissions from cattle. J. Anim. Sci. 73: 2483–2492. Jones, W. J., D. P. Nagle, Jr., and W. B. Whitman. 1987. Methanogens and the diversity of archaebacteria. Microbiol. Rev. 51:135–177. 31      King, E. E., R. P. Smith, B. St–Pierre, and A. D.Wright. 2011. Differences in the rumen methanogen populations of lactating jersey and holstein dairy cows under the same diet regimen. Appl. Environ. Microbiol. 77(16):5682–5687. Krumholz, L.R., C.W. Forsberg, and D.M. Veira. 1983. Association of methanogenic bacteria with rumen protozoa. Can. J. Microbiol. 29:676–680. Kurihara, M., M. Shibata, T. Nishida, A. Purnomoadi, and T. Terada. 1997. Methane production and its dietary manipulation in ruminants. In: R. Onodera, H. Itabashi, K. Ushida, H. Yano, and Y. Sasaki (eds.) Rumen Microbes and Digestive Physiology in Ruminants. p 199–208. Japan Sci. Soc. Press, Tokyo. Kurihara, M., T. Magner, R. A. Hunter, and G. J. McCrabb. 1999. Methane production and energy partition of cattle in the tropics. Br. J. Nutr. 81:227–234. Li, W., W. J. Powers, and G. M. Hill. 2011. Feeding distillers dried grains with solubles and organic trace mineral sources to swine and the resulting effect on gaseous emissions. J. Anim. Sci. 89:3286–3299. Livingston, G.P., and G. L. Hutchinson 1995. Enclosure–based measurement of trace gas exchange: applications and sources of error. In: Matson, P., R.C. Harris (eds.): Biogenic Trace Gases: Measuring emissions from soil and water. Blackwell Science, Oxford, UK, pp.14–51. Lovett, D., S. Lovell, L. Stack, J. Callan, M. Finlay, J. Conolly, and F. P. O’Mara. 2003. Effect of forage/concentrate ratio and dietary coconut oil level on methane output and performance of finishing beef heifers. Livest. Prod. Sci. 84:135–146 Lovett, D. K., L. Stack, S. Lovell, J. Callan, B. Flynn, M. Hawkins, and F. P. O’Mara. 2006. Effect of feeding Yucca schidigera extract on performance of lactating dairy cows and ruminal fermentation parameters in steers. Livest. Sci. 102:23–32. Lu, C. D., and N. A. Jorgensen. 1987. Alfalfa saponins affect site and extent of nutrient digestion in ruminants. J. Nutr. 117:919–927. Lyle, R. R., R. R. Johnson, J. V. Wilhite, and W. R. Backus. 1981. Ruminal characteristics in steers as affected by adaptation from forage to all–concentrate diets. J. Anim. Sci. 53:1383–1390. Machmüller, A., and H. Clark. 2006. First results of a meta–analysis of the methane emission data of New Zealand ruminants. International Congress Series. 1293:54–57 Makkar, H. P. S., S. Sen, M. Blummel, and K. Becker. 1998. Effect of fractions containing saponins from Yucca schidigera, Quillaja saponaria and Acacia auriculoformis on rumen fermentation. J. Agric. Food Chem. 46:4324–4328. Makkar, H. P. S., P. Siddhuraju, and K. Becker. 2007. A laboratory manual on quantification of plant secondary metabolites. In: H. P. S. Makkar, S. Siddhuraju, P. 32      Siddhuraju and K. Becker (eds.): Plant Secondary Metabolites. Human Press, Totowa, New Jersey, pp. 130. Mao, H., J. Wang, Y. Zhou and J. Liu. 2010. Effects of addition of tea saponins and soybean oil on methane production, fermentation and microbial population in the rumen of growing lambs. Livest. Sci. 129(1–3):56–62 McAllister, T. A., E. K. Okine, G. W. Mathison, and K. Cheng. 1996. Dietary, environmental and microbiological aspects of methane production in ruminants. Can. J. Anim. Sci. 76:231–243. Mills, J. A., J. Dijkstra, A. Bannink, S. B. Cammell, E. Kebreab, and J. France. 2001. A mechanistic model of whole–tract digestion and methanogenesis in the lactating dairy cow: model development, evaluation, and application. J. Anim. Sci. 79:1584–1579. Mills, J. A. N., E. Kebreab, C. M. Yates, L. A. Crompton, S. B. Cammell, M. S. Dhanoa, R. E. Agnew, and J. France. 2003. Alternative approaches to predicting methane emissions from dairy cows. J. Anim. Sci. 81:3141–3150. McCaughney W. P., K. Wittenberg, and D. Corrigan. 1999. Impact of pasture type on methane production by lactating cows. Can. J. Anim. Sci. 79:221–226. McGinn, S. M., K. A. Beauchemin, T. Coates, and D. Colombatto. 2004. Methane emissions from beef cattle: Effects of monensin, sunflower oil, enzymes, yeast, and fumaric acid. J. Anim. Sci. 82:3346–3356. Mosoni, P, C. Martin, E. Forano. 2011. Long–term defaunation increases the abundance of cellulolytic ruminococci and methanogens but does not affect the bacterial and methanogen diversity in the rumen of sheep. J. Anim Sci. 89:783–791. Moss, A. R., D. I. Givens, and P. C. Garnsworthy. 1995. The effect of supplementing grass silage with barley on digestibility, in sacco degradability, rumen fermentation and methane production in sheep at two levels of intake. Anim. Feed Sci. Technol. 55:9– 33. Nasri, S., H. Ben Salem, V. Vasta, S. Abidi, H. P. S. Makkar, and A. Priolo. 2011. Effect of increasing levels of Quillaja saponaria on digestion, growth and meat quality of Barbarine lamb. Anim. Feed Sci. Technol. 164:71–78. Navas–Camacho, A., M. A. Laredo, A. Cuesta, H. Anzola, and J. C. León. 1993. Effect of supplementation with a tree legume forage on rumen function. Livest. Res. Rural Dev. 5:59–73. Nelson, W.O., R. E. Brown, and R. G. Kingwill. 1960. Factors affecting ratios of CO2: CH4 in bovine rumen gas. J. Dairy Sci. 43:1654–1655. 33      Newbold C. J., B. Lassalas, and J. P. Jouany. 1995. The importance of methanogenesis associated with ciliate protozoa in ruminal methane production in vitro. Lett. Appl. Microbiol. 21:230–234. Newbold, C. J., S. M. El Hassan, J. M. Wang, M. E. Ortega, and R. J. Wallace. 1997. Influence of foliage from African multipurpose trees on activity of rumen protozoa and bacteria. Br. J. Nutr. 78:237–249. Nkrumah, J. D., E. K. Okine, G. W. Mathison, K. Schmid, C. Li, J. A. Basarab, M. A. Price, Z. Wang, and S. S. Moore. 2006. Relationships of feedlot feed efficiency, performance and feeding behaviour with metabolic rate, methane production, and energy partitioning in beef cattle. J. Anim. Sci. 84:145–153. NOAA. 2008. National Weather Service, Climate Prediction Center. Available at http://www.cpc.ncep.noaa.gov/data/indices/sstoi.indices Oda, K., H. Matsuda, T. Murakami, S. Katayama, T. Ohgitani, and M. Yoshikawa. 2000. Adjuvant and haemolytic activities of 47 saponins derived from medicinal and food plants. Biol. Chem. 381:67–74. Patra, A. K., D. N. Kamra, and N. Agarwal. 2006. Effect of plant extracts on in vitro methanogenesis, enzymes activities and fermentation of feed in rumen liquor of buffalo. Anim. Feed Sci. Technol. 128:276–291. Pen, B., C. Sar, B. Mwenya, M. Kuwaki, R. Morikawa, and J. Takahashi. 2006. Effects of Yucca schidigera and Quillaja saponaria extracts on in vitro ruminal fermentation and methane emission. Anim. Feed Sci. Tech. 129:175–186. Pen, B., C. Sar, B. Mwenya, and J. Takahashi. 2008. Effects of Quillaja saponaria extract alone or in combination with Yucca schidigera extract on ruminal fermentation and methanogenesis in vitro. Anim. Sci. J. 79:193–199. Phrompittayarat, W., K. Jetiyanon, S. Wittaya–areekul, W. Putalun, H. Tanaka, I. Khan, and K. Ingkaninan. 2011. Influence of seasons, different plant parts, and plant growth stages on saponin quantity and distribution in Bacopa monnieri. Songklanakarin. J. Sci. Technol. 33(2):193–199. Price, K. R., I. T. Johnson, and G. R. Fenwick. 1987. The chemistry and biological significance of saponins in foods and feedstuffs. Crit. Rev. Food Sci. Nutr. 26:27–135. Pinares–Patino, C. S., M. J. Ulyatt, K. R. Lassey, T. N. Barry, and C. W. Holmes. 2003. Persistence of differences between sheep in methane emission under generous grazing conditions. J. Agric. Sci. 140:227–233. Powers, W. J., S. Zamzow, and B. J. Kerr. 2007. Reduced crude protein effects on aerial emissions from swine. Appl. Eng. Agric. 23(4):539–546. 34      Puchala, R., B. R. Min, A. L. Goetsch, and T. Sahlu. 2005. The effect of a condensed tannin–containing forage on methane emission by goats. J. Anim. Sci. 83:182–186. Raich, J. W., R. D. Bowden, and P. A. Steudler. 1990. Comparison of two static chamber techniques for determining carbon dioxide efflux from forest soil. Soil Sci. Soc. Am. J. 54:1754–1757. Robertson, L.R., and G.C. Waghorn. 2002. Dairy industry perspectives on methane emissions and production from cattle fed pasture or total mixed rations in New Zealand. Proc. N.Z. Soc. Anim. Prod. 62:213–218. Santoso B, B. Mwenya, C. Sar, Y. Gamo, T. Kobayashi, R. Morikawa, K. Kimura, H. Mizukoshi, J. Takahashi. 2004. Effects of supplementing galacto-oligosaccharides, Yucca schidigera or nisin on rumen methanogenesis, nitrogen and energy metabolism in sheep. Livest Prod. Sci. 91(3):209–17. Santoso, B., B. Mwenya, C. Sar, and J. Takahashi. 2006. Ruminal fermentation and nitrogen metabolism in sheep fed a silage–based diet supplemented with Yucca schidigera or Y. schidigera and nisin. Anim. Feed Sci. Technol. 129:187–195. Santoso, B., A. Kilmaskossub, and P. Sambod. 2007. Effects of saponins from Biophytum petersianum Klotzsch on ruminal fermentation, microbial protein synthesis and nitrogen utilization in goats. Anim. Feed Sci. Technol. 137:58–68. Segal, R., M. Mansour, and D. V. Zaitschek. 1966. Effect of ester groups on the haemolytic action of some saponins and sapogenins. Biochem. Pharmacol. 15:1411– 1416. Sekine, J., S. Kondo, M. Okubo, and Y. Asahida. 1986a. A note on estimation of urinary energy from nitrogen content of urine in early weaned calves up to 6 month. Jpn. J. Zootech. Sci. 57:300-304. Sekine, J., S. Kondo, M. Okubo, and Y. Asahida. 1986b. Estimation of methane production in 6–week–weaned calves up to 25 weeks of age. Jpn. J. Zootech. Sci. 57:300–304. Singer, M. D., P. H. Robinson, A. Z. M. Salem, and E. J. DePeters. 2008. Impacts of rumen fluid modified by feeding Yucca schidigera to lactating dairy cows on in vitro gas production of 11 common dairy feedstuffs, as well as animal performance. Anim. Feed Sci. Technol. 146:242–258. Singh, J., K. Gupta, and S. K. Arora.1994. Changes in antinutritional factors of developing seeds and pod wall of fenugreek (Trigonella foenum graceum L). Plant Foods Hum. Nutr. 46:77–84. Sharp, R., C. J. Ziemer,M. D. Stern, and D. A. Stahl. 1998. Taxonspecific associations between protozoal and methanogen populations in the rumen and a model rumen system. FEMS Microbiol. Ecol. 26(1):71–78. 35      Teferedegne, B., F. McIntosh, P. O. Osuji, A. Odenyo, R. J. Wallace, and C. J Newbold. 1999. Influence of foliage from different accessions of the sub–tropical leguminous tree, Sesbania sesban, on ruminal protozoa in Ethiopia and Scottish sheep. Anim. Feed Sci. Technol. 78:11–20. Tsukamoto, C., S. Shimada, K. Igita, S. Kudou, M. Kokubun, K. Okubo, and K. Kitamura. 1995. Factors affecting isoflavone content in soybean seeds: changes in isoflavones, saponins, and composition of fatty acids at different temperatures during seed development. J. Agric. Food Chem. 43:1184–1192. US EPA. 2011. Inventory of U.S. greenhouse gas emissions and sinks: 1990–2009. Washington, DC. U.S. Environmental Protection Agency. Available at http://epa.gov/climatechange/emissions/usinventoryreport.html. Vincken, J.–P., L. Heng, A. de Groot, and H. Gruppen. 2007. Saponins, classification and occurrence in the plant kingdom. Phytochemistry. 68: 275–297. Wallace, R. J., L. Arthaud, and C. J. Newbold. 1994. Influence of Yucca schidigera extract on rumen ammonia concentrations and rumen microorganisms. Appl. Environ. Microbiol. 60:1762–1767. Wang, Y., T. A. Mcallister, C. J. Newbold, L. M. Rode, P. R. Cheeke, and K. J. Cheng. 1998. Effects of Yucca schidigera extract on fermentation and degradation of steroidal saponins in the rumen simulation technique (RUSITEC). Anim. Feed Sci. Technol. 74:143–153. Wang, Y., T. A. Mcallister, L. J. Yanke, Z. J. Xu, P. Cheeke, and K. J. Cheng. 2000. In vitro effects of steroidal saponins from Yucca schidigera extract on rumen microbial protein synthesis and rumen fermentation. J. Sci. Food Agric. 80:2114–2122. Wang, X. F., S. Y. Mao, J. H. Liu, L. L. Zhang, Y. F. Cheng, W. Jin, and W. Y. Zhu. 2011. Effect of the gynosaponin on methane production and microbe numbers in a fungus–methanogen co–culture. J. Anim. Feed Sci. 272–284 Williams, A. G. and S. G. Coleman. 1992. In: A. G.Williams and S.G. Coleman (eds.): The rumen protozoa. Springer Verlag, New York, USA, pp.77–128. Wina, E., S. Muetzel, and K. Becker. 2005. The impact of saponins or saponin– containing plant materials on ruminant production –A review. J. Agric. Food Chem. 53:8093–8105. Wina, E., S. Muetzel , and K. Becker. 2006. The dynamics of major fibrolytic microbes and enzyme activity in the rumen in response to short– and long–term feeding of Sapindus rarak saponins. J. Appl. Microbiol. 100:114–122. Wright, A. D. G., A. J. Williams, B. Winder, C. Christophersen, S. Rodgers, and K. Smith. 2004. Molecular diversity of rumen methanogens from sheep in Western Australia. Appl. Environ. Microbiol. 70:1263–1270. 36      Wright, A. D. G., C. H. Auckland, and D. H. Lynn. 2007. Molecular diversity of methanogens in feedlot cattle from Ontario and Prince Edward Island, Canada. Appl. Environ. Microbiol. 73:4206–4210. Wolin, M. J., T. L. Miller, and C. S. Stewart. 1997. Microbe–microbe interactions. In: P. N. Hobson and C. S. Stewart (eds.): The rumen Microbial ecosystem. Blackie Academic and professional, London, England, pp. 467–491. Yuan, Z.P., C. M. Zhang, L. Zhou, C. X. Zou, Y. Q. Guo, W. T. Li, J. X. Liu, and Y. M. Wu, 2007. Inhibition of methanogenesis by tea saponin and tea saponin plus disodium fumarate in sheep. J. Anim. Feed Sci. 16(Suppl. 2):560–565. Zhou, M., E. Hernandez–Sanabria, and L. Guan. 2010. Characterization of Rumen Methanogenic community variation under different diets and host feed efficiencies using PCR–DGGE analysis. Appl. Environ. Microbiol. 76(12): 3776–3786. Zhou, Y., H. Mao, F. Jiang, J. Wang, J. Liu, and C. S. McSweeney. 2011. Inhibition of rumen methanogenesis by tea saponins with reference to fermentation pattern and microbial communities in Hu sheep. Anim. Feed Sci. Technol. 166–167:93–100. . 37      CHAPTER 2 USING DIETARY SAPONIN EXTRACTS TO REDUCE METHANE EMISSIONS FROM HOSTEIN STEERS Formats in Chapter 2 to 4 are modified according to the requirements of Journal of Animal Science 38      Using dietary saponin extracts to reduce methane emissions from steers W. Li* and W. Powers* 1 , *Department of Animal Science, Michigan State University, East Lansing, MI 48824                                                              1 Corresponding author:wpowers@msu.edu  39      ABSTRACT A total of 3 experiments (Exp), in vitro and in vivo, were conducted to investigate the effects of saponin extracts from Quillaja saponaria (QS), Yucca schidigera (YS) and Camellia sinensis (TS) on CH4 emissions from steers. The in vitro Exp was carried out to determine the effects of saponin inclusions on ruminal methanogenesis and fermentation parameters. Two doses (0.5% and 2.0% of substrate DM) of each saponin were added to the mixture of rumen fluid and buffer for 0, 4, 8, 12 and 24 hr incubation. During the in vivo experiments (Exp 1 & 2), concentrations of saponin extracts added to the diets were determined based on the actual saponin content in the extract to provide 0.54 g/kg DM saponin in the diets. Exp 1 used a 3 × 3 Latin Square design with 4 replicates for each treatment, to compare the effects of saponin containing diets, QS (QS1, 1.5% DM) and YS (YS1, 0.64% DM), to a corn and corn silage based control (C1) treatment on enteric CH4 emissions. The second experiment designed using a Latin Square (2 × 2, 6 replicates) to evaluate the effect of TS (TS2, 0.25%) on enteric CH4 emissions, by comparing it to a corn and corn silage based control diet (C2). For each study, 12 Holstein steers were individually-housed in environmental rooms for 14 d per period. Methane concentrations were monitored in room exhaust air. During in vitro experiment, CH4 production was reduced in all saponin treatments at 24 hr incubation (P < 0.01). Gas production was reduced in TS0.5%, TS2.0% and YS2.0% + treatments, but was not affected in the other saponin treatments. The NH4 -N production was reduced in all saponin treatments expect QS0.5%. Acetate concentration was reduced in all treatments but 0.5% QS and 0.5% YS (P < 0.01). Except the QS0.5% treatment, all saponin treatments reduced A:P ratio compared to the control treatment, at 24hr 40      incubation (P < 0.01). During both in vivo Exp, feeding saponins to steers did not change ADG or manure excretion characteristics (P > 0.05), but feeding steers TS2 decreased DMI compared to C2 (P < 0.01). Methane emission mass, emission factors and manure excretions were not affected by dietary saponin inclusion. Results indicated that dietary supplementation of 0.54 g/kg DM saponin did not affect CH4 emissions. Key words: Quillaja saponaria, Yucca schidigera, Camellia sinensis, CH4 emissions, rumen fermentation, Holstein steer INTRODUCTION Enteric methane (CH4) fermentation from ruminants represents a substantial loss of feed energy and contributes to global climate change. According to US EPA, approximately 28% of CH4 emissions from the U.S. originate from ruminants (US EPA, 2011). Reducing ruminal enteric CH4 production will thus be significant in terms of improving feed efficiencies and moderate the impact of ruminant production on global climate change. Because 20% of the ruminal methanogen population is associated with protozoa, which aplay a critical role in interspecies hydrogen transfer (Newbold et al., 1995), reduction of protozoa population is often accompanied by lower CH4 production in the rumen (Itabashi et al., 1994). Plant saponins from Quillaja saponaria, Yucca schidigera and Camellia sinensis have been shown to suppress ruminal methanogenesis in vitro (Makkar et al., 1998; Wang et al. 1998; Pen et al., 2006; Pen et al., 2008; Guo et al., 2008; Holtshausen et al., 2009), mainly due to their toxicity against rumen protozoa (Guo et al., 2008). However, results from animal studies are not consistent with in vivo studies 41      (Santoso et al., 2004; Holtshausen et al., 2009; Mao et al., 2010). In addition, very few studies have examined the efficacy of dietary saponin inclusion to reduce CH4 emissions from beef steers. We hypothesized that CH4 emissions from beef steers could be reduced as a result of feeding saponin-containing diets. The objectives of current study were to: 1) determine the effects of saponin extracts from Quillaja saponaria, Yucca schidigera and Camellia sinensis on ruminal CH4 production and fermentation in vitro; and 2) compare the effects of saponin extracts from Quillaja saponaria, Yucca schidigera and Camellia sinensis on CH4 production and animal growth performance, in vivo. MATERIALS AND METHODS All animal procedures were approved by the Michigan State University Institutional Animal Care and Use Committee. Three experiments (Exp), one in vitro and two in vivo, were conducted at the Animal Air Quality Research Facility (AAQRF) at Michigan State University. Saponin Sources The 3 saponin extracts used in the Exp were yucca saponin (YS) which is a powder made entirely from the stem of the Yucca schidigera plant and rich in steroidal saponin (contains 8.5% saponin; Desert King International, San Diego, CA, USA); quillaja saponin (QS), which is a triterpenic saponin enriched extract from pure Chilean soap bark tree Quillaja saponaria (contains 3.6% saponin; Desert King International, San Diego, CA, USA); and tea saponin (TS), which is the whole plant saponin extract from Camellia sinensis and rich in triterpenoid saponin (contains 21.6% saponin; Ningbo Good 42      Green Science & Technology, Ningbo, ZJ, China). Saponin concentration from each product was measured by Desert King International (San Diego, CA, USA). In vitro study Experimental design. The in vitro study was conducted as a repeated completely randomized experimental design. Two doses of each saponin (0.5% and 2.0% subtract DM) were compared to a Control treatment with no saponin addition. Rumen fluid was collected from a fistulated dry cow fed on hay diet. The substrates (corn and hay) for in vitro incubation were dried at 55°C and grounded through a 1-mm screen Wiley mill. (Thomas Scientific, NJ). Prior to the incubation, 400 mg dry substrates (50 : 50, corn and hay) and saponins (0%, 0.5% or 2.0% of substrate DM) were added to a 165 ml serum bottle (15 replicates per treatment, 105 bottles in total). In addition, 15 serum bottles, three for each time point, without substrates were prepared as blanks. All bottles were pre-warmed in a water bath at 39 °C before incubation. Rumen fluid and ingesta were obtained in the morning 2 hr after feeding of the dry cow, blended and filtered by 2 layers of cheesecloth. Two volumes of buffer solution contained bicarbonate (Goering and Van Soest, 1970) was then mixed with the strained rumen fluid and maintained in a 39 °C water bath with continuous CO2 flow. The mixture, 30 ml (10 ml rumen fluid and 20 ml buffer fluid), was transferred to the pre-warmed serum bottle and flushed with CO2. Bottles, sealed with a rubber stopper and crimped by an aluminum cap to prevent any gas leakage, were incubated in a 39 °C water bath for 0, 4, 8, 12 and 24 hr. Bottles were hand-shaken every 2 hr . 43      At each time point, 3 bottles from each treatment, including blanks (24 total) were randomly chosen to terminate fermentation for gas and CH4 production (except 0 hr), + ruminal NH4 -N concentration, pH and individual VFA concentration analysis. Total VFA production was calculated as the sum of acetate, propionate and butyrate. Gas and methane production. Gas production at 4, 8, 12 and 24 hr incubation was measured using a relative pressure gauge (Model Media Gauge, SSI Technologies, WI) by inserting the attached 24 gauge needle through the rubber stopper. After the pressure measurement, approximately 10 ml gas sample from each bottle was taken from the head space through a 24 gauge needle and sealed in a syringe for immediate CH4 analysis by gas chromatography (GC, Model 2010, Shimazu, Japan). The calculation for gas production was: ∗ where Vgas is the volume of gas production at each time point, ml; VBT is the volume of serum bottle, ml; VR is the volume of rumen fluid and buffer solution, 30 ml; PEL is the atmospheric pressure in East Lansing, MI, psi and PBT are the pressure measurements from the gauge, psi. Volatile fatty acids, ammonium-N and pH. At each time point (0, 4, 8, 12, 24 hr), the pH of the incubation mixture was measured by a pH meter (Model HQ40d Portable pH meter, HACH, CO). About 5 ml of the contents from each bottle was centrifuged at 26,000 × g for 20 min, 1ml of the supernatant was saved at -20 °C for VFA analysis on high performance liquid chromatography (HPLC, Model Water 712 WISP, Millipore, 44      + MA). The remaining 25 ml were transferred to a pre-weighed tube for immediate NH4 N analysis (FOSS Tecator, MN). In vivo study General Animal Housing and Management. During each of the 2 in vivo experiments (Exp), 12 Holstein steers were housed, individually, in 12 environmentcontrolled rooms at the Animal Air Quality Research Facility at Michigan State University. Temperature was maintained at 13.75 ± 1.38 °C (Exp 1, period 1), 12.77 ± 1.17 °C (Exp 1, period 2), 18.09 ± 1.05 °C (Exp 1, period 3), 20.97 ± 0.85 °C (Exp 2, period 1) and 19.91 ± 0.58 °C (Exp 2, period 2) to remain within the thermoneutral zone of the animals. In each room, steers were confined in a 106.7 cm long × 182.9 cm wide raised stall covered with a rubber matt surface. A fiberglass feeder was placed at the front of the stall and a pan of the same width as the stall was placed at the rear side to collect both urine and feces. Fresh total mixed ration (TMR) feed was sampled by treatment and offered once daily at 16:00 h at 10% above expected DMI. Prior to feeding, orts were removed, weighed and sampled for each steer. Manure was removed and pans were cleaned once daily at 06:00 h to minimize contribution of manure-derived CH4 emissions to total CH4 emissions. A homogenous sub-sample was collected each time manure was removed. Samples for feed, orts and manure were stored at -20 °C until the end of each experimental period. At the end of each period, feed samples were composited by treatment for feed analysis. Orts and manure samples were composited by room for N content analysis. N intake was calculated for each steer as the difference between N 45      offered in diet and N left in the orts. Steers were weighed on 2 consecutive mornings prior to feeding before arriving and after leaving the rooms. Experimental Design and Dietary Treatment. Experiment 1 used a 3 × 3 Latin Square experimental design with 4 replicates for each treatment, while Exp 2 was a 2 × 2 Latin Square experimental design with 6 replicates per treatment. The length of each period was 21 d. All diets offered throughout the 2 Exp were corn-corn silage based (Table 2.1). On d 7 of each period, steers were moved into environmental-controlled rooms for 14 consecutive days for enteric CH4 measurements. Starting BW was 285 ± 9 kg (Exp 1, period 1), 305 ± 10 kg (Exp 1, period 2), 334 ± 9 kg (Exp 1, period 3), 390 ± 9 kg (Exp 2, period 1) and 411 ± 10 (Exp 1, period 2). During Exp 1 and 2, inclusion levels of the 3 products were adjusted to similar saponin concentration in order to compare the effects of different saponins at the same dietary saponin concentration, providing 0.54 g/kg dietary DM of saponin. The experimental diets in Exp 1 used a corn and corn-silage based control diet (C1), a diet containing 1.5% quillaja saponin extract of diet DM (QS1) and a third diet containing 0.64% yucca saponin extract of diet DM (YS1); 4 replicates per treatment. During Exp 2, a diet containing 0.25% tea saponin extract of diet DM (TS2) was compared to a cornsilage based control diet (C2); providing 6 replicates per treatment. Measurements of Gaseous Concentrations Twelve rooms (height = 2.14 m, width = 3.97 m, length = 2.59 m) were designed to continuously monitor incoming and exhaust concentrations of gases (Li et al., 2011). Concentrations of CH4 were measured by Innova 1412 photoacoustic analyzer (Lumasense Technologies, Ballerup, Denmark) with a detection limit at 1000 ppm. 46      Gaseous concentration monitoring of each room occurred in a sequential manner. All emission factors were calculated from emission mass which is calculated based on the emission rate. Gas emission rates were calculated as the product of ventilation rates and concentration differences between exhaust and incoming air using the following equation: ER Q C C 10 where ER is emission rate, g/min; Q is ventilation rate at room temperature and pressure, L/min; T is air temperature in room exhaust, in Kelvin; Co is gas concentration in room exhaust, mg/kg; Ci is gas concentration in the incoming air, mg/kg; MW is molecular weight of the gas, g/mol; Vm is molar volume of gas at standard condition (22.414 L/mole). Emissions in 1 full measurement cycle were estimated by multiplying the ER (g/min) with 195 min. Daily emissions were calculated as sum of the emissions in the 7 or 8 measurement cycles (as described by Li et al., 2011). Chemical Analyses Feed and orts samples were analyzed by Dairy One Forage Testing Laboratory (Dairy One, Inc. Ithaca, NY) for compositional analysis. Feed DM content was analyzed by Near Infrared Reflectance Spectroscopy (NIRS) (AOAC-991.01, 1995). Crude protein, degradable protein, NDF and ADF were analyzed by Foss NIRS systems Model 6500 with Win ISI II v1.5 (AOAC-989.03, 1996). Minerals were analyzed by microwave digestion followed by Inductively Coupled Plasma Mass Spectrometry (ICP). Energy content was determined by an IKA C2000 basic Calorimeter System (IKA Works, NC). Manure NH4-N (AOAC-928.08, 2000) and total Kjeldahl N (TKN; FOSS Tecator, 1987) 47      content was measured by distillation and digestion followed by distillation, respectively, in a Michigan State University laboratory. Statistical Analyses Data from both in vitro and in vivo studies were analyzed using mixed model procedures (SAS Institute, 2008). Results from the repeated in vitro experiments were pooled. The different experiments and serum bottles were treated as random variables. For the in vitro study, dietary effect of saponins and different concentrations were the fixed effects and the interaction between concentration and saponin types was tested in the model. A contrast statement was used to compare the least squares means differences between the Control and each saponin treatment. Orthogonal polynomial contrasts were applied to determine a linear response to saponin concentrations (0, 1.5% and 2.0% DM) for CH4 and gas production. For in vivo study, performance (DMI and ADG), excretion + (wet mass, DM mass, NH4 -N and TKN) and CH4 emissions data was analyzed using a mixed model testing the fixed effects of diet and random effects of steers and period. Day was considered as a repeated measure for DMI and manure excretion (wet mass, DM + mass, NH4 -N and TKN). Tukey’s test was applied to compare treatment differences. Significant differences among the least squares means were declared at P < 0.05. RESULTS In vitro fermentation Inclusion of QS at both concentrations did not affect gas productions during 24 hr incubation, except that at 12 hr incubation the gas production was reduced by 4% as a result of including 0.5% and 1.5% of QS (Table 2.2). At the inclusion of 0.5% of TS, 48      reduced gas production was observed only at the end of 24 hr incubation (P < 0.01) but not at other incubation periods. When TS concentration increased to 1.5%, lower gas production was observed at 12 and 24 hr incubation periods compared to Control treatment (P < 0.01). Yucca saponin, when included at 0.5%, had no effect on gas production during 24 hr incubation compared to Control treatment, while inclusion of 1.5% YS reduced gas production at 4, 12 and 24 hr incubation period (P < 0.01). Tea saponin and YS reduced CH4 production at both inclusion concentrations during every time point (Table 2.2), while QS included at 0.5% did not affect CH4 production within the first 8 hr of incubation. Linear regression relationships between saponin concentrations and CH4 productions were observed for all saponin types. + Saponins inclusion had an immediate effect on ruminal NH4 -N concentrations + (Table 2.3). When NH4 -N concentrations were analyzed at 0 hr, all saponins reduced + NH4 -N concentration by approximately 7% (Table 2.3). After 24 hr incubation, + concentrations of NH4 -N in all saponin treatments, except QS 0.5%, were lower compared to the Control treatment (P < 0.01). The pH values were not affected by saponin inclusions (Table 2.3). Across all treatments, pH was gradually decreased from the beginning of incubation to the end. Average pH values across all treatments were 7.48, 7.36, 6.97, 6.78 and 6.77 at 0, 4, 8, 12 and 24 hr incubation, respectively. At every time point, except 0 hr, 2.0% saponin treatment decreased acetate concentration and the acetate : propionate (A:P) ratio (P < 0.05), and increased propionate concentrations compared to the Control treatment (P < 0.05; Table 2.4 to 2.6). 49      At the end of 24 hr incubation, total VFA production was not affected by saponin inclusion compared to Control treatment, except in 1.5% TS total VFA was reduced by 3% (Table 2.6). Butyrate production at the end of 24 hr incubation was not affected as a result of saponin inclusion (Table 2.6). Growth performance and manure excretion Inclusion of saponin did not affect steer ADG during Exp 1 and Exp 2, respectively (P > 0.05; Table 2.7). Dry matter intake was not affected by dietary supplementation of QS or YS (P > 0.05), but was reduced as a result of TS inclusion (P < 0.01). Manure characteristics were not affected by dietary saponin during both Exp. + Average daily NH4 -N and TKN excretion mass was 33.58 and 67.68 g, respectively in Exp 1 (P > 0.05; Table 2.8) and 46.65 and 94.91 g, respectively, in Exp 2 (P > 0.05). Methane emissions Diets containing 0.54 g/kg QS or YS did not change daily CH4 emissions from steers compared to steers fed C1 diet (Exp 1). Across all treatments, average CH4 daily concentration, emission rate and mass from steers was 8.52 mg/kg, 50.70 mg/min and 79.11 g/d, respectively (P > 0.05; Table 2.9). Similarly, no differences were observed in average daily CH4 concentration (7.89 mg/kg), emission rate (57.86 mg/min) or emission mass (90.27 mg) from rooms where steers fed TS2 treatment compared to C2 (Exp 2).Although the DMI decreased in steers as a result of feeding TS2, CH4 daily emission mass per unit DMI from steers in TS2 treatment showed no difference compared to the C2 treatment. DISCUSSION 50      Ruminal fermentation and in vitro methanogenesis The well-established dose-dependent response to saponin inclusions on ruminal methanogenesis in vitro was confirmed in this study. However, not all saponins were equally effective in reducing ruminal methanogenesis. At 0.5% inclusion level, QS and YS reduced CH4 production by 9% and 8%, respectively, while TS reduced CH4 production by 14 % by the end of 24 hr incubation. At 2.0% inclusion level, supplementation of QS, TS and YS resulted in 14%, 20% and 19% less CH4 compared to Control treatment at 24 hr incubation period. The difference in CH4 reductions can be partially explained by the different saponin concentrations in the extract. Holtshausen et al. (2009) found that when saponin extracts from Yucca schidigera (6% saponin) and Quillaja saponaria (3% saponin) were added to rumen fluid at 1.5%, 3.0% and 4.5% of substrate DM, ruminal CH4 production was reduced by 8%, 12% and 26%, respectively in Yucca schidigera treatments and 6%, 11% and 12%, respectively, in Quillaja saponaria treatments. When the reduction of CH4 was corrected by actual saponin content in both extracts, Quillaja schidigera showed stronger effect against CH4 production over Yucca saponaria at 1.5% and 3.0% inclusion level. Findings from our study are in agreement with Holtshausen et al. (2009) that at both 0.5% and 2.0% inclusion level, QS was more effective in reducing CH4 than TS and YS when the reductions were adjusted by the saponin concentration in the extract. Guo et al. (2008) found that 5.3 g/kg DM crude Camellia sinensis saponin extract significantly reduced methane production by 8%, in vitro. In another study, Hu et al. 51      (2005) observed up to 26% methanogenesis was suppressed when Camellia sinensis saponin concentration increased to 40 g/kg DM. Our results are similar with both Hu et al. (2005) and Guo et al. (2008). Our results showed that adding QS and YS to rumen fluid at 0.5% did not negatively affect gas production at the end of 24 hr incubation, in vitro, which is in agreement with Lila et al. (2003), Hu et al. (2005) and Guo et al. (2008). However, reduced gas production was observed as a result of including 0.5% TS, 2.0% TS and YS. This may be explained as a toxic effect of saponin against rumen protozoa and some fibrolytic bacterium at high saponin concentrations (Hu et al., 2005; Holtshausen et al., 2009). Our results showed that over 24 hr incubation, all saponin treatments, except QS + 0.5%, produced less NH4 -N compared to Control, which is supported by many other studies (Hart et al., 2005, review; Wina et al., 2005, review). Zhou et al. (2011) found that protozoal concentration was decreased from 4.68% of total bacteria to2.66% when dietary saponin concentration increased from 0 to 3g/d. In addition, results from Valdez et al. (1986) showed that protozoa count reduced by 19% in treatment containing 77 mg/kg saponin. The predation activity of rumen protozoa proteolyzes bacteria protein, + releases NH4 -N as the end product. In addition, a reduction in protozoal number usually + leads to decreased NH4 -N concentration (Hart et al., 2005, review; Zhou et al., 2011). + Although the protozoa population was not examined in the current study, declined NH4 N concentration implies that the protozoa population is possibly reduced as a result of saponin inclusion. It is interesting to see that all saponins exerted immediate effects on 52      + ruminal NH4 -N concentrations. Similar results were also reported by Wu et al. (1994) who found both quadratic and cubic effects of yucca saponin concentration on ruminal NH3 concentrations. Yucca saponin is known to bind NH3 (Wu et al., 1994), whereas the present study suggests that both quillaja and tea saponin are also capable of binding ruminal NH3. Because of the CH4 reduction, propionate concentration and decreased A:P ratio are often observed (Makkar et al., 1998; Hu et al., 2005; Holtshausen et al., 2009). Hu et al. (2005) found that at a low saponin dose, total VFA production was increased compared to a control treatment which was in agreement with our observations during the 24 hr incubation. However, at higher concentration, TS reduced total VFA production at the end of 24 hr incubation, while VFA production in QS and YS treatment was not different from control treatment. The TS extract used in our study has a higher concentration of triterpenoid saponin (21.6%) which is 2.5 and 6 times greater than the saponin concentration in QS (8.5% of triterpenoid saponin) and YS (3.6% of steroid saponin) extracts, therefore, the stronger negative effect of TS on acetate and total VFA production could possibly be attributed to its greater concentration. In vivo study The objective of the current study was to mitigate enteric CH4 by dietary supplementation of saponins without impairing animal growth performance. Many studies have shown that dietary saponin inclusion would not affect animal performance (Aregheore, 2005; Nasri et al., 2011; Santoso et al., 2004; Hristov et al., 1999; Holtshausen et al., 2009; Depenbusch et al., 2007). In our study, ADG during either study 53      was not affected, although the DMI was decreased by feeding TS2 treatments, suggesting TS improved the feed efficiency in steers. One study from Hristov et al. (1999) demonstrated that supplementation of 20 g and 60 g of yucca saponin to heifers did not change N concentration in either urine or feces which was also observed in our experiments. Degradation of microorganisms may be decreased as a consequence of saponin inclusion, while the increased flow of microbial protein could lead to better absorption in the small intestine, resulting in no net changes in N metabolism (Lu et al. 1987). Therefore, the lack of an effect on N excretion could be due to the comprehensive microbiological and physiological effects in the rumen and lower GI tract digestion (Newbold et al., 1997; Holtshausen et al., 2009). Although a number of studies have demonstrated that ruminal methanogenesis can be reduced by dietary inclusion of saponin in vitro, results from in vivo studies are not consistent. In small ruminants, supplementation of 120 mg/kg DM Yucca schidigera to roughage-based diets in sheep (roughage:concentrate, 70:30) reduced CH4 emissions 0.75 when expressed as metabolic body weight (per kg BW ) and g/kg DMI (Santoso et al., 2004). Considerable reductions in CH4 emissions was also found in lambs receiving diet containing 3 g/d Camellia sinensis saponin, where CH4 production mass was 27.2% lower in saponin treatment compared to control treatment (Mao et al., 2010). Similar results (8.71% reduction) were reported in sheep fed 5 g/d Camellia sinensis saponins for 21 days (Yuan et al., 2007).However, in large ruminants, dietary saponin inclusions generally produce no effects on enteric CH4 production. Holtshausen et al. (2009) fed dairy cows with 1.0% quillaja or yucca saponin and observed no changes in CH4 54      emissions compared to control treatment. Likewise, in our study, saponin when included at 0.54 g/kg DM (or 10 g/kg DM quillaja saponin extract, 6.4 g/kg DM yucca saponin extract and 2.5% tea saponin extract),did not change enteric CH4 emissions. The variation among large and small ruminants may be attributed to dietary compositions where in sheep or lamb species, experimental diets were usually high in roughage while roughage made up a relatively smaller proportion of the diets in large ruminants. In addition, genetic variations among ruminant species, saponin sources and extraction methods of saponin may also be the factors contributing to the discrepancies. Comparing methane emissions from in vitro and in vivo studies Although saponins were effective in reducing CH4 production in vitro, unfortunately, the effectiveness in was not confirmed from the animal study. The divergences between in vitro and in vivo results may be explained by several factors. The in vitro fermentation technique was primarily developed to evaluate the feed digestion and N utilization in the rumen (Johnson, 1966) and isa wildly used tool for various research purposes not limited to the rumen. The validation of the in vitro technique was challenged by Moss and Givens (1997) because of the poor correlations 2 (R = 0.264) between in vitro and in vivo results. However, more recently Blümmel et al. (2005) demonstrated that CH4 production calculated from the efficiency of microbial 2 production was well correlated (R = 0.89) to measured CH4 emissions. Further study from Getachew et al. (2005) suggested the possibility of applying the in vitro fermentation technique to estimate CH4 productions under commercial conditions. 55      Unlike animal studies, in vitro experiments offer better opportunities to control error. For example, rumen fluid, obtained from one or more fistulated animals, is often composited to provide homogenous samples for in vitro incubation rather than contributing to variation, as such, genetic related variations which are commonly seen in animal studies are eliminated. In addition, substrates for fermentation are provided equally for all treatments and well-mixed in the in vitro incubation system, hence the possible differences of DMI among animals are not considered. By controlling the factors that contribute to experimental errors in animal studies, the in vitro system is very sensitive to small treatment differences. In order to avoid the influence of ingesta from donor animals, rumen fluid used in in vitro studies is usually subjected to several steps, such as filtration and straining, before mixing with the buffer solution and substrates. Elimination of some microbial species especially those attached to the ingesta is inevitable during these procedures. The in vitro system could possibly enrich certain microbial species while leaving some species uncultivable (Johnson, 1966). The changes in microbial communities from the donor animal to the in vitro system will possibly affect the microbial fermentation. In animals, rumen contents are subject to continuous wash out to lower GI tract, the dilution of substrate may therefore reduce the biological effects some dietary supplements especially those whose primarily biological effects are in the rumen, such as saponins (Lu et al., 1987). In contrast, the majority of in vitro experiments conducted to investigate enteric CH4 emissions use closed systems, thereby preventing the outflow of rumen digesta and allowing the accumulation of fermentation end products, which together might amplify the dietary effects on methanogenesis. 56      Another difference between in vitro studies and in vivo experiments is the short experimental period. Very often, the in vitro experiments are completed within 1 d. Comparably, the experimental period in animal studies is usually more than 21 d with at least 14 d of adaptation period. The short experimental duration of in vitro studies limits the day-by-day variation in CH4 production, making the system more vulnerable to small differences among dietary treatments. The acute exposure of rumen microorganisms to treatment diets during in vitro incubations could lead to drastic changes in microbial communities and CH4 production. On the other hand, in animal studies, adaptation may occur during prolonged exposure to treatment diets. In the case of dietary saponin inclusion, the adaptation of ruminal microorganisms is suggested to be one of the major reasons for the lack of effects on methanogenesis, in vivo (Wang et al., 1998, Makkar et al., 1998). In summary, the nature of in vitro systems, better control experimental errors, provide simplified fermentation conditions, offer short duration of data collections and prevent adaptations to treatments, rendering the system to be more sensitive to minor differences among treatments compared to animal studies. In vitro results are not transferable to in vivo effects. CONCLUSION The effect of saponins on CH4 emissions were not confirmed during the shortterm animal study. However, the dose-dependent response of ruminal fermentation parameters to saponin inclusion was confirmed in the current study, in vitro. Possibly 57      explanations may be related to dose and species variations. Dietary inclusion of saponin did not affect growth performance or excretion characteristics in steers. ACKNOWLEDGEMENTS Funding for this work was provided through the National Research Initiative Air Quality Program of the Cooperative State Research, Education, and Extension Service, USDA, under Agreement No. 2005-35112-15356. The authors wish to acknowledge the USDA for their support of this work. The authors would like to thank Jolene Roth, Andrew Fogiel and Mark Schilling for lab and instrumental maintenance and Desert King International (San Diego, CA) for saponin donation and composition analysis. 58      Table 2.1. Diet and nutrient composition from experimental diets (DM basis) * Exp 2 Exp 1 C1 YS1 QS1 C2 TS2 46 46 3 5 46 46 3 5 46 46 3 5 46 46 3 5 46 46 3 5 --- 0.64 -- -1.5 --- --- -- -- -- -- 0.25 50.5 12.2 10.1 27.8 0.29 1.80 50.1 11.9 11.5 26.4 0.29 1.82 Ingredients (%) DM High moisture corn Corn silage Soybean meal # Supplement 50 Saponins yucca saponin quillaja saponin tea saponin Supplement 50, % of DM TM¶ 1.4 24.9 48.3 0.3 9.6 0.2 9.6 5.1 0.7 Akey TM premix #4 Limestone Soybean meal, 48% N TM 80 Rumensin TM salt Vitamin E, 5% Urea, 45% N Potassium chloride Selenium 90 Total 100 Analyzed composition, % DM DM CP ADF NDF P § 46.9 11.6 16.5 27.4 0.27 1.82 46.2 11.8 17.0 27.8 0.28 1.79 46.8 12.1 16.4 28.1 0.27 1.83 ME (Mcal/kg) * Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. C1 = Control diet in Exp 1. 59      Table 2.1. Diet and nutrient composition from experimental diets (DM basis) con’t # ¶ Middle section of the table lists the ingredients for BFS50 supplement for all treatments Akey TM premix # 4 composition: 9% Mg, 4% S, 0.02% Co, 1% Cu, 0.09% I, 2% Fe, 4% Mn, 0.03% Se, 4% Zn, 4,400,000 IU vitamin A, 550,000 IU vitamin D, and 5,500 IU vitamin E/kg (Akey Inc., Lewisburg, OH). § Calculated value. 60      Table 2.2. Effects of Yucca schidigera (YS), Quillaja saponaria (QS) and Camellia sinensis (TS) on gas production, methane concentration at 0 and 4 hr incubation, in vitro Time (hr) Saponin Control QS TS YS # Level % DM 0 Gas (ml) 21.83 0.5 8 12 3.05 34.11 4.71 Gas (ml) 62.95 21.72 2.88 34.25 4.54 2.0 21.66 2.65 * 60.43 33.73 0.5 21.51 2.81 * 3.92 34.08 2.0 21.35 2.39 ** 4.25 60.41 62.11 33.85 0.5 21.53 2.83 * 3.77 33.48 2.0 20.54 1.17 ** 4.26 2.59 0.19 0.21 0.04 0.36 0.12 <0.01 0.33 ** 33.00 3.80 4.30 0.49 Source of variation 0.32 0.01 0.29 <0.01 0.86 0.16 0.69 0.27 <0.01 <0.01 <0.01 <0.01 SEM # 4 Saponin Level Saponin × Level Linear regression QS TS YS * CH4 (ml) Gas (ml) CH4 (ml) 0.41 0.63 0.95 CH4 (ml) 7.80 * ** * <0.01 <0.01 <0.01 Gas (ml) CH4 (ml) 83.95 9.26 7.34 * 83.45 6.60 ** 8.45 83.04 8.00 6.83 82.67 8.19 5.91 7.40 7.04 82.61 83.06 60.62 2.44 6.68 0.64 82.27 2.14 7.51 0.41 0.04 <0.01 0.02 <0.01 <0.01 0.02 0.39 <0.01 0.51 0.25 <0.01 0.49 <0.01 0.19 <0.01 <0.01 <0.01 <0.01 0.12 0.22 0.02 <0.01 <0.01 <0.01 * * 24 * * * 61.29 62.53 ** * * ** * * * * * * ** 8.53 * ** Treatments were corn and hay based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). All saponins were added on substrate DM basis. * Representing the significant difference between Control and saponin treatments (P < 0.05). 61      Table 2.2. Effects of Yucca schidigera (YS), Quillaja saponaria (QS) and Camellia sinensis (TS) on gas production, methane concentration at 0 and 4 hr incubation, in vitro, con’t ** Representing the significant difference within 2 concentrations of one saponin type and between Control and saponin treatments (P < 0.05). 62      Table 2.3. Effects of Yucca schidigera (YS), Quillaja saponaria (QS) and Camellia sinensis (TS) on gas production, methane concentration at different time points during 24hr incubation period, in vitro + Saponin Control QS TS # Level % DM 0 0 0.5 2.0 0.5 2.0 0.5 Time (hr) 8 12 4 17.92 19.70 * * 16.59 20.25 * 17.15 16.71 19.74 * 20.09 16.42 19.23 16.88 * * * ** 2.0 SEM 16.65 1.26 19.82 * 19.32 1.38 Saponin Level Saponin × Level 0.52 <0.01 0.29 <0.01 <0.01 <0.01 YS # pH NH4 -N (mg/dl) * 21.96 22.20 * 21.28 21.93 * 20.97 24 25.52 * 24.21 0 Time (hr) 4 8 12 24 30.09 7.46 7.33 6.94 6.79 6.78 7.48 7.37 7.00 6.82 6.80 23.26 29.40 * 27.81 23.41 28.31 7.46 7.33 6.91 6.81 6.80 22.34 27.90 7.48 7.40 7.01 6.72 6.73 ** * ** * * * * 24.06 28.62 21.65 * ** ** 21.01 22.88 26.16 0.57 0.96 1.14 Source of variation 0.17 0.03 0.02 <0.01 <0.01 <0.01 0.62 0.43 0.01 7.49 7.35 6.95 6.78 6.76 7.49 7.38 7.02 6.81 6.81 7.51 7.34 6.97 6.76 6.73 0.21 0.33 0.33 0.08 0.29 0.34 0.98 0.88 0.36 0.97 0.12 0.15 0.66 0.01 0.32 0.81 0.22 0.58 0.66 0.99 Treatments were corn and hay based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). All saponins were added on substrate DM basis. * Representing the significant difference between Control and saponin treatments (P < 0.05). ** Representing the significant difference within 2 concentrations of one saponin type and between Control and saponin treatments (P < 0.05). 63      Table 2.4. Effects of Yucca schidigera (YS), Quillaja saponaria (QS) and Camellia sinensis (TS) on VFA concentration at 4 hr incubation period, in vitro Saponin Level % DM 0 hr 3.66 Total (mM) 28.39 * 28.14 Acetate (mM) 19.49 * 19.21 1.80 3.66 28.20 18.81 3.42 1.81 3.75 28.25 12.87 3.47 1.81 3.71 28.59 19.27 * 19.09 17.28 11.98 3.44 1.85 3.49 28.43 19.47 2.0 17.94 1.08 12.63 0.61 3.46 0.22 28.36 1.84 Saponin Level Saponin × Level 0.16 0.72 0.17 0.07 0.51 0.13 1.00 0.96 0.99 0.04 0.20 0.06 Control QS TS YS # Total (mM) 17.90 Acetate (mM) 12.60 18.07 2.0 4 hr Propionate (mM) 3.46 Butyrate (mM) 1.84 A:P ratio 3.66 12.66 3.46 1.95 17.78 12.54 3.44 0.5 18.10 12.86 2.0 18.16 0.5 0 0.5 SEM # 1.84 3.66 0.28 0.11 Source of variation 0.50 0.40 0.43 0.89 0.43 0.67 ** * Propionate (mM) 6.55 Butyrate (mM) 2.35 6.59 * 7.10 2.34 * 2.29 6.63 2.35 * 2.41 * ** 7.10 A:P ratio 2.97 * 2.91 ** 2.65 2.90 * ** 2.69 2.37 18.93 1.52 6.60 * 7.05 0.24 2.37 0.08 2.95 ** 2.68 0.12 0.14 <0.01 0.18 0.39 <0.01 0.64 0.03 0.91 0.03 0.51 <.01 0.77 * Treatments were corn and hay based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). All saponins were added on substrate DM basis. * Representing the significant difference between Control and saponin treatments (P < 0.05). ** Representing the significant difference within 2 concentrations of one saponin type and between Control and saponin treatments (P < 0.05). 64      Table 2.5. Effects of Yucca schidigera (YS), Quillaja saponaria (QS) and Camellia sinensis (TS) on gas production, methane concentration at 8 and 12 hr incubation period, in vitro Saponin Control QS TS # Level % DM 8 hr Total (mM) Acetate Propionate Butyrate A:P ratio (mM) (mM) (mM) Total (mM) Acetate Propionate Butyrate (mM) (mM) (mM) A:P ratio 43.22 * 41.86 31.55 * 29.99 68.40 46.03 * 41.62 16.39 * 18.32 5.98 * 4.99 2.81 * 2.29 2.0 40.16 15.90 * 15.44 4.82 2.53 0 8.36 3.31 42.60 30.04 * 8.53 * 9.00 3.34 * 3.56 3.34 64.92 * 60.88 0.5 42.64 3.59 60.15 39.97 42.91 8.56 * 9.08 3.43 2.0 30.64 * 30.40 3.43 3.35 59.52 39.57 15.59 4.36 2.54 0.5 42.78 8.68 65.35 43.50 * 42.11 1.23 16.33 5.53 16.83 0.54 5.48 0.28 2.68 * 2.51 0.07 <0.01 <0.01 0.05 0.02 0.35 0.04 <0.01 <0.01 0.14 <0.01 <0.01 0.04 0.5 2.0 SEM 42.43 3.35 30.69 * 29.83 2.20 Saponin Level Saponin × Level 0.47 0.03 0.55 0.45 <0.01 0.58 YS # 12 hr * * 8.89 0.92 0.85 <0.01 0.35 3.79 * 3.54 ** * * * 3.55 3.41 * ** 3.72 3.37 0.25 0.13 Source of variation 0.19 0.85 <0.01 <0.01 0.09 0.97 * ** * * 64.42 1.85 <0.01 <0.01 0.05 * * ** * * 4.74 ** ** * 2.59 * Treatments were corn and hay based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). All saponins were added on substrate DM basis. * Representing the significant difference between Control and saponin treatments (P < 0.05). ** Representing the significant difference within 2 concentrations of one saponin type and between Control and saponin treatments (P < 0.05). 65      Table 2.6. Effects of Yucca schidigera (YS), Quillaja saponaria (QS) and Camellia sinensis (TS) on gas production, methane concentration at 24 hr incubation period, in vitro Level Saponin % DM Total (mM) # 0 Control 87.78 0.5 88.26 QS 2.0 86.97 TS YS Acetate (mM) Propionate Butyrate A:P ratio (mM) (mM) 58.68 58.65 21.29 21.59 7.81 8.02 2.75 2.71 56.21 23.31 7.45 2.41 57.73 22.10 7.39 2.60 * * * * * 0.5 87.22 2.0 85.55 54.89 23.51 7.15 2.34 0.5 88.02 58.22 22.13 7.66 2.63 2.0 87.10 0.46 7.64 0.74 2.35 0.17 0.26 0.14 0.62 0.17 <0.01 0.73 SEM # 24 hr Saponin Level Saponin × Level * ** * * * 55.77 23.69 6.02 0.91 Source of variation 0.28 0.19 0.48 0.05 <0.01 <0.01 0.81 0.74 0.91 ** * ** Treatments were corn and hay based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). All saponins were added on substrate DM basis. * Representing the significant difference between Control and saponin treatments (P < 0.05). ** Representing the significant difference within 2 concentrations of one saponin type and between Control and saponin treatments (P < 0.05). 66      Table 2.7. Growth performance from Holstein steers fed corn-corn silage based diets with different saponin sources BW kg Diets DMI kg ADG kg 6.47 0.95 6.46 6.51 0.06 0.85 0.96 0.30 0.80 0.62 ¶ * C1 QS1 YS1 SEM Diet C2 TS2 Exp 1 Period 1 Period 2 Period 3 Start End Start End Start End 304 318 338 363 283 286 288 290 308 324 332 352 284 293 302 319 331 346 9 9 10 9 9 10 Source of variation 0.91 0.86 0.91 0.88 0.83 0.48 Exp 2 Period 1 Period 2 Start End Start End 400 411 409 428 SEM * 380 9 Diet 0.14 399 9 414 10 Source of variation 0.36 0.76 b 1.05 a 7.71 429 10 7.16 0.08 1.21 0.96 <.01 0.55 0.21 Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. LC1 = Control diet in Exp 1. Exp 1, 1.5% DM of quillaja and 0.64% DM of yucca saponin; Exp 2, 0.25% DM of tea saponin. ¶ Exp 1, n=12, 4 replicates per treatment; Exp 2, n=12, 6 replicates per treatment. a, b, c Significant differences observed at the P < 0.05 probability level. 67      Table 2.8. Daily manure excretion from Holstein steers fed corn-corn silage based diets with different saponin sources Diets Wet, kg Total Daily + DM, kg Total NH4 -N, g Daily § TKN , g Total Daily Total Daily 465.47 477.24 411.11 28.63 34.53 35.47 30.73 2.09 890.58 949.66 900.53 50.07 65.99 70.34 67.01 3.72 0.26 0.26 0.67 0.67 627.36 650.81 38.50 44.81 46.49 2.70 1339.45 1318.00 65.72 95.68 94.15 4.69 0.67 0.67 0.82 0.82 ¶ * C1 QS1 YS1 SEM 159.12 167.32 151.32 6.84 11.86 12.44 11.31 0.50 0.28 0.28 C2 TS2 SEM 221.68 223.68 10.87 15.83 15.97 0.78 Diet 0.89 0.89 Diet Exp 1 29.88 2.22 33.23 2.47 30.18 2.25 2.07 0.15 Source of variation 0.46 0.46 Exp 2 45.19 3.23 44.68 3.19 2.49 0.18 Source of variation 0.88 0.88 * Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. C1 = Control diet in Exp 1. Exp 1, 1.5% of quillaja and 0.64% of yucca saponin; Exp 2, 0.25% of tea saponin. ¶ Exp 1, n=12, 4 replicates per treatment; Exp 2, n=12, 6 replicates per treatment; Exp 3, n=12, 3 replicates per treatment per period. § TKN = total kjeldahl N. 68      Table 2.9. Least squares means from CH4 emissions from Holstein steers fed corn-corn silage based diets with different saponin sources Diets Daily concentration mg/kg Daily emission rate mg/min Daily emission mass g/d Emission factors mg/kg BW g/kg DMI 252.28 263.97 261.44 16.82 12.10 12.31 12.23 0.65 0.13 0.77 229.06 221.66 5.92 11.78 12.31 0.28 0.37 0.23 ¶ Exp 1 C1* QS1 YS1 SEM Diet 0.70 LC2 LTS2 SEM * 8.41 8.61 8.53 0.55 7.91 7.87 0.38 Diet 0.71 49.52 77.26 51.43 80.24 51.17 79.82 3.08 4.82 Source of variation 0.22 0.22 Exp 2 57.95 90.40 57.78 90.14 3.80 5.93 Source of variation 0.91 0.91 Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. C1 = Control diet in Exp 1. Exp 1, 1.5% of quillaja and 0.64% of yucca saponin; Exp 2, 0.25% of tea saponin. ¶ Exp 1, n=12, 4 replicates per treatment; Exp 2, n=12, 6 replicates per treatment. a, b, c Significant differences observed at the P < 0.05 probability level. 69      LITERATURE CITED 70      LITERATURE CITED Blümmel M., Givens D. I., Moss A. R. 2005. Comparison of methane produced by straw fed sheep in open-circuit respiration with methane predicted by fermentation characteristics measured by an in vitro gas procedure. Anim. Feed Sci. Technol., 123: 379-390. Getachew, G., P. H. Robinson, E. J. DePeters, S. J. Taylor, D. D. Gisi, G. E. Higginbotham, and T. J. Riodan. 2005. Methane production from commercial dairy ration estimated using an in vitro gas technique. Anim. Feed Sci. Technol. 123:391– 402. Goering, H. K. and P. J. Van Soest. 1970. Forage and fiber analysis. Agricultural Handbook, no.379. U.S. Department of Agriculture. Guo, Y., J. Liu, Y. Lu, W. Zhu, S. E. Denman and C. S. McSweeney. 2008. Effect of tea saponin on methanogenesis, microbial community structure and expression of mcrA gene, in cultures of rumen micro–organisms. Lett. Appl. Microbiol. 47:421–426. Hart, K. J., D. R. Yáñez–Ruiz, S. M. Duval, N. R. McEwan, and C. J. Newbold. 2008. Plant extracts to manipulate rumen fermentation. Anim. Feed Sci. Technol. 147:8–35. Holtshausen, L., A. V. Chaves, K. A. Beauchemin, S. M. McGinn, T. A. McAllister, N.E. Odongo, P. R. Cheeke, and C. Benchaar 2009. Feeding saponin-containing yucca schidigera and quillaja saponaria to decrease enteric methane production in dairy cows. J. Dairy Sci. 92:2809-2821. Hu, W., Y. Wu, J. Liu, Y. G, and J. Y. 2005. Tea saponins affect in vitro fermentation and methanogenesis in faunated and defaunated rumen fluid. J. Zhejiang Univ. Sci. 6B(8):787-792. Itabashi, H., Y. Washio, A. Takenaka, S. Oda, and T. Ishibashi. 1994. Effects of a controlled fauna on methanogenesis, fiber digestion and rate of growth in calves. Proceedings of the Society of Nutrition Physiology. 3:179. Ivan, M., K. M. Koenig, B. Teferedegne, C. J. Newbold, T. Entz , L. M. Rode, and M. Ibrahim. 2004. Effects of the dietary Enterolobium cyclocarpum foliage on the population dynamics of rumen ciliate protozoa in sheep. Small Rum. Res.: 52:81–91. Johnson, R. R. 1966. Techniques and procedures for in vitro and in vivo rumen studies. J. Anim. Sci. 25:855-875. Lu, C. D., and N. A. Jorgensen. 1987. Alfalfa saponins affect site and extent of nutrient digestion in ruminants. J. Nutr. 117:919–927. Makkar, H. P. S., S. Sen, M. Blummel, and K. Becker. 1998. Effect of fractions containing saponins from Yucca schidigera, Quillaja saponaria and Acacia auriculoformis on rumen fermentation. J. Agric. Food Chem. 46:4324–4328. 71      Moss, A.R., and D. I. Givens. 1997. Methane production from a range of feedstuffs as determined in vitro using the cumulative gas production technique and compared with that measured in vivo. Proc. Brit. Soc. Anim. Sci. 194 (Abstr.). Newbold C. J., B. Lassalas, and J. P. Jouany. 1995. The importance of methanogenesis associated with ciliate protozoa in ruminal methane production in vitro. Lett. Appl. Microbiol. 21:230–234 Pen, B., C. Sar, B. Mwenya, M. Kuwaki, R. Morikawa, and J. Takahashi. 2006. Effects of Yucca schidigera and Quillaja saponaria extracts on in vitro ruminal fermentation and methane emission. Anim. Feed Sci. Tech. 129:175–186. Pen, B., C. Sar, B. Mwenya, and J. Takahashi. 2008. Effects of Quillaja saponaria extract alone or in combination with Yucca schidigera extract on ruminal fermentation and methanogenesis in vitro. Anim. Sci. J. 79:193–199. Santoso B, B. Mwenya, C. Sar, Y. Gamo, T. Kobayashi, R. Morikawa, K. Kimura, H. Mizukoshi, J. Takahashi. 2004. Effects of supplementing galacto-oligosaccharides, Yucca schidigera or nisin on rumen methanogenesis, nitrogen and energy metabolism in sheep. Livest Prod. Sci. 91(3):209–17. SAS Institute. 2008. Version 9.2. SAS Institute, Inc., Cary, NC, USA. U.S. EPA. 2011. Inventory of U.S. greenhouse gas emissions and sinks: 1990-2009. Washington, DC. U.S. Environmental Protection Agency. Available at http://epa.gov/climatechange/emissions/usinventoryreport.html. Valdez, F. R., L. J. Bush, A. L. Goetsch, and F. N. Owens. 1986. Effect of steroidal sapogenins on ruminal fermentation and on production of lactating dairy cows. J. Dairy Sci. 69:1568–1575. Wang, Y., T. A. Mcallister, C. J. Newbold, L. M. Rode, P. R. Cheeke, and K. J. Cheng. 1998. Effects of Yucca schidigera extract on fermentation and degradation of steroidal saponins in the rumen simulation technique (RUSITEC). Anim. Feed Sci. Technol. 74:143–153. Wang, X. F., S. Y. Mao, J. H. Liu, L. L. Zhang, Y. F. Cheng, W. Jin, and W. Y. Zhu. 2011. Effect of the gynosaponin on methane production and microbe numbers in a fungus–methanogen co–culture. J. Anim. Feed Sci. 272–284. Wu, Z., M. Sadik, F. T. Sleiman, J. M. Simas, M. Pessarakli, and J. T. Huber. 1994. Influence of Yucca extract on ruminal metabolism in cows. J. Anim. Sci. 72:1038−1042. 72      CHAPTER 3 EFFECTS OF SAPONIN EXTRACTS ON GASEOUS EMISSIONS FROM STEERS 73      Effects of saponin extracts on air emissions from steers 2 W. Li* and W. Powers* , *Department of Animal Science, Michigan State University, East Lansing, MI 48824                                                              2 Corresponding author:wpowers@msu.edu  74      ABSTRACT A series of experiments (Exp) were conducted to quantify the effects of saponin extracts from Quillaja saponaria (QS), Yucca schidigera (YS) and Camellia sinensis (TS) on gaseous emissions from steers. During Exp1, a control diet (C1; corncorn silage basal diet) was compared to YS1 (C1 + 0.64% dietary DM of YS) and QS1 (C1 + 1.5% dietary DM of QS); 4 replicates per treatment. During Exp 2, the control diet (C2; corn-corn silage basal diet) was compared to TS2 (C2 + 0.25% dietary DM of TS). Product inclusion levels were established to provide the same concentration of saponin compounds across studies for Exp 1 and 2. Experiment 3 compared C3 (corn-corn silage basal diet), QS3 (C3 + 1.5% QS), YS3 (C3 + 1.5% YS) and TS3 (C3 + 0.5% TS). Holstein steers (n = 12) at initial BW of 354 ± 10 kg (Exp 1), 429 ± 10 kg (Exp 2), 382 ± 16 kg (period 1, Exp 3) and 400 ± 12 kg (period 2, Exp 3) were housed, individually, in environmental rooms for 22 d per study. Gaseous emissions including methane (CH4), ammonia (NH3), hydrogen sulfide (H2S), nitrous oxide (N2O) and non-methane total hydrocarbon (NMTHC) were monitored in room exhaust air. No differences in DMI (7.54 ± 0.09 kg) and ADG (1.16 ± 0.19 kg) were observed in Exp 1 (P > 0.05). Adding TS2 to the diet improved DMI in Exp 2 (8.94 kg in TS2 vs. 8.53 in C2; P < 0.01), while ADG was not affected by diet. During Exp 3, steers fed the TS3 diet ate less (6.36 kg/d) and gained less (0.31 kg/d) compared to the other 3 treatments. Saponin inclusion did not alter daily CH4 emission per unit DMI (13.17, 10.90 and 13.21 g/kg DMI, for Exp 1, 2, and 3, respectively). Emissions of NH3 per unit N intake were not affected by diets in Exp 1 (134.89 mg/g N consumed) and Exp 3 (134.99 mg/g N consumed). Feeding TS2 reduced NH3 emission per unit of N consumed by 30% compared to C2 (P < 0.01). 75      Feeding up to 0.5% of TS failed to reduce CH4 emissions without impairing steer growth. Other gaseous emissions were not affected by TS addition. Air emissions were not affected by feeding steers with up to 1.5% YS. Feeding 1.5% QS to steers had an inconsistent effect upon NH3 emissions and no other effects upon gaseous emissions from steers in this study. Key words: Quillaja saponaria, Yucca schidigera, Camellia sinensis, air emissions, Holstein steer INTRODUCTION Environmental issues related to animal agriculture are becoming increasingly important, especially as they are related to impacts on global climate change. Enteric fermentation from ruminants produces 139.8 Tg CO2 equivalents annually, representing 28% of the total GHG emissions from the agriculture sector, according to the greenhouse gas (GHG) emission inventory from US EPA (US EPA, 2011). Beef cattle are estimated to generate 71% of total enteric CH4 fermentation from animals (US EPA, 2011). Mitigation strategies to reduce CH4 emissions from beef cattle are needed. Nutritional studies have been conducted to investigate the potential for different feed additives such as fatty acids and oils (Beauchemin et al., 2009; Mao et al., 2010), yeast products (Chung et al., 2011) and plant extract compounds (Zhou et al., 2011) to reduce enteric CH4 production. Among the category of plant extract compounds, saponins are more often studied and their suppression effects on methanogenesis via inhibition of protozoa populations have been confirmed, in vitro (Hess et al., 2003; Pen et 76      al., 2006; Guo et al., 2008). However, in vivo findings are not consistent (Santoso et al., 2004; Pen et al., 2006; Holtshausen et al., 2009; Zhou et al., 2011). The U.S. beef industry produces 25.2 billion kg of red meat, representing $51.5 billion sales during 2010 (USDA National Agricultural Statistics Service, 2011). Beef production generates approximately 500 thousand tons of N in excreta (De Wit et al., 1996), of which up to 70% of the N excreted is volatized into atmosphere as ammonia (NH3; Muck and Richards, 1983; Moreira and Satter, 2006; Hristov et al., 2009). Ammonia, along with nitrous oxide (N2O; another important greenhouse gas) and particle matter (PM) cause environmental, health and welfare issues to both human and animals (Lipfert, 1994; Pope and Dockery, 2006). In addition, the deposition of NH3 will result in soil acidification (Falkengren-Grerup, 1986). Much is known about the impacts of dietary CP concentration on NH3 emissions, but relatively less information is available about the effects of dietary saponin additions on air emissions from ruminants. It has been established in vitro that by inhibiting the protozoa population, ruminal + NH4 -N concentration decreases and microbial protein synthesis increases (Guo et al., 2008; Zhou et al., 2011). Saponin has detrimental effects on protozoa population, therefore we hypothesized that both CH4 and NH3 emissions reduced as a result of dietary inclusion of saponins. The objectives of the study were to investigate the effects of feeding steers 3 different saponin extracts from Quillaja saponaria, Yucca schidigera and Camellia sinensis on 1) CH4 and NH3 emission, in vivo; 2) potential changes in 77      hydrogen sulfide (H2S), N2O and non-methane total hydrocarbon (NMTHC) emissions that were unintended and, 3) excretion characteristics. MATERIALS AND METHODS All animal procedures were approved by the Michigan State University Institutional Animal Care and Use Committee (protocol # 01/10-004-00). General Animal Housing and Management During each of 3 experiments (Exp), Holstein steers were housed, individually, in 12 environment-controlled rooms at the Animal Air Quality Research Facility at Michigan State University. Temperature was maintained at 19.96 ± 0.91 °C during Exp 1 and 2 and was 16.69 ± 1.25 °C during Exp 3 to remain within the thermoneutral zone of the animals. In each Exp, 12 steers were used. The same steers were fed during Exp 1 and 2. Experiment 3 was conducted 3 months later therefore a new group of steers were fed. Steers, each housed individually, were confined in a 106.7 cm long × 182.9 cm wide raised stall covered with a rubber matt surface. A fiberglass feeder was placed at the front of the stall and a pan of the same width as the stall was placed at the rear to collect both urine and feces. Fresh TMR feed was sampled by treatment and offered once daily at 16:00 h at 10% above expected DMI. Prior to feeding, orts were removed, sampled by room and weighed. Manure was mixed thoroughly every morning and removed partially to maintain an equal depth of 5 cm so as to provide an emissions surface while preventing overflow of the pan. A homogenous sub-sample was collected each time manure was removed. Samples for feed, orts and manure were stored at -20 °C until the end of each Exp. Procedures minimized volatilization of manure N compounds that may have occurred during storage 78      and thawing processes. At the end of each study, feed samples were composited by treatment. Orts and manure samples were composited by room. Saponin Sources The 3 saponin products used in the Exp were yucca saponin (YS) which is a powder made entirely from the stem of the Yucca schidigera plant and rich in steroidal saponin (contains 8.5% saponin; Desert King International, San Diego, CA, USA); quillaja saponin (QS), which is a triterpenic saponin enriched extract from pure Chilean soap bark tree Quillaja saponaria (contains 3.6% saponin; Desert King International, San Diego, CA, USA); and tea saponin (TS), which is the whole plant saponin extract from Camellia sinensis and rich in triterpenic saponin (contains 21.6% saponin; Ningbo Good Green Science & Technology, Ningbo, ZJ, China). Experimental Design and Dietary Treatments Both Exp 1 and 2 were randomized one-factorial designs with 2 treatments and Exp 3 was a repeated randomized one-factorial study with 3 treatments. All diets offered throughout the 3 studies were corn-corn silage based (Table 3.1). In all Exp, steers were fed 2 wk prior to entering rooms to allow for adaptation to the new diets. Prior to starting the 2nd period of Exp 3, all animals were re-inoculated once a week for 2 consecutive weeks with rumen fluid collected from 2 dry cows fed with hay diet. Steers were fed the corn-corn silage based control diet during the inoculation period and another 2 wk to eliminate any carryover effects from the first period. Then steers were assigned, randomly, to new treatment groups and acclimated to the new treatment diets for 2 wks. Steers were weighed on 2 consecutive mornings before arriving and after leaving the 79      rooms prior to feeding. Starting BW for the 3 Exp were 354 ± 10 kg (Exp 1), 429 ± 10 kg (Exp 2), 382 ± 16 kg (period 1, Exp 3) and 400 ± 12 kg (period 2, Exp 3). The experimental diets in Exp 1 were a corn and corn-silage based control diet (C1), a diet containing C1 + 1.5% QS of diet DM (QS1), and a third diet containing C1 + 0.64% YS of diet DM (YS1); 4 replicates per treatment. During Exp 2, a corn-silage based control diet (C2) was compared to a diet containing C2 + 0.25% TS of diet DM (TS2); providing 6 replicates per treatment. Inclusion levels of the 3 products were adjusted to similar saponin concentration (0.54 g/kg dietary DM of saponin) during Exp 1 and 2 in order to compare the effects of different saponins at the same dietary saponin concentration. During Exp 3, in addition to the corn-corn silage based control diet (C3), QS (QS3) and YS (YS3) were added to the diet at the maximum inclusion rate (1.5% of dietary DM for QS and YS) and 0.5% TS (TS3) was added because steers rejected feed at the higher inclusion levels (Li et al., unpublished pre-feeding study); there were 3 replicates of each treatment. Daily N intake was calculated as the difference between N offered in diet and N remaining in orts. Nitrogen loss (N loss) was defined as the sum of N mass from manure total Kjeldahl N (TKN), gaseous ammonia (NH3-N) and gaseous nitrous oxide (N2O-N). Nitrogen loss from NO and NO2 emissions was ignored because of their minor contribution to total N losses. Measurements of Gaseous Concentrations Twelve rooms (height = 2.14 m, width = 3.97 m, length = 2.59 m) were designed to continuously monitor incoming and exhaust concentrations of gases (Li et al., 2011). During Exp 1, 2 and 3, the average ventilation rate was 298.3, 295.5 and 289.5 L/s, 80      respectively. Concentrations of NH3 was measured using a chemiluminescence NH3 analyzer with a detection limit of 0.001 ppm (Model 17i, Thermo Fisher, Franklin, MA), which is a combination NH3 converter and NO-NO2-NOx analyzer. Hydrogen sulfide (H2S) was analyzed using pulsed fluorescence SO2-H2S analyzer with a detection limit of 0.003 ppm (TEI Model 450i, Franklin, MA; error = 1% of full-scale at 1 ppm). Concentrations of CH4 (range = 0 to 100 ppm; detection limit = 0.05 ppm) and NMTHC (range = 0 to 10 ppm; detection limit = 0.02 ppm) were determined by a back-flush gas chromatography system (TEI Model 55i, Franklin, MA). Concentration of N2O (range = 0 to 50000 ppm; detection limit = 0.03 ppm) was measured using an INNOVA 1412 photoacoustic analyzer (Lumasense Technologies, Ballerup, Denmark). Through software control (LabVIEW Version 8.2; National Instruments Corp., Austin, TX), gaseous concentration monitoring of each room occurred in a sequential manner. All emission factors were calculated from emission mass which is calculated based on the emission rate. Gas emission rates were calculated as the product of ventilation rates and concentration differences between exhaust and incoming air using the following equation: ER Q 273 T C C 10 MW V where ER is emission rate, g/min; Q is ventilation rate at room temperature and pressure, L/min; T is air temperature in room exhaust, in Kelvin; Co is gas concentration in room exhaust, mg/kg; Ci is gas concentration in the incoming air, mg/kg; MW is molecular weight of the gas, g/mol; Vm is molar volume of gas at standard condition (22.414 81      L/mole). Emissions in 1 full measurement cycle were estimated by multiplying the ER (g/min) with 195 min. Daily emissions were calculated as sum of the emissions in the 7 or 8 measurement cycles (Li et al., 2011). Chemical Analyses Feed and orts samples were analyzed by Dairy One Forage Testing Laboratory (Dairy One, Inc. Ithaca, NY) for compositional analysis. Feed DM content was determined with oven drying at 55 ºC until a constant weight of sample was obtained. Feed composition was analyzed by Near Infrared Reflectance Spectroscopy (NIRS) (AOAC-991.01, 1995). Crude protein, degradable protein, NDF and ADF were analyzed by Foss NIRS systems Model 6500 with Win ISI II v1.5 (AOAC-989.03, 1996). Minerals were analyzed by microwave digestion followed by Inductively Coupled Plasma Mass Spectrometry (ICP). Energy content was determined by an IKA C2000 + basic Calorimeter System (IKA Works, NC). Manure NH4 -N laboratory (AOAC-928.08, 2000) and total Kjeldahl N (TKN; FOSS Tecator, 1987) content was measured by distillation and digestion followed by distillation, respectively, in a Michigan State University laboratory. Statistical Analyses In all Exp, performance (DMI, N intake and ADG), excretion and air emissions data were analyzed using a mixed model testing the fixed effects of diet and random effects of steers. Day was considered as a repeated measure for DMI, period within Exp 3 was treated as a random effect (SAS Institute, 2008). Tukey’s test was applied to compare treatment versus control differences. Significant differences between treatment and control least squares means were declared at P < 0.05. 82      RESULTS Growth performance Feeding steers 1.5% QS (QS1) or 0.64% YS (YS1) resulted in no difference in DMI or ADG compared to steers fed C1 (Table 3.2). Across all treatments, in Exp 1, average DMI and ADG was 7.54 ± 0.09 kg and 1.16 ± 0.19 kg, respectively. The N intake of steers was not affected by dietary QS1 inclusion (140.66 g), but decreased as a result of YS1 inclusion (129.42 g; P < 0.01) compared to steers fed C1 treatment (134.26 g; Table 3.2). Similar results were observed in steers fed 1.5% QS treatment (QS3); DMI and ADG was not different between steers fed C3 and QS3 (Table 3.2).Increasing the dietary concentration of YS to 1.5% (YS3) did not change DMI or ADG compared to feeding C3. Feeding steers QS3 resulted in a lower N intake, while N intake was not affected by feeding YS3 (Table 3.2). Dry matter intake was increased by 5% when steers were fed diets containing 0.25% TS (TS2) compared to C2 (P < 0.01; Table 3.2). Accordingly, N intake was increased by 16% in TS2 treatment (P < 0.01). Although increased DMI and N intake was observed in steers fed TS2 treatment, the ADG of steers fed TS2 did not differ from those fed C2 (Table 3.2). Decreases were observed in DMI, N intake and ADG of steers as a result of feeding TS3, while no differences were observed among steers fed QS3, YS3, and C3. Steers fed TS3 had 27% less DMI and 80% less ADG compared to steers fed the C3 diet. The N intake was 80% lower in steer fed the TS3 treatment compared to cattle fed C3 as a result of reduced DMI. 83      Excretion mass and composition When examining saponins’ effects at the same dietary concentration (0.54 g/kg DM saponins), feeding steers QS1, YS1, or TS2 did not change manure characteristics compared to control diets (Exp 1 and 2; Table 3.3). Average daily fecal DM mass was + 2.96 ± 0.21 kg for Exp 1 and 2.97 ± 0.24 kg for Exp 2; NH4 -N was 43.98 ± 2.45 g for Exp 1 and 48.66 ± 4.10 g for Exp 2; and TKN was 79.46 ± 4.83 g for Exp 1 and 98.09 ± 9.52 g for Exp 2, respectively. Feeding steers with increased concentration of YS (1.5%; YS3) or 1.5% QS did not change manure excretion characteristics during Exp 3 (Table 3.3). However, daily manure DM mass excreted was reduced approximately by 27% when TS3 was fed (0.5%; P = 0.02) compared to C3. This effect is explained by the reduced DMI observed when + the TS3 treatment was fed. Manure NH4 -N concentration was increased when steers were fed TS3. However, because TS3 treatment produced less manure DM mass, daily + NH4 -N mass excreted from steers fed TS3 was not different from steers fed C3 diet (Table 3.3). Average total Kjeldahl N (TKN) remaining in manure was not affected by dietary saponin inclusion. Nitrogen emissions Yucca and quillaja saponins. Ammonia emissions from rooms where cattle were fed YS1, QS1, and C1 diets were not different (Table 3.4). Across all treatments, average daily NH3 emission, concentration, emission rate and daily mass were 1.10 mg/kg, 11.57 mg/min and 18.04 g/d, respectively. Ammonia emission factors calculated based on BW and DMI were not affected by dietary saponin inclusion. Average daily NH3 emission 84      factors, across all treatments were 49.12 mg/kg BW and 2.40 g/kg DMI. When daily NH3 emission mass was adjusted by N intake, an 18% reduction in NH3 daily emissions was observed for the QS1 treatment compared to C1 (107.70 mg/g N consumed in QS1 vs. 131.29 mg/g N consumed in C1; P = 0.08). Feeding steers the YS3 diet did not influence NH3 emissions compared to the C3 diet (Table 3.4). Contrary to Exp 1, manure from cattle fed QS3 had 32% higher daily NH3 emission mass than cattle fed the C3 diet. The NH3 emission factor calculated on a BW basis was 32% greater from steers fed QS3 treatment than C3 treatment, whereas no differences between these treatments were observed when daily emission mass was adjusted by DMI or N consumption. Tea saponin. Feeding TS2 reduced NH3 concentration and emission factors based on BW, DMI and N consumption without affecting manure N composition compared to steers fed C2. Daily NH3 emission mass from steers fed TS2 was reduced by19% compared to C2 (P = 0.06). The NH3 emissions adjusted by BW (P = 0.03), DMI (P < 0.01) and N consumption (P < 0.01) were 20 to 30% lower from steers where TS2 was fed compared to C2. Ammonia emissions were similar when TS3 and C3 were fed to steers. Dietary inclusion of saponins showed no treatment effects on steers’ N2O emissions from animal rooms regardless of concentration or saponin type. Average daily N2O emission mass from Exp 1, 2 and 3 was 3.64, 5.37 and 1.34 mg/d, respectively (Table 3.5). N balance 85      + Quillaja saponin. Feeding steers QS1 or QS3 did not change manure NH4 -N concentrations as a proportion of TKN, TKN excreted per day or N losses from N2O emissions as a fraction of N excreted compared to control treatments (Table 3.6). When expressed as a proportion of total N excreted, less N was volatized as NH3 when QS1 was fed compared to C1. On the contrary, N loss as NH3 emissions in QS3 treatment (12.20%) accounted for a greater proportion of total N excreted compared to C3 treatment (8.89%; Exp 3; P < 0.01). Yucca saponin. Feeding YS1 to steers did not change daily manure TKN mass (Table 3.3) + or NH4 -N concentration as a proportion of TKN (Table 3.6) but increased the percentage excreted N volatized as NH3 and the volatilization of NH3 as a fraction of N + intake compared to C1 (Table 3.6). Feeding YS3 reduced NH4 -N concentration as a fraction of TKN but did not affect total daily N excretion or N losses as a proportion of N intake compared to C3. However, excreted N remaining in manure as TKN was reduced and N emissions as N2O-N was increased in the YS3 treatment. Tea saponin. When TS2 was fed to steers, no differences in N balance were observed between TS2 and C2 treatments except that feeding TS2 resulted in a smaller proportion of N lost as NH3 relative to N intake. Increasing dietary TS concentration in TS3 treatment reduced manure TKN as a percentage of total N excreted by 2 percentage units, while 30% more N was lost as NH3 by feeding steers TS3 compared to C3. CH4 emissions 86      Feeding steers up to 1.5% of QS or YS failed to reduce CH4 emissions compared to control treatment (Table 3.7). When steers were fed TS2, daily CH4 emissions showed no differences compared to C2 (Table 3.7). However, when a diet containing higher TS concentration (TS3) was fed to steers, CH4 emission mass was reduced by 31% compared to C3. Because DMI was reduced by 27% in steers fed TS3 treatment, adjusting emissions to DMI basis produced no differences in emissions between TS3 and C3 fed steers. Steers fed TS3 treatment emitted 24% less CH4 per day per kg BW compared to steers fed C3 diet. H2S and NMTHC emissions Dietary saponin inclusion did not affect H2S (Table 3.8) or NMTHC (Table 3.9) emissions regardless of inclusion concentration. Average emission mass of H2S and NMTHC was 91.15 and 1.45 mg/d, respectively. DISCUSSION Several studies have reported that feeding ruminants low concentrations of saponin extracts from Quillaja saponaria or Yucca schidigera have not caused adverse effects on animal performance (Aregheore, 2005 (goats); Nasri et al., 2011 (lamb); Santoso et al., 2004 (sheep); Hristov et al., 1999 (heifers); Holtshausen et al., 2009 (dairy cows): Depenbusch et al., 2007 (steers)). Our findings agree with those studies when we fed lower concentrations of saponins in Exp 1 and 2. The intent of our study was to investigate the effects of saponins on air emissions by feeding the highest possible concentration to steers. Our results showed that steers can be fed as high as 1.5% of yucca (1.27 g/kg steroidal saponin from Yucca schidigera) or 87      quillaja (0.54 g/kg of triterpenic saponin from Quillaja saponaria) saponin without changes in growth performance. However, it should be noted that although both Exp 1 and 3 showed that feeding steers with 1.5% QS would not affect DMI, N intake in steers fed QS3 treatment was reduced compared to the control diet while feeding steers with QS1 did not affect N intake. This effect may be the result of the standard error differences between DMI and N observations. Feeding TS2 increased DMI and decreased N excretion suggesting a possible improvement in N efficiency as proposed by Francis et al., (2002). However, DMI was 27% less when TS3 was fed to steers compared to the C3 diet. Accordingly, an 80% reduction in ADG was also observed in steers fed the TS3 diet. A similar finding was reported by Hu et al. (2006), who found that feeding Boer goats diets containing 3 g/d DM TS improved DMI and ADG, while 6 g/d DM supplementation of TS in the diets reduced DMI. Tea saponin used in our study had a strong bitter taste and is very soluble in water, thus, the reduced DMI observed in TS3 treatment could be due to its palatability (Li et al., unpublished observations). Ruminal protozoa play an important role in fiber digestion. Reduction of protozoa population as a result of inhibition by saponins leads to impaired fiber digestion in rumen, resulting in reduced ADG (Ushida and Jouany, 1990; Guo et al., 2008). Protozoa may utilize lactic acid and contribute to the buffering capacity in the rumen, preventing an abrupt drop in rumen pH (Williams and Coleman, 1992). Inhibition of the protozoa population by TS may decreases rumen pH, impairing microbial digestion and limiting nutrients available to the animal (Grummer et al., 1983). Because we did not analyze ruminal pH and microbial communities in this study these observations were not verified. 88      Emission of NH3 from beef steers are estimated between 0.9 and 19.3 kg/head/yr (Todd et al., 2007). U.S. EPA estimated that the emission factor from dry lot-housed beef steers is 11.4 kg/head/year. Comparably, our results showed that daily NH3 emissions from steers weighing between 340 kg to 450 kg ranged from 4.68 kg/head/year to 11.16 kg/head/yr. Studies with dietary supplementation of yucca saponin have achieved a 20 to 50% reduction in NH3 emissions in poultry and swine (McCrory and Hobbs, 2001) due to its NH3 binding ability (Wallace et al., 1994). However, less information is available for ruminants. Hristov et al. (1999) demonstrated that supplementation of 0.2% and 0.6% of yucca saponin containing 4.4% of steroid saponins to heifers did not change N concentration in either urine or feces. The present study fed a higher concentration of YS + to steers, but neither manure NH4 -N concentration nor NH3 emissions were affected (YS1 and YS3). This indicated that yucca saponin was inefficient in binding NH3. The lack of effect of yucca saponin on NH3 emissions could be explained by the comprehensive microbial interactions and microbial adaptation to saponins that may have occurred during our study (Newbold et al., 1997; Holtshausen et al., 2009). Because protozoa play an important role in digesting microbial cell walls, if protozoa population was reduced in the rumen as a result of dietary saponin supplementation, the amount of undigested microbial protein escaping from the rumen would be greater, leading to poorer protein digestibility in the lower GI tract, and thus outweighing the benefit of reducing protozoa populations (Van Soest, 1994). 89      The effect of feeding 1.5% QS to steers on NH3 emissions was inconsistent between Exp; NH3 emissions from steers were not affected as a result of feeding QS1 but feeding QS3 increased NH3 emission mass from steers. Steers used in Exp 3 responded differently to QS inclusion; fed at the same QS concentration the QS had different effect on N metabolism. While the animal variation is difficult to explain the N balance data showed that, as a proportion of N excreted, less N was volatized as NH3 but more N was retained in the manure (TKN) in steers fed QS1 compared to C1, while more N was lost as NH3 and less N was retained as TKN when steers were fed QS3 compared to C3. However, because we did not analyze the microbial community from rumen and lower GI tract or the digestibility of feed, the mechanism behind the observations is unclear. Less dietary N was lost and NH3 emissions per unit of N intake were reduced when TS2 was fed to steers compared to C2 treatment, suggesting an improvement in N efficiency when TS2 was fed. Finding of Hu et al. (2006) support the explanation. Boer goats fed 3 g/d tea saponin had increased protein concentration in their blood, indicating more protein was absorbed by animals. When dietary TS concentration increased to 0.5%, manure DM mass was reduced in steers fed TS3 compared to C3, attributable to the reduced DMI. Steers fed TS3 had lower DMI and N intake while daily NH3 emissions from steers were not different from steers fed C3 (Table 3.5). The N balance showed that compared to steers fed C3, a greater proportion of N was emitted as NH3 when steers + were fed TS3 and more N in the manure was in the form of NH4 -N (Table 3.6). In our 90      study, it is very likely that N metabolism was impaired when TS3 was fed, because ruminal microbial digestion declined due to lack of N and energy to support body functions, leading to greater N loss (Van Soest, 1994). Nitrous oxide emissions were not affected by dietary saponin inclusion, but N2O emissions when expressed as the proportion of N excreted (Table 3.6) during Exp 3 was increased in TS3 and YS3. Nitrous oxide is the intermediate product of nitrification from + nitrate or denitrification from NH3/NH4 . Studies show a positive correlation between + N2O emission and NH4 availability (Fukumoto et al., 2003; Heller et al., 2010). The + greater concentration of NH4 -N in the manure as a result of feeding steers with TS3 could possibly explain the increased proportion of N2O emissions compared to steers fed C3 diet. However, it is difficult to explain why steers fed YS3 had greater proportion of N2O emitted while less TKN was retained in the manure. We observed that 77% of N ingested was excreted. Twelve percent of total N intake was emitted as NH3, 51% of consumed N was retained in manure as TKN, 53% of + TKN was in the form of NH4 -N. Nitrogen lost as N2O accounted for less than 2% of total N excreted. Even if all N that was not lost was considered as retained by the animal, retained N only accounted for 23% of the total N intake. Ruminants are less efficient than monogastric animals such as swine and poultry regarding to utilize dietary CP. In beef cattle, only about 20% of total N ingested can be retained, with the rest 80% excreted in urine and feces (Farran et al., 2006; Cole and Todd, 2009). The low efficiency of N utilization is confirmed in our study. Our results are in agreement with other reports 91      (Todd et al., 2006; Todd et al., 2007; Cole and Todd 2009). However, in the current study, the N losses due to NH3 emissions were smaller (12%) and a greater portion of N was + retained in manure as NH4 -N (33%) compared to other studies where 40 to 60% of dietary N was emitted as NH3 (Todd et al., 2006; Todd et al., 2007; Cole and Todd 2009). The difference may be explained by the duration of the current study (22 d) with continuous emissions throughout the duration contributing to the mass balance. Shorter term studies may present results favoring a greater portion of excreted N emitted as NH3 due to the rapid conversion of urea N to NH3 whereas longer studies consider the mineralization of organic-bound N to inorganic N. Saponins inhibit methanogenesis in vitro (Takahashi et al., 2000; Hu et al., 2005; Holtshausen et al., 2009). Because approximately 20% of methanogens are associated with protozoa which also play an important role in inter-species hydrogen transfer for methanogenesis (Tokura et al., 1997), reduced CH4 emissions are thought to be the consequence of saponin’s toxicities towards protozoa population (Guo et al., 2008; Holtshausen et al., 2009). However, results from in vivo studies have not always been consistent, with reduced CH4 emissions observed in sheep species, only (Patra and Saxena, 2009; review). Our results indicate that dietary saponin supplementation failed to reduce CH4 emissions, except when fed at concentrations that inhibited performance. The effects of saponins are suggested to be non-permanent because of microbial adaptation or degradation of saponins in the rumen (Newbold et al., 1997; Teferedegne et al., 1999; Ivan et al., 2004). Because steers were acclimated to diets for 2 wks prior to the start of 92      the air emissions measures ruminal microbes may have already adapted to the presence of dietary saponins. In studies where 0.27% of TS (3 g/d, average DMI = 750 to 900 g) was fed to sheep and lambs, CH4 production was reduced 10% and 28%, respectively (Mao et al., 2010; Zhou et al., 2011). Feeding 0.25% TS did not change CH4 emissions in the present study, possibly due to species variation in rumen microbe populations or feeding different forage:concentrate ratios. Sheep or lambs are often fed diets containing > 60% forage (Yuan et al., 2007; Mao et al., 2010; Zhou et al., 2011), while steer and dairy cow diets contain 37 to 40% forage. Despite that, lack of CH4 emission response to saponin inclusions have occurred when steers and dairy cows were fed high forage diets (Zinn et al., 1998). Species differences between large ruminants and small ruminants should be considered the primary reason for the different results observed, not dietary forage concentration. Plant maturity, geographical region of production (Ndamba et al., 1993), and efficiency of extraction methods (Vongsangnak et al., 2004) all affect the concentration of saponins in extracts. It is impossible to compare results among studies unless actual dietary saponin concentrations or a measure of activity are provided. Results from pure-culture studies show that some rumen microbes capable of utilizing saponin can produce small amount of H2S (Gutierrex et al., 1959). Feeding saponin to steers may produce a small increase in H2S production. However, throughout the study, H2S emissions were not affected by dietary saponin inclusion perhaps because bacterial H2S production was too low for differences to be detected. Emissions of H2S 93      are considered to be correlated to the dietary S concentration (Li et al., 2011). In our study, S concentration was not different among treatments within an Exp (data not shown). Our study suggests that, in large ruminants, high dietary saponin concentrations are necessary in order to achieve a significant inhibitive effect on CH4 emissions. However, palatability of saponin may affect intake or microbial digestion thereby impairing growth performance. CONCLUSION Saponins failed to reduce CH4 emissions without affecting animal performance in. Although the response to saponin concentration in terms of CH4 production is dose dependent, higher dietary inclusion level may pose a challenge to animal’s performance as observed in steers fed TS3 treatment. Ammonia emissions adjusted for N intake were not affected by either yucca or quillaja saponin saponins however N balance and form in which N losses occurred was impacted. ACKNOWLEDGEMENTS Funding for this work was provided through the National Research Initiative Air Quality Program of the Cooperative State Research, Education, and Extension Service, USDA, under Agreement No. 2005-35112-15356. The authors wish to acknowledge the USDA for their support of this work. The authors would like to thank Jolene Roth, Andrew Fogiel and Mark Schilling for lab and instrumental maintenance and Desert King International (San Diego, CA) for product donation and composition analysis of all saponin products. 94      Table 3.1. Diet and nutrient composition from experimental diets (DM basis) Exp 1 C1 * YS1 Exp 2 QS1 C2 Exp 3 TS2 C3 YS3 QS3 TS3 Ingredients (%) DM High moisture corn Corn silage Soybean meal # Supplement 50 Total Saponins yucca saponin quillaja saponin 46 46 3 5 100 --- -1.5 --- --- --- 1.5 -- -1.5 --- -- tea saponin 0.64 --- -- -- 0.25 -- -- -- 0.5 Supplement 50, % of DM TM¶ 1.4 24.9 48.3 0.3 9.6 0.2 9.6 5.1 0.7 100 Akey TM premix # 4 Limestone Soybean meal, 48% N TM Rumensin 80 TM salt Vitamin E, 5% Urea, 45% N Potassium chloride Selenium 90 Total Analyzed composition, % DM DM CP ADF NDF P ME (Mcal/kg) * 48.9 11. 0 14.1 26.1 0.27 1.83 48.5 11.3 14.5 27.0 0.27 1.80 47.9 10.8 14.3 27.4 0.28 1.85 50.3 12.5 16.7 25.6 0.31 1.78 49.6 12.9 17.0 26.1 0.29 1.78 45.7 45.2 46.2 10.0 10.4 10.5 12.3 11.85 12.1 22.7 22.00 21.6 0.28 0.27 0.28 1.76 1.75 1.78 45.9 9.9 12.2 22.4 0.28 1.75 Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. C1 = Control diet in Exp 1. 95      Table 3.1. Diet and nutrient composition from experimental diets (DM basis), con’t # ¶ Middle section of the table lists the ingredients for BFS50 supplement for all treatments Akey TM premix # 4 composition: 9% Mg, 4% S, 0.02% Co, 1% Cu, 0.09% I, 2% Fe, 4% Mn, 0.03% Se, 4% Zn, 4,400,000 IU vitamin A, 550,000 IU vitamin D, and 5,500 IU vitamin E/kg (Akey Inc., Lewisburg, OH). 96      Table 3.2. Growth performance from Holstein steers fed corn-corn silage based diets with and ¶ without saponin addition Diets BW, kg Starting DMI, kg N intake, g ADG, kg Ending Exp 1 * 363 C1 QS1 352 YS1 SEM 346 10 Diet 0.48 C2 428 TS2 SEM 429 Diet 0.97 C3 Period 1 (kg) Start End 390 401 10 382 7.62 379 7.51 0.90 b 1.27 129.42 1.58 a 1.31 0.19 < 0.01 0.31 140.66 374 7.49 10 0.09 Source of variation 0.84 0.52 Exp 2 a 453 8.53 b 454 8.94 10 0.09 Source of variation 0.96 0.01 Exp 3 Period 2 (kg) Start End b 398 448 8.71 b QS3 366 373 401 446 8.67 TS3 389 392 401 410 6.36 YS3 SEM * b 134.26 384 16 394 16 Diet 0.72 0.64 b 400 435 8.58 12 13 0.42 Source of variation 1.00 0.24 < 0.01 a 159.72 a 1.18 184.56 1.85 b 1.18 < 0.01 1.00 0.21 c b 132.04 1.53 119.68 1.27 b a b a 96.98 0.31 134.51 2.89 1.11 0.54 < 0.01 0.04 c b Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. C1 = Control diet in Exp 1. Exp 1, 1.5% quillaja and 0.64% of yucca saponin; Exp 2, 0.25% tea saponin; Exp 3, 1.5% yucca and quillaja saponin, 0.5% of tea saponin. ¶ Exp 1, n=12, 4 replicates per treatment; Exp 2, n=12, 6 replicates per treatment; Exp 3, n=12, 3 replicates per treatment per period. a, b, c Significant differences observed at the P < 0.05 probability level. 97      Table 3.3. Daily manure excretion from Holstein steers fed corn-corn silage based diets with and ¶ without saponin addition Diets Wet, kg # Total Daily C1* QS1 YS1 SEM 297.72 304.31 306.04 26.80 13.53 13.83 13.91 1.21 Diet 0.97 0.97 C2 TS2 SEM 354.59 329.34 23.46 16.12 14.97 1.16 Diet 0.48 0.48 C3 330.07 15.00 QS3 307.94 13.99 TS3 256.42 11.66 YS3 SEM * + NH4 -N, daily DM, daily 296.39 40.29 13.47 1.83 Diet 0.25 0.25 % kg % DM § TKN , daily g % DM g 1.65 1.47 1.40 0.10 46.62 43.44 41.88 2.45 2.78 2.68 2.62 0.07 78.59 79.67 80.12 4.83 0.22 0.41 0.32 0.97 1.58 1.73 0.11 49.38 47.94 4.10 3.22 3.39 0.15 100.95 95.22 9.52 0.36 0.80 0.46 0.66 b 39.48 3.31 b 91.96 37.58 3.10 a 94.62 35.84 3.41 b b 21.70 2.85 1.34 1.13 0.32 0.02 Source of variation 0.23 0.02 0.01 74.98 38.00 2.49 3.28 0.14 91.22 7.60 0.78 0.27 0.04 Exp 1 21.05 2.83 21.68 2.99 21.15 3.06 0.67 0.21 Source of variation 0.53 0.76 Exp 2 19.29 3.11 18.83 2.82 1.00 0.24 Source of variation 0.59 0.39 Exp 3 b 20.52 3.00 b 21.77 3.00 18.77 a 2.18 1.33 1.28 1.66 b b a b Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. C1 = Control diet in Exp 1. Exp 1, 1.5% quillaja and 0.64% of yucca saponin; Exp 2, 0.25% tea saponin; Exp 3, 1.5% yucca and quillaja saponin, 0.5% of tea saponin. ¶ Exp 1, n=12, 4 replicates per treatment; Exp 2, n=12, 6 replicates per treatment; Exp 3, n=12, 3 replicates per treatment per period. § TKN = total Kjeldahl N. # Duration was 22 d in Exp1 and 2; each period in Exp 3 was 22 d. Therefore, daily excretion was calculated based on a 22 d average of total manure excreted. a, b, c Significant differences observed at the P < 0.05 probability level. 98      Table 3.4. Least squares means from NH3 emissions from Holstein steers fed corn-corn silage based diets with and without saponin addition ¶ Daily Daily Daily Diets concentration emission rate emission mass mg/kg mg/kg mg/min g/d BW Exp 1 * 1.10 11.41 17.80 48.92 C1 QS1 YS1 SEM 0.95 1.25 0.11 Diet 0.15 C2 2.11 TS2 SEM 1.80 0.11 Diet 0.04 C3 0.96a QS3 b a b 1.19 9.61 13.68 1.47 14.99 40.26 21.34 58.17 2.30 6.27 Source of variation 0.17 0.15 0.13 Exp 2 b 19.61 30.59 69.31 15.90 55.70 1.37 2.14 4.49 Source of variation 0.06 0.06 0.03 Exp 3 a 2.31 131.29 2.02 2.86 0.32 107.70 165.69 18.32 0.17 0.08 b 3.59 a b 192.03 a 2.78 0.20 134.63 10.14 < 0.01 < 0.01 35.15 a 1.67 114.42 12.33 19.23 46.58 b 2.20 158.80 32.26 a 2.07 144.21 15.56 38.47 1.97 3.52 Source of variation 0.02 0.02 ab 1.82 122.54 0.20 36.57 0.14 0.26 0.85 YS3 SEM 0.99 a 0.11 9.98 1.26 Diet 0.03 0.02 * mg/g N consumed 14.60 b a 8.07 ab a g/kg DMI 9.36 a TS3 a 24.80 Emission factors b a 12.58 a Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. C1 = Control diet in Exp 1. Exp 1, 1.5% quillaja and 0.64% of yucca saponin; Exp 2, 0.25% tea saponin; Exp 3, 1.5% yucca and quillaja saponin, 0.5% of tea saponin. ¶ Exp 1, n=12, 4 replicates per treatment; Exp 2, n=12, 6 replicates per treatment; Exp 3, n=12, 3 replicates per treatment per period. a, b, c Significant differences observed at the P < 0.05 probability level. 99     Table 3.5. Least squares means from N2O emissions from Holstein steers fed corn-corn ¶ silage based diets with and without saponin addition Diets * Emission factors Daily Daily Daily concentration emission rate emission mass mg/g N mg/kg BW g/kg DMI mg/kg mg/min g/d consumed Exp 1 C1 QS1 YS1 SEM 0.51 0.49 0.50 0.04 Diet 0.95 C2 TS2 SEM 0.50 0.50 0.03 Diet 0.94 C3 QS3 TS3 YS3 SEM 0.68 0.68 0.68 0.68 0.17 Diet 0.99 * 2.28 3.56 2.39 3.73 2.33 3.64 0.19 0.29 Source of variation 0.92 0.92 Exp 2 3.41 5.33 3.47 5.41 0.64 1.00 Source of variation 0.95 0.95 Exp 3 0.84 1.31 0.88 1.37 0.78 1.22 0.93 1.45 0.19 0.30 Source of variation 0.59 0.59 9.87 10.04 10.05 0.84 460.96 487.54 494.90 47.71 26.19 26.04 28.64 2.70 0.99 0.87 0.74 11.93 12.16 2.23 603.74 574.36 106.87 32.25 27.83 5.50 0.94 0.85 0.57 3.15 3.32 3.11 3.57 0.61 145.91 160.29 205.53 175.08 21.08 9.77 11.79 12.68 17.71 1.40 0.68 0.29 0.45 Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. C1 = Control diet in Exp 1. Exp 1, 1.5% quillaja and 0.64% of yucca saponin; Exp 2, 0.25% tea saponin; Exp 3, 1.5% yucca and quillaja saponin, 0.5% of tea saponin. ¶ Exp 1, n=12, 4 replicates per treatment; Exp 2, n=12, 6 replicates per treatment; Exp 3, n=12, 3 replicates per treatment per period. a, b, c Significant differences observed at the P < 0.05 probability level. 100     ¶ Table 3.6. Nitrogen balance from Holstein steers fed corn-corn silage based diets with and without saponin addition Diet * N intake g/d § + N excreted, g/d TKN NH3-N C1 QS1 134.26 b 78.59 140.66 b 14.81 79.67 YS1 SEM 129.42 1.58 a Diet < 0.01 C2 159.72 TS2 SEM 184.56 1.85 Diet < 0.01 C3 132.04 93.16 12.00 QS3 119.68 90.82 TS3 96.98 YS3 SEM 134.51 2.89 Diets 0.04 N2O-N b 2.95 12.34 a 3.06 80.12 4.83 17.33 0.54 b 2.99 0.22 0.97 0.11 0.91 a 100.95 25.14 4.37 b 95.22 9.52 20.89 1.99 4.34 0.81 0.66 0.13 0.98 c b a c b NH4 -N % of TKN Exp 1 59.46 54.53 TKN a 3.04 a 84.40 b 2.16 3.24 c 86.63 9.85 a 2.14 3.01 0.24 82.20 0.74 14.01 0.49 c 2.36 0.18 < 0.01 0.76 < 0.01 < 0.01 0.64 b 2.65 77.20 12.43 3.10 a 2.21 0.44 77.93 2.89 10.14 1.21 3.49 0.58 < 0.01 0.48 0.86 0.18 0.64 56.89 8.86 a 13.46 0.42 15.79 11.45 1.00 ab b a 0.96 75.46 9.41 a 0.72 8.89 1.05 78.84 13.05 0.90 12.20 10.54 0.93 0.97 14.97 1.15 12.79 0.46 1.13 0.13 a 86.62 99.07 1.50 b 54.83 b 84.69 75.13 a 53.51 b 88.28 15.81 c 52.74 10.07 0.62 0.89 0.11 86.93 0.59 9.56 0.63 1.14 0.14 < 0.01 < 0.01 0.20 < 0.01 0.11 < 0.01 < 0.01 0.02 a b b c a ab b 86.98 a 50.47 72.85 2.37 4.55 Source of variation < 0.01 < 0.01 12.26 c 11.70 b b a N2O-N † N losses, % of N excreted TKN NH3-N N2O-N b 10.92 51.18 48.99 3.37 7.22 Source of variation 0.77 0.12 Exp 3 b NH3-N 58.79 b 53.01 61.90 2.30 0.91 Source of variation 0.32 0.01 Exp 2 49.69 64.95 101   N losses, % of N intake a b b a b c a a 0.87 ab 0.97 b bc   ¶ Table 3.6. Nitrogen balance from Holstein steers fed corn-corn silage based diets with and without saponin addition * , con’t Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. C1 = Control diet in Exp 1. Exp 1, 1.5% quillaja and 0.64% of yucca saponin; Exp 2, 0.25% tea saponin; Exp 3, 1.5% yucca and quillaja saponin, 0.5% of tea saponin. ¶ Exp 1, n=12, 4 replicates per treatment; Exp 2, n=12, 6 replicates per treatment; Exp 3, n=12, 3 replicates per treatment per period. § N intake = dietary N offered – N in orts. N excreted = TKN + NH3-N + N2O-N; TKN, total Kjeldahl N remained in manure; NH3-N, N emitted as gaseous NH3; N2O-N, N emitted as gaseous N2O. The difference between N intake (g) and N excreted (g) is N retained by the growing steer. † Percentages reflected the estimated least squares means of every form of excreted N accounted; sum of the TKN, NH3-N and N2O-N did not equal to 100% due to the contributions of within treatment errors. a, b, c Significant differences observed at the P < 0.05 probability level. 102     Table 3.7. Least squares means from CH4 emissions from Holstein steers fed corn-corn ¶ silage based diets with and without saponin addition Diets * Daily concentration mg/kg Daily emission rate mg/min Emission factors Daily emission mass g/d mg/kg BW g/kg DMI Exp 1 C1 QS1 YS1 SEM 8.88 71.96 112.26 310.54 14.71 8.08 7.52 0.55 264.32 239.49 31.30 13.17 11.63 1.36 Diet 0.22 0.27 0.28 C2 TS2 SEM 9.12 9.07 0.41 211.78 215.96 14.71 11.15 10.66 0.89 Diet 0.94 0.84 0.70 C3 9.77 63.37 98.86 55.85 87.13 6.93 10.81 Source of variation 0.26 0.26 Exp 2 60.32 94.10 61.26 95.57 4.80 7.49 Source of variation 0.89 0.89 Exp 3 b 67.76 103.50 b 73.55 112.31 b b QS3 10.28 TS3 8.05 YS3 SEM 9.67 0.58 Diets <0.01 * a b a 256.97 b 13.03 281.21 b 13.77 73.23 196.18 a 12.96 105.67 9.83 9.68 Source of variation 0.07 0.03 267.23 26.28 b 13.08 0.05 0.90 49.31 b 69.45 0.83 Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. C1 = Control diet in Exp 1. Exp 1, 1.5% quillaja and 0.64% of yucca saponin; Exp 2, 0.25% tea saponin; Exp 3, 1.5% yucca and quillaja saponin, 0.5% of tea saponin. ¶ Exp 1, n=12, 4 replicates per treatment; Exp 2, n=12, 6 replicates per treatment; Exp 3, n=12, 3 replicates per treatment per period. a, b, c Significant differences observed at the P < 0.05 probability level. 103     Table 3.8. Least squares means from H2S emissions from Holstein steers fed corn-corn silage based diets with and without saponin addition Diets * Daily concentration mg/kg Daily emission rate mg/min ¶ Emission factors Daily emission mass mg/kg mg/d mg/kg BW DMI Exp 1 C1 QS1 YS1 SEM 0.005 0.077 119.53 0.33 15.03 0.004 0.004 0.001 0.17 0.25 0.10 8.37 12.04 4.22 Diet 0.42 0.50 0.53 C2 TS2 SEM 0.008 0.008 0.001 0.29 0.29 0.04 14.11 14.29 1.82 Diet 0.93 0.94 0.95 C3 QS3 TS3 YS3 SEM 0.005 0.005 0.005 0.004 0.0004 0.19 0.19 0.22 0.13 0.06 8.64 8.60 13.32 5.79 2.60 Diets 0.77 0.040 61.75 0.059 91.64 0.023 35.78 Source of variation 0.52 0.52 Exp 2 0.080 125.98 0.083 129.64 0.010 19.56 Source of variation 0.90 0.90 Exp 3 0.051 78.83 0.049 75.69 0.055 86.19 0.033 51.09 0.014 21.73 Source of variation 0.57 0.57 0.56 0.11 * Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. C1 = Control diet in Exp 1. Exp 1, 1.5% quillaja and 0.64% of yucca saponin; Exp 2, 0.25% tea saponin; Exp 3, 1.5% yucca and quillaja saponin, 0.5% of tea saponin. ¶ Exp 1, n=12, 4 replicates per treatment; Exp 2, n=12, 6 replicates per treatment; Exp 3, n=12, 3 replicates per treatment per period. 104     Table 3.9. Least squares means from non-methane total hydrocarbon (NMTHC) emissions from Holstein steers fed corn-corn silage based diets with and without saponin ¶ addition Diets * Daily concentration mg/kg Emission factors Daily emission mass g/d mg/kg BW g/kg DMI Exp 1 C1 QS1 YS1 SEM 0.07 0.07 0.07 0.003 Diet 0.21 C2 TS2 SEM 0.05 0.05 0 Diet 0.99 C3 QS3 TS3 YS3 SEM 0.06 0.06 0.06 0.06 0.01 Diets 0.15 * Daily emission rate mg/min 0.95 1.48 0.88 1.38 1.00 1.56 0.05 0.07 Source of variation 0.23 0.23 Exp 2 0.82 1.28 0.83 1.29 0.07 0.11 Source of variation 0.95 0.95 Exp 3 1.01 1.58 1.01 1.57 0.87 1.35 0.98 1.52 0.26 0.41 Source of variation 0.35 0.35 4.09 3.70 4.25 0.23 191.07 181.04 206.50 9.84 0.23 0.18 2.92 2.92 0.29 149.75 144.27 9.89 0.99 0.69 3.80 3.82 3.43 3.76 0.85 178.11 181.53 220.76 179.71 10.23 0.58 0.08 Treatments were corn and corn silage based Control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Numbers following treatment identify the Exp. eg. C1 = Control diet in Exp 1. Exp 1, 1.5% quillaja and 0.64% of yucca saponin; Exp 2, 0.25% tea saponin; Exp 3, 1.5% yucca and quillaja saponin, 0.5% of tea saponin. ¶ Exp 1, n=12, 4 replicates per treatment; Exp 2, n=12, 6 replicates per treatment; Exp 3, n=12, 3 replicates per treatment per period. a, b, c Significant differences observed at the P < 0.05 probability level. 105       LITERATURE CITED 106          LITERATURE CITED Aregheore, E. M. 2005. Effect of yucca schidigera saponin on the nutritive value of ureaammoniated maize stover and its feeding value when supplemented with forage legume (Calliandra calothyrsus) for goats. Small Rum. Res. 56:95–102. AOAC-928.08. 2000. Official methods of analysis of the association of official analytical chemists, 16th ed., AOAC, Int., Arlington, VA. AOAC-989.03. 1996. Official methods of analysis of the association of official analytical chemists, 16th ed. AOAC, Int., Arlington, VA. AOAC-991.01. 1995. Official methods of analysis of the association of official analytical chemists, 16th ed. AOAC, Int., Arlington, VA. Beauchemin, K. A., S. M. McGinn, C. Benchaar, and L. Holtshausen. 2009. Crushed sunflower, flax, or canola seeds in lactating dairy cow diets: Effects on methane production, rumen fermentation, and milk production. J. Dairy Sci. 92(5):21182127. Chung, Y. H., N. D. Walker, S. M. McGinn, and K. A. Beauchemin. 2011. Differing effects of 2 active dried yeast (Saccharomyces cerevisiae) strains on ruminal acidosis and methane production in nonlactating dairy cows. J. Dairy Sci. 94(5):2431-2439. Cole, N. A. and R. W. Todd. 2009. Nitrogen and phosphorus balance of beef cattle feedyards. Proceedings of the Texas Animal Manure Management Issues Conference Round. Rock, TX. September 29-30. Depenbusch, B. E., J. S. Drouillard, E. R. Loe, M. A. Greenquist, and M. E. Corrigan. 2007. Growth performance and carcass characteristics of finishing steers given an oral dose and daily supplement of yucca-derived sarsaponin. Prof. Anim. Sci. 23:681-685. Falkengren-Grerup, U. 1986. Soil acidification and vegetation changes in deciduous forest in southern Sweden. Oecologia 70:339-347. Farran, T. B., G. E. Erickson, T. J. Klopfenstein, C. N. Macken, and R. U. Lindquist. 2006. Wet corn gluten feed and alfalfa hay levels in dry-rolled corn finishing diets: effects on finishing performance and feedlot nitrogen mass balance. J. Anim. Sci. 84:1205–1214. FOSS Tecator. Application Note AN 300. 1987. The Determination of Nitrogen according to Kjeldahl using Block Digestion and Steam Distillation. 107          Francis, G., H. P. S. Makkar, and K. Becker. 2002. Dietary supplementation with a Quillaja saponin mixture improves growth performance and metabolic efficiency in common carp (Cyprinus carpio L). Aquaculture. 203:311–320. Fukumoto, Y., T. Osada, D. Hanajima, and K. Haga. 2003. Patterns and quantities of NH3, N2O and CH4 emissions during swine manure composting without forced aeration—Effect of compost pile scale. Bioresour. Technol. 89:109–114. De Wit, J., P. T. Westra, and A.J. Nell. 1996. Environmental impact assessment of landless livestock ruminant production systems. FAO International Agriculture Center Wageningen,The Netherlands. Gutierrex, J., R. E. Davis, and I. L. Lindahl. 1959. Characteristics of saponin-utilizing bacteria from the rumen of cattle. Appl. Microbiol., 7: 304-308. Grummer, R. R., C. R. Staples and C. L. Davis. 1983. Effect of defaunation on ruminal volatile fatty acids and pH of steers fed a diet high in dried whole whey. J. Dairy Sci. 66:1738. Guo, Y., J. Liu, Y. Lu, W. Zhu, S. E. Denman and C. S. McSweeney. 2008. Effect of tea saponin on methanogenesis, microbial community structure and expression of mcrA gene, in cultures of rumen micro-organisms. Lett. Appl. Microbiol. 47:421426. Heller, H., A. Bar-Tal, G. Tamir, P. Bloom, R. T. Venterea, D. Chen, Y. Zhang,, C. E. Clapp and P. Fine. 2010. Effects of manure and cultivation on carbon dioxide and nitrous oxide emissions from a corn field under Mediterranean conditions. J. Environ. Qual. 39: 437-448. Hess, H. D., M. Kreuzer, T. E. Díaz, C. E. Lascano, J. E. Carulla, C. R. Soliva and A. Machmüller. 2003. Saponin rich tropical fruits affect fermentation and methanogenesis in faunated and defaunated rumen fluid. Anim. Feed Sci. Technol. 109:79–94. Holtshausen, L., A. V. Chaves, K. A. Beauchemin, S. M. McGinn, T. A. McAllister, N.E. Odongo, P. R. Cheeke, and C. Benchaar 2009. Feeding saponin-containing yucca schidigera and quillaja saponaria to decrease enteric methane production in dairy cows. J. Dairy Sci. 92:2809-2821. Hristov, A. N., T. A. McAllister, F. H. Van Herk, K. J. Cheng, C. J. Newbold and P. R. Cheeke. 1999. Effect of yucca schidigera on ruminal fermentation and nutrient digestion in heifers. J. Anim. Sci. 77:2554–2563 Hristov, A. N., S. Zaman, M. Vander Pol, M. Ndegwa, L. Campbell and S. Silva. 2009. Nitrogen losses from dairy manure estimated through nitrogen mass balance and chemical markers. J. Environ. Qual. 38:2438–2448. 108          Hu, W. 2005. Effect of saponin on ruminal fermentation and methanogenesis in vitro. Ph. D. thesis: Effects of saponin on methanogenesis, ruminal fermentation and animal growth performances. Zhejiang University, Hangzhou, ZJ, China. Hu, W, J. Liu, J. Ye., Y. Wu, and Y. Guo. 2005. Effect of tea saponin on rumen fermentation in vitro. Anim. Feed Sci. Technol. 120(3-4):333-339. Hu, W, J. Liu, Y. Wu, Y. Guo, and J. Ye. 2006. Effects of tea saponins on in vitro ruminal fermentation and growth performance in growing Boer goat. Arch. Anim Nutr. 60(1):89-97. IPCC. 2001. Climate change 2001: the scientific basis. Contribution of working group I to the third assessment report of the intergovernmental panel on climate change. Cambridge University Press, Cambridge, UK. Ivan, M., K. M. Koenig, B. Teferedenge, C. J. Newbold, T. Entz, L. M.Rode and M. Ibrahim. 2004. Effect of the dietary Enterolobium cyclocarpum foliage on the population dynamics of rumen ciliate protozoa in sheep. Small Rumin. Res. 52:81–91. Li, W., W. J. Powers, D. Karcher, C. R. Angel and T. J. Applegate. 2010. Effect of DDGS and mineral sources on air emissions from laying hens. Poult. Sci. Vol. 89, E- Suppl. 1.Abstract 544. Li, W., W. J. Powers and G. M. Hill. 2011. Feeding distillers dried grains with solubles and organic trace mineral sources to swine and the resulting effect on gaseous emissions. J. Anim. Sci. (accepted April 25, 2011; manuscript number JAS3611). Lipfert, F. W. 1994. Air pollution and community health: a critical review and data sourcebook. Van Nonstrand Reinhold, New York, NY. Mao, H., J. Wang, Y. Zhou and J. Liu. 2010. Effects of addition of tea saponins and soybean oil on methane production, fermentation and microbial population in the rumen of growing lambs. Livest. Sci. 129 (1-3):56-62. McCrory, D. F. and P. J. Hobbs. 2001. Additives to reduce ammonia and odor emissions from livestock wastes: a review. J. Environ. Qual. 30:345-355. Moreira, V. R. and L. D. Satter. 2006. Effect of scraping frequency in a freestall barn on volatile nitrogen loss from dairy manure. J. Dairy Sci. 89:2579–2587. Muck, R. E. and B. K. Richards. 1983. Losses of manurial nitrogen in free-stall barns. Agric. Wastes. 7(2):65-79. Nasri, S., H. B. Salem, V. Vasta, S. Abidi, H. P. S. Makkar, and A. Priolo. 2011. Effect of increasing levels of quillaja saponaria on digestion, growth and meat quality of Barbarine lamb. Anim. Feed Sci. Technol. 164:71-78. 109          Ndamba, J., E. Lemmich, and P. Mølgaard. 1993. Investigation of the diurnal, ontogenetic and seasonal variation in the molluscicidal saponin content of Phytolacca dodecandra aqueous berry extracts. Phytochemistry. 35(1):95-99. Newbold, C. J., S. M. E. Hassan, J. Wang, M. E. Ortega, and R. J. Wallace. 1997. Influence of foliage from African multipurpose trees on activity of rumen protozoa and bacteria. Brit. J. Nutr. 78:237–249. NRC. 2003. Air emissions from animal feeding operations: current knowledge, future needs. The National Academies Press, Washington, DC. Patra, A. K. and J. Saxena. 2009. The effect and mode of action of saponins on the microbial populations and fermentation in the rumen and ruminant production. Nutr. Res. Rev. 22:204-219. Pen, B., C. Sar, B. Mwenya, K. Kuwaki, R. Morikawa, and J. Takahashi. 2006. Effects of Yucca schidigera and Quillaja saponaria extracts on in vitro ruminal fermentation and methane emission. Anim. Feed Sci. Technol. 129:175–186. Pope III, C. A. and D. W. Dockery. 2006. Health effects of fine particulate air pollution: lines that connect. J. Air & Waste Manage. Assoc. 56:709 –742. Santoso, B., B. Mwenya, C. Sar, Y. Gamo, T. Kobayashi , R. Morikawa, K. Kimura, H. Mizukoshi, and J. Takahashi. 2004. Effects of supplementing galactooligosaccharides, yucca schidigera or nisin on rumen methanogenesis, nitrogen and energy metabolism in sheep. Livest. Prod. Sci. 91:209–217. SAS Institute. 2008. Version 9.2. SAS Institute, Inc., Cary, NC, USA. Petersen, S. O., S. G. Sommer, O. Aaes, and K. Soegaard. 1998. Ammonia losses from urine and dung of grazing cattle: Effect of N intake. Atmos. Environ. 32:295–300. Takahashi, J., Y. Miyagawa, Y. Kojima, and K. Umetsu. 2000. Effects of Yucca schidigera extract, probiotics, monensin and L-cysteine on rumen methanogenesis. Asian-Aust. J. Anim. Sci. 13:499-501. Teferedegne, B., F. McIntosh, P. O. Osuji, A. Odenyo, R. J. Wallace, and C. J. Newbold. 1999. Influence of foliage from different accessions of the sub-tropical leguminous tree, Sesbania sesban, on ruminal protozoa in Ethiopia and Scottish sheep. Anim. Feed Sci. Technol. 78:11–20. Todd, R. W., N. A. Cole, and R. N. Clark. 2006. Reducing crude protein in beef cattle diet reduces ammonia emission from artificial feedyard surfaces. J. Environ. Qual. 35:404-411 Todd, R.W., N.A. Cole, and R.N. Clark. 2007. Ammonia emissions from open lot beef cattle feedyards on the southern High Plains. Proceedings of 16th Annual 110          International Emission Inventory Conference. Emission Inventories: "Integration, Analysis, Communication”. Raleigh, NC. May 14-17. Available at http://www.epa.gov/ttn/chief/conference/ei16/session5/todd.pdf Tokura, M., K. Ushida, K. Miyazaki, and Y. Kojima. 1997. Methanogens associated with rumen ciliates. FEMS Microbiol. Ecol. 22(2):137-143. U.S. EPA. 2002. Review of emission factors and methodologies to estimate ammonia emissions from animal waste handling. Available at http://www.epa.gov/nrmrl/pubs/600r02017/600r02017.pdf U.S. EPA. 2011. Inventory of U.S. greenhouse gas emissions and sinks: 1990-2009. Washington, DC. U.S. Environmental Protection Agency. Available at http://epa.gov/climatechange/emissions/usinventoryreport.html. USDA National Agricultural Statistics Service. 2011. Meat animals production, disposition, and income: 2010 summary. National Agricultural Statistics Service. Available at http://usda.mannlib.cornell.edu/usda/current/MeatAnimPr/MeatAnimPr-04-282011.pdf. Ushida, K. and J. P. Jouany. 1990. Effect of defaunation on fiber digestion in sheep given two isonitrogenous diets. Anim. Feed Sci. Technol. 29:153–159. Van Soest, P. J. 1994. Nitrogen metabolism. In: Nutritional ecology of the ruminant, 2nd ed. pp. 290-311. Cornell University Press, Ithaca, N.Y. Vongsangnak, W., Gua, J., Chauvatcharin, S. and J. J. Zhong. 2004. Towards efficient extraction of notoginseng saponins from cultured cells of Panax notoginseng. Biochem. Eng. J. 18:115–120. Wallace, R. J., L. Arthaud, and C. J. Newbold. 1994. Influence of Yucca shcidigera extract on ruminal ammonia concentrations and ruminal microorganisms. Appl. Environ. Microbiol. 60:1762–1767. Williams, A. G. and G. S. Coleman. 1992. The Rumen Protozoa. Springer-Verlag, London, UK. Yuan, Z., C. Zhang, L. Zhou, C. Zou, Y. Guo, W. Li, J. Liu, and Y. Wu. 2007. Inhibition of methanogenesis by tea saponin and tea saponin plus disodium fumarate in sheep. J. Anim. Feed Sci. 7(Suppl.2):560–565. Zinn, R. A., E. G. Alvarez, M. F. Montano, A. Plascencia, and J. E. Ramirez. 1998. Influence of tempering on the feeding value of rolled corn in finishing diets for feedlot cattle. J Anim Sci. 76:2239–2246. 111          Zhou, Y., H. Mao, F. Jiang, J. Wang, J. Liu, and C. S. McSweeney. 2011. Inhibition of rumen methanogenesis by tea saponins with reference to fermentation pattern and microbial communities in Hu sheep. Anim. Feed Sci. Technol. 166-167:93-100. 112          CHAPTER 4 EFFECT OF SAPONIN EXTRACTS, IN THE DIET OF HOLSTEIN STEERS OR ADDED DIRECTLY TO THEIR MANURE, ON GASEOUS EMISSIONS FROM THE MANURE 113          Effects of saponin extracts, in the diet of Holstein steers or added directly to their manure, on gaseous emissions from that manure W. Li* and W. Powers* 3 *Department of Animal Science, Michigan State University, East Lansing, MI, 48824                                                              3 Corresponding author:wpowers@msu.edu  114          Abstract A series of experiments (Exp) were conducted to investigate the effects of saponin extracts, in the diet of Holstein steers or added directly to their manure, on gaseous emissions from that manure. Saponin extracts added to the feed or manure were from Quillaja saponaria (quillaja saponin), Yucca schidigera (yucca saponin) and Camellia sinensis (tea saponin). During Exp 1, manure from Holstein steers fed corn and corn silage based control diet (C1) was compared to manure from steers fed control diets plus 1.5% quillaja (QS) or 0.64% yucca (YS) saponins. In Exp 2, the impact of direct application of 2% yucca (CYS, wet basis) or quillaja saponin (CQS, wet basis) to manure collected from steers fed corn and corn silage based diet (C2) on manure air emissions was investigated. In Exp 3 the effects of dietary tea saponin supplementation (TS, 0.25% DM) and direct addition (CTS, 2% wet basis) to manure collected from steers fed corn and corn-silage based diet (C3) on manure air emissions were compared in the same experiment. Gaseous emissions including methane (CH4), ammonia (NH3), hydrogen sulfide (H2S), nitrous oxide (N2O) and non-methane total hydrocarbons (NMTHC) were reported. When saponin extracts were fed, daily manure CH4 emission mass was 40.97, 58.12, and 71.49 mg/d, for YS, C1, and QS, respectively (P < 0.01). Feeding YS resulted in less (P < 0.01) daily manure NH3 emission mass than C1 and QS (318.18 vs. 391.62 and 365.54 mg/d, respectively). Daily manure H2S emission mass differed (P < 0.01) among dietary treatments (10.63, 15.16 and 21.10 mg/d for YS, C1, and QS respectively). In Exp 2 the addition of saponin extracts directly to manure did not affect any emissions monitored. Average daily emission mass of CH4, NH3 and H2S from manure was 11.92, 115          424.25 and 19.36 mg/d, respectively. Overall, the results of these experiments indicate that manure-derived gaseous emissions are altered by dietary inclusion of saponins rather than direct addition to manure. Key words: Quillaja saponaria, Yucca schidigera, Camellia sinensis, air emissions, manure INTRODUCTION The U.S. beef industry produces 25.2 billion kg of meat annually (USDA National Agricultural Statistics Service, 2011), while generating approximately 500 thousand tons of N in excreta (De Wit et al., 1996). Of this, 0-70% is emitted into the atmosphere as ammonia (NH3; Muck and Richards, 1983; Moreira and Satter, 2006; Hristov et al., 2009). Volatilization of NH3 from livestock manure accounts for 65% of annual NH3 emissions and is considered as the largest anthropogenic source (NRC, 2002). Ammonia (NH3) has received considerable attention because of its unpleasant smell, health and welfare issues for both human and animals and its contribution to fine particle matter formation (PM; Lipfert, 1994; Pope and Dockery, 2006). In addition, deposition of NH3 contributes to soil acidification (Falkengren-Grerup, 1986). Microbial fermentation from manure contributes to atmospheric greenhouse gas (GHG) emissions. According to U.S. GHG emissions inventory (U.S. EPA, 2011), the overall GHG emissions from manure has increased by 46% since 1990. In 2009, the manure-derived methane (CH4) and nitrous oxide (N2O) was 49.5 Tg CO2 and 17.9 Tg CO2, respectively. 116          Because of the concerns of air quality problems stemming from manure gaseous emissions, extensive research has been conducted to investigate possible strategies to alleviate the environmental impacts from manure, particularly for NH3 and CH4 emissions. Unfortunately, most often, only one or two gases are targeted and reported while the responses of other gases such as N2O, hydrogen sulfide (H2S) and non-methane total hydrocarbon (NMTHC) remain unknown. In addition, dietary strategies design to mitigate enteric CH4 emissions overlook the manure–derived gaseous emissions making it impossible to determine if reductions in enteric CH4 are offset by CH4 emissions postexcretion, from stored manure (Kreuzer and Hindrichsen, 2006). Saponins are glycosides of plants that can be classified into the categories of triterpenoids and steroids. Triterpenoid saponin extracts from Quillaja saponaria and Camellia sinensis have reduced CH4 emissions and NH3 concentration by ruminants via the detrimental effects on rumen protozoa (Hess et al., 2003; Pen et al., 2006; Guo et al., 2008). Steroid saponin from Yucca schidigera has been shown to reduce ruminal NH3 concentration by directly binding ruminal NH3 (Wallace et al., 1994), and inhibit ruminal methanogenesis, in vitro by indirect toxicity towards protozoa results (Wallace et al., 1994). Despite the abundant information regarding saponin effects on ruminal methanogenesis and NH3 production, little research has focused on the effects of manurederived CH4 and NH3 emissions from dietary inclusion of saponins. 117          The objectives for the current study were to investigate 1) the effects of dietary supplementation of saponin extracts from Quillaja saponaria (quillaja saponin), Yucca schidigera (yucca saponin) and Camellia sinensis (tea saponin) on manure-derived CH4 and NH3 emissions and 2) the effects of adding saponin extracts to manure on manurederived CH4 and NH3 emissions. MATERIAL AND METHODS Experimental design A total of 3 experiments (Exp 1, Exp 2 and Exp 3) were designed to investigate the effects of saponin extracts on gaseous emissions from manure. All Exp employed 12 57 L barrels for manure storage and examined the effects of treatments and day. Exp 1 and 2 were conducted as repeated studies, while Exp 3 was a single study. Fresh manure for all experiments was collected from Holstein steers who were housed individually in environmental rooms. Every morning, manure collection pans placed behind each steer were emptied and manure was composited by dietary treatments and mixed well to provide homogenous compositions. Mixed manure for each treatment was loaded for 5 consecutive days into each barrel (5.5 kg/barrel); this was followed by a 17-d emissions monitoring period (22 d total). Air flow rate was maintained at 7.22 L/min throughout all studies. Average temperature of the barrels was 15.28 °C (11.82 – 21.17 °C), 20.06 °C (17.65 – 23.20 °C) and 23.26 °C (20.82 – 25.62 °C) in Exp 1, 2 and 3, respectively. During Exp 1, manure was collected from 12 Holstein steers which a corn and corn silage based diet was fed as the control treatment (C1), a QS treatment where 1.5% DM of quillaja saponin was added to the base diet and a third diet contained 0.64% DM 118          of yucca saponin (YS) in the basal diet (Table 4.1). Manure treatments were C1, QS and YS which represented manure from steers fed C1, QS and YS diet, respectively; providing 4 replicates of each dietary treatment. Manure used in Exp 2 was collected from 4 Holstein steers where a corn and corn silage base diet was fed (Table 4.1). A total of 12 barrels were randomly assigned to one of the three treatments: 1) C2, fresh manure treatment; 2) CQS, treatment of fresh manure mixed with 2% (on wet basis) of quillaja saponin and 3) CYS, treatment of fresh manure mixed with 2% (on wet basis) of yucca saponin; 4 replicates per treatment. Manure was pre-mixed with yucca/quillaja saponin individually every day for each barrel prior to being loaded to CYS/CQS treatments. Manure used in Exp 2 was collected from 4 Holstein steers which were fed a corn and corn silage based diet (Table 4.1). A total of 12 barrels were randomly assigned to one of the three treatments: 1) C2, fresh manure treatment; 2) CQS, treatment of fresh manure mixed with 2% (on wet basis) of quillaja saponin and 3) CYS, treatment of fresh manure mixed with 2% (on wet basis) of yucca saponin; 4 replicates per treatment. Manure was pre-mixed with yucca/quillaja saponin individually every day for each barrel prior to being loaded in CYS/CQS treatments. In Exp 3, the effects of tea saponin on manure air emissions were compared to control treatment (C3) through dietary inclusion (0.25% DM; TS) and direct application (2% wet basis; CTS) of tea saponin to fresh manure (4 replicates per treatment; Table 4.1). Manure for both the C3 and the CTS treatments was collected from 6 Holstein steers fed a corn and corn silage based diet. Manure for the TS treatment was collected from 6 Holstein steers fed a corn and corn silage based diet with 0.25% (DM) tea saponin 119          supplementation. To construct the CTS treatment, manure was pre-mixed with tea saponin daily prior to filling barrels; similar methods to those used in Exp 2. Saponin sources The 3 saponin products used in the Exp were yucca saponin (YS) which is a powder made entirely from the stem of the Yucca schidigera plant and rich in steroid saponin (contains 8.5% saponin; Desert King International, San Diego, CA, USA); quillaja saponin (QS), which is a triterpenoid saponin enriched extract from pure Chilean soap bark tree Quillaja saponaria (contains 3.6% saponin; Desert King International, San Diego, CA, USA); and tea saponin (TS), which is the whole plant saponin extract from Camellia sinensis and rich in triterpenoid saponin (contains 21.6% saponin; Ningbo Good Green Science & Technology, Ningbo, ZJ, China). Air sampling Twelve plastic 57 L barrels (Interior dimensions: diameter = 30.48 cm, height = 60.96 cm) with black lids which were modified to continuously monitor incoming and exhaust concentrations of gases were used (Fogiel and Powers, 2009). Ammonia (NH3) was measured using a chemiluminescence NH3 analyzer with a detection limit of 0.001ppm (Model 17i, Thermo Fisher, Franklin, MA). Hydrogen sulfide (H2S) was analyzed using pulsed fluorescence SO2-H2S Analyzer with a detection limit of 0.003ppm (TEI Model 450i, Franklin, MA; error = 1% of full-scale at 1 ppm). Concentration of CH4 (detection limit = 0.1 ppm), NMTHC (range = 0 to 10 ppm; detection limit = 0.02 ppm) and N2O (0.03 ppm detection limit at 50,000 ppm range) was 120          measured using an INNOVA 1412 photoacoustic analyzer (Lumasense Technologies, Ballerup, Denmark). Through software control (LabVIEW Version 8.2; National Instruments Corp., Austin, TX), gaseous concentration monitoring of each barrel occurred in a sequential manner. Emission mass was calculated based on emission rate. Gas emission rates were calculated as the product of ventilation rates and concentration differences between exhaust and incoming air using the following equation: ER Q 273 T C C 10 MW V where ER is emission rate, g/min; Q is ventilation rate at room temperature and pressure, L/min; T is air temperature in room exhaust, in Kelvin; Co is gas concentration in room exhaust, mg/kg; Ci is gas concentration in the incoming air, mg/kg; MW is molecular weight of the gas, g/mol; Vm is molar volume of gas at standard condition (22.414 L/mole). Emissions in one full measurement cycle were estimated by multiplying the ER (g/min) with 195 min. Daily emissions were calculated as the sum of the emissions in the 7 or 8 measurement cycles. Daily emission mass, emission rate and concentration were reported in all studies (Li et al., 2011). Manure composition analyses Manure was sampled every day by treatment during the first five days of loading; a representative sample was taken for each barrel at the end of every experiment and + stored at -20 °C until analyzed. Samples were prepared in triplicates; manure NH4 -N 121          (AOAC, 2000) and total kjeldahl N (TKN) contents were determined by distillation and (FOSS Tecator, 1987) in a Michigan State University laboratory. Statistical Analyses In all experiments, emissions data were analyzed using a MIXED model of SAS 9.2 (SAS Institute, 2008). The model tested fixed effects of treatment, day, and the treatment × day interaction, by using period as a random variable. Exceptions were made in analyzing Exp 1 when N2O concentrations were small and close to the detection limit of analyzer, data points collected were insufficient for examining the day effect, therefore only the treatment effect was examined; and during Exp 2, when instrumental malfunction resulted in no data for N2O emissions. Tukey’s test was applied in comparing treatments differences. Significant differences among the means were declared at P < 0.05. RESULTS Methane emissions Feeding steers diets containing YS reduced manure-derived daily CH4 emission rate, while manure from steers fed QS resulted in a greater emission rate compared to the C1 (Exp 1; Table 4.2). Accordingly, the daily concentration in C1 barrels (13.77 mg/kg) was lower than QS (15.36 mg/kg) but higher than YS barrels (11.76 mg/kg; P < 0.01). Average daily emission mass in C1, QS and YS was 58.12, 71.49 and 40.97 mg/d, respectively (P < 0.01). Overall, manure CH4 emissions were not affected by direct applications of yucca (CYS) and quillaja saponins (CQS; Exp 2). Across all treatments, average CH4 122          concentration, emission rate and daily emission mass were 11.92 mg/kg, 0.026 mg/min and 39.44 mg/d, respectively (P > 0.05; Exp 2; Table 4.2). Neither dietary supplementation of TS or direct application to manure (CTS) affected manure-derived CH4 emissions (Exp 3; Table 4.2). Across all treatments, average daily emission mass, emission rate and concentration were 518.12 mg/d, 0.26 mg/min and 73.38 mg/kg, respectively. Ammonia emissions Dietary inclusion of YS reduced manure-derived NH3 (318.18 mg/d) emissions by 18% compared to C1 (391.62 mg/d) where a corn and corn silage based diet was fed (P < 0.01; Table 4.3), whereas no differences were observed between C1 and QS where 1.5% of quillaja saponin was fed to steers (365.54 mg/d). Average daily emission concentration in C1, QS and YS was 51.52, 48.39 and 44.23 mg/kg, respectively (P < 0.01). Emission rate was 0.25, 0.23 and 0.21 mg/min in Control, QS and YS, respectively (P < 0.01). Mixing manure with 2% of saponins (wet basis) in either CYS or CQS treatment did not influenced NH3 emissions compared to the C2 treatment (P > 0.05; Table 4.3). Across all treatments, average concentration, emission rate and daily emission mass were 63.27 mg/kg, 0.28 mg/min and 424.25 mg/d, respectively (P > 0.05). Daily NH3 emission mass showed no differences between TS and C3 (P > 0.05; Exp 3; Table 4.3). Daily NH3 emissions from direct application of 2% tea saponin to 123          manure (CTS) did not differ from C3 treatment (P > 0.05). Average daily emission concentration and emission rate were 67.57 mg/kg and 0.31 mg/min, respectively. Hydrogen sulfide emissions Manure-derived daily H2S emission mass from YS and QS treatment produced 50% and 28% less H2S compared to C1, respectively. Daily emission rate was 0.013, 0.010 and 0.008 mg/min in C1, QS and YS, respectively (P < 0.01; Table 4.4). During Exp 2, saponin amendments did not change H2S emissions from manure. Average daily emission mass was 17.75, 20.66 and 19.67 mg/d in C2, CQS and CYS, respectively (P > 0.05; Table 4.4). Average emission concentration and rate was 1.22 mg/kg and 0.013 mg/min, respectively across all treatments. Daily emission mass of H2S was 27% lower when manure was collected from steers fed a TS compared to C3 treatment (P < 0.01; Exp 3; Table 4.4). Adding tea saponin (CTS) to manure increased H2S emission daily mass by 34% compared to C3 treatment (P < 0.01). The same trend was observed for emission concentration, where daily concentration in C3, TS and CTS treatments was 1.62, 1.18 and 2.02 mg/kg (P < 0.01). No treatment difference was observed for emission rate. Nitrous oxide and non-methane total hydrocarbon emissions Feeding steers QS or YS did not change N2O emissions (Exp 1; Table 4.5). Across all treatments, average daily emission mass was 0.39 mg/d. Dietary inclusion of 0.25% tea saponin reduced daily N2O emission mass by 17% as compared to the control treatment (P = 0.04; Exp 3; Table 4.5), whereas mixing manure with 2% tea saponin 124          (CTS; 3.08 mg/d) resulted in no difference in N2O daily emission mass from manure when compared to the C3 treatment (2.94 mg/d). The same trends were observed in emission concentration and rate. Feeding steers with QS resulted in 21% greater manure-derived NMTHC emission mass compared to C1, while manure from steers fed YS produced 20% less NMTHC than C1 treatment (P = 0.02; Exp 1; Table 4.6). However, direct application of CYS and CQS to manure had no significant effects on daily NMTHC emissions (Exp 2; Table 4.6). Average daily emission mass throughout Exp 2 was 16.97 mg/d. Emissions of NMTHC were not affected by tea saponin treatments (TS or CTS). Average daily emission mass, concentration and rate were 13.78 mg/d, 0.78 mg/kg and 0.009 mg/min, respectively (Exp 3; Table 4.6). Manure N content Dietary supplementation of YS or QS did not change manure DM or N content + compared to C1 (Exp 1; Table 4.7). Average DM, NH4 -N and TKN were 5.77 kg, 91.39 + g and 161.80 g, respectively (P > 0.05). Dietary inclusion of TS increased manure NH4 N by 8% as compared to C3 (P < 0.01; Table 4.7). When mixing the manure with yucca (CYS), quillaja (CQS) or tea (CTS) saponins, no differences were observed in manure characteristics compared to control treatments (Exp 2 and 3; Table 4.7). DISCUSSION Manure-derived CH4 emissions reflect the chemical and microbial processes of fiber degradation from manure (Külling et al., 2002). The availability of unfermented 125          fiber in manure has a strong effect on manure CH4 emissions, whereas easily-fermented carbohydrates such as starch have limited contributions to the total manure-derived CH4 emissions (Kreuzer et al., 1986). Substitution of a forage-based diet with a concentratebased diet in beef cattle increased the manure-derived CH4 emissions by two fold (Hashimoto et al., 1981). Therefore, dietary additives that affect ruminal fiber digestion can further influence the CH4 production from manure (Hashimoto et al., 1981). Saponins have been demonstrated to reduce ruminal methanogenesis, in vitro, mainly via direct inhibition on protozoa population (Hess et al., 2003; Pen et al., 2006; Guo et al., 2008). In addition, growth of ruminal cellulolytic fungi and bacteria can be suppressed by inclusion of saponins under pure culture conditions (Makkar et al., 1995; Wang et al., 2000). These mechanisms can explain the decreased CH4 emissions found in YS treatment (Exp 1). However the day by treatment interaction indicated that the effect was dependent on the day and length of the storage. Given that diets contained the same saponin concentration (0.54 g/kg DM), it is unclear why reduced manure CH4 emissions were observed from YS treatment, whereas emissions from QS treatment produced more CH4 than C1 (Exp 1) and manure CH4 emissions from TS treatment did not differ from C3 (Exp 3). The differences in chemical structures and efficacy in terms of inhibiting ruminal protozoa and cellulolytic fungi and bacteria may have implications on manure-derived CH4 emissions but future research is needed to explain the mechanisms. 126          Production of CH4 employs anaerobic processes. Mixing saponin with manure rather than surface application provides better contact between saponin and anaerobic microbes inside the manure. In addition, if saponin can inhibit methanogenesis through reducing the availability of hydrogen provided by protozoa and other bacteria and fungi in manure, direct application of a greater concentration of saponin (2% wet basis) should be more effective than residues from diet (0.25 to 1.5% DM). As a result, a more pronounced effect of inhibited CH4 production by direct saponin application was expected in CTS treatment which contained greatest amount of saponin. However, none of the saponin treatment (CQS, CYS and CTS) showed differences in CH4 production compared control treatments (C2 and C3). The unexpected lack of effects by direct mixing of saponin with manure suggests that the microbial degradation of manure fiber may not be affected by the saponins tested in this study. On the other hand, our results support the findings that saponin is more effective on protozoa rather than methanogens (Guo et al., 2008; Zhou et al., 2011). Because, protozoa could not survive when pH is below 5.0 (Coleman and Sandford, 1979), although we did not monitor rumen protozoa population in the manure, after passing through the abomasum (pH = 2.1 to 2.2) and lower gastro-intestinal tract, the number of live protozoa in the manure should be minimal. Average CH4 emissions were similar between Exp 1 (56.86) mg/d and 2 (39.44 mg/d), whereas there was an approximately 10-fold greater emission Exp 3. It is well established that temperature has a significant effect on CH4 emissions. Several studies 127          reported that manure produced 2 to 20 times more CH4 when the manure pile temperature increased from 6 to 35 °C (Hashimoto et al., 1981; Hashimoto, 1982; Lokshina and Vavilin, 1999; Chae et al., 2008). Therefore the increases in ambient temperature between Exp 1 and 2 and Exp 3 may have increased manure temperature and contributed to increased CH4 production in Exp 3. A day effect was observed for NH3 emissions in all Exp. The significant treatment by day interactions suggested that the duration can significantly affect NH3 emissions. Saponin extract from Yucca schidigera is considered to reduce NH3 by + binding NH4 and inhibiting rumen protozoa (Kemme et al., 1993; Wallace et al., 1994). However, reports from dietary inclusion of yucca saponin on NH3 emissions are not consistent. Studies showed that dietary supplementation of yucca saponin achieved a 20 to 50% reduction in NH3 emissions in poultry and swine (McCrory and Hobbs, 2001, review). Panetta et al. (2006) reported that NH3 emissions were not affected by dietary yucca saponin inclusion in swine. This study was conducted in environmental chambers to investigate the effects of saponin extract on animal and manure-derived NH3 emissions. In other studies, saponin added to manure produced no effects on NH3 emissions (Lee et al., 2007). This is consistent with our findings. Yucca saponin was found to reduce manure-derived NH3 emissions through dietary inclusion (Exp 1), whereas yucca saponin applied to manure had no effect. Initial 128          + manure contents of NH4 -N and TKN in yucca saponin treatments (YS and CYS) were not different from the control treatments, suggesting the reduced NH3 emissions observed in the YS treatment may be because of its effect on protozoa and other microbes in the + rumen rather than a direct binding effect on NH4 . However, the effects of dietary additives on manure NH3 emissions could be short-term (Külling et al.; 2002), during long-term storage (14 wks), adaptation would occur. + Manure NH4 -N concentration in TS treatment was greater than that in C3 but the average NH3 emissions with the TS treatment in Exp 3 was not different from C3, + indicating that tea saponin may be able to bind NH4 so that lead to a slower releasing rate of NH3. Because we only examined the short-term manure storage, this effect may be reduced or eliminated in the long-term. Neither diet nor manure supplementation with quillaja saponin (QS and CQS) changed manure-derived NH3 emissions. Overall, our results indicated that saponins, of different chemical structures and application methods, had diverse effects on manure-derived NH3 emissions. Ruminal hydrogen is produced during the process of VFA production. In addition to methanogens, sulfur reducing bacteria (SRB) can also incorporate hydrogen for the - purpose of reducing SO4 with H2S the most abundant product (Biebl and Pfennig, 1977). The production of H2S can therefore be considered as a competitive pathway for methanogenesis. Saponin extract from Yucca schidigera reduced H2S emissions by 129          binding H2S or decreasing the abundance or activity of SRB, in vitro (Gibson et al., 1993; Giffard et al., 2001). Our results agreed with those findings. Dietary supplementations of all 3 saponins were found to reduce average daily H2S emissions from manure. However, yucca and quillaja saponin applied directly to manure had no effects on H2S emissions. The lack of effect observed from manure application suggests that the effects may only occur in rumen. When tea saponin was applied directly to manure (CTS), a 34% increase in H2S emissions was observed compared to C3. This again suggests that tea saponin is not as effective in manure as in the rumen and the increased H2S could possibly be due to changed manure VFA composition. Because we did not analyze the VFA or C content in manure, this hypothesis needs further confirmation. N2O is the intermediate product of nitrification from nitrate or denitrification + from NH3/NH4 . Some studies showed a positive correlation exist between N2O + emission and NH4 availability (Fukumoto et al., 2003; Heller et al., 2010). Therefore + lower initial NH4 -N concentration in TS treatment (Table 4.6) could possibly explain associated reduced N2O emissions. CONCLUSION Dietary inclusion of YS decreased manure-derived CH4, NH3 and H2S emissions but had no effect on manure N composition. Dietary supplementation of QS increased manure CH4 emissions but decreased H2S emission from manure. Manure N composition 130          and NH3 emissions were not affected by dietary inclusion of quillaja saponin. Manure + collected from steers fed 0.25% of TS produced less H2S and excreted less NH4 -N in manure, while CH4 and NH3 emissions were not affected. Our results indicated that saponins, of different chemical structures and application methods, had diverse effects on manure-derived NH3 emissions. In addition, the lack of saponin’s effects on air emissions by direct application indicates saponins are more effective as potential feed additives. ACKNOWLEDGEMENTS Funding for this work was provided through the National Research Initiative Air Quality Program of the Cooperative State Research, Education, and Extension Service, USDA, under Agreement No. 2005-35112-15356. The authors wish to acknowledge the USDA for their support of this work. The authors would like to thank Jolene Roth, Andrew Fogiel and Mark Schilling for lab and instrumental maintenance and Desert King International (San Diego, CA) for saponin donation and composition analysis. 131          Table 4.1. Ingredient and nutrient composition of diets fed to steers (DM basis) Exp 1 and 2 Control * YS Exp 3 QS Control TS 46 46 3 5 46 46 3 5 46 46 3 5 0.64 0.64 --1.5 --- --- -- 0.25 Ingredients (%) DM High moisture corn Corn silage Soybean meal 46 46 3 5 # Supplement 50 Saponins yucca saponin quillaja saponin --- tea saponin 46 46 3 5 -- -- -- Supplement 50, % of DM TM¶ 1.4 24.9 48.3 0.3 9.6 0.2 9.6 5.1 0.7 Akey TM premix # 4 Limestone Soybean meal, 48% N TM Rumensin 80 TM salt Vitamin E, 5% Urea, 45% N Potassium chloride Selenium 90 Total 100 Analyzed composition, % DM * DM CP ADF NDF P ME (Mcal/kg) 43.9 11.6 16.5 27.4 0.27 1.82 43.1 11.8 17.0 27.8 0.28 1.79 43.3 12.1 16.4 28.1 0.27 1.83 50.5 12.2 10.1 27.8 0.29 1.80 50.1 11.9 11.5 26.4 0.29 1.82 Treatments were corn and corn silage based control diet with inclusion of yucca saponin which is the saponin extract from Yucca schidigera (YS), quillaja saponin which is the saponin extract from Quillaja saponaria (QS) and tea saponin, extract from Camellia sinensis (TS). Inclusion levels of saponins were adjusted by the actual saponin 132          Table 4.1. Ingredient and nutrient composition of diets fed to steers (DM basis), con’t content in the product, adjusted actual dietary saponin concentration was 0.54 g/kg in all saponin treatments. # Middle section of the table lists the ingredients for BFS50 supplement for all treatments ¶ Akey TM premix # 4 composition: 9% Mg, 4% S, 0.02% Co, 1% Cu, 0.09% I, 2% Fe, 4% Mn, 0.03% Se, 4% Zn, 4,400,000 IU vitamin A, 550,000 IU vitamin D, and 5,500 IU vitamin E/kg (Akey Inc., Lewisburg, OH). 133          Table 4.2. Least squares means of saponin’s effect on manure CH4 emissions Average daily Average daily Cumulative average concentration emission rate daily emission mass mg/kg mg/min mg/d Ⅰ Exp 1 b C1 13.77 QS 15.36 YS SEM b Diet Day Diet × Day b 0.037 58.12 0.046 71.49 11.76 0.027 0.30 0.002 Source of variation <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 40.97 4.13 c c a a c a <0.01 <0.01 <0.01 Ⅱ Exp 2 C2 CQS CYS SEM Diet Day Diet × Day 12.34 0.028 11.73 0.025 11.70 0.025 0.40 0.002 Source of variation 0.43 0.48 <0.01 <0.01 0.17 0.54 39.36 39.99 38.98 4.46 0.99 <0.01 0.62 Ⅲ Exp 3 C3 TS CTS SEM Diet Day Diet × Day 73.33 0.25 72.12 0.27 74.69 0.26 4.64 0.02 Source of variation 0.92 0.65 <0.01 <0.01 1.00 0.96 504.23 495.91 554.23 43.87 0.59 <0.01 1.00 Ⅰ Treatments were manure collected from Holstein steers fed corn and corn silage based diet (C1) with dietary inclusion of the saponin extract from Yucca schidigera (YS; 0.64% DM) and the saponin extract from Quillaja saponaria (QS; 1.5% DM). Ⅱ Treatments were manure collected from Holstein steers fed corn and corn silage based diet (C2) and C2 with 2% inclusion (wet basis) of the saponin extract from Yucca schidigera (CYS) and C2 treatment with 2% inclusion (wet basis) of the saponin extract from Quillaja saponaria (CQS). Ⅲ Treatments were manure collected from Holstein steers fed corn and corn silage based C3 diet; manure from C3 treatment with inclusion of 2% (wet basis) the saponin extract from Camellia 134       Table 4.2. Least squares means of saponin’s effect on manure CH4 emissions, con’t sinensis (CTS) and manure from Holstein steers fed 0.25% (DM basis) of saponin extract from Camellia sinensis (TS). a, b, c Significant differences observed at the P < 0.05 probability level. 135       Table 4. 3. Least squares means of saponin’s effect on manure NH3 emissions Average daily Average daily Cumulative average concentration emission rate daily emission mass mg/kg mg/min mg/d Ⅰ Exp 1 b b C1 51.52 QS 48.39 YS SEM 44.23 0.21 16.52 0.08 Source of variation <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Diet Day Diet × Day b 0.25 0.23 b 391.62 365.54 b a a b a 318.18 133.32 <0.01 <0.01 <0.01 Ⅱ Exp 2 C2 CQS CYS SEM Diet Day Diet × Day 63.26 0.28 63.09 0.28 63.46 0.28 1.40 0.01 Source of variation 0.98 0.98 <0.01 <0.01 <0.01 <0.01 389.05 440.33 443.37 28.17 0.31 <0.01 <0.01 Ⅲ Exp 3 C3 TS CTS SEM Diet Day Diet × Day 67.42 0.31 70.98 0.33 64.30 0.29 16.06 0.10 Source of variation 0.12 0.11 <0.01 <0.01 <0.01 <0.01 460.88 481.11 455.02 158.35 0.86 <0.01 <0.01 Ⅰ Treatments were manure collected from Holstein steers fed corn and corn silage based diet (C1) with dietary inclusion of the saponin extract from Yucca schidigera (YS; 0.64% DM) and the saponin extract from Quillaja saponaria (QS; 1.5% DM). Ⅱ Treatments were manure collected from Holstein steers fed corn and corn silage based diet (C2) and C2 with 2% inclusion (wet basis) of the saponin extract from Yucca schidigera (CYS) and C2 treatment with 2% inclusion (wet basis) of the saponin extract from Quillaja saponaria (CQS). Ⅲ Treatments were manure collected from Holstein steers fed corn and corn silage based C3 diet; manure from C3 treatment with inclusion of 2% (wet basis) the saponin extract from Camellia 136       Table 4. 3. Least squares means of saponin’s effect on manure NH3 emissions, con’t sinensis (CTS) and manure from Holstein steers fed 0.25% (DM basis) of saponin extract from Camellia sinensis (TS). a, b, c Significant differences observed at the P < 0.05 probability level. 137       Table 4. 4. Least squares means of saponin’s effect on manure H2S emissions Average daily Average daily Cumulative average concentration emission rate daily emission mass mg/kg mg/min mg/d Ⅰ Exp 1 c b c C1 1.28 0.013 21.10 QS 0.91 0.010 15.16 YS SEM 0.69 0.008 0.83 0.001 Source of variation <0.01 <0.01 <0.01 <0.01 0.20 0.19 Diet Day Diet × Day b a a a b a 10.63 13.65 <0.01 <0.01 0.19 Ⅱ Exp 2 C2 CQS CYS SEM Diet Day Diet × Day 1.21 0.013 1.25 0.013 1.20 0.013 0.05 0.001 Source of variation 0.74 0.77 <0.01 <0.01 0.22 0.22 17.75 20.66 19.67 1.46 0.36 <0.01 0.17 Ⅲ Exp 3 C3 TS CTS SEM Diet Day Diet × Day 1.62b 0.018 1.18a 0.013 c 2.02 0.022 0.09 0.001 Source of variation <0.01 0.11 <0.01 <0.01 0.03 0.03 25.57b 18.45a 34.29c 1.77 <0.01 <0.01 <0.01 Ⅰ Treatments were manure collected from Holstein steers fed corn and corn silage based diet (C1) with dietary inclusion of the saponin extract from Yucca schidigera (YS; 0.64% DM) and the saponin extract from Quillaja saponaria (QS; 1.5% DM). Ⅱ Treatments were manure collected from Holstein steers fed corn and corn silage based diet (C2) and C2 with 2% inclusion (wet basis) of the saponin extract from Yucca schidigera (CYS) and C2 treatment with 2% inclusion (wet basis) of the saponin extract from Quillaja saponaria (CQS). Ⅲ Treatments were manure collected from Holstein steers fed corn and corn silage based C3 diet; manure from C3 treatment with inclusion of 2% (wet basis) the saponin extract from Camellia 138       Table 4. 4. Least squares means of saponin’s effect on manure H2S emissions, con’t sinensis (CTS) and manure from Holstein steers fed 0.25% (DM basis) of saponin extract from Camellia sinensis (TS). a, b, c Significant differences observed at the P < 0.05 probability level. 139       Table 4.5. Least squares means of saponin’s effect on manure N2O emissions Average daily Average daily Cumulative average concentration emission rate daily emission mass mg/kg mg/min mg/d Ⅰ Exp 1 C1 QS YS SEM Diet 0.21 0.0001 0.20 0.0005 0.21 0.0002 0.01 0.0002 Source of variation 0.99 0.49 0.14 0.81 0.24 0.31 0.31 Ⅱ Exp 3 b b b C3 0.48 TS 0.45 0.0016 2.44 CTS SEM 0.47 0.0020 0.06 0.0003 Source of variation <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 3.08 0.21 Diet Day Diet × Day 0.0021 a b a b 2.94 a b 0.04 <0.01 <0.01 Ⅰ Treatments were manure collected from Holstein steers fed corn and corn silage based diet (C1) with dietary inclusion of the saponin extract from Yucca schidigera (YS; 0.64% DM) and the saponin extract from Quillaja saponaria (QS; 1.5% DM). Ⅱ Treatments were manure collected from Holstein steers fed corn and corn silage based C3 diet; manure from C3 treatment with inclusion of 2% (wet basis) the saponin extract from Camellia sinensis (CTS) and manure from Holstein steers fed 0.25% (DM basis) of saponin extract from Camellia sinensis (TS). a, b, c Significant differences observed at the P < 0.05 probability level. 140       Table 4.6. Least squares means of saponin’s effect on manure NMTHC emissions Average daily Average daily Cumulative average concentration emission rate daily emission mass mg/kg mg/min mg/d Ⅰ Exp 1 C1 1.00 b 0.008 QS 1.07 b 13.79 0.011 16.81 YS SEM Diet a 0.007 0.85 0.15 0.0003 Source of variation 0.04 <0.01 b c a 11.03 3.39 0.02 Ⅱ Exp 2 C2 CQS CYS SEM Diet Day Diet × Day 1.06 0.011 1.07 0.011 1.10 0.012 0.03 0.0004 Source of variation 0.67 0.67 <0.01 <0.01 0.04 0.04 15.53 17.33 18.05 1.47 0.43 <0.01 0.44 Ⅲ Exp 3 C3 TS CTS SEM Diet Day Diet × Day 0.69 0.008 0.83 0.010 0.83 0.010 0.19 0.003 Source of variation 0.15 0.15 <0.01 <0.01 0.53 0.25 11.68 13.64 15.24 4.79 0.32 <0.01 0.50 Ⅰ Treatments were manure collected from Holstein steers fed corn and corn silage based diet (C1) with dietary inclusion of the saponin extract from Yucca schidigera (YS; 0.64% DM) and the saponin extract from Quillaja saponaria (QS; 1.5% DM). Ⅱ Treatments were manure collected from Holstein steers fed corn and corn silage based diet (C2) and C2 with 2% inclusion (wet basis) of the saponin extract from Yucca schidigera (CYS) and C2 treatment with 2% inclusion (wet basis) of the saponin extract from Quillaja saponaria (CQS). Ⅲ Treatments were manure collected from Holstein steers fed corn and corn silage based C3 diet; manure from C3 treatment with inclusion of 2% (wet basis) the saponin extract from Camellia sinensis (CTS) and manure from Holstein steers fed 0.25% (DM basis) of saponin extract from Camellia sinensis (TS). 141       Table 4.6. Least squares means of saponin’s effect on manure NMTHC emissions, con’t a, b, c Significant differences observed at the P < 0.05 probability level. 142       Table 4. 7. Initial N content of manure collected from steers fed with or without saponin supplementation (27.5 kg wet manure) DM kg + Ⅴ TKN g NH4 -N g + NH4 -N Ⅵ / TKN Ⅰ Control QS YS SEM 5.67 5.78 5.86 0.11 Diet 0.39 Exp 1 97.09 163.07 89.25 159.53 87.82 162.79 3.21 8.54 Source of variation 0.25 0.30 0.60 0.56 0.54 0.06 0.09 Ⅱ Control CQS CYS SEM 5.78 5.78 5.78 0.16 Diet 0.83 Exp 2 89.76 165.69 82.89 169.33 81.99 174.59 2.80 9.64 Source of variation 0.87 0.27 0.54 0.49 0.47 0.03 0.25 Ⅲ Exp 3 Control TS 5.61 5.49 CTS SEM 5.61 0.07 Diet 0.32 a 167.04 0.49 b 174.92 0.50 163.64 83.54 3.79 4.32 Source of variation 0.01 0.41 0.51 0.03 81.71 88.07 ab 0.38 Ⅰ Treatments were manure collected from Holstein steers fed corn and corn silage based diet (C1) with dietary inclusion of the saponin extract from Yucca schidigera (YS; 0.64% DM) and the saponin extract from Quillaja saponaria (QS; 1.5% DM). Ⅱ Treatments were manure collected from Holstein steers fed corn and corn silage based diet (C2) and C2 with 2% inclusion (wet basis) of the saponin extract from Yucca schidigera (CYS) and C2 treatment with 2% inclusion (wet basis) of the saponin extract from Quillaja saponaria (CQS). Ⅲ Treatments were manure collected from Holstein steers fed corn and corn silage based C3 diet; manure from C3 treatment with inclusion of 2% (wet basis) the saponin extract from Camellia sinensis (CTS) and manure from Holstein steers fed 0.25% (DM basis) of saponin extract from Camellia sinensis (TS). Ⅴ : TKN=total kejldahl nitrogen. Ⅵ : NH4-N to TKN ratio. a, b, c Significant differences observed at the P < 0.05 probability level. 143       LITERATURE CITED 144       LITERATURE CITED AOAC. 2000. Official methods of analysis of the association of official analytical chemists, 16th ed., Method 928.08. AOAC, Int., Arlington, VA. Biebl, H., and N. Pfennig. 1977. Growth of sulfate-reducing bacteria with sulfur as electron acceptor. Arch. Microbiol. 112:115-117. Coleman, G. S., and D. C. Sandford. 1979. The uptake and utilization of bacteria, amino acids and nucleic acid components by the rumen ciliate Eudiplodinium maggii. J. Appl. Bacteriol. 47, 409-419. Fukumoto, Y., T. Osada, D. Hanajima, and K. Haga. 2003. Patterns and quantities of NH3, N2O and CH4 emissions during swine manure composting without forced aeration—Effect of compost pile scale. Bioresour. Technol. 89:109–114. Fogiel, A. C. and W. J. Powers. 2009. Impact of amendments on air emissions from stored animal manure. ASABE Annual International Meeting. Reno, NV. June 21- 24. Paper 096123. Falkengren-Grerup, U. 1986. Soil acidification and vegetation changes in deciduous forest in southern Sweden. Oecologia 70:339-347. Fakhoury, K. J., A. J. Heber, P. Shao and J. Q. Ni. 2000. Correlation of odor detection thresholds with concentrations of hydrogen sulfide, ammonia and trace gases emitted from swine manure. 2000 ASAE Annual International Meeting, Milwaukee, Wisconsin, USA, July 912. Gibson, G.R., G. T. Macfarlane, J. H. Cummings. 1993. Sulphate reducing bacteria and hydrogen metabolism in human large intestine. Gut. 34:437–439. Giffard, C. J., S. B. Collins, N. C., R. F. Butterwick and R. M. Batt. 2001. Stoodley Administration of charcoal, Yucca schidigera, and zinc acetate to reduce malodorous flatulence in dogs. J. Am. Vet. Med. Assoc. 218:892–896. Guo, Y., J. Liu, Y. Lu, W. Zhu, S. E. Denman and C. S. McSweeney. 2008. Effect of tea saponin on methanogenesis, microbial community structure and expression of mcrA gene, in cultures of rumen micro-organisms. Lett. Appl. Microbiol. 47:421-426. Hashimoto, A. G., V. H. Varel and Y. R. Chen. 1981. Ultimate methane yield from beef cattle manure: effect of temperature, ration, constituents, antibiotics and manure age. Agric. Waste. 3:241–256. Heller, H., A. Bar-Tal, G. Tamir, P. Bloom, R. T. Venterea, D. Chen, Y. Zhang,, C. E. Clapp and P. Fine. 2010. Effects of manure and cultivation on carbon dioxide and nitrous oxide emissions from a corn field under Mediterranean conditions. J. Environ. Qual. 39:437448. 145       Hess, H. D., M. Kreuzer, T. E. Díaz, C. E. Lascano, J. E. Carulla, C. R. Soliva and A. Machmüller. 2003. Saponin rich tropical fruits affect fermentation and methanogenesis in faunated and defaunated rumen fluid. Anim. Feed Sci. Technol. 109:79–94. Hristov, A. N., S. Zaman, M. Vander Pol. M. Ndegwa, L. Campbell and S. Silva. 2009. Nitrogen losses from dairy manure estimated through nitrogen mass balance and chemical markers. J. Environ. Qual. 38:2438–2448. Kemme, P. A., A. W. Jongbloed, B. M. Dellaert and F. Krol-Kramer. 1993. The use of Yucca schidigera extract as a “urease inhibitor” in pig slurry. p. 330–335. In M.W.A. Verstegen et al. (ed.) Proc. of the 1st Int. Symp. on Nitrogen Flow in Pig Production and Environmental Consequences. EAAP Publ. 69. Pudoc, Wageningen, the Netherlands. Kreuzer, M., M. Kirchgessner, and H. L. Müller. 1986. Effect of defaunation on the loss of energy in wethers fed different quantities of cellulose and normal or steamflaked maize starch. Anim. Feed Sci. Technol. 16:233–241. Kreuzer, M. and Hindrichsen, I. K. 2006. Methane mitigation in ruminants by dietary means: the role of their methane emission from manure. In Greenhouse Gases and Animal Agriculture: An Update. C.R. Soliva, J. Takahashi, and M. Kreuzer (eds.). International Congress Series No. 1293, 199-208. Elsevier, The Netherlands. Külling, D. R., F. Dohme, H. Menzi, F. Sutter, P. Lischer and M. Kreuzer. 2002. Methane emissions of differently fed dairy cows and corresponding methane and nitrogen emissions from their manure during storage. Environ. Monit. Assess. 79: 129–150. Lee, S. R., J. K. Han, Y. U. Choi and K. Nam. 2007. Reduction of Ammonia and Hydrogen Sulfide Emission from Swine Manure Using Aqueous Foams Amended with Microorganisms and Chemical Additives. Clean Soil, Air, Water. 35(3): 230-234. Lipfert, F. W. 1994. Air pollution and community health: a critical review and data sourcebook. Van Nonstrand Reinhold. New York, NY. Makkar, H. P. S., M. Blümmel, and K. Becker. 1995. In vitro effects of and interactions between tannins and saponins and fate of tannins in the rumen. J. Sci. Food Agric. 69:481-493. McCrory, D. F. and P. J. Hobbs. 2001. Additives to reduce ammonia and odor emissions from livestock wastes: a review. J. Environ. Qual. 30:345-355. Moreira, V.R. and L.D. Satter. 2006. Effect of Scraping Frequency in a Freestall Barn on Volatile Nitrogen Loss from Dairy Manure. J. Dairy Sci. 89:2579–2587. Muck, R. E. and B. K. Richards. 1983. Losses of manurial nitrogen in free-stall barns. Agric. Wastes. 7(2):65-79. NIOSH. 2007. NIOSH Pocket Guide to Chemical Hazards. Cincinnati, Ohio: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, 146       National Institute for Occupational Safety and Health. Available at http://www.cdc.gov/niosh/npg/. NRC. 2002. The scientific basis for estimating air emissions from animal feeding operations. National Research Council. National Academy Press, Washington, DC. Pen, B., C. Sar, B. Mwenya, K. Kuwaki, R. Morikawa, and J. Takahashi. 2006. Effects of Yucca schidigera and Quillaja saponaria extracts on in vitro ruminal fermentation and methane emission. Anim. Feed Sci. Technol. 129:175–186. Pope III, C. A. and D. W. Dockery. 2006. Health effects of fine particulate air pollution: lines that connect. J. Air & Waste Manage. Assoc. 56:709 –742. Panetta, D. M., W. J. Powers, H. Xin, B. J. Kerr and K. J. Stalder. 2006. Nitrogen excretion and ammonia emissions from pigs fed modified diets. J. Environ. Qual. 35:1297–1308. SAS Institute. 2008. Version 9.2. SAS Institute, Inc., Cary, NC, USA. USDA.1997. Confined animal and manure nutrient data system: feedlot beef. Washington, DC. U.S. Department of Agriculture. Available at http://www.ers.usda.gov/Data/manure/default.asp?ERSTab=2. USDA National Agricultural Statistics Service. 2011. Meat animals production, disposition, and income: 2010 summary. National Agricultural Statistics Service. Available at http://usda.mannlib.cornell.edu/usda/current/MeatAnimPr/MeatAnimPr-04-28-2011.pdf. USEPA. 2011. Inventory of U.S. greenhouse gas emissions and sinks: 1990-2009. Washington, DC. U.S. Environmental Protection Agency. Available at http://epa.gov/climatechange/emissions/usinventoryreport.html. Wallace, R. J., L. Arthaud, and C. J. Newbold. 1994. Influence of Yucca shcidigera extract on ruminal ammonia concentrations and ruminal microorganisms. Appl. Environ. Microbiol. 60:1762–1767. Wang, Y., T. A. McAllister, L. J. Yanke, and P. R. Cheeke. 2000. Effect of steroidal saponin from Yucca schidigera extract on ruminal microbes. J. Appl. Microbiol. 88:887–896. . 147       CHAPTER 5 SUMMARY 148       EFFICACY OF SAPONINS IN REDUCING CH4 EMISSIONS Ruminants are the major contributor of CH4 emissions from agriculture sector. Methane emissions from ruminants also represent great energy loss. Saponins have been demonstrated to reduce CH4 emissions from ruminates, in vitro. These findings have also been confirmed during our in vitro study. One of the major aims for this thesis was to investigate saponin’s effect on CH4 reduction, in vivo. However, the inhibitive effect of saponins on CH4 production and the dose-dependent response were not observed during the animal study. One exception is when tea saponin was added at 0.5% of dietary DM to the steers, daily emission mass was reduced significantly when compared to control treatment (Chapter 3). Unfortunately, this effect was achieved at the expense of declined DMI and ADG. When the daily mass was adjusted on DMI basis, CH4 emissions from steers fed tea saponin treatment were not different from control treatment. In order to receive a significant response of CH4 reduction, a greater amount of saponin must be fed. We have demonstrated in our in vitro study that at 2.0% inclusion level, gas production in both yucca and tea saponin treatments was decreased, suggesting undesirable consequences such as declined fiber digestion could occur at higher dietary saponin concentration. In addition, the saponin at higher concentration can cause bad palatability of the feed which can lead to decreased DMI as we have observed from Chapter 3. When relating my research work to other studies that observed significant effects of dietary saponins on CH4 emissions it is necessary to note that most studies reporting a CH4 decrease, involved small ruminants (sheep and lamb), indicating that a species difference may exist. In addition, dietary compositions such as feed ingredients and forage to concentration 149       ration have impacts on rumen microbial populations, whereas this thesis only examined one diet type. In the future, it is necessary to examine saponin’s effects with other feed types. Although, CH4 emissions were not affected by dietary yucca or quillaja saponin inclusion, during the manure storage study, dietary inclusion of 0.64% yucca saponin decreased manurederived CH4 emissions while quillaja saponin increased manure-derived CH4 emissions. These results suggested that when examine saponins effects on CH4 emissions from ruminants, both animal-derived and manure-derived emissions should be considered. On the other hand, when saponins were mixed directly with manure, no effect on CH4 emissions was observed among all saponin treatments. These results supported those findings that the major effect of saponins in aspects of CH4 reduction was to inhibit protozoa population rather than direct affect methanogen population. On an average, CH4 emissions were 18% greater from whole animal emissions compared to enteric and rectum CH4 emissions, indicating that manure when partially removed on daily basis, contributes less than 20% to total CH4 emissions. EFFICACY OF SAPONINS IN REDUCING NH3 AND OTHER GASEOUR EMISSIONS The manure-derived NH3 emissions are one of the major concerns from animal agricultural. Some research has been conducted to investigate saponin’s effect on manurederived NH3 emissions from both ruminants and non-ruminants. However, results are not consistent. In this thesis, manure-derived NH3 emissions were monitored from emissions from a manure pan with partially removal on daily basis for 4 wk (manure source retained; Chapter 3) and from post-excretion of manure storage from tubs for 3 wk (no animal present; Chapter 4). 150       When manure was partially removed and re-mixed on a daily basis, quillaja saponin increased NH3 emissions during one of the two animal studies, whereas NH3 emissions from yucca saponin or tea saponin treatments were not different from control. Comparably, during manure storage studies (Chapter 4), the dietary yucca saponin treatment decreased NH3 emissions while quillaja and tea saponin treatments did not change NH3 emissions. These results indicated that saponins were more effective in affecting microbes in the rumen rather than + binding NH4 in the manure. The manure storage study only examined saponin’s effects on a short term basis; adaptation could possibly occur during long-term storage, eliminating saponins effects. By monitoring H2S, N2O and NMTHC emissions during the manure storage study (Chapter 4), we found that through dietary inclusion, 1) quillaja saponin reduced H2S emissions, increased NMTHC emissions but did not affect N2O emissions; 2) both NMTHC and H2S emissions were reduced in yucca while N2O was not affected and 3) tea saponin treatments reduced NMTHC, H2S and N2O emissions. In contrary, none of the gaseous emissions were affected by direct mixing saponin with manure, except an increase of H2S was observed in tea saponin treatment. Our results further supported the hypothesis that the principle biological effects of saponins occurred in the rumen rather than the manure. IMPLICATIONS AND RECOMMENDATIONS Beef steers are one of the most significant contributors of CH4 in agriculture sector. In addition, CH4 emissions from enteric fermentation and manure represent the loss of feed energy and nutrients. Therefore, reducing CH4 emissions from beef and manure will benefit both the 151       animal and the environment. Saponins of different origins, structures and concentrations were investigated for their effects on animal and manure-derived CH4 and other gaseous emissions in this thesis. Results presented through Chapter 2 to 4 imply that dietary saponin supplementation may not be a good strategy for animal-derived CH4 reduction. However, when extending the scope to manure-derived CH4 emissions, reductions observed in yucca saponin treatment (Chapter 4) suggested that dietary supplementation of steroid saponin (yucca saponin) could be more effective in the aspect of reducing manurederived CH4 emissions compared to triterpenoid saponins (tea and quillaja) or direct application. Besides, our results for the first time demonstrated that other gaseous emissions from manure, such as NMTHC, H2S and N2O emissions, could be affected by direct application of saponins, providing a new aspect of saponin application in terms of regulating manure-derived gaseous emissions. This thesis only investigated the effects of saponins on gaseous emissions from Holstein steers (280 to 500 of BW) and only one type of diet was used throughout this thesis. For future research, steers at different growth stage and fed with different dietary compositions will need to be considered. Because the genetic differences exist among different animal species, such as dairy cow, steers and lamb, et al., different response to dietary saponin inclusions can occur. In the future, studies will need to compare saponin’s effect among different animal species. 152