Citation
Effect of Fibrolytic Enzymes on the Nutritive Value of Tropical Forages and Performance of Beef Steers

Material Information

Title:
Effect of Fibrolytic Enzymes on the Nutritive Value of Tropical Forages and Performance of Beef Steers
Creator:
KREUGER, NATHAN ALFRED
Copyright Date:
2008

Subjects

Subjects / Keywords:
Ammonia ( jstor )
Cell walls ( jstor )
Digestion ( jstor )
Enzymatic treatment ( jstor )
Enzymes ( jstor )
Forage ( jstor )
Grasses ( jstor )
In vitro fertilization ( jstor )
Lignin ( jstor )
Rumen ( jstor )
City of Madison ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Nathan Alfred Kreuger. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
11/30/2006
Resource Identifier:
496613320 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

EFFECT OF FIBROLYTIC ENZYME S ON THE NUTRITIVE VALUE OF TROPICAL FORAGES AND PERFORMANCE OF BEEF STEERS By NATHAN ALFRED KRUEGER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Nathan Alfred Krueger

PAGE 3

To my lovely wife Wimberley

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank my supervisory co mmittee chair, Dr. Adegbola Adesogan, for his valuable guidance during my PhD program as well as th e rest of my supervisory committee (Dr. Charles Staples, Dr. Lynn So llenberger and Dr. Ramon Littell) for their time and dedication to my research program. I would also like to thank a ll of my lab partners and lab supervisors (Wimberley Krueger, Dervin Dean, Sam-Churl Kim, Ashl ey Hughes, Kathy Arriola, Jamie Foster, Susan Chikagwa Malunga, Joao Vendram ini, John Funk, Alvin Boning, Mustapha Salawu, Max Huisdin, Pam Miles and Sergei Sennikov) for their help during my laboratory and field activities. Special thanks are expressed to Dr. Dario Colombatto for helping me with the determination of enzyme activities. I would like to also thank my parent s (Alfred and Susan Krueger) for their continuous support and dedication toward all my a dventures in life, especially my Ph.D. Thanks go to my brother (Kenton) for all the adventurous times we have had, both good and bad, and for keeping my rugged outdoor spirit alive. Finally, I thank my soul mate, my wi fe, Wimberley Kay, who has been there through thick, thin and endless hour s in the lab. You have helped me keep it all together even when I felt it was all falling apart.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES.............................................................................................................x ABSTRACT....................................................................................................................... xi CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................4 Differences Between Temperate and Tropical Forages................................................4 Photosynthetic Pathways and Nomenclature.........................................................4 Cell Wall Components that Impede Digestion......................................................7 Differences in Cell Wall Composition of C3 and C4 Grasses..............................10 Improving the Nutritive Value of C4 Grasses.............................................................11 Chemical Treatments..................................................................................................12 Ammoniation.......................................................................................................12 NaOH Treatment.................................................................................................14 Enzyme Treatment......................................................................................................16 Cellulase..............................................................................................................17 Xylanase..............................................................................................................17 Ligninase.............................................................................................................18 Ferulic Acid Esterase Enzymes...........................................................................19 Determining Enzyme Activity.............................................................................22 Factors Affecting Enzyme Action.......................................................................24 Mode of Enzyme Action.....................................................................................25 Effect of Enzymes on Pre-I ngestion Fiber Hydrolysis........................................26 Effect of Enzyme on Post-inges tion DM and Fiber Digestibility.......................27 Effect of Enzyme Treatment on Beef Cattle Performance..................................29

PAGE 6

vi 3 THE POTENTIAL FOR INCREASING THE DIGESTIBILITY OF THREE TROPICAL GRASSES WITH A FUNGAL FERULIC ACID ESTERASE ENZYME PREPARATION.......................................................................................33 Introduction.................................................................................................................33 Materials and Methods...............................................................................................34 Enzymes and Forages..........................................................................................34 In Vitro Digestibility...........................................................................................35 Chemical Analysis...............................................................................................35 In Situ Rumen Degradability...............................................................................36 Statistical Analysis..............................................................................................37 Results........................................................................................................................ .38 Chemical Composition........................................................................................38 In Vitro Disappearance a nd Digestibility Study..................................................39 In Situ Study........................................................................................................40 Discussion...................................................................................................................41 Implications................................................................................................................44 4 EFFECT OF FIBROLYTIC ENZY ME PREPARATIONS CONTAINING ESTERASE, CELLULASE, AND E NDOGALACTURONASE ACTIVITY ON THE DIGESTIBILITY OF MATU RE, TROPICAL GRASS HAYS........................51 Introduction.................................................................................................................51 Materials and Methods...............................................................................................52 Enzymes and Forages..........................................................................................52 In Vitro Digestibility and Disappearance............................................................53 Chemical Analysis...............................................................................................54 In Situ Rumen Degradability...............................................................................54 Statistical Analysis..............................................................................................55 Results........................................................................................................................ .56 Nutritive Value of Untreated Hays......................................................................56 Cell Wall Composition........................................................................................57 In vitro Disappearance and Digestibility Study...................................................57 Enzyme Effects on In Situ Degradability............................................................58 Discussion...................................................................................................................58 Implications................................................................................................................62 5 THE EFFECT OF DIFFERENT COMBINATIONS OF FERULIC ACID ESTERASE, CELLULASE AND XYLA NASE ON THE NUTRITIVE VALUE OF MATURE BAHIAGRASS...................................................................................68 Introduction.................................................................................................................68 Materials and Methods...............................................................................................69 Forage and Enzymes............................................................................................69 Experiment 1: Enzymatic DM Disappearance....................................................69 Experiment 2: In Vitro Fe rmentation and Degradability....................................70 Statistical Analysis..............................................................................................72

PAGE 7

vii Results........................................................................................................................ .73 Experiment 1: Enzymatic DM Disappearance....................................................73 Experiment 2: In vitro Fe rmentation and Digestibility.......................................73 Discussion...................................................................................................................74 Implications................................................................................................................77 6 EFFECT OF APPLYING AMMONI A OR FIBROLYTIC ENZYMES TO BERMUDAGRASS HAY ON FEED INTAKE, DIGESTION KINETICS AND GROWTH OF BEEF STEERS..................................................................................85 Introduction.................................................................................................................85 Materials and Methods...............................................................................................86 Forage Treatments...............................................................................................86 Cattle and Diets...................................................................................................87 Sampling and Analysis........................................................................................88 In Situ Rumen Degradability...............................................................................90 Statistical Analysis..............................................................................................90 Results........................................................................................................................ .92 Forage Composition............................................................................................92 Animal Performance............................................................................................92 In Situ Rumen Degradability...............................................................................93 Discussion...................................................................................................................93 Implications................................................................................................................95 7 GENERAL SUMMARY, CONCLU SIONS AND RE COMMENDATIONS.........108 APPENDIX A DESCRIPTION OF PHENOLIC ACID ANALYSIS..............................................114 B SAS CODE USED FOR CHAPTERS 3 AND 4......................................................116 C SAS CODE USED FOR CHAPTER 5.....................................................................120 D SAS CODE USED FOR CHAPTER 6.....................................................................123 LIST OF REFERENCES.................................................................................................126 BIOGRAPHICAL SKETCH...........................................................................................138

PAGE 8

viii LIST OF TABLES Table page 3-1. Effect of enzyme application on ce ll wall and water solubl e carbohydrate (WSC) concentration of tropical grass hays, g/kg................................................................46 3-2. Effect of D740L application on the free ferulic acid and p-coumaric concentration of tropical gr ass hays, mg/g cell wall................................................47 3-3. Effect of D740L app lication on the in vitro 6-h, 24-h and 48-h rumen fluid DM disappearance and 96 h rumen-fluid pepsin digestibility of tr opical grass hays, g/kg DM...................................................................................................................48 3-4. Effect of D740L application on in vitro NDF disappearance of tropical grass hays, g/kg DM..........................................................................................................49 3-5. Effect of D740L applic ation on the in situ Dm degradability of tropical grass hays........................................................................................................................... 50 4-1. Chemical composition of ha ys (g/kg DM or as stated).............................................63 4-2. Effect of enzyme application on ce ll wall and water solubl e carbohydrate (WSC) concentration of tropical grass hays, g/kg................................................................64 4-3 Effect of D670L application on the 6-, 24and 48-h rumen fluid disappearance and 96-h rumen-fluid pepsin digestibility of tropical grass hays, g/kg DM............65 4-4. Effect of D670L application on in vitro NDF disappearance (g/kg) of tropical grass hays.................................................................................................................66 4-5. Effect of D670L applic ation on the in situ rumen DM degradability of tropical grass hays.................................................................................................................67 5-1. Effect of multienzyme cocktail composition on DM disappearance of bahiagrass hay..........................................................................................................78 5-2. Effect of enzyme treatment on con centration of volatile fatty acids produced from fermentation of bahiagrass hay in rumen fluid/ McDougals buffer for 24 h (molar %)..................................................................................................................79

PAGE 9

ix 5-3. Effect of enzyme treatment on con centration of volatile fatty acids produced from fermentation of bahiagrass in rumen fluid for 96 h (molar %)........................80 5-4. Effect of enzyme treatment on dry matter digestibility (%) of bahiagrass after 24 or 96 hr of incubation in rume n fluid and McDougals buffer..................................81 5-5 Effect of enzyme treatment on neutra l detergent fiber digestibility (%) of bahiagrass.................................................................................................................82 5-6 Effect of enzyme treatment on fermenta tion kinetics of bahiagrass incubated in rumen fluid for 24 h.................................................................................................83 5-7 Effect of enzyme treatment on fermenta tion kinetics of bahiagrass incubated in rumen fluid for 96 h.................................................................................................84 6-1 Nutrient composition of the diet fed to beef steers......................................................97 6-2. Chemical composition of the treated and untreated hays (n=6)................................98 6-3. Effect of treating bermudagrass hay with ammonia or enzymes on the voluntary intake of steers..........................................................................................................99 6-4. Effect of additive treatment on the in vi vo apparent digestibility of diets in steers.100 6-5. Effect of additive treatment on the performance of steers.......................................101 6-6. Effect of additive treatment on blood metabolites, fat depositi on and rib eye area in beef steers...........................................................................................................102 6-7. Effect of additive treatment on the in situ digestion kinetics.................................103

PAGE 10

x LIST OF FIGURES Figure page 2-1. C3 photosynthetic pathwaya (adapted from Lawlor, 2001).........................................5 2-2. C4 photosynthetic pathwaya (adapted from Lawlor, 2001).........................................6 6-1. Effect of ammonia or enzymatic trea tment of bahiagrass on dry matter intake (DMI) of growing steers (t reatment*week interaction P < 0.0001). Asterisks denote significant differences among treatments (P < 0.05)..................................104 6-2. Effect of treatment on body weight (BW) changes during the trial by week (treatment*week interaction P = 0.9816), where asteri sks denote significant differences among treatments (P < 0.05)...............................................................105 6-3. Net energy (NE) supplied by the diet components relative to NE requirements (NRC, 2000) for maintenance (NEm) and gain (NEg) of steers............................106 6-4. Metabolisable protein (MP) supplied by the diet component s relative to MP requirements for maintenance (MPm) and gain (MPg) of steers...........................107

PAGE 11

xi Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFECT OF FIBROLYTIC ENZYME S ON THE NUTRITIVE VALUE OF TROPICAL FORAGES AND PERFORMANCE OF BEEF STEERS By Nathan Alfred Krueger May 2006 Chair: Adegbola Adesogan Major Department: Animal Sciences Three experiments were conducted to evalua te the effect of applying fibrolytic enzymes to tropical forages on their nutritive va lue. An additional study investigated the effects of enzyme application on beef cattle performance. Experiment 1 determined the effects of using Depol 740L (D740) on the dige stibility of three tropical forages. The enzyme was applied to 12-wk regrow ths of Pensacola bahiagrass (BAH; Paspalum notatum ), Coastal bermudagrass (C-B; Cynodon dactylon ) and Tifton 85 bermudagrass (T-B; Cynodon sp. ) at rates of 0, 0.5, 1, 2, and 3 g/100g DM. The enzyme enhanced the chemical composition, in vitro dry matter digestibility (IVDMD), in vitro neutral detergent fiber digestibility (IVNDFD) and in situ degradability of the three hays, particularly the bahiagrass. Experiment 2 examined the effect of applying Depol 670L (D670) on the same forages as in Exp. 1. En zyme D670 enhanced the initial phase of digestion of C-B and BAH, and slightly (2-4 %) increased the extent of digestion of BAH and T-B. Experiment 3 evaluated the eff ects of the method of applying a commercial

PAGE 12

xii fibrolytic enzyme, Biocellulase A20 (A20), or ammonia to bermudagrass hay on the feed intake, digestion kinetics and growth performance of fifty Angus-Brangus steers. Fiveweek fall regrowth of bermudagrass was ha rvested and treated with either anhydrous ammonia (3 g/100g DM) or A20. The A 20 was applied at the manufacturers recommended rate of 15 g/ton of hay imme diately after harvesting (Ec), immediately before baling (Eb) or immediately before f eeding (Ef). The hays were supplemented with cottonseed meal and soy hulls at 1% of bodyw eight and fed to steers for 84 d. Ammonia and Ec treatment increased overall dry matte r intake (DMI) but di d not increased the DMI as a percent of BW, whereas Ec also in creased neutral detergent fiber (NDF) intake, but none of the treatments increased the growth performance of the steers. These experiments suggest that certain fi brolytic enzymes can improve the nutritive value of mature tropical ha ys but the response depends on forage species. An enzyme preparation was also found to be as effectiv e as ammonia at increas ing the quality of 5wk regrowth of bermudagrass hay.

PAGE 13

1 CHAPTER 1 INTRODUCTION Cattle production in the United States (US) is heavily reliant on the use of forages, which are grazed or conserved as hay or silage. In the northern parts of the United States and temperate countries where cool-season (C3) perennial grasses abound, the efficiency of conversion of consumed forage to body weight or milk is relatively low. However, this conversion is far less efficient in th e southern US and tropical and subtropical countries due to the low qua lity of the perennial C4 grasses that are adapted to these climates. Such C4 grasses are the staple diet of most of the domesticated ruminants in the tropical and subtropical regions of the world. In fact, these grasses comprise as much as eighty-five percent of the worldsÂ’ feed supply for meat, milk, and fiber production (Coleman et al., 2004). In Florida, all beef cat tle operations rely on forages as the primary source of nutrients (FASS, 2000). The total dietary inclusion rate of C4 grasses is up to 60% for lactating dairy cattle and up to 100% for beef cattle on range. The National Research Council (NRC; 1996) indicates that low quality forages like bahiagrass and bermudagrass canÂ’t meet the energy and prot ein needs of growing calves or the net energy requirements of pregnant replacement he ifers from the time of breeding to calving . This is due to the lower digestion and chemical composition of such C4 forages. Therefore, there is a pressing need to im prove the quality of Florida forages. The poor quality of these grasses has been at tributed largely to their relatively high lignin and phenolic acid concentrations (Jung and Allen, 1995). Chemical treatment has been used successfully to incr ease the feeding value of low quality forages (Brown, 1988;

PAGE 14

2 Brown and Adjei, 1995; Brown and Kunkle, 1992; Brown and Pate, 1997; Klopfenstein 1978; Kunkle et al., 1983 and 1984; Rasby et al ., 1989). Ammonia treatment is one of the most widely used chemical treatment methods. Ammonia treatment increases the crude protein (CP) concentra tion and decreases the neutral detergent fiber (NDF) fraction of forages. The latter is due to disruption of cross-linkages between the hemicellulose and lignin fractions of the forages. Feedi ng ammonia-treated forage often results in increased daily intake and av erage daily gain (ADG) by beef cattle (Brown and Kunkle, 1992; Brown and Pate, 1997). However, ther e has been limited adoption of ammonia application due to lack of in frastructure for delivering anhy drous ammonia, human health hazards posed by the alkali and the corrosive effects of ammonia on equipment (Lalman et al., 1997). In recent years, the use of exogenous fibrolytic enzymes to improve feed utilization has received a gr eat deal of research intere st (Beauchemin et al., 2003). Studies have shown increases in apparent DM digestion in situ and in vivo (Feng et al., 1996; Yang et al., 1999) and volunt ary intake (Feng et al. 199 6; Yang et al., 1999; PinosRodriguez et al., 2002) of forages. Howe ver, while some studies have reported improvements in animal performance due to treatment with enzymes (Feng et al., 1996; Beauchemin et al., 1999) others have not (Michal et al., 1996 ; ZoBell et al., 2000). These inconsistencies can be attributed to differen ces in enzyme type, pr eparation, application method and rate (Bowman et al ., 2002; Beauchemin et at., 2003), as well as the mode of enzyme action and the type of forage or con centrate to which the enzyme is added (Feng et al., 1996; Lewis et al., 1996; ZoBell et al ., 2000). Nevertheless, enzymes have been

PAGE 15

3 used to improve the utilization of diets c ontaining legumes, hays, crop residues and other feedstuffs (Beauchemin et al., 2003). Recent work suggests that cross linkages formed between ferulic acid and lignin limit microbial access to the digestible xylans in the cell walls of grasses (Hatfield, 1993; Jung and Deetz, 1993; Hatfield et al., 1997; Casler and Jung, 1999; Gr abber et al., 2000). Hatfield et al. (1997) also de monstrated that the combination of such cross-linkages and high lignin concentrations greatly redu ced the in vitro digestibility of C4 grasses. The cross-linkages are also thought to cause cell wall stiffening and growth cessation in plants (Musel et al., 1997). Ferulic acid esterase (FAE) enzymes can cleave off ferulic acid from the cell wall xylans, maki ng them more readily digestible by ruminal microbes. In C3 grasses and maize bran, fungal ferulic aci d esterase enzymes have been shown to release ferulic acid from the cross linkages and render the digestible xylans in the cell wall more susceptible to enzymatic degradati on (Faulds et al., 1995; Kroon et al., 1999). However, the majority of the research done to date on feed or forage treated with ferulic acid esterase has been carried out on temperate forage species. Therefore, little is known about effects of esterase enzymes on tropical forage quality. Therefore, this series of experiments were initiated to evaluate th e potential of using fibrolytic enzymes, especially FAE, to enhance the quality of C4 forages. The aim of these experiments was to determine the effect of different applic ation rates or mixtures of FAE and other fibrolytic enzymes on the digestibility of tropical grass hays. A further aim was to evaluate the effect of method of applying a commercial enzyme mixture to bermudagrass hay on the performance of beef cattle. .

PAGE 16

4 CHAPTER 2 LITERATURE REVIEW Differences Between Temperate and Tropical Forages Forages are the mainstay of livestock production worldwide. In the United States, forage inclusion rate in diets ranges from 40-60% in lactating dairy rations and up to 100% in beef cattle rations. Most of what is considered to be forage is, in fact, plant cell walls which are made up of polysaccharides, protein and lignin (Nevins, 1993). These plant cell walls serve as a major source of energy for ruminant animals (Aman, 1993). The dry matter (DM) digestibility and intake which determine energy supply are considerably lower in tropical grasses than in temperate grasses (Minson, 1980). This is mainly due to differences in leaf anatomy and chemical composition between the temperate (C3) and tropical (C4) species, with higher proporti ons of the less digestible tissue types being found in the C4 species (Akin and Burdick, 1975). Photosynthetic Pathways and Nomenclature The photosynthetic pathways associated with C3 and C4 grasses are different (Figures 2-1 and 2-2). In C3 grasses (Figure 2-1) CO2 enters the leaf through the stoma and diffuses into the mesophyll where rubisc o catalyzes the car boxylation of ribulose1,5-bisphosphate to form two phosphoglycerate (PGA) molecules. Phosphoglycerate is a three-carbon compound and th is is where the term C3 originates. The PGA molecules enter the Calvin cycle where th ey are converted to sugars th at are then transported to growing leaves, stems, reproduc tive structures and roots. Ribulose 1,5-bisphosphate, or

PAGE 17

5 Rubisco can also catalyze the oxidation of PGA resulting in p hotorespiration of the plant, thus releasing th e already fixed CO2. Figure 2-1. C3 photosynthetic pathwaya (adapted from Lawlor, 2001). a RuBP = Ribulose 1,5-bisphosphate ; 3PGA = 3 carbon phosphoglycerate On the other hand, C4 grasses employ two cell type s (Figure 2-2), the mesophyll and the bundle sheath, to obt ain the same results as C3 plants. Atmospheric CO2 enters through the stomata and diffuses into th e mesophyll tissue where it is fixed by phosphoenolpyruvate carboxypeptidase (PEPase) to form oxaloacetate, which is converted to malate or aspa rtate (4-carbon acids) and transp orted into the bundle sheath cells, hence the name C4. Once in the bundle sheath cells , the acid is decarboxylated and the CO2 is released and then is refixed by rubi sco in the bundle sheath and the Calvin cycle proceeds as it does in the C3 plants. Since PEPase has a higher affinity for CO2 than rubisco it is not inhibited by O2 as in the initial C3 pathway carboxylation step. Therefore, only CO2 is shuttled to the bundle shea th cells which increases the concentration of CO2 in the cell. This increased concentration causes rubisco to be supersaturated with CO2 virtually eliminati ng photorespiration.

PAGE 18

6 Figure 2-2. C4 photosynthetic pathwaya (adapted from Lawlor, 2001). a PEP = phosphoenolpyruvate; RuBP = Ribul ose 1,5-bisphosphate; 3PGA = 3 carbon phosphoglycerate Anatomical Differences Compared to C3 grasses, C4 grasses generally have a higher cell wall and lignin concentration and lower DM digestibility (Wilson and Hacker, 1987). The digestion of leaves in C4 species can be influenced by their uni que leaf structure, in particular the “Kranz”, wreath-like structur e around the vascular bundle (Wilson, 1993). This “Kranz” structure binds the epidermis to the vascular bundles and thus impe des their digestion. The Kranz structure is absent in C3 grasses and the epidermis is linked to vascular tissue only by thin-walled, easily-digested mesophyll cells (Wilson et al., 1989 ). In addition, unlike C4 grasses, C3 grasses have loosely-bound va scular bundles that contain intercellular spaces and air pockets between them and the mesophyll cells, allowing for more rapid breakdown and digestion by micr obes. Byott (1976, as cited by Wilson, 1993) stated that the air space in C3 grasses (10 to 35%) is much greater than that in C4 grasses (3 to 12%). This increased air space allows for a faster penetration of bacteria into the leaf structure resulti ng in faster digestion of C3 leaf tissue. In addition, Wilson and Hattersley (1989) reported th at differences between leaf individual structural groups

PAGE 19

7 were found consistently, with C3 grasses having greater propor tions of mesophyll (53 to 67 vs. 28 to 47%), lower proportions of bundle sheath cells (5 to 20 vs. 12 to 33%) and lower portions of vascular tissue (3 to 9 vs. 6 to 12%) than C4 grasses, such that greater proportions of thin-walled, non-ligni fied cells are present in C3 grasses. Stems tend to make up a large proportion of the rumen mat. The stems have 2 to 3 times more lignified, thick-walled vascular tissue and sclerenchyma than the leaves (Wilson, 1991). The epidermis, sclerenchyma, an d vascular tissues in the outer parts of the C4 plant stems develop qui cker than those in C3 species (Wilson, 1991). They also develop thicker and more lignified cell walls and these cell walls tend to fuse together as the stem matures to a greater extent than in C3 species (Wilson, 1991). The tissues of C4 species contain no intercellular spaces and the individual cells are linked tightly together by a highly lignified middle lamella and primary wall structure that is indigestible (Cone and Engels, 1990). Such plants contain much mo re lignified tissue in both the leaves and stems than C3 plants (Harbers, 1985). Cell Wall Components that Impede Digestion The nutritive value of C3 grasses is generally gr eater than that of C4 grasses. For example ryegrass ( Lolium multiflorum )at the boot-head stage, contains approximatley 47 to 53% neutral detergent fiber (NDF), 33 to 39% acid detergent fiber (ADF) and 8 to 12% crude protein (CP) on a dry matter (DM) basis, whereas C4 grasses, like 8-wk regrowths of bermudagrass, contain 70 to 75% NDF, 40 to 45% ADF and 6 to 8% CP on a DM basis (Ball et al., 2002) . These differences are la rgely due to the chemical composition and structural differences between C3 and C4 grasses. Mandebvu et al., (1999b) evaluated two cultivars of bermudagra ss, Coastal and Tifton 85, harvested after 2 to 7 wk of regrowth and found that for bot h cultivars, concentrations of DM, ADF and

PAGE 20

8 lignin increased with maturity whereas in vitro DM digestibility (IVDMD) decreased. The increase in lignin concentration with maturity likely hindered contact between fibrolytic enzymes in the ruminal fluid a nd cell wall polysacchari des (Mandebvu et al., 1999b). Lignin acts as a physical barri er to microbial digestion of fiber polysaccharides. Lignin is a complex structure th at is made up of three units : guaicyl units derived from coniferyl alcohol; syringyl units derived fr om sinapyl alcohol and p-hydroxyphenol units which make up a minor component of the lignin structure and are derived from pcoumaryl alcohol. These lignin units are highly interconnected and condensed together resulting in bonds that are resistant to chemi cal degradation (Grabber et al., 2004). As a plant matures the most recently deposited polysaccharides of the secondary cell wall are not lignified and the primary cell wall region is the most lignified (T erashima et al., 1993; Jung and Allen, 1995). The composition of th e lignin changes from a predominantly guaiacyl type to lignin that is high in syringyl units, resulting in lower digestibility of mature cell walls as opposed to immature cell walls (Jung and Allen, 1995). As cell wall lignification progresses from the primar y cell wall to the secondary cell wall, arabinoxylan ferulate esters of the primary cell wall begin to cross-link xylan to lignin (Iiyama et al., 1990). As the forage matu res and lignin concentr ation increases, the arabinoxylan esterified ferulates become ethe rified to the lignin compound forming cross links between lignin and the ce ll wall polysaccharides (Iiyama et al., 1990). More than 30 yr ago, Hartley (1972) reporte d that the presence of these phenolic acids were related to forage digestibility in some way. Ea rly work on free phenolic acids suggested that p coumaric acid esters were more likely to aff ect digestion than ferulate esters (Hartley,

PAGE 21

9 1972). However, more current research has shown that the majority of the p -coumeric acid is ester-linked to the lignin (Jung and Deetz, 1993) and p-coumaric ethers are only linked to lignin (Jung and Allen, 199 5), therefore they are unlikel y to affect digestion. Casler and Jung (1999) repor ted that advances in cell wall chemistry have identified ferulic acid (ligni n precursors) cross-linkages as key factors causing variation in digestibility of the C3 grasses. According to Faulds et al. (1995), ferulic acid is the most abundant hydroxycinnamic acid in the plant world. Hence, ferulates and diferulates have been recognized as important structural components of cell walls of certain plants, particularly grasses (Kroon et al., 1999). Ferulic acid is linked to the 5 C hydroxyl of L-arabinose moieties of grass xylans in th e cell wall (Kato and Nevins, 1985; MuellerHarvey et al., 1986; Grabber et al., 2000). Over 50% of the cell wall ferulates can undergo dehydrodimerization which results in the forming of a wide array of 8-coupled diferulates and 5-5` coupled diferulates (R alph et al., 1994; Grabbe r et al., 1995; 2000). These dimers are formed at the onset of lignification when the generation of hydrogen peroxide is initiated to form the per oxide-mediated coupling of p-hydroxycinnamyl alcohols which results in strong polysaccharid e-polysaccharide cross-links (Ralph et al., 1994). Arabinoxylans become extensively cro ss-linked by ferulate dimerization as well as by the incorporation of ferulate monome rs and dimers into the lignin structure (Grabber et al., 2004). This re duces the digestibili ty of cell wall polys accharides. These ferulate cross-linkages are al so thought to be the reas ons behind wall stiffening and growth cessation in plants (Musel et., al . 1997; Grabber et al., 1998). Ferulate monomers and dimers bound to arabinoxylans al so serve as initiation sites for lignin formation via ester linkages and they cross-li nk lignin to polysaccharid es in cell walls via

PAGE 22

10 ester and ether linkages (Jacque t et al., 1995; Grabber et al., 2002). Ralph et al. (1995) demonstrated that ferulates covalently bind to lignin monomers via coniferyl and sinapyl alcohol residues, resulting in the formati on of strong polysacchar ide-lignin cross-links (Ralph et al.., 1994; Grabbe r et al., 1995; 1998; 2000) that limit the enzymatic degradation of cell walls (Hatfiel d, 1993; Jung and Deetz, 1993). Differences in Cell Wall Composition of C3 and C4 Grasses Jung and Vogel (1986) evaluated two C3 and six C4 forage species and reported that as the forages matured there were incr eases in cell wall concentration and decreases in the digestibility of the forages, further noting that the C4 grasses had greater concentrations of cell wall material and that this material appeared to be less digestible at all stages of maturity. Additionally, the ligni n concentration in the cell wall material was greater in C4 species than in C3 species. Wilson and Hacker (1987) also noted lower total cell wall and lignin concentrations in the C3 species while evaluating C3 and C4 plants under various growing conditions. The concentr ations of phenolic acids have been found to be greater in C4 grasses than in C3 grasses (Jung and Deet z, 1993). Akin (1986) reported that C4 grasses contained greater concentr ations of phenolic compounds like pcoumaric acid and greater ratios of p-c oumaric acid: ferulic acid than did C3 grass species, suggesting that these compounds may be responsible for limiting the extent of degradation in C4 grasses. Most of the research on ferulate cross-linkage effects on digestibility focused on C3 forages. However, Mandebvu et al. (1999b) found that greater ester and ether-linked ferulic acid concentratio ns in coastal bermudagrass, versus Tifton 85 (T-85) bermudagrass explains the poorer di gestibility of the former. Not only are there phenolic acid differences between C3 and C4 grasses (Akin, 1986) but these differences exist within cult ivars (Mandebvu et al. 1999b).

PAGE 23

11 Improving the Nutritive Value of C4 Grasses The main goal of a forage production sy stem should be efficient, cost-effective production of high quality forage. In certain pa rts of the United St ates and tropical and subtropical countries, this is not always possi ble due to prevailing climates that favor the C4 pathway. During the winter in Florida and the dry season in trop ical and subtropical countries the quality and quantity of the forage can be a major factor limiting livestock production. Supplementation of the forage is often costly and it is impractical in many cases. Supplementation also results in a ne t incorporation of nutrients into livestock operations, which can be undesirable from an environmental standpoint. An alternative method of improving animal production is to impr ove the quality of the forage being fed. Numerous chemical and biological treatments that aim to increase the nutritive value of forages have been evaluated. These treatme nts usually attempt to modify the cell wall structure of the forage in a way that resu lts in increased intake and digestibility. Treatments with alkalis, such as NaOH or a mmonia, physically disrupt and chemically alter the cell wall structure (Turner et al ., 1990). The chemical alterations include hydrolysis of the lignin-hemicellulose matrix which is made up of phenolic acids that are covalently bound to hemicellu lase through ester linkages (M ueller-Harvey et al., 1986). The disruption of this matrix via the saponification of the ester linkages allows for the solubilization of phenolic compounds, hemice lluloses, and acetyl groups allowing for greater access to the cell wall components (Klopfenstein, 1978; Fahey et al., 1993). Treatment of low quality forages has successfu lly increased fiber digestion, intake and animal performance (Klopfenstein, 1978; Kunkle et al., 1983, 1984; Brown, 1988; Rasby et al., 1989; Brown and Kunkle, 1992; Brown and Adjei, 1995; Vagnoni et al., 1995; and Brown and Pate, 1997).

PAGE 24

12 Chemical Treatments Ammoniation One of the most widely used forms of chemical treatment of forages involves treatment with ammonia. When gas or liqui d anhydrous ammonia comes in contact with roughage, it combines with the moisture in the roughage to produce gaseous ammonium hydroxide. Ammonium hydroxide is an alkaline compound, which solubilizes hemicellulose by breaking the chemical bonds holding lignin and hemicellulose together (Lalman et al., 1997). Ammonium hydroxide al so partially breaks down the structure of cellulose by disrupting hydrogen bonds. Essentially the reac tion causes a swelling of the fiber and allows microbial cellulase to access the fiber for digestion (Church, 1988). According to Chinh et al. (1992), ammonia app lication has two functi ons: an increase in digestibility by the disruption of lignin-hemicellulose linkages and an increase in feed intake resulting from the greater supply of n itrogen to the rumen mi crobes. Studies have also reported that there is a 7.4 to 11.5 percentage unit increase for bahiagrass and 8.4 to 11.5 percentage unit increase for bermudagrass in CP concentration of the forage due to ammonia treatment (Brown, 1993, 1994). Howeve r, some of the added nitrogen from ammonia is not utilized by the ruminant animal and is either recycl ed or excreted (Rasby et. al., 1989). Ruminant microorganisms can use some of the N supplied by ammoniation because the treatment also increases energy supply to microbes from cell wall hydrolysis (Rasby et. al., 1989). Treating hay with ammonia also has non-nut ritional benefits. Cattle waste less ammoniated hay than non-treated hay (10% vs. 25% loss respectiv ely) and ammoniation allows proper conservation and utilization of wet (25 to 60% moisture) forages that would otherwise deteriorate ra pidly due to mold growth (Brown and Kunkle, 1992).

PAGE 25

13 However, forages with moisture concentra tions above 30% are not recommended to be treated with anhydrous ammonia due to likely decreased intake s of the forage. Ammonia is very hydrophilic and high moisture forage s will retain greater amounts of the ammonia resulting in a strong ammonia odor thus d ecreasing intake (Brown and Kunkle, 1992). Ideally, ammoniation should be restricted to mature, low quality forages (Lalman et al., 1997). Ammoniation of high quality forages like alfalfa ( Medicago sativa) and smallgrain hays can lead to toxicity due to consumption of the toxic compound 4methylimidazol (Jung and Allen, 1995), which is formed when the ammonia reacts with soluble sugars in the forage. The sympto ms of this condition, commonly called “crazy cow syndrome” or “bovine bonkers,” include circling, hyperexcitability and convulsions, and it can even lead to death. However, ma ture forages are low in soluble sugars and therefore pose little ri sk of causing ammonia toxicity (Lal man et al., 1997). Different methods exist for of applying a mmonia to forage. These include the stack method, high-pressure treatment, and combinati ons with other chemicals such as NaOH. The most commonly used and most practic al method would be the tarp method. The major disadvantages of ammonia treatment incl ude the caustic effect of ammonia when it is inhaled or ingested exce ssively by humans and animals, the corrosive effects of ammonia on equipment, and the requiremen t for specialized handling and delivery systems. A recent disadvantage is the curr ent high cost of ammoniation due to its synthesis from natural gas which has beco me very expensive in recent months. Ammonium hydroxide also has been applie d directly to forage producing similar results found with anhydrous treatments, but is also caustic and corro sive. Urea is a less widely used, safer method of ammonia treatment but there are several drawbacks to its

PAGE 26

14 utilization. Firstly, NH3 is released from urea only after the urea has been mixed with the forage in a silo or some other type of struct ure. The amount of urease and moisture in the forage will determine the efficacy of the treatm ent since they are both needed to cause the formation of NH3 from the urea. Several studies have demonstrated increas ed performance of animals due to forage ammoniation. Kunkle et al. ( 1983) reported that treating poo r quality, tropical forages with anhydrous ammonia (3 to 5%) increased the CP (4.2 to 11.7%) and in vitro DM digestibility of the forage (41 to 52%). Th ese researchers also showed that non-lactating beef cows consuming unsupplemented ammonia-treated forage performed as well as cows fed untreated forage plus 16% crude pr otein liquid supplements. Brown and Adjei (1995) conducted two digestion and growth trials over a two-year period, during which they sprayed round bales of hay with solutions of 0, 4, or 6% urea and fed them to beef steers. Feed intake, apparent NDF and ADF digestibility ( P < 0.05), average daily gain (ADG; P < 0.05) and gain to feed ratio ( P = 0.07) increased with increasing urea concentration. In a growth and metabolism tr ial in which steers were fed ammoniated (3% of DM) mature bahiagrass hay or a urea supplement, DM intake (DMI), digestion kinetics and ADG were increased ( P > 0.05) by ammoniation whereas urea supplementation had no effect (Vagioni et al., 1995). While ammonia treatment is effective at improving forage quality and anim al performance, the caustic nature of NH3 necessitates the use of prot ective clothing and causes excessive wear and rusting on equipment. These factors have limited the adoption of ammoniation by producers. NaOH Treatment Another widely used chemical treatmen t for forages involves using NaOH. Two different methods have been used to a pply NaOH to forages. The earlier wet

PAGE 27

15 (Beckmann) method entailed soaking the fora ge in a dilute solution of NaOH over several days. The forage then was washed to remove the NaOH residue. According to Sundstol (1988) this method increase d the OM digestibility of rye ( Secale cereale ) straw from 52.0 to 76.0 %. However, there are two ma jor drawbacks to the procedure. First, the wastewater is contaminated with NaOH and is considered a threat to the environment. Secondly, there is a cons iderable amount of digestible DM lost due to the rinsing of the treated forage prior to feeding. These draw backs led to the development of the dry treatment method. The major difference betwee n the two procedures is that in the dry method a dilute amount of NaOH is sprayed ont o the forage and the forage is not rinsed prior to feeding. The major a dvantages are that there is less labor involved in the dry method and the problem of wast ewater disposal doesnÂ’t aris e. However, there is an increased possibility of toxicity if the samp les are not uniformly treated in the dry method because excess NaOH is not removed by washing. Other chemicals have been researched as possible forage treatment agents including calcium and potassium hydroxides. In theory both of these compounds have potential as treatments and they respectively provide Ca a nd K to the animal (Berger et al., 1994). However, calcium hydroxide is ineffective as a stand-alone treatment for improving digestibility, and potassium hydroxide is too costly to be an economically feasible treatment method (Owen et al., 1984 in Berger et al., 1994). Thes e problems, along with the caustic nature of these chemicals have prevented their adoption. With the evergrowing focus on preserving the environment, chemical treatments are becoming more and more unfavored due to concerns that th ey can damage the environment. This

PAGE 28

16 emphasizes the need to find safer, more e nvironmentally-friendly methods for improving forage quality. Enzyme Treatment Enzymes are naturally occurring, benign and environmentally friendly protein molecules that catalyze specific chemical r eactions. Enzymes with specific fibrolytic activities can be specifically selected and pr oduced in mass quantities to meet the needs for forage improvement. Digestive enzymes are utilized by the animal as a means to break down complex feeds into simple sugars, animo acids and fatty acids (Kung, 2001; McAllister et al., 2001). Enzymes have been used in monogastric diets for decades, but concerted efforts to integrate enzymes into rumi nant diets started within the last decade. The enzymes included in ruminant diets are exogenous fibrolytic enzymes that degrade plant cell walls into simple sugars. The most widely evaluated enzymes are those that are effective at degrading the cel lulose and hemicellulose in the cell wall and these are referred to as cellulases and xylanases, resp ectively. The use of exogenous fibrolytic enzymes to improve feed utilization has rece ived a great deal of research interest. Studies have reported increases in in situ and in vivo DM digestion (Feng et al., 1996; Yang et al., 1999) and voluntary intake (F eng et al., 1996; Yang et al., 1999; PinosRodriguez et al., 2002), although some st udies show no improvements in animal performance (Michal et al., 1996; ZoBell et al ., 2000). The variability and inconsistencies in the results can be attributed to enzyme type, enzyme preparation, application method and quantity of enzyme used (Bowman et al., 2002; Beauchemin et al., 2003) as well as the fraction of the diet to wh ich the enzyme is applied (Fe ng et al., 1996; Lewis et al., 1996; ZoBell et al., 2000).

PAGE 29

17 Cellulase The cellulases and xylanases used commercially are typically produced in a batch type fermentation system using microorganisms of bacterial ( Bacillus spp.) or fungal ( Aspergillus spp. and Trichoderma spp.) origin (Pendleton et al., 2000 as cited by Beauchemin et al., 2003, 2004). Cellulose can be hydrolyzed through a process using various cellulases. The three major categor ies of cellulases include endogluconases, exogluconases and -glucosidases (White et al., 1993; Beauchemin et al., 2003; 2004). In general endogluconases hydrolyze cellulose chains randomly to produce oligomers with varying degrees of polymerizati on; exogluconases produce cellobiose by hydrolyzing the cellulo se chain from the non-reducing end; and -glucosidases release glucose from cellobiose and from the hydrolys is of short-chained cellulose oligomers (Bhat and Hazelwood, 2001; Beauchemin et al., 2003). Xylanase Xylans are linear polymers of xylose linked together by -1,4 bonds with varying oligosaccharide side chains of arabinose, mannose, rhamnose or glucuronic acid (White et al., 1993). The core polymer of xylan is degraded to soluble sugars by two main enzymes: xylanase and -1,4 xylosidase (Bhat and Hazelwood, 2001). Xylanases include endoxylanases, which yiel d xylooligomers, while -1,4 xylosidase yields xylose. Additionally, other hemicellulase enzymes can degrade the side chains. These enzymes include -mannosidase, -D-glucuronidase, -D-galatosidase, -L-arabinosidase, acetylxylan esterases and ferulic acid esterase (White et al., 1993; Bhat and Hazelwood, 2001). The xylan core polymer, or backbone, can al so be intertwined with cellulose via Hbonding, resulting in a highly complex structure (White et al., 1993). Murashima et al.

PAGE 30

18 (2003) demonstrated that there were synerg istic effects between xylanase and cellulase enzymes on corn cell wall degr adation. A xylanase:cellulase molar ratio of 1:2 was found to successfully degrade corn cell wa lls resulting in increased release of xylooligosaccharides and cellooligosaccharid es, but it had no effect on pure xylan or crystylline cellulose substrates. Murashima et al. (2003) postulated that the degradation of xylans in the complex structure by xylanas es could allow cellulases access to degrade embedded cellulose. Additionally, the degrada tion of cellulose in the complex structure by cellulases might help xylanase to access and degrade xylan, thus explaining the synergistic effects found in their study between xylanases and cellulases. Recent awareness of the inhi bition of cell wall digestion by cross linkages of lignin and hemicellulose via ferulic acid bridges (Hartley, 1972; Jung and Allen, 1995) has sparked a growing interest in evaluating strategi es for hydrolyzing these cross-linkages with ligninases and ferulic acid esterases. Ligninase Ligninases have been utilized by the wood and paper industries for years. However, there has been relatively little re search on the use of ligninases to improve forage quality. According to Akin et al. (1995), the chemistry of grass lignocellulose varies considerably from that of wood. Firstl y, there is a great deal less lignin in plants versus wood. Also phenolic acids in the plan t cell walls cross-link polysaccharides to the lignin polymers in plants. The cross linkages hold the polysaccharides in close proximity to the lignin polymer, preventing enzymatic hydrolysis of the polysaccharides (Jung and Deetz, 1993). Ligninases are derived from fungi, and the most effective fungi known to breakdown lignin are white rot f ungi. Some species of white rot fungi have been shown to improve the biodegradation of grasses by atta cking the aromatic constituents of lignin

PAGE 31

19 (Akin, 1993). These fungi secrete a variety of ligninase enzymes ( lignin peroxidases) that depolymerize hemicellulose and cellulose, and then fragment li gnin resulting in the increased degradability of the substrate by ru men microbes. This results in losses of carbohydrates from the substrate during lignina se hydrolysis. The digestibility of low quality forages like cereal stra ws have been improved by some white rot fungi, however the improvement of forage quality by ligni nases depends on the fungal species and crop type (Jung et al., 1992). Fahey et al. (1993) stated that li gnin is degraded by white rot fungi as a result of secondary metabolism and th at lignin is not a growth substrate for the fungi. This implies that the degradation of li gnin has to be linked to the degradation of an alternative carbon source like carbohydrates (Fahey et al., 1993). Such losses of potentially useful carbohydrates by ligninase have limited its use for improving forage quality. Ferulic Acid Esterase Enzymes The cross-linking of lignin with cell wall polysacchar ides through ferulic acid bridges is thought to be the mechanism by whic h lignin limits cell wall digestion in plants (Jung and Allen, 1995). Ferulic acid esterase s (FAE) are known to release ferulic acid (FA) from arabinose side chains of hemice llulose, which allows for further degradation of the cell wall by other polysacchardiases . However, only a few FAE -producing species of microorganisms are known to produc e esterases that can cleave the phenolic compound from the polysaccharide side chain (Faulds and Williamson, 1994). These species include Aspergillus spp. (Faulds and Williamson, 1994; Brezillon et al., 1996), Streptomyces olivochromogenes (Faulds and Williamson, 1993), Penicillium spp, Pseudomonas flourescens subsp. cellulosa (Ferreira et al., 1993) and Neocallimastix (Borneman et al., 1992). These microorga nisms have been shown to produce ferulic

PAGE 32

20 acid esterases when grown on substrates like methyl ferulate (Faulds and Williamson, 1994), sugar beet pulp (Brezi llon et at., 1996), oat spelt xylan (Faulds and Williamson, 1994, Brezillon et al., 1996), maize bran (Fauld s et al., 1995), wheat bran (Faulds and Williamson, 1994) and various other substrates. The activity of the enzymes produced from the microorganisms is dependant on th e chemical composition of the substrate on which it is grown. The addition of free ferulic acid (FA) to substrates like oat spelt xylan, which contains no detectable ferulic acid, can more than double the FAE secretion by the microorganism (Faulds et al., 1997). Faulds and Williamson (1994) used Aspergillus niger FAE (FAE-III) to release FA from wheat bran, but this effect was further enhanced in the presence of xylanase from Trichoderma viride and methyl ferulate. The effective enzymatic removal of ferulic acid from the cell wall of wheat bran or oat hulls ha s also been reported to require the addition of xylanase, which works synergistically with esterase to break the ferulic acid bonds within the cell wall (Ferreira et al., 1993; Faulds and Williamson, 1994, 1995; Kroon and Williamson, 1999; Yu et al., 2002a,b; 2003; 2005). Faulds and Williamson (1995) evaluated the release of ferulic acid from de-starched wheat bran (DSWB) by a FAE III from Aspergillus niger with or with out the addition of xylanase from Trichoderma viride over time. Ferulic acid was released when the substrate was incubated with FAE III alone, however, the addition of the xylanase pr oduced a 24-fold increase in the release of the ferulic acid. Additionally, the release of the ferulic acid from the DSWB had a linear relationship to the FAE III concentration between 0.1 and 1.3 U. Where one unit of activity is defined as th e release of one µmole of product/ min at 37oC. However, the addition of 0.3 U of xylanase increased the am ount of ferulic acid released up to 6-fold

PAGE 33

21 (Faulds and Williamson, 1995). Yu et al., ( 2002a) hypothesized that a two-step process occurs in order to efficiently release free fe rulic acid from complex cell walls. First, xylanases are required to solubilize part of the cell wall structure resulting in the formation of low molecular weight feruloylat ed compounds (Yu et al., 2002a). Then the ferulic acid esterase can act on these com pounds to release ferulic acid because ferulic acid esterase has a greater specificity for s hort chain xylo-oligomers (Bartolome et al., 1995; Faulds and Williamson 1995). The initial enzyme attack alters the chemical and physical properties of the cell wall, result ing in cleavage of the xylan backbone into feruloyl polysaccharides (Yu et al., 2002a, 2003) . It is further suggested that the synergistic effects of xylanas e and ferulic acid esterase on releasing ferulic acid from feruloyl polysaccharides makes the other cell wall polysaccharides more accessible for further attack by cellulase (Yu et al., 2003). Bartolome et al. (1995) used FAE III, Trichoderma viridae xylanase (Megazyme), and xylanases from Pseudimonas flourenscences subsp . cellulosa (XYLA), Talaromyces emersonii (XYL VII, and XYL XI), and Aspergillus (XYL I) to demonstrate that the release of ferulic acid from DSWB on the s ource of the xylanase. Additionally, the researchers analyzed the so luble fraction for low molecu lar weight carbohydrates (arabinose, glucose, xylose, xylobiose, xylotriose and xylotet raose) after 0.5 U submaximal xylanase treatment. In the pres ence of esterase, the low molecular weight carbohydrate profiles were qualitatively th e same but quantitatively greater, demonstrating that the presence of the este rase increased the initial rate of xylanase hydrolysis by 12 % for Megazyme and XYLA treatment, and by roughly 30 % for XYL

PAGE 34

22 VII, XI and XYL I, resulting in a more co mplete hydrolysis of DSWB. This further demonstrates the synergistic effects of xylanase and esterase enzymes. Purified feruloyl esterases from Aspergillus niger and Pseudomonas florescens released small amounts (1 to 9%) of ferulate monomers but not diferulates when added to barley and wheat cell walls (B artolome et al., 1997). Furthe r addition of xylanase caused a dramatic improvement in the release of fe rulates (~ 20% improvement) but elicited no improvement in the release of diferulates, though increased amounts of the 5-5` dimer was released. Yu et al. (2005) evaluated the effects of various multienzyme cocktails containing different comb inations of FAE from Aspergillus spp., xylanase from Trichoderma spp ., cellulase, endo-glucanase I and II and -glucanase on the nutritional value of oat hulls. A cocktail containing FAE (13 mU/assay), xylanase (4096 U/assay), cellulase (1024 U/assay) , endo-gluconase I and II (256 U/assay) and -glucanase (64 U/assay) gave the greatest increase in enzy matic DM disappearance of the hulls (86.3%) and increased the in vitro DM disappearance of oat hulls that were ground to 1 mm and 250 µm by 6.5% and 12.1% respectively. The au thors concluded that the addition of the multienzyme cocktail to poorly digestible f eeds before feeding has the potential to enhance ruminal digestion due to the hydrolysis of forage cell walls. Determining Enzyme Activity Due to the complex composition of plant ce ll walls, they can not be utilized as substrates in enzyme activity assays. Conse quently various other substrates are used, but the variety of substrates as well as the co mplexity of the enzyme systems produced by different microorganisms can confound enzyme activity results (Bhat and Hazelwood, 2001). Typically, enzyme activ ities are determined by colo rimetrically measuring the

PAGE 35

23 concentration of released re ducing sugars, per unit of time per unit of enzyme under defined temperature and pH conditions. Acco rding to Beauchemin et al. (2003), the two most commonly used methods for detecting reducing sugars are the copper method of Nelson/Somogyi (Somogyi, 1952) and the dinitr osalicyclic acid method (Miller, 1959). Total cellulase activity can be measured with crystalline cellulose, such as filter paper, where the measure is the total amount of gluc ose released and not a rate of release. Beauchemin et al. (2003) stated that the most commonly used s ubstrate for measuring cellulase activity is carboxymethyl cellulose (CMC) and this is often assumed to be a measure of total cellulase activity, which is no t correct. Endoglucanase activity is what is actually measured with CMC and it also can be measured with hydroxymethyl cellulose (Wood and Bhat, 1988). Exogluconase activit y can be measured with crystalline cellulose preparations like Avicel. ß-glucosidase activity is determined by measuring the release of glucose from cello biose, or the release of p -nitrophenol from p -nitrophenyl-ßD-glucoside (Bhat and Hazlewood, 2001). A ccording to Beauchemin et al. (2003) xylanase activity is measured most commonly by the release of reducing sugars from prepared xylan like oat spelt or birchwood xylan. Bhat and Hazelwood (2001) stated that xylanases in general are specific for the inte rnal ß-1,4 linkages within the xylan backbone structure and are designated as endoxylanases. These endoxyl anses are considered to be debranching or non-debranching enzymes base d on their ability to hydrolyze the xylan backbone and release the arab inose side chain (Beauchemin et al., 2003). ß-xylosidase activity can be determined using p -nitrophenyl derivatives. Ferulic acid esterase activity can be dete rmined by measuring the rate of hydrolysis of methyl ferulate, met hyl sinapinate, methyl p -coumerate, methyl caffeate, methyl

PAGE 36

24 vanillate and methyl syringate by high perf ormance liquid chromatography (HPLC) as described by Faulds and Williamson (1994). Activity of FAE can also be measured spectrophotometrically utilizing the di fferences between the free acid and the corresponding methyl ester (Fauld and Williamson, 1994; Ralet et al., 1994). The measurement of enzyme activity must be carried out under closely defined conditions because temperature, pH, ionic strength, substrate type, and substrate concentration all affect the activity of the enzyme (Beauchemin et al., 2003). Commercial cellulase and hemicellulase activ ities are typically m easured under optimum temperatures (60oC) and pH (4 to 5) for enzy me action (Coughlan, 1985 as cited by Beachemin et al., 2003). However, these condi tions are not representative of the ruminal environment where the prevailing temperature is 39oC and the pH ranges from 6.0 to 6.7 (Van Soest, 1994). The activity of commerci al enzymes destined for use on ruminant feeds should be assayed under rumen-like pH and temperature to portray the true potential of such products. Factors Affecting Enzyme Action Research on enhancing the nutritive valu e of feedstuffs or increasing animal performance with fibrolytic enzymes has produced highly variable and inconsistent results. This is partially due to factors like the type of diet being fed the type, activity and rate of enzyme utilized (Dawson and Ticarico, 1999; Siciliano-Jones, 1999) under or over supplementation of the enzyme (Beauchemin et al., 2000; Yang et al., 1999); the method of providing the enzyme to the animal, and the stability of the enzyme in the rumen (Yang et al., 2000; Sutton et al., 2003). Im provements in animal performance due to treatment usually occur when fiber digesti on is markedly improved, with the greatest responses occurring when fiber digestion is compromised and energy is the first-limiting

PAGE 37

25 nutrient in the diet (Beauchemin et at., 2003) . Therefore the leve l of performance of animals also determines enzyme efficacy. The reason for poor responses to low enzyme application rates is obvious, but that for the higher levels is less apparent. Such la tter occurrences may be attributed partly to negative feedback inhibition of enzyme action due to production of critical concentrations of a product of the enzyme-substrate in teraction. For example, fermentation of sugars produced by cell wall hydrolysis may reduce ruminal pH to levels that inhibit enzymatic cell wall hydrolysis (Adesogan et al., 2005). An alternative hypothesis is that excessive enzyme applicati on blocks binding sites for enzymes or that it might prevent substrate colonization (Beauchemin et al., 2003). Mode of Enzyme Action Exogenous fibrolytic enzymes are used typi cally to increase the nutritive value of feedstuffs in order to increase animal perfor mance. There are various modes of action of such enzymes. These include precomsumptiv e hydrolysis of a feedstuff resulting in release of soluble carbohydr ates which may enhance pala tability (Adesogan, 2005) or breakdown and removal of structural barriers that impede digestion of a particular feedstuff (Colombatto et al., 2003b; McAllis ter et al., 2001). Post consumption, exogenous enzymes can potentially increase th e rate (Adesogan, 2005) and the extent of digestion or passage (McAllister et al., 2001; Adesogan, 2005). They can change the viscosity of ruminal fluid (Hristov et al., 2000; McAllister et al., 2001) and modify the site of digestion of nutrien ts (Hristov et al., 2000). A dditionally exogenous enzymes can provide synergistic effects with rumina l microorganisms (Morgavi et al., 2000; McAllister et al., 2001) that can lead to greater improvements in feed utilization than that which can be explained by cell wall hydrolysis alone. Several of these factors cause an

PAGE 38

26 increase in the hydrolytic capacity of the rumen (Yang et al., 1999) which indirectly decreases gut fill and enhances feed intake (Adesogan, 2005). Effect of Enzymes on Pre-Ingestion Fiber Hydrolysis Fibrolytic enzymes can release reducing suga rs from feedstuffs prior to ingestion though hydrolysis of the fiber fraction. However, the amount of reducing sugars released depends on the type of feed and the type of enzyme used (McAllister et al., 2001). This is further illustrated in a study conducted to evaluate en zyme effects on pre-treatment hydrolysis and in vitro degradation of alfalf a hay and corn silage (Colombatto et al., 2003a,b). Colombatto et al. (2003a) evaluate d the effect of two commercial enzyme additions (Depol 40, Biocatalyst Ltd. and Liquicell 2500, Specialty Enzymes and Biochemicals Co.) on the hydrolysis of stems and leaves of alfalfa hay after 20 h of incubation. Both enzymes increased solubl e OM losses from both substrates, however, the increases in OM loss in the absence of rumen fluid did not fully account for all the improvement in organic matter digestibilit y (OMD) in the presence of rumen fluid, suggesting that synergistic action between exogenous and ru men fluid enzymes enhanced the degradation of the substr ates (Colombatto et al., 2003a). Colombatto et al. (2003b) conducted a larger study in which corn sila ge and alfalfa hay were treated with 22 commercial enzymes and incubated for 15 min. Soluble sugar rel ease during incubation differed with enzyme type and substrate. The protein concentration of the enzyme explained 60% of the variation in the re sponse and the remaining variability was associated with the different sp ecific activities of the enzyme products. Colombatto et al. (2003b) concluded that the pre-treatment of s ubstrates with enzymes can release reducing sugars prior to incubation in ru men fluid. They also noted th at the variability associated with an enzymeÂ’s biochemical properties and hydrolytic capacity in the absence of rumen

PAGE 39

27 fluid makes it nearly impossible to predict the enzymeÂ’s performance on a particular substrate in vitro or in vi vo (Colombatto et al., 2003b). Effect of Enzyme on Post-ingestion DM and Fiber Digestibility Feng et al. (1996) used fistul ated steers to evaluate th e effects of three enzyme preparations (E1=Alphazyme , E2=Grasszyme, and E1E2=50:50 combination of E1 and E2) applied at two rates (low, high) on five di fferent moisture levels (fresh, wilted, dried and rehydrated to fresh, wilted and dried) of smooth bromegrass ( Broma inermis ). The enzyme activities were cellulase at 11,500 and 23,300 units of hydroxyethyl cellulase per milliliter, xylanase at 300 and 5,800 IU per m illiliter, and cellobiase at 40 and 55 IU per milliliter for E1 and E2 respectively. The app lication of E1E2 at the high level to the dried forage increased DM (43.5 vs. 38.7%) and NDF (31 vs. 26%) disappearance when compared to the control treatment indicati ng that addition of fi brolytic enzymes can enhance forage digestion. Nser eko et al. (2000) treated hay individually with one of six enzymes, containing an array of enzyme activit ies, and incubated them for 0 or 2 h before autoclaving to terminate enzyme activity. Sa mples were washed to remove the soluble fraction and then incubated in rumen fluid for 12 or 48 h. Enzyme treatment increased 12and 48-h NDF degradation in the abse nce of the active exogenous enzyme and soluble fractions of the forage (Nsereko et al., 2000) demonstrating that ruminal fiber degradation was enhanced by exogenous enzyme -induced alteration of forage structure prior to incubation in rumen fluid. In a study in which an enzyme (Grassz yme, FinnFeeds International, Marlborough, Wiltshire, U.K.) was added to the forage 24 or 0 h before feeding, or to barley at 0 h before feeding or ruminally infused, DM dige stibility (DMD) was greater in the 0 h than the 24 h pre-treatment of the forage. Th e ruminal infusion treatment produced lower

PAGE 40

28 disappearance of DM and NDF at 96 h when compared to the enzyme-treated forage treatments, demonstrating that the application of the enzyme directly to the forage can improve forage digestion (Lewis et al., 1996). Yang et al. (1999), reported that addi tion of a commercial enzyme mixture containing mainly cellulase a nd xylanase enzymes (Promote, Biovance Technologies Inc., Omaha, NE) at different rates to the forage or concentrate fed to dairy cows increased feed digestion and milk production (7 % increase). This research demonstrated that the response to the enzyme supplemen tation was affected by the quantity of the enzyme that was added regardless of whether it was added to forage or concentrate. Another commercial enzyme, Fibrozyme (A lltech Inc., Nicholasville, KY), which contains predominantly xylanase increased milk production by up to 6.2 lbs per cow per day when added to the TMR of dairy cows (Dawson and Tricario, 1999). Such increases in milk production are due often to increased voluntary intake, but they also can be associated with improved rumen function (D awson and Tricario, 1999) and DM and fiber digestion. In contrast, Hristov et al. (2000) eval uated the effect of supplying increasing ruminal doses of exogenous fibrolytic enzy mes on nutrient diges tion using cannulated heifers. Enzyme treatment increased the ruminally soluble fraction and the effective degradability of DM in situ, but did not affect apparent digestibility of DM, CP and NDF. Lewis et al. (1999) fed midlactation dairy cows smooth bromegrass or orchardgrass ( Dactylis glomerata )with or without enzyme supplementation. Cows consuming the enzyme-treated forage digested more DM pe r day and produced more milk than the cows fed the control forage. However, there wa s only a numeric increase in the DMI due to

PAGE 41

29 the lack of difference in the DM and NDF dige stibilities of the enzyme-treated forage diet compared to the control diet, leading the au thors to speculate that the increased DMI could be attributed to increased passage rate. Mandebvu et al. (1999a) evaluated the effects of applying fibrolytic enzymes to 3or 6-wk regrowth of bermudagrass prior to ensiling and observed no effect on in vitro or in situ DM or NDF disappearance of the silages. Therefore, the application of e xogenous enzymes on the digestibility of forages has produced inconsistent results due to various factors discussed earlier in this chapter. Effect of Enzyme Treatment on Beef Cattle Performance Research has demonstrated that supplemen tation of feedlot cattle diets with fiberdegrading enzymes has significant potential to improve feed utilization and animal performance (Beauchemin et al., 2003). Howeve r, while addition of exogenous fibrolytic enzymes to diets of beef cattle improved animal performance in some studies (Feng et al., 1996; Beauchemin et al., 1999), it has not in others (ZoBell et al., 2000). Beauchemin et al. (1995) evaluated the eff ects of fibrolytic enzyme addition on the performance of growing steers. Steers were fed cubed alfalfa hay, timothy hay or barley silage to which a mixture of xylanase and cel lulase enzymes were applied at different rates. Low levels of enzyme application to alfalfa hay and high levels of enzyme application to timothy hay increased ADG (24 to 36%), while enzyme application to barley silage had no effect. Where it occu rred, increased ADG was due to the increased fiber and DM digestion. The differences in the responses to enzyme treatment between the forages were attributed to the differences in cell wall structural components. This demonstrates the need to match enzyme activiti es with forage or feed characteristics. Beauchemin et al. (1997) evaluated the effects of addition of enzymes with contrasting xylanase to cellulase ratios to co rn and barley-based diets on performance and

PAGE 42

30 carcass characteristics of beef cattle. Enzyme treatment of corn-based diets did not affect performance. Animals fed the barley-based diet treated with the enzyme containing high xylanase and low cellulase activity had a 6% increase in ADG and a 5% decrease in DMI, which decreased the feed to gain rati o. The digestion of the fiber via enzyme supplementation may have increased the rumina l digestion of the barley grain, since the hulls usually limit the access of microbes to th e more digestible grain endosperm (Krause et al., 1998). The higher concen tration of fiber in barley th an corn, partially explained the lack of response to enzyme addition to co rn-based diets, but the reason for varying responses to the barley based-diets was not cl ear. This research further demonstrates the importance of enzyme-feed specificity (B eauchemin et al., 1997). In another study, barley was treated with a commercial enzy me (Pro-Mote, Biovance Technologies, Inc. Omaha, NE) during the tempering process (18h prior to feeding) and fed with barley straw or silage to beef steers. Enzyme treatment increased the total tract ADF digestibility by 14% for steers fed barley st raw and by 55% for those fed barley silage when compared to the same treatments wit hout enzyme supplementation (Krause et al., 1998). This research demonstrates that di etary ingredient composition affects enzyme efficacy. Feng et al. (1996) evaluated the effects of applying enzymes (Alphazyme or 50:50 combination of Alphazyme and Grasszyme) at different times during the harvesting process of smooth bromegrass hay. Enzyme was applied immediat ely after harvesting, wilting or immediately prior to feeding (E-Dry). The EDry treatment increased NDF and ADF intakes due to increased DMI for both the hay and total diet versus the control diet. The E-Dry treatment also increased total tract DMD, NDFD, and particulate

PAGE 43

31 passage rate (31%), and decreased the rumen retention time (24%) when compared to the control treatment. The rate of in situ DM degradation of the forage tended to increase across all treatments, whereas the rate of NDF degradation increased only for E-Dry. Enzyme treatment did not affect the extent of in situ NDF degradation, suggesting that the mechanism for improvement of DM diges tion requires direct c ontact of the enzyme with the substrate prior to feeding (Feng et al., 1996). When Fibrozyme (Alltech Inc., Nicholasville, KY) was added to alfalfa or ryegrass hay in sheep diets there was an increase in digestibility of alfalfa but not ryegrass (Pinos-Rodriguez et al., 2002), further revealing the importance of matching enzymes to specific substrates (Adesogan, 2005). Fibrolytic enzyme addition does not always increase animal performance. ZoBell et al. (2000) evaluated the e ffect of an exogenous enzyme treatment containing xylanase and endoglugonase on the production and carcass ch aracteristics of finishing steers. The enzymes were applied to a TMR that cont ained alfalfa hay, grass hay, barley and finishing diet supplements. Ca rcass characteristics, ADG, DMI, of feed efficiency (FE) were not different due to enzyme addition. Additionally, Michal et al. (1996) reported that various combinations of xylanase and cellulase (1:2 and 2:1 respectively) applied to alfalfa haylage just prior to feeding increased feed intake of beef steers (8.7%) but had no effect on BW gain or FE. The inconsistencies in the re sults of feeding exogenous enzymes can be attributed to differences in enzyme type, preparation, application method and rate (Beauchemin et al., 2003; Adesogan et al., 2005) as well as fraction of the diet to which the enzyme is applied (Feng et al., 1996; Lewis et al., 1996; ZoBell et al., 2000). Nevertheless,

PAGE 44

32 enzymes have been used to improve the ut ilization of diets containing legumes, hays, crop residues and other feedstuffs (Beauchemin et al., 2003). Exogenous enzymes usually increase the rate of digestion but not the extent of digestion (Feng et al., 1996), therefore, ther e is a need to focus on enzymes that can increase the extent of digesti on, particularly in poor quality forages such as those that abound in Florida and other tropic al and subtropical countries. Most of the research on exogenous fibrolytic enzyme tr eatment of ruminant feeds has been conducted on cereal grains, byproducts and cool-season (C3) silages. There is a need for research to examine the effects of enzyme a pplication to unensiled C4 forages. Currently there is no knowledge of optimal inclusion rates of cellu lase, xylanase, and es terase enzymes for enhancing the utilization of C4 grasses in cattle. The aim of this series of experiments was to evaluate the effect of fibrolytic enzyme application on the nutritive value of tropical forages and animal performance. The specific objectives were the following: to evaluate the effect of different ap plication rates of ferulic acid esterase preparations on the dige stibility of three C4 hays (Chapters 3 & 4), to determine the effect of different combin ations of ferulic acid esterase, cellulase, and xylanase on the digestion of ba hiagrass hay (Chapter 5), and to determine the effects of various met hods of applying a commercial fibrolytic enzyme mixture or ammonia to bermudagr ass hay on the feed intake digestion kinetics and growth performance of beef steers. (Chapter 6)

PAGE 45

33 CHAPTER 3 THE POTENTIAL FOR INCRE ASING THE DIGESTIBILITY OF THREE TROPICAL GRASSES WITH A FUNGAL FERULIC ACID ESTERASE ENZYME PREPARATION Introduction Although C4 grasses are the mainstay of lives tock farming in the tropical and subtropical regions of the worl d, their nutritive value is inhe rently low, such that the performance of animals grazing such pastures is often subopt imal. The poor quality of these grasses has been attributed largely to their relatively high lignin concentrations. Recent work suggests that cross linkages form ed between hemicellulose, ferulic acid and lignin limit microbial access to the digestible xylans in the cell walls of the grasses (Hatfield, 1993; Jung and Deetz, 1993; Hatf ield et al., 1997; Ca sler and Jung, 1999; Grabber et al., 2000). Hatfield et al. (1997) also demonstrat ed that the combination of such cross linkages and high lig nin concentrations greatly reduced the digestibility of C4 grasses. This cross linking reduces the number of binding sites on the cell structure for the attachment of rumen microbes. In C3 grasses and maize bran, fungal ferulic acid esterase enzymes have been shown to release ferulic acid from the cross linkages, such that the digestible xylans in the cell wall ar e more susceptible to enzymatic degradation (Faulds et al., 1995; Kroon et al., 1999). To ev aluate the potential of using such enzymes to enhance the quality of C4 grasses, this study determined the effect of different application rates of a ferulic acid esterase enzyme (FAE) preparation on the digestibility of three C4 grass hays.

PAGE 46

34 Materials and Methods Enzymes and Forages The enzyme used for this study was De pol 740L (D740L, BioCatalyst, Pontypridd, Wales, UK). Biocatalyst stipulates that D740L is a food-grade enzyme containing 32 U/ml of ferulic acid esterase activity, wh ere one unit is the amount of enzyme which releases one micromole of ferulic acid from methyl ferulate per minute at pH 6.0 and 37oC. The enzyme was analyzed for cellulase activity with the filter-paper reducing sugar assay of Wood and Bhat (1988) and xylanase activity with the assay of Bailey et al. (1992). Cellulase activity was found to be 20.78 filter paper units (FPU) where one FPU is the equivalent amount of enzyme required to release exactly 2.0 mg of glucose from a 1 x 6 cm piece of filter paper at pH 5.5 at 39oC. Xylanase activity was found to be 8701.85 µmol/ml/min using 1% birchwood xyl an (X-0502, Sigma Chemical Company, St. Louis , MO, USA) as a standard. The units of xylanase activity ar e expressed as µmol of xylose equivalents ml/ min using a standard curve of xylose. Transverse sections of 450-kg round ba les of 12-wk regrowth of Pensacola bahiagrass (BAH) ( Paspalum notatum ), Coastal (C-B; Cynodon dactylon ) and Tifton 85 (T-85) bermudagrass ( Cynodon dactylon ) hays from the University of Florida Santa Fe Beef Research Unit were removed with a we dge cutter. Grab samples were taken from the sections and composited to give a 3-kg (a s-is) representative sample of each bale. Half of each representative sample was ground to pass through a 1-mm screen in a Wiley mill (Arther H. Thomas Company, Philadelphi a, PA, USA) and the other half was ground through a 4-mm screen for rumen degradability analysis. Half of each forage sample was ground to pass through a 1-mm screen in a Wiley mill (Arther H. Thomas Company, Philadelphia, PA, USA) and the other half was ground through a 4-mm screen for rumen

PAGE 47

35 degradability analysis. The enzyme was disso lved in 10 ml of double-distilled water and sprayed at rates of 0, 0.5, 1, 2, and 3 g/100 g DM onto 10-g (as-is) samples of ground (1 mm) forage. Control samples were sprayed wi th water alone. All treated samples were allowed to air dry for 8 h before being weighed (0.5 g) into ANKOM F57 (ANKOM Technologies, Macedon, NY) filter bags in triplicate. In Vitro Digestibility Sixteen hours later, the samples were incubated in buffered, rumen fluid for 6, 24, and 48 h using two ANKOM® Daisy II inc ubators (ANKOM Technologies, Macedon, NY). Rumen fluid was obtained before f eeding from two non-lactating, ruminally fistulated Holstein cows fed 9 kg of bahiag rass hay supplemented with 0.4 kg of soybean meal. This experiment was repeated three consecutive times to account for variations in rumen fluid activity. The T illey and Terry (1963) two-st age, rumen fluid-pepsin technique as modified by Moore and Mott (1974 ) was used to estimate 96-h in vitro DM digestibility (IVDMD). Chemical Analysis Neutral detergent fiber (NDF) and Acid de tergent fiber (ADF) concentrations were measured in the pre digested forage sample s and in vitro digestion residues using the method of Van Soest et al. (1991) in an ANKOM® 200 Fiber Analyzer (ANKOM Technologies, Macedon, NY, USA) and NDF and ADF digestibilities were calculated. Water soluble carbohydrate (WSC) concentra tion was analyzed using the anthrone method described by Ministry of Agricu lture, Fisheries, and Food (1986). Free monomeric ferulic acid and pcoumaric acid concentrations were measured in samples using procedures described by Ra lph et al. (1994). For a de tailed explanation of this procedure see Appendix A.

PAGE 48

36 In Situ Rumen Degradability The effect of Depol 740L treatment on the in situ rumen DM degradability of each of the C4 grasses was assessed using the polyester bag technique at different times. The enzyme was dissolved in 40 ml of double-di stilled water and sprayed onto 225-g samples of ground (4 mm) forage at rates of 0, 0.5, 1, 2, 3 g/100g of DM. Control samples were sprayed with water alone. All treated samples were allowed to air dry for 8 h before being weighed (5 g, as treated basis) into tared nylon bags (10 x 23 cm) with a pore size of 50 microns (Bar Diamond Inc., Parma, ID). Th e bags were tied with a rubber o-ring and rubber band and attached to a co tton rope with stainless steel clips. A calan gate key (American Calan, Northwood, NH) was attached to the end of the rope to ensure submersion of the bags in rumen fluid.Sixt een hours later, duplicate samples of each forage sample were incubated in each of the two previously described ruminallyfistulated cows for 0, 3, 6, 9, 12, 24, 48, 72, 96, a nd 120 h, resulting in four replicates per treatment for each forage for each incubati on duration. After incubation, bags were removed from the cow, rinsed with cool water and frozen. At the end of the measurement period all bags were thawed and washed in a commercial washing machine using a cool-wash cycle without soap. Bags were dried for 24 h at 60oC and residue weights determined. In order to avoid pl acing too many substrate-filled bags in the rumen, only bags for 4 time measurements a nd one forage were simultaneously incubated (40 bags maximum incubated at the same time/cow). The exponential model of McDonald (1981) was fitted to the in situ da ta using the nonlinear procedures of SAS v8 (2002, SAS Institute Inc., Cary, NC). The model is of the form: P= A + B(1-e-c(t-L))

PAGE 49

37 where P= DM degraded at time t, A= wash lo ss , B= potentially degradable fraction, A+B= total degradability, c= the rate at whic h B is degraded, t= time incubated in the rumen, and L= lag time. Statistical Analysis A 3 (forage) x 5 (enzyme rates) factorial design with three replicates per treatment was used to examine the effect of enzyme application rate on chemical composition and extent of DM and NDF in vitro digestibil ity. The data were analyzed using the GLM procedure of SAS v8 (2002). Polynomial contra sts were used to test the effect of increasing enzyme application and contrasts we re used to compare grass species (BAH vs C-B + T-85) and bermudagrass cultivars (C-B vs T-85). The model used to analyze individual treatment effects was: Yijkl = µ + Fi + Ej + Rk + FTij + FRik + TRjk + FTRijk+ Eijkl where: µ = general mean Fi = forage effect of treatment Ej = enzyme rate effect Rk = run effect FEij = enzyme x forage interaction FRjk = forage x run interaction ERik = enzyme x run interaction FERijk = forage x enzyme x run interaction Eijkl= experimental error. A 2 (cows) x 5 (enzyme rates) factorial desi gn of treatments with two replicates per treatment was used to determine the in situ DM degradability of each hay. The data from

PAGE 50

38 each hay were analyzed separately due to the fact that they were not incubated simultaneously. The data were analyzed using the GLM procedure of SAS v8 (2002). Fourth order polynomial contrasts were used to test the effect of increasing enzyme application rate. The model used to an alyze individual treatment effects was: Yij = µ + Ei + Cj + ECij + Eijk where: µ= general mean Ei = effect of enzyme rate Cj = effect of cow ECij = enzyme x cow interaction Eijk= experimental error. Significance was declared at P < 0.05 and tendencies at P < 0.10. Results Chemical Composition The DM concentrations were 909, 916 and 910 g/kg, the ash concentrations were 59, 45 and 48 g/kg of DM, and the CP concentr ations were 65, 63 and 73 g/kg of DM for BAH, C-B, and T-85 respectively. The con centration of WSC and cell wall components differed among forage species and cultivar (Tab le 3-1). Across enzyme application rates BAH had more WSC and ADF and less NDF , hemicelluose and lignin than the bermudagrasses. Also, T-85 contained a grea ter concentration of NDF, ADF and lignin, and less hemicellulose and WSC than C-B. Across the three hays, enzyme applica tion reduced NDF, ADF and hemicellulose concentrations ( P < 0.001), and increased WSC ( P = 0.012) concentration. However, the NDF, ADF, hemicellulose and lignin responses depended on forage species and cultivar

PAGE 51

39 ( P < 0.05). The reductions in NDF and hemicellu lose concentrations were linear for BAH and C-B, but not for T-85. The reductions in ADF concentration were linear for BAH, and nonlinear for the bermudagrasses. Enzyme application modified the lignin concentration of BAH and C-B ( P < 0.05), but the effects were practically in significant (Table 3-1). Free ether-linked ferulic acid concentrati ons were greater in the bermudagrasses than BAH with nearly all the differences due to T-85 (Table 3-2). Free ether-linked pcoumaric acid concentration was greater in T-85 than C-B. Free ester-linked ferulic acid concentrations were greater in BAH than in the bermudagrasses and less in T-85 than CB. In contrast free ester-linked p-coumar ic acid was greater in bermudagrasses than BAH, and greater in T-85 than C-B. Increasing the enzyme application rate li nearly increased free ether-linked ferulic acid concentration across hays. This was larg ely due to increased release of free etherlinked ferulic acid from T-85 as enzyme rate increased. The concentration of free etherlinked p-coumaric acid was unaffected by enzy me application. Release of free esterlinked ferulic acid from BAH also increased ( P = 0.002) as enzyme application rate increased. Increasing the enzyme applicati on rate increased the release of free esterlinked p-coumaric acid from BAH but did not increase thatof the bermudagrasses. In Vitro Disappearance and Digestibility Study There were few speciesrelated differences in DM disappearance of forage samples incubated for 6, 24 or 48 h, but the 6-h DM di sappearance of T-85 was greater than that of C-B. The 96-h rumen fluid-pepsin IVDMD coefficients were slightly greater in the bermudagrasses than BAH.

PAGE 52

40 Increasing the rate of enzyme application resulted in linear increases in the DM disappearance of the hays after 6 h ( P = 0.001) or 24 h ( P = 0.03) of incubation (Table 3-3). However, the response differed ( P = 0.001) for bermudagrass cult ivars at 6 h in that it was linear ( P= 0.001) for T-85 and quadratic ( P = 0.02) for C-B. Though the data suggests a linear increase in 24-h DM disappearance across forages as enzyme application increased, the res ponse only existed (P < 0.05) for BAH. There was no effect of enzyme level on the 48-h DM disappearan ce of BAH or T-85, but there was a linear ( P = 0.05) increase in the values for C-B (Table 3-3). There was a cubic increase ( P > 0.001) in 96-h rumen flui d-pepsin IVDMD across hays as enzyme application increased and the values peaked at the 2 g/kg DM rate (Table 3-3). However, the response was cubic fo r BAH and T-85, while it was quadratic for C-B. As enzyme application rate increased, there was a linear ( P = 0.02) increase in 6-h NDF disappearance (IVNDFD) of T-85, but th is trend was not evident in the other forages (Table 3-4). Increasing the enzyme application rate linearly ( P = 0.002) increased the mean 24-h IVNDFD of hays, largely due to linear ( P = 0.01) increase in the 24-h IVNDFD of BAH. Enzyme application did not affect the 48-h IVNDFD of the hays. In Situ Study Increasing the rate of enzyme applica tion increased the wash value of BAH ( P = 0.001), C-B ( P = 0.001) and T-85 ( P = 0.006; Table 3-5), and modi fied the potential and total degradable fractions and e ffective degradability of C-B ( P < 0.05). The rate of degradation was modified ( P = 0.012, quartic) for T-85 and there was a quadratic

PAGE 53

41 increase in the duration of the lag phase of C-B ( P = 0.003) with increasing enzyme application. Discussion It is well established that as forages mature there are marked decreases in CP and digestibility due to increased fiber depos ition and lignification (Jung and Allen, 1995). The bermudagrasses used in the current study were more mature a nd therefore contained less CP and more NDF, ADF and hemicellulose than the 2 to 7 wk regrowth of T-85 and C-B examined by Mandebvu et al. (1999b). Ho wever such mature forages are often fed to beef cattle in the winter in Florida, ther efore they were intenti onally chosen for this study. Bahiagrass and bermudagrasses (BERM) varied in chemical composition. Typically BAH has a greater ADF concentra tion than BERM (Evers et al., 2004). Mandebvu et al. (1999b) reported that 3 to 7 wk regrowth of T-85 contained greater concentration of NDF and ADF and less lignin free ether-linked FA and P-coumaric acid concentrations than C-B at the same maturity stages. The lower concentration of etherlinked FA in T-85 is indicative of fewer cros s-links between hemicellu lose to lignin, such that there is potentially less i nhibition of fiber digestion by microbes. In the current study C-B contained the lowest concentration of NDF, ADF and lignin followed by BAH then T-85. Hemicellulose concentration was gr eatest for C-B followed by T-85 and then BAH. The free phenolic acid c oncentrations of the forage s were lowest for the BAH followed by C-B and then T-85. The concentrations of cell wall components in the bermudagrasses in this study differed from those in the literature (Mandebvu et al., 1999a, b) due to differences in maturity. The lower quantities of lignin and greater quantities of hemicellulose in C-B meant there was less lignified tissue and greater

PAGE 54

42 amounts of hemicellulose in the forage. Consequently, C-B had greater 6-h and 96-h IVDMD than the other forages. Enzyme treatment of the forages re sulted in reductions of NDF, ADF and hemicellulose concentrations and corresponding increases in WSC concentrations. This suggests that the enzyme cleaved cell wall polys accharides, releasing sugars after 24 h of treatment but prior to incubati on in rumen fluid. This agre es with Krause et al. (1998) who suggested that the application of enzyme s to feeds could potentially solubilize NDF and ADF. Others also have not ed that a pre-feeding enzyme-f eed interaction is necessary for increases in digestion to be seen (Lewis et al., 1996; McAlliste r et al., 1999; Wang et al., 2001). Enzyme application increased the release of ether-linked ferulic acid from T-85 and release of ester-linked ferulic acid and p-coum aric acid from BAH. This is consistent with research from Anderson et al. (2005) who found that D740L application increased the release of free phenolic acids from 8-wk regrowth of T-85 and C-B as compared to buffer alone. Yu et al. (2003, 2005) demonstrat ed that in synergy with cellulases and xylanases, ferulic acid esterases can br eak linkages between ferulic acid and hemicellulose components. This makes cel l wall-bound digestible carbohydrates more accessible to fibrolytic enzymes. The releas e of phenolic acids reinforces the idea that enzyme treatment of C4 grasses prior to ingestion can increase pre-ruminal hydrolysis of the forage cell wall. The lack of effectiven ess of enzyme treatments on ether-linked pcoumaric acid can be attributed to the fact that p-coumarate ethers are linked only to lignin and do not form cross-linkages with polysaccharides (Lam et al., 1992; Jung and Allen, 1995).

PAGE 55

43 Depol 740L application increased the 6, 24, and 96 h IVDMD of the forages, but many of these responses were not due to in creased NDF disappearance. Rather, they were due to increased preruminal NDF, ADF and hemicellulose hydrolysis and release of ferulic acid. The greater DM D at 6, 24 and 48 h for C-B than T-85 can be attributed to lower concentrations of ether and ester-linked ferulic acid in C-B than T-85. This agrees with Mandebvu et at. (1999b) in th at lower concentrations of ether-linked ferulic acid can account for greater digestibility of forages. Feng et al. (1996) reported th at fibrolytic enzyme treatment resulted in greater IVDMD and IVNDFD values for smooth bromegrass ( Broma inermis) . Similar results were also reported by Varel et al. (1993) and Nsereko et al. (2000). Application of D670L to 12-wk regrowth of C-B also increased IVDMD and IVNDFD (Chapter 4), though that enzyme contained less esterase activity (7 U/ml), more cellulase (1200 U/g) and endogalacturonase activity (800 U/g) than the enzyme used in this study. The increase in the washing loss fract ion and 6-h DM disappearance were consistent across forages and is also indi cative of preruminal enzymatic hydrolysis (Lewis et al., 1996; Wang et al., 2001, 2004; Co lombatto et al., 2003a,b; Yu et al., 2005). The increased 24-h DM disappearance suggests that enzyme application may increase the rate of digestion of grasses (Yang et al., 1999), but this was not confirmed by the in situ degradability analysis at the linear, quadratic or cubic level. It is pertinent to note that a quartic ( P < 0.05) increase in degradation rate was detected occasionally as enzyme application rate increa sed, but the biological interpretation of this response is not clear. The increases in 96-h IVDMD following en zyme treatment suggests that adding fibrolytic enzymes to forage can improve the extent of fiber digest ion as noted elsewhere

PAGE 56

44 (Beauchemin et al., 1995; Feng et al., 1996; Wa ng et al., 2004). However, the magnitude of the increase was small. The IVDMD at 96 h was lower than the 48 h in vitro disappearance values largely due to losses of undigested substrate through the pores of the ANKOM bags (Adesogan et al., 2005; Wilman and Adesogan, 2000). This study supports previous work showing that mixtures of ferulic acid esterase and xylanase can break linkages between sugars and ferulic acid in the cell wall (Ferreira et al., 1993; Faulds and Williamson, 1994, 1995; Kroon and Williamson, 1999; Yu et al., 2003, 2005). Therefore, there seems to be a synergistic effect between ferulic acid esterase, xylanase and cellulose that can incr ease the digestibility a nd degradation of low nutritive value tropical forages (Yu et al., 2005). However, the improvement in the extent of digestion was marginal, therefore th e potential of using other enzyme mixtures to improve the digestibility of tropical grasses should be explored. Additionally, further research should be conducted to determine the contributions of individual enzymes in enzyme mixtures to digestibility to further elucidate the synergistic effects of enzymes on the forage digestion. Implications Application of the D740L enzyme enhan ced cell wall hydrolysis and 96-h IVDMD across hays. The 6-h IVDMD of the bermuda grasses and the 6-h IVNDFD of C-B were also increased by enzyme treatment. Increa sing the rate of enzyme application also increased the in situ wash-loss fraction and enhanced the release of WSC, ether-linked ferulic acid from T-85 and ester-linked feru lic and p-coumaric acids from BAH. These results indicate that the enzyme increased cleavage of cell wall polysaccharides and release of phenolic acids, and consequently enhanced digestibility. However the responses differed with forage species and cultivar, and they were more pronounced in

PAGE 57

45 the bermudagrass species. Since the enzyme -induced improvement in digestibility was small, further research should examine the pot ential for markedly improving the digestion of C4 grasses with other fibrolyt ic enzyme preparations.

PAGE 58

46Table 3-1. Effect of enzyme applicati on on cell wall and water soluble carbohydrate (WSC) concentration of tropical grass hays , g/kg Enzyme application rate Mean SEMEnzyme Forage effecta,b Forage*Enzyme effecta,b Item 0% 0.5% 1% 2% 3% Rate effectb BAH v. BERM C-B v. T-85 BAH v. BERM C-B v. T-85 NDF BAH 814 807 824 797 787 806 4.1 0.002 L C-B 805 804 805 798 786 800 2.5 0.001 L T-85 889 877 886 868 866 876 2.4 NS Mean 836 829 839 821 813 3.1 0.001 L <0.0001 <0.0001 0.001 0.002 ADF BAH 419 417 422 411 406 415 2.2 0.042 L C-B 362 362 360 360 357 360 1.8 0.001 Q T-85 473 463 471 460 459 465 1.8 0.022 C Mean 418 414 418 410 407 0.9 0.001 L <0.0001 <0.0001 0.019 0.047 Hemicellulose BAH 395 390 402 386 381 391 2.9 0.006 L C-B 444 442 444 438 430 440 2.8 0.001 L T-85 416 414 415 408 407 412 2.2 0.005 Q Mean 418 416 421 410 406 1.3 0.001 L <0.0001 <0.0001 0.0001 0.032 WSC BAH 22.7 21.2 23.2 27.0 20.5 22.9 3.8 NS C-B 11.5 10.7 21.8 18.4 9.9 14.4 2.3 0.004 Q T-85 5.1 5.4 8.4 5.0 5.6 5.9 0.7 0.007 Q Mean 13.0 12.4 17.8 16.9 12.0 2.6 0.012 Q <0.0001 <0.0001 NS NS Lignin BAH 61.3 60.1 65.8 58.4 66.1 62.4 0.9 0.020 C C-B 59.7 56.6 57.0 54.2 58.0 57.1 0.6 0.015 Q T-85 70.8 70.9 71.1 70.8 66.9 70.1 1.4 NS Mean 64.0 62.6 64.6 61.1 63.7 0.6 NS 0.049 <0.0001 0.002 0.003 a BERM = bermudagrass cultivars, BAH = bahiagrass, C-B = coastal bermudagrass, T-85 = Tifton 85 bermudagrass; b L = Linear, Q = Quadratic, C = Cubic

PAGE 59

47Table 3-2. Effect of D740L app lication on the free ferulic acid and p-coumaric concentration of tropical grass hays, mg/g cell wall. Enzyme application rate Mean SEMEnzyme Forage effecta,b Forage*Enzyme effecta,b Item 0% 0.5% 1% 2% 3% Rate effectb BAH v. BERM C-B v. T-85 BAH v. BERM C-B v. T-85 Ether-linked ferulic acid BAH 1.95 1.42 3.05 2.63 3.30 2.47 0.16 NS C-B 2.14 2.75 2.69 2.25 2.28 2.43 0.10 NS T-85 4.03 4.68 4.90 4.82 3.92 4.47 0.12 0.005 Q Mean 2.71 2.95 3.55 3.23 3.17 0.17 0.031 L <0.0001 <0.0001 NS NS Ether-linked p-coumaric acid BAH 0.83 2.77 1.42 1.01 1.09 1.42 0.47 NS C-B 0.92 1.05 0.99 0.94 0.93 0.96 0.02 NS T-85 2.25 2.21 2.39 2.27 1.96 2.22 0.25 NS Mean 1.33 2.02 1.61 1.41 1.33 0.38 NS NS 0.005 NS NS Ester-linked ferulic acid BAH 5.42 7.17 5.80 6.44 7.64 6.49 0.12 0.002 C C-B 5.44 6.45 5.99 5.79 5.49 5.83 0.27 NS T-85 4.66 4.41 4.14 5.06 3.93 4.44 0.07 NS Mean 5.18 6.00 5.31 5.76 5.68 0.22 NS <0.0001 <0.0001 0.016 0.049 Ester-linked p-coumaric acid BAH 5.59 7.22 6.20 6.59 7.31 6.58 0.08 0.001 C C-B 6.91 8.11 7.38 7.11 6.78 7.26 0.22 NS T-85 8.38 8.33 7.81 8.77 7.69 8.20 0.08 0.010 C Mean 6.96 7.88 7.13 7.49 7.26 0.18 NS <0.0001 <0.0001 0.032 0.015 a BERM = bermudagrass cultivars, BAH = bahiagrass, CB = coastal bermudagrass, T85 = Tifton 85 bermudagrass b L = Linear, Q = Quadratic, C = Cubic

PAGE 60

48Table 3-3. Effect of D740L a pplication on the in vitro 6-h, 24-h and 48-h rume n fluid DM disappearance and 96 h rumen-fluid pe psin digestibility of tropical grass hays, g/kg DM. Enzyme application rate Mean SEMEnzyme Forage effecta,b Forage*Enzyme effecta,b Item 0% 0.5% 1% 2% 3% Rate effectb BAH v. BERM C-B v. T-85 BAH v. BERM C-B v. T-85 6 h BAH 133 159 165 156 182 159 9.87 NS C-B 180 185 178 207 212 192 6.52 0.02 Q T-85 99 113 97 149 159 123 8.81 0.001 L Mean 137 152 147 171 184 8.52 0.001 L NS 0.001 NS 0.001 24 h BAH 403 394 417 368 404 398 13.8 0.007 L C-B 398 380 399 396 392 492 13.8 NS T-85 355 389 358 398 388 377 16.8 NS Mean 386 387 391 387 395 14.9 0.03 L NS NS 0.030 NS 48 h BAH 493 502 517 508 507 505 8.24 NS C-B 487 484 489 506 510 495 10.4 0.05 L T-85 485 481 479 485 497 485 10.9 NS Mean 488 489 495 500 505 9.91 NS NS NS NS NS 96 h BAH 398 396 406 424 405 405 5.1 0.005 C C-B 443 396 442 473 467 444 6.4 0.010 Q T-85 437 435 450 461 449 447 4.3 0.003 C Mean 426 409 433 453 440 5.2 0.001 C 0.001 NS 0.033 0.002 a BERM = bermudagrass cultivars, BAH = bahiagrass, CB = coastal bermudagrass, T85 = Tifton 85 bermudagrass b L = Linear, Q = Quadratic, C = Cubic

PAGE 61

49 Table 3-4. Effect of D740L applic ation on in vitro NDF disappearance of tropical grass hays, g/kg DM. Enzyme application rate Mean SEMEnzyme Forage effecta,b Forage*Enzyme effect a,b Item 0% 0.5% 1% 2% 3% Rate effectb BAH v. BERM C-B v. T-B BAH v. BERM C-B v. T-B 6 h BAH 26.1 48.0 36.9 41.4 50.0 40.4 10.9 NS C-B 56.3 58.3 41.0 75.3 63.9 58.9 7.74 NS T-85 41.0 34.7 29.2 61.3 72.4 47.7 8.50 0.02 L Mean 41.1 47.0 35.7 59.2 62.1 9.16 NS NS NS NS 0.01 24 h BAH 325 310 330 270 373 322 33.0 0.01 L C-B 305 280 299 300 284 294 20.4 NS T-85 310 338 304 337 331 324 19.8 NS Mean 313 309 311 302 329 25.1 0.002 L NS NS 0.03 NS 48 h BAH 434 445 438 433 423 435 10.5 NS C-B 417 415 406 434 421 419 14.6 NS T-85 460 449 449 435 454 449 12.8 NS Mean 437 436 431 434 433 12.7 NS NS NS NS NS a BERM = bermudagrass cultivars, BAH = bahiagrass, CB = coastal bermudagrass, T85 = Tifton 85 bermudagrass b L = Linear, Q = Quadratic, C = Cubic

PAGE 62

50 Table 3-5. Effect of D740L a pplication on the in situ Dm de gradability of tropical grass hays. Total degradable fraction (A+B), g/kg DM BAH 673 687 648 701 707 684 17.3 NS C-B 614 605 604 609 624 610 5.73 0.030 Q T-85 644 660 611 676 662 651 16.1 NS Fractional degradation rate (c), h-1 BAH 2.32 2.24 2.84 2.17 2.20 2.35 0.27 NS C-B 2.93 3.07 3.17 3.07 2.95 3.07 0.14 NS T-85 3.03 2.62 3.81 2.70 2.58 2.94 0.30 0.012 Qt Lag phase (L), h BAH 8.52 8.42 9.72 8.13 5.16 7.98 1.20 NS C-B 1.10 1.30 2.90 4.85 0.75 2.20 0.59 0.003 Q T-85 8.11 5.96 8.09 5.76 4.34 6.42 1.45 NS a BERM = bermudagrass cultivars, BAH = bahiagrass, C-B = coastal bermudagrass, T-85 = Tifton 85 bermudagrass b L = Linear, Q = Quadratic, C = Cubic, Qt = Quartic Enzyme application rate Mean SEM Enzyme Itema 0% 0.5% 1% 2% 3% Rate effectb Wash Value (A), g/kg DM BAH 118 110 117 132 138 123 3.85 0.001 L C-B 86 97 108 124 100 103 3.31 0.001 L T-85 66 71 74 117 90 84 6.59 0.006 C Potentially degradable fraction (B) g/kg DM BAH 554 577 532 569 569 560 17.1 NS C-B 527 507 495 485 523 506 7.80 0.001 Q T-85 578 589 537 560 572 567 20.1 NS Effective degradability, g/kg DM BAH 661 673 639 688 691 670 15.0 NS C-B 605 596 566 691 615 602 5.1 0.0240 Q T-85 634 649 604 665 651 641 14.4 NS

PAGE 63

51 CHAPTER 4 EFFECT OF FIBROLYTIC ENZYME PR EPARATIONS CONTAINING ESTERASE, CELLULASE, AND ENDOGALACTU RONASE ACTIVITY ON THE DIGESTIBILITY OF MATURE, TROPICAL GRASS HAYS Introduction Tropical grasses are the staple diet of most domesticated rumina nts in tropical and subtropical regions. Dry matter digestibility and intake of tropical C4 grasses are considerably lower than t hose of their temperate, C3 counterparts (Minson, 1980). While ryegrass (Lolium multiflorum) (C3) typically has a neutral de tergent fiber concentration (NDF) of 400 g/kg of dry matter (DM) and a lignin concentration of 48 g/kg DM (NRC 2000), C4 grasses, like 8-wk regrowthof berm udagrass, may contain 70 to 75 % NDF, 40 to 45% ADF and 6 to 8 % crude protein on a DM basis (Ball et al ., 2002). Casler and Jung (1999) identified ferulic acid cross li nkages as key factors that affect the digestibility of C3 grasses. Ferulates and diferulate s have been recognized as important structural components of plant cell walls (Kr oon et al., 1999) and ferulic acid is the most abundant hydroxycinnamic acid in the plant wo rld (Faulds et al., 1995). Such phenolic acids are more abundant in C4 grasses than C3 grasses (Akin, 1986; Kroon et al., 1999). These ferulates cross link xylans to lignin (Ralph et al., 1994; Gr abber et al., 2000) and therefore limit the enzymatic degradation of the digestible polysaccharide in the cell walls (Hatfield 1993; Jung and Deetz, 1993). The cross-linkages are also thought to cause cell wall stiffening and growth cessation in plants (M usel et al., 1997). Mandebvu et al. (1999) and Hatfield et al. (1997) reported that Ti fton 85 bermudagrass was more digestible than Coastal bermudagrass because of lower lignin and ether-linked ferulic

PAGE 64

52 acid concentrations. Ferulic acid esterase (FAE) enzymes can cleave off the ferulic acid from the cell wall xylans, making the xylans mo re readily available to ruminal microbes. The addition of FAE to other polysacchari de-degrading enzymes can increase the rate and extent of digestion of plant cell walls (Kroon and Williamson, 1999; Yu et al., 2005). So far most research on esterase applic ation to forages has been focused on C3 grasses and little is known about their potential for enhancing the digestion of C4 grasses. Previous work (Chapter 3) demonstrated e nhanced cell wall hydrolysis and in vitro DM degradability (IVDMD) following application of an enzyme preparation containing a mixture of ferulic acid esterase, xylanase a nd cellulase enzymes. However, there are other enzymes that can potentially degrade plant cell wall polysacchar ides. Therefore, the objective of this work was to evaluate the effect of different application rates of an enzyme preparation containing ferulic acid es terase (FAE) and other fibrolytic enzymes on the digestibility of NDF and DM in bermudagrass and bahiagrass hays. Materials and Methods Enzymes and Forages The enzyme preparation used for this study was Depol 670L (D670L, BioCatalyst, Pontypridd, Wales, UK). Bio catalyst stipulates that D 670L is a food-grade enzyme containing 8 U/ml of ferulic acid estera se activity, 1200 U/g cellu lase and 773 U/g of endogalactouranse activity, where the unit is the amount of enzyme which causes the release of one micromole of ferulic acid fr om methyl ferulate per minute at pH 6.0 and 37oC. The enzyme was also found to contain cellulase activity of 38.13 filter paper units (FPU) with the filter paper reducing sugar assay (Wood and Bhat, 1988), where one filter paper unit is the equivalent amount of enzy me required to releas e exactly 2.0 mg of glucose from a 1 x 6 cm piece of filter paper at pH 5.5 at 39oC. Xylanase activity was

PAGE 65

53 found to be 3682.73 µmol/ml/min using 1% birchwood xylan (X-0502, Sigma Chemical Company, St. Louis, MO, USA) as a standa rd according to Baily et al. (1992). Transverse sections of 450-kg round ba les of 12-wk regrowth of Pensacola bahiagrass (BAH) ( Paspalum notatum ), Coastal (C-B; Cynodon dactylon ) and Tifton 85 (T-85) bermudagrass ( Cynodon dactylon ) hays from the University of Florida Santa Fe Beef Research Unit were removed with a we dge cutter. Grab samples were taken from the sections and composited to give a 3-kg (a s-is) representative sample of each bale. Half of each representative sample was ground to pass through a 1-mm screen in a Wiley mill (Arther H. Thomas Company, Philadelphi a, PA, USA) and the other half was ground through a 4-mm screen for rumen degradability analysis. Half of each forage sample was ground to pass through a 1-mm screen in a Wiley mill (Arther H. Thomas Company, Philadelphia, PA, USA) and the other half was ground to pass a 4-mm screen. The enzyme was dissolved in 10 ml of double-dist illed water and sprayed at rates of 0, 0.5, 1, 2, 3 g/100 g DM onto 10-g samples of each forage. Control samples were sprayed with water alone. A 24-h enzyme-substrate interactio n period was allowed before analysis of in vitro digestion or chemical analysis. In Vitro Digestibility and Disappearance All treated samples were allowed to air dry for 8 h before being weighed (0.5 g) into ANKOM F57 (ANKOM Technologies, Mace don, NY) filter bags in triplicate. Sixteen hours later, the samples were inc ubated in buffered, rumen fluid for 6, 24, and 48 h in two ANKOM® Daisy II incubators (ANKOM Technologies, Macedon, NY) for estimation of in vitro DM disappearance. Ru men fluid was obtained before feeding from two non-lactating, ruminally-fis tulated Holstein cows fed 9 kg of bahiagrass hay supplemented with 0.4 kg of soybean meal. This experiment was repeated three

PAGE 66

54 consecutive times to account for variati ons in rumen fluid activity. The NDF concentration of post-inc ubation residues was measured and NDF disappearance (IVNDFD) was calculated. The Tilley and Te rry (1963) two-stage, rumen fluid-pepsin technique and modified by Moore and Mott ( 1974) was also used to estimate 96-h in vitro DM digestibility (IVDMD). Chemical Analysis Neutral detergent fiber (NDF) and acid dete rgent fiber (ADF) concentrations were measured in the samples and digestion resi dues using the method of Van Soest et al. (1991) modified for an ANKOM® 200 Fi ber Analyzer (ANKOM Technologies, Macedon, NY ,USA) and NDF and ADF digestibili ty were calculated. Water soluble carbohydrate (WSC) concentration was analyzed using the anthrone method described by Ministry of Agriculture, Fi sheries, and Food (1986). In Situ Rumen Degradability The effect of D670L treatment on in situ rumen DM degradability of each hay was assessed separately using the polyester bag technique. Th e enzyme was dissolved in 40 ml of double-distilled water a nd sprayed on to 225-g samples of the hays that had been ground to pass through a 4-mm screen, at rate s of 0, 0.5, 1, 2 and 3 g/100g DM. Control samples were sprayed with water alone. All tr eated samples were allowed to air dry for 8 h before being weighed (5 g as treated basi s) into nylon bags (10 x 23 cm; pore size 50 µm; Bar Diamond Inc., Parma, ID). Sixteen hours later, duplicate samples of each forage sample were incubated in each of two cows for 0, 3, 6, 9, 12, 24, 48, 72, 96, and 120 h, resulting in four replicates for each forage tr eatment at each hour. The cows used for this study were the same as those used as rume n fluid donors. The Mc Donald model (1981) was fitted to the in situ data and ruminal degradation parameters were estimated using the

PAGE 67

55 nonlinear regression procedure of SAS, vers ion 8 (2002, SAS Instit ute Inc., Cary, NC). The McDonald (1981) model is of the form: P= A + B(1-e-c(t-L)) where P= DM degraded at time t, A= wash lo ss, B= potentially degradable fraction, A+B= total degradability, c= the rate at whic h B is degraded, t= time incubated in the rumen, and L= lag time. Statistical Analysis A 3 (forages) x 5 (enzyme rates) factorial design with three rep licates per treatment was used to examine the effect of enzyme application rate on chemical composition and in vitro digestibility. The data were analyzed using the GLM procedure of SAS v8 (2002). Polynomial contrasts were used to test the effect of increasing enzyme application and contrasts were used to compare grass species (BAH v BERM) and bermudagrass cultivars (C-B v T-85). The m odel used to analyze individual treatment effects was: Yijkl: µ + Fi + Ej + Rk + FEij + FRik + ERjk + FERijk+ Eijkl where: µ= general mean Fi = forage effect of treatment Ej = enzyme rate effect Rk= run effect FEij = treatment x forage interaction FRjk = forage x run interaction ERik = enzyme x run interaction FERijk = forage x enzyme x run interaction

PAGE 68

56 Eijkl= experimental error. A 2 (cows) x 5 (enzyme rates) factorial desi gn of treatments with two replicates per treatment was used to determine the in situ degradability of each hay. The data from each hay were analyzed separately due to the fact that they were not incubated simultaneously. The data were analyzed using the GLM procedure of SAS v8 (2002). Polynomial contrasts were used to test the e ffect of increasing enzy me application rate. The model used to analyze indi vidual treatment effects was: Yij: µ + Fi + Ej + Ck + ECij + Eijkl where: µ= general mean Ei = effect of enzyme rate Cj = effect of cow ECij = enzyme x forage interaction Eij= experimental error. Significance was declared at P < 0.05 and tendencies at P < 0.10. Results Nutritive Value of Untreated Hays The crude protein of the forages rang ed from 63 to 73 g/kg DM and T-85 had greater crude protein concentration than the other hays (Table 4-1) . Neutral detergent fiber concentrations ranged from 773 to 864 g/kg DM while ADF concentrations ranged from 363 to 473 g/kg DM (Table 4-2). Forage T-85 contained the gr eatest concentration of NDF and ADF followed by BAH and C-B. Lignin concentrations ranged from 32 g/kg DM in BAH to 51 g/kg DM in C-B and T-85. The 96-h IVDMD values were greatest in

PAGE 69

57 C-B (445 g/kg DM) followed by T-85 (428 g/ kg DM) and lowest in BAH (387 g/kg DM). Cell Wall Composition The concentration of NDF differed with forage species and cultivar ( P= 0.0003, Table 4-2). The bermudagra sses contained more NDF than BAH, and T-85 contained more NDF than C-B (45g vs. 36.6 g/kg; P< 0.001). The concentration of ADF did not differ between forage species but it was greater in T-85 than in C-B ( P< 0.001). Hemicellulose concentration was greater in th e bermudagrasses than BAH, and greater in C-B than T-85. Water-soluble carbohydrate co ncentration was greate r in C-B than T-85, but unaffected by grass spp. Li gnin concentration tended to be greater in bermudagrasses than BAH, but was similar among bermudagrass cultivars. Across hays, enzyme application linearly ( P < 0.001) decreased NDF concentration and this response was unaffected by sp ecies but differed (P = 0.032) between bermudagrass cultivars. Increasing the en zyme application rate increased the ADF concentration of BAH ( P = 0.046) and T-85 ( P = 0.014), decreased the hemicellulose concentration of BAH ( P = 0.036) and increased the lignin concentration of BAH and CB. Water-soluble carbohydrate (WSC) concentrations of T-85 increased ( P = 0.044) as the enzyme application rate in creased, but those of BAH and C-B were unaffected. In vitro Disappearance and Digestibility Study There were no specie-related differences in the 6-h in vitro DM disappearance of the forages (Table 4-3), but C-B had a greate r value than T-85. Disappearance of DM in forages incubated for 24 h or 48 h were simila r across spp. and cultivars. However, 96-h IVDMD was greater in the bermudagrasses th an BAH, and similar across bermudagrass cultivars. Increasing the enzyme app lication rate resulted in a linear ( P < 0.028) increase

PAGE 70

58 in 6-h DM disappearance of BAH and C-B but not T-85. There was a tendency for a linear ( P < 0.086) increase in the 24-h DM disapp earance of C-B as enzyme application rate increased, as well as a linear (P = 0.007) increase in 48-h DM disappearance across forages. Enzyme application also produced non-linear increases (P < 0.05) in the 96-h rumen fluid-pepsin IVDMD of BAH and T-85. There were no spp.-related differences in the IVNDFD of hays incubated for 6, 24 or 48 h, but C-B had lower 24and 48-h in vitro DM disappearance values than T-85 ( P < 0.001; Table 4-4). There we re non-linear increases ( P < 0.01) in the 6-h IVNDFD concentration of BAH and C-B but not T-85 as enzyme application increased. Enzyme application did not affect the 24or 48-h IVNDFD of the hays except for a tendency for a linear (P = 0.075) increase in the 48-h IVNDFD of C-B. Enzyme Effects on In Situ Degradability Increasing the rate of enzyme applica tion increased the wash value of BAH ( P = 0.003), C-B ( P = 0.003), and T-85 ( P = 0.006; Table 4-5), slig htly increased the potentially degradable fraction of C-B ( P = 0.035), and decreased that of T-85 ( P = 0.002). Increasing the enzyme application ra te also increased the total degradable fraction of T-85 ( P = 0.036), and the effective degradability of C-B ( P = 0.044), and T85 ( P =0.042). Enzyme application did not affect th e rate of digestion of any of the hays, but increased the lag phase of T-85 (P = 0.045). Discussion The bermudagrasses used in the current study contained less CP and more NDF and ADF than the 7-wk regrowth bermudagrasse s of Mandebvu et al. (1999b) or the 10-wk regrowths of Arthington and Brown (2002) la rgely because of their maturity (12-wk regrowth). In the Southeastern Un ited States, mature poor quality C4 grasses are

PAGE 71

59 stockpiled or conserved as hay and fed to livestock during the wi nter. The advanced maturity and C4 photosynthetic pathway of such fo rages result in poo r nutritive value which limits animal performance, hence the need for improvement in quality Bahiagrass and bermudagrasses (BERM) va ry in their chemical composition. Typically BAH has a greater ADF concentratio n than the bermudagrasses (Evers et al., 2004). Mandebvu et al. (1999b) showed that 3 to 7 wk regrowth of T-85 contained greater concentration of NDF and ADF and less lignin ethe r-linked FA and P-coumaric acid concentrations than C-B at the same ma turity stages. The lower concentration of ether-linked FA in T-85 is i ndicative of fewer cross-links between hemicellulose to lignin, such that there is potenti ally less inhibition of fiber digestion by microbes. In the current study BAH contained the lowest con centration of NDF and hemicellulose, while T-85 contained the greatest NDF and C-B c ontained the greatest concentration of hemicelluose. Additionally, T-85 contained the greatest concentration of ADF followed by BAH then C-B. The phenolic acid concentrat ions of the forages were lowest for the BAH followed by C-B and then T-85. The conc entrations of cell wa ll components in the bermudagrasses in this study differed from those in the literature (Mandebvu et al., 1999a; b) due to differences in maturity. Treatment of the forages with D670L result ed in slight modifications of their cell wall concentrations. Instances where enzyme application increased cell wall concentrations probably reflect the presence of active non-fibrolytic enzymes activities or loss during handling of sugars or fine partic les hydrolyzed from the cell wall by enzymes. Krause et al. (1998) also re ported that a fibrolytic enzyme, Promote, increased the concentration of the cell wall components in whole barley ( Hordeum vuldare) grain due

PAGE 72

60 to partial hydrolysis of the cell contents pr ior to incubation in ru men fluid. The linear increase in WSC concentration of T-85 by enzyme action reflects enzymatic hydrolysis of hemicellulose prior to incubation in rumen fluid. The most noticeable effects of enzyme tr eatment on in vitro disappearance were increases of up to 27 and 40% in the 6-h in vitro DM disappear ance (IVDMD) of BAH and C-B, respectively, and slight increases of 2 and 4% in the 96-h IVDMD of BAH and T-85. The enzyme-induced increase in 6-h IVDM D can be attributed to similar increases in the 6-h in vitro NDF disappearance (Tab le 4-4). The increased 6-h IVDMD values suggest that D670L treatment in creased the initial phase of the digestion of the forages which may indicate an increased digestion ra te. Varel et al. ( 1993), Yang et al. (1999) and Colombatto et al. (2003a, b) reported that enzymes increased the rate of fiber digestion. Colombatto et al (2003b) also repo rted increases in IVDM D up to but not after 6-h of incubation, illustrating that increases in 6-h IVDMD can be indicative of increases in the rate of degradation. The 96-h IVDM D values were lower than the 48-h in vitro disappearance values largely because of losse s of undigested substrate through the pores of the bags incubated in the ANKOM vesse ls for 48 h (Wilman and Adesogan, 2000; Adesogan et al., 2005). The enzyme-induced increase in the wash-lo ss fraction supports the increase in 6 h IVDMD due to enzyme treatment and is i ndicative of pre-ruminal hydrolysis of the forage (Wang et al., 2001; 2004). The mini mal enzyme treatment effects on effective degradability are consistent with the effects of the enzyme on the extent of DM and NDF disappearance in vitro, and the e ffects of D740L on digestion of the hays (Chapter 3).

PAGE 73

61 Enzyme application did not increase the degr adation rate of the hays though values for most enzyme-treated hays were numerically greater than those fo r the control. This contradicts the earlier sugges tion that enzyme-induced increases in the 6-h IVDMD reflect increased degradation ra tes and also conflicts with pr evious reports (Feng et al., 1996; Yang et al., 1999; Nsereko et al., 2000) and the quartic ch ange in degradation rate due to D740L treatment (Chapter 3). Among ot her factors, these disc repancies are due to differences in the forage or feed type a nd enzyme composition. For instance D670L has less FAE but more cellulase and xylanase than D740L, therefore D670L would be expected to be less effectiv e at hydrolyzing phenolic cross linkages in the cell wall that impede digestion. Yu et al. (2002 a,b; 2003; 2005) had to compare several combinations of xylanase, cellulase, and esterase before fi nding one which increased the digestibility of rice straw DM by more than 10%. Colombatto et al. (2003b) also reported that the efficacy of the enzymes they tested depended on the type of feed substrate. Therefore enzyme efficacy depends on several en zymeand feed-related factors. Combinations of FAE with xylanases and cellulases have cleaved cell wall polysaccharides from C3 grasses, cereal grains and thei r by-products, but no information on effects of such enzyme combinations on the digestibility of C4 grasses was found in the literature. This st udy suggests that such enzyme comb inations can increase the initial phase of the digestion of tropi cal forages and minimally increa se the extent of digestion. The latter indicates that the FAE in the en zyme mixture did not substantially hydrolyze phenolic cross linkages in the cell wall. Theref ore, more research on conditions that are necessary to optimize the efficacy of FAE en zymes is necessary. Further research on

PAGE 74

62 ideal combinations of xylanase, cellulase and FAE for optimizing the digestion of C4 grasses is also warranted. Implications This study demonstrates that application of D670L enhanced the initial phase of the digestion of certain tropical grass hays and slightly improved the extent of digestion. However, the responses varied with forage species and cultivar and enzyme application rate. Future research should be conducted to elucidate ideal combin ations of fibrolytic enzymes for optimizing the digestion of tropical forages.

PAGE 75

63 Table 4-1. Chemical composition of hays (g/kg DM or as stated). Itema Bahiagrass Coastalb Tifton 85b DM (g/kg ) 909 916 910 Ash 59 45 48 Crude Protein 65 63 73 Lignin 32 51 51 IVDMD, 96-h 387 445 428 a DM= Dry matter, IVDMD= in vitro dry matter digestibility b Bermudagrass

PAGE 76

64Table 4-2. Effect of enzyme applicati on on cell wall and water soluble carbohydrate (WSC) concentration of tropical grass hays , g/kg. Effects Enzyme application rate Mean SEMEnzyme Foragea,b Forage*Enzymea,b Itema 0% 0.5% 1% 2% 3% Rate effectb BAH v. BERM C-B v. T-85 BAH v. BERM C-B v. T-85 NDF BAH 787 788 798 773 778 785 2.49 0.061 L C-B 794 782 782 778 777 783 2.54 NS T-85 843 858 864 850 845 852 2.21 <0.001 L Mean 808 810 816 800 800 2.43 <0.001 L <0.001 <0.001 NS 0.032 ADF BAH 424 423 427 422 424 423 1.56 0.046 L C-B 368 365 367 363 364 366 1.56 NS T-85 429 468 473 463 463 459 3.62 0.014 L Mean 407 418 422 416 417 2.39 0.002 L NS <0.001 NS 0.006 Hemicellulose BAH 363 366 372 351 354 361 1.49 0.036 Q C-B 426 418 415 414 412 417 1.41 NS T-85 414 390 392 387 382 393 3.72 NS Mean 401 391 393 384 383 3.09 0.059 Q <0.001 <0.001 NS 0.079 WSC BAH 6.0 5.7 6.7 5.8 8.6 6.6 0.81 NS C-B 8.9 6.7 7.9 7.4 9.7 8.1 1.15 NS T-85 3.6 6.4 3.7 6.1 7.7 5.5 0.48 0.044 L Mean 6.2 6.3 6.1 6.4 8.7 1.11 NS NS 0.038 NS NS Lignin BAH 32.2 35.4 39.9 33.2 51.1 38.3 2.98 0.074 C C-B 51.4 82.3 77.5 69.0 67.2 69.5 2.89 0.012 L T-85 51.2 59.5 55.9 48.5 49.8 52.9 3.83 NS Mean 44.8 59.5 57.8 49.1 56.1 4.41 NS 0.093 NS NS 008 a BERM = bermudagrass cultivars, BAH = bahiagrass, CB = coastal bermudagrass, T85 = Tifton 85 bermudagrass b L = Linear, Q = Quadratic, C = Cubic, NS = not significant

PAGE 77

65Table 4-3 Effect of D670L application on the 6-, 24and 48-h rumen fluid disappearance and 96-h rumen-fluid pepsin digestibil ity of tropical grass hays, g/kg DM. Effects Enzyme application rate Mean SEMEnzyme Foragea,b Forage*Enzymea,b Itema 0% 0.5% 1% 2% 3% Rate effectb BAH v. BERM C-B v. T-85 BAH v. BERM C-B v. T-85 6 h BAH 135 155 169 188 178 165 3.48 0.009 L C-B 176 195 204 205 224 201 3.71 0.028 L T-85 103 110 122 143 146 125 3.24 NS Mean 138 154 165 179 182 4.49 0.001 L NS <0.001 0.052 0.030 24 h BAH 367 395 389 399 408 392 6.43 NS C-B 363 410 390 413 420 399 6.73 0.086 L T-85 373 368 376 388 389 379 4.24 NS Mean 368 391 385 399 406 7.62 NS NS NS NS NS 48 h BAH 494 502 513 520 521 510 3.35 NS C-B 480 487 496 504 512 496 5.57 NS T-85 486 478 481 508 506 492 3.51 0.016 L Mean 487 489 497 511 513 5.51 0.007 L NS NS NS NS 96 h BAH 387 396 395 396 379 390 5.38 0.018 Q C-B 445 446 441 434 439 441 3.29 NS T-85 428 436 446 431 446 437 1.78 0.033 C Mean 420 426 427 420 421 3.37 NS <0.001 NS 0.064 0.038 a BERM = bermudagrass cultivars, BAH = bahiagrass, CB = coastal bermudagrass, T85 = Tifton 85 bermudagrass b L = Linear, Q = Quadratic, C = Cubic, NS = not significant

PAGE 78

66 Table 4-4. Effect of D670L applic ation on in vitro NDF disappearance (g/kg) of tropical grass hays. Effects Enzyme application rate Mean SEMEnzyme Foragea,b Forage*Enzymea,b Itema 0% 0.5% 1% 2% 3% Rate effectb BAH v. BERM C-B v. T-85 BAH v. BERM C-B v. T-85 6 h BAH 15 26 53 37 37 34 3.18 <0.001 Q C-B 46 41 51 38 59 47 2.89 0.005 Q T-85 19 17 30 41 34 28 3.57 NS Mean 27 28 45 39 43 3.21 <0.001 Q NS NS NS 0.077 24 h BAH 270 300 300 297 302 293 6.21 NS C-B 254 294 272 289 304 283 7.44 NS T-85 303 306 316 330 315 314 4.18 NS Mean 276 300 296 305 307 6.09 NS NS <0.001 0.049 NS 48 h BAH 417 422 435 434 442 430 2.99 NS C-B 397 392 403 409 422 405 4.79 0.075 L T-85 414 433 437 457 450 438 6.74 NS Mean 409 416 425 433 438 5.08 NS NS <0.001 0.095 NS a BERM = bermudagrass cultivars, BAH = bahiagrass, CB = coastal bermudagrass, T85 = Tifton 85 bermudagrass b L = Linear, Q = Quadratic, C = Cubic, Qt = Quartic, NS = not significant

PAGE 79

67 Table 4-5. Effect of D670L a pplication on the in situ rumen DM degradability of tropical grass hays. Enzyme application rate Mean SEMEnzyme Itema 0% 0.5% 1% 2% 3% rate effectb Wash Value (A), g/kg DM BAH 86 76 93 98 84 87 3.72 0.003 C C-B 88 75 70 78 97 81 1.89 0.001 Q T-85 73 82 92 119 90 91 4.99 0.006 C Potentially degradable fraction (B), g/kg DM BAH 571 570 551 544 586 565 16.2 NS C-B 525 532 537 529 515 528 6.52 0.035 Q T-85 614 607 575 591 573 592 7.95 0.002 L Total degradable fraction (A+B), g/kg DM BAH 657 646 644 642 670 652 6.27 NS C-B 612 606 607 607 612 609 5.75 NS T-85 687 689 667 712 664 683 8.00 0.036 C Effective degradability, g/kg DM BAH 282 276 308 281 277 285 15.9 NS C-B 297 304 299 292 318 302 6.46 0.044 C T-85 323 328 369 334 358 342 10.9 0.042 L Fractional degradation rate (c), %/h BAH 2.39 2.45 3.32 2.47 2.27 2.58 0.41 NS C-B 3.20 3.50 3.38 3.22 3.51 3.36 0.19 NS T-85 3.24 3.24 4.62 2.73 4.04 3.57 0.39 NS Lag phase (L), h BAH 5.75 6.30 7.54 4.56 6.20 6.07 0.57 NS C-B 3.75 4.43 4.11 3.52 4.52 4.06 0.47 NS T-85 3.83 4.18 5.25 2.65 7.81 4.74 0.93 0.045 C a BERM = bermudagrass cultivars, BAH = bahiagrass, C-B = coastal bermudagrass, T-85 = Tifton 85 bermudagrass b L = Linear, Q = Quadratic, C = Cubic, Qt = Quartic, NS = not significant

PAGE 80

68 CHAPTER 5 THE EFFECT OF DIFFERENT COMBINATIONS OF FERULIC ACID ESTERASE, CELLULASE AND XYLANASE ON THE NUTRITIVE VALUE OF MATURE BAHIAGRASS Introduction Tropical grasses are typica lly of poor quality because they contain higher concentrations of phenolic compounds and lignin than temperate grasses (Akin, 1986). Yet such C4 grasses are the mainstay of ruminant livestock production in warm climates. The cross-linking of lignin with cell wall polysaccharides th rough ferulic acid bridges is thought to be the mechanism by which lignin limits cell wall digestion in plants (Jung and Allen, 1995). Ferulic acid esterases (FAE ) are known to release ferulic acid (FA) from arabinose side chains of hemicellulose (Faulds and Williamson, 1994), which allows for further degradation of the cell wa ll by other polysaccharid ases. The addition of FAE to other cell wall-degrading enzyme s, like xylanase and cellulase, produces a synergistic effect on the de gradation of plant cell walls (Faulds and Williamson 1995; Bartolome et al., 1997; Yu et al., 2002a, ;, 2003; 2005), due to increased accessibility to digestible cell wall components by rumen micr oorganisms. Previous research (Chapters 3 and 4) showed that commercially-available FAE enzymes alsocontaining other enzyme activities, improved the initial phase of the digestion of C4 grasses and slightly increased or did not affect the extent of digestion. Therefore there is a need to determine the best ratio of FAE and other fibrolytic enzymes fo r optimizing the digesti on and utilization of C4 grasses in ruminant livestock. The objectives of this study were to determine the best combination of FAE, cellulase and xylanase for hydrolyzing mature bahiagrass in the

PAGE 81

69 absence of rumen fluid, and to evaluate the effects of the best multienzyme cocktails on the digestibility and fermentation of the forage in the presence of rumen fluid. Materials and Methods Forage and Enzymes A representative sample (3 kg) of 12-wk regrowth of Pensacola bahiagrass (BAH) ( Paspalum notatum ) was sampled from 450-kg round bales at the University of Florida, Santa Fe Beef Research Unit, Alachua, FL. Th e forage was the same bahiagrass used in Chapters 3 and 4. It contained 909 g/kg DM , 65 g/kg DM of crude protein, 814 g/kg DM NDF and 61 g/kg DM lignin. The 3-kg sa mple was ground to pass through a 1-mm screen in a Wiley mill and stored in air-tight plastic bags. Pure FAE from Clostridium thermocellum and xylanase A (XYL) from Orpinomyces spp. were obtained from Dr. Shin Li of the National Center for Agricultural Utilization Research-USDA-ARS (Peoria, IL). The FAE was prepared according to Blum et al. (2000) and had an activity of 10.5 U/ml, where one unit of activity is defined as micromoles of ferulate released per minut e from 2 mM methyl ferulate at pH 6.5 and 50oC. The XYL was prepared according to Li et al. (1997) and had an activity of 5500 U/ml, where one unit of activity is defined as micromoles of xylose released per minute from 1% oat spelt xylan substrate at pH 6.5 and 50oC. Cellulase (CEL) from Aspergillus spp. was obtained from Sigma (product # C2605, Sigma–Aldrich, St Louis, MO, USA). This product had an activity of 5259 U/g. Experiment 1: Enzymatic DM Disappearance In order to determine the best multienzyme cocktail for hydrolyzing mature BAH cell walls, CEL, XYL and FAE were mixed in different ratios a nd evaluated. Each enzyme was applied at 0, 0.5, 1 and 2 g per 100 g of forage DM and all possible

PAGE 82

70 combinations of the enzymes and the four app lication rates were eval uated (Table 6.1). Each multienzyme combination was dissolved in 25 ml of citrate-phosphate buffer (pH 6.0) and added to 100-ml beakers containing 0.5 g of bahiagrass hay in triplicate. In addition to the buffered-multienzyme mixture, 12.5 ml of double-distilled water and 12.5 ml of sodium azide solution (0.4% w/v) was added to each beaker and stirred. Control samples received 25 ml of buffer, 12.5 ml of double-distilled water, and 12.5 ml of sodium azide solution. The sodium azide solution was added to prevent microbial degradation of the substrate. All samples were incubated in a thermostatically controlled incubator at 39oC overnight (24 h) and then filt ered through preweighed filter paper (Paper no. Q8, Fisher Scientific, Pittsbur gh, PA) under low suction. The residue was rinsed with 50 ml of double-distilled water to remove the solubilised material, dried overnight at 105oC and weighed to determine the percent DM disappearance. All treatments were evaluated in triplicate and the experiment was repeated three times yielding nine replic ates per treatment. Experiment 2: In Vitro Fermentation and Degradability This experiment compared the effects of the two most promising multienzyme cocktails from the previous experiment a nd D740L (BioCatalyst, Pontypridd, Wales, UK) on the in vitro fermentation and digestion of BAH in rumen fluid. Treatments included a control (no enzyme), D740L (1g per 100 g of DM), and multienzyme cocktails 2-2-0 and 2-2-1 which contained 2% C ELL, 2% XYL and 0% or 1% FAE, of forage DM respectively. The D740L contained 32 U/ml FAE activity, 20.78 filter paper units of cellulase activity, and 8701.85 U/ml of xylanase activity as de scribed in Chapter 3. The in vitro DM degradation and ferm entation of enzyme-treated BAH was measured using the automated, wireless gas pr oduction system of Adesogan et al. (2005).

PAGE 83

71 Each enzyme cocktail was diluted into 1 ml of citrate phosphate buffer (pH 6.0) and added to 250 ml culture bottles ( # 064141B Fi sher Scientific, Pittsburgh, PA) containing 0.5 g of bahiagrass. Rumen fluid was collect ed by aspiration before the morning feeding from two non-lactating, non-pr egnant, Holstein cows fed Coastal bermudagrass hay ad libitum and 900 g/day of soybean meal. The ru men fluid was filtered through two layers of cheesecloth and immediately transported in a pre-warmed thermos flask to the laboratory where it was mixed (1:2 ratio) under a CO2 stream with a warm (39oC) McDougal's artificial saliva (Tilley and Terry 1963). For each treatment, 40 ml of the rumen fluid/artificial saliva solution was added to the contents of each culture bottle in quintuplicate, under a CO2 stream while the bottles were in a 39oC water bath. Each bottle was fitted with a pressure sensor/bottle cap assembly and placed in an incubator at 39oC. Pressure sensors were set to measure pressure hourly for durations of 24 or 96 h. After the fermentation, contents of the cultu re bottles were filter ed through Whatman no. 541 filter paper (# 09851D Fisher Scientific, Pi ttsburgh, PA). The residues were dried at 60oC overnight to determine DM digestibility and a 5 ml sub sample of filtrate was collected and frozen for later VFA analys is. The neutral detergent fiber (NDF) concentration of the substrates and digestion residues were analyzed using the method of Van Soest et al. (1991) in an ANKOM® 200 Fiber Analyzer (ANKOM Technologies, Macedon, NY, USA) and NDF dige stibility was calculated. The frozen filtrate was centrifuged at 4oC at 12000 rpm (8000 x g) for 15 min. The supernatant was analyzed for VFA using a High-Performance Liquid Chromatograph (Hitachi®, FL 7485, Tokyo, Japan) coupled to an auto sampler (Hitachi®, L 7200, Tokyo, Japan) and a UV Detector (Spectrofl ow 757, ABI Analytical Kratos Division,

PAGE 84

72 Ramsey, NJ) set at 210 nm. Twenty µl of supernatant was injected into a Bio-Rad Aminex HPX-87H (Bio-Rad Laboratories, He rcules, CA) column w ith a 0.015M sulfuric acid mobile phase and a flow rate of 0.7 ml/min at 45°C. A linear model was fitted to the 24-h gas production data using the regression procedure of SAS v8 (2002, SAS Institute Inc., Cary, NC). The model is of the form: Y = A + Bx where Y = gas pressure at time x, A= immediat ely fermentable fraction, B= rate of fermentation, x= time incubated. The exponential model of McDonald (1981) was fitted to the 96-h gas production data using th e nonlinear procedure of SAS v8 (2002; SAS Institute Inc., Cary, NC). The model is of the form: P = A + B(1-e-c(t-L)) where P = gas pressure at time t, A= immediatel y fermentable fraction , B= potentially fermentable fraction, A+B= total fermentability, c= the rate at which B is fermented, t= time incubated, and L= lag time. Statistical Analysis A completely randomized design with three replicates per treatment and three runs was used to determine the effects of the multienzyme cocktails on DM disappearance in Experiment 1. A similar design with six rep licates per treatment and two runs was used for the in vitro gas fermentation kinetics, DM and NDF digestibility and VFA production in Experiment 2. The GLM procedure of SAS v8 (2002) was used to analyze the data. The model was: Yijk: µ + Ti + Rj + TRij + Eijk where: µ = general mean

PAGE 85

73 Ti = treatment effect Rj = run effect on treatment RTij = treatment x run interaction Eijk= experimental error The DM disappearance resulting from each multienzyme treatment in Experiment 1 was compared to that of the control treat ment using the DunnetÂ’s test. Contrast statements were used to compare the gas production, VFA, and digestibility data for treated versus control sa mples in Experiment 2. Results Experiment 1: Enzymatic DM Disappearance The effects of the multienzyme cocktails on the DM disappearance of BAH are shown in Table 5-1. The greatest increas es in DM disappearance occurred with treatments numbered 22 (2-2-0) and 24 (2-2-1 ) which contained 2% of CEL and 2% of XYL with 0% or 1% of FAE, respectively. Treatment number 11 which contained 1% CEL, 2% XYL and 0.5% FAE increased the en zymatic DM disappearance of the forage to a lesser degree. Experiment 2: In vitro Ferm entation and Digestibility After 24 h of substrate inc ubation in rumen fluid all en zyme treatments decreased ( P < 0.001) acetate concentration and increased ( P < 0.05) the propionate concentration, resulting in a decrease ( P < 0.0001) in acetate to propionate ratio (A:P) (Table 5-2). Isobutyrate and iso-valerate con centrations were increased ( P < 0.01) by the 2-2-1 and D740L treatments, and all enzyme treatments increased ( P < 0.01) butyrate and valerate concentrations. Acetate, pr opionate, A:P, and butyrate con centrations were unaffected ( P > 0.05) by treatment in forages incubated for 96h (Table 5-3). However, 2-2-0 tended to

PAGE 86

74 decrease ( P = 0.0679) acetate concentration and D740L tended to decrease acetate concentration ( P = 0.0809) and increase propionate ( P = 0.0799) concentration. Treatment 2-2-1 increased the iso-butyrate, is o-valerate and valerate concentrations. Enzyme application of 221 or D740L tended to decrease ( P > 0.05) 24 h DM digestibility (DMD) whereas 220 and D740L decreased DMD at 96 h(Table 5-4). All enzyme treatments also decreased ( P < 0.05) the NDF digestibility (NDFD) of hays incubated for 24 h (Table 5-5). Treatment with 2-2-0 and D740L also decreased NDFD of samples incubated for 96 h (Table 5-5). Application of D740L decreased ( P = 0.02) the fermentation rate of samples incubated for 24 h (Table 5-6). All enzyme treatments decreased ( P < 0.0001) the lag phase of such forage samples, but 2-2-1 increased the gas pool volume ( P = 0.005). All enzyme treatments increased ( P < 0.05) the immediately ferm entable fraction of samples incubated for 96 h, but the potentially fermenta ble fraction and fermentation rates of such forage samples were unaffected by enzyme tr eatment (Table 5-7). The total fermentable fraction and lag phase of samples in cubated for 96 h were decreased ( P < 0.05) by D740L and 2-2-0, respectively. Discussion Limited research has been conducted on improving the digestibility of mature tropical hays with enzymes. Such poor quality forage is sometimes all that is available for winter-feeding of beef cattle in the south east and many tropical countries. This study indicates that the DM disappearance in the absence of rumen fluid and the fermentation of BAH in rumen fluid were improved by a ppropriate combinations of CEL, XYL and FAE, though the rate and extent of digestion in rumen fluid were not improved. The 2-21 multienzyme cocktail resulted in the gr eatest DM disappearance after 24 h of

PAGE 87

75 incubation. This supports previous work th at states that FAE and other cell walldegrading enzymes can synergistically act to disrupt the ferulic acid-cross-linked complex cell wall polysaccharides in fo rages, making cell wall-bound digestible carbohydrates more accessible to fibrolytic enzymes (Yu et al., 2002a, b; 2003; 2005). The fact that the increase in DM disappearan ce occurred in the absence of rumen fluid indicates that this enzyme mixture can hydrolyze bermudagrass cell walls prior to ingestion and digestion. Though th is suggests that the 2-2-1 treatment can be used as an additive for digestibility, the magnitude of the improvement was too small for practical adoption. Volatile fatty acid concentrations were altered by addition of multienzyme cocktails to the BAH forage. The lower A:P ratio of enzyme-treated substrate after 24-h incubation indicates that the enzymes made the fermentation more gluconeogenic as opposed to lipogenic, and hence improved the ef ficiency of the fermentation This was likely due to fermentation of sugars releas ed by enzymatic cell wall hydrolysis. This result also agrees with Dawson and Tricaric o (1999) who noted th at enzyme mixtures containing cellulase were more inclined to alter the relative proporti ons of VFA resulting in greater propionate and butyrate production an d less acetate. That enzyme effects on VFA production were less evident after 96 h of incubation reflect s the occurrence of microbial lysis which can confound VFA data. The pre-treatment of the forage with the three multienzyme cocktails prior to incubation did not increase the DMD or NDFD of the BAH after 24 h or 96 h of incubation in rumen fluid. This contradicts the effects of D740L on DM digestibility in Chapter 3. This discrepancy may be partly technique related because samples were not

PAGE 88

76 agitated during digestion in the gas production system used in this Chapter, but they were agitated to simulate peristalsis in the ANKOM digestibility technique used in Chapter 3. A more important argument is that this ex periment did not replicate the 24-h enzyme substrate interaction time used in the DM disappearance a ssessment of Experiment 1 and in Chapters 3 and 4. Lewis et al. (1996) found that ruminal infusion of enzymes produced lower disappearance of DM and NDF th an enzyme application to the forage 24h prior to feeding, and suggested that this was due to insufficient contact between the enzymes and the particulate substrate. Simila rly, McAllister et al . (1999) evaluated the effects of applying a mixture of two commerci al enzymes to silage before feeding or ruminally infusing the enzyme mixture. Th ey found that while th e application of the enzyme to the silage did not alter the DM digestibility of sheep, the ruminal infusion treatment decreased DM and NDF digestibility. In such cases, enzymes can coat substrate particles and hinde r access of microbes and thei r enzymes, thereby reducing digestion. The enzyme-induced increase in the immedi ately fermentable fraction in enzymetreated substrates agrees with enzyme effects in Chapters 3 and 4, and is indicative of preruminal enzymatic hydrolysis (Lewis et al., 1996; Wang et al., 2001, 2004; Colombatto et al., 2003a, b; Yu et al., 2005). This is supported by the shorter lag phase of enzyme-treated samples, which indicates that enzyme treatment allowed digestion to commence earlier, possibly due to pre-ruminal release of phenolic ac ids and greater or faster substrate colonization by ruminal mi crobes. That only 2-2-1 increased ( P < 0.05) the gas pool volume of substrat es incubated for 24 h suggest s that 2-2-1 was the most effective treatment, but neither this treatment nor any of the other two enzyme treatments

PAGE 89

77 improved 96-h gas pool volume. Therefore th e gas production and digestibility data indicate that the extent of substrate dige stion was not improved by enzyme treatment. This maybe attributable to the lack of ag itation of samples during the fermentation, and the lack of an adequate enzyme-substrate in teraction time prior to incubation in rumen fluid (Lewis et al, 1996; McAllister et al., 1999). Therefore, in order to ascertain the effects of the multienzyme cocktails on the nutritive value of bahiagrass future studies should ensure an adequate preincubation or preingestion enzyme-substrate interaction time, and substrates should be agitated during fermentation in vitro. Implications This study demonstrates that certain co mbinations of FAE with cellulases and xylanases are more effective at increasing the DM disappearance of bahiagrass in the absence of rumen fluid than any of the enzymes alone. Application of such multienzyme cocktails to bbahiagrass resulted in a more efficient fermentation in rumen fluid, an increased immediately fermentable fracti on and a decreased lag phase, but did not improve the extent of forage digestion. Since the latter was attributed to lack of sample agitation during the fermentation and inadequate enzyme-substrate interaction prior to substrate incubation, future stud ies that address th ese factors are needed to validate the promise of the multienzyme cocktails for improving the quality of tropical grasses.

PAGE 90

78 Table 5-1. Effect of multienzyme cocktail composition on DM disappear ance of bahiagrass hay Multienzyme cock tail composition, % of DM Treatment no.a CEL XYL FAE DM disappearance, %b Control 0 0 0 13.6 25 1 0 0 13.8 26 1 0 0.5 14.2 27 1 0 1 13.4 28 1 0.5 0 14.2 29 1 0.5 0.5 14.4 30 1 0.5 1 14.0 31 1 1 0 13.6 32 1 1 0.5 13.7 33 1 1 1 14.3 34 1 2 0 13.7 35 1 2 0.5 15.1* 36 1 2 1 14.2 37 2 0 0 13.9 38 2 0 0.5 13.8 39 2 0 1 14.9 40 2 0.5 0 13.7 41 2 0.5 0.5 14.1 42 2 0.5 1 13.3 43 2 1 0 14.7 44 2 1 0.5 14.9 45 2 1 1 15.0 46 2 2 0 15.3* 47 2 2 0.5 13.9 48 2 2 1 15.4* XYL=xylanase A, CEL= cellulase, FAE= ferulic acid esterase a Treatments 1-24 were did not differ ( P > 0.05) from the control and are not shown. b * Mean was different from control at P < 0.05

PAGE 91

79Table 5-2. Effect of enzyme treatment on concentration of vol atile fatty acids produced from fermentation of bahiagrass hay in rumen fluid/ McDougals buffer for 24 h (molar %). Treatment Acetate Propionate A:P ratio Iso-Butyrate Butyrate Iso-Valerate Valerate Control 56.0 17.6 3.2 5.5 10.7 5.6 4.4 2-2-0 53.1 18.7 2.8 6.1 11.3 5.9 4.8 2-2-1 52.6 18.4 2.9 6.5 11.5 6.2 4.8 D740L 52.1 17.9 2.9 7.0 11.8 6.3 4.8 SE 0.40 0.11 0.02 0.22 0.10 0.13 0.07 P-values Treatment effect <0.0001 <0.0001 <0.0001 0.0021 <0.0001 0.0180 0.0027 Contrasts 2-2-0 v Control 0.0002 <0.0001 <0.0001 0.0926 0.0018 0.1633 0.0032 2-2-1 v Control <0.0001 0.0001 <0.0001 0.0072 <0.0001 0.0117 0.0031 D740L v Control <0.0001 0.0401 <0.0001 0.0003 <0.0001 0.0039 0.0005 a Control, no enzyme application 2-2-0: Cellulase, Xylanase, Feruli c acid esterase concentrations were 2%, 2% and 0% of forage DM 2-2-1: Cellulase, Xylanase, Fe rulic acid esterase concentrations we re 2%, 2% and 1% of forage DM D740L: D740L applied at 1% of forage DM

PAGE 92

80Table 5-3. Effect of enzyme tr eatment on concentration of volat ile fatty acids produced from fe rmentation of bahiagrass in rum en fluid for 96 h (molar %) Treatment Acetate Propionate A:P ratio Iso-Butyrate Butyrate Iso-Valerate Valerate Control 51.3 23.5 2.2 5.8 8.4 5.8 5.2 2-2-0 49.9 24.7 2.0 5.9 8.5 5.9 5.1 2-2-1 51.1 22.9 2.3 6.3 8.7 6.1 4.9 D740L 50.0 24.5 2.1 5.9 8.6 5.9 5.1 SE 0.45 0.62 0.08 0.88 0.10 0.07 0.09 P-values Treatment effect 0.1358 0.1782 0.1701 0.0046 0.1962 0.0271 0.2599 Contrasts 2-2-0 v Control 0.0679 0.4507 0.2193 0.3646 0.7939 0.4337 0.5496 2-2-1 v Control 0.7959 0.2890 0.4528 0.0008 0.0524 0.0050 0.0576 D740L v Control 0.0809 0.0799 0.2695 0.2799 0.3393 0.4187 0.4088 a Control, no enzyme application 2-2-0: Cellulase, Xylanase, Feruli c acid esterase concentrations were 2%, 2% and 0% of forage DM 2-2-1: Cellulase, Xylanase, Fe rulic acid esterase concentrations we re 2%, 2% and 1% of forage DM D740L: Depol 740L applied at 1% of forage DM

PAGE 93

81 Table 5-4. Effect of enzyme treatment on dry matter digestib ility (%) of bahiagrass after 24 or 96 hr of incubation in ru men fluid and McDougals buffer.. Treatmenta 24 h 96 h Control 20.6 50.8 2-2-0 17.6 45.8 2-2-1 16.4 50.4 D740L 17.0 46.5 SE 1.4 0.6 P-values Treatment effect 0.2278 0.0005 Contrasts 2-2-0 v Control 0.1529 0.0004 2-2-1 v Control 0.0659 0.6895 D740L v Control 0.0943 0.0008 a Control, no enzyme application 2-2-0: Cellulase, Xylanase, Ferulic acid es terase concentrations were 2 %, 2 % and 0 % of forage DM 2-2-1: Cellulase, Xylanase, Ferulic acid esterase concentrations were 2 %, 2 % and 1 % of forage DM D740L: Depol 740L applied at 1 % of forage DM

PAGE 94

82 Table 5-5 Effect of enzyme treatment on ne utral detergent fiber digestibility (%) of bahiagrass. Treatmenta 24 h 96 h Control 19.8 53.8 2-2-0 16.3 49.0 2-2-1 13.5 52.9 D740L 12.2 49.1 SE 0.9 0.5 P -values Treatment effect 0.0008 0.0003 Contrasts 2-2-0 v Control 0.0262 0.0002 2-2-1 v Control 0.0007 0.2710 D740L v Control 0.0002 0.0003 a Control, no enzyme application 2-2-0: Cellulase, Xylanase, Ferulic acid esterase concentrations were 2 %, 2 % and 0 % of forage DM 2-2-1: Cellulase, Xylanase, Ferulic acid esterase concentrations were 2 %, 2 % and 1 % of forage DM D740L: Depol 740L applied at 1 % of forage DM

PAGE 95

83 Table 5-6 Effect of enzyme treatment on ferm entation kinetics of bahiagrass incubated in rumen fluid for 24 h. Treatmenta Rate %/h Lag, h Total Gas Pool ml/g DM Control 1.18 2.2 45.6 2-2-0 1.13 0.5 49.3 2-2-1 1.05 0.0 53.5 D740L 0.97 0.0 48.6 SE 0.05 0.2 1.5 P -valuesb Treatment effect 0.1210 <0.0001 0.0314 Contrasts 2-2-0 v Control 0.5455 <0.0001 0.1595 2-2-1 v Control 0.1329 <0.0001 0.0047 D740L v Control 0.0232 <0.0001 0.2331 a Con: Control, no enzyme application 2-2-0: 2 % Cellulase, 2 % Xylanase, 0 % Ferulic acid esterase 2-2-1: 2 % Cellulase, 2 % Xyla nase, 0.5 % Ferulic acid esterase D740L: Depol 740L applied at 1 %

PAGE 96

84 Table 5-7 Effect of enzyme treatment on ferm entation kinetics of bahiagrass incubated in rumen fluid for 96 h. Variablea, ml/g DM or as stated Treatmentb A B Gas Pool c, %/h L, h Control 0.0 171.4 174 1.6 1.2 2-2-0 3.9 178.0 179 1.5 0.5 2-2-1 5.2 155.4 167 1.5 1.1 D740L 5.6 155.3 157 1.9 0.9 SE 1.4 7.0 5.5 0.1 0.2 P values Treatment effect 0.0095 0.1045 0.0788 0.1024 0.0672 Contrasts 2-2-0 vs Control 0.0120 0.5112 0.4829 0.5518 0.0172 2-2-1 vs Control 0.0040 0.1580 0.3736 0.7761 0.7237 D740L vs Control 0.0029 0.1268 0.0436 0.0798 0.3018 a A= immediately fermentable fraction, B= potentially fermentable fraction, A+B= gas pool, c= fr actional degradation rate, L= lag phase b Con: Control, no enzyme application 2-2-0: 2 % Cellulase, 2 % Xylanase, 0 % Ferulic acid esterase 2-2-1: 2 % Cellulase, 2 % Xyla nase, 0.5 % Ferulic acid esterase D740L: Depol 740L applied at 1 %

PAGE 97

85 CHAPTER 6 EFFECT OF APPLYING AMMONIA OR FIBROLYTIC ENZYMES TO BERMUDAGRASS HAY ON FEED INTAKE, DIGESTION KINETICS AND GROWTH OF BEEF STEERS Introduction Most ruminant livestock pr oduction relies heavily on fora ge-based diets. During the winter the quality and quantity of th e forage available for grazing limits cattle productivity in the southern USA. Supplem entation of the forage is often costly, sometimes impractical and results in net im portation of nutrients to the farm. An alternative method of enhancing the productivity of cattle in the winter is to improve the quality of the forage being fed. Chemical treatment has been used successfully to improve the quality of warm-season grasses (Klopfenstein, 1978; Kunkle et al., 1983 and 1984; Brown, 1988; Rasby et al., 1989; Brow n and Kunkle, 1992; Brown and Adjei, 1995; Brown and Pate, 1997). However, there ar e several problems associated with using chemical treatments. For example the caustic nature of ammonia n ecessitates the use of special protective clothing and causes excessive wear oj equipment. In recent years, the use of exogenous fibrolytic enzymes to im prove feed digestion and utilization has received a great deal of research intere st (Beauchemin et al., 2003). Studies have reported increases in DM digestion in situ and in vivo (Feng et al., 1996; Yang et al., 1999) and voluntary intake (F eng et al. 1996; Yang et al., 1999; Pinos-Rodriguez et al., 2002) when enzymes were added to beef and dairy cattle diets. Although some studies also reported improved animal performance due to enzyme treatment of diets (Feng et al., 1996; Beauchemin et al., 1999) others ha ve not (ZoBell et al., 2000). These

PAGE 98

86 inconsistencies can be attributed to differe nces in enzyme type, preparation, activity, application rate (Bowman et al ., 2002; Beauchemin et at., 2003) mode of application, and portion of the diet to which the enzyme is applied (Feng et al., 1996; Lewis et al., 1996; ZoBell et al., 2000). The majority of the rese arch done on fibrolytic enzyme treatment of feeds has been carried out on temperate forage species and diets. Thus, little is known about the potential for improving the quality of warm-season grasses with enzymes. The three previous chapters have examined the effects of using feru lic acid esterases in combination with xylanases, cellulases and endogalacturonases on improving the degradability of tropical forages. Due to th e limited availability of these products for use in animal feeds and the costs associated w ith their current produc tion, these products are cost prohibitive for in vivo evaluation. However, there are numerous commercial enzyme products containing mainly xylanases and cellulases availabl e that have been developed for the feed industry that can provide a more economical product for evaluation. Therefore, the objec tive of this study was to ex amine the effects of various methods of applying a comm ercial fibrolytic enzyme mixture or ammonia to bermudagrass hay on the feed intake, digestion kinetics and growth performance of beef steers. Materials and Methods Forage Treatments Five-week-old fall regrowth of bermudagrass ( Cynodon spp.) was harvested from a 40-ha field owned by a local hay producer in Alachua County, FL. Before the harvesting process began, the field was divided into treatments using a measuring wheel to determine field and plot sizes. Harvesting occurred over a two consecutive-day period, where approximately 2 ha of each treatment was harvested each day and made into 18 to

PAGE 99

87 20 kg square hay bales. The forage was st ored without treatment (control) or after treatment with either a fibrolytic enzyme mixture (Biocellulase A20™ (A20), Loders Croklaan Channahon, IL, USA) or anhydrous ammonia. The enzyme contained 1400 U/g of enzyme activity, where 1 unit of activ ity is defined as the amount of enzyme required to release one micromole of glucos e reducing sugar equivalents per minute at pH 4.5 and 40oC. The enzyme was sprayed (15 g/t on) on the forage immediately after harvesting (Ec) with a New Holland 617 mower c onditioner, just before baling (Eb) after a 72-h wilt with a New Holland 385 square baler or just prior to feeding (Ef). For the Ec and Eb treatments the enzyme was applied at a flow rate of 2.7 l/min, using a tractor mounted 57-L continuous flow sprayer (FIM CO, North Sioux City, SD) fitted with a three-nozzle boom. The height of the nozzl e from the ground was set to 406 mm which allowed the spray from the boom to cover th e entire windrow. The Ef treatment was applied with a 0.5-L handheld pump type sp rayer immediately after weighing the daily forage allocation for each steer. For the am monia treatment, square bales were stacked on wooden pallets, covered with 6-mil plastic , and treated with anhydrous ammonia (3% of DM) as described by Brown (1990) and Brown and Kunkle (1992). The forage was allowed to react with the ammonia for six w eeks and then vented to release the ammonia gas. Cattle and Diets Fifty Angus-Brangus cross steers (250-270 kg BW) were stratified by body weight (BW) and randomly assigned to the five fora ge treatments from each weight group. The steers within each treatment were randomly assigned to one of seven pens such that there were seven steers in each of six pens and ei ght steers in one pen. Within each pen, the steers were randomly assigned to one of eight Calan gates for individual animal feeding.

PAGE 100

88 Steers were fed ad libitum (110% of the prev ious days intake) amounts of the hays and a basal concentrate at a constant rate of 1% (as-is) of BW twice daily (0900 h and 1500 h). The concentrate allowance was adjusted weekly as the BW changed. Diets (Table 6-1) were formulated to meet National Research Council (2000) requirements. Each forage was chopped to 15 cm lengths using a tub hay grinder (Roto Grind, model 760, Burrows enterprises, Greeley, CO) before feeding. The ingredients in the concentrate included 23% solvent extrac ted cotton seed meal and 77% soybean hull pellets. The neutral detergent fiber (NDF), acid detergent fiber (ADF), crude protein (CP), and total digestible nut rients (TDN) concentrations of the cottonseed meal and soybean hull pellets were 323, 181, 404, 664 g/kg dry matter (DM) and 625, 450, 134, 673 g/kg DM respectively. A mineral-salt mi x (UF University special Hi-CU mineral mix; Lakeland Animal Nutrition, Lakeland, FL) was offered freechoice and drinking water was offered ad libitum. The mix cont ained 13% Ca, 6% P, 21% NaCl, 0.8% K, 1% Mg, 0.4% S, 0.4% Fe, 2000 ppm Cu, 200 pp m Co, 2,200 ppm Mn, 175 ppm I, 48 ppm Se, 9,500 ppm Zn, 800 ppm F, 45,454.54 IU / kg of vitamin A and 9,090.90 IU / kg of vitamin D3. The concentrate portion of the diet was offered before the morning feeding of the hay in a separate cont ainer to each steer and it was consumed immediately by the steers. Before feeding, diet refusals from the previous dayÂ’s offering were collected. Sampling and Analysis The trial consisted of a 14-d adaptation period and an 84-d measurement period. Duplicate samples of each feedstuff were collected weekly and composited monthly. Sub-samples from each month were dried in a forced-air oven at 50oC for 72 h and analyzed for DM, ash, NDF, ADF, lignin and CP. Daily samples of the hay refusals were collected and composited on a weekly basis. Refusal sub-samples from each month were

PAGE 101

89 dried in a forced-air oven at 50oC for 72 h and analyzed for DM. Shrunk body weight was measured for two consecutive days before (days -1 and 0) and af ter the trial (days 85 and 86). Full BW was measured weekly a nd hip height (HH) wa s measured on d 0, 42 and 84. Body condition score ( BCS; scale 1-9) was visually evaluated and blood samples were collected by jugular venipuncture on d 0, 28, 56, and 84 (sodium heparin, Fisher Scientific, Pittsburgh, PA). Blood was centrifuged at 2000 x g for 20 min at 4oC and plasma was frozen. A Technicon Auto analyzer (Technicon Instruments Corp., Chauncey, NY) was used to determine concen trations of blood urea nitrogen (BUN) and glucose with modifications of the Coulombe and Favreau (1963) and Gochman and Schmitz (1972) methods respectively. U ltrasound measurements of back, rump and intramuscular fat depth and rib eye area we re taken using real-time ultrasound and an Eloka 500V system equipped with a 3.5-MHz, 17-cm transducer and superflab (Aloka USA, Inc., Wallingford, CT) to ensure proper fi t of the transducer to the curvature of the animal. In vivo apparent digestibility was estimated using chromic oxide as a marker. A gelatin (#10 Torpac Inc., Fairfield NJ) capsu le containing 10 g of chromic oxide powder was dosed twice daily via a balling gun at 0700 and 1900 h into each steer for 10 consecutive days (days 67-77). Fecal sa mples were collected at 0700 and 1900 h during the last five days of dosing for digestibility determination. Feces were dried to a constant weight at 50oC in a forced-air oven, ground to pass through a 1-mm Wiley mill (Arther H. Thomas Company, Philadelphia, PA, USA) screen and a composite sample was made from all ten fecal samples collected for each steer. Chromium concentration in feces was determined using a Perkin Elmer 5000 (Welle sley, MA) atomic absorption spectrometer according to the procedure descri bed by Williams et al. (1962).

PAGE 102

90 In Situ Rumen Degradability Subsamples taken from the monthly hay samples for each treatment were mixed and composited to give a single sample fo r in situ degradabil ity analysis. The composited samples were ground to pass a 4-mm screen using a Wiley mill screen and weighed (5 g) into 10 x 23 cm nylon bags ( pore size of 50 µm; Bar Diamond Inc., Parma, ID). Duplicate samples of forage from each treatment we re incubated in each of two ruminally-fistulated, non-lactating Holstein dairy cows for 0, 3, 6, 9, 12, 24, 48, 72, 96, and 120 h. After incubation, bags were removed from the cow, rinsed with cool water and frozen. At the end of the measurement pe riod all bags were thawed and washed in a commercial washing machine using a cool-wash cycle without soap. Bags were dried for 24 h at 60oC and residue weights determined. The cows were fed 9 kg of bahiagrass hay supplemented with 0.4 kg of soybean meal. The McDonald model (1981) was fitted to the in situ degradation data using the nonlin ear regression procedures of SAS, version 8 (2002, SAS Inst., Inc., Cary, NC). The Mc Donald (1981) model is of the form: P= A + B(1-e-c(t-L)) where P= DM degraded at time t, A = wash lo ss, B = potentially degradable fraction, A+B = total degradability, c = the rate at whic h B is degraded, t = time incubated in the rumen, and L = lag time. Statistical Analysis A randomized complete block design was used to examine the effects of the five forage treatments on the performance of the steers. The data were analyzed using the MIXED procedure of SAS v8 (2002, SAS Inst., In c., Cary, NC). To evaluate the effects of the treatments on steer BW and dry matter intake (DMI), pre-trial BW and DMI values

PAGE 103

91 were used as respective covariates. The model used to analyze individual treatment effects was: Yijk: µ + Ti +Sk (i) Wj + TWij + Eijkl where: µ= general mean Ti = effect of treatment Sk (i) = effect of kth steer in treatment i Wj = effect of week of feeding TWij = effect of treatment across weeks of feeding Eijk= experimental error. A repeated measures statement was used fo r the analysis of feed intake, BW, blood glucose and BUN concentrations and body c ondition score. The most appropriate covariate structure was examined. Where appropriate the SAS slice command was used to examine treatment differences at each week for each dependent variable. To further separate differences among treatments at each week, the PDIFF statement was used. Contrast statements were used to compare the forage treatments against the control forage. The in situ degradability data were an alyzed using the GLM procedure of SAS v8 (2002). Contrast statements were used to compare each treatment to the control. The model used to analyze indivi dual treatment effects was: Yijk: µ + Ti + Cj + Eijk where: µ= general mean

PAGE 104

92 Ti = effect of treatment Cj = effect of cow Eijk= experimental error. Significance was declared at P < 0.05 and tendencies at P < 0.10. Results Forage Composition Ammonia-treated hay had a numerically ( P > 0.005) greater CP concentration and lower NDF, hemicellulose and lignin concentr ations than the other hays, which had similar concentrations of thes e analytes (Table 6-2). Animal Performance Steers fed NH3-treated and Ec-treated hay had greater intakes of total dietary and hay DM and total dietary NDF th an steers fed the control hay ( P < 0.05; Table 6-3). However, there was no increase in the DMI as a percent of BW (P > 0.05). Steers fed Ec also had greater hay NDF intake than steers fed the control hay ( P < 0.05). Intakes of total or hay DM or NDF were unaffected by Eb or Ef treatment. There were no treatment effects on total DM intake as a percent of BW. Figure 6-1 shows that the DMI of the steers fed Ecand NH3-treated hay was generally greater than that of steers fed the control hay, though such diffe rences were detected ( P < 0.05) at weeks 1, 5, 10 and 11. The decrease in DMI across treatments in weeks 11 and 12 may have been due to dosing with chromic oxide during this period. All treatments increased (P < 0.011) the DM digestibility (DMD) of the hays (Table 6-4). Digestibility of NDF ( NDFD) and CP (CPD) were increased ( P < 0.01) by NH3, Ec, and Eb treatments.

PAGE 105

93 Steers had similar BW in the first ten weeks of the trial, but those fed NH3and Ectreated hays had greater (P < 0.05) BW than the others in th e last two weeks (table 6-2). The initial and final BW, ADG, BCS and HH of the steers were not affected by treatment ( P > 0.05, Table 6-5), though steers fed Ecand NH3treated hays had numerically greater (6 %) ADG than those fed the cont rol diet. Rib eye area, back, rump and intramuscular fat depth and plasma BUN and glucose concentrations were also similar among treatments (Table 6-6). In Situ Rumen Degradability The wash fraction of the hays was decreased ( P < 0.1) by all enzyme treatments (Table 6-7). Ammonia treatment increased ( P < 0.001) the potentially degradable fraction, the total degradable fraction and the effective degradability of the forage, but enzyme treatment had no effect on these meas ures. The fractional degradation rate and lag phase were unaffected by treatment. Discussion The forage used in this study was similar in maturity and composition (Table 6-2) to the untreated bermudagrass forage used by Galloway et al. ( 1993) and Mandebvu et al. (1999b). Ammonia treatment increased the CP concentration of the forage and reduced NDF, hemicellulose and lignin concentrations as in other studies (Kunkle et al., 1983; Brown, 1988; Brown and Kunkle, 1992; Brown a nd Pate, 1997). Enzyme treatment had no similar effects on the hays. Ammonia and Ec treatment increased the DMI of the hays, partly because these treatments increased DMD and NDFD. Fe ng et al. (1996) also found that enzyme treatment increased the DMI and DMD of smooth bromegrass ( Broma inermis) by beef steers. The greater DMI of the ammoniated fo rage also may have been due to its greater

PAGE 106

94 CP concentration. Unlike other treatments , Ec treatment also increased NDF intake suggesting that, like ammonia, Ec treatment caused some preingestive hydrolysis of the cell walls. Preingestive cell wall hydrolysis provides additional readily fermentable carbohydrates that may promote substrate co lonization by ruminal bacteria (Wang and McAllister 2002), increase pa latability (Adesogan, 2005), and reduce gut fill. These factors probably contributed to the increase in DMD and NDFD due to Ec treatment. This study therefore suggests that Ec trea tment is as effective as ammoniation at improving the quality of bermudagrass. Enzyme treatment is a more practically attractive treatment for producers than ammoniation, since enzymes are benign and they donÂ’t present the handling and deliver y challenges of ammoniation. The superior effect of the Ec treatmen t versus Eb treatment on DMI and DMD is likely due to greater moisture availability for enzyme dispersal and activity in the forage treated with Ec versus that treated with E b. The greater difference in BW of steers fed the Ec or NH3treated hay versus control hay in th e last two weeks of the trial suggests that a longer measurement period ma y have revealed that Ec and NH3 treatment increased animal performance. Unlike enzyme treatment, ammoniation increased the CP intake and CP digestibility of the hay, sugges ting a potential increase in the available nitrogen from this treatment. However, the BUN and ADG data indicate that the additional nitrogen did not promote growth or cause excess blood N c oncentrations. Therefore, most of the additional N provided by the NH3-treated forage was probably excreted in the urine. Despite improving DMI and DMD, Ec and NH3 treatments did not increase animal performance. Figures 6-3 and 6-4 show that al l the hays provided adequate intake of net

PAGE 107

95 energy (NE) and metabolisable protein (MP) for maintenance, but none of them provided enough NE or MP for growth. The addition of the concentrate supplement allowed all treatments to meet or exceed the steers tota l NE and MP requirements, suggesting that differences between treatments may have been more apparent if the hays had been fed alone. Though hay is typically fed to beef cattle without supplementation in south Florida, supplement was added to mimic the normal practice of supplementation in central and north Florida. When the in situ digestion kinetics were evaluated, ammoniation was the only treatment that increased the potentially degrad able and total degradable fractions as well as the effective degradability of the forage yet it did not improve th e performance of the steers. This can be attributed to the fact that the forage was a 5-wk regrowth containing greater than 5% CP and 45% TDN making it a poor candidate for ammoniation (Rasby et al., 1989). The potential improvement in ADG from such medium quality forage is marginal (Brown and Kunkle, 1992). Enzy mes typically increase only the rate of degradation and not the extent of forages and feeds (Feng et al., 1996). However, in this study the degradation parameters were not affected by enzyme treatment. This contradicts the improvement in vivo digestibility of the forages, and may reflect losses of fine, undigested particles through the pores of nyl on bags used in the in situ degradability trial. Implications This study shows that applyi ng Biocellulase A20 immediat ely after harvesting is as effective as ammoniation at increasing DMI and DMD of 5-wk regrowth bermudagrass in beef steers. These improvements in inta ke and digestion did not affect animal performance largely because of the presence of the concentrate suppl ement in the ration.

PAGE 108

96 Future work should examine the effect of this enzyme on similar maturity or more mature bermudagrass without concentrate supplementation.

PAGE 109

97 Table 6-1 Nutrient composition of the diet fed to beef steers. Componentb Treatmenta CP, % NDF, % TDN, % BCP, g/d NE, Mcal/d MP, g/d Control 14.364.2 61.0 537 9.8 594 NH3 15.862.7 64.4 660 11.7 696 Ec 13.665.7 63.1 661 11.5 679 Eb 14.365.4 63.1 567 10.3 612 Ef 14.066.1 60.9 579 10.9 623 a NH3 = Ammonia treatment, Ec = Enzyme applied at cutting, Eb = enzyme applied at baling, Ef = enzyme applied at feeding b BCP = bacterial crude protein, NE = net energy, MP = metabolisable protein

PAGE 110

98 Table 6-2. Chemical composition of th e treated and untreated hays (n=6). Variablea, g/kg DM Treatmentb NDFADFHemiLignin CP Control 741 349 392 54 109 NH3 699 348 350 47 139 Ec 743 347 396 54 106 Eb 757 353 404 62 111 Ef 761 345 415 50 109 a NDF = neutral detergent fiber, ADF = acid detergent fiber, Hemi = hemicellulose, CP = crude protein b NH3 = Ammonia treatment, Ec = Enzyme applied at cutting, Eb = enzyme applied at baling, Ef = enzyme applied at feeding

PAGE 111

99 Table 6-3. Effect of treating bermudag rass hay with ammonia or enzymes on the voluntary intake of steers. Intake kg d-1 Total DMI Total Hay Total Hay Total Hay Treatmenta % BW DM DM NDF NDF CP CP Control 2.33 7.28 4.49 4.88 3.42 1.06 0.50 NH3 2.50 8.64 5.91 5.79 3.72 1.30 0.75 Ec 2.31 8.38 5.51 5.69 4.08 1.14 0.61 Eb 2.55 6.90 4.06 4.61 3.47 1.07 0.46 Ef 2.53 7.82 5.09 5.35 3.79 1.10 0.57 SE 0.12 0.31 0.31 0.21 0.21 0.03 0.03 P values Treatment effect 0.428 0.005 <0.00010.001 0.221 <0.0001 <0.0001 Contrasts Control vs NH3 0.309 0.002 0.001 0.003 0.324 <0.0001 <0.0001 Control vs Ec 0.931 0.008 0.008 0.004 0.037 0.009 0.019 Control vs Eb 0.180 0.359 0.269 0.348 0.862 0.312 0.338 Control vs Ef 0.217 0.169 0.112 0.090 0.239 0.215 0.134 a NH3 = Ammonia treatment, Ec: Enzyme applied at cutting, Eb: Enzyme applied at baling, Ef: Enzyme applied at feeding

PAGE 112

100 Table 6-4. Effect of additive treatment on the in vivo apparent digestibility of diets in steers. Variablea g/kg Treatmentb DMD NDFD CPD Control 504 510 445 NH3 644 608 705 Ec 631 624 592 Eb 595 597 543 Ef 539 534 471 SE 19.2 18.1 17.7 P values Treatment effect 0.001 0.001 0.001 Contrasts Control vs NH3 0.001 0.001 0.001 Control vs Ec 0.001 0.001 0.001 Control vs Eb 0.001 0.001 0.001 Control vs Ef 0.011 0.087 0.117 a DMD: Dry matter digestibil ity, NDFD: Neutral detergen t fiber digestibility, CPD: Crude protein digestibility b NH3 = Ammonia treatment, Ec = Enzyme at cutting, Eb = enzyme at baling, Ef = enzyme at feeding

PAGE 113

101 Table 6-5. Effect of additive treatment on the performance of steers. Variable Treatmenta Full BW, kg Shrunk initial BW, kg Shrunk final BW, kg ADG, kg/d BCSb HHc, cm Control 320 257 361 1.21 5.28 122.6 NH3 323 250 365 1.29 5.25 122.6 Ec 324 253 361 1.28 5.23 122.3 Eb 320 258 361 1.21 5.19 122.5 Ef 322 254 359 1.23 5.29 121.6 SE 2.18 9.13 10.65 0.05 0.58 0.53 P values Treatment effect 0.614 0.831 0.968 0.559 0.713 0.918 Contrasts Control vs NH3 0.294 0.595 0.682 0.275 0.733 0.961 Control vs Ec 0.238 0.409 0.948 0.245 0.491 0.783 Control vs Eb 0.945 0.810 0.922 0.924 0.255 0.942 Control vs Ef 0.563 0.552 0.759 0.843 0.925 0.419 a NH3 = Ammonia treatment, Ec: Enzyme applied at cutting, Eb: Enzyme applied at baling, Ef: Enzyme applied at feeding b BCS: Body condition score scale 1-9 c HH: Hip height.

PAGE 114

102 Table 6-6. Effect of additive treatment on blood metabolites, fat deposition and rib eye area in beef steers. Variablea Treatmentb BUN mg/dl GLU mg/dl REA cm2 FT cm RF cm IMF % Control 12.4 83.8 45.2 1.3 0.98 3.76 NH3 11.6 83.6 46.0 1.3 0.91 3.72 Ec 12.5 83.0 42.9 1.4 0.94 3.90 Eb 12.4 84.0 42.7 1.3 0.88 4.06 Ef 11.5 83.7 43.8 1.3 0.84 3.70 SE 0.6 1.5 1.6 0.06 0.05 0.31 P values Treatment effect 0.521 0.989 0.216 0.579 0.341 0.910 Contrasts Control vs NH3 0.314 0.938 0.644 0.484 0.336 0.922 Control vs Ec 0.816 0.696 0.167 0.442 0.607 0.759 Control vs Eb 0.994 0.901 0.140 0.647 0.148 0.490 Control vs Ef 0.269 0.976 0.399 0.589 0.060 0.882 a BUN= Blood urea nitrogen; GLU= Glucose; REA= rib eye area; FT= 12th-13th rib juncture fat; RF= rump fat; IMF= intramuscular fat b NH3 = Ammonia treatment Ec: Enzyme app lied at cutting, Eb: Enzyme applied at baling, Ef: Enzyme applied at feeding

PAGE 115

103 Table 6-7. Effect of additive treatme nt on the in situ digestion kinetics Variablea, g/kg DM or as stated Treatmentb A B A+B c, %/h L, h ED Control 154 469 623 4.75 5.1 626 NH3 160 621 781 5.14 4.2 777 Ec 128 504 633 4.27 5.4 629 Eb 133 518 652 3.84 3.6 642 Ef 141 485 626 4.09 4.4 622 SE 4.6 20.7 20.2 0.4 0.6 16.9 P values Treatment effect 0.002 0.002 0.001 0.278 0.408 0.001 Contrasts Control vs NH3 0.409 0.001 0.001 0.538 0.416 0.001 Control vs Ec 0.002 0.269 0.738 0.457 0.701 0.882 Control vs Eb 0.014 0.155 0.364 0.163 0.164 0.507 Control vs Ef 0.071 0.613 0.925 0.309 0.488 0.883 a A= wash loss, B= potentially degradable fraction, A+B= total degradable fraction, c= fractional degradation rate, L= lag phase, ED= effective degradability b Ec: Enzyme applied at cutting, Eb: Enzyme applied at baling, Ef: Enzyme applied at feeding

PAGE 116

104 0 2 4 6 8 10 12 123456789101112 Control NH3 Ec Eb EfWeekDMI, kg * * * * 0 2 4 6 8 10 12 123456789101112 Control NH3 Ec Eb EfWeekDMI, kg 0 2 4 6 8 10 12 123456789101112 Control NH3 Ec Eb EfWeekDMI, kg 0 2 4 6 8 10 12 123456789101112 Control NH3 Ec Eb EfWeekDMI, kg * * * * Figure 6-1. Effect of ammonia or enzymatic treatment of bahiagrass on dry matter intake (DMI) of growing steers (t reatment*week interaction P < 0.0001). Asterisks denote significant differences among treatments (P < 0.05). a NH3 = Ammonia treatment, Ec = Enzyme applie d at harvest, Eb = Enzyme applied at baling, Ef = enzyme applied at feeding

PAGE 117

105 260 280 300 320 340 360 380 123456789101112 Control NH3 Ec Eb EfWeekBW, kg * * 260 280 300 320 340 360 380 123456789101112 Control NH3 Ec Eb EfWeekBW, kg 260 280 300 320 340 360 380 123456789101112 Control NH3 Ec Eb EfWeekBW, kg * * Figure 6-2. Effect of treatment on body weight (BW) changes during the trial by week (treatment*week interaction P = 0.9816), where asteri sks denote significant differences among treatments (P < 0.05). a NH3 = Ammonia treatment, Ec = Enzyme applie d at harvest, Eb = Enzyme applied at baling, Ef = enzyme applied at feeding

PAGE 118

106 0 1 2 3 4 5 6 7 8 9 10 11 12 13 ControlNH3EcEbEf NE hay NE concentrateMcal/d NEg NEm 0 1 2 3 4 5 6 7 8 9 10 11 12 13 ControlNH3EcEbEf NE hay NE concentrateMcal/d NEg NEm Figure 6-3. Net energy (NE) supplied by the diet components relative to NE requirements (NRC, 2000) for maintenance (NEm) and gain (NEg) of steers.

PAGE 119

107 0 100 200 300 400 500 600 700 800 900 1000 ControlNH3EcEbEf MP hay MP Concentateg/d MPg MPm 0 100 200 300 400 500 600 700 800 900 1000 ControlNH3EcEbEf MP hay MP Concentateg/d MPg MPm Figure 6-4. Metabolisable prot ein (MP) supplied by the diet components relative to MP requirements for maintenance (MPm) and gain (MPg) of steers.

PAGE 120

108 CHAPTER 7 GENERAL SUMMARY, CONCLUSI ONS AND RECOMMENDATIONS Temperate forages have higher nutritive values than tropi cal forages, mainly due to the differences in cell wall anatomy and dige stibility. These differences also reduce intake and performance of cattle fed tropical forages instead of temperate forages. Chemical treatments have been used successfully to improve forage nutritive value, but their adoption has been limited due to the caus tic nature of such chemicals, which poses human health hazards and corrodes equipm ent. Enzyme treatments have produced inconsistent improvements in the quality of temperate forages, but their effects on tropical forages have not been el ucidated. Four experiments we re initiated to evaluate the effect of fibrolytic enzyme application on the nutritive value of tropical forages and animal performance. The objective of experiments 1 and 2 was to evaluate the effect of different application rates of exogenous fibrolytic enzymes containing ferulic acid esterase on the digestibility of three tropical grass hays. The hays used were 12-wk regrowths of Pensacola bahiagrass (BAH) ( Paspalum notatum ) and Coastal (C-B; Cynodon dactylon ) and Tifton 85 (T-85) bermudagrass ( Cynodon sp. ). Experiment 1 involved the use of Depol 740L (BioCatalyst, Pontypridd, Wales, UK) which contains 32 U/ml of ferulic acid esterase activity, 20.78 FPU of cellulase activity and 8701.85 µmol/ml/min of xylanase activity. Experiment 2 involved the use of Depol 670L (D670L), from the same company, which contained 8 U/ml of feru lic acid esterase activity, 773 U/g of endogalactouranse activity, 38.13 FPU cellu lase activity and 3682.73 µmol/ml/min

PAGE 121

109 xylanase activity. The enzymes were applie d to the forages at rates of 0, 0.5, 1, 2, 3 g/100g DM in triplicate. Treated hays were analyzed for chemical composition, in vitro DM and NDF disappearance, in vitro digestibi lity as well as in situ DM degradability. In Experiment 1, increasing the applica tion rate of D740L decreased the NDF, ADF and hemicellulose concentrations of the three hays. Enzyme application also enhanced the release of wate r-soluble carbohydrates (WSC), fe rulic acid and p-coumaric acid from certain hays. This indicates in creased cleavage of ce ll wall polysaccharides and release of phenolic acids, respectively. A pplication of the enzyme also enhanced the 6-h and 96-h in vitro DM disappearance of the bermudagrasses and all forages, respectively. Therefore, D740L treatment improved the nutri tive value of the three hays, though improvements in the extent of digestion were marginal. These increases in nutritive value were more pronounced in bermudagrasses than in bahigrass. In Experiment 2, increasing the applica tion rate of D670L decreased the NDF, ADF, and hemicellulose concentrations of BA H, but produced inconsistent effects on those of C-B and T-85. Enzyme applicati on increased 6-h DM and NDF disappearance of the three forages, but only increased the IVDMD of BAH and T-85. Increasing the enzyme application rate increased the rate and extent of degradation of T-85, and decreased the lag phase of the hays. This st udy demonstrates that application of D670L enhanced the initial digestion of all hays and slightly increase d the extent of digestion of BAH and C-B hays. Experiments 1 and 2 demonstrate that the application of fibrolytic enzymes can improve the DM digestion of tropical grass hays . Such improvements were greater at the earlier stages of digestion and they were more consis tent and pronounced for D740L

PAGE 122

110 versus D670L. These studies demonstrate that enzymes can be used to improve the nutritive value of mature tropical forage s but the response varies with enzyme composition, activity and forage type. Si nce the magnitude of the enzyme-induced increases in nutritive value were small, furt her research was warranted on combinations of FAE and other fibrolytic enzymes that would optimi ze the digestion of tropical grasses. The objectives of Experiment 3 were to determine the ideal combination of FAE, cellulase and xylanase for optim izing the hydrolysis of mature bahiagrass in the absence of rumen fluid, and enhancing the digestion of the forage in rumen fluid. The first aspect of this experiment evaluated the effect s of the different combinations of FAE , xylanase (XYL) and cellulase (CEL) on DM disappearance of a 12-wk regrowth of bahiagrass hay that was incubated in a buffered enzyme so lution for 24 h. The second aspect of the experiment evaluated the effects of treatment with Depol 740L or the two most promising multienzyme cocktails from the enzymatic disappearance study on the in vitro digestibility and fermentation of the bahiagrass hay. The most promising multienzyme cocktails from the enzymatic DM disappearance study were those that contained 2% cellulo se, 2% xylanase and no FAE (2-2-0) and 2% cellulose, 2% xylanase and 1% FAE (2-2-1). Application of these multienzyme cocktails and Depol 740L decreased or did not affect the rate and extent of in situ degradation, and decreased the extent of in vitro gas production and DM and NDF digestibility. However, enzyme application reduced the lag phase be fore fermentation commenced, increased the immediately fermentable fraction of the forages and increase d propionate production while decreasing acetate production after 24 h of incubation in rumen fluid. The resultant

PAGE 123

111 decrease in the acetate to propi onate ratio indicates that en zyme treatments improved the efficiency of fermentation of the forage. The increase in the immediately fermentable fraction and decrease in the lag phase indicate that the enzyme enhanced preruminal hydrolysis of the forage. That enzyme treatm ent did not affect the fermentation rate and or the extent of gas production and NDF or DM digestion, was attri buted to inadequate agitation of samples during fermentation and inadequate enzyme-substrate interaction prior to incubation. Therefore, future resear ch should address these factors to validate the promise of the multienzyme cocktails for improving the nutritive value of bahiagrass. Future work should also investigate how the response to multienzyme cocktail treatment varies with tropical forage spp and maturity. Animal experiments should also be used to validate the in vitro experiments. The objective of Experiment 4 was to dete rmine the effects of method of applying a fibrolytic enzyme mixture (Biocellulase A 20) or ammonia to bermudagrass hay on feed intake, digestion kinetics and growth performance of beef st eers. Five-wk fall regrowth of bermudagrass ( Cynodon sp.) was harvested and stored as hay in 18-20 kg square bales. The forage was treated with either anhydr ous ammonia (3 g/100g DM) or a fibrolytic enzyme mixture, Biocellulase A20™ (A 20) Loders Croklaan (Channahon, IL, USA). Enzyme A20 contained 1400 U/g of enzyme activity and it was applied at the manufacturers recommended rate of 15 g/ton to bermudagrass immediately after it was cut (Ec), immediately before it was baled (E b) or immediately before feeding (Ef). Anhydrous ammonia was applied at 3% DM to a sealed stack of baled hay. Fifty AngusBrangus steers (250-270 kg) were fed the hays supplemented (1% BW) with cottonseed meal and soy hulls for 84 d.

PAGE 124

112 Ammonia treatment increased CP concen tration and reduced NDF, hemicellulose and lignin concentrations of the forage, but enzyme treatments had no similar effects. Ammonia and Ec treatment increased DM in take by 19% and 15% and DM digestibility by 28% and 25%, respectively. In addition, thes e treatments also increased NDF intake by 19% and 17% and NDF digestibility by 19 % and 22%, respectively. However, none of the treatments altered the ADG of the st eers because supplementation with the concentrate masked the benefi cial effects of Ec and NH3 on animal performance. The results from this study suggest that A20 was as effective as amm onia at increasing the quality of bermudagrass hay. Future experi ments should investigate the effects of Ec treatment of bermudagrass hay on the ADG of beef steers fed unsupplemented diets. Since a positive BW response to Ec and amm onia treatment was only evident in wks 1112 of the trial, future experiments should also examine longer measurement periods (120 d). This set of experiments demonstrated that fibrolytic enzyme preparations containing esterase, cellulose and xylanase improved the nutritive value of mature forages tropical forages. However, th e improvement was small, probably costprohibitive and dependent on the composition and application rate of the enzyme preparation, and forage type. A commercial fibrolytic enzyme produced as much improvement in the quality of 5-wk regrowth of bermudagrass hay as ammoniation, but neither of the treatments increased ADG. Therefore, these studies indicate that more work needs to be done before beef producers can be advised to use enzymes to enhance the quality of tropical forages. Since the re sponse to enzyme treatment in this study and the literature varies with forage type a nd maturity, enzyme com position and activity and

PAGE 125

113 method of enzyme delivery, further research on the interaction of these factors and enzyme efficacy is required before defin itive recommendations can be developed for improving tropical forages with enzymes. Further research requirements include standardized methods of re porting enzyme activities and determination of enzyme activities at a pH and temperat ure that are similar to those that prevail in forages or animals. Since most of this work focused on mature, C4 grasses, future studies should also examine the potential for improving the quality of less mature tropical forages with fibrolytic enzymes.

PAGE 126

114 APPENDIX A DESCRIPTION OF PHENOLIC ACID ANALYSIS Ester-linked phenolic acids were measur ed by adding 2 M NaOH (2.4 ml) to 75 mg of bahiagrass for 20 h at 25oC in 6 ml teflon vials (Savil lex Corp.) under nitrogen. TwoHydroxycinnamic acid (75 µl of a 1 mg/ml solution in 2 N NaOH) was added as an internal standard. The mixture was then aci dified (pH<2.0) with concentrated (12 M) HCL (0.42 ml). Liberated ester-linked phenolic acids were extracte d into diethyl ether (2.5 ml x 4 times). Ether extracts were combined and reduced to dryness with N2. Dried extracts were silylated with pyridine ( 10 µl) and BSFTA (40 µl) for 30 min at 60oC. Ether-linked phenolics were extracted using the acidified samples from the esterlinked extraction. Acidified samples in teflon vial are flushed with N2 to remove additional ether. Internal standard was once ag ain added to the vial at the same level as for the ester-linked analysis, as well as 12M NaOH (degassed with N2). Vials were flushed with N2 and then capped with the teflon vial cap follwed by amber overcaps (Savillex Corp.) to limit the swelling of th e caps during heating. The vials were then placed in large teflon bombs (47 mm X 12 cm, Savillex Corp.) with 3 ml of water in the bottom of the bomb. Bombs were closed us ing bomb wrenches (Savillex Corp) and placed in a 2L beaker for stability and then placed in 170oC mechanical oven for 2 h. Samples were then allows to cool to room temperature, followed by an additional 15 min cooling by putting cool water in the 2L beak er where bombs are sitting. Samples were quantitatively transferred to screw cap vials (Pyrex 9826) rinsing the teflon vial with MilliQ several times (2ml x 3 times). Samp les were acidified using 12 M HCL (1.5ml) to

PAGE 127

115 a pH< 2.0. Samples were then ether extracte d (4 ml x 4 times) and derivitized in the same manner as the ester-linked samples. Trimethylsilylated deri vatives of phenolic acids were separated by gas chromatogra phy (Varian CP3800, Varian Inc. Cary, NC) using a 25 mm x 0.33 mm ID x 4 µm film thic kness column (Varian, inc. P/N CP7739) and a flame ionization detector. Samples we re hand injected, by a single person, with a 10 µl syringe using a hot sandwich technique onto the GC that was fitted with a 1177 type injector set at 300oC, split on for 1 min, with a split ratio of 50:1 and a constant injector flow rate of 1ml/ min. Parameters were as follows: Injector= 300oC, Detector= 315oC, Oven= 220oC for 1 minute, then ramps 220-240oC at 4oC/min and holds at 240oC for 6 min.

PAGE 128

116 APPENDIX B SAS CODE USED FOR CHAPTERS 3 AND 4 The SAS code used to analyze the pre-in vitro chemical composition, IVDMD and IVNDFD. Data; INFILE 'U:\research\Trial 1 2002 03\I NVITRO SAS\TABLES GRAPHS\ALL Pre Invitro.csv' delimiter=',' firstobs=2; INPUT EXP TRT REP ENZ GRASS Pr eNDF PreADF PreLig PreHemi; FORAGE= .; If TRT=1 or TRT=2 or TRT=3 or TRT=4 or TRT=5 then FORAGE =1; If TRT=6 or TRT=7 or TRT=8 or TRT =9 or TRT=10 then FORAGE =2; If TRT=11 or TRT=12 or TRT=13 or TR T=14 or TRT=15 then FORAGE =3; ENZYME =.; If TRT=1 or TRT=6 or TRT=11 then ENZYME =1; If TRT=2 or TRT=7 or TRT=12 then ENZYME =2; If TRT=3 or TRT=8 or TRT=13 then ENZYME =3; If TRT=4 or TRT=9 or TRT=14 then ENZYME =4; If TRT=5 or TRT=10 or TRT=15 then ENZYME =5; If forage=2 or forage=3 then delete; PROC SORT; BY TRT; PROC PRINT; proc means; PROC GLM; CLASS rep forage enzyme; MODEL PreNDF PreADF PreLig PreHem i= EXP FORAGE ENZYME exp*forage exp*enzyme forage*enzyme exp*forage*enzyme/SS1 SS3; contrast 'bahia v berm' forage -2 1 1; contrast 'coast v tift' forage 0 -1 1; contrast 'lin' enzyme -2 -1 0 1 2; contrast 'quad' enzyme 2 -1 -2 -1 2; contrast 'cubic' enzyme -1 2 0 -2 1; contrast 'quartic' enzyme 1 -4 6 -4 1; contrast 'bah v ber x lin' forage*enzyme 4 2 0 -2 -4 -2 -1 0 1 2 -2 -1 0 1 2; contrast 'bah v ber x quad' forage*enzyme -4 2 4 2 -4 2 -1 -2 -1 2 2 -1 -2 -1 2;

PAGE 129

117 contrast 'bah v ber x cubic' forage*enzyme 2 -4 0 4 -2 -1 2 0 -2 1 -1 2 0 -2 1; contrast 'bah v ber x quart' forage*enzyme -2 8 -12 8 -2 1 -4 6 -4 1 1 -4 6 -4 1; contrast 'coast v tift x lin' forage*enzym e 0 0 0 0 0 2 1 0 -1 -2 -2 -1 0 1 2; contrast 'coast v tift x quad' forage*enzyme 0 0 0 0 0 -2 1 2 1 -2 2 -1 -2 -1 2; contrast 'coast v tift x cubic' forage*enz yme 0 0 0 0 0 1 -2 0 2 -1 -1 2 0 -2 1; contrast 'coast v tift x quart' forage*enzym e 0 0 0 0 0 -1 4 -6 4 -1 1 -4 6 -4 1; LSMEANS FORAGE ENZYME/STDERR pdiff; LSMEANS Forage*enzyme/STDERR pdiff; run;

PAGE 130

118 The SAS Code used to generate the in situ parameters using the Proc NLIN procedures of SAS. OPTIONS NOCENTER; data D740; infile 'C:\Documents and Settings\Nathan\Desktop\Research\In situ 2003\SAS\D670insitu.csv' delimiter=','; input hr deg id ; proc print; run; Proc NLIN data=d740 method=mar quardt MAXITER=400 noitprint; by id; PARAMETERS a=100 to 200 by 5 b=500 to 700 by 5 c= 0.008 to 0.05 by 0.005 L= 1 to 12; BOUNDS a <= 200; BOUNDS b <= 1000; BOUNDS c < 1; BOUNDS L <= 12; MODEL deg=a; if (hr>L) then mo del.deg=a+b*(1-exp(-c*(hr-L))); Output out=nlinout p=pred r=resi parms=a b c L sse=errors; run;

PAGE 131

119 The SAS code for analyzing the in situ parameters generated from the NLIN procedures of SAS. Data; INFILE 'C:\Documents and Settings\N athan\Desktop\Research\In situ 2003\SAS\Depol670 combined insituSAS.csv' delimiter=',' firstobs=2; INPUT TRT COW ENZ REP GRAS S a b ab c l AA BB PP ED; FORAGE= .; If TRT=1 or TRT=2 or TRT=3 or TRT=4 or TRT=5 then FORAGE =1; If TRT=6 or TRT=7 or TRT=8 or TRT =9 or TRT=10 then FORAGE =2; If TRT=11 or TRT=12 or TRT=13 or TR T=14 or TRT=15 then FORAGE =3; ENZYME =.; If TRT=1 or TRT=6 or TRT=11 then ENZYME =1; If TRT=2 or TRT=7 or TRT=12 then ENZYME =2; If TRT=3 or TRT=8 or TRT=13 then ENZYME =3; If TRT=4 or TRT=9 or TRT=14 then ENZYME =4; If TRT=5 or TRT=10 or TRT=15 then ENZYME =5; if forage=3 or forage=2 then delete; PROC SORT; BY TRT; PROC PRINT; proc means; run; PROC GLM; CLASS rep forage enzyme; MODEL a b ab c l AA BB PP ED = COW FO RAGE ENZYME forage*enzyme /SS1 SS3; contrast 'bahia v berm' forage -2 1 1; contrast 'coast v tift' forage 0 -1 1; contrast 'lin' enzyme -2 -1 0 1 2; contrast 'quad' enzyme 2 -1 -2 -1 2; contrast 'cubic' enzyme -1 2 0 -2 1; contrast 'quartic' enzyme 1 -4 6 -4 1; contrast 'bah v ber x lin' forage*enzyme 4 2 0 -2 -4 -2 -1 0 1 2 -2 -1 0 1 2; contrast 'bah v ber x quad' forage*enzyme -4 2 4 2 -4 2 -1 -2 -1 2 2 -1 -2 -1 2; contrast 'bah v ber x cubic' forage*enzyme 2 -4 0 4 -2 -1 2 0 -2 1 -1 2 0 -2 1; contrast 'bah v ber x quart' forage*enzyme -2 8 -12 8 -2 1 -4 6 -4 1 1 -4 6 -4 1; contrast 'coast v tift x lin' forage*enzym e 0 0 0 0 0 2 1 0 -1 -2 -2 -1 0 1 2; contrast 'coast v tift x quad' forage*enzyme 0 0 0 0 0 -2 1 2 1 -2 2 -1 -2 -1 2; contrast 'coast v tift x cubic' forage*enz yme 0 0 0 0 0 1 -2 0 2 -1 -1 2 0 -2 1; contrast 'coast v tift x quart' forage*enzym e 0 0 0 0 0 -1 4 -6 4 -1 1 -4 6 -4 1; LSMEANS FORAGE ENZYME Fora ge*Enzyme/STDERR pdiff; run;

PAGE 132

120 APPENDIX C SAS CODE USED FOR CHAPTER 5 The SAS code used to analyze enzymatic DM disappearance. data; input TRT run rep dmdiss; cards; proc print; proc glm; class TRT rep run; model dmdiss = TRT run TRT*run ; *lsmeans TRT/pdiff; means TRT/dunnett; run;

PAGE 133

121 SAS code used to estimate the gas fermentation parameters at 24 h of incubation. OPTIONS NOCENTER; data gasprod; infile 'C:\Documents and Settings\Natha n\Desktop\Research\Experiment 4 Ratio studies\GP Volumes Data & SAS\24h GP pre ssures per gram substrate.csv' delimiter=',' firstobs=2; input id hr pressure; proc print; run; Proc REG data=gasprod; by id; model pressure=hr; run; SAS code used to generate the gas ferm entation parameters at 96 h of incubation. OPTIONS NOCENTER; data gasprod; infile 'C:\Documents and Settings\Natha n\Desktop\Research\Experiment 4 Ratio studies\GP Volumes Data & SAS\GP R un 3 3-16-06\GP run 3 per g DM 3-14-06 96h.csv' delimiter=',' firstobs=2; input id hr pressure; proc print; Proc NLIN data=gasprod met hod=marquardt maxiter=500 noitprint; by id; PARAMETERS A=0 to 20 by 1 b=0 to 50 by 1.0 c= 0.05 to 2.50 by 0.05 L= -10 to 40; BOUNDS A <= 20; BOUNDS b <= 50; Bounds c < 3; Bounds L < 41; MODEL pressure=a; if (hr>L) then m odel.pressure=a+b*(1-exp(-c*(hr-L))); Output out=nlinout p=pred r=resi parms=a b c L sse=errors; run;

PAGE 134

122 SAS code used to analyze the gas pr oduction DMD, NDFD, VFA production and parameter analysis at 24 and 96 h of incubation. Data; INFILE 'C:\Documents and Settings\Natha n\Desktop\Research\Experiment 4 Ratio studies\GP DMD SAS\GP DM D SAS 24h modified for enz.csv' delimiter=',' firstobs=2; INPUT TRT run rep DMD; proc sort; by TRT; proc print; proc glm; class rep trt run; model DMD=TRT run TRT*run; contrast '1 v 2' TRT -1 1 0 0; contrast '1 v 3' TRT -1 0 1 0; contrast '1 v 4' TRT -1 0 0 1; lsmeans trt/ STDerr pdiff ; run

PAGE 135

123 APPENDIX D SAS CODE USED FOR CHAPTER 6 SAS code used to analyze ADG, shrunk BW, feed to gain ratio, in vivo analyses. Data; INFILE 'C:\Documents and Settings\Nat han\Desktop\Research\Beef Feeding Trial\SAS\ADGdata.csv' delimiter=',' firstobs=2; INPUT Steer TRT Pen ADG block; Proc sort; by TRT Steer; Proc print; Proc mixed; class Steer TRT Pen Block ; model ADG=TRT / ddfm=kr; random pen block; lsmeans TRT; contrast 'NH3 v Conrol' TRT 1 -1 0 0 0; contrast 'Ec v Conrol' TRT 1 0 -1 0 0; contrast 'Eb v Conrol' TRT 1 0 0 -1 0; contrast 'Ef v Conrol' TRT 1 0 0 0 -1; run;

PAGE 136

124 SAS code used to analyze repeated measures like BCS, plasm BUN and glucose, intakes, hip height, ultrasound me asurements and chemical composition of diets. Data; INFILE 'C:\Documents and Settings\Nat han\Desktop\Research\Beef Feeding Trial\SAS\BW.csv' delimiter=',' firstobs=2; INPUT Steer TRT Week Pen BW initbw block; Proc sort; by TRT Steer; Proc print; Proc mixed; class Steer TRT Week Pen Block ; model BW=TRT Week TRT* Week initbw / ddfm=kr; random pen block; repeated Week/ sub=steer type=arh(1); lsmeans TRT Week TRT*week / slice=Week; contrast 'control v others' TRT -4 1 1 1 1; contrast 'NH3 v others' TRT 1 -4 1 1 1; contrast 'Ec+Eb+Ef v Control' TRT 3 0 -1 -1 -1; contrast 'NH3 v Conrol' TRT 1 -1 0 0 0; contrast 'Ec v Conrol' TRT 1 0 -1 0 0; contrast 'Eb v Conrol' TRT 1 0 0 -1 0; contrast 'Ef v Conrol' TRT 1 0 0 0 -1; contrast 'NH3 v Ec+Eb+Ef' TRT 0 -3 1 1 1; contrast 'Ec+Eb v Ef' TRT 0 0 -1 -1 2; contrast 'Ec v Eb' TRT 0 0 -1 1 0; run;

PAGE 137

125 SAS code used to generate in situ parameters using the NLIN procedures . OPTIONS NOCENTER; data hayinsitu; infile 'C:\Documents and Settings\Nat han\Desktop\Research\Beef Feeding Trial\SAS\hayinsitu.csv' delimiter=','; input ID hr DEG; proc print; run; Proc NLIN data=hayinsitu met hod=marquardt MAXITER=400 noitprint; by id; PARAMETERS a=100 to 200 by 5 b=500 to 700 by 5 c= 0.008 to 0.05 by 0.005 L= 1 to 12; BOUNDS a <= 200; BOUNDS b <= 1000; BOUNDS c < 1; BOUNDS L <= 12; MODEL DEG=a; if (hr>L) then model.deg=a+b*(1-exp(-c*(hr-L))); Output out=nlinout p=pred r=resi parms=a b c L sse=errors; run; SAS code used to analyze in situ parameters. Data; INFILE 'C:\Documents and Settings\Nat han\Desktop\Research\Beef Feeding Trial\SAS\insitu parameters.csv' delimiter=',' firstobs=2; INPUT TRT COW REP a b ab c l AA BB AABB P96 ED; PROC PRINT; run; PROC GLM; CLASS rep trt; MODEL a b ab c l AA BB AABB P96 ED = COW TRT /SS1 SS3; contrast 'NH3 v Conrol' TRT 1 -1 0 0 0; contrast 'Ec v Conrol' TRT 1 0 -1 0 0; contrast 'Eb v Conrol' TRT 1 0 0 -1 0; contrast 'Ef v Conrol' TRT 1 0 0 0 -1; LSMEANS TRT/STDERR pdiff; run;

PAGE 138

126 LIST OF REFERENCES Adesogan, A. T. 2005. Improving forage quality a nd animal performance with fibrolytic enzymes. Proc. Florida Ruminant Nutriti on Symposium, University of Florida, Gainesville, FL. p 91-109. Adesogan, A. T., N. K. Krueger, and S. C. Kim. 2005. A novel, wireless, automated system for measuring fermentation gas production kinetics of feeds and its application to feed characterization. Anim. Feed Sci. Technol. 123-124: 211-223. Akin, D. E. 1986. Interaction of ruminal bact eria and fungi with southern forages. J Anim. Sci. 63: 962-977. Akin, D., L. Rigsby, A. Sethuraman, W. Mo rrison, 3rd, G. Gamble, and K. Eriksson. 1995. Alterations in struct ure, chemistry, and biodegradability of grass lignocellulose treated wi th the white rot fungi Ceriporiopsis subvermispora and C yathus stercoreus . Appl. Environ. Microbiol. 61: 1591-1598. Akin, D. E. 1993. Perspectiv es of cell wall biodegrad ation-session synopsis. p 73-82. In:Forage cell wall structur e and digestibility. H. G. Jung, D. R. Buxton, R. D. Hatfield and J. Ralph (Eds.). ASA, CSSA, SSSA, Madison, WI. Akin, D. E., and D. Burdick. 1975. Percentage of tissue types in tropical and temperate grass leaf blades and degradation of ti ssues by rumen microorganisms. Crop Sci. 15: 661-668. Aman, P. 1993. Composition and structure of ce ll wall polysaccharides in forages. p 183196. In:Forage cell wall structur e and digestibility. H. G. Jung, D. R. Buxton, R. D. Hatfield and J. Ralph (Eds.). ASA, CSSA, SSSA, Madison, WI. Anderson, W. F., J. Peterson, D. E. Akin, and W. H. Morrison III. 2005. Enzyme pretreatment of grass ligno cellulose for protein highvalue co-products and an improved fermentable substrate. A ppl. Biochem. Biotech. 121-124: 303-310. Arthington, J., and W. Brown. 2002. Effect of maturity on quality measures of four common Florida pastures. University of Florida, IFAS Extension, Beef Reports Series No. N-02297. Ball, D. M., C. S. Hoveland, and G. D. Lacefield. 2002. Warm season grasses. p 28-40. In:Southern forages. PPI and FAR, Norcross, Georgia.

PAGE 139

127 Bailey, M. J., P. Biely, and K. Poutanen. 1992. Interlaboratory tes ting of methods for assay of xylanase activit y. J. Biotech. 23: 257-270. Bartolome, B., C. Faulds, P. Kroon, K. Wa ldron, H. Gilbert, G. Hazlewood, and G. Williamson. 1997. An Aspergillus niger esterase (ferulic aci d esterase III) and a recombinant Pseudomonas fluorescens subsp. Cellulosa esterase (Xy1D) release a 5-5' ferulic dehydrodimer (dif erulic acid) from barley and wheat cell walls. Appl. Environ. Microbiol. 63: 208-212. Bartolome, B., C. Faulds, M. Tuohy, G. P. Hazelwood, H. J. Gilber t, and G. Williamson. 1995. Influence of different xylanases on the activity of ferulic acid esterase on wheat bran. Biotechnol. A ppl. Biochem. 22: 65-73. Beauchemin, K. A., D. Colombatto, and D. P. Morgavi. 2004. A rationale for the developement of feed enzyme products fo r ruminants. Can. J. Anim. Sci. 84: 23-26. Beauchemin, K. A., D. Colombatto, D. P. Morgavi, and W. Z. Yang. 2003. Use of exogenous fibrolytic enzymes to improve f eed utilization by ruminants. J. Anim. Sci. 81: E37-47. Beauchemin, K. A., S. D. M. Jones, L. M. Rode, and V. J. H. Sewalt. 1997. Effects of fibrolytic enzymes in corn or barl ey diets on performance and carcass characteristics of feedlot diets. Can. J. Anim. Sci. 77: 645-653. Beauchemin, K. A., L. M. Rode, and D. Karren. 1999. Use of feed enzymes in feedlot finishing diets. Can. J. Anim. Sci. 79: 243-246. Beauchemin, K. A., L. M. Rode, M. Maekawa, D. P. Morgavi, and R. Kampen. 2000. Evaluation of a nonstarch polysaccharidase f eed enzyme in dairy cow diets. J. Dairy Sci. 83: 543-553. Beauchemin, K. A., L. M. Rode, and V. J. H. Sewalt. 1995. Fibrolytic enzymes increase fiber digestibility and growth rate of st eers fed dry forages. Can. J. Anim. Sci.: 641644. Berger, L. L., G. C. Fahey, L. D. Bourqui n, and E. C. Titgemeyer. 1994. Modification of forage quality after harvest. p 922966. In:Forage quality, evaluation, and utilization. G. C. Fahey, M. Collins, D. R. Mertins and L. E. Moser (Eds.). ASA, CSSA, SSSA, Madison, WI. Bhat, M. K., and G. P. Hazelwood. 2001. Enzymology and other characteristics of cellulases and xylanases. p 11-60. In:Enzym es in farm animal nutrition. M. R. Bedford and G. C. Partridge (Eds .). CABI Publishing, New York, NY. Blum, D. L., I. A. Kataeva, X.-L. Li, and L. G. Ljungdahl. 2000. Feruloyl esterase activity of the clostridium thermocellum cellulosome can be attributed to previously unknown domains of xyny and xynz. J. Bacteriol. 182: 1346-1351.

PAGE 140

128 Borneman, W., L. Ljungdahl, R. Hartley, and D. Akin. 1992. Purification and partial characterization of two feruloyl es terases from the anaerobic fungus Neocallimastix strain mc-2. Appl. Envir on. Microbiol. 58: 3762-3766. Bowman, G. R., K. A. Beauchemin, and J. A. Shelford. 2002. The proportion of the diet to which fibrolytic enzymes are added a ffects nutrient digestion by lactating dairy cows. J. Dairy Sci. 85: 3420-3429. Brezillon, C., P. A. Kroon, C. B. Faulds, G. M. Brett, and G. Williamson. 1996. Novel ferulic acid esterases ar e induced by growth of Aspergillus niger on sugar-beet pulp. Appl. Microbiol. Biotech. 45: 371-376. Brown, W. F. 1988. Maturity and ammoniation effects on the feedi ng value of tropical grass hay. J. Anim. Sci. 66: 2224-2232. Brown, W. F. 1993. Results of 1992-93 hay amm oniation demonstrations in Florida. Available:http://rcrec-ona.ifas.ufl.ed u/or10-93.html. Accessed:November 20, 2005. Brown, W. F. 1994. Results of on-farm 199394 hay ammoniation demonstrations in Florida. Available:http:// rcrec-ona.ifas.ufl.edu/or894.html. Accessed:November 20, 2005. Brown, W. F., and M. B. Adjei. 1995. Urea a mmoniation effects on the feeding value of guineagrass ( Panicum maximum ) hay. J. Anim. Sci. 73: 3085-3093. Brown, W. F., and W. E. Kunkle. 1992. Impr oving the feeding value of hay by anhydrous ammonia treatment. University of Flor ida, IFAS Extension. EDIS Reports. Available http://edis.ifas.ufl.edu/ BODY_AA203 Accessed: October 30, 2002. Brown, W. F., and F. M. Pate. 1997. Cottonseed meal or feather meal supplementation of ammoniated tropical grass hay for year ling cattle. J. Anim. Sci. 75: 1666-1673. Byott, G. S. 1976. Leaf air space systems in C3 and C4 species. New Phytol. 76: 285-299. Casler, M. D., and H.-J. G. Jung. 1999. Select ion and evaluation of smooth bromegrass clones with divergent lignin or etherified ferulic acid concentration. Crop Sci. 39: 1866-1873. Chinh, B. V., L. E. Ly, N. H. Tao, P. V. Thin, and T. R. Preston. 1992. Ammoniated rice straw or untreated straw supplemented with a molasses urea block for growing sindhi x local cattle in Vietnam. Available:http://www.fao.org/ag/AGA/ AGAP/FRG/lrrd/lrrd4/3/chinh.htm. Accessed:May 20, 1998. Church, D. C. 1988. The ruminant animal . Prentice-Hall, Englewood Cliffs, NJ.

PAGE 141

129 Coleman, S. W., J. E. Moore, and J. R. Wilson. 2004. Quality and utilization. p 267-308. In:Warm season grasses. L. E. Moser, B. L. Burson and L. E. Sollenberger (Eds.). ASA, Madison, WI. Colombatto, D., F. L. Mould, M. K. Bhat , and E. Owen. 2003a. Use of fibrolytic enzymes to improve the nutritive value of ruminant diets: A biochemical and in vitro rumen degradation assessment. Anim. Feed Sci. Technol. 107: 201-209. Colombatto, D., D. P. Morgavi, A. F. Fu rtado, and K. A. Beauchemin. 2003b. Screening of exogenous enzymes for ruminant diets: Relationship between biochemical characteristics and in vitro ruminal degradation. J. Anim. Sci. 81: 2628-2638. Cone, J. W., and F. M. Engels. 1990. Influen ce of growth temperature on anatomy and in vitro digestibility of maize tissu es. J. Agric. Sci. 114: 207-212. Coughlan, M. P. 1985. The properties of fungal and bacterial cellulases with comment on their production and application. Bi otechnol. Genet. Eng. Rev. 3: 39-109. Coulombe, J.J., and L. Favreau. 1963. A new simple semi micro method forcolorimetric determination of urea. Clin. Chem. 9: 102-108. Dawson, K. A., and J. M. Tricarico. 1999. The use of exogenous fibrolytic enzymes to enhance microbial activities in the rumen and performance of ruminant animals. Proc. 15th annual symposium on biotechnology in the feed industry, Loughborough, Leics, UK. p 303-312. Evers, G. W., L. A. Redmon, and T. L. Provin. 2004. Comparison of bermudagrass, bahiagrass, and kikuyugrass as a sta nding hay crop. Crop Sci. 44: 1370-1378 . Fahey, G. C., L. D. Bourquin, E. C. Titgem eyer, and D. G. Atwell. 1993. Post harvest treatment of fibrous feedstuffs to im prove there nutritive value. p 715-766. In:Forage cell wall structur e and digestibility. H. G. Jung, D. R. Buxton, R. D. Hatfield and J. Ralph (Eds.). ASA, CSSA, SSSA, Madison, WI. FASS. 2000. Florida agricultur al statistics service. http://www.nass.usda.gov/Statis tics_by_State/F lorida/index.asp . Accessed 4/20/01. Faulds, C. B., R. P. deVries, P. A. Kroon, J. Visser, and G. Williamson. 1997. Influence of ferulic acid on the producti on of feruloyl esterases by A spergillus niger . FEMS Microb. Letters 157: 239-244. Faulds, C. B., P. A. Kroon, L. Saulnier, J.-F. Thibault, and G. Williamson. 1995. Release of ferulic acid from maize bran and derived oligosaccharides by Aspergillus niger esterases. Carb. Poly. 27: 187-190. Faulds, C. B., and G. Williamson. 1994. Purificat ion and characterization of a ferulic acid esterase (FAE-III) from Aspergillus niger : Specificity for the phenolic moiety and binding to microcrustalline cellu lose. Microbiol. 140: 779-787.

PAGE 142

130 Faulds, C. B., and G. Williamson. 1995. Releas e of ferulic acid from wheat bran by a ferulic acid esterase (FAE-III) from Aspergillus niger . Appl. Microbiol. Biotechnol. 43: 1082-1087. Feng, P., C. W. Hunt, G. T. Pritchard, a nd W. E. Julien. 1996. Effect of enzyme preparations on in situ and in vitro degr adation and in vivo di gestive characteristics of mature cool-season gra ss forage in beef steers. J. Anim. Sci. 74: 1349-1357. Ferreira, L. M., T. M. Wood, G. Williamson, C. Faulds, G. P. Hazlewood, and G. W. Black. 1993. A modular esterase from Pseudomonas fluorescens subsp. Cellulosa contains a non-catalytic cellulose-binding domain. Biochem. J. 294: 349-355. Galloway DL, S., A. L. Goetsch, L. A. Forster, Jr, A. R. Patil, W. Sun, and Z. B. Johnson. 1993. Feed intake and digestibility by cattle consuming bermudagrass or orchardgrass hay supplemented with soybean hulls and(or) corn. J. Anim. Sci. 71: 3087-3095. Grabber, J. H., R. D. Hatfield, J. Ral ph, J. Zon, and N. Amrhein. 1995. Ferulate crosslinking in cell walls isolated from ma ize cell suspensions. Phytochem. 40: 10771082. Grabber, J. H., R. D. Hatfield, J. Ra lph,. 1998. Diferulate cross-links impede the enzymatic degradation of non-lignified maize walls. J. Sci. Food Agric. 77: 193200. Grabber, J. H., J. Ralph, and R. D. Ha tfield. 2000. Cross-linking of maize walls by ferulate dimerization and incorporation into lignin. J. Agric. Food Chem. 48: 61066113. Grabber, J. H., J. Ralph, and R. D. Hatfie ld. 2002. Model studies of ferulate-coniferyl alcohol cross-product formation in pr imary maize walls: Implications for lignification in grasses. J. Agric. Food Chem. 50: 6008-6016. Grabber, J. H., J. Ralph, C. Lapierre, and Y. Barriere. 2004. Genetic and molecular basis of cell-wall degradability. I. Lignin-cell wall matrix inte ractions. C. R. Biologies 327: 455-465. Gochman, N., and J.M. Schmidz. 1972. Applica tion of a new peroxide indicator reaction to the specific, automated determination of glucose with glucose oxidase. Clin. Chem. 18: 943-952. Harbers, L. H. 1985. Ultrasonic utilization of plants by herbivores. Food Microstruct. 4: 357-364. Hartley, R. D. 1972. P-coumaric and ferulic acid components of cell walls of ryegrass and their relationships wit lignin and dige stibility. J. Sci. Food Agric. 23: 13471354.

PAGE 143

131 Hatfield, R. D. 1993. Cell wall polysaccharide interactions and de gradability. p 285-313. In:Forage cell wall structur e and digestibility. H. G. Jung, D. R. Buxton, R. D. Hatfield and J. Ralph (Eds.). ASA, CSSA, SSSA, Madison, WI. Hatfield, R. D., P. Mandebvu, and J. W. West. 1997. A comparison of Tifton 85 and coastal bermudagrass cell walls. U. S. Dair y Forage Research Center Summaries. Madison, WI: 49-50. Hristov, A. N., T. A. McAllister, and K. J. Cheng. 2000. Intraruminal supplementation with increasing levels of exogenous polys accharide-degrading enzymes: Effects on nutrient digestion in cattle fed a barley grain diet. J. Anim. Sci. 78: 477-487. Iiyama, K., T. B. T. Lam, and B. A. Stone. 1990. Phenolic acid bridges between polysaccharides and lignin in wheat internodes. Phytochem. 29: 733-737. Jaucquet, G., B. Pollet, and C. Lapierre. 1995. New ether-linked ferulic acid-coniferyl alcohol dimers identified in grass st raws. J. Agric. Food Chem. 1995: 2746-2751. Jung, H. G., and M. S. Allen. 1995. Characteristics of plant cell walls affecting intake and digestibility of forages by rumi nants. J. Anim. Sci. 73: 2774-2790. Jung, H. G., and D. A. Deetz. 1993. Cell ligninfication and degradability. p 315-346. In:Forage cell wall structur e and digestibility. H. G. Jung, D. R. Buxton, R. D. Hatfield and J. Ralph (Eds.). ASA, CSSA, SSSA, Madison, WI. Jung, H. G., F. R. Valdez, A. R. Abad, R. A. Blanchette, and R. D. Hatfield. 1992. Effect of white rot basidiomycetes on chemical co mposition and in vitro digestibility of oat straw and alfalfa stems. J. Anim. Sci. 70: 1928-1935. Jung, H. G., and K. P. Vogel. 1986. Influence of lignin on digestibility of forage cell wall material. J. Anim. Sci. 62: 1703-1712. Kato, Y., and D. J. Nevins. 1985. Isolati on and identification of O-(5-O-feruloylarabinofuranosyl)(1-->3)-O-xylopyranosyl-(1-->4) --xylopyranose as a component of zea shoot cell-wa lls. Carb. Res. 137: 139-150. Klopfenstein, T. 1978. Chemical treatment of crop residues. J. Anim. Sci. 46: 841-848. Krause, M., K. A. Beauchemin, L. M. Rode, B. I. Farr, and P. Norgaard. 1998. Fibrolytic enzyme treatment of barley grain and sour ce of forage in high-grain diets fed to growing cattle. J. Anim. Sci. 76: 2912-2920. Kroon, P. A., M. T. Garcia-Conesa, I. J. Filingham, G. P. Hazlewood, and G. Williamson. 1999. Release of ferulic acid de hydrodimers from plant cell walls by feruloyl esterases. J. Sc i. Food Agric. 79: 428-434. Kroon, P. A., and G. Williamson. 1999. Hydroxyci nnamates in plants and food: Current and future perspectives. J. Sci. Food Agric. 79: 355-361.

PAGE 144

132 Kung, L. 2001. Direct-fed microbials for dair y cows and enzymes for lactating dairy cows: New theories and applications . Proc. 12th Annual Florida Ruminant Nutrition Symposium, Ga inesville, FL. p 22-42. Kunkle, W. E., F. F. Henteges and J. G. Wasdin. 1984. Effects of anhydrous ammonia treatment of hay and supplemental protei n on the performance of lactating beef cows. Florida Beef Research Reports. Available http://www.animal.ufl.edu/extension/beef /pubs.shtml Accessed: October 30, 2002. Kunkle, W. E., R. Goff and J. Durrance. 1983. Anhydrous ammonia treatment of low quality hay. Florida Beef Research Reports. Available http://www.animal.ufl.edu/extension/beef /pubs.shtml Accessed: October 30, 2002. Lalman, D., G. Horn, R. Huhnke, and L. Redmon. 1997. Ammoniation of low quality roughages. Available http://www. Osuextra.Com. Accessed: May 20,1998 Lam, T., T. Bach, K. Iiyama, and B. A. Stone. 1992. Cinnamic acid bridges between cell wall polymers in wheat and phalaris internodes. Phytochem. 31: 1179-1183. Lewis, G. E., C. W. Hunt, W. K. Sanchez, R. Treacher, G. T. Pritchard, and P. Feng. 1996. Effect of direct-fed fibrolytic enzy mes on the digestive ch aracteristics of a forage-based diet fed to beef steers. J. Anim. Sci. 74: 3020-3028. Lewis, G. E., W. K. Sanchez, C. W. Hunt, M. A. Guy, G. T. Pritchard, B. I. Swanson, and R. J. Treacher. 1999. Effect of direct-f ed fibrolytic enzymes on the lactational performance of dairy cows. J. Dairy Sci. 82: 611-617. Li, X., H. Chen, and L. Ljungdahl. 1997. M onocentric and polycentric anaerobic fungi produce structurally related cellulases and xylanases. Appl. Environ. Microbiol. 63: 628-635. Mandebvu, P., J. W. West, M. A. Froetschel, R. D. Hatfield, R. N. Gates, and G. M. Hill. 1999a. Effect of enzyme or microbial tr eatment of bermudagrass forages before ensiling on cell wall composition, end products of silage fermentation and in situ digestion kinetics. Anim. Feed Sci. Technol. 77: 317-329. Mandebvu, P., J. W. West, G. M. Hill, R. N. Ga tes, R. D. Hatfield, B. G. Mullinix, A. H. Parks, and A. B. Caudle. 1999b. Comparison of Tifton 85 and coastal bermudagrasses for yield, nutrient traits, intake, and digestion by growing beef steers. J. Anim. Sci. 77: 1572-1586. McAllister, T. A., A. N. Hristov, K. A. Beauchemin, L. M. Rode, and K.-J. Cheng. 2001. Enzymes in ruminant nutrition. p 273-298. In:E nzymes in farm animal nutrition. M. R. Bedford and G. G. Partri dge (Eds.). CABI Publishing.

PAGE 145

133 McAllister, T. A., S. J. Oos ting, J. D. Popp, Z. Mir, L. J. Yanke, A. N. Hristov, R. J. Treacher, and K.-J. Cheng. 1999. Effect of exogenous enzymes on digestibility of barley silage and growth performance of feedlot cattle. Can. J. Anim. Sci.: 353360. McDonald, L. 1981. A revised model for the es timation of protien degradability in the rumen. J. Agric. Sci. 96: 251-252. Michal, J. J., K. A. Johnson, R. J. Treacher, C. T. Gaskins, and O. Sears. 1996. The impact of direct fed fibrolytic enzymes on the growth rate and feed efficiency of growing beef steers and heifers. J. Anim. Sci. 74 (Suppl. 1): 758. Miller, G. L. 1959. Use of dinitrosalicylic acid reagent for determination of reducing sugar. Anal. Chem. 31: 426-428. Ministry of Agriculture, Food and Fisheries. 1986. The analysis of agricultural materials. Reference book 427, HMSO, London. Minson, D. J. 1980. Nutritional differences be tween tropical and temperate pastures. p 143-157. In:Grazing animals. F. H. W. Morley (Ed.). Elsevier, Amsterdam. Moore, J. E. and G. O. Moore. 1974. Recove ry of residual organic matter from in vitro digestion of forages. J. Dairy Sci. 57:1258-1259. Morgavi, D. P., K. A. Beauchemin, V. L. Nsereko, L. M. Rode, A. D. Iwaasa, W. Z. Yang, T. A. McAllister, and Y. Wang. 2000. Synergy between ruminal fibrolytic enzymes and enzymes from Trichoderma longibrachiatum . J. Dairy Sci. 83: 13101321. Mueller-Harvey, I., R. D. Hartley, P. J. Harris, and E. H. Curzon. 1986. Linkage of pcoumaroyl and feruloyl groups to cell-wall polysaccharides of barley straw. Carb. Res. 148: 71-85. Murashima, K., A. Kosugi, and R. H. Doi. 2003. Synergistic effects of cellulosomal xylanase and cellulases from C lostridium cellulovorans on plant cell wall degradation. J. B acteriol. 185: 1518-1524. Musel, G., T. Schindler, R. Bergfeld, K. Ruel , G. Jaucquet, C. Lapierre, V. Speth, and P. Schopfer. 1997. Structure and disruption of lignin and primary and secondary cell walls of maize coleoptiles analyzed by chemical and immunogological probes. Planta 201: 146-159. N.R.C. 1996. p 185-205. In:Nutrien t requirements of beef cattl e. 7th. Rev. Ed. National Academy Press, Washington, D.C. Nevins, D. J. 1993. Analysis of forage cell wall polysaccharides. p 105-129. In:Forage cell wall structure and digestibility. H. G. J ung, D. R. Buxton, R. D. Hatfield and J. Ralph (Eds.). ASA, CSSA, SSSA, Madison, WI.

PAGE 146

134 Nsereko, V. L., D. P. Morgavi, L. M. Rode , K. A. Beauchemin, and T. A. McAllister. 2000. Effects of fungal enzyme preparat ions on hydrolysis and subsequent degradation of alfalfa hay fiber by mixe d rumen microorganisms in vitro. Anim. Feed Sci. Technol. 88: 153-170. Owen, E., T. Klopfenstein, and N. A. Urio. 1984. Treatment with other chemicals. p 248275. In: Straw and other fibrous by-products as feed. F. Sundstol and E. Owen (Eds.). Elsevier Science Publishe rs, Amsterdam, The Netherlands. Pendleton, B. 2000. The regulat ory environment. p 49-55. In:Direct-fed microbial, enzyme and forage additive compendium. S. Muirhead (Ed.). The Miller Publishing Company, Minnetonka, MN. Pinos-Rodriguez, J. M., S. S. Gonzalez, G. D. Mendoza, R. Barcena, M. A. Cobos, A. Hernandez, and M. E. Ortega. 2002. Eff ect of exogenous fibrolytic enzyme on ruminal fermentation and digestibility of al falfa and rye-grass hay fed to lambs. J. Anim. Sci. 80: 3016-3020. Ralet, M.-C., C. B. Faulds, G. Williams on, and J.-F. Thibault. 1994. Degradation of feruloylated oligosaccharides from suga r-beet pulp and wheat bran by ferulic acid esterases from A spergillus niger . Carb. Res. 263: 257-269. Ralph, J., J. H. Grabber, and R. D. Hatfiel d. 1995. Lignin-ferulate cr oss-links in grasses: Active incorporation of ferulate polysacchar ide esters into ryegrass lignins. Carb. Res. 275: 167-178. Ralph, J., S. Quideau, J. H. Grabber, a nd R. D. Hatfield. 1994. Identification and synthesis of new ferulic acid dehydrodimers present in grass cell walls. J. Chem. Soc., Perk. Trans. I.: 3485-3498. Rasby, R., I. Rush, J. Ward and T. Kl opfenstein. 1989. Ammonia treatment of low quality forages. Nebraska Coopera tive Extension EC89-265. Available http://ainrpubs.unl.edu/beef/ec265.htm Accessed: May 20, 1998. Siciliano-Jones. 1999. Strategies for coping with poor forage digestibi lity in dairy rations: Response to enzyme supplements. Pr oc. Proc. 15th Annual Symp. Biotechnology in the Feed Industry., Loughborough, Leics, UK. p 87-96. Somogyi, M. 1952. Notes on sugar determination. J. Biol. Chem. 195: 19-23. Sundstol, F. 1988. Straw and other fibrous byproducts. Livestock Prod. Sci. 19: 137-158. Sutton, J. D., R. H. Phipps, D. E. Beever, D. J. Humphries, G. F. Ha rtnell, J. L. Vicini, and D. L. Hard. 2003. Effect of method of application of a fibrolytic enzyme product on digestive processes and milk pr oduction in Holstein-Friesian cows. J. Dairy Sci. 86: 546-556.

PAGE 147

135 Terashima, N., F. Fukushima, L.-F. He, a nd K. Takabe. 1993. Comprehensive model of the lignified plant cell wall. p 247-27 0. In:Forage cell wall structure and digestibility. H. G. Jung, D. R. Buxton, R. D. Hatfield and J. Ralph (Eds.). ASA,CSSA, SSSA, Madison, WI. Tilley, J. M. A., and R. A. Terry. 1963. A twostage technique for in vitro digestion of forage crops. J. British Grassland Soc. 18: 104-111. Turner, N. D., G. T. Schelling, L. W. Green e, and F. M. Byers. 1990. Modification of in vitro an in situ dry matter digestibility of forages with chemical treatment. J. Prod. Agric. 3: 83-87. Vagnoni, D. B., W. M. Craig, R. N. Gates, W. E. Wyatt, and L. L. Southern. 1995. Monensin and ammoniation or urea supplem entation of bermudagrass hay diets for steers. J. Anim. Sci. 73: 1793-1802. Van Soest, P. J. 1994. Nutritional ecology of the ruminant. 2nd ed. Cornell University Press, Ithica, NY. Van Soest, P. J., J.B. Robertson, and B.A. Lewis. 1991. Methods for dietary fiber, neutral detergent fiber and non-starch polysacchar ides in relation to animal nutrition. J Dairy Sci. 74: 3568-3597. Varel, V. H., K. K. Kreikemeier, H. G. Jung, and R. D. Hatfield. 1993. In vitro stimulation of forage fiber degrad ation by ruminal microorganisms with Aspergillus oryzae fermentation extract. Appl. Environ. Microbiol. 59: 3171-3176. Wang, Y., and T. A. McAllister. 2002. Investigation of exogenous fibrolytic enzyme activity on barley straw using in vitro in cubation. J. Anim. Sci. 80 (Suppl. 1): 316. Wang, Y., T. A. McAllister, L. M. Rode, K. A. Beauchemin, D. P. Morgavi, V. L. Nsereko, W. Z. Iwassa, and W. Z. Ya ng. 2001. Effects of an exogenous enzyme preparation on microbial protei n synthesis, enzyme activit y, and attachment to feed the rumen simulation technique (Rusite c). British L. Nutrition 85: 325-332. Wang, Y., B. M. Spratling, D. R. ZoBell, R. D. Wiedmeier, and T. A. McAllister. 2004. Effect of alkali pretreatment of wheat st raw on the efficacy of exogenous fibrolytic enzymes. J. Anim. Sci. 82: 198-208. Williams, C.H., D.J David, and O. Iismaa. 1962. The determination of chromic oxide in feces samples by atomic absorption spectrophoto metry. J. Agric. Sci. 59: 381385. Wilman, D. and Adesogan, A. 2000. A comp arison of filter bag methods with conventional tube methods of determining the in vitro digestibility of forages. Anim. Feed Sci. and Technol., 84: 33-47.

PAGE 148

136 White, B. A., R. I. Mackie, and K. C. Doerner. 1993. Enzymatic hydrolysis of forage cell walls. p 455-484. In:Forage cell wall structur e and digestibility. H. G. Jung, D. R. Buxton, R. D. Hatfield and J. Ralph (Eds.). ASA, CSSA, SSSA, Madison, WI. Wilson, J. R. 1991. Plant structures: Ther e digestive and physical breakdown: Recent advances on the nutrition of herbivores . Y. W. Ho, H. K. Wong, N. Abdullah and Z. A. Tajuddin (Eds.). Malaysian Soci ety of Animal Production, Kuala Lumpur, Malaysia. Wilson, J. R. 1993. Organization of forage plant tissues. p 1-32. In:Forage cell wall structure and digestibility. D. R. B. H. G. Jung, R. D. Hatfield, and J. Ralph (Ed.). ASA-CSSA-SSSA, Madison, WI. Wilson, J. R., D. E. Akin, M. N. McLeod, a nd D. J. Minson. 1989. Particle size reduction of the leaves of tropical and a temperate gr ass by cattle. Ii. Relation of anatomical structure to the process of leaf breakdo wn through chewing and digestion of leaf breakdown through chewing and digestion. Grass and Forage Sci. 44: 65-75. Wilson, J. R., and J. B. Hacker. 1987. Compar ative digestibility a nd anatomy of some sympatric c3 and c4 arid zone grasse s. Aust. J. Agric. Res. 38: 287-295. Wilson, J. R., and P. W. Hattersley. 1989. Anatom ical characteristics and digestibility of leaves of Panicum and other grass genera with c3 and different types of c4 photosynthetic pathway. Aust. J. Agric. Res. 40: 125-136. Wood, T. M., and M. K. Bhat. 1988. Methods for measuring cellulase activities. p 87112. In:Methods in enzymology. W. A. W ood and S. T. Kellogg (eds.). No. 160. Academic Press, London, UK. Yang, W. Z., K. A. Beauchemin, and L. M. Rode. 1999. Effects of an enzyme feed additive on extent of diges tion and milk production of l actating dairy cows. J. Dairy Sci. 82: 391-403. Yang, W. Z., K. A. Beauchemin, and L. M. Rode. 2000. A comparison of methods of adding fibrolytic enzymes to lactating cow diets. J. Dairy Sci. 83: 2512-2520. Yu, P., D. D. Maenz, J. J. McKinnon, V. J. Racz, and D. A. Christensen. 2002a. Release of ferulic acid from oat hulls by Aspergillus ferulic acid esterase and Trichoderma xylanase. J. Agric. Food Chem. 50: 1625-1630. Yu, P., J. J. McKinnon, and D. A. Christense n. 2005. Improving the nu tritional value of oat hulls for ruminant animals with pretre atment of a multienzyme cocktail: In vitro studies. J. Anim. Sci. 83: 1133-1141. Yu, P., J. J. McKinnon, D. D. Maenz, A. A. Olkowski, V. J. Racz, and D. A. Christensen. 2003. Enzymatic release of reducing suga rs from oat hulls by cellulase, as influenced by Apergillus ferulic acid esterase and Trichoderma xylanase. J. Agric. Food Chem. 51: 218-223.

PAGE 149

137 Yu, P., J. J. McKinnon, D. D. Maenz, V. J. Racz, and D. A. Christensen. 2002b. The interactive effects of enriched sources of Aspergillus ferulic acid esterase and Trichoderma xylanase on the quantitative rel ease of hydroxycinnamic acids from oat hulls. Can. J. Anim. Sci. 82: 251-257. ZoBell, D. R., R. D. Wiedmeier, K. C. Olson, and R. Treacher. 2000. The effect of an exogenous enzyme treatment on production and carcass characteristics of growing and finishing steers. Anim. Feed Sci. Technol. 87: 279-285.

PAGE 150

138 BIOGRAPHICAL SKETCH Nathan Alfred Krueger was born in Flores ville, Texas and graduated Karnes City High School in 1995. Growing up he actively pa rticipated in 4-H and FFA while helping run his familyÂ’s custom hay baling business, commercial seed dist ribution facility, and livestock operations. After high school, he attended Texas A&M University where he actively participated in the Saddle and Sirloin club as well as the Beef Cattle Show Team. Also during this time Nathan had the opportuni ty to work as a student worker in the ruminant nutrition section of the Animal Sciences Department, which helped to develop his desire for conducting research and teach ing. In 1999, he received his Bachelor of Science degree from Texas A&M University. In 1999, Nathan was accepted into the Master of Science program in the Agricultural Education department at Texa s A&M University. During this time he was given the opportunity to assist in teaching courses as well as hone his teaching skills. While working on his masterÂ’s degree, Nathan was an active member of the Agricultural Education Graduate Student Society and a me mber of Gama Sigma Delta. He graduated with his Master of Science in 2001. In 2001, Nathan was accepted into the Ph.D. program in the department of Animal Sciences at the University of Florida under the guidance of Dr. Adegbola Adesogan. While pursuing his PhD in ruminant nutrition he was given the opportu nity to assist in teaching and to guest lecture in several of the beef cattle nutrition courses and graduate level techniques courses. While at the Florid a, Nathan was also involved in the Animal

PAGE 151

139 Science Graduate Student Association, Gr aduate Students United and a member of Gamma Sigma Delta.