45 131311111 11111111 1111 33 1111311'111sg11 31:13 1333: “3333133 41112111131133111 11 1"1111' 3113333 1 1' 1 3'3": 1111112‘1'1=31='1"11l‘1 1.1.13 11.,31. . -. 3 11111173|3~|1331331133111313113‘1131'11111 1 3111.111 '1‘ -.' 1 11 H ., 33, 1| ... 1313111111 52¢: 3‘. :.‘..,._ ... .— W. .J—w“ v.“ ”—0 .— '1 1111111'13'31' 111'1'33113.‘ '1133; 11‘13'33333 | ‘1: 1111111111'131111211"'5fl13%33:3.:33333 1|". 131131113111113.3131:31.113111311111112331-1 11111333313 ”1111:3433 1111’ 131111 11-33311' 111333131. .1. 111 .3 1133133113 "1" 1111.1|"111111,1|1 13111311311 31W“ 11-3 11’ 131|1|j131 11' ,31333333 111‘ =1“ ,1 133119.1' 1”“ 11111311 “‘ 1111' 13333331 1.31% 1111113 11131333 1111 111111 3__a. 3'11 .11 1333313131111”: 1313? 1 1 11‘ 1'" 111111 33 331111 1311' 1111113111 11 .— ”a I I -. 3:4?“ 3 3, n“... - - . .2333... .- ..._...- _._.__.. .. . 1 ___._..._ .47: 11 I 1 11'1 11 1'13 111131 -9 .flw .._.- _.‘ ”.4... M...— :L; —.'..:...—' .1... - My 5331. 111151- 13111111113111.1111 ~_ 1311111 11111113131313 3- H-”* .33 .,,. .“' 1 . ..,. ’ ' .. . ‘ ' 4 -’ . - _ - ’. 3 ”,3; l" l -L I. ~ , _ . . J I. . t.) - . 3.. . - ~ ,, -13. . V._ E L;"“" ‘-'- '0 H. «.¢rv a u “w- .w-O“: ____«, ' 4. 1| ..-—~ 3:.— .- :.."'-:"~" . . .0 7‘1.“ 1‘1 111'”- 1'11} - .- .,...:.-—-o—a-— on '— . .ao -._¢- .. CorvGIC$ u . 1377. "o- ' 21. 1| 3 '1" ' -' '3‘ ‘ 111:1: ‘1; ~-: 511 : 111-" , .= 11111 1 ..1.==1.1=,1.. 11 2.311111 '; ’ 115W 1 111333333 1 1&311'111. ‘; . 2- 31' 111111 31 33‘111313:3 .. 1111111:13-113133- :3. .. 1 :- - . . -- , . =~'.,~,1 ' ,7 1 1 i" ' ' 1| .211111 :11z1‘1'131'é 1" 1.111 : 1" 11 11111111111111 1 1.1 z. ‘ an 11 z . 11111191111111 1311,. 11,91 - s:.1.;11'1'11.‘.=31:1=1éi .33 ' .3 _ 3 .1 ' 13-1.! 11:33 2331323: :331131? 3‘33; "11'11|1.1:§1|'o|1'.2 3333.3:3 : 33 _: .. .1 .1 - 3311111“ 5 , 315:1"31151‘1313: 13333 311.3%; :3. '- ':' . 3'31,- “I! 3 -3 ~3 33 3. 11353.11 11‘111'41'1111‘1111‘11'11' ’ 3... 3:513:33? igggR'I-3333.3._3:533-1‘ v 21' ' 11 - . 31 11111111.: 11111111; “111111.; - z . 1111121 11 11v “ ' ‘ 1 E11111-112111111-1111 1v" 2‘. 112.1111: 111. 33-3 3 3 3 3 3'31} '3‘3 .-31 '-'-: 33311333 3?. 111111'1'1' 9%?! £113 31! f: .3. ' 11111111111111.111'111'1 1 .~ 111211111 : 2:. 11: -3 “111133111. .113: 2211-131: 1311,; ,3 2 23:11:31 11=131 ‘11“? 111 - 7" “‘ 3‘ 21111113? 4113111111111111.: 1111.1E 1.311.131.1151 3:111. 11:11,. 1'1. ' .tii "1'; 1'11'11 { '. ":1' ' L 1'. ' 1' a "'03: i-.'§::31." .11": .1"'1'11i¥311‘11-"':1:111: ;-= 1:11 11 1 1.1.1 11.11. ~ .. - - 11111111111111: 1‘ 5-. = "-1' 1111-1111- « ..||1%1é' "111‘ 1111111“- “"111 3L1 .1133:11:1'31153121311213113331111: '5 , ".1111 ,1 .‘_:;* 1?... 3 '.=""§‘=|"E;~:‘-’"; 1x33'1i'1fi1‘1,f",:"|-_'.3, 5; .3. 1:1,"111111 “11‘; 1. 22111!- - 1-1111-1-‘111112 111 1'1 :1: "3 '1 1' 133191-33 ' 3:! . 33,3 131" 4’}: o.» r W . - ~ - ‘1!” "oil- “(BE-9.1.3.3"33 ‘5:- .43.. .3 3 11333313133l313333311133‘331 33333 3 = 11 1"" new 111.3131; 31333111111 1,! :11 11' r11 33333311135333=33_3111 1W1 -'131313‘:i‘31.11"1" 1 333 1.1.1.. .m»... , __,v,.--.-. .vuh- — “A ._. .‘tg..- -. o n ‘q a ' ' 1.. . ......- 3 .5 . - 23‘ . .43.}: V I . . ”1.42" 1-23, .3 23': “'13. .. 4:. 31311331111111 3 1133 '11;1,1 =11 ‘ .31 1'- ' 333 3 1111 131113131 1.13:, 3 .. .. : =“- 3333 3 3,331.73 ' . t3'3 ':‘,' 33513313 13133 1 ‘ 39.3- 1.3 11... 3333332333233 333313133 333.333 3 33:1333 ‘ :3 11'1 116411333313; i3231'111133333.33.1111"1 13333333313333: 311111“?! (331333 3113133333 33313-121313uj3133-313 335331133311 3.33.333 . 3 . ~ ' _ ‘. 111 1131“ 1‘.- '== .1 . 11111111111121; . ‘ -= 113311: wcbt ‘V y... «I ‘ u - - ‘ A .5“. -. “£- gg. n;- .310: ¢ 2%.. «1% ""i!‘ "r- 11‘, "= '1 333‘ .._3333 £333 ‘3 33 - ~33 3’ 3 331%)“. 5g3fl’3'3n‘131 3 11333 i=3m~=113§ns3:" "113131 33113133333331 1111 1111;. .311 . - ..?!1 3 331131“ ‘33:: 1 1 11 1111111111 111111111 111;;- ‘11 1111 11191—131112; 3: 1, Jfl-K‘K'dvdI311W'1'111131igi': 3.1131813111311311‘1311 fl“ 1’ FE 313-13333'13313113-111111 333'131'5'11 a1] :35,th 31.333311“ 3 :33: «'31? 1 ‘2', 3:11)! - 1”» 11:31 ‘ . flux .1‘ ;-: ‘1 13 ‘5' N. :13:\3 C 1' : 1" 2.2111 1.1.11;- 1' “11.1111‘ . .-.~= 11:11.'-‘~=;:"1|331' -=1 19.:- 1111 =11 . - . 1.11.31.13th 33-31333 33:3I33 33‘“ if” 3 ‘3 33.1315. . ' 1111111111111 11311411,.«1115133. 3.111111 3111111111 11'111‘1 331311111111 “$311111 11111-111 11.12-11.11:- 11.11111. 1 - > 9 d‘ 4L-m-1‘31’ M124 .. ; l' .0 ,msu) Ill”llllllllllllllIlllllllllllllllllll 3 1293 10672 6825 Michigan State L ...‘.:.._...’:,? ll This is to certify that the thesis entitled THE EFFECT OF IONOPHORES, GLYCOPEPTIDES AND THEIR COMBINATION ON CULTURABLE RUMINAL BACTERIA AND VARIOUS RUMINAL PARAMETERS presented by Patty Sue Dickerson has been accepted towards fulfillment of the requirements for Master of Science degree in Animal Science Major professor Date 5/16/86 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES- —_ RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. Hg? '8 w ”)7 l, ‘ -. . — l 1“” 2(28’3. fiwomfi‘ n; . i W 15 ”WE murmurfl a: THE EFFECT OF IONOPHORES ' GLYCOPEPTIDES AND THEIR COMBINATION ON CULTURABLE RUMINAL BACTERIA AND VARIOUS RUMINAL PARAMETERS BY Patty Sue Dickerson THESIS Submitted to Michigan State University in partial fulfillmemt of the requirement for the degree of MASTER OF SCIENCE Department of Animal Science 1986 7'0 5’0 .9 :7 ABSTRACT THE EFFECT OF IONOPHORES, GLYCOPEPTIDES AND THEIR COMBINATION ON CULTURABLE RUMINAL BACTERIA AND VARIOUS RUMINAL PARAMETERS BY Patty.Sue Dickerson Ionophores and glycopeptides were investigated in vitro and in vivo to determine their effects on bacterial growth, fermentation patterns and protein degradation. In the in vitro study pure cultures of ruminal bacteria were utilized to determine the minimum concentration of ionophores and an ionophore-glycopeptide combination necessary to alter growth of these organisms. Of the compounds examined monensin and lasalocid were the most effective in inhibiting bacterial growth. In the in vivo study four cannulated steers were used in) investigate fermentation shifts, alterations in protease and deaminase activity and bacterial growth rates in the rumen due to addition of an ionophore, a glyc0peptide or a combination to the diet. Of the compounds examined, only the narasin- actaplanin combination significantly altered the volatile fatty acid profile. With respect to alterations in bacterial protease and deaminase activity and bacterial growth, the compounds examined were ineffective in significantly changing any of these parameters. ACKNOWLEDGMENTS . Although it is often hard to state ones true appreciation in a few lines, I feel that recognition to those who made this project successful is a necessity. I hope, however, that those mentioned know that my graditude goes far beyond my words. First I would like to sincerely thank Dr. W.G. Bergen for his guidance, advice and continued support throughout this project. Next I would like to express my appreciation to both Dr. M. Bennink and Dr. M.T. Yokoyama for contributing their time and valuble expertise in this area. I am deeply grateful to Dr. Douglas Bates, for his expertise in this area and his willingness to share it with me. Also, sincere graditude is extended to Liz Rimpau for the many hours she donated to this study. Without her dedication I would have been at a loss. I also wish to thank Scott Barao and Sally Johnson for their aid throughout this experiment. My most sincere and grateful thanks go to the people who not only contributed their time but their emotional support. I would like, therefore, to recognize my brother Larry, and my dear friends, Kris Johnson, Marilyn Loundenslager, Gary Weber and Bill Rumpler. Their friendship has made this time at Michigan State University not only pleasant but enjoyable. ii TABLE OF CONTENTS LIST OF TABLES.........................................Vii LIST OF FIGURES........................................ ix INTRODUCTION........................................... LITERATURE REVIEW...................................... 1 2 Diet and Rumen Ecosystem Interaction................ 2 Nitrogen Metabolism....;............................ 6 Overview......................................... 6 Protein Hydrolysis............................... 7 Deamination...................................... 11 Protein Synthesis................................ l6 Ionohores........................................... 20 Feedlot Performance.............................. 20 Lactic Acidosis.................................. 22 Bacterial POpulation............................. 23 Fermentation..................................... 25 Mode of Action................................... 28 MATERIALS AND METHODS.COCOOOOOOOOOOOOOOOO...0.0.0.0.... 31 In Vitro Study Overview............................. 31 Media Preparation................................... 33 In Vivo Study Overview.............................. 35 Laboratory Preparation.............................. 36 Protease Analysis................................... 39 Deaminase Analysis.................................. 40 RNA Determination................................... 41 Protein Determination............................... 42 Volatile Fatty Acid Analysis........................ 43 Ammonia Nitrogen Analysis........................... 45 Statistical Analysis................................ 45 RESULTSCOOCOCOOCCCOOCOO0.0......OOOOOOOOOOOOOOOOOOOOOOO 46 In Vitro Experiment................................. 46 In Vivo Experiment.................................. 49 Volatile Fatty Acid Concentrations............... 49 Protease Activity of Adherent and Free Floating Bacteria..................................... 54 iii iv Cell Free Supernatant Proteolytic Activity....... Deaminase Activity of Adherent and Free Floating BacteriaOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0...... RNA/Protein Ratios for Adherent and Free Floating BaCteriaeocoo-coco00.000.000.000.00.0.0000... Rumen Ammonia LeveISOOIOCOOOOOCOOOOOOOOO000...... DISCUSSIONOOOOOOOOOOOOOOO...OIOOOOOOOOOOOOOOOOOO0...... In Vitro Experiment................................. In Vivo Experiment.................................. Volatile Fatty Acids............................. Protease and Deaminase Activity in the Rumen..... RNA to Protein Ratios............................ Rumen Ammonia Nitrogen........................... Relationship of VFAs, Ammonia Concentrations, Protein Degradation and Bacterial Growth in the Rumen................................. LITERATURE CITEDOO0.0.0.0...O0....OOOOOOOOOOOOOOOOOOOOO 59 59 63 67 72 72 75 75 77 81 85 86 LIST OF TABLES Table Page 1 Fermentation Patterns Of Concentrate And Roughage DietSOOO0.00...OOOOOOOOOOOOOOOOOOOO... 5 2 Pure Culture Ruminal Bacteria.................. 32 3 Complex Carbohydrate Rumen Fluid Media......... 34 4 Composition Of Basal Diets..................... 37 5 Lowry Reagents................................. 44 6 Minimal Inhibitory Concentrations For Ruminal BacteriaOOOOOOCOOOOOOOI...OOOOOCOOOOOOOOOOOOOOO 47 7 Ruminal Acetate Concentration Across Treatment With Time After FeedingOOOOOOOOOO0.000000000000 50 8 Ruminal Propionate Concentration Across Treatment With Time After Feeding.............. 52 9 Overall Treatment Means For Butyrate Concentration Across Treatment................. 53 10 Pooled Treatment Means For Butyrate Concentration With Time After Feeding.......... SS 11 Overall Treatment Means For Protease Activity For Adherent Rumen Bacteria.................... 56 12 Pooled Treatment Means For Protease Activity For Adherent Bacteria With Time After Feeding.. 57 13 Overall Treatment Means For Protease Activity For Free Floating Rumen Bacteria............... 58 14 Pooled Treatment Means For Protease Activity For Free Floating Bacteria With Time After FeedingOOOOOOOOOOOOOIOOOOOOOOCOOOOOOOOCOOOOOIOI 60 15 16 17 18 19 20 21 22 23 vi Overall Treatment Means For Protease Activity For Cell Free Supernatant...................... Pooled Treatment Means For Protease Activity For Cell Free Supernatant With Time After Feeding.....OOO0....0..OCCOOOOOOOOOCOOOOOOOOOOO Overall Treatment Means For Deaminase Activity For Adherent Rumen Bacteria.................... Pooled Treatment Means For Deaminase Activity For Adherent Bacteria With Time After Feeding.. RNA/Protein Means For Adherent Bacteria Across Treatment With Time After Feeding.............. Overall Treatment Means for RNA/Protein For Free Floating Bacteria......................... Pooled Treatment Means For RNA/Protein For Free Floating Bacteria WIth Time After Feeding. Overall Treatment Means For Ruminal Ammonia Nitrogen Concentration......................... Pooled Treatment Means For Rumen Ammonia Concentration With Time After Feeding.......... 61 62 64 66 68 70 71 LIST OF FIGURES Figures Page 1 Schematic Of Separation Of Whole Rumen contentSOOOOOOOOO0......OOOOOOOOOOOOOOOOO..0. 38 2 RNA/Protein Ratios For Adherent Bacteria Across Treatments With Time After Feeding.... 83 INTRODUCTION With monensinfs overwhelming success in the feedlot industry, other fermentation manipulators have been developed to compete with monensin. To date the only other semi-successful feed additive of this type is lasalocid but it falls acdistant second to monensin in the market place. Along with the continual introduction of new compounds an increasing interest towards understanding these compunds action in the rumen has arisen. Although it is well recognized that monensin alter the rumen bacterial population which shifts the volatile fatty acid profile and leads to an increase in feed efficiency in the animal, there is mounting evidence that there are other underlining effects of ionophores. Raun et a1. (1976) spectulated that these compounds have a protein sparing effect in the rumen which probably contributes to the increased efficiency. With this baseline understanding there is a need to investigate these new compounds to determine their effects in the rumen. Thus, the following investigations were designed to examine the effects both ionophores and glycopeptides have on ruminal bacteria and to investigate their action in the rumen. LITERATURE REVIEW DIET AND RUMEN ECOSYSTEM INTERACTION The microbial population inhabiting the rumen is influenced in large measure, by the dietary constituents with the carbohydrate source being the most influential. Hemicellulose, cellulose, pectin, starch and sugars are the carbohydrates which are typically present in the rumen. Hemicellulose, cellulose and pectin are cell wall constituents and starch and sugars are associated with the cell contents. Typical roughage-type diets fed to ruminants contain large amounts of cell wall constituents in comparison to soluble carbohydrates while grain or concentrate diets are the reverse. Under roughage feeding regime then, the predominant bacterial strains inhabiting the rumen are the fiber digesters (e.g. Ruminococcus albus, Ruminococcus flavefaciens, Butyrivibrio fibrisolvens and Bacteroides succinogenes (Hungate, 1966; Schwartz and Gilchirst, 1975). These ruminal organisms are closely associated with the solid fraction of the digesta (Hobson and Wallace, 1982). Forsberg and Lam (1977) showed that seventy-seven percent of the diaminopimelic acid (DAPA), a compound found exclusively in procaryotic organisms, is associated with the particulate mattemu Using radioactive labeled cellulose, Rasmussen et.a1.(1983) determined that 2 3 the extent of attachment is regulated by the substrate supply. Therefore, on a diet consisting mainly of roughages the feed particles are saturated with adherent bacteria. Cheng and Costerton (1980) postulated that the attachment is a specific and rapid process“ 'Under high roughage feeding conditions the cellulolytic bacteria are not, however, the only bacteria present in the rumen. A secondary class of organisms exist which survive and proliferate through the utilization of intermediates and/or end-products produced by the primary population (Schwartz and Gilchrist, 1975). With the shift to high concentrate diets «Lg. alpha linked carbohydrates) the rumen ecosystem shifts from a cellulolytic dominated population to a amylolytic population. A large proportion of high concentrate diets consists of readily fermentable carbohydrates (starch and sugars) which support amylolytic organisms like Streptococcus bovis, Bacteroides amylophilus, Bacteroides ruminicola and Selenomonas ruminantium (Hungate, 1966). Cellulolytic bacteria are characteristic of slow growing organisms while the amylolytic bacteria are generally rapid growing organisms. With the availability of readily fermentable carbohydrates the cellulolytics can not compete with the rapid growing organism and thus, they lose their dominating position in the rumen ecosystem under these conditions (Hungate, 1966). Overall fermentation on high concentrate diet in comparison to the fermentation of high roughage diet is 4 illustrated in Table 1. Concentrate diet supports a larger population of organisms than a roughage diet (Ogimoto and Imai, 1981). The pH range for a high concentrate diet is 5.75-5.80 while a roughage diet range is 6.0-7.0 (Ogimaoto and Imai, 1981). Total volatile fatty acid (VFA) production increases with the increase of cereal grains in the diet (Orskov, 1982). In fact, the proportions of the VFAs shift towards a higher propionate production. This primarily results from a higher concentration of lactic acid utilizers which convert lactate to propionate (Mackie and Gilchrist, 1979). The protozoal population of the rumen, also, differs with alternate substrates. Purser (1959) showed protozoa numbers are considerably lower in animals fed a high concentrate diet as compared to a high roughage diet. This difference is primarily due to the lower ruminal pH which accompanies the feeding of readily fermentable carbohydrates (Purser, 1959). With respect to the rumen fermentation there seems to be a clear divison between high roughage diets and high concentrate diets. Corn silage, a prevalent feedstuff in Michigan, is a mixture on a dry matter basis of approximately fifty percent forage and fifty percent cereal grain (Goodrich and Plegge, 1984). It seems reasonable then, to rank corn silage as an intermediate feedstuff between high roughage and cereal grain diets in respect to the overall microbial population inhabiting the rumen. Therefore, it appears that a corn silage diet should TABLE 1. FERMENTATION PATTERNS OF CONCENTRATE AND ROUGHAGE DIETS. Concentrate Roughage Maize Barley Timothy Lucerne pH 5.76 5.96 6.61 6.81 ammonia (mg/dliter) 3.2 28.7 8.2 16.2 acetate UnM) 34.8 42.8 53.8 64.3 propionate UnM) 28.7 23.7 12.3 17.7 butyrate hnM) 3.8 11.1 4.7 5.3 6 support both cellulolytic and amylolytic bacteria in relatively equal pr0portions and all ruminal parameters should be between the aforementioned extremes. NITROGEN METABOLI SM Overview Ruminal nitrogen metabolism is a series of complex reactions which are conducted by the ndcmobiota inhabiting the system. Dietary proteins, as well as endogenous proteins, are degraded first to amino acids and short peptides in the rumen. Through further catabolism the amino acid are deaminated and/or decarboxylated to form branched- chain volatile fatty acids, carbon dioxide and ammonia. These products are then, utilized by the microorganisms as growth factors in respect to the two former and as a nitrogen sources in the case of the latter. This entire process of protein degradation and metabolism is carried out by a wide range of organisms interacting with one another. Examination of ruminal nitrogen metabolism is as complex as the reactions and organisms involved. Therefore, to facilitate the discussion the subject shall be subdivided into three major categories; 1) protein’hydrolysis, 2) intermediate catabolism and 3) microbial protein synthesis or cell growth. Protein Hydrolysis Protein hydrolysis is the first step in the overall processing of dietary protein. The hydrolysis, according to Nugent and Mangan (1973), is the rate limiting step in ruminal protein degradation. Through the use of uniformally radioactive labelled leaf fraction I protein these workers showed proteolysis exhibits first order kinetics. Earlier work conducted by Henderickx and Martin (1969) and Robson and Wallace (1982) indicated that the rate of proteolysis is dependent on the proteins solubility. Recent research, however, supports a more complex viewpoint. Although many proteins, like bovine serum albumin and ovalbumin, are soluble in water their rate of degradation in the rumen as compared to casein, a relatively insoluble protein, is much slower (Robson and Wallace, 1982). According to Mangan (1972) the rate of proteolysis of casein is approximately 4.6 pecent per hour while the rate of degradation of ovalbumin is considerably slower. Mangan (1972) attributed this to the cyclic structure of the ovalbumin. This molecule lacks both the terminal amino and carboxyl ends which inhibits the attachment of the exoproteases, thus, decreasing the rate of proteolysis. Nugent and Mangan (1978) showed that the disruption of disulfide bridges present in bovine albumin by the addition of dithiothreitol increases albumin proteolysis several fold. By comparing the degradation of a variety of soluble and insoluble proteins 7 8 by the ninhydrin method (which measures liberated amino acids), Mahadevan and coworkers (1980) provided further evidence that solubility alone is not a strong enough criteria for the rate of protein degradation and that structural characteristics like disulfide bonds are important. Wallace and Kopecney (1983) indicated through the use of azocasein that secondary and tertiary structures influence the overall rate of protein hydrolysis. The rate of proteolysis, therefore, is determined by many factors other than solubility lite. disulfide bridges, tertiary structures and availability of end terminal groups of the moleculeo (Bergen and Yokoyama, 1977; Hobson and Wallace, 1982). Factors other than the molecular characteristics of a protein determine the extent to which the protein is degraded in the rumen. Dietary regime plays an important role in the overall hydrolysis of a protein. Nugent and others (1983) demonstrated when changing from a hay- concentrate diet to a lucerne diet the rate of proteolysis of casein, leaf fraction I protein and bovine serum albumin increases. Earlier reports by Blackburn and Hobson (1960) showed proteolytic activity varies with diet but the change is not apparent immediately after switching the diets. The change is, therefore, an adaptive response to the availability of protein as a substrate. Through the isolation of strong proteolytic bacteria Hazlewood and coworkers (1983) confirmed that an increase in ruminal 9 proteolysis occurs when fresh fodder diets are substituted for dry diets. In washed cell preparations, however, Annison (1956) and Warner (1956) reported activity is independent of the diet. Recently, Siddon and Paradine (1983) when comparing cereal and forage diets reported higher activity is associated with cereal diets when casein is used as the substrate. Compiling this information with the earlier studies Siddon and Paradine (1983) concluded that the higher activity is probably a function of microbial numbers rather than an increase in the activity of proteolysis which suggests that bacterial protease activity'is constitutive (Hungate, 1966). The rumen microbial system is mainly'a sacchrolytic population. Few, if any, true proteolytic bacteria have been isolated from the rumen. The majority of the proteolytic activity is linked to organisms which have been already classified as major ruminal inhabitants. Primarily then, ruminal bacteria ferment.dietary carbohydrates forlenergy and utilize proteolytic activity to sequester nitrogen for use in microbial protein production or growth. Brock et a1. (1982) reported that the proteolytic activity in the rumen is associated twenty-five percent with the fluid fraction and seventy-five percent with the particulate fraction, indicating protease activity is possessed by all ruminal inhabitants. While attempting to isolate and characterize proteolytic ruminal bacteria Fulghum and Moore (1963) discovered that the majority of organisms isolated from IO ruminal ingesta tested positive for proteolytic activity. Blackburn and Hobson (1960) showed all fractions of rumen microbial population, large and small bacteria as well as protozoa, possess proteolytic activity. The relative amounts of activity, however, varies between the bacterial fractions and protozoa fraction. Blackburn and Hobson (1960) expressed their data on activity per weight bases which illustrates that the large bacteria have the highest activity between the separated fractions. The specific activity is six to ten times higher in bacteria fraction when compared to the protozoa fraction (Brock et al., 1982). The organisms which are highly proteolytic are primarily gram negative (Kopecny and Wallace, 1982) like Butyrivibrio sp., Succinivibrio spp., Selenomonas ruminantium var. lactilytica, Lachnospira multiparus, Bacteroides ruminicola, Borrelia spp., and Bacteroides amlephilus. Streptococcus bovis, although a gram positive organisms is also highly proteolytic (Fulghum and Moore, 1963; Brock et al., 1982). The proteolytic activity of the rumen is then, primarily due to bacterial species which are known to already occur in large numbers in the rumen. Therefore, it seems that this process as a whole is a secondary mechanism in the major species of the ruminal bacteria and suggests that the importance of proteolysis is more related to the cell growth rather than the survival of the organisms. The process of protein hydrolysis is carried out by 11 proteolytic enzymes which are associated with the cell wall (Nugent and Mangan, 1978). With the aid of nonionic detergents and density gradient centrifugation of the bacterial membranes, Kopencny and Wallace (1982) determined that the largest proteolytic activity is associated with the cell coat or capsular material of the bacteria. Some activity is contained in the intracellular material, however, these protease enzymes are primarily important in turnover of cellular protein rather than dietary proteins. The only instance these endoenzymes become involved in the degradation of dietary proteins is when cell lysis occurs (Goldberg and St. John, 1976; Kopecny and Wallace, 1982). Further investigation showed that the-predominant proteolytic enzymes of the rumen bacteria are serine-type proteases (Kopecny and Wallace, 1982; Forsberg et al., 1984). Forsberg et a1. (1984) reported that approximately sixty to seventy percent of the proteolytic enzymes are serine type but this probably changes with diet. Deamination Ruminal intermediates of total protein hydrolysis, amino acids and small peptides, are found at very low levels in rumen fluid. These intermediate metabolites, therefore, must be rapidly catabolized to other constituents (McDonald, 1952; Lewis, 1955; Annison, 1956). Pilgrim et al. (1970), through the uses of N-lS, showed amino acids are directly 12 incorporated into the bacteria and the proportion of direct microbial utilization of the amino acids is diet dependent. Feeding a low energy:low nitrogen diet the bacteria incorportate eighty percent of their nitrogen in the form of ammonia. The ammonia available to the organisms arises from degradation of dietary proteins and the ammonia recycling process. When high energyzhigh nitrogen diets are fed the percentage of ammonia utilized for microbial protein synthesis declines to less than sixty-five percent. This suggests that amino acids are directly used in the production of microbial protein when high amounts of both energy and nitrogen are available. Nolan and Leng (1972) further investigated amino acids utiliztion by ruminal bacteria and revealed similar information. For diets containing twenty to twenty-six percent crude protein, twenty percent of the microbial protein is derived directly from the incorporation of dietary amino acids. Similar findings have been reported by other as well (McMeniman et al., 1976; Salter et al., 1979; Armstrong and Weekes, 1983). Under typical feeding practices where neither energy nor nitrogen is limiting only twenty percent or less of microbial protein is derived from amino acids directly and since eighty-two percent of ruminal microorganisms require ammonia for growth (Bryant and Robinson, 1962) the majority of the amino acids formed through hydrolysis of dietary protein must be further degraded. As Mathison and Milligan reported, Portugal and Sutherland (1963) used of carbon 13 labelled amino acids to show that ten percent of the amino acids in microbial protein arose from direct incorporation while Weller et a1. (1962) reported up to eighty percent of dietary plant nitrogen is found in microbial protein. Again the above findings illustrate carbon and nitrogen of dietary proteins are separated during degradation.‘Therefore, the primary nitrogen involved in de novo synthesis of microbial protein is in the form of ammonia. Under normal feeding regimes ammonia nitrogen is not present in the diet, unless urea is fed, therefore, the ammonia must come from further catabolism of the amino acids. The precise amino acid degradation pathway predominating in the rumen has not been fully determined. Two possible pathways for amino acid catabolism by microorganisms have been elaborated. First, a non-oxidative deamination reaction resulting in the formation of volatile fatty acids and ammonia has been reported in Megasphaera elsdenii (Prins, 1977). It appears that a single enzyme is responsible for this non-oxidative reaction (Lewis and Elsdens, 1955; Walker, 1958; van den Hende et al., 1963). The Stickland reaction, a coupled oxidation-reduction reaction between two amino acids which produces carbon dioxide,eunmonia and volatile fatty acids (Prins, 1977) is the second prOposed pathway. Only certain amino acids have been found to be involved in this reaction. According to Barker (1961) the most activity involved amino acids are alanine, leucine, isoleucine, glycine, proline, l4 hydroxyproline and ornithinen Other amino acids have been implicated in the Stickland reaction but the relative rates of their utilization are not as high as for the above mentioned amino acids. It should be noted that the majority of the work in this area has been conducted with clostridia and this pathway has not been investigated in ruminal organisms. More importantly, the Stickland reaction has a pH optimum of about 6.0 while the non-oxidative deaminase pathway required pH is around neutrality (Prins, 1977). Under normal conditions the pH in the rumen ranges from 6.7 to 7.0 (near neutrality), therefore, the primary route for amino acid degradation must be via the non-oxidative deamination pathway. Ruminal microorganisms in an overall sense have been studied in respect to the effect diet has on the deamination process. The information, however, is quite limited. Basically, it has been shown that ruminal deaminase activity varies with diet. El-Shazly (1952) reported that in washed suspensions of rumen microorganisms deamination depends strongly on the diet. This dependence might be linked to types of amino acids found in the dietary proteins or to the type of bacteria population inhabiting the rumen at that time. Mangan (1972) demonstrated that different amino acids are degraded at different rates. Lewis and Emery (1962) reported that amino acids such as serine , cysteine, aspartic acid, arginine and threonine are dissimilated when added to either strained rumen fluid or washed cell 15 suspensions. The percent degraded is lower in the washed cell suspensions than the strained rumen fluid. This is probabily due to the lack of important co-factors which absent in the cell preparation. Chalupa (1976) showed that with incubations of mixed rumen microbial populations methionine and valine are degraded slowly while arginine and threonine are degraded much more rapidly and other essential amino acids fall into an intermediate group between these two extremes rates. Chalupa (1976) indicated that in vivo rates of degradation are much more rapid than in vitro rates, again implying the absence of important co-factors. Siddon and Paradine (1981) compared deaminase activity between cereal and forage dietsu Cereal diets exhibited a two-fold higher activity than the forage diets. These results may be related to the types of amino acids found in each diet or to the increased number of bacteria found with cereal diets or to the predominating species present under the feeding conditions. Since the information on the overall deamination process in anaerobic organisms and the rumen environment is limited it is necessary for futher investigation of this area before any firm conclusions may be reached. Therefore, experiments, in vitro as well as in vivo, need to be conducted to determine the overall existence and importance of the individual deaminase pathways in the rumen, Once a full understanding of this process from a bacterial stand point the manipulation of this process may be investigated. Protein Synthesis Through the utilization of protein breakdown intermediates, as well as the final products, the rumen microbiota synthesizes cellular protein. To date there seems to be a general consensus that the primary nitrogen source incorporated into the cell islammonia.as mentioned previously. However, there is evidence that bacteria in the rumen can and do utilize free amino acids and small peptides in synthesizing protein. Bryant and Robinson (1962) using freshly isolated rumen bacteria demonstrated that eighty-two percent of isolates could survive on media containing ammonia as the sole protein source while fifty-six percent of the isolates could use either ammonia or casein hydrolysate. They also established that twenty-five percent of the isolates have an obligate requirement for ammonia and only six percent require amino acids. Similar results have been reported by Stevenson (1978). Earlier work by Bryant and Robinson (1961) showed that Ruminococcus flavefaciens and Ruminococcus albus require ammonia as a sole nitrogen source regardless of the presence or absence of amino acids. Bacteroides succinogenes also requires ammonia.tolachieve maximal growth rate but can utilize amino acids to some extent. Maeng and coworkers (1976) determined that for maximal microbial cell yield and volatile fatty acid 16 17 production with mixed rumen bacteria grown in batch culture a seventy-five percent urea and twenty-five percent preformed protein on a nitrogen basis must be present. These studies suggest that some bacteria may incorporate preformed nitrogen compounds into their cellular protein. Teather et al. (1984) illustrated that the bacteria population increases seventy percent when a combination of urea-silage or soybean meal alone is fed as compared to urea alone. The primary species enhanced under these dietary conditions are Bacteroides ruminicola, Lactobacillus, Bifidobacterium spp., Fusobacterium spp., Butyrivibrio fibrisolvens, Megasphera elsdenii and Lactobacillus spp. which have been shown to have a essential requirement for amino acids (Allison, 1970). Maeng and coworkers (1976) provided information that the amount of amino acid incorporation into microbial cells is highest immediately after feeding and then rapidly declines. Pittman and Bryant (1967) determined that Bacteroides ruminicola possesses a general system for the uptake of peptides with these peptides are rapidly hydrolyzed during and after uptake. The primary function of these peptides appears to supply amino acids for the production of microbial protein. It appears that within the vast microbial population present in the rumen ammonia, amino acids and even peptides are incorporated into microbial protein. In fact it may be that the slower growing adherent bacteria (i.e. 18 cellulolytics) (Hungate,1966) use primarily ammonia as a nitrogen source while the free fraction incorporates either ammonia or the other intermediates depending on the availability of the compounds (Hungate, 1966). Microbial protein synthesis and efficiency values for a typical dilution rate during the day for both the free and adherent papulation have been reviewed by Bergen and coworkers (1982). Depending on the diet, the dilution of the liquid phase of the rumen may vary from .04-.12 per hour while the solid phase exhibits dilution values from .02-.09 per hour _ which suggests that the corresponding specific growth rates for specific phase associated bacteria must be greater than .1 and .05 per hour, respectively. Since protein metabolites are quickly degraded to ammonia it seems likely that only the bacteria with a high growth rate will be able to use these compounds. Nolan and others (1976) estimated that thirty to eighty percent of the dietary nitrogen passes through the ammonia pool before its incorporation into microbial protein. They inferred that carbohydrate availability might be partially responsible for this variation. Different carbohydrate are degraded at different rates in the rumen, thus, the energy microorganisms derive from the breakdown of these compounds is staggered. In other words, the bacteria which obtain energy from readily available carbohydates can use this energy to synthesize protein. Since at this point, the dietary protein has not been totally reduced.tx> ammonia, the incorporation of l9 peptides and amino acids seems likely for these species. The carbohydrates which exhibit a lag in digestion become an available energy source much later in the fermentation process. .At this point the dietary protein.is:strictly in the form of ammonia except that which can be considered as escape protein and the organisms which derive energy from these carbohydrates primarily incorporate ammonia into their cellular protein. The process by which the rumen bacteria transport amino acids and peptides is not well understood but the transport process of ammonia incorporation is well defined. Assimilation of ammonia into microbial protein can follow two distinct processes dependent on the ammonia concentration present. When ammonia concentration is low the high affinity enzyme glutamine synthase comes into play, however, at high concentrations of ammonia a low affinity enzyme dominates. Smith and Bryant (1979) used Selenomonas ruminantium to investigate ammonia transport and reported glutamine synthase pathway recycles glutamate via glutamine which eventally leads to the formation of two glutamates. This is at the expense of four moles of ATP. The glutamate dehydrogenase pathway consumes one less ATP per mole of ammonia assmilated in glutamate through the use of alpha- ketoglutarate. Since these two pathway exist the ability of the rumen microbiota to assimilate ammonia into microbial protein can occur irrespective of ammonia concentration present in the rumen. IONOPHORES Feedlot Performance Generally, the effects of different ionOphores on feedlot cattle performance are comparable. Many researchers have characterized the effects of these feed additives with the most investigated ionophore being monensin (Perry et al., 1976; Raun et al., 1976; Steen et al., 1978; Hanson and Klopfenstein, 1979; Perry et a1. 1979; Perry et al., 1983). Feedlot performance of cattle supplemented with lasalocid has been outlined by Bartley and coworkers (1979). Salinomycin, narasin and laidlomycin butyrate, all experimental compounds, have also been tested in the feedlot (McClure et al., 1980; Potter et a1. 1976;.Spires and Algeo, 1983, respectively). The overall effect of ionophores in feedlot cattle has been summarized by Owens (1980). When carbohydrates are highly available in the diet ionophores depress feed intake without an accompanied decline in body weight gain, therefore, overall feed efficiency'(eug. feed/gain ratio) is improved. Dyer and coworkers (1980) as well as Ferrell and others (1982) evaluated monensin and lasalocid when fed in conjunction with high concentrate diets and reported similar results. Feed efficiency of the animals increased in both of these experiments primarily because dry matter intake was 20 21 reduced. Including an ionophore in a high roughage diet yields a similar endpoint as with the high concentrate diet (e.g. enhances feed conversion), however, under these conditions average daily gains are increased without alteration in feed consumptions (Johnson et al. 1979; Owens, 1980; Brown et al., 1982). With corn silage diets supplemented with ionophores there seems to be a discrepancy in the literature to which parameter(s) are really altered. Perry et a1. (1983) illustrated with corn silage diets that monensin supplementation decreased dry matter intake without any apparent effect on average daily gains. Brown and coworkers (1982) observed increased average daily gains and stationary feed intakes when corn silage diets are supplemented with lasalocid. IonOphore supplemented cattle are the same as non- supplemented animals in respect to carcass characteristics (Dyer et al., 1980; Johnson et al., 1979; Thompson and Riley 1980; Perry et al. 1983; Rioni and Bittante, 1983). The cattle receiving feed additives, therefore, grade similarly tolthe1non-supp1emented cattle and bring a similar market price. The overall feed input per pound of gain or lean tissue, however, is considerably reduced with supplementation. This translates into an overall savings to the feedlot owner since feed consumption in the feedlot is less over the entire process with the supplemented animals. Lactic Acidosis Besides their ability to increase feed conversion ionophores decrease the incidence of lactic acidosis in feedlot cattle. Lactic acidosis can be described as a fermentation disorder which occurs as a result of grain engorgement. The basic symptoms of lactic acidosis are lacticacidemia, acid-base imbalance, rumen stasis, diarrhea, dehydration, sytemic acidosis and, in acute forms of the disease, cardiovascular and respiratory failure (Huber, 1976; Dennis et al., 1981a & b). These symptoms are attributed to the increase in lactate producing bacteria in the rumen (e.g. Streptococcus bovis) and the subsequent drop in ruminal pH (Dennis et al., 1981a & b). When readily fermentable carbohydrates become abruptly abundant the rapidly growing bacteria, generally lactic acid producers, become the predominating organisms and alter the fermentation process which results in a subsequent pH decline. Coinciding with this rise in lactate producers and decreased gflh the lactic acid utilizing bacteria disappear which further contributes to the accumulation of lactic acid (Counottee, 1978/1979). As the pH approaches 5.5 other gram negative organisms, as well as the protozoa, decrease in number while Steptococcus species and other lactate producers numbers continue to increase. Eventually the pH drops below 5.0 and the predominant organism becomes a Lactobacillus spp“. At this point the metabolic 22 23 alterations caused by excess lactic acid are apparent in the host. With the feeding of ionOphores in feedlot diets the problem of lactic acidosis is greatly depressed. Primarily, these feed additives inhibit the growth of lactic acid producers in the rumen without altering the lactate utilizing populations, therefore, preventing lactic acid accumulation. Nagaraja et al. (1981) illustrated the ability of ionophores to prevent lactic acidosis with the use of intraruminal administration of glucose. Nagaraja et al. (1981) concluded from this experiment that addition of ionophores to rations during the switch over help prevent lactic acidosis. Bacterial Population Ionophores like other antibiotics inhibit the growth of certain microoganisms. These compounds when feed to cattle alter the ruminal bacteria population which ultimately leads to a shift in the ruminal fermentation end— products. Extensive pure culture research has been conducted in order to determine the antibiotic sensitive strains in the rumen. Chen and Wolin (1979) determined that the ruminal microbes sensitive to antibiotics, monensin and lasalocid, are primarily the gram positive microoganisms “Lg. Ruminococcus albus and Ruminococcus flavefaciens). Chen and Wolin (1979) observed that gram negative species 24 (e.g. Bacteriodes succinogenes, Bacteroides ruminicola and Selenomonas ruminantium) are relatively resistant to these compounds. In general they concluded that gram positive organisms illustrate a high sensitivity to ionophores while gram negative bacteria are resistant to the compound. They reported, however, that Butyrivibrio fibrisolvens, a gram negative bacteria which possesses a gram-positive like cell- wall structure (Cheng and Costerton, 1977), is suppressed by the antibiotics. The microorganisms for which ionophores select for are succinate producers and succinate utilizers (eug. propionate producers). The inhibited organisms are the carbohydrate fermenters which produce formate, acetate, butyrate and hydrogen as their end-products. With this reduction of these specific end-products the survival of bacteria dependent upon these compounds is reduced. Chen and Wolin (1979) suggested that the reason methanogenesis is reduced with the presence of monensin is due to a reduction in the availability of substrate (eug. hydrogen) rather than a specific toxic affect on the cells. Another group of feed additives similar to the ionophores are the glyCOpeptide antibiotics. One of the more investigated glycopeptide antibiotics is avoparcin, a growth promotant for broiler chickens (Lesson et al., 1984), and ruminants (Chalupa et al., 1980). Froetshel and coworkers (1983) demonstrated that avoparcin causes similar shifts in the rumen population as monensin. Specifically, avoparcin inhibits many gram positive bacteria directly 25 while indirectly, through reduced hydrogen availability, decreased methane producers. Gram negative flora, like with monensin, are virtually unaffected by avoparcins presence (Stewart et al., 1983). Stewart et a1. (1983) suggested that avoparcin alters the ruminal microbial population in a similar manner to monensin but is required in higher concentration than monensin to yield the similar results. Fermentation It has been demonstrated that ionophores increase the molar proportion of propionic acid while decreasing acetate and butyrate proportions but does not influence the total volatile fatty acid (VFA) production (Thornton et al., 1976; Chalupa et.al., 1980; Richardson.et.al., 1976). ,As stated before methane production is reported to be reduced by an indirect effect on the methanogenic bacteria (Hungate et al., 1966; Chalupa et al., 1980; Richardson et al., 1976; Chen and Wolin, 1979). Glycopeptides, on the other hand, cause a similar shift in VFA concentration but these compounds also reduce total VFA concentration (Froetschel et al., 1983). The mechanism controlling the VFA shift has been attributed primarily to the selection for succinate—forming organisms (e.g. Bacteroides succinogenes and Bacteroides ruminicola) and propionate-producer (e.g. Selenomonas ruminantium) which decarboxylates succinate to propionate 26 (Chen and Wolin, 1979). Romatowski and coworkers (1979) illustrated in batch culture that monensin increases the succinate decarboxylating capacity of mixed rumen bacteria. Chalupa and others (1980) reported that the increase in propionate production is due to increased activity through the acrylate pathway. These findings, according to Chalupa et a1. (1980), suggest that the enhanced propionate production seen when monensin and other ionophores are present is probably a result of both population shift and increased enzyme activity. Methanogenesis is reduced in cultures exposed to ionophores. The alteration in methane production is a result of the selection by the ionophore against hydrogen producing organisms (e.g. Ruminococcus albus) (Chen and Wolin, 1979; Van Nevel and Demeyer, 1977). Therefore, the ionophore effect on methane production is a secondary characteristic and not a direct metabolic inhibitions of these organisms by the ionophore (Van Nevel and Demeyer, 1977). Blaxter and Waiman (1964) suggested that propionate can be more efficiently utilized by the ruminant. The primary reason is propionate has a lower heat increment than acetate. As reported by Rowe et al. (1981) twenty percent more metabolizible energy is available when the shift in VFA production has been accomplished. Rowe et a1. (1981) and Richardson et a1. (1979) supported the idea that a diet containing monensin is more efficiently utilized than an 27 unsupplemented diet. Bull et al. (1970), Johnson et al. (1972), Orskov et al. (1979) and Byers (1980), however, demonstrated that acetate and propionate are energetically the same and are utilized with similar efficiency for growth. The increase in performance, therefore, may not be totally due to the shift in VFA production. Raun and cowokers (1976) agreed that propionate fermentation is more favorable than an acetic and butyric acid fermentation from a energy stand point because carbon conservation. Still this phenomenon cannot account for all of the increase performance which occurs when monensin is fed to feedlot cattle. Raun et al. (1976) suggests that a suppression of deaminase and protease activity might contribute to the enhanced performance. Poos et al. (1979), Owens et al. (1980), Isichei (1980) and Van Nevel and Demeyer (1977) reported increased non-ammonia nitrogen reaching the lower gut with monensin supplementation which further supports Raun et al. (1976) hypothesis that ionophores are protein sparing. The increase in dietary protein reaching the lower digestive tract observed in the above studies ranged form twenty-two to fifty-five percent (Bergen and Bates, 1984). Since the range of ”by-pass" protein is quite large this might suggest that the process is governed by diet. With ruminal protein and amino acid degradation inhibited or reduced the major site of protein digestion is shifted to the lower gut which might account for some of the increased performance seen in feedlot cattle. One must remember, 28 however, that this shift could be detrimental if the dietary protein is of low quality. Since microbial protein is reduced and protein by-pass elevated with monensin the dietary protein should be of higher quality than need be in a diet without monensin. Van Nevel and Demeyer (1977) and Bartley et al. (1979) used in vitro techniques in determining that monensin and lasalocid decrease bacterial growth in cultures not previously exposed to either of the compounds. Herod et al. (1979) showed.that bacterial growth isrunzdepressed with monensin feeding. In mixed cultures, total and net growth, is reduced with supplementation of monensin while the amount of substrate fermented is approximately the same. Thus the microbial growth efficiency is reduced (Van Nevel and Demeyer, 1977). These workers incubated unadapted mixed cultures in casein and monensin and observed a lower protein degradation and ammonia production as compared to the control.‘This suggests that monensinfs effect in vivo may extend far beyond the simple population and VFA shift. However, as stated all these experiments were conducted using unadapted mixed ruminal bacteria and therefore the results may all indicate the effects of monensin when first added to the animals diet.i Further investigation of the area is necessary. Mode of Action Monensin and other carboxylic polyether ionophores have been described as cation-H ion antiporters (Harold et al., 1972). The direction of the tranport and degree of cation exchange is determined by the chemical gradients which exist (Bergen and Bates, 1984). The affinity of the particular cation varies between ionophores. Monensin is a sodium/hydrogen transporter (Pressman.et a1”.1976) while lasalocid prefers potassium, calcium and sodium. The chemiosmotic hypothesis, postulated by Peter Mitchell, states that metabolic energy is conserved at the level of the membrane as an electrochemical gradient of hydrogen ions (Harold, 1972). This gradient can be established through the passing of an electron down the electron transport chain or by the extrusion of a hydrogen ion during the hydrolysis of ATP (e.g. ATPase). In the case of hydrogen extrusion, the established gradient exerts a force on the electron and pulls it back into the cell. This causes a dissipation of the chemical gradient and may be coupled with movement of important metabolic substrates or ATP production (Bergen and Bates, 1984). Both bacterial and mammalian cells have been shown to utilize this electrosmotic energy to transport amino acids, sugars and other ions (Eddy, 1978; Rosen and Kashket, 1978; Booth and Hamilton, 1980). Rumen bacteria are obligate anaerobes dependent primarily on substrate phosphorylation for ATP synthesis 29 30 (Hungate, 1966). Some ruminal bacteria have a partial cytochrome system which contributes to the ATP supply (Prins, 1977). The ATP produced in bacteria is primarily used toward the production of the electrochemical gradient which in turn is used for secondary transport. As stated before, monensin effects are dictated by the concentration gradients of hydrogen and sodium ions. The ion in the highest concentration will be dissipated while the other, whether it be against its gradient, will be driven in the opposite direction. In the rumen environment the hydrogen gradient is greater than the sodium gradient. Thus the hydrogen electrochemical gradient which the cells have established will be dissipated. The cell then will be required to increase its ATP utilization to correct this problem. It is believed that organisms with electron transport chains can more easily adjust to this phenomenon because they can extrude hydrogen electrons via the electron transport chain. These organisms, therefore, are not required to use their ATP reserves to create a new proton gradient. In other words, monensin places an addition stress upon the microbiota inhabiting the rumen. Those organisms capable of adjusting to the increased ATP demand will survive while those unable to change will perish. Also, the organisms which survive may be unable to reach their maximum growth potential since energy usually partitioned towards growth, protein synthesis, will be shunted towards maintenance. Although it seems apparant that ionophores as 31 well as other similar compounds cause this response in ruminal bacteria the specific metabolic changes in the bacteria have not been fully explained. Therefore, a great need for further investigation into this area is needed. MATERIALS AND METHODS In Vitro Study Overview Eight pure cultures of rumen microorganisms were utilized in an experiment to determine the effects of four different fermentation manipulators «Lg. monensin, narasin, lasalocid and actaplanin) on microbial growth. This compounds were a gift from Elanco. The rumen bacteria involved in this study were Selenomonas ruminantiunn(Sr), Megasphera elsdenii (ATCC19169), Bacteroides ruminicola (B4), lBacteroides succinogenes ($85), IButyrivibrio fibrisolvens (DI), Ruminococcus flavefaciens (C94), Ruminococcus albus (7), and Streptococcus bovis (24). These ruminal organisms were grown in batch culture in a complex- carbohydrate rumen fluid media (Table 2) which contained either 0.0, 0.5, 2.5, 5.0, 10.0, or 20.0 ppm of the respective compound. On two consecutive weekends four of the eight microorganisms were carried through the following experiment to determine the minimal inhibitory concentration (MIC) of the compounds. Some ruminal bacteria are readily resistant to these compounds while others are either adaptively resistant or totally non—resistant an adaptation period seemed necessary to study these organisms in a restricted time period. One week prior to the commencement of the experiment, the organisms were adapted to each of the compounds at the 0.5 ppm level. Following the adaption period, fresh 0.5 ppm batch culture tube were inoculated 32 33 TABLE 2. PURE CULTURE RUMINAL BACTERIA. ORGANISM GRAM STAIN TYPE Selenomonas ruminantium (Sr) negative Not inhibited Megasphera elsdenii (ATCC19167) negative Not inhibited Bacteroides ruminicola (B4) negative Adapts Bacteroides succinogenes ($85) negative Adapts Butyrivibrio fibrisolvens (D1) negative Inhibited Ruminococcus flavefaciens (C94) positive Inhibited Ruminococcus albus (7) positive Inhibited Streptococcus bovis (24) positive. Slowed growth 34 with the organism and incubated at 39 C overnight for subsequent use the following day in the actual trial. The organisms which were unadaptable to the compounds were simply obtained from a fresh control batch cuture. At the onset of the trial the fresh cultures were transferred to the experimental test tube and incubated at 39 C in an incubation room. The incubation period lasted approximately forty hours during which optical density (OD) reading were taken every two hours for the first eighteen hours and at hours nine, fourteen and twenty-one for the remaining twenty-one hours. Optical densities were read from a Spectrophotometer 80 at a wave length of 660 nm. Media Preparation An anaerobically prepared complex-carbohydrate rumen fluid medium was used in this experiment (Table 3). Seventy-five milliliters (ml) of deionized water were added to a mixture containing twenty-five ml of clarified rumen fluid, 3.75 ml of both Mineral I and Mineral II, 0.1 ml of resazurin solution,(L1 ml hemin solution,(L1 m1 volatile fatty acid solution, 0.05 gram of cellobiose, dextrose, maltose and yeast extract and 0.2 gram of trypticase. The resulting mixture was adjusted to pH 6.7 with 6 N NaOH, followed by the addition of five ml of an eight percent sodium carbonate solution and the appropriate amount of 35 TABLE 3. COMPLEX CARBOHYDRATE RUMEN FLUID MEDIA. CONSTITUENTS ml/lOO ml Deionized water 75.0 Clarified rumen fluid 25.0 Mineral I 3.75 Mineral II 3.75 Resazurin solution 0.10 Hemin solution 0.10 Volatile Fatty Acid mixture 0.10 Cellobiose 0.05 g Dextrose 0.05 g Maltose 0.05 9 Yeast extract 0.05 g Trypticase 0.20 g pH to 6.7 with 6 N sodium hydroxide and bubbled under oxygen-free carbon dioxide 8% sodium carbonate solution 5.00 Cysteine chloride 2.00 stoppered, wired and autoclaved for fifteen minutes For more complete details consult Holdeman et a1”.1977. 36 antibacterial compound. The medium was placed in round bottom flask and brought to a boil under oxygen-free carbon dioxide. Following boiling the medium was allowed to equilibrate under oxygen-free carbon dioxide for an additional fifteen minutes before the aseptic addition of two ml of cysteine chloride solution. Then nine ml of the reduced medium were transfered anaerobically to a test tube (13 x 150 mm), stoppered and autoclaved at 121 C with 15 lb pressure for fifteen minutes to insure the media were free of unwanted bacterial contaminates. Upon commencement of the trial one ml of adapted culture was added to the each broth. A11 transfers were conducted according to the Hungate technique (Hungate, 1950) for culturing rumen anaerobes. In Vivo Study Overview Four ruminal cannulated Holstein steers weighing approximately 500 kg were used in a single 4 x 4 changeover repeat measure Latin square design to determine the effects of specific feed additives on the major ruminal fermentation processes. Steers were individually housed in pens approximately .fifteen by twenty meter in the metabolism room at Michigan State Beef Cattle Research Center. During the twenty-one day adaption and five day collection period the steers received a diet composed of corn silage and soybean meal supplement (Table 4L. The supplement was also 37 used as the treatment vehicle. The steers, during each block, received either: 1) soy, 2) soy plus 12.5 ppm monensin, 3) soy plus 33 ppm actaplanin and 4) soy plus 36 ppm narasin-actaplanin combination. Upon termination of each block the steers were given a three day rest period during which they received the control diet. Ten days prior to and during the collection periods the steers were fed at twelve hour intervals. Samples were collected over five consecutive days (one per day) to construct a 0, 2, 4, 7, 10 hours post feeding cycle. At each sampling period whole rumen contents were placed in a thermos and transported to the laboratory for further analysis. At the same time nineteen ml of strained rumen fluid was collected, fixed with one ml of saturated mercuric chloride and stored at 4 C for use later in the volatile fatty acid and ammonia nitrogen determination. Laboratory Preparation Freshly obtained whole rumen contents were squeezed through two layers of cheesecloth to separate the liquid and solid fractions (Figure 1). The strained rumen fluid (SRF) was saved for further fractionation to obtain free floating rumen bacteria. The solid residue remaining was resuspended.in an equal volume of cold anaerobic dilution solution (Bucholtz, 1972) to SRF. The resulting combination was mixed and strained through two layers of cheesecloth to 38 TABLE 4. COMPOSITION OF BASAL DIETS INGREDIENTS % DIET DM Corn, aerial pt, w-ears, w-husk, ensiled, well eared 88.7 mx 50% mm 30% dry matter Soybean, seeds, meal solv-exted 11.3 Crude protein content of diet equalled 12%. Adequate amount of vitamins and minerals were added. Supplement was used as carrier of experimental feed additives. 39 Whole Rumen Contents (0,2,4,7,10 hrs after feeding) Strained Rumen Pellet‘ strained through two layers of cheesecloth Fluid (SRF) Solids resuspended in anaerobic dilution solution Rumen Fluid Extract I Solids (RFE) (discard) centrifuge at 121 x g fifteen minutes (removes feed particles and protozoa) (discard) Pellet resuspended 0.1 M potassium phosphate buffer Supernatant I centrifuge at 45,000 x g fifteen minutes (removes bacteria from solution) I Supernatant II protease assay Washed-cell suspension protease assay deaminase assay RNA determination protein determination FIGURE 1. SCHEMATIC OF SEPARATION OF WHOLE RUMEN CONTENTS. 40 separate the liquid and solid fractions. The fluid obtained was classified as rumen fluid extract (RFE) and was retained for further processing to obtain adherent ruminal bacteria. After separation the SRF and RFE were processed by differential centrifugation. The first spin (121 x g) removed feed particles and protozoa while the second spin (45,000 x g) fractionated the first supernatant into a supernatant and a bacterial pellet. The final supernatant was retained for protease activity determination and the pellet was resuspended in ten ml of 0.1 M potassium phosphate buffer and vortexed until a homogeneous washed cell supension (WCS) was produced. The WCS was analysized for protease and deaminase activity, as well as, RNA and protein content. Protease Analysis Protease activity was determined on the free floating and adherent bacterial fractions by the method of Brock et al. (1982). Two ml of the WCS were added to one ml of a three percent azocasein solution. The sample was incubated in a 39 C water bath for ninty minutes. The incubation period was completed by the addition of 0.3 ml of fifty percent TCA. Immediately following the addition of the precipitant the sample was placed on ice for approximately thirty minutes, and then centrifuged at 27,000 x g for fifteen minutes. Two ml of the resulting supernatant were 41 added to two ml of an 1 N NaOH solution, vortexed for approximately thirty seconds and allowed to sit at room temperature for fifteen minutes prior to reading at 440 nm on the Gilford spectrophotometer. Cell-free supernatant was handled similarly to the WCS in determining the protease activity. With the supernatant, however, one ml of sample was combined with one ml of 0.1 molar potassium phosphate buffer and one ml of the three percent azocasein solution. Blanks were prepared in a similar manner to the experimental samples except TCA additions preceded the addition of the WCS and cell-free supernatant. For each set of samples a standard curve was constructed using final concentrations of .006, .012, .018, and .024 units of (Sigma Type P5380 Bacillus subtilis alkaline protease, crystallized and lypholized, 12 units/mg) protease enzyme activity per ml to calculate the proteolytic activity present in the three fractions. Deaminase Analysis A modified version of the Broderick and Balthorp (1979) procedure was used in the determination of deaminase activity. The original assay system required strained rumen fluid while the modified version replaced SRF with WCS. Two ml of the suspension were added to a solution composed of 0.4 m1 of an 1.8 percent casein hydrolysate and 1.6 ml of a 0.1 M potassium phosphate buffer. Blanks were prepared with 42 two ml of the potassium phosphate buffer rather than two ml of WCS. The sample was incubated for three hours. One ml of a fifty percent TCA solution was added to terminate the reaction before the sample was placed on ice. After thirty minutes the sample was centrifuged at 27,000 x g for fifteen minutes with the resulting supernatant being assessed for total amino acid remaining by the alpha amino nitrogen analysis (Palmer and Peters, 1969). Thus, 0.2 ml of the supernatant was added to a solution containing 1.6 ml of 0.05 M sodium borate and 0.2 ml of twenty-five percent solution of fresh trinitrobenzene sulfonic acid. The combination was incubated at 39 C for twenty minutes followed by addition of two ml of’l N hydrochloric acid. The absorbance was read on the Gilford spectrophotometer at 420 nm. Citrulline was used to constructed a standard curve. RNA Determination Two ml of WCS and two ml of ten percent TCA were combined in a twevle m1 centrifuge tube and placed on ice for thirty minutes. The bacterial pellet was harvested via centrifugation (27,000 X g) for fifteen minutes. The WCS pellet remaining was frozen and used later for RNA and protein determination. RNA was determined in WCS pellets according to Schneider (1957). Ribonucleic acids were extracted by 43 hydrolyzing the pellet in six ml of five percent TCA for thirty minutes in a 97 C water bath. The sample was allowed to cool prior to centrifugation at 45,000 x g for twenty minutes. The supernatant was carefully decanted and placed into an apprOpriately labeled test tube. Two ml of the supernatant was added to two ml of the orcinol reagent (one ml of orcinol per 100 ml of hydrochloric acid containing 0.5 g of ferric chloride). The sample was covered heated for twenty minutes in a 97 C water bath, allowed to cool to room temperature and then, read at 660 nm on the Gilford spectrophotometer. Standards were prepared in a similar manner from a stock solution composed of Sigma ribonucleic acid prepared from Torula Yeast. Concentrations of the standards utilized in the construction of the standard curve were .008, .010, .040, .100, and .200 mg/ml. Protein Determination The pellet remaining after the RNA extraction was completely hydrolyzed in three m1 of an 1 N NaOH by placing them in a 97 C boiling water bath for five minutes to extract the protein. Hartree (1972) modification of the Lowry procedure (1951) was used to determine the amount of protein present in the bacterial fractions. Since the Lowry procedure is only linear within a certain range of protein concentration 0.1 ml of the extract was mixed with 0.9 ml of water to insure protein concentration was within 44 the designated range. The diluted extract was then combined with 0.9 ml of reagent A (Table 5) and incubated for ten minutes in a 50 C water bath. Following the incubation the sample was cooled and 0.1 ml of reagent B (Table 5) was added. The sample was vortexed and placed at room temperature for approximately ten minutes. Finally, three ml of fresh Folin-phenol reagent (Table 5)‘were added and the final combination was vortexed and placed in a 50 C water bath for ten more minutes. Following a cooling period the sample was read on the Gilford spectrophotometer at 650 nm. Bovine serum albumin 'was used to create a standard curve . Volatile Fatty Acid Analysis Five ml of strained rumen fluid were placed in a twelve ml centrifuge tube containing one ml of fresh twenty- five percent meta-phosphoric acid, mixed and placed at room temperature for thirty minutes. This step deproteinized the sample. The sample was vortexed for thirty seconds and centrifuged at 15,000 x g for fifteen minutes. The supernatant was transferred to injection vials and capped. Two microliters of the extract was analyzed by a Hewlett- Packard gas-liquid chromatograph (Model 5840A). A stainless steel (183 cm by .32 cm) Chromosorb column (10% SP 1200 and 1% phosphoric acid 80/100 WAW mesh) (Supelco Inc., Bellefonte, PA) was used. The following conditions were 45 TABLE 5. LOWRY REAGENTS. Reagent A: Two 9 potassium sodium tartrate (sodium potassium tartrate) and 100 g sodium carbonate are dissolved in 500 ml of 1 N sodium hydroxide and diluted with water to 1 liter. Reagent B: Two 9 potassium sodium tartrate and one g c0pper sulfate are dissolved in ninety ml water and ten ml 1 N sodium hydroxide is added. Phenol Reagent: Use commercially prepared Harleco Item No. 2690. (Phenol Reagent, Folin and Ciocalteu). Dilute one part plus ten part water. Diluted solution will be stable for several weeks. ~46 programed into the GC: column temperature, 125 C; injection temperature, 170 C; flame ionization detector temperature, 175 C and carrier gas flow (helium) 25 ml/min. .A Supelco VFA standard was used to determine the concentration of acetate, propionate and butyrate present in the rumen fluid sample. Total run time for each sample was approximately 20 minutes. AmmoniaflNitrogen.Analysis Five ml of the mercuric chloride fixed rumen fluid were placed in a twelve ml centrifuge tube containing one ml of nine normal sulfuric acid. The resulting mixture was vortexed for thirty seconds and centrifuged at 45,000 x g for fifteen minutes. The supernatant was transfered to a suitable vial and analyzed on a Technicon Auto Analyzer II. A stock solution (one mg N/ml) of ammonia sulfate was prepared and used to construct a standard curve. Statistical Analysis The vivo study was statistical analyzed. by Genstat analysis for a repeat measure Latin square design. Bonferroni T statistic was used for the treatment comparisons. RESULTS In Vitro Experiment Growth curves, OD vs time, were constructed for each organism with respect to the various antibacterial agents and their varying concentrations. The curves for an organism within a particular treatment were compared to determine the minimum concentration (i.e. .5, 2.5, 5, 10, or 20 ug/ml) required to inhibit the organisms growth. This concentration will be referred to as the minimum inhibitory concentration or MIC. The antibacterial agents (Table 6) used in the experiment were ineffective against S. ruminantium, M. elsdenii, and B. ruminicola, three gram negative organisms. Lasalocid, however, did increase S. ruminantium's lag phase at all the examined concentrations. A similar shift in the lag phase was observed when B. ruminicola was exposed to the full spectrum of antibacterial agents. B. succinogenes and B. fibrisolvens, two other gram negative organisms, were sensitive to the antibiotics. Both monensin and narasin completely inhibited B. succinogenes at the .5 ug/ml level while 2.5 ug/ml of lasalocid was required for similar results. Narasin-actaplanin.combination.did not change the growth rate of B. succinogenes. As for B. fibrisolvens, monensin and lasalocid inhibited growth at .5 47 48 TABLE 6. MINIMAL INHIBITORY CONCENTRATIONS FOR RUMINAL BACTERIA GRAM MON- LASA- COMBIN- BACTERIA STAIN ENSIN LOCID NARASIN ATION S. rumintium - NI N1 N1 NI M. elsdenii - NI NI NI NI B. ruminicola - NI NI NI NI B. succinogenes - ILS) 112.5) 1L5) NI B. fibrisolvens - I(.5) I(.5) I(2.5) I(S) R.flavefaciens + I(2.5) I(.5) I(20) I(2.5) IL albus + I(2.5) I(.S) I(2.5) I(S) S. bovis + I(.5) I(.5) 1(5) I(5) Calculated from growth curves (OD vs. time). NI means not inhibited. I means inhibited. 49 ug/ml while narasin and the combination caused growth inhibition at the 2.5 and 5.0 ug/ml respectively, These concentrations were 5 and 10 fold higher than the effective concentrations for monensin and lasalocid. The gram positive organisms examined, contrary to many of the gram negative organisms, showed sensitivity to the antibacterial agents. R. flavefaciens demonstrated sensitivity to lasalocid at .5 ug/ml while monensin and the combination caused inhibition at a concentration five times (2.5 ug/ml) that of lasalocid. Narasin was relativily ineffective at low concentrations but caused complete growth inhibition of R. flavefaciens at 20 ug/ml. R. albus was susceptible to lasalocid at .5 ug/ml concentration while monensin and narasin concentrations had to be elevated to 2.5 ug/ml before growth inhibition occurred. Narasin- actaplanin combination, the least effective against R. albus, inhibited its growth at 5 ug/ml. Sensitivity of S. bovis to monensin and lasalocid was observed at .5 ug/ml and complete growth inhibition was detected at a 10 fold higher concentration when narasin and the combination were present. In summary under these experimental conditions the gram negative species, except for B. succinogenes and B. fibrisolvens, were resistant to the antibacterial agents while the gram positive species showed a level dependent sensitivity to the compounds. It appeared that within the sensitive species monensin and lasalocid were generally more - 50 effective inhibitors of growth than either narasin or the narasin-actaplanin combination. In Vivo Experiment Volatile Fatty Acid Concentrations Ruminal volatile fatty acid concentrations (mM) were measured to examine feed additive effects on ruminal fermentation end-products. Ruminal acetate concentrations (mM) for the four treatments with time after feeding are presented in Table 7. For all treatments only the narasin-actaplanin combination (combination) significantly differed (P<.05) from the other treatments. At all sampling times the acetate concentration in the combination fed animals was significantly lower (Pk.05) than in the other treatment groups. The basic pattern for acetate concentration as well as the other volatile fatty acids, was a cyclic pattern based upon the feeding regime. Initially, the concentration would start at a prefeeding level, increase and plateau at a peak level and then decline to the prefeeding level just prior to refeeding. In general across the four treatments the acetate concentration followed this cyclic pattern. For both the control and actaplanin diets acetate concentration significantly increased (P<.05) between the prefeeding and two hours postfeeding concentration. This upper concentration was maintained through the 7 hour sample and 51 TABLE 7. RUMINAL ACETATE CONCENTRATIONS ACROSS TREATMENT WITH TIME AFTER FEEDING TIME T0 T2 T4 T7 T10 TREATMENT ------------------- mM ----------------------- Control 52.3 65.4 60.4 64.7 56.9 Monensin 52.3 65.2 65.1 58.2 54.7 Actaplanin 44.7 55.7 55.4 56.3 45.0 Combination 38.9 50.7 51.4 45.3 41.0 SEM 3.0 Means of four animal. Overall treatment means differ (P<.05). Row means with different superscripts differ (P<.05). 52 by 10 hours after feeding the acetate concentration had significantly declined (P<.05) to the prefeeding value. The monensin and combination treatments followed a similar pattern except the plateau stretched only until the four hour sample. By the seven hour sampling period a significant (P<.05) decrease in ruminal acetate concentration was observed with both the monensin and combination treatments. Propionate concentrations (mM) are represented in Table 8. The combination had a significantly higher (P<.05) ruminal propionate concentration across all sampling times in comparison to the other treatments. The general trend within treatments was similar to that observed with ruminal acetate concentration. For the control, monensin and actaplanin treated animals the prepionate concentration significantly increased (P<.05) between the prefeeding and two hours postfeeding values. This increase was followed by a gradual, although nonsignificant, decline over the next ten hour period. As for the combination treated animals the ruminal propionate concentration significantly increased (P<.05) initially, plateaued and declined significantly (P<.05) between seven and ten hours after feeding. The four feed additives contained in this experiment did not significantly alter the butyric acid concentrations (Table 9L.The treatment means, therefore, were pooled and examined with respect to changes in ruminal butyrate concentrations over the feeding cycle. Ruminal butyrate mean concentrations significantly changed over time (P<.05) 53 TABLE 8. RUMINAL PROPIONATE CONCENTRATIONS ACROSS TREATMENT WITH TIME AFTER FEEDING. TIME T0 T2 T4 T7 T10 TREATMENT ------------------- mM ----------------------- Control 18.0 24.4 21.6 22.0 18.8 Monensin 18.3 24.1 23.4 21.6 19.5 Actaplanin 18.9 25.0 24.7 24.5 22.0 Combination. 28.4 34.0 36.1 34.9 29.5 SEM 1.4 Means of four animal. Overall treatment mean differs from other treatment (P<.05). Row meanswithdifferent superscriptsdiffer (P<.05). TABLE 9. OVERALL TREATMENT MEANS FOR BUTYRATE CONCENTRATION ACROSS TREATMENT TREATMENT ------------------- m4 ....................... _ Control 12.0 Monensin 11.9 Actaplanin 11.8 Combination 8.4 SEM 1.3 Means for four animals 55 (Table 10). The butyric acid concentration increased significantly (P<.05) between prefeeding and two hours postfeeding, then remained constant for the next five hours at which point it significantly declined (P<.05) to the prefeeding level by ten hours after feeding. Protease Activity of Adherent and Free Floating Bacteria The adherent and free floating bacteria were examined to ascertain antibacterial treatment effects and feeding effects on bacterial protease activity. Adherent bacterial protease activity exhibited no significant treatment effects (Table 11). The antibacterial compounds were ineffective in altering proteolytic enzyme activity of the adherent bacterial population. Treatment means were pooled to examine time after feeding effects on protease activity (Table 12%. Significant time effects were not evident under these experimental conditions, as protease activity (u units of protease activity/ mg protein) remained constant over the feeding cycle. Although there seemed to be a slight numerical decline in proteolyic activity of the adherent bacteria at the four hour sample, this fluctuation was not significant. Similarly, significant treatment differences in protease activity were not observered in the free floating bacterial fraction (Table 13),thus, the pooled treatment means were used as before to probe for significant time 56 TABLE 10. POOLED TREATMENT MEANS FOR BUTYRATE CONCENTRATION WITH TIME AFTER FEEDING TIME ------------------- mu ........................ 0 8.8 2 12.5 4 12.5 7 11.6 10 9.7 SEM .6 Means for four animals. Means with different superscripts differ (P