This item is only available as the following downloads:
1 EFFECTS OF BACTERIAL INOCULANTS AND MYCOTOXIN ADSORBENTS TO IMPROVE THE QUALITY AND SAFETY OF FEED INGREDIENTS By OSCAR CEZAR MULLER QUEIROZ A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011
2 2011 Oscar Cezar Muller Queiroz
3 To my parents, Ana Maria Muller and Marivaldo Alves Queiroz, for all the support and love they gave me especially in the last few years when I was far away from home
4 ACKNOWLEDGMENTS I would like to thank my advisor, Dr. Adegbola Adesogan, for his supervision and help during my Ph.D. program. I was always very impressed with his professionalism and I will strive to follow his example in the future. I would also like to thank Dr. Charles Staples for being always ready to help me. I really enjoyed the opportunities I had to work with him. I would like to acknowledge how thankful I am to Dr. Linda Young, who despite her position as Chair of her department and her busy agenda always found time to sit and explain statistical concepts and strategies to Dr. Adesogan and me. I would also like to thank Dr. Williams for being a great professor She is the kindest professor I have met. I am also grateful to Dr. Jose Santos for his efforts, support, and constructive criticism of my research. I acknowledge my co workers (Sam Churl Kim, Jamie Foster, Kathy Arriola, Juan Romero, Miguel Urbano Zarate, Max Huisden, Jan Kivipelto, Lucas Giordano, Junghun Han, Joao Daniel, Maria Fernanda, and Andre Pedroso) for their help during my laboratory and field activities. My most sincere thanks to all the good friends whom I met at the University of Florida, including all members of the Brazilian mafia in the Animal Sciences Department including Lincoln Zotarelli, Flavio Avila, Jose Francisco Figueiredo and family, Sylvia Moraes, and Davi Araujo as well as my Peruvian friend, Miriam Garcia. I would like to acknowledge two instit utions that had an important role in my development as a person and as a professional. The first one is the Escola Superior de Agricultura Luiz de Queiroz, Piracicaba, Brazil, where I learned to love science and the second is Rep. K labouco, Piracicaba, Brazil, where I learned to work for the collective improvement of humanity.
5 I would like to express my deepest gratitude to my parents: Marivaldo and Ana, to my sisters: Helena and Elisa, and to the rest of my family, especially my dear grandma Helena for their support and guidance through the years. Finally, I would like to thank my girlfriend Belen, for being there for me and loving me even during the long challenging times of writing this dissertation. She personifies the best attributes that I have ev er imagined in people. Thanks to God for giving me the opportunity to live and to enjoy this unique experience.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 9 LIST OF FIGURES ........................................................................................................ 10 ABSTRACT ................................................................................................................... 11 CHAPTER 1 INTRODUCTION .................................................................................................... 14 2 LITERATURE REVIEW .......................................................................................... 18 The Importance of Silage ........................................................................................ 18 Ens iling Phases ...................................................................................................... 19 Initial Aerobic Phase ......................................................................................... 19 Main Active Fermentation Phase ...................................................................... 20 Stable Fermentation Phase .............................................................................. 20 The Feed out Phase ......................................................................................... 21 Silage Microbiology ................................................................................................. 22 Primary Fermentation ....................................................................................... 22 Secondary Fermentation .................................................................................. 24 Enterobacteria ............................................................................................ 24 Clostridia .................................................................................................... 24 Yeasts ........................................................................................................ 25 Silage Additives ...................................................................................................... 2 6 Bacterial Additives ............................................................................................ 26 Homolactic bacteria ................................................................................... 27 Heterolactic bacteria .................................................................................. 32 Chemical Additives ........................................................................................... 38 Fermentation inhibitors .............................................................................. 38 Enhancers of aerobic stability .................................................................... 40 Mode of Action of Bacterial Inoculants .................................................................... 42 Acid Induced Inhibition ..................................................................................... 42 Antifungal Inhibition .......................................................................................... 42 BacteriocinMediated Inhibition ........................................................................ 44 Pathogenic Agents in Silage ................................................................................... 45 B acillus ............................................................................................................. 45 Listeria .............................................................................................................. 46 Clostridia .......................................................................................................... 48 Enterobacteria .................................................................................................. 50 Importance of Mycotoxins in Silage ........................................................................ 51
7 Effects of Specific Mycotoxins .......................................................................... 53 D eoxynivalenol ........................................................................................... 53 Fumonisin .................................................................................................. 54 Zearalenone ............................................................................................... 55 Aflatoxin ..................................................................................................... 56 Mycotoxin Detoxification Methods .......................................................................... 58 Enterosorbents ................................................................................................. 58 3 CONTROL OF E. COLI O157:H7 IN CORN SILAGE WITH OR WITHOUT VARIOUS INOCULANTS: EFFICACY AND MODE OF ACTION ........................... 62 Introduction ............................................................................................................. 62 Materials and Met hods ............................................................................................ 63 Harvesting, Inoculation and Ensiling ................................................................ 63 Aerobic Stability ................................................................................................ 65 Survival of E. coli 0157:H7 in Silages Exposed to Aerobic Conditions ............. 65 Laboratory Analyses ......................................................................................... 66 Preparation of silage extrac ts ..................................................................... 66 Microbial enumeration ................................................................................ 66 Chemical analysis ...................................................................................... 67 Antibacterial activity ................................................................................... 67 Statistical Analysis ............................................................................................ 67 Results .................................................................................................................... 68 Anaerobic Phase .............................................................................................. 68 Aerobic Phase .................................................................................................. 68 Antibacterial Activity ......................................................................................... 69 Discussion .............................................................................................................. 69 Conclusions ............................................................................................................ 72 4 EFFECT OF A DUAL PURPOSE INOCULANT ON THE QUALITY AND NUTRIENT LOSSES FROM CORN SILAGE PRODUCED IN FARM SCALE SILOS ..................................................................................................................... 78 Introduction ............................................................................................................. 78 Materials and Methods ............................................................................................ 79 Silage Production ............................................................................................. 79 Laboratory Analysis .......................................................................................... 80 Statistical Analysis ............................................................................................ 82 Results and Di scussion ........................................................................................... 83 Conclusions ............................................................................................................ 88 5 RELATIONSHIP BETWEEN RUST INFESTATION AND THE QUALITY AND SAFETY OF CORN SILAGE TREATED WITH OR WIT HOUT A BACTERIAL INOCULANT ........................................................................................................... 97 Introduction ............................................................................................................. 97 Materials and Methods ............................................................................................ 98
8 Silage, Treatments and Design ........................................................................ 98 Laboratory Analysis ........................................................................................ 100 Statistical Analysis .......................................................................................... 101 Results and Discussion ......................................................................................... 102 Conclusions .......................................................................................................... 107 6 EFFECT OF ADDING A MYCOTOXIN SEQUESTERING AGENT ON MILK AF LATOXIN M1 CONCENTRATION AND THE PERFORMANCE AND IMMUNE RESPONSE OF DAIRY CATTLE FED AN AFLATOXIN B1 CONTAMINATED DIET ........................................................................................ 115 Introduction ........................................................................................................... 115 Materials and Methods .......................................................................................... 116 Cows, Treatments and Design ....................................................................... 116 Analytical Procedures ..................................................................................... 117 Statistical Analysis .......................................................................................... 119 Results and Discussion ......................................................................................... 120 Conclusions .......................................................................................................... 126 7 SUMMARY, GENERAL CONCLUSIONS AND RECOMENDATIONS ................. 132 LIST OF REFERENCES ............................................................................................. 139 BIOGRAPHICAL SKETCH .......................................................................................... 158
9 LIST OF TABLES Table page 3 1 Dry matter, organic acids, and yeast and mold values of corn forage inoculated with Escherichia coli O157:H 7 (EC) alone or EC and commercial bacterial inoculants and ensiled for 82 days ....................................................... 74 4 1 Effect of inoculant treatment on the quantity and chemical composition of good silage removed daily from silos ................................................................ 90 4 2 Effect of inoculant treatment on the quantity and chemical composition of spoiled silage removed daily from silos and the associated nutrient and energy losses ..................................................................................................... 91 4 3 Effect of inoculant treatment on fermentation indices and temperature during ensiling in good corn silages ............................................................................. 92 4 4 Effect of inoculant t reatment on fungal counts, aerobic stability and temperature of good corn silage ....................................................................... 93 5 1 Chemical composition of untreated and inoculated corn forage with different levels of rust infestation .................................................................................... 109 5 2 The relationship between the severity of rust infestation and measures of the quality of the fermentation, nutritive value, safety, and shelf life of corn silage 110 5 3 Effect of rust infestation and inoculant application on chemical composition of corn silage ........................................................................................................ 111 5 4 Effect of rust infestation and inoculant application on fermentative parameters, microbial counts, aerobic stability and mycotoxins in corn silage 112 5 5 Effect of rust severity and inoculant application on mycotoxin concentration of corn silage ........................................................................................................ 113 6 1 Ingredient and chemical composition of the experimental diet ......................... 127 6 2 Effect of dietary addition of aflatoxin B1 with or without low or high doses of a mycotoxin binder on the performance of dairy cows ......................................... 128 6 3 Effect of dietary addition of aflatoxin B1 with or without low or high doses of a mycoto xin binder on the aflatoxin M1 concentration in the milk ......................... 129 6 4 Effect of dietary addition of aflatoxin B1 with or without low or high doses of a mycotoxin binder on markers of the innate imm une response .......................... 130
10 LIST OF FIGURES Figure page 2 1 Homofermentative and heterofermentative pathways for fermentation of glucose to lactate in silage ................................................................................. 61 3 1 Changes in pH in corn forage inoculated with 5 105 log cfu/g of E. coli O157:H7 ( EC) or EC plus bacterial inoculants and ensiled for different durations ............................................................................................................. 75 3 2 Effect of inoculation with 5 105 cfu/g of E. coli O157:H7 (EC) or EC and bacterial inoculants at ensiling on aerobic stability of corn silages ensiled for 82 d ays ............................................................................................................... 76 3 3 Effect of reinoculation of corn silages with 1 106 cfu/g of E. coli O157:H7 (EC) 144 h ours after silo opening (day 82) on pH and E. coli counts (EC; log cfu/g) of silages treated with EC or EC and bacterial inoculants at ensiling ........ 77 4 1 Changes in dry matter concentration of spoiled silage with time ........................ 94 4 2 Changes in lactic acid concentration of corn silage wi th time ............................. 95 4 3 Changes in acetic acid concentration of corn silage with time ............................ 96 5 1 Corn plants affected by different severities of Southern Rust. .......................... 114 6 1 Effect of dietary addition of aflatoxin B1 with or without low or high doses of a mycotoxin binder on clearance of aflatoxin M1 from the milk of dairy cows ...... 131
11 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFECTS OF BACTERIAL INOCULANTS AND MYCO TOXIN ADSORBENTS TO IMPROVE THE QUALITY AND SAFETY OF FEED INGREDIENTS By Oscar Cezar Muller Queiroz August 2011 Chair: Adesogan T. Adegbola Major: Animal Sciences Ensiling is an important preservation method that allows storage of forages harvested at maturity stages that optimize their productivity and nutritive value for use at times when forages are dormant or unavailable. Consequently, silage is one of the major components of the diet of ruminan t livestock in many countries. However, reductions in f orage nutritive value and safety often occur during ensiling. To study methods to minimize such problems, three experiments were conducted to evaluate the effects of additives on the fermentation, shelf life, microbiology, pathogenicity, and toxicity of co rn silage. Because of the occasional presence of mycotoxins in silage and challenges with existing detoxification methods, a fourth experiment was conducted to evaluate the effect of dietary addition of a mycotoxinsequestering agent on measures of health, immune response, and performance of dairy cows fed a diet contaminated with a mycotoxin. The objective of Experiment 1 was to evaluate the effect of applying different bacterial inoculants containing homofermentative, heterofermentative, or both types of bacteria on the survival of pathogenic E. coli O157:H7 in corn silage during the anaerobic fermentation and aerobic exposure stages of silage production. Escherichia
12 coli O157:H7 did not survive after the pH dropped below 4.0 during ensiling. Unlike others, only inoculants containing Lactobacillus buchneri increased the concentration of acetic acid. Because of the antifungal properties of acetic acid, such inoculants prevented the growth of yeasts and molds, kept the pH below 4 after ensiling, and thereby prevented the growth of the pathogen after ensiling. Experiment 2 aimed to determine effects of applying an inoculant containing homofermentative and heterofermentative bacteria on the fermentation, nutritive value, aerobic stability, and nutrient losses from corn silage produced in farm scale silos. Applying the inoculant made the fermentation more heterolactic, by increasing acetic acid concentrations, tended to inhibit the growth of yeasts, and substantially reduced the amount of spoilage and the assoc iated energy and nutrient losses. The objective of Experiment 3 was to examine the association between increasing severity of Southern rust infestation and fermentation, nutritive value, and safety of corn silage, and to determine if inoculant application could mitigate adverse effects of rust infestation on silage quality and safety. The nutritive value and fermentation of corn silage were reduced drastically as the severity of rust infestation increased. Uninoculated corn silage with the highest level of rust infestation had 200 times more aflatoxin than the Food and Drug Admi nistration Action (FDA) Level. Application of the bacterial inoculant improved the nutritive value, fermentation, and shelf life of rust infested corn silage and prevented accumulation of aflatoxin in the silage. The objectives of Experiment 4 were to determine the effect of feeding two doses of a mycotoxin adsorbent on milk aflatoxin M1 concentrations and performance and immune response of dairy cows fed a diet contaminated with aflat oxin B1. Feeding the
13 high dose of the adsorbent kept the aflatoxin concentration of the milk below the FDA action level, whereas feeding low dose of the adsorbent or the toxincontaminated diet alone did not. Both doses of the adsorbent prevented the heightened inflammatory stress response and decrease in milk production and quality caused by feeding the toxin without the adsorbent.
14 CHAPTER 1 INTRODUCTION Silage is defined as the material produced by the controlled fermentation of a crop of high moisture content (McDonald, 1991). The main objective of producing silage is to harvest crops at maturities that optimize their quality and productivity and preserve them to feed animals during winter or dry seasons (Pahlow et al., 2003) or for year round feeding o f housed ruminant livestock. This method of forage conservation has been adopted in many countries and has replaced traditional methods such as haymaking. In the US, silage is fed year round to most dairy cattle because the variability in quantity and qual ity of pastures, climatic challenges, difficulties of milking cows on pasture, and large herd sizes make it more challenging to raise dairy cows on pastures. The area of corn planted for silage in the US in 2010 was 13.7 million ha, which is six times greater than the estimated planted area in 1994 (Wilkinson and Bolsen ,1996; USDA, 2011). Quality silage production is dependent on specific fermentation pathways, without which growth and prevalence of spoilage and pathogenic bacteria and toxigenic fungi oc cur Fenlon and Wilson (2000) reported an increase in the population of Escherichia coli O157:H7 a Shiga toxinproducing, acid tolerant bacteria, in poorly fermented ryegrass silage. This bacterium is a major food safety concern and contamination of food by this bacterium has led to several health scares, deaths extensive recalls of various food items, and culminated in closure of major food companies in the US and around the world (Mead and Griffin, 1998; Rangel et al., 2005). Another food safety concern with poorly made silage is the production of mycotoxins. Gonzalez Pereyra et al. (2008) noted that poorly prepared corn silage samples were contaminated with a high concentration (60 g/kg DM) of aflatoxin. This toxin is a mutagenic, carcinogenic,
15 and tox ic metabolite produced by Aspergillus species and it can be transferred from the diet of dairy cows to their milk and ultimately to consumers (Creppy, 2002; CAST 2003). Aflatoxin is responsible for causing 25,000 to 155,000 new cases of cancer around the w orld and is estimated to have cost the US economy approximately $350 million in 2010 (Wu, 2010). Poor fermentation and aerobic deterioration are the main sources of silage spoilage, nutrient losses from silage (McDonald, 1991; Tabacco et al., 2011), and si lage pathogenicity. Feeding diets containing increasing levels of spoiled silage has linearly reduced feed intake and animal performance, and predisposed cows to various health complications (Whitlock et al., 2000; Whitlow and Hagler, 2005). Because of th e widespread need for and use of silage in US dairy farms, research in the last few decades has focused on development of technologies to improve silage quality but little attention has been focused on food or feed safety aspects of silage production. Vari ous scientific advances have been made in the use of additives and inoculants to improve silage fermentation and shelf life (Kung et al., 2003; Huisden et al., 2009). Homolactic bacteria inoculants, which synthesize lactic acid as the sole product of hexos e fermentation, have been used successfully to rapidly increase the acidity in silages to values ( < 4) that reduce the growth of fermentationimpairing bacteria such as Enterobacteria and Clostridia (Pahlow et al., 2003; Filya et al., 2007). Heterolactic bacterial inoculants have been developed t o ferment hexose and lactate to antifungal acids like acetic acid, which inhibit the growth of spoilage and toxigenic fungi and hence increase the shelf life of forages (Huisden et al., 2009). However, silage qualit y improvement efforts have focused almost exclusively on improving the fermenta tion and shelf life of silage. Consequently, no proven methods of destroying
16 silage pathogens or detoxifying silage mycotoxins by applying additives at the time of ensiling exis t. Rather, the main strategy for preventing adverse effects of ingesting mycotoxin contaminated silage or feed ingredients involves dietary addition of mycotoxin sequestering agents. The interaction of the toxin and binder result s in a macromolecular compl ex that cannot be absorbed by the animal (Phillips, 1999). Mycotoxin sequestering agents do not detoxify because they do not change the concentration of the toxin in the diet; however, because they prevent absorption of the toxin, they decrease its adverse effects on the animal and its transfe r to milk (Diaz et al., 2004). The main objectives of this dissertation w ere to examine potential effects of different inoculants on the fermentation, aerobic stability, nutrient and dry matter (DM) losses of silage, a s well as effects on the pathogenicity and toxicity of silage. An additional objective was to evaluate the efficacy of using different doses of a mycotoxin sequestering agent on the performance, immune response, and milk aflatoxin concentration of dairy cows fed a diet contaminated with aflatoxin. In order to achieve these objectives four experiments were performed. Experiment 1 investigated if different bacterial inoculants containing homofermentative bacteria, heterofermentative bacteria, or both types of bacteria could curtail the growth of pathogenic E. coli O157:H7 in corn silage during the anaerobic and aerobic stages of silage production. A second objective of this experiment was to examine whether pH independent antibacterial activity against E. coli O157:H7 existed in the inoculants and persisted in inoculated silages. Experiment 2 examined the effects of applying a dual purpose inoculant containing homolactic and heterolactic bacteria selected based on the results of Experiment 1, on the fermentation, nutritive value, aerobic stability, and nutrient losses from corn silage
17 produced in farm scale silos. Experiment 3 examined the relationship between increasing severity of Southern rust infestation and the quality and safety of corn silage, and whether the dual purpose inoculant used in Experiment 2 could mitigate adverse effects of Southern rust on the quality and safety of corn silage. Experiment 4 determined the effect of adding two doses of a mycotoxin adsorbent on milk aflatoxin M1 (AFM1) concentrations and the performance and immune response of dairy cows fed a diet contaminated with aflatoxin B1 (AFB1).
18 CHAPTER 2 LITERATURE REVIEW The Importance of Silage Silage is defined as the material produced by controlled fermentation of a crop of high moi sture content McDonald (1991). The main rationale for preserving forages is to be able to harvest and store them at a growth stage that optimizes their productivity and nutritive value and use them at periods when forages experience little or no growth, ve ry often during the winter. Ensilage, the technique of making silage, has been practiced for over 3000 years; however, interest in this process only became widespread in the latter part of ninete enth century (McDonald, 1991). The importance of this practic e is evident from its increasing adoption by producers. In western European countries, silage has been the main forage preservation method in the last few decades, far exceeding the production of hay (Wilkinson et al., 1996). The area of corn planted for s ilage in the US in 2010 was 13.7 million ha, which is six times greater than that in 1994 (Wilkinson and Bolsen, 1996; USDA, 2011). Countries such as Brazil, Argentina, Chile, Australia, Mexico, and New Zealand experienced a rapid increase in silage produc tion in the 1990s (Wilkinson et al., 2003) mainly because of improvement s in ensiling technologies (Kaiser and Evans, 1997; Muhlbach, 1998; ). These technologies have facilitated adherence to the basic concepts of silage making, which are 1) to obtain the anaerobic conditions necessary to promote rapid fermentation and, consequently enhance the rate of acidification of silage and thereby impair the growth of undesirable microorganisms; 2) to densely pack the forage into silos immediately after harvest and r apidly achieve and maintain anaerobic conditions for the ensiling duration, and 3) to manage the silage during the aerobic feedout phase in a way that minimizes spoilage,
19 nutrient depletion, and mycotoxin production. To understand these concepts and how n ew technologies can improve the quality of silage, the process of ensiling will be discussed under four main chronological phases: aerobic, active fermentation, stable fermentation, and feed out phases (Weinberg and Muck, 1996). Ensiling Phases Initial Aer obic Phase This phase is characterized by use of the oxygen trapped within the forage mass and is frequently accompanied by an increase in temperature of the ensiled material. Ongoing respiration by harvested plants and microorganisms results in nutrient o xidation and heat generation during this phase. Obligate and facultative aerobic microorganisms continue to utilize hexose sugars and the presence of oxygen sustains the growth of molds, yeasts and undesirable bacteria that may later produce toxins or predispose the silage to the growth of pathogens at feed out. Plant enzymes, proteases, and carbohydrases initiate the decomposition of proteins and carbohydrates to amino acids and soluble sugars respectively (Pahlow et al., 2003). Oxygen also delays the dec rease in pH necessary for ensiling by reducing the rate of plasmolysis of plant cells (Greenhill, 1964), which releases the intracellular fluid needed by lactic acid bacteria (LAB) for synthesis of lactic acid. All these processes negatively affect silage quality and this initial phase can last for hours in well compacted silage or days if oxygen is present within the silage matrix. Thus, optimizing silage density at silo filling and immediately sealing silos is necessary to expel residual oxygen (McDonald, 1991) and prevent further oxygen ingress into the silage at ensiling.
20 Main Active Fermentation Phase Th e fermentation phase begins when the oxygen in the silage mass is greatly reduced after the initial aerobic phase and it is characterized by the fermentation of water soluble carbohydrates (WSC) by bacteria particularly LAB to acids that decrease the pH. Obligate and facultative anaerobic LAB as well as Clostridia, Enterobacteria, Bacilli and yeasts compete for fermentation substrates like WSC, but in most cases, LAB prevail and become dominant whereas the population of other microorganisms tends to decrease (Pahlow et al., 2003). However, if the pH decreases slowly, Enterobacteria can outcompete LAB (Lin, 1992; Kizilsimsek et al., 2007). Th e fermentation phase can last for several days or weeks depending on the crop used (McDonald, 1991). Pedroso et al. (2005) reported active fermentation in sugarcane silage during the first 15 days (d) of fermentation causing the WSC concentration and pH to decrease by 14% (DM basis) and 1.55 units, respectively. Once active fermentation was over on d 15 of ensiling, no further decreases in pH or WSC were detected during the 165d ensiling period. Common external signs during the active fermentation phase are the produc tion of effluent, formation of gases, and shrinkage of the silage mass (Pahlow et al., 2003) because of increased cell plasmolysis, formation of CO2, and dry matter losses, respectively. Stable Fermentation Phase During this phase, the low pH and fermentat ion product accumulation reduce the activity of most microorganisms in the silage. Even the population of LAB undergoes a threefold logarithmic reduction (Pahlow et al., 2003). Nevertheless, acidtolerant enzymes continue to caus e a slow and continuous release of WSC by hydrolyzing polysaccharides as long as anaerobic conditions are maintained. Kleinschmit and Kung
21 (2006a) noted that the concentration of WSC in corn forage ensiled for an extended period (361 d) fluctuated with time. Silages treated with LA B achieved the lowest concentration of WSC after 28 d of fermentation; however, by d 361, the concentration was greater. This is probably caused by hydrolysis of complex carbohydrates by acidtolerant enzymes. The authors also reported that treatment with Lactobacillus buchneri slightly increased concentrations of acetate (1% DM) and 1, 2propanediol (0.4% DM) from d 56 to 361 of fermentation. This low rate of synthesis of fermentation products is characteristic of microorganisms surviving the stable phase because the prevailing conditions hinder optimization of their metabolic processes (Oude Elferink et al., 2001). Acid tolerant yeast species also can become less active and survive this phase as can Clostridia and Bacilli, which can bec o me dormant endospor es (Pahlow et al., 2003). The Feedout Phase The feed out phase begins when the silo is opened and the silage is exposed to air. Once oxygen penetrates the silage mass, dormant facultative aerobic organisms within the silage grow and multiply, as do airbor ne microorganisms that penetrate the silage. This phase can be divided into two main stages (Oude Elferink., 2011). The first one is characterized by metabolism of organic acids by yeast or bacteria (McDonal d, 1991). Ohyama et al. (1971) isolated acidtole rant yeasts of the genera Candida, Hansenula and Pichia which are involved in initiating the aerobic deterioration of corn and grass silages and can use organic acids as a substrate. Oxidation of acids by these fungi increases the pH, making the silage more conducive for the growth of less acidtolerant aerobic microorganisms that characterize the second stage of aerobic deterioration. The succession of microorganisms in these phases is well represented by
22 two thermal peaks on aerobic stability curves. The first one is by acidtolerant yeasts or bacteria and the second by later growth of molds (McDonald, 1991). In addition to acidtolerant yeasts, acetic acid bacteria (AAB) can also use lactic and acetic acids as a substrate. The complete oxidation of organic acids by these bacteria was studied by Spoelstra et al. (1988) who reported that AAB also are capable of causing the first stage of aerobic deterioration in corn silage. Aerobic deterioration is a major cause of nutrient loss in silage and is also often responsible for silage potential pathogenicity and toxicity, which can lead to poor animal performance, diseases and death (Whitlow and Hagler, 2005; Gonzalez Pereyra et al., 2008). Consequently, concerted efforts have been directed at minimizing aerobic deterioration in the last few decades and these have culminated in development of a newer group of silage inoculants. Silage Microbiology Fermentation is the process responsible for the preservation of forage during ensiling. To describe the associated mi crobiology, the process is divided into primary and secondary fermentation (Pahlow et al., 2003). Primary Fermentation Primary fermentation is caused by LAB, which acidify the silage mainly by synthesizing lactic acid, and thereby inhibiting the growth o f undesirable microorganisms that cause secondary fermentation. Several bacteria are classified as LAB but those from genera Lactobacillus, Pediococcus, Leuconostoc, Enterococcus, Lactococcus, and Streptoco cc us are found most frequently in silages. The LAB are obligate fermenters and they generally but not always use carbohydrates as the source of energy (Oude Elferink et al., 2001; Pahlow et al., 2003). Their presence in the
23 epiphytic microbial population of crops is probably because of their defensive rol e against reactive oxygen species ; for instance, their high manganese concentration works as an intracellular oxygen scavenger (Archibald and Fridovich, 1981). Because they are fermenters, LAB perform well under anaerobic conditions and oxygen can be lethal for most of those that do not secrete catalase (Codon, 1987). Most LAB will grow at temperatures between 25 and 40oC; however some strains can su rvive more extreme conditions. Mulrooney and Kung (2008) reported that as incubation temperatures increased from 30 to 45oC for 6 h the viability of several LAB decreased, however L. plantarum MTD/1 was thermotolerant and its viability was maintained even at 45oC. Under anaerobic conditions, LAB dominate the microbial population and decrease forage pH from 6 an d above to approximately 4 by fermenting a wide variety of substrates and synthesizing lactic acid as well as other products (Pahlow et al., 2003). Because of the fermentative differences within the group, LAB have been classified by the type of fermentation they promote (McDonald, 1991; Pahlow et al., 2003) into homofermentative or heterofermentative types. Homofermentative LAB ferment hexose sugars to pyruvate almost exclusively by the glycolytic pathway followed by reduction of pyruvate to lactic acid (Figure 2 1, part A). However, they cannot ferment pentoses since they do not have phosphoketolase, which is necessary for the synthesis of acetyl phosphate and glyceraldehyde3 phosphate (White, 2007). Facultative heterofermentative LAB utilize the same p athway as homofermentative LAB to ferment hexoses, as they secrete aldolase, which is necessary for the synthesis of 3phosphoglyceraldehyde an d dihydroxy acetone phosphate. However they also secrete
24 phosphoketolase and can therefore ferment pentoses. Obligate heterofermentative LAB are characterized by their fermentation of hexoses to other products in addition to lactic acid (Figure 21, part B). Secondary Fermentation Enterobacteria Enterobacteria species are well known for their capacity to decarbox ylate and deaminate amino acids during secondary fermentation of silage. Those commonly isolated from silage are gram negative, rodshaped, often motile, catalaseexpressing bacteria capable of reducing NO3 (Pahlow et al., 2003) Most Enterobacteria are st rictly dependent on carbohydrates for growth under anaerobic conditions (Pahlow et al ., 2003) and are not pathogenic. However poor fermentation may cause the establishment of E.coli O157:H7 (Fenlon and Wilson, 2000) on silage and increase the chances of fe ed and food contamination with the pathogen. In general, Enterobacteria are very sensitive to climatic changes, which cause fluctuation in the epiphytic population of plants; nevertheless, their minimum population is frequently 100 times higher than that of LAB (Pahlow et al., 2003). At the beginning of the initial aerobic phase, Enterobacteria often multiply to numbers (108 to 1010 cfu/g) that may exceed those required for LAB dominance during the active fermentation phase (Lindgren et al., 1985). This ca pacity to compete with beneficial LAB for nutrients can cause extensive proteolysis and DM losses (McDonald, 1991); therefore, their presence in silage is highly undesirable. Clostridia Clostridia species are important participants in the secondary fermentation in silages and they are as undesirable as Enterobacteria species. Clostridial growth is
25 optimized by conditions, which lead to poor fermentation such as low DM at harvest, limiting WSC concentration, high buffering capacity, and the presence of resi dual oxygen in the silage mass (McDonald, 1991). Clostridial species are gram positive obligate anaerobes capable of forming dormant endospores that survive adverse conditions (Madigan et al. 2009). Clostridial growth is possible at pH of 4.6 to 7.0 and f rom 3.5 to 50oC (Kendall, 2006). Strains frequently isolated from silage include saccharolytic types that usually ferment carbohydrates such as C. tyrobutyricum and C. butyricum and proteolytic types like C. sporogenes (Pahlow et al. 2003). The main negat ive effects of Clostridia on the quality of silage are degradation of amino acids, which increases NH3 concentration and synthesis of butyric acid, in particular as well as acetic, propionic, isobutyric, and 2methyl and 3 methyl butyric acids (McDonald, 1991). Clostridia also may synthesize biogenic amines and CO2 via decarboxylation of amino acids (Hui and Sherkat, 2006). The butyric acid and biogenic amines typically decrease silage intake (Dulphy and Van Os, 1996) Such biogenic amines reduced ruminal function and decreased DM intake by up to 25% in steers fed contaminated alfalfa silage (Phunstok et al., 1998). Yeasts Yeasts are eukaryotic microorganisms that mainly grow as a single cell (Brock, 2009). Yeasts belonging to the Candida and Hansenula genera can initiate aerobic deterioration of silage by using lactate as a substrate during the final aerobic phase, but yeasts also have the capacity to cause secondary fermentation under anaerobic conditions. When present in the epiphytic population of crops with high WSC concentrations such as sugarcane, yeasts ferment hexoses to ethanol primarily via an alcoholic fermentation. Coexistence of yeasts and heterofermentative LAB cause
26 considerable DM losses and result in silages of poor quality with high ethan ol concentrations (Pedroso et al., 200 5 ; Avila et al., 2009). Jonsson and Pahlow (1984) noted that under anaerobic conditions, fermentative, nonlactate assimilating Saccharomyces cerevisiae are predominant, however under aerobic conditions, lactateassimi lating yeasts are dominant. Dominance of the microbial population by LAB can curtail secondary fermentation but the outcome depends on the DM concentration of the silage and the pH or the acidification rate. Silages with low DM concentrations require lower pH to be anaerobically stable (Pahlow et al., 2003). Ensuring a fast acidification of silage is the main mechanism for preventing the growth of the Clostridia and Enterobacteria that cause secondary fermentation in silage. Silage A dditives Three of the m ain silage making goals are preventing secondary fermentation, establishing and maintaining anaerobic conditions during ensiling, and preventing entry of oxygen into the silage mass during the final aerobic phase. Silage additives are often employed to address these issues ; however a wide range of additives are available and they have different effects and modes of action. The ensuing section describes the main additives currently used for silage preservation. Bacterial Additives Bacterial additives or inoculants contain selected strains of bacteria that ferment sugars to either lactic acid, which causes a rapid pH decline and, thereby preserves the forage, or to antifungal acids that inhibit the growth of spoilagecausing fungi. These bacteria are classi fied as homolactic and heterolactic types, respectively. Both types
27 can be used to improve silage quality but they have different roles and act on different phases of the ensiling process. Homola c tic bacteria The first recorded use of homolatic bacteria to preserve plant products occurred at the beginning of the twentieth century when French workers applied cultures of bacteria to preserve sugar beet pulp (Watson and Nash, 1960). Homolactic bacteria ferment glucose into lactic acid via a pathway that is e nergetically efficient because no energy containing byproducts are produced. The transformation of 1 mole of glucose to 2 moles of lactic acid and 2 moles of adenosine triphosphate (ATP) via the EmbdenMeyerhof pathway yields high recoveries of energy (99. 3 % ) and DM (100%; Kung et al., 2003; White, 2007). Lactobacillus p lantarum : The use of homolactic bacteria was common at the end of 1970s (Kung et al., 2003). At that time, most silage inoculants were developed based on the criteria of Whittenbury (1961) who recommended that inoculant bacteria should grow vigorously, be able to dominate the microbial population during fermentation, and be homofermentative and acid tolerant in order to produce significant amounts of lactic acid and rapidly decrease the pH. The microorganism that fit all these criteria was Lactobacillus plantarum which is still the bacteria most commonly found in commercial silage inoculants. Lactobacillus plantarum is a gram positive rod shape d bacterium commonly isolated from fermented fo od and silages (Kung et al 200 3 ). It is tolerant to aerobic conditions due to the high intracellular content of manganese polyphosphate, which works as an oxygen scavenger to lower the amount of reactive oxygen species (Archibald and Fridovich, 1981). The latter survival mechanism of L. plantarum and its
28 physiological and biochemical properties make it a perfect candidate for dominating the fermentation and meeting Whittenburys other criteria for silage inoculants. Lactobacillus plantarum was classified previously as an obligate homofermentative bacterium based on its ability to ferment one mole of glucose to 2 moles of lactic acid via the energetically efficient homofermentative EmbdenMeyerhof pathway. However, L. plantarum is now classified as a facul tative heterofermenter because when glucose is lacking, it shifts the fermentation away from that based on exclusive lactic acid production to one based on fermenting pentoses to lactic acid, CO2, and acetic acid via the less efficient heterofermentative pathway (Holzer et al., 2003). Inadequate glucose supply also reduces the concentration of fructose1, 6 bisphosphate, an essential activator of lactate dehydrogenase (Pahlow et al., 2003). The new classification method is based on phylogenetic comparison o f 16 ribosomal ribonucleic acid (16S rRNA) and is more accurate than the traditional physiological and biochemical system. Nevertheless, L. plantarum is still considered a homolactic bacteria when glucose availability does not limit its fermentative activi ty. Thus, in silages with adequate glucose concentrations, L. plantarum usually synthesizes lactate exclusively, and the lactate reoxidizes NADH, thus allowing the EmbdenMeyerhof pathway to be continuously repeated during the metabolism of carbohydrates ( McDonald, 1991). Lactobacillus plantarum has been used successfully to decrease the pH of silages particularly those with high buffering capacity. Filya et al. (2007) reported that application of three strains of L. plantarum reduced the pH of alfalfa from 5.08 in the Control silage to an average of 4.43 in inoculated silages and increased the lactate to acetate ratio from 2.88 to 6.53. Conaghan et al. (2010) showed that L. plantarum
29 increased lactic acid concentrations in ryegrass silages from 73 (Control) to 95 g/kg of DM and reduced the pH from 4.5 to 4.0. Pediococcus p entosaceus : Pediococcus pentosaceus is a homolatic, gram positive, facultatively anaerobic, nonspore forming bacterium that often is used as a silage inoculant. Like Lactobacillus plantarum it is acid tolerant and is capable of synthesizing L and D forms of lactic acid (Garvie, 1986; Axelsson, 1998). However, it grows more actively than L. plantarum and most other silage bacteria when the pH is 5 to 6.5 and residual oxygen is still present during the early stages of the fermentation (Kung et al., 2003 ; McDonald, 1991). Pediococcus strains also can grow well at high DM concentrations and under low water activity (Tanaka and Ohmomo, 2000). These characteristics allow Pediococcus strains to st art the acidification of silage during the initial aerobic phase at early stages of fermentation when Lactobacillus strains grow less vigorously due to the high pH. Thus, some inoculants contain both P. pentosaceus and L. plantarum in order to exploit thei r complementary growth niches and thereby increase the rate of acidification during the early and later stages of the fermentation (Lin et al., 1992). Cocconcelli et al. (1991) used DNA (deoxyribonucleic acid) probes to verify the colonization of corn sil age by P. pentosaceus and L. plantarum. The authors verified that the population of Pediococcus pentosaceus was maximized after an exponential increase during the first 12 h of ensiling, whereas dominance by Lactobacillus plantarum occurred only after 48 h Cai et al. (1999) inoculated alfalfa and ryegrass silages with isolates of Pediococcus acidilactici or Pediococcus pentosaceus at 25 or 48oC. The authors noted that the quality of silage kept at 25oC was improved by both strains as evidenced by decreases in DM and gas losses and products of
30 secondary fermentation such as NH3 and butyric acid. Similar but less pronounced trends were evident at 48oC, which suggests that P. pentosaceus may be an unsuitable homolactic bacteria for silage produced in subtropic al or tropical areas. Enterococcus faecium : Enterococcus faecium is an important producer of lactic acid during the initial stages of fermentation like P. pentosaceus These bacteria are gram positive facultatively anaerobic cocci that synthesize Llactic acid and can grow at pH 4.5 to 9.6 (Cai, 1999). Cai (1999) isolated 48 strains of lactic acid bacteria from forage crops, from which 4 isolated strains of Enterococcus and two commercial strains of Lactobacillus were applied to alfalfa and grass silage. Si lages inoculated with the commercial additives had lower pH, butyric acid, ammoniaN, gas production, and DM losses compared with Control silages, but those inoculated with the Enterococcus strains did not. The authors concluded that Enterococcus species cannot improve silage fermentation likely because they were unable to grow at pH below 4.5. Nevertheless, like P. pentosaceus, Enterococcus spp. are used in homolactic inoculants containing L. plantarum to 1) dominate the initial fermentation phase and quic kly initiate the pH decrease and thereby prevent the onset of secondary fermentation and 2) to decrease the pH to levels conducive for the growth of L plantarum. Filya et al. (2000) applied Enterococcus faecium and Lactobacillus plantarum to fresh and wil ted wheat silage and reported that a pH of 5 was achieved after 15 and 70 d in Control silages versus 1 and 5 d in inoculated silages. Filya et al. (2007) studied the effect on the fermentation of alfalfa silage of inoculants containing 2 strains of E. fae cium that were applied separately and mixtures of E. faecium and L. plantarum, or E. faecium, L. plantarum, and P. pentosaceus Combination of E. faecium with L. plantarum reduced the pH, increased
31 lactic acid concentration, and decreased ethanolic fermentation, however addition of P. pentosaceus did not increase the effectiveness of the inoculant compared to using E. faecium and L. Plantarum One of the two inoculants containing only E. faecium improved the fermentation but did not reduce ethanol concentr ation likely due to secondary fermentation by Enterobacteria and or yeasts, whereas the other one did not have any effects when compared to the Control treatment. This study shows the complementary effects of L. plantarum and E. faecium on silage fermentation. The effects of inoculants containing homolactic acid bacteria on silages were reviewed by Kung and Muck (1997). Homolactic bacteria inoculants frequently cause a rapid pH decrease by increasing lactic acid concentration and thereby reducing proteolys is, deamination, and the potential for ethanolic, butyric, or acetic acid dominated fermentations, such that secondary fermentation is decreased and DM recovery is increased (Kung et al., 2003). Therefore, one of the main benefits of inoculation with hom ofermentative bacteria is to reduce losses of energy, nutrients, and DM associated with secondary fermentation. However, homolactic bacteria do not reduce the risk of aerobic deterioration of silage. Kung and Muck (1997) reported that homolactic inoculants do not affect and in fact sometimes worsen the aerobic deterioration of inoculated silages. This is largely because lactic acid is not a strong antifungal agent and therefore, it does not usually inhibit the growth of spoilage fungi. In contrast, the propionic and acetic acids produced during heterolactic fermentation (Moon, 1983 ; Huisden et al., 2009) are strong antimycotic agents that inhibit spoilage fungi and thereby reduce silage deterioration.
32 The term aerobic stability is used frequently to express how long silage remains without signs of microbial deterioration under aerobic exposure. During the aerobic exposure phase, acidtolerant yeasts that use lactic acid as a growth substrate indirectly increase the pH to levels that are ideal for molds and o ther spoilage and pathogenic microbes that worsen silage deterioration (Adesogan and Queiroz, 2009; Queiroz et al., 2011). Respiration by such microbes results in rapid metabolism of nutrients and increased DM loss (Henderson et al., 1979; Cai 1999). Poor aerobic stability does not occur only on the silo face but also within the silo (Pitt and Muck, 1993). Air can penetrate up to 4 meters into the silage mass, which implies that silage can begin to deteriorate for days before it is exposed on the silo face and fed (Parsons, 1991). Whittenburys criteria for the ideal bacterial silage inoculant did not account for the inability of lactic acid to reduce fungal growth and aerobic deterioration, perhaps because aerobic deterioration problems were less extensive and common in the 1960s due to the smaller silo sizes that were prevalent. With the large silos in current use, the need to maintain or increase aerobic stability is very critical. Heterolactic inoculants were developed to address this need. Heterolact ic bacteria Heterolactic bacteria produce lactic acid and additional products such as ethanol, CO2, and acetic acid during hexose fermentation (Oude Elferink et al., 2001). This is because these bacteria lack fructose diphosphate aldolase, thus glucose 6p hosphate is fermented to 6phosphogluconate rather than fructose6 phosphate (Kung et al., 2003). Adding heterolactic bacteria instead of homolactic bacteria to silage can increase DM and energy losses. For instance, respective losses of up to 27 and 1.7% were reported for heterolactic fermentation, whereas with homolactic fermentation, no DM
33 loss occurs and only about 0.7% of energy is lost (McDonald, 1991). Nevertheless, the heterofermentative pathway is attractive because it generates antifungal agents such as acetate or propionate (Oude Elferink et al., 2001; Krooneman, 2002), which are powerful antifungal agents that have increased the aerobic stability of corn, sorghum, and ryegrass silages inoculated with heterolactic inoculants (Kung and Ranjit, 2001; Tabacco et al., 201 1; Driehuis et al., 2001; Huisden et al., 2009). The acetate produced by heterofermentative bacteria can also curtail yeast induced ethanolic fermentation by inhibiting fungal growth in forages with high sugar concentrations. Lactobacillus buchneri : Lactobacillus buchneri is the most commonly used heterofermentative bacteria in silage inoculants. Avila et al. (1999) reported that when added to sugarcane silage, L. buchneri increased the acetic acid concentration from 17 g/kg DM in th e Control silage to 43 g/kg of DM, reduced ethanol concentration by 33 g/kg of DM, reduced yeast counts by 4log cfu/g, and increased aerobic stability by 36 hours. Lactobacillus buchneri is a gram positive, rodshaped, nonspore forming and anaerobic bac terium. It is one of the most commonly used LAB in silage inoculants due to its capacity to synthesize acetic acid even under low pH. Oude Elferink et al. (2001) described the pathway used by L. buchneri to convert lactic acid to acetic acid, 1,2propanedi ol, and traces of ethanol under anoxic conditions. They also reported that the conversion of lactate to these products is dependent on environmental conditions, such as pH and temperature. All the L. buchneri strains they evaluated degraded lactate when temperatures were increased 15 to 25oC, however when temperatures were further increased to 30oC, only one strain degraded lactate and at 35oC, none had this effect. The degradation of lactic acid was also dependent upon environmental pH. At pH
34 5.8, the conc entration of lactate was steady for 200 h, while reduction of pH to 4.3 and 3.8 increased the rate of lactate degradation. Lactobacillus buchneri has been used to increase aerobic stability in corn, barley, alfalfa, sorghum, sugarcane, grass and other sil ages (Filya, 2003; Huisden et al., 2009; Pedroso et al., 2005). Kleinschmit and Kung (2006b) performed a metaanalysis with 33 studies to understand the effect of L. buchneri on corn, grass and small grain silages. The authors reported that L. buchneri inc reased acetate concentration, decreased lactate concentration, and consequently reduced yeast counts. However, the latter benefit occurred at the expense of a relatively small increase in DM loss caused by the inefficiency intrinsic to heterolactic ferment ations. Effects of L. buchneri on corn silage were also dose dependent with doses >105 being more effective than < 105. The effect of L. buchneri on grass and small grain silages in the meta analysis was different from that on corn silages. Treatment wit h L. buchneri increased ethanol and propionic acid concentrations of grass and small grain silages and yeast not detected in most of such silages; consequently no relationship existed between yeasts and acetate concentration. All the other effects of L. bu chneri application on corn silage such as improved aerobic stability were also evident in grass and small grain silages. The fact that application of L. buchneri increased aerobic stability in grass and small grain silages without affecting yeast counts is possibly because the acetic acid produced by the inoculant decreased the growth of other spoilage fungi or bacteria after aerobic exposure. That L. buchneri application increases propionic acid in silages is desirable, because propionic acid has stronger antifungal properties than acetic acid. However, L.
35 buchneri treatment has not increased consistently propionic acid concentration. Combining propionic and acetic acids result in a synergistic antifungal effect, which results in increased aerobic stabilit y. Driehuis et al. (1999) reported that corn silage treated with increasing concentrations of L. buchneri had increasing concentrations of acetic, propionic and 1 propanol instead of the 1, 2propanediol often synthesized in other studies. Corn silage inoc ulated with 1x106 cfu/g of L. buchneri contained 113 mmol of propionic acid/kg of DM, which represents a tenfold increase over the value for the Control silage. This increase in propionic acid and the a threefold increase in acetic acid resulted in an aerobic stability of 792 h versus 42 h in Control silages. The authors hypothesized that 1, 2propanediol was being converted to 1propanol and propionic acid by Lactobacillus buchneri or another microorganism. This hypothesis was later confirmed by Krooneman et al. (2002), who isolated two facultative anaerobic, heterofermentative, novel strains named Lactobacillus diolivorans These strains can co exist in corn silage inoculated with L. buchneri and convert 1,2propanediol to 1propanol and the desirable pr opionic acid. The need for coexistence of these bacteria is probably why effects of L. buchneri on propionic acid are inconsistent in the literature. Propionic acid concentrations would only be increased by L. buchneri treatment in locations where L. diol ivorans is part of the natural epiphytic population of crops or if it is added in to the L. buchneri inoculant. Lactobacillus b revis: Lactobacillus brevis is a rod shaped, heterofermentative, gram positive bacterium frequently found in aerobically stable s ilages (Holzer et al., 2003). It is also known for its ability to synthesize acetic acid from fermentation of carbohydrates (Holzer et al., 2003) and thereby increase aerobic stability of silages
36 (Wang and Nishino, 2008). Danner et al. (2003) examined effects of different bacterial strains on the fermentation and aerobic stability of corn silage. The authors reported that silage treated with L. brevis had 28.6 g/kg DM of acetic acid, 73% more than that in the Control silage and 42 h longer aerobic stability than the Control silage. Nevertheless, L. buchneri was more effective than L. brevis at increasing acetic acid concentration (55.3 g/kg of DM) and aerobic stability (274 h). This study and the more frequent and successful use of L. buchneri for increasi ng silage aerobic stability, indicate that L. brevis is less effective for this purpose. Propionibacteria f reudenreichii : Propionibacteria freudenreichii is a gram positive, nonspore forming, oxygen tolerant bacterium found in dairy products and silage (W oolford, 1975). Propionibacteria can synthesize propionic acid, which is a potent antimycotic agent (Moon, 1983). Merry and Davies (1999) reviewed the effects of Propionibacteria on silage and their role in biological control of aerobic spoilage. They con cluded that the evidence that inoculation with P. freudenreichii increases the propionic acid concentration of silages is equivocal. This may be because Propionibacteria are not competitive when the pH decreases quickly in silages with adequate fermentable carbohydrate concentrations and LAB populations such as corn silage. Arriola et al. (2011) reported that adding an inoculant containing P. freudenreichii and P. Pentosaceus did not affect propionic acid concentration, the fungal population, or aerobic stability of corn silages. Similar results were observed by Kung and Ranjit (2001), Taylor et al. (2002) and Pedroso et al. (2010) when they applied P. freudenreichii, L. plantarum, and P. pentosaceus together.
37 Correct use of bacterial inoculants requires an understanding of the specific purposes of different inoculants. As a general rule, homolatic bacteria inoculants are used to improve silage fermentation, whereas heterolactic bacteria inoculants are used to increase aerobic stability. This general con cept was illustrated by the work of Tabacco et al. (2011) who compared effects of applying homolactic L. plantarum or heterolactic L. buchneri to corn and sorghum silages. Silages inoculated with L. buchneri had higher pH, lower lactic acid concentration, greater acetic acid concentration and DM loss (0.8%), and greater aerobic stability (690% and 572% greater, respectively) than those in untreated silages. The DM losses during 14 d of aerobic exposure were 15% for L. buchneri treated corn silage versus 44 and 42% for Control and L. plantarum treated silages. The complementary roles of homolactic and heterolactic bacteria in silage fermentation have led to the development of inoculants containing both types of bacteria in order to improve the fermentati on and aerobic stability of the silage. Such Dual purpose or Combo inoculants have been used successfully to improve the preservation of corn, alfalfa, sorghum, and bermudagrass silage (Filya, 2003; Sch midt et al., 2009; Schmidt and K ung, 2010). Kleins chmit and Kung (2006 a ) reported that corn silages treated with a mixture of L. buchneri (4 x 105 cfu/g) and P. pentosaceus (1 x 105 cfu/g) had lower concentration of NH3, greater concentration of acetate, and greater aerobic stability compared with the unt reated silages after 361 d of fermentation. Filya (2003) showed that corn and sorghum silages treated with a dual purpose inoculant had lower NH3 from d 2 of ensiling until d 90 (silo opening). Inoculation also increased
38 acetate concentration in both silag es and thereby reduced the yeast population and increased aerobic stability. Chemical Additives Chemical additives are applied to silage to totally or partially reduce microbial activity during fermentation or feed out, therefore they can be classified as inhibitors of fermentation or enhancers of aerobic stability. Fermentation inhibitors Sulfuric a cid : Inorganic acids have been used as silage additives since the late 1800s (Watson & Nash, 1960), however their use is declining due to their hazardous and ex tremely corrosive potential (Kung et al., 2003). The inorganic acids most frequently used for forage preservation are sulfuric, hydrochloric, and phosphoric acids, which are very strong acids capable of decreasing silage pH immediately after application. T hese acids do not have any specific antimicrobial activity rather their mode of action is direct acidification (Woolford, 1978). Due to their lack of antimicrobial activity, low to moderate rates of application (<2L/ton of fresh weight) are not effective at eliminating microorganisms tolerant to acidic conditions such as yeasts and coliforms (Chamberlain & Quig, 1987). The latter authors noted that applying 0, 2, 4, and 6 L of sulfuric acid per ton of fresh weight of perennial ryegrass ( Lolium perenne L. ) d ecreased pH, and lactic, acetic, and butyric acids and also increased residual WSC and ethanol concentrations. The increase in ethanol was due to the presence of yeasts capable of surviving pH lower than 3.5. The effect of sulfuric acid and other inorgani c acids on animal performance is inconsistent. O Kiely et al. (19 9 6) reported that DM intake (DMI) and liveweight gain increased in beef cattle fed grass silage treated with 2.3 L/t of sulfuric acid instead of
39 the untreated silage. However, O Kiely (1996) reported that no improvement in weight gain occurred when beef heifers were fed grass silage treated with 2.5 L/t of sulfuric acid. Reasons for such discrepancies are unclear. Formic acid: Formic acid is an organic acid that inhibits fermentation by p romoting acidification of silage and by conferring specific antimicrobial activity. Formic acid has a pronounced inhibitory effect against Clostridia and Enterobacteria, but yeasts are more tolerant to the acid (McDonald et al., 1991). Formic acid is stronger than other organic acids frequently found in silage such as propionic, lactic, acetic, and butyric acids but it promotes a moderate rate of acidification compared with inorganic acids (Kung et al., 2003). Thus, it is common to use mixtures of formic ac id with inorganic acids for forage preservat ion (Kung et al., 2003). Formic acid application is typically associated with decreased production of fermentation products like lactic, acetic, and butyric acids, increased residual WSC concentration, and reduce d ammonia nitrogen concentration (Barry et al., 1978). Waldo et al. (1978) reported that DMI (9.7 vs. 8.6 kg/d) and milk yield (17.7 vs. 17.1 kg/d) increased when dairy cattle were fed formic acid treated ryegrass silage instead of untreated silage. Keady and Murphy (1996) demonstrated that formic acidtreated ryegrass silages had lower ammonia nitrogen concentration (43 vs. 61 g/kg of total N), greater aerobic stability, and lower DM digestibility than the untreated silage. These authors detected no differ ence in DMI or milk yield due to inoculation, however milk protein and fat concentrations were greater in cows fed formic acid treated silage. Formic acid treatment does not frequently increase aerobic stability of silages because yeasts can survive its ac idifying and antimicrobial effects (McDonald et al., 1991).
40 Enhancers of aerobic stability Propionic acid: Propionic acid is an organic acid that is used as a silage additive primarily because of its strong antimycotic effects. Propionic acid is fungicidal or fungistatic but these properties are only exhi bited by the undissociated form which increases proportionally as pH decreases (Lambert and Stratford 1999). Inhibition of the growth of yeasts and molds by this acid concomitantly increases aerobic stability because these organisms are responsible for initiating and promoting silage spoilage, respectively. Kung et al. (1998) applied 0.1 or 0.2% of propionic acid to wholeplant corn and reported that the only effect on the fermentation was an increase in the concentration of the acid, whereas the treatment increased aerobic stability by more than 90 hours. Kung et al. (2004) reported that silages treated with 0.1 or 0.2% of a propionic acidb ased additive had a 10 to 100 fold decrease in yeast counts and i ncreases in aerobic stability of 39 and 57 h, respectively. Combining propionic acid with other acids can result in a synergistic inhibition of the growth of yeasts and molds (Moon, 1983). Jung (1972) showed that a mixture of propionic and formic acids resulted in better preservation of corn silage than the use of propionic acid alone. Silage containing both acids had greater aerobic stability and lower protein degradation than silage containing the individual acids. Kung et al. (1998) reported that a combination of propionic acid and other silage additives such as acetic, benzoic and citric acids, ammonia, and sorbate resulted in increased aerobic stability of corn silage by of 278 h. Benzoic acid : Benzoic acid also enhances aerobic stability by its ant ifungal effect (Woolford, 1975). Pedroso (2003) applied 0.5 to 2.0 g/kg of sodium benzoate to sugarcane, which is well known for its ethanolic fermentation due to high WSC
41 concentrations that predispose to the growth of spoilage and ethanol producing yeast s. Benzoic acid treatment increased in vitro DM digestibility (IVDMD; 45.4 vs. 49.8% DM) and aerobic stability (65 vs. 79h), and decreased ethanol concentration (3.83 vs. 2.52% DM). Pedroso et al. (2006) showed that heifers fed sugarcane silage treated with 0.1% of benzoate had greater feed efficiency than heifers fed untreated silages (7.6 vs. 9.4 kg of DM/kg of live weight). Lattemae and Lingvall (1996) reported that grass silage treated with sodium benzoate had less fungal counts and greater aerobic stabi lity than the untreated silage (Lattemae & Lingvall, 1996). Urea and ammonia : Urea is frequently classified as a nutrient silage additive because it is a source of nitrogen for ruminal fauna, however the main purpose of treating silage with urea is to obtain ammonia, which enhances aerobic stability via its alkalinity. Urea contains 46% nitrogen and it can be converted to ammonia by urease enzymes in silage. The main reasons why producers use urea instead of ammonia for forage preservation are the equipme nt needed, expenses involved, and danger associated with application of ammonia. Once the urea is transformed into ammonia, the caustic and antimycotic effect of ammonia curtails fungal growth because in alkaline environments, uncharged NH3 diffuses throug h fungal membranes and kills the microorganism by metabolic poisoning (Kung et al., 2003). The same authors reported that in general, ammoniatreated silages have a higher pH, lower lactic to acetic ratio, lower residual water soluble carbohydrates, and gr eater nitrogen, aerobic stability and digestibility than untreated silages. The reason why digestibility is increased by ammoniation is that the alkali hydrolyzes linkages between hemicelluloses and lignin in the plant cell wall. However, ammoniation can r educe the rate of pH decrease in silage.
42 This factor coupled with the alkalinity can favor the growth of Enterobacteria and Clostridia that cause secondary fermentation. A further problem is that under high temperatures, ammonia reacts with sugars in forag es to form 4methylimidazole, which causes hyperexcitability in cattle (Kerr et al., 1987). Mode of Action of Bacterial Inoculants Acid Induced Inhibition The main function of bacterial silage inoculants is to produce organic acids, which reduce the pH and or inhibit the growth of spoilage causing fungi. Homofermentative bacteria are primarily involved in the former and are considered stimulators of fermentation whereas, heterofermentative bacteria are involved in the latter and classified as aerobic stabil ity enhancers. Adding homofermentative LAB considerably increases the lacti c acid concentration of silage. Although lactic acid has weak antifungal effects, it is a strong acid that can lead to rapid acidification of the silage because of its low dissociat ion constant (3.08; Madigan et al., 2009). Acidification to pH of 4 or below is recommended in silages because most undesirable microbes are less active at this pH range (Kung et al., 2003). Lindgren et al. (1985) reported that the following pH values rest rict microbial activity: 4.6 for Enterobacteria, 4.2 to 5.0 for Clostridia 4.5 to 5.0 for Bacilli, 5.0 for Streptomyces and 2.0 for yeasts and molds. Therefore, acidification of silage to pH of 4 inhibits all undesirable microorganisms except yeasts and m olds Antifungal Inhibition The ability of yeasts to survive acidic environments allows them to worsen silage fermentation and to initiate aerobic spoilage. This is because these organisms can make the fermentation ethanolic instead of homolactic and they c an metabolize lactic
43 acid, thereby increasing silage pH to the threshold that is conducive for other spoilage and pathogenic microbes (Pedroso et al., 2005; Gonzalez Pereyra et al., 2008). The inability of lactic acid or even strong inorganic acids to prev ent the growth of fungi in silages suggests that mechanisms other than pH reduction should be used to inhibit their growth (Chamberlain and Quig, 1987; Kung et al., 2003). Consequently, heterofermentative bacteria that synthesize antifungal acids like propionic and acetic acid (Moon 1983) are being used increasingly to inhibit spoilage fungi. The antifungal effect of relatively weak acids like propionic and acetic acid is caused mainly by their undissociated form, which can diffuse through the plasma membr ane of fungi into the cytoplasm (Lambert and Stratford, 1999). Once inside the cell, the neutral intracytoplasmic pH causes the acid to dissociate into charged ions. To avoid accumulation of hydrogen ions and cytoplasmic acidification, the cell utilizes a H+ATPase pump to remove the excess ions. Expenditure of energy for this process reduces the energy available for growth and multiplication of the fungi (Lambert and Stratford, 1999). The proportion of free undissociated acid is dependent on the extracellular pH and on the acid dissociation constant (pKa). The pKa of acetic and propionic acid is 4.8, which implies that in well made silages with a pH of about 4, more than 50% of ions should return to the undissociated form and allow the cycle described by Lambert and Stratford (1999) to continue. The percentage of undissociated acids increases as the pH drops, indicating that combining strong acids with acetic and propionic acid would intensify their antifungal action. Inoculants containing combinations of he terolactic and homolactic bacteria exploit this concept (Kung et al., 2003) as do mixtures of strong and weak acids.
44 Bacteriocin Mediated Inhibition In addition to the modes of action described above, silage inoculants may also contain bacteriocins, which are small proteins that inhibit or kill closely related bacteria species or even different strains of the same species (Yildirim, 2001). Bacteriocins synthesized by one bacterial strain can bind to specific receptors on the membrane of susceptible cells. T he receptors are proteins whose normal function is to transport substances, growth factors, or micronutrients from the outer membrane (Madigan et al., 2009). Bacteriocins also can have deleterious effects on DNA and RNA of target cells. Lactobacillus buchneri for example produces buchnericin, a bacteriocin that inhibits the growth of select species of Listeria, Bacillus, Micrococcus, Enterococcus, Streptococcus, Lactobacillus, Leuconostoc and Pediococcus genera (Yildirim, 2001). Buchnericin is bactericidal because it makes susceptible cells lose high amounts of intracellular potassium ions and become more permeable to onitrophenol D galactopyranoside ( Yildirim et al., 2002). Sparo et al. (2006 ) characterized enterocin MR99, a bacteriocin produced by Ente rococcus and isolated from corn silage, which has bactericidal effects on strains of Listeria monocytogenes Staphilococcus aureus and bovine mastitis agents and it had bacteriostatic effects on E. coli Marcinakova et al. (2008) inoculated grass silage w ith a bacteriocinproducing strain of Enterococcus faecium and reported that the population of Listeria species was reduced after 105 days of ensiling However, it was not clear whether the latter was caused by the pH decline during ensiling or by the bact eriocin itself. In general, little is known about effects of bacteriocins on silage but the limited data available indicates that bacteriocins from silage inoculant bacteria can inhibit
45 pathogenic bacteria (Kung et al., 2003) However, the effects of bac teriocins on silage fermentation and aerobic stability have not been properly characterized. Other inhibitors of undesirable microorganisms in silage produced by inoculant bacteria include benzoic acid and mevalonolactone, which is produced by Lactobacillus plantarum ( Niku Paavola et al., 1998) The authors observed that the combination of these compounds produced by L. plantarum and lactic acid had synergistic effect to reduce Fusarium species. Sorbic, formic and nitric acids or their salts also have inhib ited undesirable organisms in other studies (Kung et al., 2003). Pathogenic Agents in Silage Pathogenic microorganisms can be found in silage and some produce toxic compounds. The presence of such organisms in silage is due to their tolerance of acidic con ditions and inadequate management during the fermentation or aerobic exposure phases. Poor management of silage during the latter phase causes aerobic spoilage, which can predispose to the growth of opportunistic pathogenic microorganisms such as Listeria and Bacillus species. Pathogenic organisms such as Clostridia and Enterobacteria may also thrive if silage is poorly managed during the initial aerobic phase and the subsequent anaerobic phase of ensiling. Bacillus Bacillus are gram positive, facultativel y anaerobic, sporulating bacteria. Their growth can be decreased by environmental factors such as low water activity (<0.935) and pH (<4.8) and high temperature (>30oC) (Quintavalla and Parolari, 1993) but their endospores tolerate harsh environmental temperatures. Bacillus cereus is particularly notorious because its spores can pass through the digestive tract intact, contaminate the milk of dairy cows, survive pasteurization temperatures, and decrease the shelf life
46 of milk and cream (Christiansson et al. 1999; Pahlow et al., 2003). Furthermore, enterotoxins produced by this bacteria cause foodborne illnesses, notably emesis and diarrhea ( Ankolekar et al., 2009). Diarrheal types, also cause abdominal pain and are more commonly associated with milk from co ntaminated silage. Bacillus can dominate other silage microorganisms in certain instances such as under high temperatures (Lindgren et al., 1985), in big bale silages or after treatment with formaldehyde (Barry et al., 19 78) or antibiotics (Woolford et al .,1982). They are also more common in corn and small grain cereal silages (McDonald et al., 1991; Adesogan and Quieroz, 2009). Bacillus spp. are unable to initiate aerobic deterioration of silages, however they can be present at later stages after yeasts i ndirectly increase the pH by metabolizing lactate (Holden, 1989). However, Bacillus spp. also can produce bacteriocin like substances that inhibit the growth of yeasts and thereby increase aerobic stability of silages ( Pahlow et al., 2003). Goodman et al (1995) demonstrated that B. subtilis and B. licheniformis are capable of produce zymocin, a bacteriocin that could impair the growth of yeasts and molds. However, factors that determine whether Bacillus spp. decrease or enhance aerobic stability are unc lear. Listeria Listeria are opportunistic gram positive bacteria that cause high mortality rates and a wide range of diseases in immunocompromized animals and humans including meningitis, encephalitis, septicemia, gastroenteritis, mastitis and abortions (McDonald et al., 1991; Adesogan and Queiroz, 2009). Listeria monocytogenes (formerly called Bacterium monocytogenes) is the main causative agent and the main source of contamination of ruminants in spoiled silage (Wiedman, 2003) although some others strai ns such as L. ivanovii can infect ruminants (Sleator et al., 2009). This facultatively
47 anaerobic bacterium is ubiquitous in nature because it can tolerate refrigeration temperatures, low water activity, and a wide range of pH above 5.0 (Tienungoon et al., 2000). Listeria normally resides in decaying plant matter in the soil but is also associated with the gastrointestinal tract in a number of animals. On farms it is commonly found in baled silages because of their relatively low density, high pH, and high surface area to mass r atio (McDonald et al., 1991). In well prepared bunker silages, L. monocytogenes only thrives in areas exposed to a prolonged, low rate of oxygen infiltration because they can be dominated by more aerobic organisms like yeasts and molds and they dont survive the low pH conditions in well made silage (McDonald et al., 1991). Using rybotyping techniques, Ry ser et al. (1997) demonstrated that 77% of haylages with pH of 5 to 6 had some ribotype of L. monocytogenes or L. innocua. They also verified that 18% of all 210 silage samples were contaminated by the pathogen. In Britain, outbreaks of listeriosis are associated more commonly with sheep particularly around parturition because sheep are fed more commonly baled silages and cattle are mor e resistant though they are often asymptomatic carriers (McDonald et al., 1991; Villar et al., 2007). Fecal shedding increases the prevalence of Listeria infection among small ruminants but is a less effective method of recontamination in cattle. The pathogen can be transmitted from contaminated silages into milk and according to Sanaa and Mnard (1994) the presence of L. monocytogenes in silage increases the risk of i t s presence in milk by a factor of 20. Fortunately, the pathogen is destroyed by adequate pasteurization of milk but it may survive in soft cheeses and dairy products, which are not subjected to such treatments (Adesogan and Quieroz, 2009). Zoonotic trans mission is also possible by the
48 consumption of meat and raw foodstuffs infected with the pathogen (Swaminathan & Gener Smidt, 2007) Clostridia Clostrida are gram positive, mostly obligately anaerobic, sporulating bacteria that thrive in silages with low WSC concentrations particularly when plant moisture (>70%) and buffering capacity and the prevailing pH (>4.6) and temperature (>30oC) are high. Consequently, they often dominate the fermentation of unwilted legumes ensiled without additives (McDonald et al., 1991) Those commonly found in silage are saccharolytic types that ferment sugars and organic acids (e.g. C. butyricum and C. tyrobutyricum ) and others that ferment both sugars and protein (e.g. C. sporogenes and C. perfringens ), but those that ferment amino acids exclusively are uncommon in silages (Pahlow et al., 2003). Clostridial presence in silage is mainly from soil contamination or slurry application. When slurry is applied to forages before ensiling, it is common to find high Clostridial spore n umbers even when the environment is not ideal for the multiplication of Clostridia in silages ( Weissbach, 1993) Clostridium butyricum is the most frequently isolated species in silage. This bacterium and others in the same genus can considerably worsen si lage fermentation (McDonald et al., 1991) because such saccharolytic Clostridia derive energy by fermenting sugars and lactate into butyric acid, CO2 and H2 Although the antifungal properties of butyric acid can enhance aerobic stability, its pungent acri d odor typically depresses intake and therefore reduces performance of ruminant livestock (Adesogan and Queiroz, 2009) The depletion of lactate by saccharolytic Clostridia increases the pH and provides a more conducive environment for growth of proteolyt ic Clostridia that deaminate and catabolize amino acids into fatty acids. Consequent increases in the ammonia
49 concentration and protein solubility of silages make them less suited for high producing cattle and enhance environmental pollution from livestock operations (Adesogan and Queiroz, 2009) Furthermore, biogenic amines such as cadaverine, glucosamine, histamine, putrescine, and tyramine can be produced during Clostridial proteolysis in silages (Adesogan and Queiroz, 2009). Many of these putrefactiona ssociated compounds are malodorous and unpalatable, therefore they reduce feed intake by livestock (Neumark, 1967; Neumark and Tadmor, 1968), but some are potentially toxic. For instance histamine is lethal at high doses and when injected intravenously at low doses, it stopped ruminal motility and eructation in sheep (Dain et al., 1955). An added complication with Clostridia is that their spores can be transmitted from silage into milk or dairy products. In cheese, spores can develop unsightly outgrowths and continued butyric fermentation can lead to the formation of gas pockets that can double the size or cause cracks to appear in cheese, a phenomenon called late blowing of cheese (Cocolin et al., 2004). The large quantities of butyric acid produced result in a rancid odor and tainted flavor. Clostridium perfringens also causes enteric syndromes characterized by abdominal pain and diarrhea resembling B. cereus diarrhea ( Adesogan and Queiroz, 2009). One of the main reasons why inoculation with homolactic LAB has been successful is that it curtails Clostridial secondary fermentation by rapidly reducing the pH and increasing the concentration of bacteriocinlike substances (Pahlow et al., 2003; Pedroso et al., 2010). Arriola et al. (2011) demonstrated that a homolactic inoculant containing a mixture of E. faecium P. pentosaceus, and L. plantarum was more effective at inhibiting the growth of Clostridia than the Control silage or silages treated
50 with a heterolactic inoculant containing L. buchneri alone, dual purpose inoculants containing P. pentosaceus and either L. buchneri or Propionibacteria freudenreichii or a homolactic inoculant containing L. plantarum and P. pentosaceus. Yet the treatment with the greatest inhibitory effect on Clostridia had one of the slowest rates of pH decline, indicating that the mode of action was not based on direct acidification. Thuault et al. (1991) demonstrated that C. tyrobutyricum could be inhibited by substances other than hydrogen peroxide and lactic acid from LAB. Other tr eatment methods to curtail the growth of Clostridia include using reduced oxygen permeability film to cover bunker silos ( Borreani and Tabacco, 2008), wilting forages and using chemical additives such as nitrite ( Pahlow et al., 2003) Enterobacteria Ente robacteria are gram negative facultatively anaerobic bacteria. Epiphytic Enterobacteria including Erwinia herbicola and Rahnella aquitilis often dominate fresh crops, but these are superseded by others like Escherichia coli, Hafnia alvei and Serratia font icola during ensiling (Driehuis and Elferink, 2000). Although Enterobacteria actively compete with LAB in the early stages of ensiling, they are inhibited once the pH drops below 4.5 (Pahlow et al., 2003). Those that survive ensiling can start growing acti vely when the pH increases after aerobic deterioration (Driehuis and Elferink, 2000). Like Clostridia, Enterobacteria deaminate and decarboxylate amino acids in silages, thereby enhancing ammonia and biogenic amine production and increasing the risk of depressed intake and ineffiecient N utilization by livestock. Escherichia coli O157:H7, a shigatoxin producing gram negative bacteria is the most notorious of the Enterobacteria. It has emerged as an important cause of food borne disease. In children and the elderly, it initially causes acute bloody diarrhea but
51 this may evolve into hemolytic uremic syndrome, a severe illness characterized by anemia and kidney failure. Cattle are the main reservoir of E. coli O157:H7 and the pathogen may be present in feces, m ilk, and feed of dairy cows (Armstrong et al., 1996; Mechie et al., 1997; Chapman et al., 1997; Lynn et al., 1998). Silage can be contaminated with E. coli O157:H7 via manure or irrigation water (Weinberg et al., 2004) but the pathogen disappears from cont aminated silages when the pH drops below 4 5 (Bach et al., 2002; Chen et al., 2005; Pedroso et al., 2010). However, the pathogen has been found in decaying commercial silages with relatively high pH values and it survived for three weeks in grass silages (pH 4 to 4.6) contaminated with the pathogen (Reinders et al., 1999). Therefore, it is critical that silage pH is kept below 4 during and after ensiling to prevent the growth of the pathogen. Importance of Mycotoxins in Silage Mycotoxins are secondary m etabolites secreted by molds mostly belonging to the Aspergillus Penicillium and Fusarium genera (Yiannikouris and Jouany, 2002). The ubiquitous nature of mycotoxins and the severity of their effects on human health make them a major food safety concern. The Food and Agriculture Organization (FAO) estimates that 25% of all crops are contaminated with mycotoxins (CAST, 1989). Direct costs of disposal of condemned food and feed ingredients and indirect costs of regulatory enforcement and quality control meas ures caused by fungal toxin contamination in the USA were estimated at approximately $1.4 billion (CAST, 2003). The social importance of mycotoxins is evident from the 25,000 to 166,000 new cases of liver cancer annually caused by aflatoxins (Liu and Wu, 2010). Mycotoxins also negatively affect domestic animals causing suppression of the immune system, imbalance of hormonal function, and reduction of nutrient utilization,
52 which result in decreased performance and eventually death (Whitlow and Hagler, 2005). For the farm system, the costs associated with feeding animals contaminated diets occur in the form of poor animal performance and milk disposal due to high concentration of toxins (Applebaum et al., 1982; Masoero et al., 2007; Kutz et al., 2009). Milk with aflatoxin concentrations above 0.5 g/kg are considered illegal to be marketed in USA, Brazil, Argentina, and countries belonging to the southern common market, whereas in the European Union, the maximum permissible concentration is 0.05 g/kg (FDA, 200 0; ANVISA, 2002; EFSA, 2004). Anaerobiosis and low pH inhibit the growth of most fungi, thus ensiling is an effective strategy to prevent the growth of molds and mycotoxin production from many field and storage fungi. However, these conditions provide a m ore conducive environment for the growth of acidtolerant and low oxygentolerant species, such as Penicillium roqueforti Aspergillus fumigatus Byssochlamys nivea and various Fusarium species (Pahlow et al., 2003). Gonzalez Pereyra et al. (2008) noted that inadequate management of silage can impair fermentation, promote aerobic conditions within the silage mass, and favor the growth of fungi that are normally less tolerant of acidic or anaerobic conditions such as Aspergillus flavus Furthermore, once s ilos are opened, the ensuing aerobic conditions often allow the growth of spoilagecausing, yeasts, which metabolize lactate to CO2 and thereby increase the silage pH. This elevated pH predisposes toxigenic fungi to grow actively during the feedout phase, particularly in poorly managed silages. Mycotoxin contamination is a ubiquitous problem due to the high adaptive capacity of toxigenic fungi, which allows them to proliferate in all stages of feed production such
53 as on the field and at harvest, storage, and feed out. Nevertheless, various techniques can be used to prevent or at least minimize contamination of feeds with mycotoxins and effective detoxification methods also exist. Effects of Specific Mycotoxins More than 400 mycotoxins are known to occur naturally, however, only a few of them have been extensively studied (Whitlow and Hagler, 2005). Mycotoxins that are frequently present in silages and feed ingredients include deoxynivalenol (DON), zearalenone (ZEA), fumonisin, and roquefortine C (Dr i ehuis et al., 2008). Aflatoxin can also be present in silages made in hot and humid environments. Deoxynivalenol Deoxynivalenol is a toxin produced by Fusarium species such as F. graminearum, F. sporotrichioides, F. culmorum, F. poae, F. roseum, and F. tricinc tum Deoxynivalenol is also known as vomitoxin because it tends to cause emesis in swine. Other than vomiting, DON causes feed refusal, diarrhea, reproductive problems, and eventually death. The effect of DON in dairy cattle is not well established but dec reases in animal performance in dairy herds have been associated with the toxin (Whitlow et al., 1994). Deoxynivalenol also has been related to altered rumen fermentation (Seeling et al., 2005) and reduced flow of protein to the duodenum (Danicke et al., 2005). In a survey on the presence of mycotoxins in feed ingredients, Driehuis et al. (2008) reported that corn silage was the main source of DON and ZEA in diets of dairy cattle. These authors reported an average concentration of DON in corn and grass silage samples of 550 g/kg and a maximum concentration of 1,250 g/kg. They estimated that these high DON concentrations could result in intakes of 8.4 g of DON/kg of BW or 5 mg per cow/day. After reviewing several studies, DiCostanzo et al. (1995) concluded that beef
54 cattle are able to tolerate up to 21 mg/kg of DON. Charmley et al. (1993) demonstrated that a contamination level of 6 mg of DON /kg of diet DM did not adversely affect milk yield or cause carry over of the toxin into milk. Data available on adv erse effects of DON on dairy cattle is limited, and insufficient to allow the establishment of a maximum tolerance level, however the guidance value for DON is 5 mg/kg in Western Europe (European Commission 2006 ; Driehuis et al., 2008) and the advisory guidelines stipulated by the Food and Drug Administration (FDA) in the US are 10 and 5 mg/kg DM of diet for beef and dairy cattle, respectively (FDA, 2010). The concern about the intake of DON by dairy cattle is related to its potential negative effect on animal he alth and production. However, due to the low transfer of this toxin from the diet to milk, DON is not considered a contaminant that reduces the safety of dairy products (EFSA, 2004). Fumonisin Fumonisins are mainly produced by two species of Fusari um F. verticilloides (formerly F. moniliforme ) and F. proliferatum (Whitlow and Hagler, 2005). The structural formula of fumonisin is similar to sphingosine, which is a component of sphingolipids that abound in nerve tissue. The toxicity of this mycotoxin results from interruption of sphingolipid biosynthesis, therefore it causes paralysis, nervousness, and ataxia in horses, and pulmonary edema in swine (Marasas et al., 1988; Ross et al., 1990; Diaz and Bo ermans, 1994). The toxin tends to be less aggressiv e in ruminants than monogastrics, however fumonisin has been shown to be hepatoxic and nephrotoxic to calves fed 1 mg/kg of body weight of the toxin (Mathur et al., 2001). The later authors reported hepatocellular apoptosis and renal tubular necrosis withi n 7 d of dosing the toxin. Similar results were found in a study with beef calves supplemented with 148 mg/kg of total fumonisin in the diet for 31 d (Osweiler et al., 1993). Diaz et al. (2000)
55 demonstrated that dosing of 100 mg/kg of fumonisin reduced feed intake and milk yield in dairy cattle. Nevertheless, carry over of the toxin into milk is minimal (Scott et al., 1994). Gonzalez Pereyra et al. (2008) reported that fumonisin levels in corn silage varied from 340 to 2490 g/kg of DM. These authors also reported that samples from different locations in the silo had different concentrations of the toxin and those from the top layer and sidewalls had high pH values, which could favor development of the toxin. Fusarium species tolerate low winter temperatur es and colonize crop residues such as maize stalks and rice and wheat stubble, and these become major sources of inocula as temperatures increase in early spring (Binder, 2007). Fungal spores also become airborne during the rainy season and can travel long distances causing mycotoxin contamination epidemics (Binder, 2007; Lanier et al., 2010). The FDA advisory guideline for fumonisin is 15 mg/kg of DM of diet for lactating dairy cows (FDA, 2010). Zearalenone Zearalenone is an estrogenic metabolite produced by several species of Fusarium such as F. graminearum, F. culmorum, and F. crookwellense (Saeger et al., 2003). Zearalenone can cause numerous reproductive problems including hyperestrogenism, mammary gland enlargement, and vaginitis (Diekman and Green, 1992). Ruminants are less susceptible than pigs or chickens due to conversion of ZEA to its hydroxyl zearalenol is three to four times more estrogenic than ZEA, its lower negative effects of ZEA (Fink Gremmels, 2008). Despite their lower susceptibility to this toxin, studies report that high intakes of the toxin may negatively affect dairy cattle. For
56 instance, conception rates were decreased by 25% in dairy heifers receiving 12.5 mg/kg of ZEA (Weaver et al., 1986). Whitlow and Hagler (2005) reported a 30% incidence of ZEA contamination in 461 corn silage samples in the US and the average contamination concentration was 525 g/kg of DM. Driehuis et al. (2008) reported ZEA contamination in 13 and 50% of grass and corn silage samples in Denmark with average concentrations of 180 and 146 g/kg, respectively. Reed and Moore (2009) reported that sorghum silage contained 660 g/kg of ZEA and alfalfa contained up to 79.80 mg/kg. The main concerns with this toxin focus on its negative effect on animal health and reproduction because like fumonisin the degree of transfer into m ilk is considered negligible (Seeling et al., 2005). There are no zearalenone action limits, guidance, or advisory levels established by FDA at this time. The guidance value for zearalenone in Europe is 500 g/kg (European Commission, 2006). Aflatoxin Afla toxin is a mutagenic, carcinogenic, and toxic secondary metabolite produced Aspergillus flavus A. parasiticus and A. nomius (Creppy, 2002). Due to the ability of Aspergillus to grow over a broad range of temperatures and humidities, aflatoxin is a ubiquit ous contaminant of food and feed ingredients worldwide (Phillips, 1999). Feeding aflatoxin contaminated diets to lactating cows reduces their health and performance and also causes transfer of the toxin to milk and dairy products (Diaz et al., 2004). The s ymptoms associated with aflatoxin ingestion or inhalation include inappetence, lethargy, ataxia, enlargement of the liver, liver cancer decreased rumen motility, and reduced feed efficiency and milk production (Mathur et al., 1975; Guthrie and Bedell, 1979 ; Whitlow and Hagler, 2005). Aflatoxin also reduces the immune
57 response to opportunistic diseases and interferes with the efficiency of vaccines (Diekman and Green, 1992; Marin et al., 2002). However, these symptoms are not specific to aflatoxicosis, which makes precise diagnosis of the condition difficult (Co ulomb e, 1993). Garret et al. (1968) demonstrated increased liver weight in beef cattle fed a diet containing 100 g/kg of aflatoxin. Similar concentrations also decrease d animal production and compromi se d animal health (Patterson and Anderson, 1982). Applebaum et al. (1982) reported that highproducing animals were more sensitive to diets contaminated with impure aflatoxin than those contaminated with pure aflatoxin B1 (AFB1). The incidence of aflatoxi n is relatively low in silages compared to other mycotoxins, probably due to low tolerance of Aspergillus flavus and A. parasiticus to the acidic and anaerobic silage environment. However, high concentrations of aflatoxin can occur in poorly made or managed silages or in silages made with diseased corn plants (Queiroz et al., 2009). Gonzalez Pereyra et al. (2008) reported an aflatoxin B1 (AFB1) concentration of 156 g/kg of DM in corn silage stored in a trench type silo without proper sealing. Richard et al (2009) observed aflatoxin concentrations of up to 60 g/kg of DM in corn silage. Aflatoxin also tends to be more common in silages made in hot, humid areas, which facilitate the growth of the fungal source of the toxin. The rate of aflatoxin transfer to the milk varies from 1 to 6% (EFSA, 2004). Because of the severity of aflatoxin effects on human health, it is the only one of the 400 known mycotoxins that has an Action Level regulated by governmental agencies. The aflatoxin Action Level established by t he FDA for fluid milk is 0.5 g/kg and it is 20 g/kg for feed ingredients offered to dairy cattle.
58 Mycotoxin Detoxification Methods Despite the existence of several notable advances in pre and post harvest fungal inhibition technologies, feed contamination with mycotoxins is still very common. Silage traditionally represents a significant proportion of the diet of dairy cows and its year round, widespread use in some countries and seasonal use at times of low pasture availability in others, emphasize the i mportance of ensuring that mycotoxin contamination of silage is minimized. Yet no method effectively or completely detoxifies silages contaminated with mycotoxins. However, numerous products that improve aerobic stability via their antimycotic action may i ndirectly decrease the risk or level of mycotoxin contamination of silage. Dilution of contaminated ingredients is an effective strategy for detoxifying contaminated grains. However, it is less appropriate for contaminated silage because it would probably reduce the feedout rate and cause slower utilization of the silage on the silo face. These factors could lead to further mold growth and intensification of the contamination problem (Whitlow and Hagler, 2005). Contaminated grains can also be detoxified t hrough biological, physical, or chemical post harvest methods but for silage, the most practical and effective detoxification strategy is perhaps by using enterosorbents. Enterosorbents Enterosorbents are substances that can bind to toxins in the gastroint estinal tract of animals, reducing their bioavailability and associated toxicities (Philips, 1999). They represent an alternative method to avoid the inaccessibility or high cost of physical detoxification methods and the hazards of chemical methods.
59 Hydr ated sodium calcium aluminosilicates (HSCAS) are clay based products that form a stable complex with AFB1, which cannot cross the luminal membrane in the gastrointestinal tract (Spotti et al., 2005). In vitro evaluation of the binding capacity of HSCAS dem onstrated that 80 to 99% of aflatoxin present in various solutions (buffer, water, or rumen fluid) was bound to the clay (Moschini et al., 2008). In vivo studies with HSCAS have shown less complete, but nevertheless useful levels of detoxification. Diaz et al. (2004) evaluated different sequestering agents and reported a wide range of efficacy at reducing AFM1 in milk (31 to 65%) when a diet containing 100 g/kg of AFB1 was fed to lactating dairy cows. Kutz et al. (2009) reported a 44 to 48% aflatoxin reduc tion in milk when mycotoxin binders were used in the diet. These results emphasize the importance of carefully selecting and screening potential sequestering agents before evaluating them in animals under practical farming conditions. Satisfactory decontam ination has been achieved with the use of inorganic sequestering agents such as HSCAS but these have discriminating affinity to aflatoxin, having no effect on other toxins. For instance, the inclusion of HSCAS in diets has not changed the estrogenic effect of zearalenone (Bursian et al., 1992). When used at 0.5 or 1% of the diet, HSCAS did not mitigate the negative effects of DON on the daily gain of nonruminant animals (Patterson and Young, 1993). Other types of enterosorbents such as activated carbon, gl ucomannan, and peptidoglycans also have been evaluated. Diaz et al. (2004) studied the effect of bentonites, esterified glucomannan (yeast cell wall), and activated carbon as sequestering agents to reduce AFB1 absorption and transfer to milk. The authors r eported that esterified glucomannan fed at 0.05% of diet DM was similarly effective as
60 sodium bentonites fed at 1.2%. Both products decreased milk aflatoxin concentration by 58.5% and 64.6%, respectively. In the same experiment, activated carbon showed no significant reduction in aflatoxin M1. The results of using these binders are fairly dependent on the extent of aflatoxin contamination in the diet. For instance, the same esterified glucomannan product that reduced milk AFM1 concentration when dietary AFB1 concentration was 55 g/kg of diet DM (Diaz et al., 2004) had no effect on milk aflatoxin when dietary concentration of AFB1 was 100 g/kg of diet (Kutz et al., 2009). Cholestyramine, an insoluble quaternary ammonium anion exchange resin was used (2.5g/k g) to decrease the toxic effect of 6 mg/kg of zearalenone in mice (Underhill et al. 1995) Avantaggiato et al. (2005) stated that cholestyramine effectively binds to zearalenone and fumonisins. The authors also reported that activated carbon was the only absorbent, out of 21 products, capable of binding deoxynivalenol and nivalenol.
61 A B Figure 2 1. Homofermentative (A) and heterofermentative (B) pathways for fermentation of glucose to lactate in silage ( Reprinted with permission from Dr. Kenneth Todar ; Todar, 2011)
62 CHAPTER 3 CONTROL OF E. COLI O157:H7 IN CORN SILAG E WITH OR WITHOUT VA RIOUS INOCULANTS: EFFICACY AND MODE OF ACTION Introduction Escherichia coli O157:H7, a shigatoxinproducing gram negative bacteria, has emerged as an important cause of food borne diseases since it was first isolated in 1982 (Riley et al., 1 983). Cattle are the main reservoir of E. coli O157:H7 and the pathogen may be present in feces, milk, and feed of dairy cows (Mechie et al., 1997; Chapman et al., 1997; Lynn et al., 1998). Corn silage is one of the most widely used components of dairy cow rations and it can be contaminated with E. coli O157:H7 via application of manure or irrigation water (Weinberg et al., 2004). Escherichia coli O157:H7 was eliminated from ensil ed, contaminated wheat and corn forage when the pH decreased below 5.0 (Chen et al., 2005) Elimination of the pathogen has been hastened by treatment with a bacterial inoculant that accelerated the pH decline (Bach et al., 2002). Weinberg et al. (2004) al so confirmed that ensiling is effective in eliminating E. coli sp. in contaminated forages but noted that these bacteria may develop in spoiled silages, which often have high pH. To date, no studies have examined if bacterial inoculants that increase aerobic stability and maintain low pH after aerobic exposure such as L. buchneri (Filya et al., 2003; Adesogan et al., 2004; Huisden et al, 2009) can also prevent the growth of E. coli on aerobically exposed silages. Many lactic acid and propionic acidproducing bacteria produce substances with antibacterial activity such as peroxides and bacteriocins (Jack et al., 1995; Meile et al., 1999). Gollop et al. (2005) reported that antibacterial activity independent of pH was present in many LAB based inoculants but t he antibacterial activity was not imparted to
63 some of the silages treated with the bacteria. Little is known about the antibacterial activity of bacterial silage inoculants against E. coli O157:H7. The first objective of this study was to evaluate the eff ectiveness of bacterial inoculants containing homofermentative or heterofermentative bacteria, and both types of bacteria, at controlling E. coli O157:H7 in corn silage during the anaerobic and aerobic stages of silage production. A second objective was to examine if pH independent antibacterial activity against E. coli O157:H7 existed in the inoculants and persisted in inoculated silages. Material s and Methods Harvesting, Inoculation and Ensiling The experiment was conducted at the Department of Animal Sci ences, University of Florida from October 2007 to July 2008. Corn forage was harvested at the 50% milk line stage (30% DM) with a forage harvester (Claas Jaguar 900, Claas of America LLC., Columbus, IN) adjusted to achieve a 19mm theoretical length of cut The following treatments were applied in triplicate to the forage: 1) distilled water (Control); 2) 5 105 cfu/g of E. coli O157:H7 (ATCC 43894 EC); 3) EC and 1 106 cfu/g of Pediococcus pentosaceus 12455 and Propionibacterium freudenreichii (EC+ BII); 4) EC and 1 106 cfu/g of Lactobacillus buchneri 40788 (EC+LB); 5) EC and 1 106 cfu/g of L. buchneri 40788 and P. pentosaceus 12455 (EC+B500) The E. coli O157:H7 was supplied by ABC Research Corporation (Gainesville, FL), whereas bacteria in treatments 3, 4, and 5 were commercial inoculants, Biotal Plus II (BII), Buchneri 40788 (LB), and Buchneri 500 (B500), respectively (Lallemand Animal Nutrition, Milwaukee, WI). The E. coli O157:H7 was grown in trypticase soy broth (Becton, Dickinson and Co., Sparks, MD, USA) for 24 h at 37oC. Bacteria were sedimented by centrifugation (3000 g for 10
64 min), washed three times, and resuspended in buffered peptone water (BPW; pH 7.2; Oxoid, Basingstoke, UK). Cells were adjusted to an optical density of 0.5 at 6 40 nm (Jasco V 530 Spectrophotometer, Jasco Inc. Easton, MD) to achieve 1 108 cfu/ mL in the suspension (Bach et al., 2002). The population of E. coli in the suspension was verified by spread plating on sorbitol MacConkey agar (SMAC, Oxoid, Basingstoke, U K) supplemented with 0.05 mg/L of cefixime and 2.5 mg/ L of potassium tellurite (CT, Oxoid, Basingstoke, UK) to yield CT SMAC. Suspensions of the pathogen were diluted in distilled water and applied to achieve 5 105 cfu/g at ensiling. The commercial inoculants were suspended in distilled water and applied at the rate of 3 mL /kg of fresh forage to achieve 1 106 cfu/g in the forage and the same amount of distilled water was applied to Control. The number of bacteria in the commercial inoculants was determi ned previously by serial dilutions in BPW, spread plating in de Man Rogosa Sharpe ( MRS) agar and incubating for 24 h at 35oC (Difco; Becton, Dickinson and Company, Sparks, MD). The forage for each treatment (25 kg) was spread onto a plastic sheet and the treatment was applied using a manual sprayer. Afterwards, the four ends of the plastic sheet were drawn together and the forage was tumbled for 3 min to ensure even distribution of the inoculum. Approximately 1.5 kg of forage from each treatment was manually compacted into a 14 21 cm, thick walled plastic bag. Each treatment was prepared in triplicate and ensiled for 3, 7 or 31 days at 20oC to give a total of 45 bag silos. Samples from these bags were used to determine changes in E. coli O157:H7 counts and pH over time. In addition, 4.5 kg of forage from each treatment were manually compacted in triplicate into 20L mini silos lined with thick walled plastic bags
65 and ensiled for 82 d at 20oC at an approximate density of 500 kg/m3. On d 82, mini silos wer e opened and contents were mixed thoroughly and representatively subsampled for analysis of pH, lactate, VFA, aerobic stability, and E. coli O157:H7, yeast, and mold enumeration. Additional samples also were taken for reinoculation with E. coli. Aerobic St ability Approximately 1.3 kg of d82 silage from each treatment replicate was transferred to a plastic bag within an opentop styrofoam container. Silage boxes were covered with two layers of cheese cloth to avoid dehydration and stored at 20oC for 144 h. Silage temperature was monitored with thermocouples (Campbell Scientific, Inc., Utah) placed in the geometrical center of the silage mass in each box. Thermocouples were connected to a data logger (CR 1000 data logger; Campbell Scientific, Inc, Utah) that recorded silage temperatures at 30min intervals. Aerobic stability was defined as the number of hours the temperature in the silages remained stable before rising more than 2C above room temperature. Survival of E. coli 0157:H7 in Silages Exposed to Aer obic Conditions This aspect of the experiment determined if treatment with the commercial inoculants at ensiling would prevent the growth of E. coli O157:H7 in silages contaminated with the pathogen after silos were opened. Approximately 1 kg of silages sa mpled at silo opening and those aerated for 144 h in the aerobic stability assay were transferred into separate plastic bags and reinoculated with 1 105 cfu/g of E. coli O157:H7. Silage pH and E. coli counts were determined 24 h after reinoculation.
66 Laboratory Analyses Preparation of silage extracts Silage extracts for enumeration of E. coli analysis of VFA, pH and determination of anti bacterial activity in the silages were prepared by blending 15 g of silage and 135 mL of distilled water in a stomac her (UL LabBlender 400 Seward Laboratory, London, UK) for 2 min. For VFA determination, the extracts were filtered through two layers of cheese cloth and frozen ( 10oC) after addition of 3 mL of 50% sulfuric acid per 100 mL of extract. The extracts for enumeration of yeasts and molds were prepared by blending 25 g of silage with 225 mL of distilled water for 2 min in the stomacher. Microbial enumeration For enumeration of E. coli O157:H7 in the suspensions prepared to inoculate silages and in silage ex tracts, serial 10fold dilutions were prepared in BPW followed by spread plating on CT SMAC agar. Plates were incubated for 24 h at 35oC. Colonies on CT SMAC were screened by API 20E System (BioMerieux, Inc, Hazelwood, MO) and confirmed as E. coli O157:H 7 by latex agglutination using O157 and H7 antisera (Difco, Detroit, MI). The number of bacteria in suspensions of commercial inoculants was determined by serial dilutions in BPW and spread plating on MRS agar (Difco; Becton, Dickinson and Company, Sparks, MD). Plates were incubated for 24 h at 35oC. For yeast and mold enumeration, silage extracts were serial diluted in BPW and spread plated on Rose Bengal agar (Oxoid, Basingstoke, UK) supplemented with 1.0 mL /L of dichloran 0.2% in ethanol (Ultra Scientifi c, Kingstown, RI) and 0.1 g/L of chloramphenicol (Fisher Biotech, Fair Lawn, NJ). Plates were incubated for 5 days at 27oC. For all microbial counts, a range of dilutions was prepared in duplicate and plates yielding 30 to 300 colonies were counted.
67 Chem ical a nalysis Dry matter content in the initial forage and silages were determined by drying samples in a forced air oven at 55oC for 48 h. The pH of silage extracts was measured using a digital pH meter (Accumet AB15 Fisher Scientific). Lactate and VFA concentrations were analyzed by High Performance Liquid Chromatography as described Muck and Dickinson (1988). Antibacterial activity Antibacterial activity against E. coli O157:H7 in pure cultures of the inoculants and d 82 silage extracts was determine d by the Kirby Bauer disc diffusion test. Petri plates containing 1% of the E. coli O157:H7 inoculum (108 cfu/ mL ) were prepared on CT SMAC agar. Bacteria in the commercial inoculants were grown in MRS broth at 35oC for 24 h. Cell free supernatants were pre pared by centrifuging the cultures at 3000 x g for 10 min at 5oC. The pH of the supernatants was adjusted to 5.0 with 2 N NaOH. Similary silage extracts were centrifuged and the cell free supernatants adjusted to pH 5. Two paper discs (6 mm diameter; Cat N o. 231039, Becton & Dickinson, Sparks, MD ) were immersed into the supernatants for 15 s using flamesterilized forceps, and placed on the surface of CT SMAC agar plates containing the indicator microorganism. Plates were prepared in duplicate and incubate d for 24 h at 35oC. Zones of inhibition around the paper discs were measured with a ruler. Statistical Analysis The experiment had a completely randomized design with five treatments and t hree replicates per treatment. Data were analyzed with the General Linear Model procedure of SAS (SAS Institute Inc., Cary, NC) and a model including the treatment effect. In addition, the model used to analyze changes in pH and E. coli during ensiling
68 included time and treatment time. Differences between means were determined using the Tukey test. Significant differences were declared if P < 0.05. Results Anaerobic Phase The pH of all silages decreased to below 4 within 3 d of ensiling and remained low at all subsequent ensiling durations (Figure 3 1). The pH at final si lo opening (d 82) was greater ( P < 0.05) in silages treated with commercial inoculants (3.72) than in Control and EC silages (3.52). Escherichia coli was not detected in silages after any of the ensiling durations. Silages treated with inoculants containing L. buchneri (EC+LB and EC+B500) had lower lactate and greater acetate concentrations than other treatments (Table 3 1). The EC+LB silage had less ( P < 0.05) DM than Control, EC, and EC+BII silages. Propionate and butyrate concentrations were low or undetectable in all silages. Applying the inoculant containing propionic bacteria (EC+BII) did not result in greater propionic acid concentration versus other treatments. Aerobic Phase Yeasts and molds were not detected in silages treated with L. buchneri in oculants but they were present in other silages. Treatment with L. buchneri inoculants improved aerobic stability by at least 115% compared with the Control, but treatment with EC or EC+BII did not affect aerobic stability (Figure 3 2). Escherichia coli O1 57:H7 was not detected in silages reinoculated with the pathogen at final silo opening (d82) and exposed to the air for 24 h, probably because pH remained below 4 in all silages during that period (mean =3.78 + 0.13). However, the pathogen was present in all silages reinoculated 144 h after silo opening except in
69 that treated with EC+B500 (Figure 33 ). One day after reinoculation, Control, EC, and EC+BII silages had relatively high pH values (4.71, 5.67, and 6.09 ) and E. coli counts (2.87, 6.73, and 6.87 l og cfu/g, P < 0.05) whereas those treated with L. buchneri inoculants had low pH values (< 4) and undetectable (EC+B500) or low E. coli counts (1.97, log cfu/g; EC+LB). Counts of the pathogen were at least 10fold less in silages treated with L. buchneri inoculants versus other treatments. Antibacterial Activity The pH corrected cell free supernatants from pure cultures of each commercial inoculant produced a 2.5mm zone of inhibition against E. coli O157:H7, whereas no zone was evident from that of the C ontrol. In addition, no inhibition zone against the pathogen was produced by cell free supernatants from extracts of d82 silages treated with these inoculants. Discussion The fact that E. coli O157:H7 was undetected in silages during the anaerobic phase is most probably attributable to the inhibitory low pH resulting from formation of ferme ntation acids during ensilage. A bacterial inoculant that accelerated the rate of lactate accumulation and the resulting pH decline reduced (7 vs. 15 d) the period req uired for elimination of E. coli O157:H7 from barley silage contaminated with the pathogen (Bach et al., 2002). In this study as in others using corn and grass silage (Byrne et al., 2002; Chen et al., 2005) E. coli was eliminated during ensiling even in t he absence of inoculation because a low pH was achieved rapidly. Inoculants containing L. buchneri have improved aerobic stability of different silages ( Driehuis et al., 2001; Weinberg et al., 2002, Kleinschmit et al., 2005) In this study, inoculation with L. buchneri inoculants increased the aerobic stability of corn
70 silage by at least 115%. This effect is attribu table to the inhibitory effect of acetate produced by L. buchneri on spoilagecausing fungi (Driehuis et al., 2001; Ranjit et al., 2002). Yeasts were controlled by acetic acid concentrations above 5.6 g/L of culture medium (Woolford, 1975). As in this study, inoculation with L. buchneri typically results in acetate concentrations ranging from 36 to 50 g/kg of DM (Driehuis et al., 2001; Taylor et al., 2002). Propionic acid is the most effective antimycotic agent among short chain fatty acids at pH 5 or lower (Woolford, 1975) ; therefore, inoculants containing propionic acid bacteria have been tested for their ability to improve aerobic stability. Merry and Davies (1999) indicated that such bacteria have not inhibited consistently the growth of yeasts and molds or improved aerobic stability because they do not grow well when ensiling conditions are conducive to a rapid decline in pH. Previous studies indicated that inoculation with P. freudenreichii (1 104 cfu/g) with or without homolactic bacteria did not affect propionic acid concentration, yeast and mold counts, or aerobic stability of corn and barley silages (Ranjit et al., 2002; Taylor et al., 2002). Likewise, treatment with the inoculant containing P. pentosaceus and P. freudenreichii did not affect th ese measures in this study. Although antibacterial activity against E. coli was evident in cultures of the commercial inoculants, it was not detected in silages treated with the inoculants. Gollop et al. (2005) also reported that pure cultures of different strains of L plantarum, Enterococcus faecium and L. buchneri inhibited the growth of Micrococus luteus and Pseudomonas aeruginosa, but only some of the extracts of silages treated with these bacteria had antibacterial activity independent of pH. Antibact erial activity is dependent
71 on several factors including water activity, pH, temperature, and VFA concentration, and it varied with crop species and crop moisture concentration in the study of Gollop et al. (2005). The antibacterial activity of inoculant bacteria can be exploited to enhance forage preservation and food safety; therefore future work should determine the source of such activity in silage inoculants, and devise methods to ensure its persistence against spoilagecausing or pathogenic microorganisms in silages. Fenlon and Wilson (2000) reported that E. coli O157:H7 numbers in poorly fermented artificially inoculated silages grew from 103 to 106 log cfu/g during the first 7 d of ensilage when the pH stayed above 5. Similarly, in this study, E. c oli counts in samples reinoculated with the pathogen after 144 h of aerobic exposure increased from 105 to over 106 cfu/g when the pH was above 5. Weinberg et al. (2004) showed that though low pH achieved during ensilage eliminates E. coli from contaminate d forages, the pathogen can be found in decaying parts of commercial silages with high pH. In agreement, in this study, E. coli was undetected in silages when pH was less than 4 during ensiling. Furthermore, growth of E. coli in silages reinoculated with t he pathogen after 144 h of aerobic exposure depended on pH. High populations were present in Control, EC and EC+BII silages, which had high pH (> 4), whereas they were at least about 10fold lower in silages treated with L. buchneri inoculants which had pH < 4. These results indicate that preventing aerobic deterioration and the attendant pH increase can prevent or minimize the growth of E. coli O157:H7 on silages contaminated with the pathogen during the feedout phase. Maintenance of the low pH achieved at silo opening for 144 h by inoculants containing L. buchneri is probably attributable to the inhibitory effect of the acetate
72 produced by L. buchneri on yeasts that metabolize lactate and thereby increase the pH in aerobically exposed silages. Our data suggest that L. buchneri inoculants can be used to maintain low pH (< 4) in silages during the feedout phase and thereby curtail the growth of E. coli O157:H7, from extraneous sources that might contaminate the forage. The E. coli strain used in this experiment (ATCC 43894) was reported to have low acid tolerance (Benjamin and Datta, 1995), indicating its susceptibility to inhibition during acidic silage fermentation. However, E. coli O157:H7 may acquire acid tolerance and acid tolerant strains could conc eivably survive the low pH achieved in silages. Nevertheless, all studies we reviewed on the subject indicate that the ensiling process in wellmanaged silages is efficient at eliminating E. coli O157:H7 (Bach et al., 2002; Byrne et al., 2002; Chen et al., 2002). Future work should examine if other factors such as bacteriocins, competition with other microorganisms, and high organic acid concentrations contribute to elimination of the pathogen when forages are ensiled. Conclusions This study shows that E. c oli O157:H7 was eliminated from forages contaminated with 5 log cfu/g of the pathogen when the pH dropped to 4 within 3 d of ensiling. Inoculant treatment did not affect E. coli O157:H7 elimination during ensiling but unlike other treatments, application of L. buchneri inoculants eliminated yeast and mold populations, increased aerobic stability, and kept the pH below 4 for the duration of the 144h aerobic exposure period. Reinoculation of silages with E. coli O157:H7 after 144 h of aerobic exposure resul ted in relatively high ( P > 0.05) E. coli O157:H7 counts of 2.87, 6.73, and 6.87 log cfu/g for the Control, EC and EC+BII silages, respectively. However counts in EC+LB silages (1.96 log cfu/g) were low and the pathogen was not detected in EC+B500 silages This suggests that L. buchneri inoculants can be used to
73 curtail the growth of E. coli in silages contaminated with the pathogen at the feedout stage. All pure cultures of commercial bacterial inoculants exhibited antibacterial activity independent of pH against E. coli O157:H7 but the activity did not persist in the treated silages, suggesting that E. coli O157:H7 elimination from the silages was mediated by pH reduction.
74 Table 31. Dry matter, organic acids, and yeast and mold values of corn forage inoculated with Escherichia coli O157:H7 (EC) alone or EC and commercial bacterial inoculants and ensiled for 82 days Attributes Control EC 2 EC+BII 3 EC+LB 4 EC+B500 5 SE 6 DM 1 % 26.3 ab 26.6 ab 27.1 a 25.0 c 25.8 bc 0.34 Lactate, % DM 2.69 2.10 2.12 0.48 b 1.10 b 0.31 Acetate, % DM 2.42 c 1.73 c 1.98 c 4.77 3.81 b 0.35 Propionate, % DM 0.26 0.00 0.00 0.00 0.01 0.2 Butyrate, % DM 0.21 0.00 0.00 0.00 0.00 0.11 Yeasts and molds, log cfu/g 2.13 ab 5.67 6.30 0.00 b 0.00 b 1.7 a,b,cMeans within a row with different super scripts differ ( P < 0.05). 1DM = Dry matter; cfu/g = colony forming units per gram. 2EC = 5 105 cfu/g Escherichia coli O157:H7 3BII = 1 106 cfu/g of P. pentosaceus and Propionibacterium freudenreichii 4LB = 1 106 cfu/g of L. buchneri 5B500 = 1 1 06 cfu/g of Pediococcus pentosaceus and Lactobacillus buchneri 6SE = standard error.
75 Figure 31. Changes in pH in corn forage inoculated with 5 105 log cfu/g of E. coli O157:H7 (EC) or EC plus bacterial inoculants1 and ensiled for different durations. 1BII = P. pentosaceus + P. freudenreichii ; LB = L. buchneri ; B500 = P. pentosaceus + L. buchneri Treatment day S.E. and P value for pH = 0.33 and 1.00, respectively; *. Treatment day S.E. and P value for E. coli counts = 0.001 and 0.00, respectivel y
76 Figure 32. Effect of inoculation with 5 105 cfu/g of E. coli O157:H7 (EC) or EC and bacterial inoculants1 at ensiling on aerobic stability of corn silages ensiled for 82 d. 1BII = P. pentosaceus + P. freudenreichii ; LB = L. buchneri ; B500 = P. p entosaceus + L. buchneri S.E. = 4.11. Bars with different letters differed ( P < 0.05)
77 Figure 33. Effect of reinoculation of corn silages with 1 106 cfu/g of E. coli O157:H7 (EC) 144 h after silo opening (d 82) on pH and E. coli counts (EC; log cfu/g) of silages treated with EC or EC and bacterial inoculants1 at ensiling. 1BII = P. pentosaceus + P. freudenreichii ; LB = L. buchneri ; B500 = P. pentosaceus + L. buchneri S.E. values for pH and E. coli data were 0.41 and 1.04, respectively. Similarly shaded bars with different letters differed ( P < 0.05)
78 CHAPTER 4 EFFECT OF A DUAL PUR POSE INOCULANT ON THE QUALITY AND NUTRIE NT LOSSES FROM CORN SIL AGE PRODUCED IN FARM SCALE SILOS Introduction Bacterial inoculants have been used to dominate the epiphy tic microbial population and thereby improve the fermentation, shelf life and quality of silages (Filya et al., 2007; Pedroso et al., 2010). During ensiling, fermentation of sugars into lactic acid by homofermentative lactic acid bacteria, such as Pediococ cus pentosaceus causes a rapid pH drop which inhibits growth of undesirable microorganisms (Filya et al., 2006). Heterofermentative lactic acid bacteria ferment sugars into lactic acid as well as other VFA (McDonald et al., 1991). Lactobacillus buchneri is a heterofermentative bacterium capable of converting lactate to acetate, which is a potent antimycotic agent and is metabolized to a lesser extent than lactate by aerobic microorganisms (Oude Elferink et al., 2001; Kung, Jr. and Ranjit, 2001). The increas e in acetate due to inoculation with L. buchneri decreases the growth of spoilagecausing yeasts and molds, thereby enhancing silage aerobic stability (Kleinschmit and Kung, Jr., 2006 b ; Kristensen et al., 2010; Arriola et al., 2011). Recently, dual purpos e inoculants containing homofermentative and heterofermentative bacteria have been marketed as Combo inoculants that improve the fermentation and aerobic stability of silage. The beneficial effects of dual purpose inoculants have been proven in numerous studies (Driehuis et al., 2001; Weinberg et al., 2002; Huisden et al., 2009). However such studies examined silage prepared in laboratory silos. Small scale silos offer ideal conditions for the fermentation process. They are essential tools to model the e ffect of inoculants on silage fermentation and aerobic stability because this approach is less costly and labor intensive compared with
79 using farm scale silos. Nevertheless, it is important to validate inoculant efficacy results obtained with small scale s ilos in farm scale silos because the imperfect conditions of the latter are more challenging for proper fermentation (Mari et al., 2009 ). Also, the relatively small amount of silage used in mini silos prevents accurate estimation of inoculant effects on nu trient and other losses during the aerobic feedout phase, which can last for several months on farms. Therefore, studies demonstrating inoculant efficacy using farm scale silos are indispensable to validate the effectiveness of i noculants in mini silo stud ies. Little is known about effects of dual purpose inoculants on the fermentation of corn silage prepared using farm scale silos, and less is known about their effects on silage quality and preservation during the feedout phase in such silos. The objectiv e of this study was to examine the effects of applying a dual purpose inoculant on the fermentation, nutritive value, aerobic stability, and nutrient losses from corn silage produced using farm scale silos. We hypothesized that inoculation would improve the fermentation and aerobic stability of the silages and reduce associated nutrient and energy losses during the feedout phase. Materials and Methods Silage Production A corn hybrid (Dekalb 6970, Monsanto, St. Louis, MO) was grown at the Dairy Unit, Univer sity of Florida, Hague, FL, harvested at 34% DM ( milk line stage ) and chopped to achieve a 19mm theoretical length of cut using a forage harvester fitted with an 8row corn head (Claas Jaguar 980, Claas of America LLC., Columbus, IN). Corn plants were treated without (Control) or with a dual purpose inoculant (LB500) that supplied 1 x 105 cfu of Pediococcus pentosaceous 12455 and 4 x 105 cfu of
80 Lactobacillus buchneri 40788 per gram of fresh forage (Lallemand Animal Nutrition, Milwaukee, WI). In order t o ensure that similar forage was used for both treatments, the inoculant was sprayed on alternate 8row wide swaths of forage with a sprayer mounted on the harvester and the respective swaths were alternately packed into Ag bag silos (Ag Bag, A Miller St. Nazianz, Inc. Company, St Nazianz, WI) with a Versa 1012 bagger (Versa Corporation, Astoria, OR). Therefore, each bag containing the Control silage was filled immediately after filling the preceding one with the inoculated silage. This may have allowed mi nor inoculation of the forage initially packed into Control bags, therefore, silage in the first 5 m from the front of Control and inoculated bags w ere not used for the experiment. Inoculation and packing was completed on the same day and 4 replicate 45me tric ton bags of forage were prepared for each treatment. Wireless sensors programmed to record temperature data hourly were placed at approximately the same location in each bag during packing. After 166 d of ensiling, the bags were opened and silage was removed from the face at the rate of 500 kg/d, separated into good and spoiled (visibly moldy, darkened, slimy or heating) portions daily by the same three observers, and weighed for 35 d. Temperature sensors were retrieved during unloading of the silage and the data was used to calculate average, minimum, and maximum temperatures and time to achieve maximum temperature during ensiling. Laboratory Analysis Daily samples of good and spoiled silage were analyzed for DM by drying in a forcedair oven at 60oC for 48 h. Additional samples were collected on d 0, 7, 14, 21, 28, and 35 after opening of silage bags and immediatelly analyzed for yeast and mold counts and aerobic stability or stored at 20oC for subsequent determination of chemical
81 composition, ferm entation product profile, and gross energy. Nutrient and energy losses were quantified as the product of the dry weight of the good or spoiled silage removed and the concentration of the nutrient in the silage. Frozen w eekly silage samples were thawed, dried at 60C for 48 h in a forced air oven, ground to pass the 1mm screen of a Wiley mill (A. H. Thomas, Philadelphia, PA), and analyzed for DM (105oC for 16 h) and ash (512oC for 8 h). Concentrations of NDF and ADF were measured using the method of Van Soest et al. (1991) in an Ankom 200 Fiber Analyzer (Ank om Technologies, Macedon, NY). Heat amylase and sodium sulfite were used in the NDF assay. Nitrogen was determined by rapid combustion using a Macro elemental N analyzer (Vario MAX CN, model ID 25.005003; Elementar, Hanau, Germany) and CP was calculated as N x 6.25. Water solu ble carbohydrate concentration was determined by the anthrone reaction assay (Ministry of Agriculture, Fisheries and Food, 1986), and ammoniaN was measured by distillation (AOAC, 1985 ). An adiabatic bomb calorimeter (1261 Isoperibol, Parr Instrument Compa ny, Moline, IL) was used to calculate the gross energy concentration of spoiled samples from both treatments. Aerobic stability was measured on weekly samples by placing 2 kg of good silage i n an opentop polystyrene box. Temperature sensors (HOBO temperat ure data logger 64 k, Onset computer corporation, Cape Cod, MA) were place at the geometric center of each silage sample and data was recorded every 30 min for 7 d. Four additional sensors were placed in the room to record ambient temperature. Silages were covered with 2 layers of cheesecloth to prevent drying. Aerobic stability was denoted as the length of time that elapsed before silage and ambient temperatures differed by more
82 than 2C (Huisden et al., 2009). Also, temperature data was plotted against hours of aerobic exposure to calculate the area under the aerobic stability curve. Silage extract was prepared by mixing 25 g of corn silage with 225 mL of 0.1% peptone water in a stomacher for 3 min. The solution was filtered through 2 layers of cheesecloth and an aliquot was immediately used for yeast and mold counts as described by Schmidt and Kung Jr. (2010). The pH of silage was measured using a pH meter (Corning Model 12, Corning Scientific Instruments, Medfield, MA). An aliquot of 2 m L of silage extr act was centrifuged at 2000 x g for 15 min and the supernatant was filtered with a 0.22 m syringe filter and used for quantification of lactic acid and VFA with a h igh performance liquid chromatograph (Merck Hitachi, Elite Lachrom HTA, Tokyo, Japan) coupled to a UV Detector (Merck Hitachi L2 400) set at 210 nm. The column was a BioRad Aminex HPX 87H (Bio Rad Laboratories, Hercules, CA 9454) with 0.015 M sulfuric acid mobile phase and a flow rate of 0.7 mL /min at 45C (Arriola et al., 2011). Statistical A nalysis The experiment had a completely randomized design with two treatments and repeated measurements over time for each four replicates per treatment. Data were analyzed with the MIXED procedure of SAS (SAS Institute Inc., Cary, NC) and a model includin g treatment, time, and treatment x time effects. The model included a repeated statement and the ar (1) covariance structure was used to account for sampling from the same bag over time. Differences between means were determined using the PDIFF procedure o f SAS, which differentiates means based on the F protected least significant difference test. Significant differences were declared at P < 0.05 and tendencies at P>0.05<0.10.
83 Results and D iscussion The effects of inoculation on chemical composition and q uantity of silage classified as good is shown in Table 41. The quantity of good silage removed from Control and inoculated silos was similar (156 vs. 159 kg of DM/d, P = 0.23), however the proportion of good inoculated silage was greater than that of the good Control silage (96.6 vs. 92.2%, P = 0.004). Inoculant application did not change ( P > 0.05) concentrations of DM (34.8 vs. 34.1%), CP (9.55 vs. 9.47% DM), NDF (41.7 vs. 41.3 % of DM), NFC (40.7 vs. 41.1% DM) or WSC (3.46 vs. 3.30% of DM) in good silage. Huisden et al. (2009) reported that except for slightly decreasing WSC concentration, the inoculant used in this study did not affect the chemical composition of silage made from the same corn hybrid used in this study. Mari et al. (2009) also reported t hat no difference existed between the chemical composition of silage made using farm scale silos that had been inoculated without or with L. buchneri alone. The quantity (12.88 vs. 5.69 kg/d, P = 0.002) and proportion (7.83 vs. 3.39% P = 0.004) of spoile d silage was greater in Control versus inoculated silages (Table 42). Therefore, the lower proportion of good silage in Control versus inoculated silages was due to the greater proportion of spoiled sil age in the respective silages. That inoculation reduced the quantity of spoiled silage by about 56% is important because feeding spoiled silage can predispose cows to reduced performance and increase ingestion of pathogenic organisms. Whitlock et al. (2000) reported that ingestion of increasing quantities o f spoiled silage linearly reduced DMI and NDF digestibility in cattle and Bolsen and Bolsen (2006) suggested that these reductions could reduce daily milk production from 1.3 to 2.3 kg/d. Spoiled silage also can harbor molds that directly cause diseases s uch as a spergillosis, f armers lung, and h emorrhagic bowel syndrome and
84 indirectly cause diseases and reduced performance via mycotoxin production (Adesogan and Queiroz, 2009). In addition, spoiled silage can contain pathogenic bacteria like Listeria monocy togenes Bacillus cereus, and E.coli O157:H7, which can cause diseases like meningitis, encephalitis, septicemia, gastroenteritis, mastitis and abortions and reduce the safety, flavor, and shelf life of dairy products (McDonald et al., 199 1 ; Adesogan and Queiroz, 2009; Pedroso et al., 2010). Inoculated silages had a lower DM concentration than Control silages 35 d after silos were opened for unknown reasons (inoculation x time interaction, P = 0.07; Table 4 2; Figure 41). Inoculant application did not af fect ( P > 0.1) the mean gross energy value (20.5 vs. 20.5 KJ/g) or mean concentrations of DM (29.8 vs. 31.6% DM), CP (10.1 vs. 10.4 % of DM), NDF (47.5 vs. 47.1 % of DM), ADF (27.9 vs. 28.7 % of DM) or ash (5.01 vs. 5.05% of DM) in spoiled silage. Therefore, inoculation did not affect the composition of good or spoiled silages but it reduced the amount of silage that became spoiled. Consequently, inoculation reduced losses of CP (0. 23 vs. 0. 92 kg/d, P = 0.03), NDF (1.34 vs. 4.12 kg/d, P = 0.04), ADF (0.80 vs 2.53 kg/d, P = 0.04), ash (0.03 vs. 0.13 kg/d, P = 0.03), and gross energy (1.80 vs. 7.69 kJ/g, P = 0.02). Inoculation decreased the pH of silages (3.91 vs. 3.99; P = 0.01) but all values were below the threshold of four which indicates adequate fermentation (Table 43). Mean lactic acid concentration did not differ between Control and inoculated silages (7.63 vs. 7.86% of DM; P = 0.63), although that of the Control silage tended to be greater 28 d after silo opening (inoculation x time interaction, P = 0.09; Figure 42) but not before or after. Mean acetic acid concentration tended to be greater in inoculated silages (5.11 vs. 3.56 % of DM, P = 0.07), therefore the lactate to acetate ratio tended to
85 be lower in such silages (1.58 vs. 2.53, P = 0.08). The se results were due to a rapid increase in acetic acid concentration in inoculated silages during the first 7 d of the feedout phase (Figure 43) as well as a tendency for greater concentrations of the acid in inoculated versus Control silages at 14 and 21d (inoculant x treatment interaction, P = 0.09; Table 43 ). That inoculation increased acetate contribution is attributable at least partly to the presence of L. buchneri which ferments sugars and lactate to acetate (Oude Elferink, 2001). Application of L actobacillus buchneri alone to corn silages often increased acetate concentration at the expense of lactate concentration (Kleinschmit and Kung, 2006b ) but this difference is not always evident or statistically significant as when L. buchneri was used in c ombination with homofermentative bacteria (Kleinschmit and Kung, 2006a ; Huisden et al., 2009) as in this study. It is interesting to note that acetic acid concentrations of Control and inoculated silages did not differ statisti cally until 7 d after opening This delay suggests that aerobic organisms may have contributed to the greater acetate production in inoculated silages because effects of L. buchneri on acetic acid concentrations typically occur during ensiling (Driehuis et al., 1999; Oude Elferink, 2001). Acetic acid bacteria can oxidize ethanol and lactate to acetate and they have been implicated in initiating spoilage in corn silage in some studies (Klienschmit and Kung, 2006 b ). In addition to increasing the acetic acid concentration of silages, L. b uchneri may also increase indirectly the propionic acid concentration because it can convert lactic acid into 1, 2propanediol, which is converted to propionic acid when L. diolivorans is present (Krooneman et al., 2002). That inoculation did not affect pr opionic acid concentration in this study may reflect the absence of L. diolivorans in the silages. The
86 concentrations of butyric acid were similar in Control and inoculated silages (0.19 vs. 0.25% of DM, P = 0.52) and the values were slightly above the thr eshold of 0.04% which may indicate that Clostridial fermentation occurred (Pahlow et al., 2003 ; Arriola et al., 2011). T emperatures recorded during fermentation were typical of those in appropriately packed and s ealed silages (Kung Jr., 2008). Maximum ensi ling temperature did not differ between treatments (30.1 vs. 28.9oC, P = 0.51) and these temperatures were achieved within the first 3 d of ensiling irrespective of treatment (53.8 vs. 49.7 h, P = 0.86). The first three ensiling days represent the period w hen residual oxygen in the silage mass is used up by obligate and facultative microorganisms during the initial part of the aerobic phase (Pahlow et al., 2003). The minimum temperature (20.1 vs. 20.6oC, P = 0.57) and temperature range (10.1 vs. 9.3oC, P = 0.76) did not differ between treatments. However, mean ensiling temperature tended to be greater in Control versus inoculated silages (22.7 vs. 22.1oC, P = 0.08), suggesting that inoculation reduced wasteful heat production from microbial activity during ensiling. Inoculation reduced yeast and mold counts (2.59 vs. 4.62 cfu/g, P = 0.01; Table 44) likely reflecting the antifungal activity of the acetic acid, which tended to increase with inoculation. Aerobic stability was numerically, but not statistical ly increased (14.7 vs. 9.5 h, P = 0.71) by inoculation. Aerobic stability is typically increased when L. buchneri alone is applied to corn silage. Based on 23 published experiments, Kleinschmit and Kung, Jr. (2006 b ) used a metaanalysis to demonstrate that application of L. buchneri alone increased acetic acid concentration in corn and small grain silages, reduced the yeast and mold population, and increased aerobic stability. Several subsequent studies
87 have also confirmed the efficacy of using L. buchneri inoculants to increase silage aerobic stability (Huisden et al., 2009; Mari et al., 2009). Nevertheless, a f ew contradictory studies exist. For instance, numerical nonstatistical increases ( P > 0.1) in aerobic stability were detected when L. buchneri alone or L. buchneri and P. pediococcus were applied to corn silage in a mini silo study (Arriola et al., 2011). Similarly, inoculation with L. buchneri alone did not improve the aerobic stability of bermudagrass silage largely because Control silages had undergone a secondary Clostridial fermentation, which made them stable by increasing the concentration of butyric acid (Adesogan et al., 2004). Schmidt and Kung, Jr. (2010) reported that inoculation with L. buchneri alone improved the fermentation and aerobic stability of silages made at 3 of 5 locations, further indicating that factors other than the prevailing fungal population can influence the efficacy of L. buchneri inoculants. Such factors include the duration of ensiling and initiation of spoilage by ac etic acid bacteria rather than yeasts (Pahlow et al., 2003; Kleinschmit and Kung, Jr. 2006b), high epiphytic counts of L. buchneri (Arriola et al., 2011), and differences among corn hybrids (Kang et al., 2009). No treatment difference in aerobic stability was detected when defined as the time that elapsed before silage and ambient temperatures differed by more than 2oC. However, other important but less commonly used measures of aerobi c stability warrant evaluation. For instance, spoilage bacteria and fung i produce CO2 as a fermentation byproduct (Spoelstra et al., 1988), therefore, CO2 production has been used as a measure of aerobic stability in some studies (Ashbell et al., 2002; Weinberg et al., 2011). Other studies have simply graphically illustrated c hanges in temperature over
88 time for inoculated and Control silages (Salawu et al., 2001; Adesogan et al., 2004). In the current study, total heat accumulation over time during the aerobic feedout phase was reduced by about 16% by inoculation (1916 vs. 2212 oC h, P = 0.02). This aerobic stability index is particularly useful for detecting effects of inoculation in studies where heat accumulation over time is substantially different between Control and inoculated silages that start heating at approximately the same time. Maximum post ensiling temperature (43.1 vs. 43.4 oC, P = 0.85), silo face temperature (36.18 vs. 34.86oC, P = 0.42), and temperature range (17.6 vs. 19.1oC, P = 0.32) were not affected by treatment but minimum post ensiling temperatures tended to be reduced by inoculation (25.5 vs. 24.3oC, P = 0.09). The short aerobic stability and discrepancy among measures of heat production and aerobic stability in the silages suggest that the inoculant was not as effective at increasing the bunk life of t he silage as in other studies. This is attributable to factors such as a relatively low packing density (approximately 168 kg/m3) in the bags, which made it challenging to maintain a straight silo face and to achieve a feedout rate of at least 15 to 30 cm of silage per day (Pitt and Muck, 1993). Furthermore, the prevailing subtropical conditions, which are ideal for the growth of spoilage causing yeasts and molds (Oude Elferink et al., 2001; Adesogan, 2009) may have contributed to poor aerobic stability response. Conclusions This study showed that application of a dual purpose inoculant did not affect the nutritive value of good or spoiled corn silage made using farm scale silos but reduced the amount and proportion of spoiled silage by over 50%, and thereby reduced the associated energy and nutrient losses from the silage. Inoculation reduced the yeast and mold population and consequently reduced the total heat accumulation during the
89 aerobic feedout phase. Most beneficial effects of the inoculant were attr ibuted to its tendency to make the fermentation more heterolactic by increasing the concentration of acetate.
90 Table 41. Effect of inoculant treatment1 on the quantity and chemical composition of good2 silage removed daily from silos Control Inoculan t SEM 3 P value treatment P value time P value treatment x time Quantity, kg of good silage DM/d 156 159 2.12 0.23 0.20 0.65 Percentage of good silage, % 92.2 96.6 1.08 0.004 0.76 0.71 Chemical composition of good silage DM, % 34.1 34.8 0.96 0 .60 0.11 0.51 CP, % 9.47 9.55 0.21 0.79 0.01 0.58 ADF, % 24.9 24.0 0.64 0.29 0.21 0.85 NDF, % 41.3 41.7 1.20 0.81 0.12 0.49 NFC 4 % 41.1 40.7 1.16 0.83 0.06 0.55 WSC, % 3.30 3.46 0.27 0.63 0.61 0.40 Ash, % 3.45 3.29 0.12 0.36 0.60 0.86 1Control = no inoculant added, Inoculant = 1 x 105 cfu/g of P. pentosaceus and 4 x 105 cfu/g of L. buchneri 2Silage that was not visibly moldy, dark, heating or slimy 3SEM = standard error of mean. 4Calculated as NFC = 100 [CP+ ash+ fat (NRC, 2001 values) +NDF]
91 Table 42. Effect of inoculant treatment1 on the quantity and chemical composition of spoiled2 silage removed daily from silos and the associated nutrient and energy losses Control Inoculant SEM 3 P value treatment P value time P value treatment x time Quantity of s poiled silage, kg /d 12.88 5.69 1.53 0.002 0.40 0.49 Percentage of spoiled silage 7.83 3.39 1.08 0.004 0.76 0.71 Chemical composition of spoiled silage DM, % 31.6 29.8 1.36 0.36 0.12 0.07 CP, % 10.4 10.1 0.22 0.35 0.02 0.32 AD F, % 28.7 27.9 0.67 0.41 0.22 0.67 NDF, % 47.1 47.5 0.71 0.63 0.05 0.90 Ash, % 5.05 5.01 0.41 0.94 0.03 0.63 GE, KJ/g 20.5 20.5 0.08 0.99 0.02 0.60 Energy and nutrients lost in spoiled silage CP kg/d 0.92 0.23 0.20 0.03 0.02 0.39 ADF, kg/d 2 .53 0.80 0.54 0.04 0.01 0.51 NDF, kg/d 4.12 1.34 0.88 0.04 0.01 0.51 Ash, kg/d 0.13 0.03 0.03 0.03 0.02 0.30 GE, KJ/d 7699 1809 168 0.02 0.02 0.29 1Control = no inoculant added, Inoculant = 1 x 105 cfu of P. pentosaceus and 4 x 105 of L. buchneri 2Silage that was visibly moldy, dark, heating or slimy 3SEM = standard error of mean.
92 Table 43. Effect of inoculant treatment1 on fermentation indices and temperature during ensiling in good2 corn silages Control Inoculant SEM 3 P value treatment P val ue time P value treatment x time pH 3.99 3.91 0.20 0.01 0.04 0.25 Ammonia nitrogen % of DM 1.25 1.58 0.2 0.34 0.01 0.13 Lactic acid, % of DM 7.86 7.63 0.32 0.63 0.60 0.09 Acetic acid, % of DM 3.56 5.11 0.50 0.07 0.05 0.10 Lactate: acetate 2.53 1.58 0 .32 0.08 0.19 0.20 Propionic acid, % of DM 0.94 1.10 0.07 0.17 0.15 0.86 Butyrate, % of DM 0.19 0.25 0.06 0.52 0.70 0.54 Mean temperature, o C 22.70 22.13 0.22 0.08 NA 4 NA Maximum temperature, o C 30.16 28.89 1.41 0.51 NA NA Minimum temperature, o C 20.0 5 20.62 0.67 0.57 NA NA 1Control = no inoculant added, Inoculant = 1 x 105 cfu of P. pentosaceus and 4 x 105 of L. buchneri 2Silage that was not visibly moldy, dark, heating or slimy. 3SEM = standard error of mean. 4NA = not applicable
93 Table 44. Eff ect of inoculant treatment1 on fungal counts, aerobic stability and temperature of good2 corn silage Control Inoculant SEM 3 P value treatment P value time P value treatment x time Yeasts and molds, log cfu/g 4.62 2.59 0.62 0.01 0.08 0.29 Aerobic stabi lity (h) 9.5 14.7 10.6 0.71 0.63 0.35 Area under the temperature curve 2212 1916 81 0.02 0.01 NA 4 1Control = no inoculant added, Inoculant = 1 x 105 cfu of P. pentosaceus and 4 x 105 of L. buchneri 2Silage that was not visibly moldy, dark, heating or s limy 3SEM = standard error of mean. 4NA = not applicable.
94 Figure 41. Changes in dry matter concentration of spoiled silage with time. = P <0.05.
95 Figure 42. Changes in lactic acid concentration of corn silage with time. = P <0.05
96 Fi gure 43. Changes in acetic acid concentration of corn silage with time. = P <0.05, ** = 0.05
97 CHAPTER 5 RELATIONSHIP BETWEEN RUST INFESTATION AND THE QUALITY AND SAFE TY OF CORN SILAGE TREAT ED WITH OR WITHOUT A BACTERIAL INOCULANT Introduction Southern corn rust is an aggressive disease of corn caused by Puccinia polysora Underw. Pustules produced during the growth of the fungus cause severe damage to the leaf cuticle causing increased transpiration, premature leaf desiccation, and reduced photosynthesis and nutrient translocation (Lucas, 1998). The fungus can kill the hosts and yield losses of up to 45% have been reported in cases with severe epiphytotic contamination in the US (Rodriguez Ardon et al., 1980) and other areas of the world (Rhind et al., 1952; Liu and Wang, 1999). The disease has several similar symptoms and effects to common rust, which is caused by Puccinia sorghi ; however, cool wet conditions favor Common rust whereas hot humid conditions favor Southern rust. Despite concerted research efforts that have produced resistant commercial hybrids, the unpredictable occurrence of Southern rust and its fast and aggressive development make it a real threat to dairy operations in the Southeast that rely on corn silage for dietary energy and fiber. Infestation of corn plants by Puccinia polysora can affect negatively the chemical composition of corn forage. Johnson et al. (1997) reported increases of DM, NDF, and ADF and a decrease in DM digestibility of corn forage infested by Southern rus t. Yet, mycotoxins are important food safety hazards that are commonly produced after the growth of opportunistic fungi on previously stressed or challenged plants (Whitlow and Haggler, 2005). Undesirable microorganisms can survive during ensiling in inadequately packed silage particularly in the top layer or around the sides of bunkers or in air pockets in the
98 silage mass. Such areas can contain sufficient oxygen to favor the survival or growth of mycotoxin producing fungi. High concentrations of aflatox ins, which are carcinogenic secondary metabolites of A. parasiticus and A. flavus were detected in aerobically exposed corn silages (Gonzales Pereyra, 2008). Dual purpose inoculants containing homofermentative and heterofermentative bacteria are used to improve silage fermentation and aerobic stability, respectively (Driehuis et al., 2001; Filya, 2003; Adesogan et al., 2004). To our knowledge, no study has evaluated if such inoculants can improve the quality and safety of diseaseinfested silage. Yet the safety of feeding silage to animals depends on the bacterial and fungal population particularly because fungi may introduce significant amounts of harmful mycotoxins to animals (Mall m an et al., 2009). This study aimed to examine the relationship between increasing severity of rust infestation on corn plants and the fermentation, nutritive value, aerobic stability, and safety of the resulting silage. A second objective was to determine if bacterial inoculation could mitigate adverse effects of rust infest ation on the quality, aerobic stability, and safety of corn silage. We hypothesized that rust infestation would decrease the quality of corn silage and inoculation would mitigate adverse effects of the disease on measures of silage quality and safety. Mate rials and Methods Silage, Treatments and Design A corn hybrid was grown on a 65ha field in August 2007 as a summer or second planting silage crop. The corn was infested with Southern rust at tasselling and treated with Abound fungicide (Syngenta Crop Prot ection, LLC, Greensboro, NC) by aerial application using a crop duster airplane. However, uneven coverage of the fungicide
99 allowed rust to persist in some parts of the field. Consequently, in some areas, plants were completely disease free (no rust, NR) whereas in other areas leaves in the lower half (medium rust, MR) or all leaves (high rust, HR) were infested as shown by dense occurrence of small orange pustules on the leaves as well as yellowish brown discoloration (Figure 51). Corn forage from representative parts of the field with each rust classification were harvested when the DM of NR plants was 40% DM, and chopped using a Jaguar 900 forage harvester (Claas of America Inc., Omaha, NE). The forages were separated into piles and either treated with 100 m L of deionized water (Control) or 100 m L of a solution of Buchneri 500 inoculant (LB500 Lallemand Animal Nutrition, Milwaukee, WI) that delivered 1 105 cfu/g of Pediococcus pentosaceus 12455 and 4 105 cfu/g of Lactobacillus buchneri 40788 The inoculant also contained enzymes resulting from Trichoderma reesei fermentation and their minimum activities were 400 mg of glucanase), 1.140 mg of glucose/min/g (xylanase), and 18 mg of glucose/min/g (galactomannanase). Four replicates of each treatment were ensiled in 20 L laboratory silos at a density of 250 kg of DM/m3 for 97 days at ambient temperature (25oC) in an enclosed barn. Each of the 24 silos was sealed with airtight plastic lids and lids were secured with heavy duty duct tape. Samples of freshly treated unensiled forage were stored at 20C for chemical analys e s. Weights of empty and full silos at ensiling were recorded for estimation of DM loss. At silo opening, silo weights were recorded, and the top slimy or moldy layer of silage was disposed of where present, as is the norm in practice. The remaining forage was transferred to a 20 L plastic bag for proper mixing, after which subsamples were collected for analysis of nutritive value, microbial counts, aerobic stability, and mycotoxins.
1 00 Laboratory Analysis Corn forage and silage samples were dried at 60C for 48 h in a forcedair oven. Dried samples were ground to pass the 1mm screen of a Wiley mill (A. H. Thomas, Philadelphia, PA) and analyzed for DM (105oC for 16 h) and ash (512oC for 8 h). Concentrations of NDF and ADF were measured in an ANKOM 200 Fiber Analyzer (Ankom Technologies, Macedon, NY ) using the method of Van Soest et al. (1991). Heat amylase and sulfi te were used in the NDF assay. Nitrogen (N) was determined by rapid combustion using a Macro elemental N analyzer (Vario MAX CN, model ID 25.00 5003; Elementar, Hanau, Germany) and CP was calculated by multiplying N concentration by a factor of 6.25. In vitro DM digestibility (IVDMD) was measured with the method of Van Soest et al. (1966) and NDFD was estimated after analysis of dried samples and residues from the IVDMD assay for NDF. Silage samples were also analyzed for alflatoxin, zearalenone, deoxynivalenol, and fumonisin by thinlayer chromatography (Scott et al., 1970) at Dairyland L aboratories, Arcadia, WI). Silage extract was prepared by mixing 25 g of corn silage with 22 5 mL of 0.1% peptone water in a stomacher for 3 minutes. The solution was filtered through 2 layers of cheesecloth and an aliquot was immediately used for fungal counts as described by Schmidt and Kung Jr. (2010). Yeasts and molds were enumerated after ser ial 10 fold dilutions by spreadplating on malt extract agar (Oxoid CM59, Oxoid Inc.). Silage pH was measured using a pH meter (Corning Model 12, Corning Scientific Instruments, Medfield, MA). An aliquot of 100 m L of silage extract was stored at 20C aft er its pH was reduced to 2 with 50% (wt./vol.) sulfuric acid. Subsequently, such samples were used to quantify lactate and VFA with a high performance liquid chromatograph (Hitachi, FL 7485, Tokyo, Japan) coupled to a UV Detector (Spectroflow 757, ABI
101 Ana lytical Kratos Division, Ramsey, NJ) set at 210 nm. The column was a BioRad Aminex HPX 87H (Bio Rad Laboratories, Hercules, CA 9454) with a 0.015M sulfuric acid mobile phase and a flow rate of 0.7 mL /min at 45C (Arriola et al., 2011). Aerobic stability was measured by placing 2 kg of silage in an open top polystyrene box. Temperature sensors (HOBO temperature data logger 64k, Onset computer corporation, Cape Cod, MA) were placed in the geometric center of each silage sample and temperatures were recorded every 30 min for 14 d. Four additional sensors were placed in the room to record ambient temperature. Silages were covered with 2 layers of cheesecloth to avoid drying. Aerobic stability was denoted by the time that elapsed before the difference between silage and ambient temperature was above 2 C (Huisden et al., 2009). Samples of unensiled corn forage also were analyzed for DM, CP, NDF, ADF, IVDMD, NDFD, alflatoxin, zearalenone, deoxynivalenol, and fumonisin as described above. Statistical Analysis Thi s experiment was analyzed as a completely randomized design with 3 ( severity of rust: NR, MR, and HR) 2 (inoculation treatments: inoculated vs. Control) factorial arrangement of treatments. The REG procedure of SAS (Version 9. 2 SAS Institute Inc., Cary, NC) was used to evaluate relationships between rust levels and the measures of the fermentation, nutritive, shelf life, and safety of corn silage. The GLM procedure of SAS (Version 9.2, SAS Institute Inc., Cary, NC) and a model containing inoculant, rust a nd interactions of these terms, was used to analyze the data. Mean separation was performed using the PDIFF procedure of SAS, which employs the Fishers F protected least significant difference test. Polynomial contrasts were used to examine effects of
102 inc reasing severity of rust infestation on measures of silage quality and safety. Significance was declared at P < 0.05 and tendencies at 0.1 > P > 0.05. Results and Discussion The chemical composition of the freshly treated corn forage is shown in Table 51. The chemical composition of the forage was similar to that of corn forage in other studies in Florida (Kim and Adesogan, 2006 ; Huisden et al., 2009) except that concentrations of NDF and ADF of MR and HR silages were greater. As indicated by the adjust ed r2 value, rust infestation accounted for ( P < 0.05) most of the variability in DM concentration (82%), DM loss (76%), pH (76 %), and DMD (65%), almost half of that in NDF concentration (41 %) and between 20 and 24% of that in ADF, butyrate, and NDFD (Tabl e 52) Weak relationships (adjusted r2 < 20 % ) also existed ( P < 0.05) between the level of rust infestation and mold counts (15%) and concentrations of acetate (14%) and ammoniaN (15%). The strong relationship between DM concentration and the level of rust infestation was visually evident as increasing rust infestation made the crops drier. It is noteworthy that rust infestation was also highly correlated with pH, DM loss, and DMD, which are some of the most important measures of silage quality. The respective slopes and adjusted r2 of these relationships indicate that as rust infestation increased, pH and DM concentration increased, and IVDMD decreased. Consequently, increasing rust infestation worsened the fermentation of corn silage, increased the assoc iated losses, and reduced the nutritive value of the silage. Although no relationship was found between rust infestation and aerobic stability, it is pertinent to note that mold counts ( P = 0.03, adjusted r2 = 0.15) and aflatoxin concentration ( P = 0.07, adjusted r2 = 0.09) had weak but significant relati onships with rust infestation. This is likely because rust infestation by Puccinia
103 polysora allowed opportunistic Aspergillus molds to infest corn plants and produce the mycotoxin. Like lodging, and insect injury, disease infestation often allows opportunistic fungi to invade plant tissues and produce mycotoxins (Whitlow and Haggler, 2005; Adesogan and Quieroz, 2009). The chemical composition, IVDMD and NDFD of silages are shown in Table 53. Increasing rust infestation increased ( P < 0.001, quadratic) the DM concentration of the silages and resulted in a 20percentage unit difference between NR and HR silages. Crude protein and ash concentrations decreased and then increased ( P < 0.001, quadratic) with increasing rust infestation, with the nadir occurring in MR silage. The NDF and ADF concentrations increased linearly ( P <0.01), whereas IVDMD decreased linearly ( P < 0.001) as rust infestation increased. The decreases in the nutritive value of corn silage with increasing rust infestation reflect the effects of the disease on corn forage and indicate that ensiling did not mitigate adverse effects of rust infestation on silage nutritive value. Puccinia polysora induces a premature desiccation of the plant (Rodrig uez Ardon et al., 1980), which could explain the adverse effects of increasing rust infestation on forage and silage DM concentration and nutritive value in this study. Johnson et al. (1997) reported that the DM, NDF and ADF concentrations of 5 corn hybri ds were increased by Southern rust infestation leading to decreased IVDMD. Potter (1987) and Wilson et al. (1991) reported that rust infestation decreased the digestibility of two cultivars of ryegrass and pearl millet, respectively. In agreement, this stu dy shows that rust infestation decreased the digestibility of corn forage and silage, largely by increasing the relative proportion of fiber in the silage. Johnson et al. (1997) reported that rust infestation had no effect on the CP concentration of corn plants.
104 Although a quadratic CP response to increasing rust infestation was detected in this study, the differences in CP concentration were very small. Based upon a literature review Dimmock and Gooding (2002) stated that the effect of rust on wheat protein is variable even though rust may have a deleterious impact on N accumulation and partitioning within the plant. The progressive increase in fiber concentration as rust infestation increased is similar to findings of Johnson et al. ( 1997) and is attributable to premature desiccation of the plant or increased lignin deposition (Hammerschmidt, 1984), which is a defense mechanism against the invading fungus and the associated disease (Ride, 1978 ). Increases in fiber contamination with rust infestation may have also been caused by utilization of digestible nutrients in the plant by the fungus. Inoculation slightly reduced ( P < 0.001) the DM concentration and increased ( P = 0.02) the ADF concentration of silages, perhaps reflecting the use of water soluble ca rbohydrates as substrates by the bacteria in the inoculant. Inoculation also increased the NDFD of silages ( P = 0.023), reflecting the action of the fibrolytic enzymes in the inoculant as well as those released by L buchneri (Kang et al., 2009 ). All sila ges had pH equal to or less than 4, indicating that the fermentations were satisfactory (Table 54 ). However, increasing rust infestation increased the pH values (quadratic, P < 0.0 01). The latter occurred largely because increasing rust infestation progre ssively decreased concentrations of lactate, acetate, propionate, and total VFA in Control forages but these trends were mostly reversed by inoculation (rust inoculant interaction, P < 0.01). These results demonstrate perhaps for the first time that rust infestation worsens the fermentation of corn silage but inoculant application can mitigate
105 such effects. It is also noteworthy that inoculation was most effective at improving the fermentation of HR silages, probably because uninoculated HR silages had had fewer epiphytic bacteria or less fermentable substrate. AmmoniaN concentrations decreased to a nadir in MR silages as rust infestation increased in Control silages but increased progressively in inoculated silages (rust inoculant interaction, P < 0.00 1). Butyrate was only detected in HR silages and Control HR silages had slightly less ( P = 0.023) butyr ate than inoculated HR silages. Butyrate is primarily produced by Clostridia in silages with relatively high moisture concentrations, slow acidification rates or high final pH values (> 4.8). Clostridia typically increases proteolysis and deamination in silages resulting in greater ammonia N concentrations (McDonald, 1991). In this study, inoculation decreased the DM concentration, increased the pH, and tended to increase ammoniaN concentration. Therefore, inoculation may have increased Clostridial growth. Nevertheless, the difference in butyrate and ammoniaN concentrations between inoculated and Control silages were minimal and these disadvantages were ou tweighed by the benefits of inoculation. Yeasts were not detected in the silages and inoculation did not affect DM losses. Mold counts decreased (quadratic, P < 0.05) as the level of rust infestation increased likely because of the concomitant increases in DM concentration reduced the water activity of the silage and thereby curtailed growth of molds. Aerobic stability increased (quadratic, P < 0.01) as the level of rust infestation increased reflecting the corresponding decrease in mold counts. Inoculat ion had no effect ( P = 0.13) on mean mold counts though counts were 3.5 times greater in Control HR silages versus inoculated HR silages (3.40 vs. 0.95 cfu/g). The latter explains the 75% improvement in
106 the aerobic stability of Control versus inoculated HR silages (44 vs. 77 h). The increase in aerobic stability with increasing rust infestation was more pronounced in inoculated than Control silages (rust inoculant interaction, P = 0.01) further justifying inoculation of the silages. Aflatoxin was only det ected in the Control HR silage (rust x inoculant interaction, P = 0.10) and the level (5.2 mg/kg) was 260 times greater than the Action Level of 20 g/ kg stipulated by the US Food and Drug Administration for dietary ingredients for dairy cattle. Consequently, the Control HR silage would be unsafe to feed. High levels of aflatoxin in corn silage are notable but not unique. Gonzales Pereyra et al. (2008) reported an aflatoxin concentration of 156 g/ kg in corn silage. Yet silage is expected to have low conc entrations of aflatoxin because the toxin is primarily produced by Aspergillus flavus and A. parasiticus which have low tolerance for the anaerobic conditions and low pH in silage (Holmquist et al., 1983). However, the latter factors do not prevent the survival of these organisms in silage. Garon et al. (2006) successfully isolated A. parasiticus from corn silage samples with a high aflatoxin concentration (34 g/kg). That no toxin was found in the inoculated HR silage suggests that inoculation prevented accumulation of the toxi n in the rust infested silage. Future research should validate this finding and determine effects of rust infestation on populations of A. parasiticus and A. flavus on corn forages before and after ensiling. No deoxynivalenol or f umonis in was detected in the silages. However, unlike rust infested silages, Control and inoculated NR silages had high concentrations of zearalenone (0.6 5 vs. 0.47 mg/kg). The FDA does not have action, advisory or
107 guidance levels established for zearalenone; however, the concentrations of zearalenone in this study were below the guidance level (2000 g/ kg) stipulated by the European Commission legislation (European Commission, 2006). Nevertheless, the presence of zearalenone in the uninfested corn silages indicates that summer or second planting corn silage produced in Florida could be infested with this mycotoxin, perhaps because the hot, humid conditions allow the fungi that produce this mycotoxin to thrive. No data was found in the literature on how inoculation affects the quality of diseaseinfested silage. However, effects of bacterial inoculation on the quality of ensiled healthy corn plants are well documented (Pedroso et al., 2010; Schmidt and Kung 2010; Arriola et al., 2011). Huisden et al. (2009) reported that the aerobic stability of corn silage was increased by the same inoculant used in this study (LB500; 143 vs. 389 h). Schmidt and Kung (2010) demonstrated that L. buchneri alone or in combination with Pediococcus pentosaceus increased the conce ntration of acetate, which decreased the population of spoilage fungi and increased aerobic stability of corn silage. In this study, mean concentrations of antifungal acids (acetate, and butyrate) were increased by inoculation and hence mean aerobic stabil ity was increased. The inoculated HR silage had the greatest acetate concentration, the lowest mold counts, and the best aerobic stability. These results reflect the ability of the L. buchneri in the inoculant to synthesize the antifungal organic acid, ac etate by using lactate as a substrate (Oude Elferink et al., 2001). Conclusions Rust infestation was positively correlated with pH, DM concentration, and DM recovery and negatively correlated with IVDMD and it explained most of the variability in these measures. Therefore, increasing rust infestation reduced the nutritive value,
108 worsened the fermentation, and inc reased losses from the silage. Silages with the highest level of rust infestation had aflatoxin concentrations that exceeded the FDA Action Level for livestock feeds by 260%, implying that such silage is unsafe to feed. Inoculation with the dual purpose inoculant decreased most negative effects of rust infestation on the nutritive value and fermentation of the silage, prevented accumulation of afla toxin in the silage, and increased aerobic stability.
109 Table 5 1. Chemical composition of untreated and inoculated corn forage with different levels of rust infestation (% of DM)1 Control Inoculant NR MR HR NR MR HR DM 39.3 40.7 58.2 39.0 42.1 58.6 CP 7.40 7.41 8.07 7.96 6.82 8.32 NDF 44.7 49.1 55.1 45.3 51.1 54.6 ADF 25.5 29.3 32.9 26.4 28.4 31.2 ASH 4.43 3.76 4.79 4.66 3.86 4.40 1Control = no inoculant added, Inoculant = 1 x 105 cfu of P. pentosaceus and 4 x 105 of L. buchneri NR = no ru st level; MR = medium rust level; HR = high rust level.
110 Table 52. The relationship between the severity of rust infestation and measures of the quality of the fermentation, nutritive value, safety, and shelf life of corn silage Parameter Intercept Slope R 2 Adjusted 1 R 2 RMSE P value pH 3.46 0.16 0.77 0.76 0.07 <.0001 NH 3 N, 0.16 0.05 0.19 0.15 0.08 0.03 Lactate 3.86 0.24 0.04 0.001 0.97 0.32 Total VFA 4.75 0.07 0.003 0.04 1.16 0.80 Acetate 1.06 0.31 0.17 0.14 0.56 0.04 Propionate 1.15 0.11 0.12 0.08 0.26 0.08 Butyrate 2.94 0.42 0.24 0.20 0.64 0.01 DM loss 3.46 0.16 0.77 0.76 0.07 <0.001 Molds, log cfu/g 0.16 0.50 0.19 0.15 0.08 0.03 Aerobic stability, h <0.001 0.32 0.04 0.002 0.97 0.32 DM 25.3 9.88 0.83 0.82 3.74 <0.001 CP 9.03 0.11 0 .06 0.02 0.38 0.25 NDF 41.2 3.06 0.44 0.41 2.95 0.001 ADF 22.0 1.88 0.24 0.20 2.84 0.02 Ash 4.89 0.20 0.11 0.07 0.48 0.11 IVTDMD 2 72.6 4.55 0.66 0.65 2.76 <0.001 NDFD 3 45.0 3.00 0.28 0.24 4.17 0.02 Aflatoxin 1.74 1.30 0.13 0.09 2.79 0.07 1Adjust ment of the r2 value for degrees of freedom following analysis of variance 2IVTDMD = 48 h in vitro true dry matter digestibility, %. 3NDFD = NDF obtained after 48h of in vitro true dry matter digestibility, %.
111 Table 53 Effect of rust infestation an d inoculant application on chemical composition of corn silage (% of DM or as stated)1 Control Inoculant SEM2 Inoculant Rust Inoculant x Rust Rust level NR MR HR NR MR HR Linear Quadratic DM 38.0 40.6 58.3 37.2 39.6 56.5 0.29 <0.001 <0.001 0.24 <0.001 <0.001 CP 9.4 8.8 9. 6 9.3 8.8 9.6 0.11 0.82 0.01 0.96 0.05 <0.001 NDF 44.1 47.7 48.5 43.4 48.8 51.3 1.50 0.38 0.002 0.51 <0.001 0.30 ADF 23.1 25.1 25.3 25.3 25.4 30.7 1.21 0.02 0.02 0.13 0.006 0.42 Ash 4.7 4. 1 4.5 5.1 4.0 4.8 0.14 0.44 0.001 0.35 0.06 0.01 IVTDMD3 66.9 63.2 60.1 68.9 64.2 57.5 1.41 0.93 <0.001 0.25 <0.001 0.80 NDFD4 38.1 39.8 36.2 43. 4 45. 7 33. 0 1. 3 0.02 <0.001 0.004 0.001 0.001 1Control = no inoculant added, Inoculant = 1 x 105 cfu of P. pentosaceus and 4 x 105 of L. bu chneri NR = no rust level; MR = medium rust level; HR = high rust level. 2SEM = standard error of mean. 3IVTDMD = 48 h in vitro true dry matter digestibility, %. 4NDFD = NDF obtained after 48h of in vitro true dry matter digestibility, %.
112 Table 54 Ef fect of rust infestation and inoculant application on fermentative parameters, microbial counts, aerobic stability and mycotoxins in corn silage (% of DM or as stated)1 Control Inoculant SEM2 Inoculant Rust Inoculant x Rust Rust level NR MR HR NR MR H R Linear Quadratic pH 3.65 3.71 3.97 3.68 3.69 4.00 0.02 0.314 <0.001 0.340 <0.001 <0.001 NH3N 0.29 0.19 0.25 0.17 0.27 0.42 0.03 0.074 0.001 <0.001 0.002 0.034 NH3 % total N 3.05 2.18 2.55 1.87 2.98 4.24 0.28 0.07 0.007 <0.001 0.004 0.175 Lac tate 4.99 4.02 2.28 2.27 3.17 3.50 0.21 <0.001 0.006 <0.001 0.004 0.077 Total VFA 6.37 4.59 4.02 4.04 5.31 6.31 0.36 0.441 0.740 <0.001 0.916 0.449 Acetate 1.53 1.35 1.38 1.27 1.89 2.67 0.20 0.006 0.022 0.005 0.007 0.615 Propionate 1.16 0.90 0.58 0. 94 0.91 0.74 0.05 0.694 <0.001 0.004 <0.001 0.253 Butyrate 0.00 0.00 0.51 0.00 0.00 0.71 0.03 0.023 <0.001 0.009 <0.001 <0.001 DM loss, % 1.13 1.26 0.93 1.07 0.97 0.78 0.24 0.415 0.514 0.904 0.334 0.536 Molds, log cfu/g 5.24 4.96 3.40 4.93 5.20 0.95 0.64 0.127 <0.001 0.117 <0.001 0.018 Aerobic stability, h 26.0 27.5 44.0 27.5 23.8 77.3 3.77 0.05 <0.001 0.012 <0.001 0.005 1Control = no inoculant added, Inoculant = 1 x 105 cfu of P. pentosaceus and 4 x 105 of L. buchneri NR = no rust level; MR = m edium rust level; HR = high rust level. 2SEM = standard error of mean.
113 Table 55 Effect of rust severity and inoculant application on mycotoxin concentration of corn silage (mg/kg of DM)1 Items Control Inoculant SEM2 Inoculant Rust Inoculant x Rust NR MR HR NR MR HR Aflatoxin 0.00 0.0 5.20 0.00 0.0 0.00 1.50 0.109 0.109 0.109 Zearaleneone 0.646 0.0 0.0 0.471 0.0 0.0 0.080 0.303 <0.001 0.303 Deoxynivalenol 0.00 0.0 0.00 0.00 0.0 0.00 0.00 NA3 NA NA Fumonisin 0.00 0.0 0.00 0.00 0.0 0.00 0. 00 NA NA NA 1Control = no inoculant added, Inoculant = 1 x 105 cfu of P. pentosaceus and 4 x 105 of L. buchneri NR = no rust level; MR = medium rust level; HR = high rust level. 2SEM = standard error of mean. 3NA = Not applicable
114 A B C Figure 51 Corn plants affected by different severities of Southern Rust A) no rust, B) medium rust and C) high rust.
115 CHAPTER 6 EFFECT OF ADDING A M YCOTOXIN SEQUESTERING AGENT O N MILK AFLATOXIN M1 CONCENT RATION AND THE PERFO RMANCE AND IMMUNE RESPONSE OF DAIRY CATTLE FED AN AFLATOXI N B1 CONTAMINATED DIET Introduction Aflatoxins are secondary metabolites produced by Aspergillus parasiticus A. flavus and the rare A. nomius (Creppy, 2002). They occur in nature as the B1 (AFB1), B2, G1, G2, and M1 forms (Masoero et al., 2007). Aflatoxins can negatively affect animal health, performance and reproduction if consumed in sufficient quantities (Whitlow and Haggler 2005). The toxic and carcinogenic M1 form, which results from conversion of AFB1 by hepatic metabolism, ca n be secreted into milk (Masoero et al., 2007). Due to the high amount of milk and milk products consumed by humans, keeping the concentration of aflatoxin M1 (AFM1) in milk within safe levels is critical. To avoid the risk of aflatoxin ingestion and intox ication, agencies around the world have established acceptable limits for aflatoxin concentration in milk and feeds. In the US, the Food and Drug Administration (FDA) stipulated Action Levels for aflatoxin in raw milk and lactating cow feeds are 0.5 and 20 concentration of milk (Creppy, 2002). The goal of keeping feed and milk aflatoxin concentration below these limits can be difficult to achieve because mycotoxinproduc ing molds infect crops and grains before and after harvesting ( Garcia Lopez and Phillips ., 1999; Gonzales Pereyra et al., 2008 ; Richard et al., 200 9 ). Furthermore, damage from insects, hail, lodging, and diseases (Queiroz et al., 2009) can predispose plants to mycotoxin contamination. Many post harvest treatments are used for detoxifying feeds contaminated with mycotoxins including thermal inactivation, irradiation, fermentation, ammoniation, and
116 nixtamalization (Lopez Garcia and Phillips ., 1999). Most of these methods are costly, time consuming or partially effective (Kurtz et al., 2009), and most are impractical for detoxification of the large quantities of feed used by many US dairies. Studies have demonstrated that dietary addition of adsorbent clays i s a promising and effective way to prevent aflatoxin intoxication by livestock on farms (Stroud, 2006). However, few ofsuch studies have examined the effect of the clay dose on efficacy in dairy cows and fewer still have examined dose effects on markers of the immune response in dairy cows. The objectives of this study were to determine the effect of adding two doses of a montmorillonite hydrated sodium calcium aluminosilicate clay based mycotoxin adsorbent on milk aflatoxin M1 (AFM1) concentrations and the performance and immune response of dairy cows fed a diet contaminated with aflatoxin B1 (AFB1). Materials and Methods Cows, Treatments and Design Animals used in this study were cared for according to protocols approved by the University of Florida Institutional Animal Care and Use Committee Eight lactating Holstein cows in late lactation (295 45 DIM) were stratified by milk production and randomly assigned to one of four treatments arranged in a balanced, replicated 4 x 4 Latin square design. Cows wer e housed in an opensided, freestall barn bedded with sand and equipped with Calan gates (American Calan Inc.) for individual feeding and misters and fans to minimize heat stress. The following treatments were investigated: 1) Control diet (C), 2) Toxin diet (T) containing C and 75 g/kg of AFB1 in the TMR, 3) Low clay (LC) diet containing T and the adsorbent added at 0.2% of the TMR DM, and 4) High clay diet (HC) containing T and the adsorbent added at 1% of the TMR DM. The
117 adsorbent was Calibrin A from A mlan International, Chic ago, IL. Diets were formulated to meet or exceed the nutrient requirements of Holstein cows in late lactation producing 20 kg/d of milk (NRC, 2001). The ingredient and chemical compositions of the basal TMR fed to all cows are shown in Table 6 1. Cows were fed the toxin and adsorbent based on an estimated average DMI of 23 kg/d, resulting in a daily intake of 1725 g of AFB1 for T, LC, and HC treatments, and 46 and 230 g of adsorbent in the LC and HC diets, respectively. Dietary AFB1 was obtained from an Aspergillus parasiticus (NRRL 2999) culture at the University of Missouri Diagnostic Laboratory and it contained 640 mg/kg of AFB1, 22 mg/kg of aflatoxin B2, 333 mg/kg of aflatoxin G1, and 3 mg/kg of aflatoxin G2. Appropriate doses of the adsorbent were mixed into the TMR and fed daily to cows fed LC and HC diets. The AFB1 was added only to diets on d 6 to 9 of each period and d 10 to 12 and 1 to 5 of the current and subsequent periods were for clearance of the toxin from the cows mil k. Therefore, the toxin was dosed for 4 d and it cleared from the system over 8 d. The daily dose of AFB1 was divided into 2 portions and each was mixed with 20 mL of molasses and 400 g of corn silage to facilitate consumption. The mixture was fed to cows in a plastic container before the rest of the TMR was fed at the a.m. and p.m. feeding times (0700 and 1700 h). Intake of the TMR was restricted to 95% of that in a twoweek pretrial period when a common diet was fed to all cows. This feeding strategy ensured complete consumption of the clay adsorbent and AFB1 in the relevant diets and prevented contamination of equipment and the feed bunk with the toxin. Analytical Procedures Cows were milked twice a day at 0100 and 1300 h and milk weights were recorded. Two milk samples were collected from a.m. and p.m. milkings on d 5, 9, 10,
118 11, and 12 in each period. Samples were analyzed by Southeast Dairy labs (Belleview, FL) for fat, protein and somatic cell counts (SCC) using a Bentley 2000 Near Infrared Reflectanc e Spectrophotometer (Bentley Instruments Inc., Chaska, MN). Somatic cell scores were generated as described by Norman et al. (2000) for statistical analysis of SCC. Values for 3.5% fat correct milk yield were calculated according to the equation: [(0.4324 milk yield) + (16.218 milk fat yield)] (NRC, 2001). Milk AFM1 concentration was quantified using the radioimmunoassay test (CHARM II test, Charm sciences Inc., Maiden, MA) described by Diaz et al. ( 2004), which had been validated against High Performance Liquid Chromatography measurements in a ring test (Salter et al., 2006). Weights of feed offered and refused by each cow were recorded daily. Dietary ingredients (corn silage, alfalfa hay, and concentrate mix) were sampled representatively daily. Four composites of the daily samples in each period were subsampled, dried at 60C for 48 h in a forcedair oven, ground to pass the 1mm screen of a Wiley mill (A. H. Thomas, Philadelphia, PA), and analyzed for DM (105oC for 16 h) and ash (512oC for 8 h). Conc entrations of NDF and ADF were measured using the method of Van Soest et al. (1991) in an Ankom 200 Fiber Analyzer (Ankom Technologies, Macedon, NY). Nitrogen was determined by rapid combustion using a Macro elemental N analyzer (Vario MAX CN, model ID 25. 005003; Elementar, Hanau, Germany) and CP was calculated as N x 6.25. Blood samples (10 mL ) were collected at 5 and 9 d from the coccygeal vein into vacutainer tubes containing sodium heparin anticoagulant (Becton, Dickinson and Co., Franklin Lakes, NJ) stored on ice during transport, centrifuged at 2,500 g for 20 min at 4C to separate plasma, and stored at 20C until analyzed. Plasma haptoglobin concentrations were determined by measuring
119 haptoglobin/hemoglobin complexing based on differences in peroxidase activity (Makimura and Suzuki, 1982). Plasma ceruloplasmin oxidase activity was measured in duplicate samples using the colorimetric procedure described by Demetriou et al. (1974). Plasma fibrinogen concentrations were determined using a kit (Sigma procedure No. 880; Sigma Diagnostics, St. Louis, MO) as described by Arthington et al. (2003). Neutrophil phagocytic activity was measured by monitoring the uptake of E.coli particles labeled with a pH sensitive dye from pHrodo E. coli BioParticles Conjugate (Invitrogen Life Sciences, Carlsbad, CA). Briefly, 40 L of pHrodo E.coli were added to a 100 L aliquot of blood containing no more than 5 103 cell/ L The solution was incubated for 2 h at 37oC and cell membranes were disrupted by incubation in 2.5 m L of lysis buffer for 15 min at room temperature. The suspension was centrifuged at 2000 g for 5 min, and washed with fluorescenceactivated cell sorting (FACS) buffer before a final centrifugation at 2000 g for 5 min. The pellet was kept on ice and phagocytotic activity was measured using a flow cytometer with a 488 nm excitation wavelength (FACSort; Becton Dickinson, San Jose, CA). Neutrophil adhesion molecules, CD62 (Lselectin) integrin) were quantified by flow cytometry as described by Silvestre et al. (20 11). Complete blood counts (CBC) were conducted at the University of Florida Clinical Pathology Laboratory using a Bayer Advia 120 hematology analyzer (Bayer Avia 120 hematology analyzer; Bayer Diagnostics (Siemens), Deerfield, IL). Statistical Analysis The experiment consisted of a balanced, replicated Latin square design with 4 treatments and 4 periods. The MIXED procedure of SAS (Version 9.2 SAS Institute Inc., Ca ry, NC) was used to analyze the data. The model for analyzing the data included treatment, period, square, treatment x period, and treatment x square interactions, and
120 cow within square as the random effect. Data from blood and milk samples collected on d 5 were used as covariates for analyzing data obtained during the toxindosing period. Blood data were not normally distributed, therefore they were log transformed and analyzed with the GLIMMIX procedure of SAS. The PDIFF statement of SAS was used to compa re Least Squares means. Contrast statements were used to compare the Control treatment to each of the other treatments. Significance was declared at P < 0.05 and tendencies were declared if P > 0.05 < 0.10. Results and Discussion Dry matter intake and mi lk yield were not affected by dietary treatments ( P > 0.05) and averaged 19.2 kg/d (Table 62). Feeding diet T tended to reduce FCM yield (19.0 vs. 20.8 kg/d, P = 0.08). Kutz et al. (2009) detected no changes in DMI or milk yield 1/kg of TMR DM was fed to dairy cows in early to midlactation for 7 1 /kg of TMR DM for 11 days, but the toxin decreased DMI by 1.5 kg of DM/d compared to cows no supplemented with AFB1. Whitlow and Hagler (2005) suggested that diets this study also reduced 3.5 % FCM yield. Stroud (2006) noted that dosing cows with AFB1 Milk composition was reduced by feeding the toxin. Compared to diet C, diet T reduced milk protein concentration (3.28 vs. 3.36% P = 0.01) and milk fat yield (0.67 vs. 0.74 kg/d, P = 0.04). A previous study demonstrated binding of aflatoxin to some fractions of milk protein but no changes in milk composit ion were reported (Barbiroli et
121 al., 2007). Smith et al. (1994) did not detect changes in milk fat and protein percentages 1/kg of diet DM to goats. Kutz et al. (2009) reported that milk protein and fat percentage were unaffected when dairy cows were fed an aflatoxincontaminated diet. The reason why the toxin adversely affected milk quality in thi s study but not others may be partly due to the combination of different aflatoxins fed in the current study. Natural dietary mycotoxin contamination of diets typically results in more adverse effects on cows than purified mycotoxins because of synergistic effects of different toxins ( Applebaum et al., 1982). The similarity in milk quality and FCM yield measures among diets C, LC, and HC indicate that both doses of the adsorbent prevented adverse effects of the toxin on milk production and composition. Conc entrations of AFM1 in milk from cows fed diets T, LC, and HC (0.57, 0.64 and P < and concentrations were lower ( P < 0.05) in cows fed HC versus those receiving diet T or LC (Table 63). Relative to diet T, diet HC reduced the milk AFM1 concentration by 20% but diet LC did not reduce the value. Milk AFM1 concentrations were greater than were lower. Therefore, dietary addition of HC kept the milk safe and legal but addition of LC did not, indicating the ineffectiveness of the low dose of the adsorbent. Other studies have also reported that mycotoxin binders reduced the AFM1 concentration of mil k. Diaz et al. (2004) compared effects of adding activated carbon, esterified glucomannan, calcium bentonite and 3 sodium bentonite products at 1.2% of diet DM on concentrations of AFM1 1. Respective reductions in milk AFM1 concentrations were 59, 31, 65, 50 and 61% Kutz
122 et al. (2009) evaluated the effect of adding two hydratedsodium calcium aluminosilicates (Novasil plus, En gelhard Corp, New Jersey USA and Solis, Novus International Inc., St. Charles, MO) or an esterified glucomannan (MTB 100, Alltech, Nicholasville, KY) product at 0.5% of diet DM on concentrations of AFM1 in milk of cows 1 concentration by 45 and 48%; however, the esterified glucomannan caused a reduction of only 4%. The effects of the strategy used to administer the toxin and adsorbent on performance of cows and milk AFM1 concentration are unknown. Adding the toxin to ingredients fed before the rest of the TMR was offered achieved the objective of minimizing contamination of equipment and ensuring complete consumption of toxin, but may have limited binding of the toxin by the adsorbent if the outflow rate of the toxin from the rumen was rapid. A similar approach was used by Kutz et al. (2009) to ensure complete ingestion of the toxin and minimize equipment contamination. However, Diaz et al. ( 2004 ) and Stroud ( 2006) mixed the toxin and adsorbent with the rest of the ingredients in the TMR, which is a more practically relevant strategy. That these authors changed their dosing strategy reflects the importance of minimizing equipment contamination with the toxin and the associated health and safety risks. Excretion and transfer of aflatoxin into the milk were lower in cows fed HC than those fed LC but cows fed T had similar values to those fed LC or HC. Therefore, addition of either dose of the adsorbent did not reduce the transfer and excretion of the toxin in milk This may have been because the dosing method did not maximize binding of the toxin by the adsorbent. Nevertheless, the numerical trend for lower transfer and
123 excretion of the toxin for diet HC versus T or LC is consistent with the lower AFM1 concentration in milk of cows fed HC versus T or LC. The LC adsorbent was less effective than HC at preventing transfer and excretion of AFB1 to milk because adsorption of aflatoxin by aluminosilicate binders happens in a dosedependent manner (Sarr 199 2 ). Even though application rates of mycotoxin adsorbents to ruminant diets range from 0.5 to 1.2% of dietary DM ( Har vey et al., 1991 ; Stroud 2006 ; Kutz et al., 2009), low rates have not been effective consistently at mitigating effects of aflatoxin on the performance, health and milk AFM1 concentrations of dairy cows. The rate of transfer of dietary aflatoxin into milk was reported to be between 0 and 4% by Sieber and Blanc (1978, as described by V an Egmond 1989) and between 1 and 6% by the European Food Safety Authority (EFSA, 2004). Lafont et al. (1980) fed daily doses of 0.09, 0.18, 0.86, or 2.58 mg to cows in early and late lactation producing 20 L/d of milk on average and the res pective transfer rates were 0.78 and 0.22%. The transfer rates in late lactation and the milk production level are similar to those in this study, which also involved cows in late lactation. Frobish et al. (1986) reported that high producing cows have greater transfer rates than cows with m oderate to low milk production. Cows in this study and that of Lafont et al. (1980) also had lower transfer rates than the mean value of 2% reported for cows producing 33.8 kg/d in a similar study (Kutz et al., 2009). In fact, after reviewing 14 studies on effects of adsorbents on aflatoxin transfer, Stroud (2006) reported that the mean transfer rate was about 1% as in this study, when diets were dosed with up to 150
124 Treatment effects on immune responses are summarized in Table 6 4. Haptoglobin concentrations in plasma were greater ( P < 0.01) in cows fed diet T than other diets, whereas values for cows fed diets C, LC and HC did not differ. Stimulation of the acutephase response due to inflammatory stress is characterized by increased secretion of acutephase proteins such as ceruloplasmin and haptoglobin (Bertoni et al., 2008). Haptoglobin is often used in ruminants as a biomarker to identify immu nechallenged animals (Heegaard et al., 2000 ; Arthington et al., 2003). Hiss et al. (2004) demonstrated the sensitivity of haptoglobin to immune stressors by showing that blood haptoglobin concentration was increased 11.3 folds after 12 h of intramammary a dministration of lipopolysaccharides. That blood haptoglobin concentrations were elevated when diet T was fed but not when diets C, LC, or HC were fed indicates that both doses of the adsorbent prevented the increased innate immune inflammatory stress resp onse caused by the toxin. F ibrinogen and ceruloplasmin are also acute phase protein markers of the innate immune response, yet their concentrations did not differ among treatments. This is partly because these markers are often less sensitive and respond l ess consistently than haptoglobin to inflammatory stressors (Arthington et al., 2003). Furthermore, because concentrations of these proteins were only measured at a single time point, 96 h after toxin dosing was initiated, important trends in the early res ponse of ceruloplasmin and fibrinogen to the toxininduced stress may have been undetected. The fluorescence intensity observed when evaluating the activity of adhesion molecules indicates the presence of receptors for integrin or Lselectin on the leukocyte membrane ( Murphy et al., 2008). Both are adhesion molecules that facilitate
125 the interaction between leukocytes and the endothelium, which is needed to allow diapedesis, which is further cellular extravasation of leukocytes to the site of infection (Kenneth et al., 2008). In this study, feeding diet T instead of C tended to increase the integrin (131 versus 220, P = 0.1), which agrees with the haptoglobin response. Collectively, the response to haptog integrin indicate that the toxin increased inflammatory stress, but feeding the adsorbent prevented this problem. Dietary treatments did not affect the percentage of phagocytotic neutrophils or neutrophil phagocytosis (Table 64). The technique used to evaluate neutrophil phagocytosis is based on using E.coli cells stained with a pH sensitive die, which becomes fluorescent when bacterial cells are exposed to the acidic environment within the leukocyte phagosome. This technique is more convenien t, rapid, and less laborious than more traditional techniques (Silvestre et al., 2011). However, it does not involve using dihydrorhodamine 123 or other oxidase peroxidase markers, therefore it cannot account for oxidative burst, which is the capacity of n eutrophils to destroy the phagocytized bacteria. The pattern of clearance of AFM1 in the milk after toxin dosing was terminated on d 9 is shown in Figure 61. No AFM1 was detected in the milk of any cow by d 12. Clearance of the toxin within 3 d agrees wi th published data. Frobish et al. (1986) reported that after 3 to 4 d of AFB1 removal from diets, the AFM1 concentration in milk should be reduced to normal levels. The studies of Applebaum et al. (1982) and Diaz et al. (2004) also confirmed that AFM1 clea red from the milk of animals receiving an AFB1contaminated diet within 4 d. Masoero et al. (2007) detected that within 24 h of toxin withdrawal from the diet, milk AFM1 levels had dropped below the FDA legislative limit
126 (0.5 g/kg) and this decrease was s ignificant ( P < 0.05) within 48 h of withdrawing the toxin from the diet. In this study, no AFM1 was detected in the milk of cows fed diet HC 24 h after the toxin was withdrawn from the diet. Therefore, the high dose of the adsorbent kept the milk legal a nd safe when the toxin was included in the diet and cleared toxin residues from the milk within a shorter period than the low dose. Conclusions Feeding AFB1 tended to reduce FCM yield, reduced milk fat yield and milk protein concentration, increased the in nate immune response, and increased milk AFM1 concentration to levels that exceeded the FDA legislative limit. The low and high doses of the mycotoxin adsorbent prevented the adverse effects of the toxin on the immune response, milk quality, and FCM yield but only the high dose kept the AFM1 concentration of the milk below the FDA Action Level.
127 Table 6 1 Ingredient and chemical composition of the experimental diet Ingredient composition % of DM Corn silage 40.9 Alfalfa hay 8.05 Wet brewers grains 5. 56 Distillers grains 6.97 Dried citrus pulp 3.34 Ground c orn 18.2 SoyPlus 1 4.03 Soybean meal 5.26 Sugarcane molasses 3. 95 Mineral and vitamin mix 3.74 Chemical composition DM, % 46.7 Ash, % of DM 5.8 Crude protein, % of DM 15.5 Neutral det ergent fiber, % of DM 39.1 Acid detergent fiber, % of DM 20.4 Aflatoxin B1, g/kg ND2 Afltoxin B2, g/kg ND Aflatoxin G1, g/kg ND Aflatoxin G2, g/kg ND Deoxynivalenol, mg/kg ND T 2, mg/kg ND Zearalenone, mg/kg ND 1Mineral mix contained 23.7% CP 9.7% Ca, 8.0% Na, 6.7% K, 2.4% Mg, 0.4% S, 0.9% P, 2886 mg/kg of Mn, 3092 mg/kg of Zn, 886 mg/kg of Cu, 339 mg/kg of Fe, 31 mg/kg of Co, 30 mg/kg of I, 17.0 mg/kg of Se. 147,756 IU of vitamin A/kg, 787 IU of vitamin E/kg (DM basis). 2ND concentrations of respective toxin were bel ow lower detection limits (5 g/kg for aflatoxins, and 0.5 mg/kg for deoxynivalenol, T 2, and zearalenone.
128 Table 62. Effect of dietary addition of aflatoxin B1 (AFB1) with or without low (LC) or high (HC) doses of a mycotoxin bi nder1 on the performance of dairy cows Item Control AFB1 toxin LC+T HC+T SEM Contrast P values (C) (T) (LC) (HC) C vs. T C vs. LC C vs. HC DMI, kg/d 20.3 18.0 18.1 20.4 1.24 0.12 0.14 0.96 Milk yield, kg/d 19.5 18.9 19.9 19.1 1.11 0.44 0.60 0.60 3.5 % FCM, kg/d 20.8 19.0 20.5 19.4 0.79 0.08 0.80 0.12 Milk protein, % 3.36ab 3.28c 3.35 b 3.41a 0.095 0.01 0.72 0.08 Milk fat, % 3.75 3.78 3.68 3.69 0.180 0.77 0.55 0.61 Milk protein, kg/d 0.63 0.60 0.67 0.66 0.05 0.62 0.59 0.65 Milk fat, kg/d 0.74a 0.67b 0.73ab 0.69ab 0.03 0.04 0.73 0.09 SCC x 1000/ mL 272 147 194 260 112 0.25 0.35 0.73 a, b,c Means within rows with no common superscript differed significantly ( P < 0.05) 1Produced by Amlan International, Chicago, IL.
129 Table 63. Effect of dietary addition of aflatoxin B1 (AFB1) with or without low (LC) or high (HC) doses of a mycotoxin binder1 on the aflatoxin M1 (AFM1) concentration in the milk Item Control AFB1 toxin LC+T HC+T SEM Contrast P values (C) (T) (LC) (HC) C vs. T C vs. LC C vs. HC Con centration, g/kg 0c 0.57a 0.64a 0.46b 0.04 <0.001 <0.001 <0.001 Excretion, g/d 0c 10.1ab 12.9a 8.7b 1.55 <0.001 <0.001 <0.001 Transfer, % 0c 0.56ab 0.75a 0.50b 0.05 <0.001 <0.001 <0.001 Reduction, % 100a 0c 0.5c 20b 7.05 <0.001 <0.001 <0.001 Clearance rate, g//h 0c 0.011ab 0.013a 0.009b 0.0008 <0.001 <0.001 <0.001 a, b, c Means within rows with no common superscript differed significantly (P<0.05) 1Produced by Amlan International, Chicago, IL.
130 Table 64. Effect of dietary addition of aflatoxin B1 (AFB1) with or without low (LC) or high (HC) doses of a mycotoxin binder1 on markers of the innate immune response Item Control AFB 1 toxin LC+T HC+T SEM Contrast P values (C) (T) (LC) (HC) C vs T C vs LC C vs. HC Acute phase proteins Cerulo plasmin, mg/100m L 21.0 21.5 20.2 22.2 1.18 0.74 0.64 0.48 Haptoglobin, arbitrary unit1 14.4b 22.0a 14.8b 16.0b 1.98 0.01 0.87 0.56 Fibrinogen, mg/100m L 270 301 278 275 25 0.38 0.81 0.89 Median fluorescence intensity of neutrophils and neutrophil adhesio n molecules integrin (CD18) 131.1 219.7 154.7 138.2 32 0.10 0.55 0.85 L selectin (CD62) 822.4 951.6 977.8 912.5 129 0.42 0.29 0.52 Neutrophils 100 113 110 102 14 0.29 0.42 0.87 Neutrophil phagocytosis, % 82.6 79.6 82.2 83.0 3.0 0.45 0.91 0.92 a, b,c Means withi n rows with no common superscript differed significantly (P<0.05) 1Produced by Amlan International, Chicago, IL.
131 Figure 61. Effect of dietary addition of aflatoxin B1 with or without low (LC) or high (HC) doses of a mycotoxin binder on clearance o f aflatoxin M1 (AFM1) from the milk of dairy cows
132 CHAPTER 7 SUMMARY GENERAL CONCLUSIONS AND RECOMENDATIONS The main objective of this series of studies was to evaluate the effect of different inoculants on the fermentation, aerobic deterioration, DM and nutrient losses, pathogenicity, and toxicity of corn silages A n additional objective was to evaluate the efficacy of using a montmorillonitebased adsorbent to prevent adverse effects of dietary AFB1 on the performance and innate immune response of dair y cows and the transfer of dietary mycotoxins into milk. In order to achieve these objectives four studies were performed. The first study aimed to evaluate the effectiveness of three commercial bacterial inoculants at controlling E. coli O157:H7 in corn s ilages during ensiling and feedout phases of silage production. A second objective was to determine if the inoculants exhibited and transferred antibacterial activity against E. coli O157:H7 to the silages. Chopped corn forage was ensiled after treatment w ith:1) distilled water (negative Control); 2) 5 105 cfu/g of E. coli O157:H7 (EC, positive control); 3) EC and 1 106 cfu/g of Pediococcus pentosaceus and Propionibacterium freudenreichii (EC+BII); 4) EC and 1 106 cfu/g of Lactobacillus buchneri (EC +LB); 5) EC and 1 106 cfu/g of L. buchneri and P. pentosaceus (EC+B500). Escherichia. coli O157:H7 w as not detected in silages after their pH dropped below 4. Applying inoculants containing L. buchneri resulted in a more heterolactic fementation, which i ncreased acetic acid concentrations, reduced counts of spoilage fungi and thereby increased aerobic stability compared with control treatment. After 144 h of aerobic exposure only inoculants containing L. buchneri kept silage pH below 4 and thus prevented reestablishment of the pathogen. Applying EC+BII containing P. freudenreichii did not improve aerobic stability or inhibit the growth
133 of E.coli O157:H7. All pure cultures of bacterial inoculants exhibited antibacterial activity, independent of pH against E coli O157:H7, but this trend did not persist in inoculated silages, suggesting that E. coli elimination from silages was mediated by pH reduction. This experiment demonstrated that application of inoculants containing L. buchneri prevented establishment of E.coli O157:H7 on silage by keeping the pH low during the aerobic exposure phase. This effect was attributed to the antimycotic effect of the acetic acid produced by L. buchneri, which prevented the growth of yeasts that initiate spoilage by increasing the pH during aerobic exposure by metabolizing lactic acid. This suggests that application of L. buchneri inoculants at ensiling may prevent establishment of E. coli O157:H7 on silages during the feedout stage. Therefore, such inoculants may help to limit or inhibit the growth of E. coli O157:H7 in a TMR containing a mixture of ingredients contaminated with the pathogen and L. buchneri treated silage. This may help to prevent the spread of the pathogen among dairy cows, which are known as the primary reservoir. Future research should validate these suggestions and examine if L. buchneri inoculants can increase E. coli shedding in cows and inhibit the growth of other pathogens that require neutral pH to thrive such as Listeria monocytogenes, Bacillus aureus Clostridia spp., Klebsiella spp., and other Enterobacteria. Future work should also determine if inoculants can inhibit the growth of E. coli O157:H7 on silages that do not ferment as readily as corn silage because of high buffering capacities or low sug ar concentrations such as legume silages and t ropical grasses, respectively. Due to the discovery of antibacterial activity against the pathogen
134 by pure cultures of all inoculants, future work should also identify the source of such activity and develop methods of ensuring that it persists during ensiling. The objective of the second study was to evaluate the effects of a dual purpose inoculant, which was the most promising treatment in the previous study, on the fermentation, quality and aerobic stability of corn silages made in farm scale silos. This trial was conducted because little information exists on effects of using dual purpose inoculants to preserve silages in farm scale silos. Corn forage was harvested at 34% DM and treated without (Control) or 6 cfu/g of Lactobacillus buchneri and Pediococcus pentosaceus Forty five metric tons of corn forage were packed alternately into each of four replicate silo bags per treatment and ensiled for 166 days. During the feedout, silage was removed f rom the bags, separated into good and spoiled (darkened, moldy, slimy, or heating) silage, and weighed daily for 35 d. Weekly samples were analyzed for chemical composition, aerobic stability and fungal counts. Inoculation did not affect the chemical composition of the spoiled or good silage but decreased the quantity and percentage of spoiled silage in the bags by over 50%, which resulted in a decrease of associated nutrient and energy losses. Inoculated silages tended to have a more heterolactic fermentat ion and therefore had lower yeast counts. However, aerobic stability was not different across treatments. The main deduction from this experiment was that treatment with the dual purpose inoculant markedly reduced nutrient and energy losses in the silages by inhibiting the growth of spoilage fungi and thereby curtailing the amount of spoiled silage. To our knowledge this is the first time the latter has been shown in farm scale silos. These reductions in losses would directly improve profitability on dairi es and increase
135 environmental stewardship and food safety by reducing the amount of discarded and toxigenic or pathogenic silage, respectively. That most measures of temperature were numerically but not statistically reduced by inoculation suggests that there may have been insufficient power to detect differences between treatments in aerobic stability or that the inoculant could not increase this measure under the relatively low packing density used and the prevailing hot, humid conditions, which predispos e to the growth of spoilage organisms. Therefore, this study should be repeated with more replicates to conclusively indicate if the aerobic stability of silages made in farm scale silos can be improved by inoculation. Plants grown under environmental str ess are more susceptible to the growth of opportunistic epiphytic fungi that produce mycotoxins and cause undesirable fermentations. The third experiment aimed to evaluate 1) if increasing levels of southern rust infestation on corn plants reduces the nutr itive value and safety of corn silage and 2) if application of the dual purpose bacterial inoculant used in previous experiments could mitigate adverse effects of rust infestation on silage quality. Corn plants with three levels of rust infestation, none ( NR), medium (half of the plant was infested, MR) and high rust (whole plant was infested, HR) were harvested at 40% DM. The forages were chopped and ensiled without (Control) or with application of a bacterial inoculant (INO) at a rate that delivered 1 105 cfu/g of Pediococcus pentosaceus 12455 and 4 105 cfu/g of Lactobacillus buchneri 40788. Each treatment was ensiled in quadruplicate 20 L laboratory silos for 97 days. As the level of rust infestation increased, concentrations of silage DM and NDF incr eased, whereas in vitro DM digestibility decreased by 9%. The NDF digestibility of NR and MR silages were similar among untreated silages or
136 inoculated silages but corresponding values for HR silages were lower particularly in inoculated silages. Concentra tions of lactate decreased with increasing rust infestation in Control silages, but this trend was reversed in inoculated silages. Mold counts of NR and MR silages were similar among Control or inoculated silages; whereas corresponding values in HR silages were lower particularly in inoculated silages. Consequently, aerobic stability was greater in HR silages than NR or MR silages among Control or inoculated silages. Aflatoxin was only detected in Control HR silages and the level (5.2 mg/kg) exceeded the Fo od and Drug Administration (FDA) Action Level for feeds. Absence of the toxin in inoculated HR silages suggested that inoculation prevented accumulation of the toxin. This study demonstrated for the first time that rust infestation severely reduced the nutritive value and fermentation of corn silage but dual purpose inoculants could mitigate or reduce such adverse effects. Further, high levels of rust infestation resulted in higher levels of aflatoxin in uninoculated corn silages than the FDA Action level, but inoculated silages were free of the toxin. Therefore, such inoculants can be used to improve the fermentation, nutritive value, and sa fety of rust infested silages. These benefits are probably attributable to the antimycotic effects of the acetic acid produced by L. buchneri which prevented the growth of aflatoxinproducing Aspergillus molds. Future studies are needed to confirm the latter, and to determine if the mode of action is acetate induced inhibition or binding of aflatoxins to cell walls of b acteria in the inoculant. Additional studies should examine if such inoculants can be used to prevent adverse effects of other diseases and plant stressors such as insect damage and lodging on the quality and safety of corn silage.
137 The occurrence of mycot oxins in the silage in the previous study and the high corn silage intake by dairy cows in the USA highlighted the need to examine the efficacy of using adsorbents to minimize adverse effects of mycotoxins on dairy cows. The fourth trial aimed to examine t he effects of adding two doses of a mycotoxin adsorbent on milk aflatoxin M1 (AFM1) concentration and the performance and immune response of dairy cows fed a diet contaminated with aflatoxin B1 (AFB1). Eight multiparous lactating cows were used in an exper iment with a duplicated 4 x 4 Latin square design with 12d periods. Treatments included the following: 1) Control diet (C); 2) Toxin diet (T) containing C and 75 g/kg of AFB1; 3) Low clay (LC) diet containing T and 0.2% of a clay based aflatoxin binder ( Calibrin A Amlan International, Chicago, IL); and 4) Highclay diet (HC) containing T and 1% of the binder. In each period, the toxin was dosed on d 6 to 9, whereas the clay was fed every day. Milk production and DMI were recorded daily and milk was sampl ed twice daily on d 5, 9, 10, 11, and 12 in each period. Blood samples were collected on days 5 and 9 of each period. Dietary treatments did not affect DMI, milk yield, or feed efficiency (kg FCM /kg DMI). Feeding T instead of C tended to reduce 3.5% FCM y ield, reduced milk fat yield and milk protein concentration. Concentrations of AFM1 in milk of cows fed the T and LC diets were similar and greater than those of cows fed the HC or C diets. Haptoglobin concentration in plasma was greater and integrin ex pression tended to be greater in cows fed diet T instead of C, but values for cows fed LC, HC and C did not differ. Feeding diets T and LC resulted in greater milk AFM1 concentrations than the FDA Action Level but feeding diets HC and C did not. Therefore, diet T made the milk unsafe, increased the innate immune response, tended to reduce FCM yield, and reduced milk quality. Feeding the HC or LC diets
138 prevented adverse effects of the toxin on the innate immune response and FCM yield and milk quality, but on ly the HC diet kept milk AFM1 concentrations below the unsafe threshold. This trial demonstrated that dietary addition of enterosorbents at an appropriate dose is an effective strategy to detoxify diets or silages contaminated with aflatoxin and thereby ensure the safety of milk from cows fed such diets or silages. The study also showed that enterosorbents can also reduce adverse effects of aflatoxin ingestion on milk yield and quality and they can prevent the energetically expensive immune response caused by the toxin. Future studies should compare different types of enterosorbents, determine optimum dose rates for each one, and examine long term effects of their inclusion in diets on the performance, health, and reproduction of dairy cows. The main concl usions from this series of studies are that in addition to traditionally known effects such as improving the fermentation, DM and nutrient recovery, and aerobic stability of silages, inoculants can be used strategically to mitigate the pathogenicity and toxicity of silages, and thereby prevent the spread of pathogens. These benefits provide additional justification for inoculant use for forage conservation that will improve profitability and environmental stewardship on dairies, enhance the safety of dairy products, and improve herd and human health.
139 LIST OF REFERENCES Adesogan, A. T., M. B. Salawu, A. B. Ross, D. R. Davies, and A. E. Brooks. 2003. Effect of Lactobacillus buchneri L. fermentum or Leuconostoc mesenteroides inoculants or a chemical additive on the fermentation, aerobic stability and nutritive value of crimped wheat grains. J. Dairy Sci. 86:1789 1796. Adesogan, A. T., N. A. Krueger, M. B. Salawu, D. B. Dean, and C. R. Staples. 2004. The influence of treatment with dual purpose inoculants or s oluble carbohydrates on the fermentation and aerobic stability of bermudagrass. J. Dairy Sci. 87:3407 3416. Adesogan, A. T., and O. C. M. Queiroz. 2009. Silage pathogenicity and implications for the ruminant production chain. Pages 225 241 in Proceedings o f the International Symposium on Forage Quality and Conservation. M. Zopollato, G. B. Muraro and L. G. Nussio, FEALQ, Piracicaba, SP. Ankolekar, C., T. Rahmati, and R. G. Labbe. 2009. Detection of toxigenic Bacillus cereus and Bacillus thuringiensis spores in US rice. Intl. J. Fd. Microbiol. 128:460 466. ANVISA. 2002. Regulamento tcnico mercosul sobre limites mximos de aflatoxinas admissveis no leite, no amendoim, no milho. Accessed May 20, 2011. http://www.anvisa.gov.br/legis/resol/2002/274_02rdc.htm A.O.A.C., Association of Official Analytical Chemists 1985. Official Methods of Analysis 14th ed. Association of Official Analytical Chemists, Washington, DC, USA. ALPHA. American Public Health Association 1985. Standard Methods for the Exami nation of Water and Wastewater. 16th ed. American Public Health Association, Washington, DC, USA. Applebaum, R. S., R. E. Brackett, D. W. Wiseman, and E. L. Marth. 1982. Responses of dairy cows to dietary aflatoxin: feed intake and yields, toxin content, and quality of milk of cows treated with pure and impure aflatoxin. J. Dairy Sci. 65:1503 1508. Archibald, F. S., and I. Fridovich. 1981. Manganese and defenses against oxygen toxicity in Lactobacillus plantarum J. Bacteriol. 145:442 451. Armstrong, G. L., J. Hollingsworth, and J. G. Morris. 1996. Emerging foodborne pathogens: Escherichia coli O157:H7 as a model of entry of a new pathogen into the food supply of the developed world. Epidemiol. Rev. 18: 29 51. Arriola, K. G., S. C. Kim, and A. T. Adesogan. 2011. Effect of applying inoculants with heterolactic or homolactic and heterolactic bacteria on the fermentation and quality of corn silage. J. Dairy Sci. 94:1511 1516.
140 Arthington, J. D., S. D. Eicher, W. E. Kunkle, and F. G. Martin. 2003. Effect of transportation and commingling on the acute phase protein response, growth, and feed intake of newly weaned calves. J. Anim. Sci. 81:1120 1125. Ashbell, G., Z. G. Weinberg, Y. Hen, and I. Filya. 2002. The ef fects of temperature on the aerobic stability of wheat and corn silages. J. Ind. Microbiol. Biotechnol. 28:261 263. Avantaggiato, G., M. Solfrizzo and A. Visconti. 2005. Recent advances in the use of adsorbent materials for detoxification of Fusarium mycot oxins. Food Addit. Contam. 22:379 388. Avila, C. L. S., J. C. Pinto, H. C. P. Figueiredo, and R. F. Schwan. 2009. Effects of an indigenous and a commercial Lactobacillus buchneri strain on quality of sugar cane silage. Grass and For. Sci. 64:384 394. Axels son, L. 1998. Lactic acid bacteria: classification and physiology. Pages 1 72 in: Lactic Acid Bacteria: Microbiology and Functional Aspects, 2nd ed. S. Salminen and A. Von Wright, Marcel Dekker Inc, New York. Bach, S. J., T. A. McAllister, J. Baah, L. J. Y anke, D. M. Veira, V. P. J. Gannon, and R. A. Holley. 2002. Persistence of Escherichia coli O157:H7 in barley silage: effect of a bacterial inoculant. J. Appl. Microbiol. 93:288 294. Barbiroli, A., F. Bonomi, S. Benedetti, S. Mannino, L. Monti, T. Cattaneo, and S. Lametti. 2007. Binding of aflatoxin M1 to different protein fractions in bovine and caprine milk. J. Dairy Sci. 90:532 540. Barry, T. N., J. E. Cook, and R. J. Wilkins. 1978. The influence of formic acid and formaldehyde additives and type of harv esting machine on the utilization of nitrogen in lucerne silages: 1. The voluntary intake and nitrogen retention of young sheep consuming the silages with and without intraperitoneal supplements of DL methionine. J. Agric. Sci. 91:701 715. Benjamin, M. M., and A. R. Datta. 1995. Acid tolerance of enterohemorrhagic Escherichia coli. Appl. Environ. Microbiol. 61:1669 1672. Bertoni, G., E. Trevisi, X. Han, and M. Bionaz. 2008. Effects of inflammatory conditions on liver activity in puerperium period and conseq uences for performance in dairy cows. J Dairy Sci. 91:3300 3310. Binder, E. M. 2007. Managing the risk of mycotoxins in modern feed production. Anim. Fd. Sci. 133:149 166 Bolsen, K. K., and R. Bolsen. 2006. A trouble shooter for common silage problems. Pr oc. Southwest Nutrition and Management Conference, Temple, AZ.
141 Borreani, G., and Tabacco, E. 2008. Low permeability to oxygen of a new barrier film prevents outgrowth of butyric acid bacteria in farm corn silage. J Dairy Sci. 91:4272 4281. Bursian, S. J., R. J. Aulerich, J. K. Cameron, N. K. Ames, and B. A. Steficek. 1992. Efficacy of hydrated sodium calcium aluminosilicate in reducing the toxicity of dietary zearalenone to mink. J. Appl. Toxicol. 12:85 90. Byrne, C. M., P. O'Kiely, D. J. Bolton, J. J. Sher idan, D. A. McDowell, and I. S. Blair. 2002. Fate of Escherichia coli O157:H7 during silage fermentation. J. Food Prot 65:1854 1860. Cai, Y. 1999. Identification and characterization of Enterococcus species isolated from forage crops and their influence o n silage fermentation. J. Dairy Sci. 82:2466 2471. Cai, Y., Y. Benno, M. Ogawa, and S. Kumai. 1999. Effect of applying lactic acid bacteria isolated from forage crops on fermentation characteristics and aerobic deterioration of silage. J. Dairy Sci. 82:520 526. CAST, Council for Agricultural Science and Technology. 1989. Mycotoxins: economics and health risks. Task Force Report No. 116. Ames, Iowa. CAST, Council for Agricultural Science and Technology. 2003. Mycotoxins: risk in plant, animal and human syst ems. Task Force Report No. 139. Ames, Iowa. Chamberlain, D. D., and J. Quig. 1987. The effect of the rate of addition of formic acid and sulphuric acid on the ensilage of perennial ryegrass in laboratory silos. J. Sci. Food Agric. 38:217 228. Chapman, P. A., C. A. Siddons, A. T. Cerdan Malo, and M. A. Harkin. 1997. A 1 year study of Escherichia coli O157 in cattle, sheep, pigs and poultry. Epidemiol. Infect. 119:245 250. Charmley, E., H. L. Trenholm, B. K. Thompson, D. Vudathala, J. W. G. Nicholson, D. B. Prelusky, and L. L. Charmley. 1993. Influence of levels of deoxynivalenol in the diet of dairy cows on feed intake, milk production, and its composition. J. Dairy Sci. 76: 3580 3587. Chen, Y., S. Sela, M. Gamburg, R. Pinto, and Z. G. Weinberg. 2005. Fate of Escherichia coli during ensiling of wheat and corn. Appl. Env. Microbiol. 71:5163 5170. Christiansson, A., J. Bertilsson, and B. Svensson. 1999. Bacillus cereus spores in raw milk: Factors affecting the contamination of milk during the grazing period. J. Dairy Sci. 82:305 314.
142 Cocconcelli, P. S., E. Triban, M. Basso, and V. Bottazzl. 1991. Use of DNA probes in the study of silage colonization by Lactobacillus and Pediococcus strains. J. Appl. Microbiol. 71:296 301. Cocolin, L., N. Innocente, M. Biasutti, and G. Comi. 2004. The late blowing in cheese: a new molecular approach based on PCR and DGGE to study the microbial ecology of the alteration process. Intl. J. Fd. Microbiol. 90:83 91. Codon, S. 1987. Responses of lactic acid bacteria to oxygen. FEMS Micr obiol. Rev. 46:269 280. Conaghan, P., P. OKiely, and F. P. OMara. 2010. Conservation characteristics of wilted perennial ryegrass silage made using biological or chemical additives. J. Dairy Sci. 93:628 643. Coulombe Jr, R. A. 1993. Biological action of mycotoxins. J. Dairy Sci. 76:880 891. Creppy, E. E. 2002. Update of survey, regulation and toxic effects of mycotoxins in Europe. Toxicol. Lett. 127:19 28 Dain, J. A., A. L. Neal, and R.W. Dougherty. 1955. The occurrence of histamine and tyramine in rumen ingesta of experimentally overfed sheep. J. Anim. Sci. 14:930 935. Dnicke, S., K. Matthus, P. Lebzien, H. Valenta, K. Stemme, K. H. Ueberschr, E. Razzazi Fazeli, J. Bhm, and G. Flachowsky. 2005. Effects of Fusarium toxin contaminated wheat grain on nut rient turnover, microbial protein synthesis and metabolism of deoxynivalenol and zearalenone in the rumen of dairy cows. J Anim. Physiol. Anim. Nutr. 89:303 315. Danner, H., M. Holzer, E. Mayrhuber, and R. Braun. 2003. Acetic acid increases stability of s ilage under aerobic conditions. Appl. Environ. Microbiol. 69:562 567. Demetriou, J. A., P. A. Drewes, and J. B. Gin. 1974. Ceruloplasmin. Pages 857 865 in Clinical Chemistry. C. Cannon and J. W. Winkelman, ed. Harper and Row, Hagerstown, MD. Diaz, D. E., B A. Hopkins, L. M. Leonard, W. M. Hagler, Jr, and L. W. Whitlow. 2000. Effect of fumonisin on lactating dairy cattle. J. Dairy Sci. 83:1171. (Abstr.) Diaz, D. E., W. M. Hagler, J. T. Blackwelder, J. A. Eve, B. A. Hopkins, K. L. Anderson, F. T. Jones, and L. W. Whitlow. 2004. Aflatoxin binders II: Reduction of aflatoxin M1 in milk by sequestering agents of cows consuming aflatoxin in feed. Mycopathologia 157:233 241. Diaz, G.J., and H.J. Boermans. 1994. Fumonisin toxicosis in domestic animals: A review. Vet Human Toxicol. 36:548 555.
143 DiCostanzo, A., L. Johnston, H. Windels and M. Murphy. 1995. A review of the effects of molds and mycotoxins in ruminants. Prof. Anim. Scient. 12:138 150. Diekman, M. A., and M. L. Green. 1992. Mycotoxins and reproduction in domestic livestock. J. Anim. Sci. 70:1615 1627 Dimmock, J. P. R. E and M. J. Gooding. 2002. The influence of foliar diseases, and their control by fungicides on the protein concentration in wheat grain: a review. J. Agric. Sci. 138:349 366. Driehuis, F., S. J. W. H. Oude Elferink, and S. F. Spoelstra. 1999. Anaerobic lactic acid degradation during ensilage of whole crop maize inoculated with Lactobacillus buchneri inhibits yeast growth and improves aerobic stability. J. Appl. Microbiol. 87:583 594. Driehui s, F. and S. Elferink. 2000. The impact of the quality of silage on animal health and food safety: A review. Vet. Q. 22:212 216. Driehuis, F., S. J. W. H. Oude Elferink, and P. G. Van Wikselaar. 2001. Fermentation characteristics and aerobic stability of g rass silage inoculated with Lactobacillus buchneri with or without homofermentative lactic acid bacteria. Grass For. Sci. 56:330 343. Driehuis, F., M. C. Spanjer, J. M. Scholten, and M. C. te Giffel. 2008. Occurrence of mycotoxins in feedstuffs of dairy c ows and estimation of total dietary intakes. J. Dairy Sci. 91:4261 4271. Dulphy, J. P., M. Van Os. 1996. Control of voluntary intake of precision chopped silages by ruminants: a review. Repr. Nutr. Dev. 36:113 135. EFSA, European Food Safety Authority 20 04. Opinion of the scientific panel on contaminants in food chain on a request from the commission related to aflatoxin B1 as undesirable substances in animal feed. European Food Safety Authority, EFSA Q 2003 0.35. European Commission. 2006. Commission rec ommendation of 17 August 2006 on the presence of deoxynivalenol, zearalenone, ochratoxin A, T 2 and HT 2 and fumonisins in products intended for animal feeding. Off. J. Eur. Union 229:7 8. FDA. 2010. Guidance for Industry and FDA: Advisory Levels for Deoxy nivalenol (DON) in Finished Wheat Products for Human Consumption and Grains and Grain By Products used for Animal Feed. Accessed May 20, 2011. www.fda.gov/Food/GuidanceComplianceRegulatoryInformation/GudianceDocume nts/NaturalToxins/ucm120184.htm Fenlon, D. R., and J. Wilson. 2000. Growth of Escherichia coli O157 in poorly fermented laboratory silage: a possible environmental dimension in the epidemiology of E. coli O157. Lett. Appl. Microbiol. 30:118 121.
144 Filya, I., G. Ashbell, Y. Hen, and Z. G. Weinberg. 2000. The effect of bacterial inoculants on the fermentation and aerobic stability of whole crop wheat silage. Anim. Fd. Sci. Technol. 88:39 46. Filya, I. 2003. The effect of Lactobacillus buchneri and Lactobacillus plantarum on the fermentation, aerobic stability, and ruminal degradability of low dry matter corn and sorghum silages. J. Dairy Sci. 86:3575 3581. Filya, I., E. Sucu, and A. Karabulut. 2006. The effects of Propionibacterium acidipropionici and Lactobacillus plantarum applied at ensiling, on the fermentation and aerobic stability of low dry matter corn and sorghum silages. J. Ind. Micro. Biotech. 33: 353 358. Filya, I., R. E. Muck, and F. E. Contreras Govea. 2007. Inoculant effects on alfalfa silage: Fermentation products and nutritive value. J. Dairy Sci. 90:5108 5114. Fink Gremmels, J. 2008. Mycotoxins in cattle feeds and carry over to dairy milk: A review. Food Addit. Contam. 25:172 180. Frobish, R. A., D. Bradley, D. D. Wagner, P. E. Long Bradley, and H. Hairston.1986. Aflatoxin residues in milk of dairy cows after ingestion of naturally contaminated grain. J. Food prot. 49:781 785. Garcia Lopez, R., D. L. Park, and T. D. Phillips. 1999. Integrated Mycotoxin Management System. Pages 1 15 in Third Joint FAO/WHO/UNEP International Conference of Mycotoxins, Tunis, Tunisia. Garon, D., E. Richard, L. Sage, V. Bouchart, D. Pottier, and P. Lebailly. 2006. Mycoflora and multimy cotoxin detection in corn silage: experimental study. J. Agric. Food Chem. 54: 3479 3484. Garret, W. N., H. Heitman, Jr., and A. N. Booth.1968. Aflatoxin toxicity in beef cattle. Proc. Soc. Exp. Biol. Med. 127:188 190. Garvie, E. I. 1986. Genus Pediococcus. Pages 1075 1079 in Bergey's Manual of Systematic Bacteriology. Vol. 2, 9th ed. Williams and Wilkins, Baltimore, MD. Gollop, N., V. Zakin, and Z. G. Weinberg. 2005. Antibacterial activity of lactic acid bacteria included in inoculants for silage and in sil ages treated with these inoculants. J. Appl. Microbiol. 98:662 666. Gonzalez Pereyra, M. L., V. A. Alonso, R. Sager, M. B. Morlaco, C. E. Magnoli, A. L. Astoreca, C. A. R. Rosa, S. M. Chiacchiera, M. Dalcero, and L. R. Cavaglieri. 2008. Fungi and selected mycotoxins from pre and postfermented corn silage. J. Appl. Microbiol. 104:1034 1041.
145 Goodman, S. A., P. J. Warner, and C. Orr. 1995. A novel bacterial inoculant to reduce aerobic deterioration of silages. Page 452 458 in Proc. Alltechs Annual Symposiu m, 11th. Nottingham University press, UK. Greenhill, W. L. 1964. Plant juice in relation to silage fermentation. II factors affecting the release of juice. J. Brit. Grassl. Soc. 19:231 236. Guthrie, L.D., and D. M. Bedell. 1979. Effects of aflatoxin in cor n on production and reproduction in dairy cattle. Proc. Ann. Mtg. US Anim. Health Assoc. 83:202 204 Hammerschmidt, R. 1984. Rapid deposition of lignin in potato tuber tissue as a response to non pathogenic fungi on potato. Physiol. Plant Pathol. 24:33 42. Harvey, R. B., T. D. Phillips, J. A. Ellis, L. F. Kubena, W. E. Huff, H. D. Petersen. 1991. Effects on aflatoxin M1 residues in milk by addition of hydrated sodium calcium aluminosilicate to aflatoxin contaminated diets of dairy cows. Am. J. Vet. Res. 52: 1556 1559. Heegaard, P. M. H., D. L. Godson, M. J. M. Toussaint, K. Tjrnehj, L. E. Larsen, B. Viuff, and L. Rnsholt. 2000. The acute phase response of haptoglobin and serum amyloid A (SAA) in cattle undergoing experimental infection with bovine respirat ory syncytial virus. Vet. Immunol. Immunopathol. 77 :151 159. Henderson, A. R., J. M. Ewart, and G. M. Robertson. 1979. Studies on the aerobic stability of commercial silages. J. Sci. Food Agric. 30:223 228. Hiss, S., M. Mielenz, R. M. Bruckmaier, and H. Sa uerwein. 2004. Haptoglobin concentrations in blood and milk after endotoxin challenge and quantification of mammary Hp mRNA expression. J. Dairy Sci. 87:3778 3784. Holden, A. N. G. 1989. Some effects of silage inoculants on aerobic stability of grass silag e. Dissertation Abst. Int., Sci. Engr. 49:41 10, 4150B. Holmquist, G. U., H. W. Walker, and H. M. Stahr. 1983. Influence of temperature, pH, water activity and antifungal agents on growth of Aspergillus flavus and A parasiticus J. Food Sci. 48:778 782. Ho lzer, M., E. Mayrhuber, H. Danner, and R. Braun. 2003. The role of Lactobacillus buchneri in forage preservation. Trends Biotechnol. 21:282 287. Hui, Y. H.; F. Sherkat. 2006. Handbook of food science, technology, and engineering. Vol. III. 6th. Ed. CRC pre ss., Melbourne, Australia. Huisden, C. M., A. T. Adesogan, S. C. Kim, and T. Ososanya. 2009. Effect of applying molasses or inoculants containing homofermentative or heterofermentative bacteria at two rates on the fermentation and aerobic stability of cor n silage. J. Dairy Sci. 92:690 697.
146 Jack, R. W., J. R. Tagg, and B. Ray. 1995. Bacteriocins of gram positive bacteria. Microbiol. Rev. 59:171 200. Johnson, J. C., R. N. Gates, G. L. Newton, J. P. Wilson, L. D. Chandler, and P. R. Utley. 1997. Yield, compos ition, and in vitro digestibility of temperate and tropical corn hybrids grown as silage crops planted in summer. J. Dairy Sci. 80: 550 557. Jonsson, A., and G. Pahlow. 1984. Systematic classification and biochemical characterization of yeasts growing in gr ass silage inoculated with Lactobacillus cultures Anim. Res. Develop. 20:7 22. Jung, J. 1972. The action of various silage additives. Wirtschaftseigene Futter. 18: 178 185. Kaiser, A., and M. J. Evans. 1997. Forage conservation on Australian dairy farms. A nimal Industry Rep. 3. NSW Agriculture, Wagga Wagga, Australia. Kang, T. W., A. T. Adesogan, S. C. Kim, and S. S. Lee. 2009. Effects of an esterase producing inoculant on fermentation, aerobic stability, and neutral detergent fiber digestibility of corn s ilage. J. Dairy Sci. 92:732 738. Keady, T. W. J. and J. J. Murphy. 1996. Effects of inoculant treatment on ryegrass silage fermentation, digestibility, rumen fermentation, intake and performance of lactating dairy cattle. Grass and For. Sci. 51:232 241. Ke rr L. A., Groce A. W., Kersting K. W. 1987. Ammoniated forage toxicosis in calves. J. Am. Vet. Med. Assoc. 191: 551 552. Kendall, P. 2006. Food Safety: Botulism. Accessed Feb. 20, 2011. http://www.ext.colostate.ed u Kennedy D. G., S. A. Hewitt, J.D. McEvoy, J. W. Currie, A. Cannavan, W. J. Blanchflower, and C. T. Elliot.1998. Zeralenol is formed from Fusarium spp. Toxins in cattle in vivo. Food Addit. Contam. 15:393 400. Kiessling K. H., H. Pettersson, K. Sandholm, and M. Olsen. 1984. Metabolism of aflatoxin, ochratoxin, zearalenone, and trichothecenes by intact rumen fluid, rumen protozoa, and rumen bacteria. Appl. Environ. Microbiol. 47:1070 1073. Kim, S. C., and A.T. Adesogan. 2006. Influence of ensiling temperat ure, simulated rainfall, and delayed sealing on fermentation characteristics and aerobic stability of corn silage. J. Dairy Sci. 89: 3122 3132 Kizilsimsek, M., R. J. Schmidt, and L. Kung. 2007. Effects of a mixture of lactic acid bacteria applied as a freeze dried or fresh culture on the fermentation of alfalfa silage. J. Dairy Sci. 90:5698 5705.
147 Kleinschmit, D. H., R. J. Schmidt, and L Kung. 2005. The effects of various antifungal additives on the fermentation and aerobic stability of corn silage. J. Dair y Sci. 88:2130 2139. Kleinschmit, D. H., and L. Kung. 2006a. The effects of Lactobacillus buchneri 40788 and pediococcus pentosaceus r1094 on the fermentation of corn silage. J. Dairy Sci. 89:3999 4004. Kleinschmit, D. H., and L. Kung. 2006b. A meta analy sis of the effects of Lactobacillus buchneri on the fermentation and aerobic stability of corn and grass and small grain silages. J. Dairy Sci. 89:4005 4013. Kristensen, N. B., K. H. Sloth, O. Hojberg, N. H. Spliid, C. Jensen, and R. Thogersen. 2010. Effec ts of microbial inoculants on corn silage fermentation, microbial contents, aerobic stability, and milk production under field conditions. J. Dairy Sci. 93:3764 3774. Krooneman, J., F. Faber, A. C. Alderkamp, S. J. H. W. Oude Elferink, F. Driehuis, I. Clee nwerck, J. Swings, J. C. Gottschal, and M. Vancanneyt. 2002. Lactobacillus diolivorans sp. nov., a 1,2 propanediol degrading bacterium isolated from aerobically stable maize silage. Int. J. Syst. Evol. Microbiol. 52:639 646. Kung, L., and R. E. Muck. 1997. Animal response to silage additives. Pages 200 210 in Silage: Field to feedbunk. NRAES 99. Proc. N. American Conf. Hershey, PA. 11 13 Feb. Northeast Reg. Agric. Eng. Serv., Coop. Ext., Ithaca, NY. Kung Jr., L A. C. Sheperd, A. M. Smagala, K. M. Endres, C. A. Bessett, N. K. Ranjit, and J. L. Glancey. 1998. The effect of preservatives based on propionic acid on the fermentation and aerobic stability of corn silage and a total mixed ration. J. Dairy Sci. 81:1322 1330. Kung, L. and N. K. Ranjit. 2001. The effect of Lactobacillus buchneri and other additives on the fermentation and aerobic stability of barley silage. J. Dairy Sci. 84:1149 1155. Kung, L., M. R. Stokes, and C. J. Lin. 2003.Silage additives. Pages 31305 3060 in Silage Science and Technology. D. R. Buxton, R. E. Muck, and J. H. Harrison, ed. ASA CSSA SSSA, Madison, USA. Kung Jr., L C. L Myers, J. M. Neylon, C. C. Taylor, J. Lazartic, J. A. Mills, and A. G. Whiter. 2004. The effects of buffered propionic acid based additives alone or combined w ith microbial inoculation on the fermentation of high moisture corn and whole crop barley. J. Dairy Sci. 87:1310 1316. Kung, L. 2008. Silage Temperature: how hot is to hot? Dairy Newsletter Accessed May 10, 2011. http://www.qualitysilage.com/silage troubleshooting/PDF/SilageTemps How Hot is Too Hot.pdf
148 Kutz, R. E., J. D. Sampson, L. B. Pompeu, D. R. Ledoux, J. N. Spain, M. Vazquez Anon, and G. E. Rotti nghaus. 2009. Efficacy of Solis, Novasilplus, and MTB 100 to reduce aflatoxin M1 levels in milk of early to midlactation dairy cows fed aflatoxin B1. J. Dairy Sci. 92:3959 3963. Lafont, P., J. Lafont, S. Mousset, and C. Frayssinet. 1980. Etude de la contamination du lait de vache lors de lingestion de faibles quantites daflatoxine. Ann. Nutr. Alim. 34:699 708. Lambert, R. J., and M. Stratford. 1999. Weak acid preservatives: modeling microbial inhibition and response. J. Appl. Microbiol. 86:157 164. Ltt eme, P. and P. Lingvall. 1996. Effect of hexamine and sodium nitrite in combination with sodium benzoate and sodium propionate on fermentation and storage stability of wilted and long cut grass silage. Swed. J. Agri. Res. 26:135 146. Lanier, C., E. Richar d, N. Heutte, R. Picquet, V. Bouchart, and D. Garon. 2010. Airborne molds and mycotoxins associated with handling of corn silage and oilseed cakes in agricultural environment. Atmos. Environ. 44:1980 1986. Lin, C. 1992. Epiphytic microflora on alfalfa and corn; Lactic acid bacteria succession during the pre ensiling and ensiling periods; and the effect of additives on microbial succession on silage fermentation. Ph.D. diss. Kansas State University,. Manhattan, KS. Lindgren, S., K. Petterson, A. Johnsson, P. Lingvall, and A. Kaspersson. 1985. Silage inoculation. Selected strains, temperatures, wilting and practical application. Swed. J. Agric. Res. 15:9 18. Liu, Y., and J. Wang. 1999. Southern corn rust occurred in Hebei Province in 1998. Plant Prot. 25:53 58. Liu, Y., and F. Wu. 2010. Global burden of aflatoxin induced hepatocellular carcinoma: A risk assessment. Environ. Health Perspect.118:818 8 24. Lucas, J. A. 1998. Plant Pathology And Pathogens, 3rd ed. Blackwell Science, Oxford. Lynn, T. V., D. D. Hanco ck, T. E. Besser, J. H. Harrison, D. H. Rice, N. T. Stewart, and L. L. Rowan. 1998. The occurrence and replication of Escherichia coli in cattle feeds. J. Dairy Sci. 81:1102 1108. Madigan, M. T., J. M. Martinko, P. V. Dunlap, and D. P. Clark. 2009. Biolog y of Microorganisms. 12th ed. Pearson Benjamin Cummings., San Francisco CA. Makimura, S., and N. Suzuki. 1982. Quantitative determination of bovine serum haptoglobin and its elevation in some inflammatory diseases. J. Vet. Sci. 44:15 21.
149 Mallmann, C. A. P. Dilkin, A. O. Mallmann, and D. Tyska. 2009. Mycotoxin: impacts and control strategies. Pages 269 280 in Proceedings of the International Symposium. on Forage Quality and Conservation. M. Zopollato, Muraro, G. B, Nussio, L. G., FEALQ, Piracicaba, SP, B razil. Marasas, W. F. O., T. S. Kellerman, W. C. A. Gelderblom, J. A. W. Coetzer, P. G. Thiel, and J. J. Van Der Lugt. 1988. Leucoencephalomalacia in a horse induced by Fumonisin B1 isolated from Fusarium moniliforme. Onderstepoort J. Vet. Res. 55:197 203. Marcinakova M., A. Laukova, M. Simonova, V. Strompfova, B. Korenekova, and P. Na. 2008. A new probiotic and bacteriocin producing strain of Enterococcus faecium EF9296 and its use in grass ensiling. Czech J. Anim. Sci., 53:336 345. Mari, L. J., R. J. Schm idt, L. G. Nussio, C. M. Hallada, and L. Kung. 2009. Short communication: An evaluation of the effectiveness of Lactobacillus buchneri 40788 to alter fermentation and improve the aerobic stability of corn silage in farm silos. J. Dairy Sci. 92:1174 1176. M arin, D. E., I. Taranu, R. P. Bunaciu, F. Pascale, D. S. Tudor, N. Avran, M. Sarca, I. Cureu, R. D. Criste, V. Suta, and I. P. Oswald. 2002. Changes in performance, blood parameters, humoral and cellular immune response in weanling piglets exposed to low doses of aflatoxin. J. Anim. Sci. 80:1250 1257. Masoero, F., A. Gallo, M. Moschini, G. Piva, and D. Diaz. 2007. Carryover of aflatoxin from feed to milk in dairy cows with low or high somatic cells count. Anim. 9:1344 1350. Mathur, C. F., R. C. Smith, and G. E. Hawkins.1975. Growth and morphology of Streptococcus bovis and of mixed rumen bacteria in the presence of aflatoxin B1, in vitro. J. Dairy Sci. 59:455 458. Mathur, S., P. D. Constable, R. M. Eppley, A. L. Waggoner, M. E. Tumbleson, and W. M. Haschek. 2001. Fumonisin B1 is hepatotoxic and nephrotoxic in milk fed calves. Toxicol. Sci. 60:385 396. McDonald, P., N. Henderson, and S. Heron. 1991. The Biochemistry of Silage. 2nd ed. Chalcombe Publications, Bucks, UK. Mead, P. S., and P. M. Griffin. 1998. E scherichia coli O157:H7. The Lancet 352:1207 1212. Mechie, S. C., P. A. Chapman, and C. A. Siddons. 1997. A fifteen month study of Escherichia coli O157:H7 in a dairy herd. Epidemiol. Infect. 118 (Suppl.1):17 25. Meile, L G. Dasen, S. Miescher, M. Stier li, and M. Teuber. 1999. Classification of propionic acid bacteria and approaches to applied genetics. Lait. 79:71 78.
150 Merry, R. J., and D. R. Davies. 1999. Propionibacteria and their role in the biological control of aerobic spoilage in silage. Lait. 79:49 164. Ministry of Agriculture, Fisheries, and Food. 1986. The Analysis of Agricultural Materials. Reference book 427. HMSO, London, UK. Moon, N. J. 1983. Inhibition of the growth of acid tolerant yeasts by acetate, lactate and propionate and their synerg istic mixtures. J. Appl. Bacteriol. 55:454 460. Moschini, M., A. Gallo, G. Piva, and F. Masoeo. 2008. The effects of rumen fluid on in vitro aflatoxin binding capacity of different sequestering agents and in vivo release of sequestered toxin. Anim. Feed Sc i. Technol. 147:292 309. Muck, R. E., and J. T. Dickerson. 1988. Storage temperature effects on proteolysis in alfalfa silage. Trans. ASAE. 31:1005 1009. Muhlbach, P. 1998. Potential for silage production in Brazil. Pages 135 144 in Proc. Biotechnology in the Feed Industry Symposium. T.P. Lyons and K.A. Jacques (ed.), Nottingham University Press, Nottingham, UK. Mulrooney, C. N. and L. Kung. 2008. Short Communication: The effect of water temperature on the viability of silage inoculants. J. Dairy Sci. 91: 236 240. Murphy, K., P. Traves, M. Walport. 2008. Janeways Immunology.7th rev. ed. Garland Sci., Abingdon, UK. National Research Council. 2001. Nutrient requirements of dairy cattle. 7th rev. ed. Natl. Acad. Sci., Washington, DC. Neumark H. 1967. On the areas of the stomach of sheep that are sensitive to formic acid and histamine. J. Agric. Sci. 69:297 303. Neumark, H. and Tadmor, A. 1968. The effect of histamine combined with formic acid on food intake and rumen motility, when infused into the omasum o f a ram. J. Agric. Sci. 79:267 270. Niku Paavola, M. L., A. Laitila, T. Mattila Sandholm, and A. Haikara. 1999. New types of antimicrobial compounds produced by Lactobacillus plantarum J. Appl. Microbiol. 86:29 35. Norman, H. D., R. H. Miller, J. R. Wrigh t, and G. R. Wiggans. 2000. Herd and state means for somatic cell count from dairy herd improvement. J. Dairy Sci. 83:2782 2788. Ohyama, S., Masaki, A. Takigawa, and T. Morichi. 1971. Studies on various factors affecting silage fermentation. Synergistic ef fect of inoculation of Lactobacillus plantarum and addition of glucose on silage quality. Jap. J. Zootech. Sci. 42 :1 8.
151 Okiely, P. 1996. Performance of beef cattle offered grass silages made using bacterial inoculants, formic acid or sulfuric acid. Ir. J. Agric. Food. Res. 35:1 15. Osweiler, G. D., M. E. Kehrli, J. R. Stabel, J. R. Thurston, P. F. Ross, and T. M. Wilson. 1993. Effects of fumonisin contaminated corn screenings on growth and health of feeder calves. J. Anim. Sci. 71:459 466. Oude Elferink, S J. W. H., J. Krooneman, J.C. Gottschal, S. F. Spoelstra, F. Faber, and F. Driehuis. 2001. Anaerobic conversion of lactic acid to acetic acid and 1,2 propanediol by Lactobacillus buchneri. Appl. Environ. Microbiol. 67:125 132. Oude Elferink, S. J. W. H., F. Driehuis, J. C. Gottschal, and S. F. Spoelstra. 2011. Silage fermentation processes and their manipulation. Accessed June 10, 2011. http://www.fao.org/docrep/005/x8486e/x8486e09.htm Pahlow, G., R. E. Muck, F. Driehuis, S. J. W. H. Oude Elferink, and S. F. Spoelstra. 2003. Microbiology of ensiling. Pages 31 93 in Silage Science and Technology (Agronomy Series No. 42). D. R. Buxton, R. E. Muck, and H. J. Harrison, eds. American Society of Agronomy. Madison, WI Parsons, D. J. 1991. Modeling gas flow in a silage clamp after opening. J. Agric. Eng. Res. 50:209 218. Patterson, D. S. P., and P. H. Anderson. 1982. Recent aflatoxin feeding experiments in cattle. Vet. Rec. 110:60 61. Patterson, R., an d L.G. Young. 1993. Efficacy of hydrated sodium calcium aluminosilicates, screening and dilution in reducing the effects of mold contaminated corn in pigs. Can. J. Anim. Sci. 73:615 624. Pedroso, A. de F. 2003. Chemical additives and bacterial inoculants on control of losses and quality of sugar cane ( Saccharum officinarum L.) Silage. Ph.D. Thesis. Escola Superior de Agricultura Luiz de Queiroz., Piracicaba, Brazil. Pedroso, A. F., L. G. Nussio, S. F. Paziani, D. R. S. Loures, M. S. Igarasi, R. M. Coelho, I H. Packer, J. Horii, and L. H. Gomes. 2005. Fermentation and epiphytic microflora dynamics in sugar cane silage. Sci. Agric. 62:427 432. Pedroso, A. F., L. G. Nussio, W. Barioni Junior, A. A. Rodrigues, D. R. A. S. Loures, F. Campos, J. L. Ribeiro, L. J. Mari, M. Zopollatto, M. Junqueira, P. Schmidt, S. F. T. Paziani, and J. Horii. 2006. Performance of Holstein heifers fed sugarcane silages treated with urea, sodium benzoate or Lactobacillus buchneri Pesqui. Agropecu. Bras. 41:649 654. Pedroso, A. F., A. T. Adesogan, O. C. M. Queiroz, and S. K. Williams. 2010. Control of Escherichia coli O157:H7 in corn silage with or without various inoculants: Efficacy and mode of action. J. Dairy Sci. 93:1098 1104.
152 Phillips, T. D. 1999. Dietary clay in chemoprevention of aflatoxin induced disease. Toxicol. Sci. (Suppl. 1 ) 52:118 126. Phuntsok, T., M. A. Froetschel, H. E. Amos, M. Zheng, and Y. W. Huang. 1998. Biogenic amines in silage, apparent post ruminal passage, and the relationship between biogenic amines and diges tive function and intake by steers. J. Dairy Sci. 81:2193 2203. Pitt, R. E., and R. E. Muck. 1993. A diffusion model of aerobic deterioration at the exposed face of bunker silos. J. Agric Engr. Res. 55:11 26. Potter, L. R. 1987. Effect of crown rust on regrowth, competitive ability and nutritional quality of perennial and Italian ryegrasses. Plant Path. 36:455 461. Queiroz, O. C. M., A. T. Adesogan, and S. C. Kim. 2009. Can bacterial inoculants improve the quality of rust infested corn silage? J. Anim. Sc i. 87 (Suppl. E). p 543, Abstr. 663. Queiroz, O. C.M., M. B. Rabaglino, A. T. Adesogan. 2011. Silage pathogenicity and implications for the ruminant production chain. Pages 225 241 in Proceedings of the International Symposium on Forage Quality and Conserv ation. M. Zopollato, Muraro, G. B., Nussio, L. G., FEALQ, Piracicaba, SP, Brazil. Quintavalla, S. and G. Parolari. 1993. Effects of temperature, aw and pH on the growth of Bacillus cells and spores: a response surface methodology study. Intl. J. Food Micro biol. 19:207 216. Rangel, J. M., P. H. Sparling, C. Crowe, P. M. Griffin, and D. L. Swerdlow. 2005. Epidemiology of Escherichia coli O157:H7 o utbreaks, United States 1982 2002. Emerging Infect. Dis.11:603 60 9. Ranjit, N. K., C. C. Taylor, and L Kung. 200 2. Effect of Lactobacillus buchneri 40788 on the fermentation, aerobic stability and nutritive value of maize silage. Grass and For. Sci. 57:73 81. Reed, K. F. M., and D. D. Moore. 2009. A preliminary survey of zearalenone and other mycotoxins in Australi an silage and p asture. Anim. Product. Sci. 49: 696 703. Reinders, R. D., P. G. H. Bijker, and S. J. W. H. Oude Elferink. 1999. Growth and survival of vertoxigentic E. coli. Pages 18 27 in Conferences proceedings on Survival and growth of verocytotoxigenic E. coli Athens: Agricultural University of Athens, Athens. Rhind, D., J. M. Waterson, and F. C. Deighton. 1952. Occurrence of P. polysora in west Africa. Nature 169:631 6 32.
153 Richard, E., N. Heutte, V. Bouchart, and D. Garon. 2009. Evaluation of fungal contamination and mycotoxin production in maize silage. Anim. Feed Sci. Technol. 148:309 320. Ride, J. P. 1978. The role of cell wall alterations in resistance to fungi. Ann. Appl. Biol. 89:302 306. Riley, L W., R. S. Remis, S. D. Helgerson, H. B. McGee, J. G. Wells, B. R. Davis, R. J. Hebert, E. S. Olcott, L. M. Johnson, N. T. Hargrett, P. A. Blake, and M. L. Cohen. 1983. Hemorrhagic colitis associated with a rare Escherichia coli serotype. N. Engl. J. Med. 308:681 685. Rodriguez Ardon, R., G. E. Scott, and S. B. King. 1980. Maize yield losses caused by southern maize rust. Crop Sci. 20:812 814. Ross P. F., P. E. Nelson, J. L. Richard, G. D. Osweiler, L. G. Rice, R. D. Plattner, and T. M. Wilson. 1990. Production of fumonisins by Fusarium monilifornie and Fusarium proliferatlem isolates associated with equine leukoencephalomalacia. and pulmonary edema syndrome in swine. Appl. Environ. Micro. 56:3225 3226. Ryser, E. T., S. M. Arimi., and C. W. Donnelly. 1997. Effects of pH on distribution of Listeria riboty pes in corn, hay, and grass silage. Appl. Env iromental Microbiol. 63: 3695 3697. Saeger, S.D., L. Sibanda, and C. V. Peteghem. 2003. Analysis of zearalenone and alpha zearalenol in animal feed using high performance liquid chromatography. Anal. Chim. Acta 478:137 143. Salawu, M. B., E. H. Warren, and A. T. Adesogan. 2001. Fermentation characteristics, aerobic stability and ruminal degradation of ensiled pea/wheat bi crop forages treated with two microbial inoculants, formic acid or quebracho tannins. J. S ci. Food Agric. 81:1263 1268. Salter, R., D. Douglas, M. Tess, B. Markovsky, S. J. Saul. 2006. Interlaboratory study of the charm ROSA safe level aflatoxin M1 quantitative lateral flow test for bovine raw milk. J. A.O.A.C. Int. 89:1327 1334. Sanaa, M., and F. J. L. Mnard. 1994. Contamination du lait cru par L monocytogenes: Origines, facteurs de risque, prvention, Rec. Md. Vet 170 :437 442. Sarr, A. B. 1992. Effects of phyllosilicate clay on the metabolic profile of aflatoxin B1 in Fischer 344 rats. Tox icol. Lett. 75:145 151. Schmidt, R. J., W. Hu, J. A. Mills, and L. Kung. 2009. The development of lactic acid bacteria and Lactobacillus buchneri and their effects on the fermentation of alfalfa silage. J. Dairy Sci. 92:5005 5010.
154 Schmidt, R. J., and L. Ku ng. 2010. The effects of Lactobacillus buchneri with or without a homolactic bacterium on the fermentation and aerobic stability of corn silages made at different locations. J. Dairy Sci. 93:1616 1624. Scott, P. M., J. W. Lawrence, and W.V. Walbeek .1970. Detection of mycotoxins by thin layer chromatography: Application to screening of fungi extracts. Appl. Microbiol. 20:839 842. Scott, P. M., T. Delgado, D. B. Prelusky, H. L. Trenholm, and J. D. Miller. 1994. Determination of fumonisin in milk. J. Environ. Sci. Health. B29:989 998. Seeling, K., S. Danicke, K. H. Ueberschar, P. Lebzien, and G. Flachowsky. 2005. On the effects of Fusarium toxin contaminated wheat and the feed intake level on the metabolism and carryover of zearalenone in dairy cows. Food Addi t. Contam. 22:847 855. Silvestre, F. T., T. S. M. Carvalho, P. C. Crawford, J. E. P. Santos, C. R. Staples, T. Jenkins, and W. W. Thatcher. 2011. Effects of differential supplementation of fatty acids during the peripartum and breeding periods of Holstein cows: II. Neutrophil fatty acids and function, and acute phase proteins. J. Dairy Sci. 94:2285 2301. Sleator, R. D,D. Watson, C. Hill, C. G. M. Gahan. 2009. The interaction between Listeria monocytogenes and the host gastrointestinal tract. Microbiol., 155 : 2463 2475. Smith, E. E., T. D. Phillips, J. A. Ellis, R. B. Harvey, L. F. Kubena, J. Thompson, and G. Newton. 1994. Dietary hydrated sodium calcium aluminosilicate reduction of aflatoxin M1 residue in dairy goat milk and effects on milk production and com ponents. Anim. Sci. 72:677 682. Sparo, M. D., M. S. Castro, P. J. Andino, M. V. Lavigne, C. Ceriani, G. L. Gutirrez, M. M. Fernndez, M. C. De Marzi, E. L. Malchiodi, and M. A. Manghi. 2006. Partial characterization of enterocin MR99 from a corn silage i solate of Enterococcus faecalis J. Appl. Microbiol. 100:123 134. Spoelstra, S. F., M. G. Courtin, and J. A. C. Van beers. 1988. Acetic acid bacteria can initiate aerobic deterioration of whole crop maize silage. J. Agric. Sci. 111:127 132. Spotti, M., M. L. Fracchioll, F. Caloni, and G. Pompa. 2005. Aflatoxin B1 binding to sorbents in bovine ruminal fluid. Vet. Res. Commun. 29:507 515. Stroud, J. S. 2006. The effect of feed additives on aflatoxin in milk of dairy cows fed aflatoxin contaminated diets. MS t hesis. North Carolina State Univ., Raleigh. Swaminathan, B., and P. Gerner Smidt. 2007. The epidemiology of hum an listeriosis. Microb. Inf. 9: 1236 1243.
155 Tabacco, E., F. Righi, A. Quarantelli, and G. Borreani. 2011. Dry matter and nutritional losses during aerobic deterioration of corn and sorghum silages as influenced by different lactic acid bacteria inocula. J. Dairy Sci. 94:1409 1419. Tanaka, O., and S. Ohmomo.2000. Effect of inoculation with Pediococcus sp M 9 on ensiling. Grassl. Sci. 46:153 157. Taylo r, C.C., N. J. Ranjit, J. A. Mills, J. M. Neylon,. and L Kung Jr. 2002. The effect of treating whole plant barley with Lactobacillus buchneri 40788 on silage fermentation, aerobic stability, and nutritive value for dairy cows. J. Dairy Sci. 85:1793 1800. Tienungoon, S., D. A. Ratkowsky. T. A. McMeekin, and T. Ross. 2000. Growth limits of Listeria monocytogenes as a function of temperature, pH, NaCl and lactic acid. Appl. Environ. Microbiol. 66 :4979 4987. Todar, K. 2011. Lactic Acid Bacteria. Accessed Jan. 29, 2011. www.textbookofbacteriolagy.net Underhill, K. L., B. A. Rotter, B. K. Thompson, D. B. Prelusky, and H. L. Trenholm. 1995. Effectiveness of cholestyramine in the detoxification of zearalenone as determined in mice. Bull. Environ. Contam. Toxicol. 54:128 134. USDA 2011. USDA economics, statistics and market information system. Accessed May 11., 2011. http:// usda.mannlib.cornell.edu/usda/current/CropProd/CropProd 05 11 2011.pdf Van Egmond, H. P. 1989. Aflatoxin M1: Occurrence, toxicity, regulation. Pages 11 55 in: H.P. Van Egmond, ed., Mycotoxins in Dairy Products Elsevier Applied Science, London, UK. Van So est, P. J., R. H. Wine, and L. A. Moore. 1966. Estimation of the free digestibility of forages by the in vitro digestion of cell wails. Page 438 in Proc. 10th Intl. Grassl. Congr., Helsinki, Finland. Van Soest, P. J., J. B. Robertson, and B. A. Lewis. 1991. Methods for dietary fiber, neutral detergent fiber and non starch polysaccharides in relation to animal nutrition. J Dairy Sci. 74:3568 3597. Vilar, M. J., E. Yus, M. L. Sanjuan, F. J. Dieguez, and J. L. Rodriguez Otero. 2007. Prevalence of and risk factors for Listeria species on dairy farms. J. Dairy Sci. 90:5083 5088. Waldo. D. R. 1978. The use of direct acidification in silage production. Pages 117 179 in Fermentation of Silage A review. M.E. McCullough ed. NFIA, Des Moines, IA.
156 Wang, F., and N. Nishino. 2008. Ensiling of soybean curd residue and wet brewers grains with or without other feeds as a total mixed ration. J. Dairy Sci. 91:2380 2387. Watson, S. J., and Nash, M. J. 1960. Pages 319 395 in The conservation of grass and forage crops, Oliver and Boyd, Edinburgh, London. Weaver, G. A., H. J. Kurtz, J. C. Behrens, T. S. Robison, B. E. Seguin, F. Y. Bates, and C. J. Mirocha 1986. Effect of zearalenone on the fertility of virgin dairy heifers. Am. J. Vet. Res. 47:139 5 1397. Weinberg, Z. G., and R. E. Muck. 1996. New trends and opportunities in the development and use of inoculants for silage. FEMS Microbiol. Rev. 19:53 68. Weinberg, Z. G., G. Ashbell, Y. Hen, A. Azrieli, G. Szakacs, and I. Filya. 2002. Ensiling whole crop wheat and corn in large containers with Lactobacillus plantarum and Lactobacillus buchneri J. Ind. Microbiol. Biotechnol. 28:7 11. Weinberg, Z. G., G. Ashbell, Y. Chen, M. Gamburg, and S. Sela. 2004. The effect of sewage irrigation on safety and hyg iene of forage crops and silage. Anim. Feed Sci. Technol. 116:271 280. Weinberg, Z. G., P. Khanal, C. Yildiz, Y. Chen, and A. Arieli. 2011. Ensiling fermentation products and aerobic stability of corn and sorghum silages. Grassl. Sci. 57:46 50. Weissbach F., H. Honig, E. Kaiser. 1993. The effect of nitrate on the silage fermentation, Pages 122 123 in Proc. 10th Int. Conf. Silage Research, Dublin City University, Dub lin, Ireland. White ,D. 2007. The physiology and biochemistry of prokaryotes. Vol. 3rd edn. Oxford University Press., Oxford, New York. Whitlow, L.W., R. L. Nebel, and W. M. Hagler, Jr. 1994. The association of deoxynivalenol in grain with milk production loss in dairy cows. Pp. 131 139. In: G. C. Llewellyn, W. V. Dashek and C. E. O Rear. (Eds.) "Biodeterioration Research 4". Plenum Press, New York. Whitlock, L. A., T. J. Wistuba, M. K. Seifers, R. V. Pope, and K. K. Bolsen. 2000. Effect of level of sur face spoiled silage on the nutritive value of corn silage diets. J. Dairy Sci. 83(Suppl. 1):110. (Abstr.) Whitlow L. W., and W. M. Hagler. 2005. Mycotoxins in dairy cattle: occurrence, toxicity, prevention and treatment. Proc. Southwest Nutr. Conf.124 138. Whittenbury, R. 1961. An investigation of the lactic acid bacteria. Ph.D. thesis. Univ. of Edinburg, UK.
157 Wiedmann, M. 2003. An integrated science based approach to dairy food safety: Listeria monocytogenes as a model system. J. Dairy Sci. 86:1865 1875. Wi lkinson, J. M., J. Hill, and J. D. Leaver. 1996. Effect of swath treatment and period of wilting on field losses, ensiling characteristics and feeding value of grass silage. Pages 46 47 in Proc. Int. Silage Conf., 11th. D. I. H. Jones ed. Aberystwyth, UK Wilkinson, J. M., and K. K. Bolsen. 1996. Production of silage and hay in Europe and North America. Pages 42 43 in Proc. Int. Silage Conf., 11th. D. I. H. Jones ed. Aberystwyth, UK. Wilkinson, J. M., K. K. Bolsen, and C. J. Lin. 2003. History of silage sc ience and technology. Pages 1 30 in Silage Science and Technology (Agronomy Series No. 42). D. R. Buxton, R. E. Muck, and H. J. Harrison, eds. American Society of Agronomy. Madison, WI Wilson, J. P., R. N. Gates, and W. W. Hanna. 1991. Effect of rust on y ield and digestibility of pearl millet forage. Phytopathol. 2:233 236 Woolford, M. K. 1975. Microbial screening of food preservatives, cold sterilants and specific antimicrobial agents as potential silage additives. J. Sci. Food Agric. 26:229 237. Woolfo rd, M. K. 1978. Antimicrobial effects of mineral acids, organic acids, salts and sterilizing agents in relation to their potential as silage additives. J. Br. Grassl. Sci. 33:131 136. Woolford, M. K., K. K. Bolsen, and L. A. Peart. 1982. Studies on the aer obic deterioration of whole crop cereal silages. J. Agric. Sci 98: 529 535. Wu, Felicia. 2010. Mycotoxins, animal health, and economic impacts in the present and future. Pages 43 50 in Proc. World Nutrition Forum. Vol. 1. E. M. Binder, ed. Nottingham university press, Salzburg, Austria. Yiannikouris, A., and J. P. Jouany. 2002. Mycotoxins in feeds and their fate in animals: a review. Anim. Res. 51:81 99. Yildirim, M. 2001. Purification of buchnericin LB produced by Lactobacillus buchneri LB. T.J. Biol. 25: 5 9 65. Yildirim, Z ., Y. K. Avsar, and M. Yildirim. 2002. Factors affecting the adsorption of buchnericin LB, a bacteriocin produced by Lactocobacillus buchneri Microbiol Res. 157:103 107.
158 BIOGRAPHICAL SKETCH Oscar Queiroz was born in Sao Paulo, Brazil. He enrolled at the University of Sao Paulo, Escola Superior de Agricultura Luiz de Queiroz in Brazil in 1999 and earned his Bachelor of Sciences in a gricultural e ngineering 5 years later. This university is the oldest and perhaps the best institution for studying Agricultural Sciences in Brazil. Oscar obtained his Master of Sciences degree at the same university in 2007. His thesis focused on developing new methods to improve the quality and aerobic stability of sugarcane silage to facilitate its use as a livestock feed in Brazil. In August 2007, Oscar became a Doctor of Philosophy candidate in the Department of Animal Sciences at the University of Florida under Dr. Adegbola Adesogans supervision. While pursuing his doctoral research, Oscar also participated in or coordinated several additional research projects on various areas such as using fibrolytic enzymes to improve the digestion of forages and feeds and the performance of dairy cattle, evaluating effects of different films for covering bunker silos on the quality and shelf life of silage, determination of the influence of the culture medium, temperature, and duration on the growth of yeasts and molds on silage, examination of effects of the amount of silage and container type on the aerobic stabilit y and shelf life of silage, examination of effects of species and bacterial inoculation on the fermentation and nutritive value of tropical grasses etc. Oscars long term plan is to become a faculty member at a university. His immediate plan is to hone an d improve the skills and knowledge he acquired during his Ph.D. program in a postdoctoral research position where he can fulfill his passion to improve technologies involved in making silage.