because they can not be further hydrogenated (Harfoot and Hazlewood, 1997). ...... (Lee et al., 2003a), or higher concentration of soluble sugars in fresh plants ...
DETERMINING THE FACTORS THAT CAUSE HIGHER CONCENTRATION OF CONJUGATED LINOLEIC ACID (CLA) IN MILK FAT OF DAIRY COWS FED FRESH ALFALFA VERSUS ALFALFA HAY
DISSERTATION
Presented in Partial Fulfillment of the Requirement for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University
By Claudio Vaz Di Mambro Ribeiro, M.S. *****
The Ohio State University 2005
Dissertation Committee:
Approved by
Professor Maurice L. Eastridge Professor Jeffrey L. Firkins Professor Normand St-Pierre Professor William Weiss
______________________ Adviser Graduate Program in Animal Sciences
ABSTRACT
Much attention has been given to conjugated linoleic acid (CLA) due to its anticarcinogenic properties. Dairy products are one of the major sources of CLA in human diets. Dietary management of dairy cows to increase CLA in milk may be beneficial for the dairy industry. Several researchers have reported that grazing animals produce a higher percentage of CLA in milk fat when compared to animals fed TMR diets; however, the reason for this observation is uncertain. In the first two experiments, the effect of storage on the profile of fatty acids (FA) from fresh alfalfa was determined. The samples were stored at -20 oC for 60 days and the profile of FA was estimated in Experiment 2 based on the results from Experiment 1. After flash-freezing alfalfa samples with liquid N just after harvesting, the percentage of total FA did not change. I proposed that the higher sucrose concentration in fresh forages results in higher vaccenic acid (VA) flow to the duodenum, the main factor responsible for high milk CLA. Experiment 3 was conducted to test the hypothesis that supplemental sucrose increases the percentage of VA in the effluent of continuous culture fermenters. The ii
pattern of biohydrogenation (BH) of FA from fresh alfalfa or alfalfa hay supplemented with three concentrations (0, 4, and 8%) of sucrose was studied at a constant pH of 6.2. Four continuous culture fermenters were used in a 4 X 4 Latin square design. Fresh alfalfa increased total BH of FA compared to hay. Vaccenic acid concentration in the effluent was almost three times higher for fresh alfalfa compared to alfalfa hay and was the predominant 18:1 isomer, whereas trans-10 18:1 was 50% lower for fresh alfalfa. Total BH and BH of linoleic and linolenic acids linearly decreased with sucrose. The effects of sucrose on BH and concentration of VFA may have been caused by a shift in microbial population by mechanisms that are independent of pH. In Experiment 4, the interaction effect between sucrose and pH on BH was evaluated in vitro. I also developed a multi-compartmental model to estimate pool size and flux of vaccenic acid during BH of FA in fresh alfalfa. There was no effect of sucrose addition on the disappearance rate of 18:2 and 18:3. Also, there was no effect of sucrose on the estimated BH of vaccenic acid. In conclusion, sucrose is not the main cause of the higher CLA in milk fat of dairy cows fed fresh as opposed to preserved forage. However, the pH-independent effect of sucrose on BH should be further investigated by quantifying the changes in microbial species. Using compartmental kinetics to study BH can provide new information on the rate constants of the intermediates of BH.
iii
Dedicated to my parents, sister, and family.
iv
ACKNOWLEDGMENTS
I would like to express my deepest appreciation to my advisor, Dr. Maurice Eastridge, for his support, patience, and friendship. I sincerely appreciate the opportunity and guidance he gave me during my Ph. D. program. I wish to thank my committee members, Drs. Maurice Eastridge, William Weiss, Jeffrey Firkins, and Normand St-Pierre for their valuable inputs since the beginning of my graduate program. I am also grateful to Dr. Jeffrey Firkins and Dr. Normand St-Pierre for their encouragement, stimulating discussion, and comments. The knowledge I acquired by interacting with the above mentioned professors could not be possibly obtained by just reading books and publications and it also goes beyond scientific knowledge. I am deeply thankful to my colleagues in the office and laboratory. Carine Reveneau, John Sylvester, and Sanjay Karnati provided intellectually stimulating discussions and a friendly place to work. Many of my very good friends helped me to feel at home. I would like to especially thank Dr. Donald Palmquist and Maria Sol Morales for their friendship, guidance, and support during the many times I was in Wooster. They welcomed me in their home and treated me as their own son.
v
I do not know how to put in words how fortunate I am by having the opportunity to meet and share part of my life with Ana Gonzalez. I am grateful for her love, patience, support, and optimism. She helped me a lot during difficult times and, I learned a great deal from her. Although we were very far apart, I am very grateful to my whole family, especially my parents and sister. Their support, guidance, encouragement, and love were the basis for my strength and perseverance to pursue my Ph. D. in a foreign country. I consider myself blessed by God for my physical and mental health, family and friends. I am also thankful for the opportunities, obstacles, and amazing human beings I have encountered during my life.
vi
VITA October, 21, 1973...................Born – Belo Horizonte, Brazil 2000 ....................................... M.S. Animal Sciences, University of Sao Paulo/ESALQ 2000 – present ........................Presidential Fellowship, The Ohio State University
PUBLICATIONS
1. Ribeiro, C.V.D.M., and M.L. Eastridge. 2004. Relationship of rate of appearance of vaccenic acid and pH during in vitro biohydrogenation of linolenic acid in alfalfa hay. J. Dairy Sci. (Suppl 1):37 (Abstr.) 2. Bucci, P.B., M.L. Eastridge, and C.V.D.M. Ribeiro. 2004. Effects of NDF from alfalfa hay, grass hay, straw, and whole cottonseed on performance of lactating cows. J. Dairy Sci. (Suppl 1):466 (Abstr.) 3. Ribeiro, C.V.D.M., and M.L. Eastridge. 2004. Comparison of biohydrogenation of fatty acids in lyophilized forage and air dried forage with sucrose additions. J. Dairy Sci. (Suppl 1):38(Abstr.) 4. Ribeiro, C.V.D.M., A.V. Pires, I. Susin, J.M. Simas, and R. C. Oliveira Jr. 2004. Effect of corn grain substitution by pear millet grain (Pennisetum americanum) on the diet of lactating dairy cows. Rev. Soc. Bras. Zootec. 5:1351-1359.
vii
5. Ribeiro, C.V.D.M., M.L. Eastridge, and D.L. Palmquist. 2003. Evaluation of the profile of fatty acids extracted from fresh alfalfa. J. Dairy Sci. (Suppl 1): 285 (Abstr.) 6. Eastridge, M.L., and C.V.D.M. Ribeiro. 2003. Formulating dairy rations for concentrations of fat versus fatty acids. J. Dairy Sci. 86:3806 (Presented at Midwest ADSA) 7. Dutra, A. R., A. C. de Queiroz, J. C. Pereira, S. C. Valadares Filho, J. T. L. Thiebaut, F. N. Matos, and C. V. D. M. Ribeiro. 1997. Effects of fiber levels and protein sources on nutrient intakes and digestion in steers. Rev. Soc. Bras. Zoot. 4:787-796. 8. Dutra, A. R., A. C. de Queiroz, J. C. Pereira, S. C. Valadares Filho, J. T. L. Thiebaut, F. N. Matos, and C. V. D. M. Ribeiro. 1997. Effects of fiber levels and protein sources on the synthesis of microbial nitrogen compounds in steers. Rev. Soc. Bras. Zoot. 4:797-805.
FIELDS OF STUDY Major Field: Animal Sciences Ruminant Nutrition
TABLE OF CONTENTS
Page Abstract ............................................................................................................................. ii Dedication ........................................................................................................................ iv Acknowledgments ............................................................................................................. v Vita .................................................................................................................................. vii List of Tables .................................................................................................................. xii List of Figures ................................................................................................................. xv List of Abbreviations .................................................................................................... xviii Chapters:
1. INTRODUCTION .......................................................................................................... 1 2. REVIEW OF LITERATURE ......................................................................................... 4 BIOHYDROGENATION PROCESS........................................................................... 4 Lipolysis.................................................................................................................. 4 Biohydrogenation.................................................................................................... 5 FACTORS AFFECTING BIOHYDROGENATION ................................................... 7 Concentration of Unsaturated Fatty Acids.............................................................. 7 Lipolysis Rate ......................................................................................................... 8 Ruminal pH........................................................................................................... 10 Ruminal Microorganisms...................................................................................... 11 CLA AND MILK FAT................................................................................................ 12 EFFECT OF FORAGE ON MILK CLA ..................................................................... 14 SOLUBLE SUGARS AND RUMINAL FERMENTATION ..................................... 19 3. EVALUATION OF EXTRACTION NUMBER AND STORAGE ON THE PERCENTAGE OF FATTY ACIDS FROM FRESH ALFALFA .............................. 22 ix
ABSTRACT................................................................................................................ 22 INTRODUCTION ...................................................................................................... 23 MATERIAL AND METHODS .................................................................................. 25 Reagents................................................................................................................ 25 Experiment 1......................................................................................................... 25 Experiment 2......................................................................................................... 26 Fatty Acid Analyses.............................................................................................. 27 Statistical Analysis................................................................................................ 27 RESULTS ................................................................................................................... 29 Experiment 1......................................................................................................... 29 Experiment 2......................................................................................................... 29 DISCUSSION ............................................................................................................. 30 CONCLUSIONS......................................................................................................... 35 4. NUTRIENT DIGESTIBILITIES AND BIOHYDROGENATION OF FATTY ACIDS FROM FRESH ALFALFA OR ALFALFA HAY PLUS SUCROSE IN CONTINUOUS CULTURE ........................................................................................ 40 ABSTRACT................................................................................................................ 40 INTRODUCTION ...................................................................................................... 41 MATERIAL AND METHODS .................................................................................. 43 Treatments and Experimental Design................................................................... 43 Continuous Culture Operation .............................................................................. 44 Sample Collection and Analysis ........................................................................... 45 DNA Extraction and PCR..................................................................................... 46 Phylogenetic analysis based on RIS-Length Polymorphism ................................ 46 Statistical Analysis................................................................................................ 47 RESULTS ................................................................................................................... 48 Digestibility Measures .......................................................................................... 48 Volatile Fatty Acids .............................................................................................. 48 Fatty Acids and Biohydrogenation ....................................................................... 49 RIS-LP Analysis ................................................................................................... 50 DISCUSSION ............................................................................................................. 51 Digestibility........................................................................................................... 51 Volatile Fatty Acids .............................................................................................. 53 x
Biohydrogenation of Fatty Acids.......................................................................... 55 Intermediates of Biohydrogenation....................................................................... 59 CONCLUSIONS......................................................................................................... 61 5. EFFECT OF SUCROSE, PH, AND FORAGE CONSERVATION ON THE IN VITRO BIOHYDROGENATION OF FATTY ACIDS ........................................................... 70 ABSTRACT................................................................................................................ 70 INTRODUCTION ...................................................................................................... 71 MATERIAL AND METHODS .................................................................................. 73 Treatments and Incubation Procedure .................................................................. 73 Analysis of Fatty Acids......................................................................................... 75 Statistical Analyses ............................................................................................... 76 Model 1: single pool, first-order kinetic model .................................................... 77 Model 2: multiple pools, first-order kinetic model............................................... 78 RESULTS AND DISCUSSION ................................................................................. 79 Effects of pH on Biohydrogenation ...................................................................... 79 Forage Source and Biohydrogenation................................................................... 81 Sucrose and Biohydrogenation ............................................................................. 83 Intermediates of Biohydrogenation....................................................................... 83 Kinetics of Biohydrogenation Using SAAM II .................................................... 86 Limitations of the Modeling Approach................................................................. 89 CONCLUSIONS......................................................................................................... 90 6. OVERALL DISCUSSION ......................................................................................... 105 FRESH FORAGES ................................................................................................... 106 KINETICS OF RUMINAL BIOHYDROGENATION ............................................ 108 THEORIES ............................................................................................................... 112 Lipases ................................................................................................................ 112 Protozoa .............................................................................................................. 113 Fresh Forage........................................................................................................ 114 BIBLIOGRAPHY...................................................................................................... 116 APPENDICE............................................................................................................. 131
xi
LIST OF TABLES Table 3.1
Page Least square means of fatty acid composition from fresh alfalfa stored at - 20 °C............................................................................39
4.1
Nutrient composition of fresh alfalfa and alfalfa hay .......................................63
4.2
Nutrient digestibility in continuous culture fermenters receiving fresh alfalfa or alfalfa hay supplemented with three concentrations of sucrose ........64
4.3
Proportions of volatile fatty acids in continuous culture fermenters receiving fresh alfalfa or alfalfa hay supplemented with three concentrations of sucrose..................................................................................65
4.4
Biohydrogenation (BH) of 18:1, 18:2, 18:3 and total BH of fatty acids in continuous culture fermenters receiving fresh alfalfa or alfalfa hay supplemented with three concentrations of sucrose ........66
4.5
Fatty acid profile in effluent from continuous culture fermenters receiving fresh alfalfa or alfalfa hay supplemented with three concentrations of sucrose..................................................................................67
xii
4.6
Fatty acid profile of the intermediates of biohydrogenation in effluent from continuous culture fermenters receiving fresh alfalfa or alfalfa hay supplemented with three concentrations of sucrose .........................................................................................................68
5.1
Nutrient composition of fresh alfalfa and alfalfa hay .......................................92
5.2
Fractional rates (k) of disappearance of 18:2 and 18:3 (h-1) during in vitro incubation of fresh alfalfa and alfalfa hay at high or low pH and with or without supplemental sucrose determined by a single pool, first-order kinetic model (SAS) ........................102
5.3
Areas under the curve (mg.h) of individual fatty acids during in vitro incubation with strong (SB) or weak (WB) buffers of fresh alfalfa and alfalfa hay with 0, 4, and 8% sucrose addition...103
5.4
Areas under the curve (mg.h) of the intermediates of biohydrogenation during in vitro incubation with strong (SB) or weak (WB) buffers of fresh alfalfa and alfalfa hay with 0, 4, and 8% sucrose addition .........................................................................104
D.1
Fatty acid profile of protozoa and bacteria collected from the omasum of dairy cows ....................................................................................134
E.1
Rate constants and standard error (in parenthesis) of in vitro BH of three different initial amounts 14C-labeled linoleic acid estimated by SAAM II ....................................................................................................136 xiii
LIST OF FIGURES Figure 3.1
Page Percentage of 18:0, 18:1, 18:2, and 18:3 fatty acids from the total 18-carbon fatty acids recovered from each extraction..............................36
3.2
Percentage of 18:3 fatty acids from the total fatty acids as the number of extractions increases.......................................................................37
3.3
Change in the ratio of fatty acids among extractions........................................38
4.1
Comparison of bacterial community structure in fermenters samples..............69
5.1
Change in pH during in vitro biohydrogenation of fatty acids from alfalfa hay and fresh alfalfa incubated with strong (SB) or weak buffers (WB) .......................................................................................93
5.2
In vitro biohydrogenation curves of vaccenic acid with strong (A) or weak (B) buffers ..........................................................................................94
5.3
In vitro biohydrogenation curves of trans-9, trans-10, trans-12, trans-13, and cis-15 octadecenoic fatty acids from fresh alfalfa (A) and alfalfa hay (B) incubated with a strong buffer ...........................................95
5.4
Model of in vitro ruminal biohydrogenation of fatty acids from fresh alfalfa from 0 to 12 h incubated with strong (A) or weak (B) buffers..................................96 xiv
5.5
Changes in the in vitro vaccenic acid pool ......................................................98
5.6
In vitro disappearance of the 18:3 pool estimated by a single pool, first-order kinetic model (SAS) and a multiple pools, first-order kinetic model (SAAM II) from incubations of fresh alfalfa with strong (A) or weak (B) buffers..........................................................................99
5.7
Representation of the temporal change of the pool size (mg) of cis-9, trans-11, cis-15 octadecatrienoic acid plus trans-11, cis15 octadecadienoic acid (A; 18:3-Int pool) and CLA (B) estimated from a multiple pools, first order kinetic model (SAAM II) during in vitro biohydrogenation of fresh alfalfa with strong (SB) or weak (WB) buffers ................................100
5.8
Estimation of the instantaneous flux (mg/h) to and from the vaccenic acid pool from a multiple pools, first-order kinetic model (SAAM II) during in vitro biohydrogenation of fresh alfalfa with strong (A) or weak (B) buffers.......................................101
A.1
Cluster analysis of bacterial profiling using DDGE .......................................131
B.1
Diagram of a multi-compartmental model from SAAM II applied to samples taken over time using continuous culture.........................132
C.1
Image of a protozoal cell taken from the omasum of a grazing cow..............133
xv
E.1
Diagram of a multi-compartmental model from SAAM II applied in vitro incubation of 14C-labeled linoleic acid ..................................135
F.1
Temporal change of peak areas and identification of octadecenoic fatty acids during in vitro biohydrogenation .......................138
xvi
LIST OF ABBREVIATIONS
BCVFA = Branched-Chain Volatile Fatty acids BH = Biohydrogenation CLA = Conjugated Linoleic Acid DGDG = Digalactosyldiglyceride DR = Disappearance Rate FA = Fatty Acids FAME = Fatty Acid Methyl Esters GC = Gas Chromatrograph H:IP = Hexane:isopropanol IP = Isopropanol LOX = Lipoxygenase MGDG = Monogalactosyldiglyceride N = Nitrogen NSC = Nonstructural Carbohydrates PUFA = Polyunsaturated Fatty Acids RIS-LP = Ribosomal intergenic spacer length polymorphism SCD = Stearoyl-CoA Dessaturase SFA = Saturated Fatty Acids xvii
TFA = Total Fatty Acids TMR = Total Mixed Ration UFA = Unsaturated Fatty Acids VA = Vaccenic Acid VFA = Volatile Fatty Acids
xviii
CHAPTER 1
INTRODUCTION Conjugated linoleic acid (CLA) is a collective term to describe one or more positional and geometrical isomers of linoleic acid (18:2), the cis-9, cis-12octadecadienoic acid (Parodi, 1997). Much attention has been given to CLA due to its anticarcinogenic properties and ability to alter lipid metabolism (Ha et al., 1987; Parodi, 1997; Blankson et al., 2000; Ip et al., 2002; Martin and Valeille, 2002). Because synthesis of CLA occurs in the rumen during biohydrogenation (BH) by microorganisms, dairy products are the major sources of CLA in human diets (Kelly et al., 1998a). The ruminal BH process is responsible for changing the profile of dietary FA, increasing the proportion of saturated fatty acids (SFA) reaching the duodenum. During this process, many CLA isomers and trans-18:1 FA are synthesized, absorbed, and become a constituent of the FA in meat and milk of ruminants. Understanding BH and factors involved in its regulation may allow for greater dietary manipulation, increase product quality, and consumer acceptance of ruminant products. The cis-9, trans-11-octadecadienoic acid (cis-9, trans-11 18:2) is the major CLA found in ruminant products (Jahreis et al., 1997). It can be hydrogenated to trans-11 18:1 (vaccenic acid; VA), the major trans-18:1 FA in the rumen. Forages contain a higher percentage of linolenic acid (18:3) that can be biohydrogenated to VA, but not to CLA; 1
however, VA is also an important precursor of the synthesis of CLA in the mammary gland (Griinari et al., 2000; Kay et al., 2004) and human adipose tissues (Turpeinen et al., 2002). Therefore, consumption of VA-rich foods also may be beneficial to human health (Banni et al., 2001; Turpeinen et al., 2002). Several studies have shown that some dietary modifications may affect the population of ruminal microorganisms, leading to a shift in ruminal biohydrogenation (Czerkwaski et al., 1966; Chalupa et al., 1967; Latham et al., 1972; Gerson et al., 1985; Van Nevel and Demeyer, 1995; Van Nevel and Demeyer, 1996). Level and type of supplemented oil (Kelly et al., 1998a; Dhiman et al., 1999b; Franklin et al., 1999; Lawless et al., 1999; Donovan et al., 2000; Mir et al., 2000), underfeeding (Jiang et al., 1996; Stanton et al., 1997), and forage to concentrate ratio (Jiang et al., 1996; Griinari et al., 1998; French et al., 2000) are some dietary factors known to affect the CLA and trans-18:1 concentration in milk. The concentration of CLA in milk can vary from 5.04 to 11.28 mg/g of fat (Jiang et al., 1996) or be higher than 33 mg/g of fat (Kelly et al., 1998b), depending on the level and type of supplemental fat in the diet or whether the animals are fed total mixed rations (TMR) or fresh forages. Grazing animals produce a higher percentage of CLA in milk fat when compared to animals fed TMR diets (Jahreis et al., 1997; Kelly et al., 1998b; Dhiman et al., 1999a; French et al., 2000). The exact mechanism by which pasture increases CLA in milk has yet to be elucidated. The higher concentration of soluble sugar in fresh forages has been hypothesized to be the cause of the greater concentration of CLA in milk fat of grazing animals; this higher concentration of soluble carbohydrates leads to a more active biohydrogenation from Butyrivibrio fibrisolvens (Kelly et al., 1998b; French et al., 2000), 2
the major bacterium responsible for the first step of ruminal biohydrogenation (Polan et al., 1964). However, there are limited data regarding why pasture is associated with high levels of CLA in grazing cows. Therefore, the factors affecting level of CLA isomers and their metabolites during ruminal biohydrogenation should be elucidated in order to achieve a desirable profile of these FA in milk and meat of ruminants used for human consumption. The proportion of the intermediates of BH reaching the mammary gland may cause milk fat depression (Griinari et al., 1998; Bauman and Griinari, 2000) and, therefore, change energy partitioning; a low VA to trans-10 18:1 ratio is associated with milk fat depression. Because the ruminal concentrations of the intermediates of BH are determined by their respective rates of appearance and disappearance, developing additional tools to study the kinetics of BH and understand how dietary factors can alter those rates is also important to prevent milk fat depression and improve energy status of transition cows. Additionally, little is known about the BH of 18:3 from fresh forages and why cows fed fresh forages have a higher concentration of the intermediates of BH in milk and rumen fluid. Hence, understanding the BH of fresh alfalfa may help to modulate BH towards a higher VA flow to the duodenum. The major objective of this set of experiments is to study and improve the methods used to estimate BH rates of FA from fresh forages and to test the hypothesis that nonstructural carbohydrates affect biohydrogenation towards CLA and/or VA synthesis, thus increasing the percentage of those FA in milk fat of dairy cows. A second objective is to improve the techniques to study kinetics of BH.
3
CHAPTER 2
REVIEW OF LITERATURE
BIOHYDROGENATION PROCESS
The biohydrogenation process is responsible for the increase in the proportion of saturated fatty acids (SFA) reaching the duodenum of ruminants and also for the ruminal synthesis of conjugated linoleic acid (CLA) and vaccenic acid (VA) that are eventually incorporated in meat and milk of ruminants. The BH process consists of two distinguished events: lipolysis and biohydrogenation itself (hydrogenation).
Lipolysis Because the free carboxyl group is required for hydrogenation of unsaturated FA (UFA), lipolysis is a prerequisite to biohydrogenation (Chalupa and Kutches, 1968). Esterified lipids are hydrolyzed extensively by microbial lipases after consumption, causing the release of constituent FA (Jenkins, 1993).
4
Although protozoa also may be involved in this process (Wright, 1961), Anaerovibrio lipolytica, a Gram-negative bacterium, is considered of major importance as a lipolytic agent in the rumen (Henderson, 1975). Rate of lipolysis is an important factor in determining the products of BH (Wu and Palmquist, 1991) because it alters the release and subsequent concentration of FA that undergo hydrogenation. Because the concentration of some FA can inhibit their hydrogenation (Noble et al., 1974), the rate of lipolysis may alter the concentration of their intermediates. Noble et al. (1974) observed 49.3% of unesterified FA after 2 hours and 46.1% after 3 hours during incubation with 0.9 and 1.33 mg/ml of esterified linoleic acids (trilinoein), respectively. These authors also observed a slower and less complete hydrogenation of 18:2 with increasing concentration of trilinolein. They concluded that the mechanism of BH of trilinoelin-derived 18:2 may differ from that for free 18:2, which demonstrates that the form of supplemented FA to ruminants has an impact upon BH.
Biohydrogenation The BH is an intracellular process comprising two separated systems, where the first step is the hydrogenation of 18:2 to a monoene FA, and the second step is the saturation of the monoene to stearic acid (Polan et al., 1964). The first step of BH of 18:2 to stearic acid comprises two reactions; the first one is the production of a conjugated acid (cis-9, trans-11 18:2), which it is extremely rapid compared to the overall BH (Kepler et al., 1966). The second reaction produces a more saturated FA, vaccenic acid (Kepler et al., 1966; Mills et al., 1970; Noble et al., 1974; Fellner et al., 1995), and also produces cis-11 18:1 in lower levels (Fellner et al., 1995). Linoleic acid (cis-9, cis-12), 5
for the most part, was converted to the conjugated cis/trans 18:2 acid within 0.5 hours, and as the incubation continued, there was an accumulation of the trans-octadecenoic acid (trans-18:1) at the expense of the conjugated cis/trans 18:2 (Kepler et al., 1966). The second step of BH is the conversion of cis/trans 18:1 to stearic acid, which increases rapidly after high levels of cis/trans 18:1 FA accumulate (Polan et al., 1964) and may require synergism of rumen microorganisms to be completed (Polan et al., 1964; Mills et al., 1970). Polan et al. (1964) observed that pure culture of B. fibrisolvens produced monoenoic acid from 18:2, but mixed rumen bacteria produced a complete SFA (stearic acid). These reactions appear to have substrate control rates and may also be regulated by different factors. Kellens et al. (1986) indicated that the isomerase reaction from 18:2 to conjugated intermediate acid is more active than the next reaction, an inhibited reductase step, due the high concentration of the diene after 7 hours of 18:2 incubation. Additionally, Noble et al. (1974) observed that an accumulation of large amounts of VA was preceded in all instances by the accumulation of cis-9, trans-11 18:2, which increased with the increasing initial concentration of 18:2 and could be associated with the decrease in the rate of appearance of VA. These authors suggested that factors other than the amount of 18:2 may be involved in the inhibition of the conversion of 18:1 to 18:0, since they observed a persistent inhibition after the concentration of 18:2 had been decreased to a negligible amount. However, this persistent inhibition could be attributed to the readaptation of the microorganisms that may have been affected by high concentrations of 18:2.
6
The BH of 18:3 also produces trans-18:1 FA. The first reaction is an isomerization, followed by a hydrogenation, producing cis-9, trans-11, cis-15 18:3 and trans-11, cis-15 18:2, respectively. The last FA can be BH either to trans-11, trans-15, or cis-15 18:1 FA. The trans- and cis-15 18:1 FA are considered to be true products of BH because they can not be further hydrogenated (Harfoot and Hazlewood, 1997).
FACTORS AFFECTING BIOHYDROGENATION
Dietary nutrients alter microbial species numbers making it difficult to determine precisely how and which factors affect BH. Also, synergism of factors may occur making it more difficult to access their relationship in this process. Any factors changing viability of microorganisms or their metabolic pathways may alter BH by modifying the rate and extent of lipolysis and/or hydrogenation.
Concentration of Unsaturated Fatty Acids The concentrations of UFA (mainly 18:2 and 18:3) present in the medium affects the rate of BH and the concentration of the intermediate acids. Complete hydrogenation of 18:2 to stearic acid is diminished with increasing 18:2 concentration (Polan et al., 1964; Noble et al., 1974; Kellens et al., 1986). The 18:2 acts as a competitive inhibitor for the hydrogenation of the monoenoic acid (Polan et al., 1964). Harfoot et al. (1973) 7
reported that 18:2 prevents hydrogenation of oleic acid to stearic acid by irreversible inhibition rather than by simply competing for hydrogen; however, the continuous supply of readily fermentable substrate may reduce the inhibitory effect of 18:2 upon hydrogenation. Because the second step of BH is dependent on cross-feeding, the group of microorganisms that account for reduction of cis/trans 18:1 to stearic is probably susceptible to UFA concentration. Body (1976) reported that the increase in the concentration of 18:3 caused a decrease in the synthesis of stearic acid, similarly to the effect of 18:2 concentration (Harfoot et al., 1973; Noble et al., 1974). At low concentrations, the conversion of 18:2 to monoenoic acid was much faster than at higher concentrations at equal cell densities (Kellens et al., 1986); thus, the rate and extent of lipolysis in the rumen are important factors to consider when studying BH. As the level of 18:2 increases, the conversion to stearic acid decreases, with a higher occurrence of the monoenoic fraction (Polan et al., 1964).
Lipolysis Rate The amount of UFA available for BH is related to the form of FA supplementation (free FA or triglyceride) and lipolysis rate. Rumen microorganisms are adapted to metabolize and tolerate esterified FA to a greater extent than free FA, and the rates of hydrolysis may not be sufficient to liberate an amount of free FA that could affect the extent of BH (Noble et al., 1974). A shift in the number of lipolytic microorganisms may alter significantly the rate of BH. Reduction of lipolysis may also contribute to the reduction in BH (Latham et al., 8
1972). The reduced lipolytic activity in low-roughage diets is probably related to the reduced number of lipolytic vibrios (Latham et al., 1972) and hydrogenation may be increased by the number of ciliate protozoa (Chalupa et al., 1967). Wright (1961) reported a considerable reduction of lipolysis by adding penicillin and terramycin and little effect of neomycin and streptomycin in rumen liquor of cows; however, a similar degree of reduced hydrogenation of UFA was observed with all antibiotics. The extent to which UFA escape BH appears to depend on microbial growth conditions that influence rates of lipolysis and BH (Jenkins, 1993). The form of FA affects the concentration of the intermediates of BH. Noble et al. (1974) demonstrated a lower production of CLA when 18:2 was incubated as triglyceride compared to free FA. The mechanism of BH of trilinolein-derived 18:2 may differ from that for free 18:2 (Noble et al., 1974). These authors concluded that there are distinct differences in the extent of BH and the nature of the intermediates produced during the BH of 18:2 when supplied in free or esterified forms. They reported a greater BH with trilinolein incubated in vitro than when approximately equivalent amounts of 18:2 were present as the free acid. Grazing
animals
ingest
a
higher
percentage
of
18:3
FA
as
monogalactosyldiglyceride (MGDG) and digalactosyldiglyceride (DGDG). Because plant lipases (galactolipases and phospholipases; Butler and Bailey, 1973) are activated after harvesting, the amount of free 18:3 inside the plant tissues during ruminal fermentation may be higher than in the fresh forage. However, the impact of the free 18:3 derived from the lipolysis of plant enzymes associated with bacterial enzymes upon ruminal fermentation is still uncertain. 9
Ruminal pH Ruminal pH is known to affect the microbial population and, therefore, fermentation pattern. Feeding grain to ruminants may decrease lipolysis, resulting from low ruminal pH (Jenkins, 1993). Also, changing a high-fiber diet to high-starch diet may decrease BH due to decreasing number of cellulolytic species (Gerson et al., 1985). Van Nevel and Demeyer (1996) reported a decrease in lipolysis at pH 5.9-6.0. Those authors reported that, at similar pH, incubations with 80 mg of soybean oil had higher inhibition of lipolysis than 40 mg. Grain feeding decreases rumen BH and increases unsaturation of carcass fat and milk (Jenkins, 1993); however, lipolytic activity was independent of the presence of hay (Van Nevel and Demeyer, 1996). Griinari et al. (1998) observed a decrease in the rumen pH and a higher proportion of propionate when cows were fed a low fiber diet. The authors reported that UFA and low fiber diets increased the concentration of trans-10 in milk fat. Although the bacterial species responsible for synthesizing trans-10 18:1 is not known, the effect of pH and UFA on the bacterial population is the major cause of the inhibition of BH and change in the ratio of VA to trans-10 18:1. Although lipolysis can be inhibited at low pH, BH of liberated polyunsaturated FA (PUFA) was influenced to a much lower extent. Only pH values lower than 6.3 seem to inhibit lipolysis (Van Nevel and Demeyer, 1996). Accordingly, Qiu (2004) observed an inhibition of BH when the pH changed from 6.5 to 5.8 in a dual-flow continuous culture system. The author reported greater flow of CLA and VA when the pH dropped.
10
Whether the lower pH values affect the population of lipolytic microorganisms and/or enzyme activity is yet uncertain. Also, energy readily available to rumen microflora might affect lipolysis rate (Gerson et al., 1985). Diets with higher passage rate and low readily available energy source could modify hydrogenation rate.
Ruminal Microorganisms Although protozoa may play a role in BH, bacteria are the most important and studied ruminal microorganisms (Harfoot and Hazlewood, 1997). There are two major groups of bacteria responsible for BH: group A and group B (Kemp and Lander, 1976). Group A biohydrogenates 18:2 to trans-18:1, while group B biohydrogenates trans-18:1 to 18:0; yet, both groups participate in the first two steps of the BH of 18:3 (Harfoot and Hazlewood, 1997). Therefore, synergism between both groups of bacteria is required for a complete BH (Polan et al., 1964; Mills et al., 1970). Yet, some bacteria species are able to carry out the BH of UFA to 18:0 (van de Vossenberg and Joblin, 2003). Among the bacteria species capable of BH, B. fibrisolvens (Polan et al., 1964; Kepler et al., 1966) and Fusocillus babrahamensis (White et al., 1970; Hazlewood et al., 1976) represent the most important bacteria in group A and B, respectively. Dietary and ruminal factors that alter the proportion of both groups of bacteria can inhibit BH and, therefore, decrease or increase the accumulation of trans-18:1 FA and CLA in the rumen. Although protozoa may not be as important as bacteria during BH of UFA, their exact role is uncertain. Chalupa and Kuthches (1968) argued that rumen protozoa contribute significantly to BH of FA in the rumen because they possess a hydrogenation potential similar to that of bacteria. Oligotrich protozoa were capable of converting 11
approximately 95% of the
14
C-labeled 18:2 to octadecenoic and stearic acids, and
Holotrich protozoa were able to promote very little BH. Conversely, Harfoot et al. (1973) observed no effect on the pattern of BH when protozoa were removed before incubation in those experiments in which oleic acid was the major end product of hydrogenation. The authors suggested that rumen protozoa might be affected only by 18:2 and not by the intermediates of BH or that, after 6 hours of incubation, the FA are taken up into the bacterial cells and, therefore, do not come into contact with the protozoa. Also, lowroughage diets decreased the number of ciliate protozoa (Latham et al., 1972). Impure mixtures of ruminal ciliates have been reported to hydrogenate UFA from 18:2 to stearic acid (Gutierrez et al., 1962). Those authors showed that pure cultures of ciliates could hydrogenate UFA. Rumen protozoa either directly or indirectly may influence protein, carbohydrate, or lipid metabolism in ruminants (Klopfenstein et al., 1966). These authors indicated that the major effect of protozoa in the rumen on the plasma lipid distribution is a relative shift from 18:2 to oleic acid. Synergistic effects between rumen bacteria and protozoa and the influences of rumen liquor per se on various aspects of rumen microbial metabolism are not completely understood.
CLA AND MILK FAT
The CLA in milk originates from CLA produced during the BH process (Harfoot and Hazlewood, 1997) and also from VA through ∆9-desaturase (or stearoyl-CoA desaturase, SCD) activity in the mammary gland (Griinari et al., 2000; Kay et al., 2004). 12
Because higher amounts of UFA inhibit the last step of BH, supplemental fat containing either 18:2 or 18:3 favors the increase of CLA in milk fat of dairy cows. For instance, adding soybean oil or cottonseed oil increased milk CLA percentage compared with control (Dhiman et al., 1999b). Furthermore, fish oils have also been shown to increase CLA in milk (Donovan et al., 2000; Chilliard et al., 2001). Additionally to higher UFA in the diet, ruminal pH can alter the proportion of intermediates of BH and inhibit the hydrogenation of trans-18:1 FA to 18:0. Griinari et al. (1998) reported an interaction of UFA and rumen pH on the percentage and profile of trans-18:1 FA. The authors observed a greater increase in trans-10 18:1 milk fat with low fiber diets with higher amounts of UFA; this interaction also caused milk fat depression. The milk fat depression observed with low-fiber diets has been proposed to be related to higher concentration of trans-18:1 FA in the rumen (Erdman, 1996). The UFA have detrimental effects on some ruminal microbes (Palmquist and Jenkins, 1980; Jenkins, 1993), which may affect the extent of BH. The 18:2 and 18:3 have been shown to inhibit the final step of BH, i.e., reduction of trans-18:1 acid to stearic acid (Polan et al., 1964; Harfoot et al., 1973; Body, 1976; Kellens et al., 1986). Also, as the concentrations of 18:2 and 18:3 increase, the conversion to stearic acid decreases with a higher occurrence of the monoenoic fraction (Polan et al., 1964; Body, 1976). Romo et al. (1996) observed a decrease in milk fat percentage of cows abomasally infused with VA. Loor and Herbein (1998) argued that CLA, but not VA, might inhibit the desaturation of stearic acid in the mammary gland. However, some studies showed that there may be another specific isomer associated with milk fat depression (Jiang et al., 1996; Griinari et al., 1998; Lawless et al., 1999; Baumgard et al., 2000; Baumgard et al., 13
2001) besides cis-9, trans-11 18:2. The increase of trans-10 18:1 in milk was related with milk fat depression induced by diets low in fiber (Griinari et al., 1998). This FA is probably one intermediate of BH of the trans-10, cis-12 18:2, the CLA isomer responsible for inhibiting SCD (Park et al., 2000; Baumgard et al., 2001). Baumgard et al. (2000) observed a decrease in milk fat yield (44%) and percentage (42%) when cows were abomasally infused with 10.3 g/d of trans-10,cis-12 18:2. The pathway by which trans-10, cis-12 18:2 is formed is not known. Griinari et al. (1998) showed that a diet high in unsaturated fat and low in fiber increased the concentration of trans-10 18:1 in milk fat and decreased milk fat percentage. This inhibitory effect on BH may be associated with a decrease in number of cellulolytic species (Gerson et al., 1985).
EFFECT OF FORAGE ON MILK CLA
Grazing cows have a higher concentration of milk CLA (Jahreis et al., 1997; Kelly et al., 1998b; Dhiman et al., 1999a; French et al., 2000). The BH of 18:2 has both CLA and VA as intermediates. However, during ruminal BH of 18:3, which is the major FA in forages, only VA is synthesized (Harfoot and Hazlewood, 1997). Because VA is a precursor for cis-9, trans-11 CLA by the action of SCD in grazing cows (Kay et al., 2004), the higher concentration of CLA in milk from those cows could be attributed to the higher concentration of VA reaching the mammary gland. 14
Kelly et al. (1998b) observed twice as much CLA concentration in milk fat from cows on pasture compared with the control group. This finding concurs with that made by Dhiman et al. (1999a), who observed a linear increase in the percentage of CLA in milk fat of dairy cows as the amount of pasture increased. The treatments were one-third pasture, two-thirds pasture, and all pasture. As the amount of pasture increased, the proportion of 18:3 increased and 18:2 decreased in the diets. The authors pointed out that cows fed the all pasture diet had 500% more CLA in milk fat compared to a mixed diet (Experiment 1). There was no difference in milk fat percentage among treatments. Kelly et al. (1998b) concluded that the higher CLA percentage in milk fat of cows on pasture was due to the rapidly fermented sugar and soluble fiber found in fresh forages, which may create ruminal conditions that increase the outflow of CLA from the rumen. Additionally, ruminal factors that increase the outflow of VA from the rumen lead to an increase in CLA in milk fat by the action of SCD in the mammary gland. French et al. (2000) argued that the amount of soluble fiber and sugar in the fresh grass influenced somehow the fermentation of Butyrivibrio fibrisolvens, and it may have been responsible for the higher levels of CLA found in muscle fat of steers fed grass. Kelly et al. (1998b) commented that other factors that may have influenced the higher percentage of CLA in milk fat of cows on pasture were passage rate, fluid dilution rate, meal size, and feeding frequency. None of the experiments above evaluated either the changes on the profile of FA flowing out from the rumen or the amount of sugar in the pasture plants. Kelly et al. (1998b) observed 19% less DMI of cows fed pasture compared with a mixed diet, causing a reduction in body weight. They also observed an increase (30%) in intake of 18:1 FA and a consumption of 3 times more 18:3 from cows on the pasture 15
treatment. Cows on pasture lost weight, leading to FA mobilization. One could suggest that trans-18:1 FA and CLA mobilized from body fat reserves may have contributed to an increased CLA concentration in the milk. In another experiment (Experiment 4), Dhiman et al. (1999a) concluded that the percentage of hay in the diet does not influence milk CLA percentage. The authors suggested that the possible cause was the lower amounts of 18:2 when grass was dried and stored. Jiang et al. (1996) studied the effect of the percentage of forage and restriction of diet on milk CLA. The control diet, fed in restricted amount, had 50% concentrate, and the trial diets (restriction and ad libitum) had 65%. They found a difference in CLA percentage in milk fat due to the percentage of forage in the diet (control vs. restricted trial diet): the animals fed the restricted diet had twice more CLA in milk fat (11.28 mg/g) than the control group (5.04 mg/g) and also higher CLA than the unrestricted diet (6.56 mg/g). These results are in opposition to those of Stanton et al. (1997), who observed that higher levels of CLA in milk fat were due to higher DM allowances of grass. Timmem and Patton (1988) observed an increase in 18:1 and a decrease in 4:0 and 16:0 fatty acids in milk of dairy cows fed a restricted diet, which may support the results from Jiang et al. (1996). Stanton et al. (1997) argued that restricted diets may cause a higher mobilization of fat stores, affecting the FA profile of milk fat. Thus, in experiments with dairy cows in negative energy balance, the percentage of VA and CLA in fat reserves may have an impact on milk CLA concomitantly to the dietary factors. Jiang et al. (1996) found that VA in milk fat followed the tendency of CLA among treatments, and it was also higher for the unrestricted diet than the control group. 16
The authors concluded that the higher amount of starch in the diets was responsible for the increased CLA and trans-18:1 in milk, since a higher percentage of grain in the diet inhibits the final step of BH (Palmquist and Schanbacher, 1991). No explanation was made as to why the restricted diet produced more CLA than the ad libitum treatment. Conversely, Dhiman et al. (1999a) found no difference in CLA in milk fat between diets containing 66.6 or 98.2% grass hay. However, they did find a higher percentage of trans18:1 in milk fat of cows fed the 66.6% grass hay treatment. The control and trial diets from Jiang et al. (1996) had different amounts of some feed ingredients that may have altered either the ruminal microflora and the profile of FA flowing from the rumen. For example, the control group diet had 10% beet pulp molasses, while the trial diets had 3% molasses. The concentration of VA in milk may be more correlated with the amount of forage in the diet, while CLA is more variable. The major difference between pasture and TMR relative to FA composition of milk is the higher levels of stearic acid and 18:1 FA in milk of cows fed pasture (Timmem and Patton, 1988). The difference for FA profile between grass and grass silage is that grass tissues contain a lower proportion of SFA and a higher proportion of UFA (mainly 18:3) than grass silage (French et al., 2000). The difference among experiments regarding the percentage of CLA in milk fat and levels of forage in the diets may be due to expression of SCD. This enzyme plays an important role on the amount of CLA in milk fat (Griinari et al., 2000). Factors other than the amount of trans-18:1 arriving at the mammary gland may have affected the activity of SCD and produced different results. For instance, 18:2 and PUFA are known to decrease the SCD activity in rats and humans (Jeffcoat et al., 1979; Ntambi, 1999). French et al. 17
(2000) found higher levels of PUFA, 18:3, and CLA in muscle fat of steers fed grass compared with silage or grass plus concentrate. However, no difference was found among treatments for 18:1 and 18:2 FA in muscle fat. The percentage of 18:3 in grass was about 30 times higher than in the concentrate, which lead to a higher n-3/n-6 ratio in muscle fat. Therefore, it is difficult to ascertain how much CLA found in milk fat is originally from the ruminal BH or from the activity of SCD in the mammary gland, because the profile of FA flowing out from rumen was not measured. The higher amount of CLA in milk fat of grazing animals has been proposed to result from the higher levels of soluble sugars in fresh forages (Kelly et al., 1998b; French et al., 2000). Levels of sugars in grasses average 17% of DM (McDonald, 1981), with fructosans being the major one followed by sucrose, fructose, and glucose (McDonald, 1981; Woolford, 1984). Conversely, the level of sugar in a TMR is about 1.5 to 3.0% of DM (Hoover and Miller-Webster, 1998). Some temperate forages have high percentage of protein (Van Soest, 1994), most of it being soluble protein (McDonald, 1981), which may lead to a higher degradability rate. Therefore, both nutrients may be responsible for a more active BH towards the production of VA by B. fibrisolvens. Yet, B. fibrisolvens is more tolerant to low pH than other celullolytic bacteria (Russell and Dombrowski, 1980), which may favor its competition for nutrients.
18
SOLUBLE SUGARS AND RUMINAL FERMENTATION
The effect of soluble sugar on performance of ruminants depends on the level of sugar, available N, and buffering capacity of the diet. Charmley et al. (1991) observed no effect of supplemental sucrose at 10% of DMI on ruminal pH of sheep. Piwonka et al. (1994) observed a rapid utilization of supplemental dextrose (5.6% of DM) in the rumen, because the baseline level of glucose was reached 1 hour post-feeding. No effect on ruminal pH was observed; however, the dextrose-supplemented diet had 48% NDF plus buffers. Solomon et al. (2000) tested the effect of dry citrus pulp pellets versus corn grain on CLA percentage in milk fat of dairy cows. They observed neither an effect on milk yield nor milk fat percentage. However, DMI decreased with citrus pulp. They also observed an effect of the citrus pulp diet on CLA in milk fat, and no interaction was found between citrus pulp and addition of full fat extruded soybean. Piwonka et al. (1994) found higher bacterial N flow to the duodenum with supplemental dextrose compared with the control diet. The profile of bacterial lipids may affect percentage of CLA in milk fat, if supplemental soluble sugar significantly increases duodenal bacteria flow. Total VFA concentration increased with dextrose supplementation, but no difference was found for concentration of individual VFA. However, molasses increased the proportion of butyrate in the rumen (Friggens et al., 1998). Whether soluble sugars increase BH by increasing ciliate protozoa is not known. Soluble sugars are the main carbohydrate substrate for holotrichs (Van Soest, 1994); 19
however, total protozoa number, holotrichs, and entodinomorphs did not change (percentage of total) with dextrose supplementation compared with the control diet (Piwonka et al., 1994). Sucrose supplementation (10% of silage DMI) decreased the daily rumen ammonia concentration in sheep (Charmley et al., 1991). The authors argued that this effect was probably related to the increase in microbial protein synthesis. Firkins and Oldick (1997) commented that cows might be able to efficiently utilize between 4 to 8% molasses by adjusting nonstructural carbohydrate (NSC) and degradable protein. The major soluble sugars in fresh grass are fructose, glucose, sucrose, and fructosans. Molasses, as a source of soluble sugar in diets of dairy cows, would be useful to mimic the sugar in the pasture. Molasses has approximately 63% sugar (Friggens et al., 1998), almost entirety sucrose. Fresh forages also contain about 25% CP of which 70 to 90% is true protein (Woolford, 1984). The synchronism between soluble sugars and protein of forages may be responsible for a higher initial bacterial growth, leading to a more active BH and production of VA, the precursor for CLA synthesis in the mammary gland. The increase in sugar to soluble protein ratio from 1:1 to between 2 and 3:1 increases the microbial protein production by 25% (Hoover and Miller-Webster, 1998). Furthermore, vitamins and minerals in pasture may have their effect on the overall ruminal fermentation. In summary, the level of CLA and VA may be affected by two major groups of factors. The first one is the chemical factors, such as level and type of FA in the diet, forage to concentrate ratio, source of forage (pasture versus conserved forage), and level of intake. The second are physical factors, such as passage rate, ruminal pH, meal size, 20
and frequency of feeding. The reasons by which cows fed fresh forages have higher concentration of CLA in milk fat may be due to either chemical or physical factors, or both. There are some disagreements about how some of these factors affect CLA in milk.
21
CHAPTER 3
EVALUATION OF EXTRACTION NUMBER AND STORAGE ON THE PERCENTAGE OF FATTY ACIDS FROM FRESH ALFALFA
ABSTRACT
Quantity and profile of fatty acids from fresh forage samples may be altered by repeated extractions and during storage by oxidation of unsaturated fatty acids. In Experiment 1, fatty acids from fresh alfalfa were extracted with hexane:isopropanol (H:IP, 3:2 v/v) in three sequential extractions, and the percentage and profile of fatty acids from each of the three extractions were determined. In Experiment 2, fresh alfalfa samples were stored for 0, 30, or 60 d and the profile of fatty acids was also determined. Samples of fresh alfalfa were randomly harvested and immediately submerged in liquid nitrogen. For the first extraction (Experiment 1), approximately 5 g of the frozen alfalfa was mixed with 18 ml of H:IP per gram of material. The second and third extractions were done by adding H:IP to the pellet (3 ml/g of the original sample weight), mixing for 2 min, and then centrifuging. The effect of storage on the profile of fatty acids was tested (Experiment 2) using the results from Experiment 1. Repeated extractions increased 22
the percentage of total fatty acids recovered from the DM. The concentration of total fatty acids in the alfalfa after three extractions was 4.0%. The first, second, and third extractions resulted in 92.67, 4.77, and 2.56% of the total fatty acids extracted, respectively. There was no effect of extraction on the proportion of 16:0, 18:0, 18:1, and 18:2 fatty acids. However, the proportion of 18:3 in the extract decreased from the first to the second extraction. After flash-freezing the forage with liquid N just after harvesting, the storage of fresh alfalfa at - 20 oC for 60 d did not change the percentage of total fatty acids. The results of this experiment revealed that the profile of fatty acids can vary with the number of extractions performed and that storing fresh alfalfa samples after flashfreezing with liquid N does not alter the percentage of 18:3 in the samples. The higher amount of 18:3 in the first extraction may reflect the higher proportion of linolenic acid in the more easily extracted plant fractions.
INTRODUCTION
An accurate assessment of the profile and percentage of total fatty acids (TFA) in fresh forage is crucial when studying the biohydrogenation (BH) of fatty acids (FA) from fresh plants. After harvesting, loss of lipids occurs due to oxidation of unsaturated fatty acids (UFA) during wilting (Dewhurst et al., 2002) and ensiling (Makoni and Shelford, 1993) through the activity of plant lipases (Butler and Bailey, 1973) and lipoxygenases (Gardner, 1991; Fried, 1993; Feussner et al., 1995).
23
Moreover, oxidation of UFA also alters the ratio of saturated FA (SFA) to UFA when samples are processed (storage, drying, and grinding). Therefore, appropiate storage and post-harvest handling are essential to accurately determine the percentage of FA in fresh forages. Most commonly used solvents for lipid extraction from fresh tissues are based on mixtures of chloroform:methanol (Folch et al., 1957; Bligh and Dyer, 1959). Yet, the chloroform:methanol procedures have suffered criticism regarding their wash step and loss of some lipid classes (Nelson, 1991). Hexane:isopropanol (H:IP; 3:2, v/v) can be used as an alternative solvent to extract lipids from biological matrices because it is less toxic, extracts less non-lipid fraction, and offers simpler handling procedures (Hara and Radin, 1978; Radin, 1981). Quantitative analysis of TFA is time consuming and consists of combining sequential extractions from each sample; an alternative approach is to correct for the amount of TFA recovered in the first extraction after the method has been validated. However, we were not aware of any study evaluating the profile of FA from individual sequential extractions using H:IP. Our objectives were to test the hypothesis that the number of extractions using H:IP (3:2, v/v) alters the profile of fatty acids from fresh alfalfa and that flash-freezing alfalfa after harvesting and storage of the samples at -20 oC for 60 days does not decrease the percentage of UFA in the plant tissues.
24
MATERIAL AND METHODS
Reagents Hexane and isopropanol (IP) were purchased from Fisher Scientific (Optima grade, Pittsburgh, PA). Fatty acid methyl esters (FAME), heptadecanoic acid (17:0), and nonadecanoic acid (19:0) were from NuCheck Prep, Inc. (Elysian, MN).
Experiment 1 Alfalfa samples (prebloom/bloom) were obtained from a greenhouse at the Ohio Agricultural Research and Development Center (Wooster, OH). Alfalfa samples were harvested randomly and immersed immediately in liquid nitrogen using a metal grid. Samples were weighed and immersed in H:IP and stored at 8 °C in the dark until extraction. Separate samples were used to determine DM (6 h at 105 °C). Samples were extracted 3 times and each extract was separately analyzed. For the first extraction, approximately 5 g of the frozen alfalfa was mixed with 18 ml of H:IP per gram of material (Radin, 1981). Bottles containing samples and solvent were put in ice and homogenized until totally fragmented. Samples were then centrifuged for 10 minutes at 10,000 x g, the supernatant was collected in a volumetric flask, and the solvent was evaporated, being careful to avoid total dryness. The second and third extractions were done by adding H:IP to the pellet (3 ml/g of the original sample weight), mixing for 2 min, and then centrifuging for 10 min at 10,000 g.
25
Samples were maintained immersed in solvent at all times and kept in the dark at low temperature when possible. Methylation of FA from the concentrated extracts was performed as described by Sukhija and Palmquist (1988) with some modifications (Palmquist and Jenkins, 2003); 2 ml of benzene containing 17:0 acid (2 mg/ml) and 3 ml of 10% HCl were added to each tube.
Experiment 2 The evaluation of storage time of frozen samples on the concentration and profile of FA from fresh alfalfa was studied. Only one extraction was used and correction were made when necessary accordingly to the results from Experiment 1. Alfalfa samples (prebloom/bloom) were obtained from the greenhouse at the Ohio Agricultural Research and Development Center (Wooster, OH). To assess the effect of oxidation on the profile of FA from fresh alfalfa during storage, samples were randomly harvested and submitted to the following treatments: T0) immediately frozen in liquid nitrogen, weighed, and immersed in H:IP; S) same as 0 but using only the stems of the plants; L) same as 0, using only the leaves; T30) same as 0, but stored at -20 °C for 30 d; and T60) same as 0, but stored for 60 d. Approximately 5 g of the frozen alfalfa was immersed in 18 ml of H:IP per gram of material (Radin, 1981). Bottles containing samples and solvent were put on ice and homogenized until totally fragmented. Samples were then centrifuged for 10 min at 10,000 g, the supernatant was collected in a volumetric flask, and the solvent was evaporated, being careful to avoid total dryness. Samples were maintained immersed in solvent and kept in the dark at 8 °C at all times until FA analysis. 26
Fatty Acid Analyses Experiment 1: The FAME were analyzed by GLC (Hewlett-Packard, Model 5890; Palo Alto, CA) using a 30-m x 0.25-mm i.d., 0.2-µm film thickness SP-2880 fused silica capillary column (Supelco, Bellefonte, PA). This system was coupled to a mass spectrometer (Thermo Finnigan Trace 2000) and a data system (Xcalibur 2000; Thermo Finnigan, San Jose, CA). Experiment 2: Methylation of FA from the concentrated extracts was performed as described by Sukhija and Palmquist (1988) with some modifications; 2 ml of benzene containing 19:0 (2 mg/ml) and 5 ml of 10% HCl were added to each tube. The FA were extracted once, and recovery of individual FA was corrected based on the previous extraction procedure (Experiment 1). The FAME were analyzed by GLC (HewlettPackard, Model 5890; Palo Alto, CA) using a 100 m x 0.25 mm i.d., 0.2 µm film thickness SP-2380 fused silica capillary column (Supelco, Bellefonte, PA). Nitrogen was used as the carrier gas. Detector and injector temperatures were set at 250 and 220 oC, respectively, and the split ratio was set at 100:1. Oven temperature was set for 160 oC for 10 min, increased by 3.0 oC/min to 180 oC and held for 60 min, increased by 5.0 oC/min to 220 oC and held for 50 min, and decreased by 20 oC/min to 160 oC for 1 min.
Statistical Analysis Experiment 1: Data were analyzed as a completely randomized design, with repeated measures in space using the mixed procedure of SAS (SAS, 2004). The Protected Fisher’s LSD test was used to compare treatment means. Significance was declared at P < 0.05. The model was: 27
Yij =µ + Ti + sj + εij, where i=1,2,3 and j= 1,2...7;
Y = effect of extraction treatment i and sample j; Ti = fixed effect of extraction i; sj = random effect of sample j; and εij = error term
Experiment 2: Data were analyzed as a completely randomized design using the MIXED procedure of SAS (SAS, 2004), according to the following model:
Yi = µ + Ti + εi, where:
Yi = dependent variable; µ = overall mean; Ti = fixed effect of treatment i; and εi = residual error.
Orthogonal contrasts were used to test for significance of linear and quadratic effect of storage and also to compare the percentage of FA between stems and leaves of fresh alfalfa. Significance was declared at P < 0.05.
28
RESULTS
Experiment 1 The GC-mass spectometer was used to detect artifacts that may be produced during the extraction/methylation procedure. Trace amounts of isopropyl esters (data not shown) were detected in the samples; their origin may be either from the plant tissues or synthesis during the methylation step or both. The percentage of TFA in the alfalfa samples averaged 4.0% across all samples (data not shown). Repeated extractions increased (P < 0.01) the percentage of TFA recovered from the samples. The TFA for first, second, and third extractions were 92.67, 4.77, and 2.56% of the total, respectively. The proportions of 18-carbon FA in each extract are shown in Figure 3.1. There was no difference (P > 0.05) in the percentage of 16:0, 18:0, 18:1, and 18:2 among extractions. However, the first extract contained a higher percentage of 18:3 (P < 0.01; Figure 3.2) and, thus, a lower ratio 16 to 18:3 (P < 0.01; Figure 3.3). Also, the ratio of saturated to unsaturated FA tended (P < 0.10) to increase from the first to the third extraction (Figure 3.3).
Experiment 2 The concentration (mg/g DM) of FA did not change (P < 0.05) during storage; however, the percentage of 16:0 linearly decreased with storage time (Table 3.1). Among the UFA, 18:1 and 18:2 tended (P > 0.10) to linearly decrease during storage; conversely, there was no effect of storage on the proportion of 18:3 in fresh alfalfa. The ratios SFA/UFA and 16:0/18:3 were not affected (P > 0.05) by storage. 29
The proportion of individual and total FA was always higher for leaves compared to stems, except for 18:2, which had the same proportion for both tissues. The ratio 16:0/18:3 was higher (P < 0.05) for stem and the ratio SFA/UFA did not differ (P > 0.05) between leaves and stems.
DISCUSSION
Because lipids, non-polar compounds, are associated with polar compounds, such as protein, to form biological membranes, solvents used to extract lipids from animal and plant tissues must have a combination of polar and non-polar organic solvents. The H:IP solvent was used as an alternative to more toxic and most used solvents (Folch et al., 1957; Bligh and Dyer, 1959) to quantitatively extract FA from plant tissues. The use of H:IP as a solvent system to quantitatively extract FA from fresh forages has a few advantages over chloroform:methanol: it extracts most of the FA in the first extraction (93% observed in this trial), IP helps to inhibit enzymatic activity and reduce loss of FA (Butler and Bailey, 1973), and IP and hexane are less toxic than methanol and chloroform, respectively (Nelson, 1991). The percentage of FA extracted from the first extraction could be used as a correction factor to quantify FA from forages in future experiments using the same methodology and just one extraction. Most of the plant FA is concentrated in the chloroplasts; linolenic acid is the major FA in plant tissue, constituting about 75 to 95% of FA from galactolipids in alfalfa leaf tissue (Hawke, 1973). The higher values for 18:3 30
in the first extract reflect the higher proportion of 18:3 in the more easily extracted plant fraction and also show that H:IP is a solvent able to quantitatively extract phospholipids and galactolipids (more polar molecules), triglycerides, and FA. This characteristic is essential when studying FA esters from biological membranes. Furthermore, IP inhibits loss of UFA resulting from enzymatic peroxidation by plant enzymes (Butler and Bailey, 1973). Oxidation of UFA is a major concern when studying FA from fresh forages. It is more likely to occur to UFA than SFA, and the higher the unsaturation of a specific FA, the more prone it is to oxidation. Using IP or flash-freezing with liquid N are methods that inhibit those enzymes; use of liquid N also avoids loss of soluble sugars (Lindroth and Koss, 1996) by respiration. Oxidation can occur either by enzymatic or non-enzymatic reactions. The enzymatic reaction is primarily caused by activation of lipases (galactolipases and phospholipases; Butler and Bailey, 1973) and lipoxigenases (LOX) from forages after harvesting (Fuller et al., 2001). There are many inter- and intra-species enzyme isoforms (Gardner, 1991; Fuller et al., 2001); although the exact purpose of those enzymes has not been completely elucidated, they are triggered after damage and infection of the plant tissues and are also associated with senescence of the plant (Gardner, 1991; Feussner et al., 1995). After harvesting, there is an increase in the amount of free FA (Dewhurst et al., 2003) and a decrease in the amount of UFA (Dewhurst et al., 2002). The 18:2 and 18:3 FA from plant tissues are the primary target of plant lipases (Butler and Bailey, 1973) and 31
LOX (Gardner, 1991; Feussner et al., 1995), resulting in the formation of hydroxyperoxides (Gardner, 1991; Makoni and Shelford, 1993; Feussner et al., 1995; Nunez et al., 2001) that are broken down even further to volatile organic compounds, such as hexenol (Croft et al., 1993; Dewhurst et al., 2003). The products of LOX cannot be used by the animals as a source of energy. The non-enzymatic reaction is dependent on the presence of oxygen (Galliard et al., 1974) and can be triggered by elevated temperature, metals (Galliard et al., 1974), light, and other free radicals present in the tissue (Min and Lee, 1996; Min, 1998). Freezing samples to stop oxidation is a common practice, but it may not completely avoid oxidation; there is still enzymatic (Nelson, 1991) and non-enzymatic oxidation (Min and Lee, 1996; Min, 1998), even at low temperatures caused by singlet oxygen and/or free radicals. Because most of the 18:3 is found in the chloroplast membrane, there is a possibility that this FA is more protected against non-enzymatic oxidation by ascorbic acid, vitamin E, and glutathione located in the chloroplasts (Makoni and Shelford, 1993). However, enzymatic oxidation is most likely to occur in the photosynthetic tissues, and chlorophyll is a known photosensitizer responsible to produce singlet oxygen in the presence of light (Lee and Min, 1991). Because liquid N and IP inhibit plant lipases and samples were kept at low temperature and in the dark, the tendency of linear effect of storage on the oxidation observed for 18:2 (P = 0.10; Table 3.1) may have resulted from non-enzymatic reactions.
32
Alfalfa leaves had a higher amount of all individual FA and total FA compared to stems. However, no effect was observed for the amount of 18:2 between leaves and stems. Dewhurst et al. (2001) commented that leaf proportion is important when determining FA concentration. This observation is consistent with the location of FA (18:2 and 18:3) in plant tissues. Leaves contain a higher amount of chloroplast, in which the monogalactosyldiglycerides (MGDG) and digalactosyldiglycerides (DGDG) are richer in 18:3 compared to 18:2 (Butler and Bailey, 1973; Hawke, 1973); thus, the difference in location of those FA may have influenced their oxidation. Van der Veen and Olcott (1967) reported 30% loss of 18:3 when alfalfa was kept at room temperature for 24 hours. The authors concluded that the most critical period to protect oxidation of FA from forages is immediately after harvesting. The ratio SFA/UFA was similar between leaves and stems. Because most of the FA in plants are a constituent of cellular membranes, I speculate that this ratio represents membrane fluidity that is kept similar between plant tissues. Furthermore, the ratio of 16:0/18:3 was double for the stem tissue, consistent with the 70% lower 18:3 from stem compared to leaf tissues. Solvents can absorb oxygen if left standing open. Organic solvents used to store FA should be freed of oxygen by redistillation and/or purging high purity N through the solvent (Nelson, 1991). The use of organic solvents is a good way to prevent autooxidation; however, solvent systems containing alcohol may react with the UFA, decreasing their proportion (Dr. Huang, Adjunct Professor, Dept. Human Nutrition, OSU; personal communication). Fried (1993) reported that formaldehyde, phenol, and other preservatives should not be used to store lipids. The percentages of total and individual 33
FA from oven-dried alfalfa samples were higher than for fresh alfalfa (data not shown). This observation is not consistent with the detrimental effect of high temperatures on oxidation of FA; thus, I speculated that the loss of FA occurred during storage of fresh plant tissues using H:IP. A better approach may be using H:IP to stop enzymatic oxidation, extract lipids from the plant, evaporate it, and add hexane (or other hydrocarbon) before storing the samples for future analysis (Guil-Guerrero et al., 2001); additionally, the use of antioxidantes during the extraction and/or storage would prevent oxidation. Conserved forages have a lower amount of total FA and a different FA profile than fresh forages, both results of oxidation after harvesting and loss of leaves during wilting. Dewhurst et al. (2002) showed that wilting decreases the proportion of TFA, with 18:3 being the most affected. Makoni and Shelford (1993) reported a 50% loss of lipids by ensiling fresh alfalfa. Because BH occurs with UFA, preserving their proportion in the samples to be studied is critical; the concentration of UFA in the samples affects the rate of synthesis of intermediates of the BH (VA and CLA) and also their concentration in the rumen fluid. Additionally, ruminant products contain higher percentages of 18:3 (Duckett et al., 1993) and CLA (Kelly et al., 1998b; Dhiman et al., 1999a) when animals are grazing compared to those fed conserved forages. In vitro studies of BH of FA from fresh forages must minimize the impairment of the loss of FA after harvesting. Moreover, the loss of leaves during wilting decreases the
34
proportion of UFA and total FA. Thus, I adopted this methodology to conserve FA in harvested alfalfa to study ruminal BH of FA from fresh alfalfa.
CONCLUSIONS
The percentage of 18:3 in fresh alfalfa decreased with increasing number of extractions and tended to change the ratio of SFA/UFA. Each laboratory should evaluate the profile and percentage of TFA from each extraction before using a single extraction to determine TFA. Unless corrections are made for the percentage and profile of FA, more extractions are needed to precisely determine the amount and profile of individual FA. Tissue samples extracted by H:IP should be free of alcohol during storage and processed as soon as possible to avoid lipid oxidation. After flash-freezing the forage with liquid N just after harvesting, the storage of fresh alfalfa at - 20 oC for 60 d did not change the percentage of TFA.
35
100%
18:03
80%
18:02
60% 40%
18:01
20%
18:00
0% 1
2
3
Extractions
Figure 3.1: Percentage of 18:0, 18:1, 18:2, and 18:3 fatty acids from the total 18-carbon fatty acids recovered from each extraction. Values for extraction 1 were 8.7, 4.7, 14.4, and 72.9%, respectively; extraction 2: 10.8, 6.5, 18.6, and 64.0%, respectively; and extraction 3: 19.0, 10.8, 21.1, and 49.1%, respectively.
36
60 40 20 0
Extractions
first
second
third
Figure 3.2: Percentage of 18:3 fatty acids from the total fatty acids as the number of extractions increases. The first extraction resulted in the highest value (P < 0.01; 50.7%) and no difference was found (P > 0.10) between the second (29.9%) and the third (21.0%) extractions.
37
5
0
1
2
3
a 16/18:3 0.51
0.86a
1.67b
a sat/uns 0.42
1.87b
2.82b
Figure 3.3: Change in the ratio of fatty acids among extractions. The third extraction yielded the highest values for 16:0/18:3 and saturated to unsaturated ratio (sat/uns). Means in the same row with different superscript differ (P < 0.01).
38
Table 3.1: Least square means of fatty acid composition from fresh alfalfa stored at -20 o
C. (Experiment 2).
Treatments Probability of Contrasts1
Days stored Fatty acids
0
30
60
Leaves Stems
SE
Linear Quadratic L vs S
--------------mg/g of DM--------------16:0
4.61
4.69
3.84
5.56
3.39
0.24
0.02
0.12
< 0.01
18:0
0.94
1.04
0.81
1.39
0.56
0.11
0.31
0.22
< 0.01
18:1
0.80
0.83
0.58
1.01
0.53
0.10
0.08
0.25
< 0.01
18:2
3.81
3.70
3.26
3.48
4.02
0.26
0.10
0.60
0.17
18:3
7.30
8.17
7.68
9.22
2.86
0.83
0.70
0.49
< 0.01
16/18:3
0.64
0.62
0.51
0.61
1.20
0.08
0.17
0.69
< 0.01
SFA/UFA2
0.49
0.50
0.42
0.53
0.55
0.05
0.26
0.39
0.70
Total 17.7 18.8 16.4 20.9 11.5 1.1 0.32 0.17 < 0.01 Linear and quadratic effect of days stored and the difference between leaves (L) and stems (S) on the profile of fatty acids. 1
2
Ratio between saturated fatty acids (SFA) to unsaturated fatty acids (UFA).
39
CHAPTER 4
NUTRIENT DIGESTIBILITIES AND BIOHYDROGENATION OF FATTY ACIDS FROM FRESH ALFALFA OR ALFALFA HAY PLUS SUCROSE IN CONTINUOUS CULTURE
ABSTRACT
The pattern of biohydrogenation of fatty acids from fresh alfalfa or alfalfa hay supplemented with three concentrations (0, 4, and 8%) of sucrose was studied at a constant pH of 6.2. Four continuous culture fermenters were used in a 4 X 4 Latin square design. Effluent was collected from each of the four fermenters during the last three days of each 10-day period. Nutrient digestibility, VFA, and fatty acids in the effluent were measured. Flow of bacterial OM and NDF and ADF digestibilities were higher for fresh alfalfa than alfalfa hay. Fresh alfalfa increased total biohydrogenation of fatty acids more than hay. Vaccenic acid concentration in the effluent was almost three times higher for fresh alfalfa compared to alfalfa hay and was the predominant 18:1 isomer, whereas trans-10 18:1 was 50% lower for fresh alfalfa. Therefore, the ratio of vaccenic acid to trans-10 18:1 was higher for fresh alfalfa. 40
True OM digestibility of alfalfa hay tended to linearly decrease with sucrose supplementation. However, microbial efficiency and flow of bacterial OM (grams per day) linearly increased with sucrose addition. There was no change in total VFA concentration; however, proportion of acetate linearly decreased and proportion of butyrate linearly increased with sucrose addition. Total biohydrogenation and biohydrogenation of linoleic and linolenic acids linearly decreased with sucrose. Concentration of trans-12 18:1 linearly increased, whereas the concentrations of trans-10 and trans-9 18:1 linearly decreased with sucrose supplementation. Sucrose addition linearly decreased stearic acid concentrations in the effluent; however, there was no effect of sucrose on total trans fatty acid concentration. Sucrose may be more detrimental to the last step of biohydrogenation of VA. Bacterial community structure in the fermenter effluent samples was examined based on PCR amplicons containing the ribosomal intergenic spacer (RIS) and its flanking partial 16S ribosomal RNA gene. Cluster analysis showed no distinct banding patterns, though treatments tended to group together. The effects of sucrose on biohydrogenation and concentration of VFA may have been caused by a shift in microbial population by mechanisms that are independent of pH.
INTRODUCTION
Dairy products are one of the major sources of conjugated linoleic acid (CLA) in the human diet, and cis-9, trans-11 octadecadienoic acid (cis-9, trans-11 18:2) 41
contributes the greatest proportion of all CLA isomers (Parodi, 1997). Much attention has been given to CLA because of its anticarcinogenic properties (Parodi, 1997). Dietary management of dairy cows to increase CLA concentration in milk may be beneficial for human health and the dairy industry. The
CLA
in
milk
originates
from
CLA
produced
during
ruminal
biohydrogenation (BH) of linoleic acid and from desaturation of vaccenic acid (trans-11 18:1; VA) in the mammary gland. Most of the CLA in milk fat originates from VA (Kay et al., 2004); thus, most efforts have addressed ways to increase VA flow from the rumen. The factors affecting the flow of CLA isomers and trans-18:1 fatty acids (FA) to the duodenum during ruminal BH need to be elucidated in order to increase the CLA concentration in milk and meat of ruminants used for human consumption. Several researchers have reported that CLA is higher in milk from grazing animals when compared to animals fed TMR diets (Jahreis et al., 1997; Kelly et al., 1998b; Dhiman et al., 1999a; French et al., 2000). However, there are limited data addressing the basis for high concentration of CLA in milk from grazing cows. The higher concentration of CLA in milk fat from grazing animals could be a result of higher concentration of octadecatrienoic acid (18:3) in fresh forage, specific plant chemicals (Lee et al., 2003a), or higher concentration of soluble sugars in fresh plants (Kelly et al., 1998b). The concentrations of sugars are higher in fresh forages. Storing forages as silage or hay can decrease the concentration of FA (Boufaied et al., 2003a) and sugars in the forages (Van Soest, 1994). Traditional TMR diets have lower proportions of these 42
nutrients compared to fresh pasture. For instance, the concentration of sugar in a TMR is about 1.5 to 3.0% of DM (Hoover and Miller-Webster, 1998), whereas the concentration of sugars in fresh grasses averages 17% of DM, with fructosans being the major one, followed by sucrose, fructose, and glucose (Woolford, 1984). However, the effect of soluble sugars on BH has not been studied yet. The dual flow continuous culture system was used in the present study to assess the effect of diet on flow of BH intermediates without the confounding effects of variable or unpredictable passage rate, pH, fluid dilution, meal size, and feeding frequency. Because grazing cows have a higher concentration of VA in milk and almost 50% of the sucrose is lost during hay making, we hypothesized that fresh forage would increase concentration of VA in the effluent compared to the same forage in hay form and that this difference would be diminished by adding sucrose to the hay diet by changing the bacterial community profile
MATERIAL AND METHODS
Treatments and Experimental Design Fresh alfalfa samples (prebloom/early bloom) were obtained from a greenhouse at the Ohio Agricultural Research and Development Center (Wooster, OH). Samples were harvested randomly and immersed immediately in liquid nitrogen using a metal grid.
43
Samples were freeze-dried, ground at 1 mm, and stored at -20 oC. Purchased alfalfa hay was also ground at 1 mm, and stored at -20 oC. The percentage of nutrients from each forage source is shown in Table 4.1 Four continuous culture fermenters were used (Hoover et al., 1976) in a 4 X 4 Latin square design. Each period consisted of 7 d for adaptation and 3 d for sample collection. A priori analysis of sucrose from both forages showed that 4% of supplemental sucrose added to the hay treatment would match the percentage of sucrose in the fresh alfalfa; therefore, the treatments were: 1) fresh alfalfa, 2) alfalfa hay, 3) alfalfa hay plus 4% sucrose (as a percentage of DM), and 4) alfalfa hay plus 8% sucrose.
Continuous Culture Operation Ruminal inoculum was obtained from two ruminally cannulated Holstein cows fed a TMR diet. Within 20 min of collection, the inocula from both cows were pooled, squeezed through two layers of cheesecloth, and divided among the four fermenters. Diets were fed (100 g of DM) continuously throughout the day to provide steady state conditions. Sucrose was mixed once daily with hay prior to the initiation of the 24-h feeding cycle. The liquid and solid dilution rates were maintained at 10 and 5.5%/h, respectively, by regulation of filtrate removal rates and buffer input. The pH was maintained at 6.2 ± 0.1 by automatic administration of 5 N NaOH or 3 N HCl. The fermenters were continuously purged with N2 to maintain anaerobiosis, and the temperature was held at 39 oC. Temperature, acid and alkali use, filtrate flow, and pH were recorded and recalibrated, if needed, every 6 h. Solid and liquid effluents were weighed once daily to determine flow rates. 44
Sample Collection and Analysis One daily sample of effluent (1 L) was taken on d 8, 9 and 10 and composited by fermenter for analysis. The contents were blended using a Waring blender for 30 sec, and the effluent was strained through 2 layers of cheesecloth. Differential centrifugation was used to first separate the feed from the bacteria (15 min at 500 x g) and then to separate the bacteria from the liquid supernatant (15 min at 23,300 x g). An aliquot of the effluent was acidified using 3 ml of 6 N HCl per 50 ml of sample to stop fermentation prior to analysis of VFA (Firkins et al., 1990). Effluent and bacteria samples were frozen, freezedried, and stored at -20 oC. Both samples were analyzed for N by micro-Kjeldahl (AOAC, 1990), and effluent samples were analyzed for NDF and ADF (Van Soest et al., 1991). Forage sources were also analyzed for sucrose (Hall, 2000). The FA in the dried forages and effluent samples, stored at -20 oC, were methylated with 0.5 M sodium methoxide (10 min at 50 oC), followed by 5% methanolic HCl (10 min at 80 oC). Nonadecanoic acid (19:0) was used as an internal standard. Retention times and response factors were determined with methyl ester standards purchased from Nu-Check Prep (Elysian, MN; cat. No. GLC-60) and Matreya, Inc. (Pleasant Gap, PA; FIM-FAME-7). The 18:1 FA that were not available commercially (trans-6/8, trans-9, trans-12, trans-13, trans-15, cis-12, and cis-15) were identified by order of elution (Molkentin and Precht, 1995). The FA methyl esters were separated by GLC. The column was a fused silica capillary (SP-2380, 100 m x 0.25 mm id x 0.2-µm film thickness; Supelco Inc., Bellefonte, PA). Nitrogen was used as the carrier gas. Detector and injector temperatures were set at 250 and 220 oC, respectively, and the split ratio was set at 100:1. Oven temperature was set at 160 oC for 10 min, increased by 3.0 45
o
C /min to 180 oC, held for 60 min, increased by 5.0 oC /min to 220 oC, held for 50 min,
and decreased by 20 oC /min to 160 oC for 1 min. The BH of FA was calculated as described by Wu and Palmquist (1991).
DNA Extraction and PCR One sample per fermenter per period was taken for DNA extraction. and frozen at –80°C. Total community DNA was extracted from the effluent samples using the bead beating method (Yu and Mohn, 1999) followed by purification of the DNA using a QIAamp column (Qiagen, Valencia, CA). The DNA samples were used in PCR amplification with primers S926f (5'-CTYAAAKGAATTGACGG-3') and L189r (5'TACTGAGATGYTTMARTTC-3'). The resultant ribosomal DNA-RIS PCR products contain the complete RIS and parts of the flanking ribosomal DNA (rDNA) genes (ca. 600 bp of 16S rDNA and 190 bp of 23S rDNA). The PCR conditions were as follows: initial denaturation at 94°C for 3 min, annealing at 45°C for 1.5 min, and extension at 72°C for 2.5 min. Subsequent cycles consisted of a 1.5-min denaturation step at 94°C, a 1.5-min annealing step at 45°C, and a 2-min extension step at 72°C. After 30 cycles, there was a final 7-min extension step at 72°C.
Phylogenetic analysis based on RIS-Length Polymorphism PCR amplified rDNA-ribosomal intergenic spacer (RIS) fragments were separated on a 4% polyacrylamide (37:1) gel, and the gel was stained with GelStar nucleic acid stain (BioWhittaker, Inc., Walkersville, MD). The ribosomal intergenic spacer length polymorphism (RIS-LP) banding patterns were documented using a 46
FluorChem Imaging System (Alpha Innotech Corporation, San Leandro, CA). The gel image was then exported into the GelCompar II analysis software in the BioNumerics package (BioSystematica, Devon, UK) for cluster analysis and dendrogram construction. The UPGMA method was used by the software for dendrogram construction.
Statistical Analysis Data were analyzed by the PROC MIXED procedure of SAS (SAS, 2004), according to the following model:
Yijk = µ + Ti + cj + pk + εijk, where:
Yijk = dependent variable for treatment i on fermenter j and period k; µ = overall mean; Ti = fixed effect of treatment i; cj = random effect of fermenter j; pk = random effect of period k; and εijk = residual error associated with the ijk observation.
Pre-planned contrasts were used to compare the treatment effects: 1) fresh alfalfa versus alfalfa hay without sucrose, 2) linear effect for concentration of sucrose, and 3) quadratic effect for concentration of sucrose. Significance was declared at P < 0.05.
47
RESULTS
Digestibility Measures True OM digestibility was 14% higher (P = 0.09) in fresh alfalfa than alfalfa hay (Table 4.2). The digestibility of N tended (P = 0.11) to be higher and digestibilities of the NDF and ADF fractions were higher for fresh alfalfa than hay. Flow of bacterial OM was higher for fresh alfalfa; however, no difference was found on efficiency of microbial protein synthesis between alfalfa sources. Addition of sucrose tended (P = 0.09) to decrease true OM digestibility. There was no effect of sucrose on NDF or ADF digestibilities; however adding 8% sucrose numerically decreased NDF digestibility of alfalfa hay by 14%. Sucrose addition did not affect true N digestibility. However, there was a positive linear effect of sucrose concentration on efficiency of microbial protein synthesis and flow of bacterial OM in the effluent.
Volatile Fatty Acids The VFA concentrations are presented in Table 4.3. There was a quadratic trend (P = 0.12) for the concentration of total VFA to linearly decrease with increasing sucrose level. Addition of sucrose linearly decreased proportion of acetate. However, no effects were observed on proportion of propionate or the acetate to propionate ratio.
48
Butyrate and valerate proportions linearly increased with increasing sucrose addition, whereas proportions of isobutyrate and isovalerate linearly decreased as sucrose level increased. Only proportions of isobutyrate and isovalerate were affected by alfalfa source, being higher for fresh alfalfa.
Fatty Acids and Biohydrogenation Total BH and the BH of oleic, linoleic, and linolenic acids were higher for fresh alfalfa than alfalfa hay with no supplemental sucrose. Increasing sucrose concentration linearly decreased total BH and BH of linoleic and linolenic acids, but there was no difference in the BH of oleic acid (Table 4.4). The concentrations of 16:0, cis 18:1, 18:2, and 18:3 FA in the effluent were lower for fresh alfalfa compared to alfalfa hay. Percentages of 18:0 and trans FA were higher for fresh alfalfa (Table 4.5). Addition of sucrose linearly increased the percentage of palmitic, linoleic, and linolenic acids, and linearly decreased the percentage of 18:0, reflecting lower BH. There was no effect of sucrose on the percentage of total cis and trans 18:1 FA in the effluent. However, sucrose addition tended (P = 0.11) to have a quadratic effect on the percentage of CLA, with the lowest CLA proportion observed at the intermediate level of sucrose. Concentrations of trans-6/8, VA, trans-13, trans-15, and cis-15 18:1 FA were higher in the effluent of fermenters receiving fresh alfalfa compared to those given alfalfa hay with no supplemental sucrose (Table 4.6). However, concentrations of trans10, cis-9, and cis-11 18:1 FA were lower for those receiving fresh alfalfa.
49
The concentration of trans-9 tended (P = 0.09) to be higher and cis-12 tended (P = 0.15) to be lower for fresh alfalfa compared to hay. No difference (P > 0.05) in the concentration of trans-12 18:1 FA was observed between the two forage sources. Addition of sucrose linearly decreased the proportions of trans-6/8 and trans-10 18:1 FA. The trans-9 and cis-12 18:1 FA tended (P < 0.15) to decrease with increasing sucrose addition. The trans-12 was the only intermediate of BH that linearly increased with sucrose addition. Addition of sucrose had no effect on the concentrations of VA, trans-13, trans-15, cis-9, cis-11, or cis-15 18:1 FA.
RIS-LP Analysis Distinct RIS-LP banding patterns were obtained from gel electrophoresis of the PCR amplicons (Figure 4.1). Cluster analysis showed that the treatments tended to group together, but no distinct pattern was detected because samples taken in the same period tended to cluster together irrespective of the treatment group the samples were collected from. However, all the sucrose treatments (hay plus 4 and 8% sucrose) grouped together. RIS-LP followed by cluster analysis based on pair-wise comparison of the banding patterns obtained from gel electrophoresis of rDNA-RIS amplicons provides a reliable and fast method for comparison of microbial community profiles in environmental samples.
50
The banding patterns suggest that sucrose supplementation had a strong influence on bacterial community structure. Because each period was initiated by a new inoculation, and the fermentations were controlled for pH and continuous feeding of forages, periods had also an effect on the banding patterns. However, minor alterations in the community profile might not be easily detected because of the presence of other bacterial species in larger numbers. The use of genera- and species specific primers would allow more specific monitoring of the minor population changes.
DISCUSSION
We reported the isolated effect of sucrose on BH and FA concentrations in the effluent of continuous fermenters since pH, passage rates, and nutrient composition did not vary among the hay treatments. Fresh alfalfa was used as a positive control to show how VA and other intermediates of BH accumulate in the effluent and also to compare the differences in BH attributed to forage conservation.
Digestibility Although there is a large variation in ruminal NDF digestibilities reported in the literature, the NDF digestibilities observed in this trial are in agreement with those form a previous study (Bach et al., 1999). The variation observed in NDF digestibility among
51
experiments using continuous culture fermenters may be caused by differences in forage sources and maturity, pH, inoculum, and liquid and solid passage rates (Qiu et al., 2004). True OM digestibilities for hay and hay plus 4% sucrose were very similar, whereas the value for hay plus 8% sucrose was 16% lower than hay alone. Sucrose addition has been reported to reduce the rates of fiber fermentation in situ (Huhtanen and Khalili, 1992). Addition of 8% sucrose negatively affected true OM digestibility of hay, most likely reflecting the numerical decrease in NDF and ADF digestibilities by a pHindependent mechanism. A decrease in NDF digestibility by addition of soluble sugars with pH higher than 6.2 has been previously reported (Piwonka and Firkins, 1996). Some cellulolytic bacteria are able to utilize soluble sugars as well as fiber (Huhtanen and Khalili, 1992), changing their metabolism accordingly to substrate availability (Russell and Baldwin, 1978). We postulate that sucrose decreased nutrient digestibility by shifting either bacterial ecology or metabolism, or both. Russell and Baldwin (1979) reported alterations in bacterial growth rates with a combination of different carbohydrates. Therefore, sucrose addition may have favored growth of bacteria species that have faster growth rates when fermenting this sugar. Furthermore, sucrose may decrease fiber digestibility by decreasing fibrolytic enzyme activity (Huhtanen and Khalili, 1992; Piwonka and Firkins, 1993), possibly explaining the numerical decrease in fiber digestibility in this trial.
52
Bacterial OM flow increased 14% and microbial efficiency increased 35% from 0 to 8% sucrose. A similar increase in efficiency has been reported in Holstein heifers fed forage supplemented with dextrose (Piwonka et al., 1994). We reported higher values for microbial efficiency (Table 4.2), likely caused by the lower values observed for true OM digestibilities. Meng et al. (1999) observed a higher microbial efficiency for nonfibrous carbohydrate sources compared to fibrous sources in continuous culture fermenters at higher dilution rates (0.15 and 0.20/h). However, these observations were different at lower dilution rates. Supplemental sucrose provided a more readily-available energy source, contributing to higher efficiency and bacterial OM flow. Nutrients were more digestible in fresh alfalfa than hay. Harvest of alfalfa to represent the selection of the immature plants during grazing conditions resulted in a lower percentage of NDF compared to hay. Higher nutrient digestibilities for fresh alfalfa resulted in higher bacterial OM flow. Sucrose content of fresh alfalfa was two percentage units higher than for alfalfa hay (Table 4.1). A possible synchronism between N and sucrose release from the plant matrix and fermentation may have contributed to the almost 2-fold increase in true N digestibility for fresh alfalfa. Also, fresh forages have higher concentrations of other rapidly fermented sugars that are lost during wilting and higher concentration of more digestible protein (Van Soest, 1994).
Volatile Fatty Acids Total VFA concentrations observed in this experiment were similar to previous reports that used continuous culture fermenters (Bach et al., 1999) and grazing animals 53
(Reis and Combs, 2000). The reason for the quadratic effect of sucrose addition on VFA concentration is not certain; such a quadratic response was not observed elsewhere in this trial, except for a quadratic trend observed for the percentage of CLA in the effluent. Degradation of fiber in the rumen is associated with a higher molar proportion of acetate to propionate and higher methane production (Van Soest, 1994). Changes in fermentation patterns likely reflect shifts in bacterial population that respond to changes in fermentable substrates. The decrease in acetate concentration as sucrose concentration increased is consistent with the numerically lower NDF digestibility observed for the highest concentration of sucrose. The change in molar proportion of butyrate was the most pronounced of all, with a 22% increase in concentration at the highest concentration of sucrose. Crawford et al. (1980) also reported reciprocal changes in acetate and butyrate concentrations by varying liquid dilution rate. Addition of sucrose also linearly increased the concentration of valerate, whereas isobutyrate and isovalerate concentrations were decreased. When taken together, these observations support our hypothesis that addition of sucrose caused a shift in the bacterial population. Molar proportion of butyrate has been reported to increase with addition of soluble sugars (Piwonka and Firkins, 1996) and concentrate feeds (Reis and Combs, 2000). Soluble sugars are the main carbohydrate substrate for holotrichs (Van Soest, 1994). However, continuous culture fermenters are associated with a decrease in protozoal population and even greater reductions in protozoa numbers were observed with solid dilution rates higher than 0.04/h (Hoover et al., 1976). Therefore, the higher butyrate concentration observed in this experiment was not likely caused by an increase in protozoal cells. The higher butyrate concentration could be attributed to a change in 54
fermentation pathways to accommodate the higher flux of hydrogen from the rapidly fermented sugar source (Piwonka and Firkins, 1996). Also, the higher molar proportion of butyric acid could be attributed to its synthesis from lactate by Megasphaera elsdenii (Klieve et al., 2003). No difference was observed for the major VFA between fresh alfalfa and hay; however, isobutyrate and isovalerate were higher for fresh alfalfa compared to hay whereas these branched-chain VFA (BCVFA) linearly decreased with sucrose addition. Valerate was not different between fresh and hay but linearly increased with sucrose. The BCVFA resulted from fermentation of branched-chain amino acids. The change in BCVFA likely is related to the difference in N digestibility observed in this trial.
Biohydrogenation of Fatty Acids The BH values observed in this experiment are similar to those reported by Qiu et al. (2004) and Wu and Palmquist (1991). However, other researchers have reported BH values for 18:2 and 18:3 higher than 80% for in vitro (Loor et al., 2003) and in vivo studies (Sackmann et al., 2003). Source of FA, DM intake, concentration of FA in the diet, and ruminal pH are some of the factors that may contribute to the differences in estimates of BH among experiments. The decrease in BH of FA in alfalfa hay by sucrose addition may be associated with a decrease in cellulolytic microbes. Any shift in bacterial population in favor of noncellulolytic species could inhibit the BH of FA, because cellulolytic bacteria play a major role in ruminal BH (Harfoot and Hazlewood, 1997).
55
The linear increase in efficiency of microbial protein synthesis with sucrose addition may be partially associated with a decrease in BH. Biohydrogenating bacteria hydrogenate FA as an anti-toxic mechanism (Harfoot and Hazlewood, 1997), which may increase their maintenance energy cost. If BH is inhibited, the additional energy cost of maintenance (used to induce genes and synthesize proteins to hydrogenate the double bonds of FA) could be diverted to more growth-efficient processes; thereby, improving microbial efficiency as observed in this study. The same rationale could be used to explain the higher BH for fresh alfalfa compared to hay, because NDF digestibility was 20% higher in fresh alfalfa. The BH of linoleic and linolenic acids was 14.6 and 32.9% higher, respectively, for fresh alfalfa when compared with alfalfa hay with no supplemental sucrose. Higher BH of 18:2 and 18:3 also has been reported for fresh grass compared to hay (Boufaied et al., 2003b). The composition of individual FA between grass and legumes is variable, depending on season, species, and N fertility (Boufaied et al., 2003a). Nutrient digestibility was higher, especially the fiber fraction, for fresh alfalfa. The lower degradability observed for hay may have physically trapped the FA within the organic matrix, decreasing release and, therefore, BH of 18:2 and 18:3 (Boufaied et al., 2003b). Conversely, the flow of bacterial OM was 14% lower for hay, indicating that, for a similar bacterial mass, the biohydrogenating microflora were much more active in fresh alfalfa.
56
The BH of 18:3 from fresh plants has been reported to be in the range of 80 to 90% (Loor et al., 2003; Sackmann et al., 2003). The lower BH of 18:3 for fresh alfalfa observed in the present study could be attributed to the lower digestibility of the OM fraction observed in our trial. If the plant matrix has a lower digestibility, release of FA as substrate for BH would be diminished. The higher CLA percentage in milk of grazing cows is largely a result of endogenous synthesis by ∆9-desaturase, with intestinally absorbed VA as a substrate (Kay et al., 2004). The BH of 18:3 results in VA as an intermediate but not CLA (Harfoot and Hazlewood, 1997). High concentrations of VA in the duodenal digesta have been reported for ruminants consuming diets having either 18:2 or 18:3 FA as the major unsaturated FA (UFA) (Sackmann et al., 2003). Therefore, factors regulating VA synthesis and flow out of the rumen must be defined to improve our understanding of ruminal BH. The fresh alfalfa diet had a 59% higher percentage of VA in the effluent, consistent with our hypothesis. This was most likely due to the higher percentage of UFA in fresh forage (Table 4.1). Although there was a linear decrease in total BH and percentage of 18:0 in the effluent as sucrose concentration increased, no differences were detected in the percentage of VA and total trans FA. Thus, supplemental sucrose decreased total BH and disappearance of 18:2 and 18:3, while decreasing, more than proportionally, the appearance of 18:0. These effects caused similar proportions of trans FA to be observed in the effluent of all the hay treatments.
57
Whereas total BH was decreased by 20% when 8% sucrose was added, synthesis of VA only decreased 11% (data not shown). Therefore, sucrose may be more detrimental to the last BH step (synthesis of 18:0 from trans 18:1 FA) than to the isomerization/reduction step of 18:2 and 18:3 to trans 18:1 FA. The concentration of VA in the effluent of continuous culture fermenters fed the fresh alfalfa was lower than that reported by Loor et al. (2003). These authors reported ca. 16% of VA in the effluent from red clover and orchardgrass; whereas, we observed 3 to 9% VA. Our lower values were likely a result of: 1) lower OM digestibility observed in this trial, 2) type of forage and conservation method, 3) passage rate, and 4) constant pH. Feeding an early vegetative state of alfalfa could have increased the amount of VA formed because of its higher concentration of UFA. Because grasses have a higher proportion of 18:3 than legumes (Boufaied et al., 2003a), research using grasses in continuous culture reported higher concentrations of VA in the effluent (Loor et al., 2003). Also, alfalfa has both a lower proportion of total FA and 18:3 compared with red clover (Boufaied et al., 2003a). The concentration of VA in the effluent of the hay treatments was half of that for the fresh alfalfa. Conserved forages (hay) have a lower concentration of FA (Boufaied et al., 2003a), which decreases the amount of substrate for BH, thus diminishing the concentration of VA in the rumen.
58
Higher solid and liquid passage rates may increase the flow of intermediates of BH (Qiu et al., 2004) (i.e., a higher flow rate decreases the opportunity for complete BH). In the experiment of Loor et al. (2003), the liquid and solid passage rates were 0.18/h and 0.07/h, respectively. Those values were higher than used in this trial (0.10/h and 0.05/h). Also, diurnal variation in ruminal pH of grazing animals with values falling below 6.2 (Kolver and de Veth, 2002) may inhibit BH of trans 18:1 to 18:0 (Qiu et al., 2004). The result would be higher VA concentration in the ruminal fluid.
Intermediates of Biohydrogenation The proportions of the intermediates of BH did not follow a specific pattern with sucrose addition. The cause for a linear decrease in trans-6/8, trans-9, and trans-10, and an increase in trans-12 18:1 FA was not apparent. Changes in the duodenal flow of trans-9, trans-12, and trans-10 varied with the proportion of forage in the diet (Sackmann et al., 2003). Also, the initial concentration of UFA affects the rates of BH for 18:2 and 18:3 (Czerkawski, 1967). The initial concentration of UFA in the fermenters was less than 1 mg/ml. Therefore, we could isolate the effect of sucrose on the parameters studied from the effects of pH and high initial amounts of 18:2 and 18:3. Yu and Mohn (2001) developed a composite method for investigating bacterial community structure in an aerated lagoon. The method is based on analyses of PCR amplicons containing the RIS and its flanking partial 16S rRNA gene. Community structure similarity was determined based on RIS length polymorphism. The 16S-23S rDNA-RIS region has a highly variable length, and can hence be used as a marker to 59
distinguish different bacterial species. We postulated that there was a shift in the microbial population which affected nutrient digestibilities, BH, and VFA concentrations. The RIS-LP analysis may resolve different genera or species of bacteria as influenced by sucrose. We observed that sucrose caused a change in the bacterial community although there was a period effect influencing the analysis. This shift in bacterial population may have been associated with the observed effects of sucrose on some of the intermediates of BH. Different trans-18:1 isomers (positions 6 to 16) can be formed from cis-9 18:1 (Mosley et al., 2002). However, there is not data on the literature regarding how all cis and trans isomers are formed during BH or how the major ruminal microbial species involved. Sucrose linearly decreased the proportion of trans-10 18:1 in the effluent, resulting in a 26% increase in the ratio of VA to trans-10 18:1. Milk fat depression has been associated with an increase in trans-10 18:1 in milk (Griinari et al., 1998). This fatty acid is probably an intermediate of BH of trans-10, cis-12 18:2, the CLA isomer responsible for inhibiting ∆9-desaturase (Park et al., 2000). The exact pathway and microbial species involved in the BH of 18:2 to trans-10 18:1 is not known. However, manipulation of ruminal BH towards an increase in the ratio of VA to trans-10 18:1 would benefit herds experiencing milk fat depression. We observed a higher ratio of VA to trans-10 18:1 in the effluent of fresh alfalfa compared to hay (33.7 vs. 6.73). Sackmann et al. (2003) observed a linear decrease in this ratio as the proportion of forage increased in the diet of steers. These observations agree with the proposed BH pathway in which VA is a major intermediate of BH of 18:3 (Hazlewood et al., 1976). Those authors also observed synthesis of cis- and trans-15 18:1 60
from BH of 18:3. Besides the much higher proportion of those two FA in the effluent of the fresh alfalfa treatment (Table 4.6), trans-13 18:1 was the second most prevalent intermediate of BH. We are not aware of any proposed BH pathway that accounts for the formation of this FA. Loor et al. (2003) have also reported trans-13 18:1 as the second major 18:1 FA in the effluent of continuous culture fermenters. There is still a need to understand how the major factors (diet composition, ruminal parameters, and population of microorganisms) of ruminal BH are correlated.
CONCLUSIONS
Sucrose does not increase VA concentration during BH of hay; however, our data support the hypothesis that addition of sucrose changed microbial populations; as concentration of sucrose increased, total BH decreased by a pH-independent process. Synthesis of VA and trans FA were not changed by sucrose addition. The final step of BH, reduction of trans 18:1 to 18:0, may be more sensitive to the suppressing effect of sucrose supplementation. The VA was the major octadecenoic FA in the effluent of all treatments, followed by trans-13 18:1. Adding sucrose to hay shifted the intermediates of BH independently of pH and concentrations of UFA. Fresh alfalfa had a higher digestibility of NDF and ADF and tended to have a higher true OM digestibility than alfalfa hay. Addition of sucrose tended to decrease true OM digestibility of hay and improve efficiency of microbial protein synthesis. Sucrose addition decreased acetate and increased butyrate concentrations. Total BH of FA from 61
fresh alfalfa was higher than for alfalfa hay. Future research will link the sucrose-induced changes in BH with changes in metabolites and a shift in microbial species in the rumen.
62
Table 4.1: Nutrient composition of fresh alfalfa and alfalfa hay.
Item
Fresh Alfalfa
Alfalfa Hay
DM
21.2
91.5
------------------- % of DM --------------------OM
90.5
92.2
CP
23.3
20.2
NDF
31.1
37.9
ADF
23.5
26.8
Sucrose
8.64
6.78
Total fatty acids (FA)
2.28
1.07
Individual FA
------------------- % of FA -------------------
16:0
20.8
30.9
18:0
3.40
5.41
cis-9 18:1
2.65
3.78
18:2
17.0
19.7
18:3
48.6
24.1
63
Table 4.2: Nutrient digestibility in continuous culture fermenters receiving fresh alfalfa or alfalfa hay supplemented with three concentrations of sucrose.
Treatment Fresh Alfalfa Item
Hay + Sucrose (%) 0
4
8
Probability of Contrasts SE
Fresh vs Hay
Linear Quadratic
Digestibility, % OM1
46.1
39.5
40.0
33.0 2.6
0.09
0.09
0.25
NDF
39.1
31.2
31.9
26.8 2.6
0.03
0.18
0.29
ADF
40.2
29.7
30.3
25.0 2.4
0.01
0.28
0.15
NBN
48.1
29.5
23.8
24.5 5.6
0.11
0.36
0.50
Bacterial OM, g/d
14.8
12.3
12.5
14.4 0.8
0.02
0.05
0.25
Efficiency 37.6 1 True OM digestibility.
36.7
36.3
56.4 6.2
0.92
0.04
0.17
2
3
2
Non-bacterial nitrogen.
3
Grams of microbial N produced per kilogram of OM truly digested.
64
Table 4.3: Proportions of volatile fatty acids in continuous culture fermenters receiving fresh alfalfa or alfalfa hay supplemented with three concentrations of sucrose.
Treatment Fresh Hay + Sucrose Alfalfa (%) Item Total VFA, mM
100.9
Probability of Contrasts Fresh vs Linear Quadratic Hay
0
4
8
SE
97.2
82.8
88.5
6.2
0.58
0.22
0.12
-------- mol/100 mol ------Acetate (A)
73.6
74.8
72.8
71.3
0.9
0.38
0.02
0.81
Propionate (P)
14.8
13.5
14.1
14.9
0.8
0.28
0.26
0.96
Butyrate
8.3
8.7
10.1
11.2
0.4
0.22
< 0.01
0.43
Isobutyrate
0.58
0.38
0.25
0.14 0.04
< 0.01
< 0.01
0.73
Valerate
1.37
1.61
1.98
2.20
0.1
0.20
< 0.01
0.65
Isovalerate
1.31
1.02
0.75
0.37 0.10
0.05
< 0.01
0.60
A:P
4.99
5.58
5.20
4.92 0.33
0.23
0.18
0.90
65
Table 4.4: Biohydrogenation (BH) of 18:1, 18:2, 18:3 and total BH of fatty acids in continuous culture fermenters receiving fresh alfalfa or alfalfa hay supplemented with three concentrations of sucrose.
Treatment Fresh Alfalfa BH1
Hay + sucrose (%) 0
4
66.0
45.0
41.2
35.9 2.5 < 0.01
0.01
0.77
18:1 cis-9 28.4
12.8
13.4
6.3
0.36
0.52
18:2
55.3
47.2
43.8
38.8 3.3 < 0.01
< 0.01
0.66
18:3
78.4
52.6
50.1
44.5 2.3 < 0.01
0.02
0.50
Total BH
1
8
SE
Probability of Contrasts Fresh Linear Quadratic vs Hay
5.2
0.05
Percentage of fatty acids that disappeared corrected for the proportion of 18 carbon fatty
acids fed and in the effluent (Wu and Palmquist, 1991).
66
Table 4.5: Fatty acid profile in effluent from continuous culture fermenters receiving fresh alfalfa or alfalfa hay supplemented with three concentrations of sucrose.
Treatment Fresh Hay + sucrose (%) Alfalfa Item1
0
4
8
SE
Probability of Contrasts Fresh vs Hay Linear Quadratic
16:0
21.1
33.6
34.8
35.8
0.8
< 0.01
0.04
0.88
18:0
39.3
27.6
24.8
22.0
1.9
< 0.01
0.05
0.99
trans 18:12
14.7
6.0
7.2
6.2
1.3
< 0.01
0.88
0.43
cis 18:13
4.66
6.31
6.04
6.15 0.43 < 0.01
0.87
0.60
18:2
8.2
12.3
12.8
13.8 0.82 < 0.01
0.03
0.68
CLA4
0.65
0.81
0.57
0.93 0.14
0.44
0.53
0.11
18:3
11.4
13.4
14.8
15.1 0.54
0.01
0.03
0.45
UFA5
39.6
38.7
40.4
42.2 2.03
0.67
0.14
0.96
SFA6
60.4
61.2
59.6
57.8 2.03
0.67
0.14
0.96
1
Percentage of total (16 and 18-carbon fatty acids) long chain fatty acids.
2
Total trans 18:1 fatty acids.
3
Total cis 18:1 fatty acids.
4
Geometric and positional isomers of conjugated linoleic acid.
5
Unsaturated fatty acids.
6
Saturated fatty acids. 67
Table 4.6: Fatty acid profile of the intermediates of biohydrogenation in effluent from continuous culture fermenters receiving fresh alfalfa or alfalfa hay supplemented with three concentrations of sucrose. Treatment Fresh Alfalfa Item1
Hay + sucrose (%) 0
4
8
SE
Probability of Contrasts Fresh Linear Quadratic vs Hay
trans 18:1 trans-6/8
0.48
0.34
0.24
0.06
0.05
0.04
0.01
0.42
trans-9
0.33
0.24
0.19
0.12
0.03
0.08
0.07
0.84
trans-10
0.27
0.55
0.33
0.33
0.13
< 0.01
< 0.01
0.71
trans-11
9.12
3.74
4.11
3.01
1.03
< 0.01
0.60
0.49
trans-12
0.67
0.65
0.90
1.30
0.08
0.80
< 0.01
0.42
trans-13
2.77
0.84
0.89
1.04
0.22
< 0.01
0.35
0.94
trans-15
1.08
0.43
0.83
0.48
0.24
0.03
0.75
0.24
cis -9
2.01
3.85
3.76
4.01
0.24
< 0.01
0.63
0.51
cis -11
0.93
1.42
1.31
1.41
0.13
< 0.01
0.96
0.38
cis-12
0.47
0.61
0.57
0.48
0.10
0.13
0.15
0.77
cis-15
1.34
0.38
0.36
0.30
0.07
< 0.01
0.31
0.79
cis 18:1
1
Percentage of the total long chain fatty acids. 68
2 0
4 0
6 0
8 0
1 0 0
0% – P1 0% - P2 4% - P2 8% - P1 8% - P3 8% - P2 4% - P3 8% - P4 4% - P1 4% - P4 FA – P3 FA – P4 0% – P3 FA – P1 FA – P2 0% - P4 Ladder Ladder
Figure 4.1: Comparison of bacterial community structure in fermenters samples; dendogram shows cluster analysis performed based on samples similarities of the communities. P = period 1 to 4; 0%, 4% and 8% = percentage of sucrose supplementation on the hay treatments; FA = fresh alfalfa. The length of scale depicts the percentage similarities between different lanes.
69
CHAPTER 5
EFFECT OF SUCROSE, pH, AND FORAGE CONSERVATION ON THE IN VITRO BIOHYDROGENATION OF FATTY ACIDS
ABSTRACT An in vitro experiment was performed to determine differences in biohydrogenation (BH) of fatty acids (FA) from fresh alfalfa and alfalfa hay when supplemental sucrose and media pH were varied. We also developed a multi-compartmental model to estimate pool size and flux of vaccenic acid (VA) during BH of FA in fresh alfalfa. Alfalfa samples were inoculated with rumen fluid using two incubation buffers. Samples were incubated for 0, 1, 2, 3, 4, 6, 9, and 12 h; pH was measured and tubes were put in ice and stored until analysis. Disappearance rates (DR) of linoleic acid (18:2) and linolenic acid (18:3) were estimated by PROC NONLIN of SAS (single pool, first-order kinetic model) and SAAM II (multiple pools, first-order kinetic model). Both methods gave similar estimates for the DR of 18:2 and 18:3 as well the temporal pool size of VA. The DR of 18:2 and 18:3 was slower (P < 0.05) for both forage sources when substrates were incubated in the weak buffer (WB). However, DR of 18:3 was higher from fresh alfalfa than alfalfa hay.
70
There was a trend (P = 0.11) for an interaction between buffer and forage source for the DR of 18:3. There was no effect (P > 0.05) of sucrose addition on the DR of 18:2 and 18:3. Also, there was no effect on the BH of VA estimated by multiple pools, firstorder kinetic model. Because we could estimate fluxes, as well as mass of the VA pools, more information is generated from the data when using multiple pools compared to single pool, first-order kinetic model.
INTRODUCTION The ruminal BH process is responsible for alteration of dietary FA in the rumen, causing a decrease in the proportion of unsaturated FA (UFA) and the appearance of conjugated linoleic acid (CLA) isomers and trans FA in the duodenum. The 18:2 is the precursor of the synthesis of CLA and trans FA during BH. Dairy products are the major source of CLA in human diets (Kelly et al., 1998a), and cis-9, trans-11 CLA contributes the greatest proportion of the CLA isomers in the human diet (Parodi, 1997). Much attention has been given to CLA because of its potential beneficial effects on metabolism and health (Parodi, 1997; Ip et al., 2002). Grazing cows produce milk with a higher concentration of CLA (Jahreis et al., 1997; Kelly et al., 1998c; Dhiman et al., 1999a; French et al., 2000); however, limited data address the basis for this increased concentrations. Kay et al. (2004) reported that the majority of milk CLA from cows fed pasture was derived from the desaturation of trans11 18:1 (vaccenic acid) by stearoyl-CoA desaturase in the mammary gland. Whereas desaturation of VA in the mammary gland contributes to the majority of CLA in milk fat 71
(Griinari et al., 2000; Kay et al., 2004), less is understood concerning factors that regulate the amount of VA produced in the rumen. Understanding BH and the factors involved in its regulation may allow for greater control of the ruminal BH, increased product quality, and increased consumer acceptance of dairy products. The higher amount of CLA in milk fat from grazing cows has been proposed to be a result of the higher levels of soluble sugars in fresh plants compared with preserved forages (Kelly et al., 1998c; French et al., 2000). Ensiling and drying decrease the amount of FA, especially 18:3, (Chilliard et al., 2000; Boufaied et al., 2003a) and sugars in forages (Van Soest, 1994). Concentrations of sugars in grasses average about 17% of DM; fructosans occur in major amounts, followed by sucrose, fructose, and glucose (McDonald, 1981; Woolford, 1984). The 18:3 is the major FA in plant tissue, constituting about 75 to 95% of FA from monogalactosyldiglycerides in alfalfa leaf tissue (Butler and Bailey, 1973). During ruminal BH of 18:3, VA but not CLA is an intermediate (Wilde and Dawson, 1966; Harfoot and Hazlewood, 1997). Additionally, Kay et al. (2004) showed that the majority of milk CLA is synthesized from VA in grazing cows. The higher concentration of CLA in grazing cows must be a consequence of the higher amounts of VA reaching the mammary gland; thus, most research effort has addressed ways to increase VA flow from the rumen.
72
Kolver and de Veth (2002) reported mean ruminal pH lower than 6.2 for cows fed pasture-based diets. A lower pH may inhibit the final step of BH, favoring formation of monoenoic trans 18:1 FA (Qiu et al., 2004). Therefore, the interactions among pH, concentrations of 18:3, and sucrose may play a role in the overall BH process towards VA accumulation and flow to the duodenum of grazing animals. We postulated that BH rate of UFA in fresh alfalfa is greater than in alfalfa hay and that decreasing pH will decrease the BH rate independently of the forage preservation method. We also postulated that there is an interaction between pH and sucrose supplementation which influences BH of 18:3 from alfalfa hay.
MATERIAL AND METHODS
Treatments and Incubation Procedure An in vitro experiment was performed to determine differences in BH of FA from fresh alfalfa and alfalfa hay when supplemental sucrose and media pH were varied. Analysis of sucrose from both forages showed that 4% (w/w) of supplemental sucrose added to the hay treatment would match the percentage (w/w) of sucrose in the fresh alfalfa; therefore, the treatments were: 1) fresh alfalfa, 2) alfalfa hay, 3) alfalfa hay plus 4 % sucrose, and 4) alfalfa hay plus 8% sucrose. The percentages of nutrients from each forage source are shown in Table 5.1.
73
Fresh alfalfa samples (prebloom/early bloom) were obtained from a greenhouse at the Ohio Agricultural Research and Development Center (Wooster, OH). Samples were harvested randomly and immersed immediately in liquid nitrogen using a metal grid. Samples were freeze-dried, ground to 1 mm, and stored at -20 oC. Purchased alfalfa hay also was ground to 1 mm and stored at -20 oC. Each substrate was incubated (0.5 g of dried forage) with a strong (SB) or a weak (WB) buffer in two sets (blocks) of incubations. The pH was varied by using different molarities of buffer (SB, 0.4 M/L NaHCO3; WB, 0.2 M/L NaHCO3); sodium molarities of the buffers were equalized by adding NaCl to the weak buffer (Piwonka and Firkins, 1996). Ruminal fluid was obtained from 2 ruminally-cannulated dairy cows, one consuming an alfalfa hay diet and the other a TMR diet consisting of 60% (w/w) forage (alfalfa hay plus corn silage), allowing for a greater diversity of microorganisms to be present during the incubation. Ruminal contents were strained through 2 layers of cheesecloth to allow small particles to remain in the inoculum. The ruminal fluids from the 2 cows were mixed and taken to the laboratory under anaerobic conditions at 39 ºC. The ruminal fluid was blended to dislodge particle-associated bacteria and strained through 8 layers of cheesecloth. We have previously observed that this procedure allows small particles to be present in the inoculum. The inoculum was divided into two equal volumes and either SB or WB was added at 20% of total volume. Inocula were gassed with CO2 and blended while 30 mL was transferred to each of the centrifuge tubes containing the various substrates.
74
Each set of 8 treatments was incubated for 0, 1, 2, 3, 4, 6, 9, 12, and 24 h. These times were chosen to allow identification of peak concentrations of CLA and VA in early hours and concomitantly with a lower standard error of means previously reported (Ribeiro and Eastridge, 2004). After incubation, the pH was measured and the tubes were placed on ice to stop fermentation. The tube contents were transferred to aluminum pans, frozen, freeze-dried, and kept at -80 ºC until analyses.
Analysis of Fatty Acids The FA of substrates and residues of incubation were methylated with 2 mL of 0.5 M/L sodium methoxide (10 min at 50 oC) followed by 3 mL of 5% methanolic HCl (10 min at 80 oC) as described by Krame et al. (1997). Methyl esters were separated by GLC using a HP 5890 Series II gas chromatograph (Hewlett Packard Co., Palo Alto, CA). The column was a fused silica capillary (SP-2560, 100 m x 0.25 mm id x 0.2-µm film thickness; Supelco Inc., Bellefonte, PA). Helium was used as carrier gas. Detector and injector temperatures were 260 oC, and split ratio was 80:1. Oven temperature was 166 oC for 39 min, increased by 10.0 oC /min to 240 oC, held for 10 min, increased by 3.0 o
C /min to 245 oC, and held for 10 min. The temperature program used was optimized to
separate most of the 18:1 FA in the first isothermal range as described by Molkentin and Precht (1995).
75
Nonadecanoic acid was used as an internal standard. Retention times and response factors were determined with methyl ester standards purchased from Nu-Check Prep (Elysian, MN; cat. No. GLC-60) and Matreya, Inc. (Pleasant Gap, PA; FIM-FAME-7). The 18:1 FA that were not available commercially (trans-6/8, trans-9, trans-12, trans-13, trans-15, cis-12, and cis-15) were identified by order of elution as shown by Molkentin and Precht (1995).
Statistical Analyses Data were first analyzed as a complete randomized block design by PROC MIXED of SAS (SAS, 2004) according to the following model: Yijk = µ + Ai + cj + Tk + ATik + εijk, where:
[1]
Yijk = dependent variable for treatment i on block j and time k; µ = overall mean; Ai = fixed effect of treatment i; i=1, 2, 3, 4, 5, 6, 7, 8; cj = random effect of block j; j=1, 2; Tk = fixed effect of time k; k= 1, 2, 3, 4, 5, 6, 7, 8, 9; ATik = interaction of treatment i with time k; and εijk = residual error associated with the ijkth observation.
In model [1], there is no functional form imposed on the effect of time and time by treatment interaction.
76
Model 1: single pool, first-order kinetic model To characterize this structure, the first-order kinetics model of ∅rskov and McDonald (1979) was fitted with the NLIN procedure of SAS (2004) using the least squares subclass means from [1] as observations. The model for disappearance of 18:3 and 18:2 (Equation 2) was as follows: Y = C + Pe -k (T - L) + εi
[2]
Y = amount (mg/tube) of FA at time t; C = pool of unavailable FA; P = pool of potentially available FA; k = fractional rate of FA disappearance (h-1); T = incubation time (h); L = lag (h); εi = residual error associated with the ith observation.
The parameter estimates of each k and L were compared by orthogonal contrasts to test the effects of pH, sucrose (linear and quadratic), forage sources and interactions between pH and sucrose, and pH and forage source. Significance was declared at P < 0.05.
77
Model 2: multiple pools, first-order kinetic model The time effect can be modeled alternatively by using the more complex kinetics diagram with multiple pools presented in Figure 5.4. To do so, the least squares means of model [1] were used as observations to estimate fractional rates of transfer among FA pools using SAAM II Software (SAAM, 1997). The rates of disappearance of the potentially available fractions of 18:3 and 18:2 were calculated, to decrease the variation of the estimated parameters resulting from SAAM II. Observations (least square means of [1]) were weighted by the reciprocals of their standard errors. The concentrations of the intermediates of BH from 0 to 12 h were used also in SAAM II to estimate the fractional passage rates of the pools. Arrows represent links among pools. The decision to specify direct transfers among specific pools was made using two sequential steps: 1) if there were published data showing a direct link for the pools, and 2) if the addition of a pool or transfer produced a smaller value of the Akaike (AIC) information criterion (Cobelli and Foster, 1998): AIC = WRSS + 2P, where: WRSS = weighted residual sum of squares; P = number of unknown model parameters.
For instance, there is evidence that cis-9, trans-11 18:2 (one of the CLA isomers) is the first product of the BH of 18:2 (Kepler et al., 1966; Noble et al., 1974); therefore, the pool of linoleic acid has an arrow to the CLA pool. The Rosenbrock integrator method was used and the optimization was performed using a variance model based on relative data and a forward derivative. 78
RESULTS AND DISCUSSION We reported the effect of sucrose and media buffer on the kinetics of BH of FA from fresh alfalfa and alfalfa hay. Fresh alfalfa was used as a positive control to establish standard values for VA and other intermediates of BH accumulation in the effluent and also to compare differences in BH that could be attributed to forage conservation. The use of fresh alfalfa with a lower percentage of NDF compared to hay was done intentionally to represent the selection of the immature parts of the plant by grazing cows.
Effects of pH on Biohydrogenation The temporal patterns of pH in incubation media with the two buffers are in Figure 5.1. The mean pH for the SB was 6.7 and did not fall below 6.4, whereas the mean pH for the WB was 6.3, and it dropped to below 6.2 after 5 h of incubation. Lower pH is associated with an impairment of lipolysis and BH, and lower fiber digestibility (Qiu et al., 2004). A pH of 6.2 is considered to be the lower threshold for optimum fiber digestion (Grant and Weidner, 1992), although the period of time that pH is less than 6.2 may be more significant in inhibiting growth of cellulolytic bacteria (Calsamiglia et al., 2002). Because we did not separate free and esterified FA in the incubated tubes, the BH values reported represent the disappearance of 18:2 and 18:3 by hydrogenation after their release from glycerides by lipolysis. Therefore, we used the term DR to represent the BH of 18:2 and 18:3 reported in this experiment. The DR of 18:2 and 18:3 are shown in Table 5.2. Lag times for both 18:2 and 18:3 were not different from zero. Zero lag times 79
have been reported previously with incubation of forages (Boufaied et al., 2003b) and oil seeds (Enjalbert et al., 2003). Conversely, Troegeler-Meynadier et al. (2003) reported lag times between 1 and 2 h for soybean oil. Lag time is related to the time needed to release the glyceride from its matrix and is dependent on fat source (Beam et al., 2000) and may also be associated with DM digestibility. The UFA of fresh plants are located mainly in the chloroplasts as monogalactosyldiglycerides and digalactosyldiglycerides (Butler and Bailey, 1973; Hawke, 1973), and hydrolysis of plant galactolipids has been reported to be 78 to 95% complete after 4 h of incubation (Dawson and Hemington, 1974). Rapid hydrolysis of galactoglycerides and triglycerides (Beam et al., 2000) may not cause a lag time to be observed. Unesterified FA at 0 h in the incubation tubes would have contributed to any differences in lag time observed among experiments. The DR of 18:2 and 18:3 was slower (P < 0.05) when substrates were incubated in the WB, independent of the forage source. Others have shown that lower pH decreases lipolysis (Van Nevel and Demeyer, 1996) and the extent of BH and increases the proportion of monoenoic FA in the medium (Van Nevel and Demeyer, 1996; Griinari et al., 1998; Troegeler-Meynadier et al., 2003). Qiu et al. (2004) reported a 50 % decrease in total BH when the pH changed from 6.5 to 5.8 in continuous culture. The DR was almost always higher for 18:3 than for 18:2 and was 74% higher when fresh alfalfa compared with hay was incubated with the WB. The area under the curve (AUC) was estimated for each FA (Tables 5.3 and 5.4) to compare effects of sucrose, pH, and forage source on the mean amount of individual FA during in vitro BH. Although some information is lost when using AUC (i.e., curve shape and zenith), it allows comparisons of the mean concentration of specific FA during 80
the 24-h incubation period. The AUC values for both 18:2 and 18:3 FA were higher (P < 0.05) with WB incubations, a result of the lower DR for both FA. The buffers had no effect on AUC for 18:0.
Forage Source and Biohydrogenation The DR of 18:3 was higher from fresh alfalfa than alfalfa hay (P < 0.05; Table 5.2). Boufaied et al. (2003b) reported higher rates of 18:3 BH for fresh grass compared with hay. Those authors argued that, because dry hay may have a lower DM digestibility, the FA could have been trapped physically within the organic matrix, decreasing rate of release and, therefore, decreasing BH of 18:2 and 18:3. There was a trend (P = 0.11) for an interaction between buffer and forage source for the DR of 18:3. The WB decreased both 18:2 and 18:3 DR; however, the DR of 18:3 from fresh alfalfa decreased 12% compared to a 62% decrease from alfalfa hay. Therefore, the decreased DR of 18:3 with lower pH is less pronounced for fresh alfalfa than for alfalfa hay. The decrease in BH with lower pH may be associated with a decrease in the numbers of cellulolytic bacteria (Wales et al., 2004), because cellulolytic bacteria play a major role in ruminal BH and these are known to be decreased at lower pH (Harfoot and Hazlewood, 1997). Moreover, low pH inhibits lipolysis (Van Nevel and Demeyer, 1996), decreasing BH by diminishing the availability of free UFA.
81
Qiu et al. (2004) observed reduced cellulolytic bacterial numbers with reduced BH of 18:2 in continuous culture. Thus, we postulated that fresh alfalfa maintains the number of cellulolytic (biohydrogenating) species even at decreased pH. This is consistent with higher NDF digestibility of fresh alfalfa compared to alfalfa hay at lower ruminal pH. Kolver and Veth (2002) reported that the pH of grazing cows may be as low as 5.8. Because low pH may increase the proportion of intermediates of BH (Qiu et al., 2004), the diurnal variation in pH of grazing ruminants and the high 18:3 concentration of pasture could increase the flow of VA out of the rumen. The interaction between buffer and forage source on the AUC for 18:3 was significant (P < 0.05; Table 5.3), which is consistent with the trend (P = 0.11) of the same interaction on the DR of 18:3 (Table 5.2). Also, the AUC of both 18:2 and 18:3 FA during the incubation were higher for fresh alfalfa than for alfalfa hay, even though the BH rates for fresh alfalfa were higher. We speculated that tubes with fresh alfalfa would have higher amounts of 18:2 and 18:3 FA compared to tubes with hay (Table 5.3). Accordingly, Boufaied et al. (2003b) estimated that 18:3 flow from the rumen would be higher with fresh grass than with hay. Because the ensiling process decreases the absolute and relative amounts of UFA of fresh forages (Chilliard et al., 2000; Boufaied et al., 2003a), the percentage of FA in fresh alfalfa was twice as high for alfalfa hay (Table 5.1). Therefore, one of the most expected results when cows are not grazing is reduced concentration and outflow of 18:3 FA from the rumen.
82
Sucrose and Biohydrogenation There was no effect (P > 0.05) of sucrose addition on the DR of 18:2 and 18:3 (Table 5.2). This observation did not support our hypothesis that sucrose would inhibit BH. However, adding 8% sucrose to hay in the SB treatment decreased the disappearance rates of 18:2 and 18:3 by 38 and 22%, respectively. We postulated that sucrose incubated with WB would result in greater impairment of BH than when incubated with SB, but we observed no interaction (P > 0.05) between sucrose and buffer source on the DR of 18:2 and 18:3. Wales et al. (2004) observed equal nonstructural carbohydrate digestion in continuous culture with pH at 5.6 and 6.1. The authors concluded that the bacterial population responsible for sugar digestion is insensitive to low pH. Therefore, low pH or supplemental sugars could increase the competitiveness of the sugar-digesting bacterial population. This shift in microbial population may decrease the numbers of biohydrogenating bacteria, most likely a result of substrate competition to cause a reduction in total BH. More research is needed to characterize specific bacterial groups with their role in the ruminal BH.
Intermediates of Biohydrogenation Because the proportion of 18:2 in forages is low, the amount of CLA in the incubation tubes was too low to be measured accurately; therefore, CLA concentrations were not reported. The BH curves of VA for all treatments are shown in Figure 5.2. The concentration of VA was highest with the fresh alfalfa treatment, an expected result because there was 79% more 18:3 in fresh alfalfa than in alfalfa hay and the majority of the VA originates from BH of 18:3 when forage is the only source of FA. 83
Although the DR of 18:3 decreased when incubated with WB, the shapes of the VA curves were similar for fresh alfalfa for both buffers. If pH was lower in WB incubations for a prolonged time, the growth of cellulolytic species could have been impaired (Calsamiglia et al., 2002); thus, we postulated that a greater difference in the mean pH between SB and WB would result in more pronounced differences of BH between buffer treatments than was observed. The in vitro BH curve of VA reflects relative rates of appearance (synthesis) and disappearance (hydrogenation) during the early incubation. During in vitro BH of FA from fresh alfalfa, the VA concentration peaked with the same magnitude (no difference of AUC) at 7 h for the SB and 9 h for the WB, followed by a decrease in its concentration until 24 h. Similar BH patterns of VA during in vitro incubation have been observed by Noble et al. (1974), who reported peaks of VA of 0.32 and 0.65 mg/mL of 18:2 at 1 and 2 h during the in vitro BH, respectively. Qiu et al. (2004) reported no effect of pH on the proportion of VA in the effluent of continuous culture fermenters. The authors speculated that the lack of effect of pH on VA flow was a result of a continuous input of linoleic acid. In our hay treatments, the VA curves did not increase sharply, as observed for the fresh alfalfa treatment, perhaps because the initial amount of UFA available for BH was much lower with hay. Troegeler-Meynadier et al. (2003) observed a higher rate of appearance of the intermediates of BH when a higher amount of linoleic acid was used.
84
The BH curves for trans-9, trans-10, trans-12, trans-13, trans-15, and cis-15 18:1 from the in vitro incubation of fresh alfalfa and alfalfa hay with no supplemental sucrose are shown on Figure 5.3. These showed similar behavior within forage source. During the in vitro BH of FA from the fresh alfalfa treatments, the trans-13 18:1 FA sharply increased until 9 h. The concentrations of trans-15 and cis-15 18:1 FA increased until 4 h and then slowly decreased after 9 h. Only forage source and the interaction between forage source and buffer affected (P < 0.05; Table 5.4) the AUC of VA. The AUC for VA was 39 and 46% higher for fresh alfalfa than for alfalfa hay with the SB and WB, respectively. However, the AUC of VA from fresh alfalfa was only 2% higher for the SB compared to the WB treatment. When fresh alfalfa was incubated with the WB, the DR of 18:2 and 18:3 were affected more (decreased 44% and 12%, respectively) than the BH of VA. This observation was also similar for hay without sucrose. The AUC of trans-12, trans-13, trans-15, and cis-15 18:1 FA were higher (P < 0.05; Table 5.4) for fresh alfalfa than for alfalfa hay. The AUC of trans-10 18:1 was higher (P < 0.05) for the WB than for SB, which is consistent with the higher concentration of trans-10 18:1 in milk fat reported with lower ruminal pH (Griinari et al., 1998). The AUC for cis-15 and trans-15 were similar within forage source, suggesting that these are synthesized in similar proportions independently of sucrose addition and pH. The AUC of trans-13/14 18:1 FA was second highest of all 18:1 intermediates. Loor et al. (2003) reported that VA was the primary intermediate of BH, followed by trans-13 18:1 FA during continuous culture fermentation of grasses.
85
The GLC method used in this trial did not allow the separation of trans-13 and trans-14 FA. The synthesis of both FA from 18:3 BH has been reported previously (Ward et al., 1964; White et al., 1970); however, the exact pathway and microorganisms involved are not known.
Kinetics of Biohydrogenation Using SAAM II Most of the research with in vitro BH used nonlinear regression to estimate the disappearance rates of 18:2 and 18:3, whereas none have reported BH rates of VA and other intermediates of BH (Beam et al., 2000; Enjalbert et al., 2003; Troegeler-Meynadier et al., 2003). The factors that control the concentration of VA during ruminal BH are significant because VA plays a central role in BH (Harfoot and Hazlewood, 1997) and it is the major source of CLA in milk of confined (Griinari et al., 2000) and grazing cows (Kay et al., 2004). Therefore, we developed a multi-compartmental mathematical model to estimate the fractional rates of VA appearance and disappearance to characterize the influence of dietary factors on pool size and flux of VA during in vitro BH. The representation of the FA pools used to estimate the fractional rates by SAAM II during ruminal BH is shown in Figure 5.4. Because the pool size was low for many of the intermediates during the BH of FA from hay, the estimation of fractional rates of transfer among pools became unreliable or did not solve; thus, only the fresh alfalfa data were used. The CLA isomers and the intermediates of 18:3 conversion to VA were not measured; however, we could estimate the temporal changes of the CLA pool and used one FA pool for conversion of 18:3 to VA (18:3-Int pool; Figure 5.4) to represent cis-9, trans-11, cis-15 octadecatrienoic acid plus trans-11, cis-15 octadecadienoic acid. 86
The BH rates estimated by SAAM II are in Figure 5.4. The BH rates for 18:2 and 18:3 disappearances were lower (P < 0.05) for the WB than for SB, consistent with the rates estimated by SAS. Comparison of the estimated potential DR of 18:2 and 18:3 by SAS and SAAM II was performed by paired t-test and is illustrated in Figure 5.6; there was no difference (P > 0.05) between methods (data not shown). We observed that the 18:3 pool was depleted faster when alfalfa was incubated with SB compared to WB, contributing to the earlier peak for VA with this treatment. The temporal changes in the CLA and 18:3-Int pool sizes are illustrated in Figure 5.7. The estimated BH curve of CLA followed the same pattern as reported previously (Noble et al., 1974; Kellens et al., 1986), and similar to the VA curve, but peaking earlier, which is consistent with precursor/product kinetics. The 18:3-Int pool increased more sharply when alfalfa was incubated with the SB. Also, the size of the 18:3-Int pool decreased faster than the CLA pool, regardless of the buffer used, because the BH of 18:3 was faster than for 18:2. The WB treatment slowed BH of the 18:3-Int pool, and after 4 h, the 18:3-Int pool size was greater for the WB treatment, consistent with the lower BH rate of 18:3 and might be consistent also with the later peak of VA observed for the WB. We were able to estimate the DR (hydrogenation) of VA using SAAM II (Figure 5.4). We are not aware of any other research reporting these rates without using isotopes. There was no difference (14.1 and 12.5 %/h; P > 0.05) in the DR of VA between the WB and SB, although the DR of 18:2 and 18:3 were decreased by the WB. To measure the accuracy of SAAM II in estimating pool size of VA, we compared the VA curves estimated by SAS and SAAM II software (Figure 5.5). The coefficient of variation for the VA curves was 3.8 and 5.2% for the SB and WB, respectively. Appearance (synthesis) of 87
VA was greater than disappearance until ca. 6 and 9 h for SB and WB, respectively. These represent the times that VA concentration peaked for each buffer. Because a multiple pools, first-order kinetic model (SAAM II) estimates fluxes, as well as mass of the VA pools, it generates more information from the data than a single pool, first-order kinetic model (SAS). The flux (mg/h) to and from the VA pools for the fresh alfalfa treatments is in Figure 5.8. The convergence of the input and output curves of the VA pool represents the time of peak VA concentration. The flux from the VA pool was ca. 27% higher for the SB between 3 to 6 h of incubation. Because the DR of 18:3 was greater for the SB compared to the WB, the flux to the VA pool peaked earlier for the SB. The last step of BH is assumed to be most sensitive to pH because trans-18:1 FA accumulate when high grain diets are fed (Griinari et al., 1998). However, the WB decreased DR of 18:2 and 18:3 without affecting the mean concentration and DR of VA. A shift to alternate pathways of BH that do not include VA could explain the lack of response of VA concentration to lower pH. For instance, WB decreased (P < 0.05; Figure 5.4) the BH of other 18:1 FA pool (trans-9, trans-10, trans-12, trans-13) to 18:0 and the hydrogenation of the 18:3-Int pool to the other 18:1 FA pool (43.7 and 13.3%/h for SB and WB, respectively). Consequently, the lower rate of disappearance of 18:2 and 18:3 caused a more pronounced decrease in the flux of the UFA to other trans FA while maintaining almost similar fluxes to VA.
88
Limitations of the Modeling Approach Although using a multi-compartmental model to estimate BH is a superior approach to model the dynamics of FA metabolism in the rumen, there are constraints and assumptions that limit interpretation. One constraint is that all rates are correlated (data not shown), and any error in estimating one rate will affect others. Because 18:0 is the final product of BH, any error in estimating the precedent pools will be reflected in the 18:0 pool. Further, any intermediate of BH and microbial FA synthesis not taken into account that contribute to the synthesis of 18:0 will increase the error of estimating 18:0 pool size. Although some intermediates of BH of oleic acid have been reported (Ward et al., 1964; Mosley et al., 2002), the pathways and microbial species for BH of oleic acid have not been defined. Therefore, defining direct links between pools may be compromised. For example, we assumed that cis- and trans-15 18:1 was hydrogenated to 18:0; however, Harfoot and Hazlewood (1997) considered these to be true end products of 18:3 BH, based on research by Body (1976) and White et al. (1970). However, it is possible that cis- and trans-15 18:1 are hydrogenated directly to 18:0. Including hydrogenation of cis- and trans-15 18:1 to 18:0 in the model improves the fit of the model (AIC value), because the concentration of both decreased after 9 h of incubation. Alternatively, these may have been isomerized to other 18:1 FA (Mosley et al., 2002; Proell et al., 2002) and then hydrogenated to 18:0; however, this option did not solve. Our data demonstrate the need to define the intermediates of BH, as well the fractional rates of transfer among pools as dietary management is changed and microbial
89
populations shift. Such knowledge would greatly improve understanding of the complex reactions that occur in the rumen and that affect the profile of FA in ruminant products.
CONCLUSIONS Culturing rumen contents at lower pH decreased DR of 18:2 and 18:3, independent of the forage source. There was a trend for an interaction between incubation pH and forage source for DR of 18:3, with a greater decrease in the DR of 18:3 for alfalfa hay than for fresh alfalfa. Sucrose did not affect DR of 18:2 and 18:3. Adding 8% sucrose to hay at higher pH decreased the DR of 18:2 and 18:3 by 38 and 22%, respectively. There was no interaction between sucrose and incubation pH on the DR of 18:2 and 18:3. Compartmental analysis of BH of FA in fresh alfalfa showed that the mean concentration and hydrogenation rate of VA were not affected by pH, whereas the last step of BH at lower pH was slower for other trans-18:1 FA, and the mean concentration of trans-10 18:1 was higher. The concentration of trans- and cis-15 18:1 FA declined after 9 h of incubation, most likely due to hydrogenation to 18:0, which is an observation not previously reported. Using AUC to compare treatment effects on the concentration of BH intermediates during in vitro BH may be used to investigate dietary factors that influence BH. We quantified changes in the rates of BH of trans-18:1 and VA, which is the key regulatory step determining the amount of VA leaving the rumen and affecting milk CLA. The multi-compartmental model analysis allowed factors affecting the DR of 18:2 90
and 18:3 to be defined, as well as the synthesis and hydrogenation rates of other BH intermediates. We showed the magnitude of BH pathways other than through VA, that may explain variations of VA concentration as incubation conditions change. Because we could estimate fluxes, as well as mass of the VA pools, more information is generated from the data when using multiple pools compared to single pool, first-order kinetic model.
91
Table 5.1: Nutrient composition of fresh alfalfa and alfalfa hay.
Item
Fresh Alfalfa
Alfalfa Hay
DM
20.1
90.9
--------------------------- % of DM ----------------------------OM
91.1
93.5
N
3.2
2.9
NDF
31.3
39.9
Total fatty acids (FA)
2.1
0.8
Individual FA
--------------------------- % of FA -----------------------------
16:0
21.4
29.9
18:0
3.9
5.6
cis-9 18:1
3.2
4.0
cis-11 18:1
0.48
0.84
trans 18:1
-
0.04
18:2
17.1
19.0
18:3
46.8
25.7
92
7.20 7.00
SB WB
6.80
pH
6.60 6.40 6.20 6.00 5.80 0
5
10
15
20
Hours
Figure 5.1: Changes in pH during in vitro biohydrogenation of fatty acids from alfalfa hay or fresh alfalfa incubated with strong (SB) or weak buffers (WB).
93
A
ALF
0%
4%
8%
SEM=0.073 1.60
mg/flask
1.20
0.80
0.40
0.00 0
3
6
9
12
15
18
21
24
Time, h
B
SEM=0.113 1.60
ALF
0%
4%
8%
mg/flask
1.20 0.80 0.40 0.00 0
3
6
9
12
15
18
21
24
Time, h
Figure 5.2: In vitro biohydrogenation curves of vaccenic acid with strong (A) or weak (B) buffers. Treatments were fresh alfalfa (ALF) and alfalfa hay with addition of 0, 4, and 8% sucrose. 94
A
SEM = 0.091 0.70
Trans-9
Trans-10
Trans-12
Trans-13
Trans-15
Cis-15
0.60
mg/flask
0.50 0.40 0.30 0.20 0.10 0.00 0
3
6
9
12
15
18
21
24
Time, h
B
SEM = 0.016 0.70
Trans 9
Trans 10
Trans 12
Trans 13
Trans 15
Cis 15
0.60
mg/flask
0.50 0.40 0.30 0.20 0.10 0.00 0
3
6
9
12
15
18
21
24
Time, h
Figure 5.3: In vitro biohydrogenation curves of trans-9, trans-10, trans-12, trans-13, and cis-15 octadecenoic fatty acids from fresh alfalfa (A) and alfalfa hay (B) incubated with a strong buffer. 95
A
18:2
43.8 (±0.2)*
18:3
18:3_int
27.4 (±0.7)*
CLA 36.7 (±19.8)
4.6 (±0.2)Ŧ
10.9 (±7.4)
43.7 (±7.3)*
2.9 (±0.2)
t15
c15 t9
t10 t12 t13 0.8 (±0.5)
VA 20.2 (±3.0)* 9.0 (±0.9)*
3.9 (±0.5)
cis-18:1
12.5 (±2.1)
Stearic
Figure 5.4: Model of in vitro ruminal biohydrogenation of fatty acids from fresh alfalfa from 0 to 12 h incubated with strong buffer (A) and weak buffer (B). Boxes represent fatty acid pools and arrows represent rates (%/h) of transfer between pools during biohydrogenation with standard error in parenthesis. Linolenic acid (18:3), linoleic acid (18:2), CLA isomers (CLA), cis-9,trans-11, cis-15 octadecatrienoic acid plus trans11,cis-15 octadecadienoic acid (18:3-Int), vaccenic acid (VA), cis-9 octadecenoic acid (cis-18:1), cis-15 octadecenoic acid (cis-15), trans-15 octadecenoic acid (trans-15), trans-9 plus trans-10 plus trans-12 plus trans-13 octadecenoic acids (t9, t10, t12, and t13), and stearic acid (Stearic). * P < 0.05 Ŧ
P < 0.10 (Continue) 96
Figure 5.4 Continue
B
18:2
23.5 (±0.9)*
30.3 (±0.6)*
18:3
18:3_int
CLA 25.9 (±27.2)
3.6 (±0.5)Ŧ
12.2 (±11.5)
13.3 (±3.5)*
2.6 (±0.6)
t15
c15 t9
t10 t12 t13 3.0 (±3.6)
VA 6.2 (±4.1)* 5.6 (±0.6)*
6.2 (±3.9)
cis-18:1
14.1 (±3.7)
97
Stearic
A
LSMEANS SAAM
1.80 1.60 1.40 mg/flask
1.20 1.00 0.80 0.60 0.40 0.20 0.00
0
2
4
6
8
10
12
Time, h
B
LSMEANS 1.80
SAAM
1.60 1.40 mg/flask
1.20 1.00 0.80 0.60 0.40 0.20 0.00
0
2
4
6
8
10
12
Time, h
Figure 5.5: Changes in the in vitro vaccenic acid pool: least square means from randomized complete block design analysis (observed) and SAAM II (estimated) from incubations of fresh alfalfa with strong (A) or weak (B) buffers. The root mean square error for curves A and B is 0.036 and 0.045, respectively.
98
A SAAM II SAS
4 3.5 3 mg/falsk
2.5 2 1.5 1 0.5 0 0
2
4
6
8
10
12
Time, h
B
SAAM II
3.5
SAS
3
mg/flask
2.5 2 1.5 1 0.5 0 0
2
4
6
8
10
12
Time, h
Figure 5.6: In vitro disappearance of the 18:3 pool estimated by a single pool, first-order kinetics model (SAS) and a multiple pools, first-order kinetics model (SAAM II) from incubations of fresh alfalfa with strong (A) or weak (B) buffers.
99
A 1.4
SB
1.2
WB
mg/flask
1 0.8 0.6 0.4 0.2 0 0
3
6
9
12
Time, h
B SB
0.6
WB 0.5
mg/flask
0.4 0.3 0.2 0.1 0 0
3
6
9
12
Time, h
Figure 5.7: Representation of the temporal change of the pool size (mg) of cis-9, trans11, cis-15 octadecatrienoic acid plus trans-11, cis-15 octadecadienoic acid (A; 18:3-Int pool) and CLA (B) estimated from a multiple pools, first-order kinetic model (SAAM II) during in vitro biohydrogenation of fresh alfalfa with strong (SB) or weak (WB) buffers.
100
A Disapearance
0.35
Appearance
0.3
mg/h
0.25 0.2 0.15 0.1 0.05 0 0
3
6
9
12
Time, h
B Disappearance
0.3
Appearance
0.25 mg/h
0.2 0.15 0.1 0.05 0 0
3
6
9
12
Time, h
Figure 5.8. Estimation of the instantaneous flux (mg/h) to and from the vaccenic acid pool from a multiple pools, first-order kinetic model (SAAM II) during in vitro biohydrogenation of fresh alfalfa with strong (A) and weak (B) buffers.
101
102 0.378
0.201
0.156
WB
0.338
0.250
SB
0.063
0.058
SEM2
0.05.
buffer (F x B), and linear and quadratic effect of the interaction between sucrose and buffer (S x B).
versus all alfalfa hay treatments), linear (L), and quadratic (Q) effects of sucrose (S), interactions between forage and
Orthogonal contrasts to test the effects of buffer (B; all WB versus all SB treatments), forage (F; all fresh alfalfa
8.8 2.6 3.0
* * * * * *
SxB L Q
3
6.7 1.9 3.0
NS * NS NS NS NS
F
Mean SEM across treatments.
8.5 2.6 2.4
0.07 3.07 1.59 0.23 0.70 0.17
B
2
7.1 1.9 3.0
1.4 5.8 14.8 3.3 5.7 2.7
SEM2
Strong (SB) and weak (WB) buffers.
8.1 2.4 2.7
1.5 6.9 14.8 3.7 7.0 3.1
7.4 2.3 5.9
1.5 6.0 15.5 3.4 6.2 2.9
8.7 3.0 5.8
1.5 6.4 14.3 3.5 6.3 2.9
1.9 6.5 26.0 4.6 12.1 5.8
1.8 7.1 25.4 4.9 13.7 6.3
1.6 6.2 15.8 3.5 6.2 2.9
1.5 6.0 12.6 3.1 5.7 2.7
Fresh Alfalfa WB SB
Contrasts3 S FxB L Q
1
Item trans 18:1 trans-9 trans-10 VA trans-12 trans-13 trans-15 cis 18:1 cis-9 cis-11 cis-15
Treatments1 Hay + sucrose (%) 0 4 8 WB SB WB SB WB SB
(SB) and weak (WB) buffers of fresh alfalfa and alfalfa hay with 0, 4, and 8% sucrose addition, estimated by SAS.
Table 5.4: Areas under the curve (mg.h) of the intermediates of biohydrogenation during in vitro incubation with strong
CHAPTER 6
OVERALL DISCUSSION
Four major experiments were conducted to study and improve the methods used to estimate BH rates of FA from fresh forages. In Experiment 1 and 2 (Chapter 3), I reported a 92% recovery of FA from fresh alfalfa samples in the first extraction using H:IP. Moreover, there were no changes in the percentage of TFA from fresh alfalfa stored at - 20 oC for 60 d after flash-freezing the samples with liquid N after harvesting. In Experiment 3 (Chapter 4), addition of sucrose decreased total BH by a pHindependent process without changing the percentage of VA in the effluent. However, in Experiment 4 (Chapter 5), sucrose did not affect BH of 18:2 and 18:3. Additionally, the multi-compartmental model used in this experiment allowed: 1) the estimation of BH rates of most of the FA; and 2) the comparison of the shift in magnitude of alternate pathways of BH, besides going through VA, that might have resulted in the same VA concentration with lower DR of 18:2 and 18:3 for the WB. The multi-compartmental approach to model ruminal BH could generate new data to improve the prediction of ruminal metabolism of FA.
105
Theories and laboratory analysis not commented on before were also explored in this section. To integrate and also isolate important points regarding the objective of my research, I divided my discussion in 3 major topics: 1) Fresh Forages: impact of oxidation on the concentration of FA from fresh forages and future research on the post-harvesting storage of fresh plant tissues; 2) Kinetics of Ruminal Biohydrogenation: discussion of the novel approaches and results of my research and possible methods to overcome the limitation of in vitro studies of BH; and 3) Theories: discussion of other possible theories behind the higher concentration of CLA in milk fat from grazing cows.
FRESH FORAGES
The study of digestibility and BH of FA from fresh forages requires proper care to avoid loss of nutrients; plant enzymes are triggered after harvesting, causing loss of soluble sugars, proteins, and FA. Because the initial profile and amount of FA greatly affect BH, I developed and improved a method of inhibiting post-harvesting loss of FA from fresh alfalfa in our laboratory conditions. This methodology was applied to all samples used in my research. From Experiment 1, I recommend using H:IP to extract plant lipids instead of other more widely used solvents, such as the procedure of Bligh and Dyer (1959), which uses toxic solvents and may cause low recoveries of acidic phospholipids and 106
phophatidylinositol. Although the use of H:IP may be a better approach, I do not recommend storing samples with isopropanol; an alternative procedure would be evaporating isopropanol and storing the samples with hexane until after the extraction step and the use of antioxidants would also help in the prevention of oxidation. Additionally, using boiling ethanol or water at 95 oC to stop enzymatic activity could be done if not using isopropanol as a solvent. Independently of the solvent used, samples should be kept in the dark, at low temperature, and analyzed as soon as possible to avoid oxidation of FA. Alcohol, such as isopropanol, can react with double bonds of FA during storage. I could not assess the impact of storing fresh alfalfa samples with H:IP on the profile of UFA (Experiment 2). Therefore, future research should test the effect of storage time of fresh forage samples using H:IP or hexane (positive control) on the percentage of 18:3 and the ratio 16: 0 to 18:3. Animals fed conserved forages have a lower proportion of CLA in milk compared to grazing animals (Dewhurst et al., 2003). Research studying BH of FA from fresh forages should avoid post-harvesting losses of FA by flash-freezing samples with liquid N and storing the dry samples at low temperatures (as in Experiment 2). Because propagation of free radicals and formation of more stable peroxides occur (Perkins, 1996), even at low temperature, storing the dry samples in at least - 20 oC is the best approach.
107
The accuracy of estimating the profile of FA from fresh forages will depend on how much of the free radical formation is inhibited. In Experiment 2, fresh forages were stored for 60 d at - 20 oC after frozen with liquid N; however, I do not know if 90, 120, or more days of storage would decrease the percentage of 18:3; that is a factor of how soon plant enzymes are inhibited after harvesting. The literature is consistent about the higher BH rate of 18:3 compared to 18:2 and the inhibitory effect of the hydrogenation of trans-18:1 to 18:0 as the amount of unesterified UFA in the rumen increases. Hence, the higher percentage of 18:3 in fresh plants was one of the first theories explaining why grazing cows have a higher percentage of milk CLA compared to TMR-fed cows. However, Dhiman (2000) showed similar concentrations of CLA in milk fat of dairy cows fed diets high in 18:2 or 18:3.
KINETICS OF RUMINAL BIOHYDROGENATION
Because acquiring representative samples of FA from the rumen to study ruminal BH is difficult, in vitro experiments are useful to study kinetics of BH (Moate et al., 2004). However, there are limitations when extrapolating the results to in vivo situations. In Experiment 3 (Chapter 4), sucrose inhibited the BH of UFA but not in Experiment 4 (Chapter 5), contrasting the differences in the effect of bolus and continuous feeding of soluble sugars on in vitro fermentation studies. Sucrose is rapidly fermented (Weisbjerg et al., 1998) and would be quickly degraded with a bolus dose, whereas a continuous infusion would maintain a certain amount of sucrose during the 108
entire fermentation. Because sucrose can be more efficiently utilized for growth for some bacterial species (Russell and Baldwin, 1979) that do not play a role in BH, these species would predominate in the presence of sucrose. Such a shift in bacterial population could lead to a substrate competition (most likely N) and discrimination against biohydrogenating bacteria. Therefore, a continuous dose of sucrose would be more prone to inhibit BH than a bolus dose during in vitro fermentation. Furthermore, the effect of sucrose on BH was assessed in Experiment 3 using continuous culture with 7 d of adaptation. This adaptation period may have allowed bacteria species that are more efficient in utilizing sucrose to grow in numbers (Slyter and Putnam, 1967), diluting the effect of inoculum on the results. A shift in microbial ecology against biohydrogenating bacteria was the major cause of the sucrose-induced inhibition on BH; the change in microbial population is also presented in Appendix A. Conversely, the results in Experiment 4 (in vitro) were greatly influenced by inoculum (diet, cow, and time after feeding), contributing to the difference in results observed between both experiments for the effect of sucrose on BH. Experiment 4 also revealed that low pH inhibited DR of UFA; I hypothesize that the low ruminal pH inhibits BH, causing an increase in trans-18:1 FA in the rumen by direct and indirect mechanisms. Low pH decreases the BH of UFA, causing a higher percentage of UFA to accumulate and, therefore, an inhibition of the hydrogenation of trans-18:1 FA to 18:0. Additionally, the low pH could directly decrease the numbers of the microorganisms responsible for the last BH step. Either way, the result is a higher decrease in the hydrogenation of trans-18:1 FA than the BH of UFA.
109
The concentration of intermediates of BH depends on their rates of appearance (synthesis) and disappearance (isomerization or hydrogenation); thus, a method that estimates both rates is more useful to study dietary effects on the overall BH process. Additionally, alternative pathways may take place, making necessary the estimation of flux and rates of the intermediates. For instance, the WB in Experiment 4 decreased the isomerization and hydrogenation rates of the 18:3-Int and trans-18:1 pools, respectively. Using SAAM II to estimate BH rates and integrate FA pool sizes is a novel and improved approach to study kinetics of BH. I reported the rates of BH of VA, the most limiting and central step in the overall BH, was about 13%/h. I modeled the data reported by Noble et al. (1974) using SAAM II and found a similar value (11.7%/h) for the hydrogenation of VA compared to when 0.65 mg/ml of
14
C-labeled 18:2 was incubated with rumen fluid
(Appendix E). Moate et al. (2004) estimated the BH rate of trans-18:1 to be 22.8 %/h. The authors emphasized that no attempt was made to model the rate constants of most of the FA during ruminal BH and that there is a lack of data in the literature regarding the BH rates of trans-18:1 FA. The ratio between the hydrogenation rate of VA and the DR of UFA (primarily 18:2 and 18:3) may be the most important factor regulating the proportion of VA flowing out of the rumen and could be used to compare feedstuffs and dietary effects on the probability to increase CLA concentration in milk and/or cause milk fat depression. The multi-compartment method estimates both and also the hydrogenation rate of other intermediates of BH.
110
Factors that increase the BH of 18:3 and/or decrease the hydrogenation of VA are strong candidates to increase CLA in milk fat. A shift in the ratio between the BH rates of 18:3 and VA is a reflection of a change in the bacterial population (groups A and B) in the rumen fluid. Experiment 3 had lower BH rates as sucrose supplementation increased. Cluster analysis of the banding profiles grouped the sucrose treatments together for both methods suggesting that sucrose altered the bacterial populations (Figure 4.1 and Appendix A). A better approach would be to identify the species responsible for this change in the ratio of BH rates; therefore, future research should isolate bands and categorize the bacteria species that may be responsible for treatment-induced changes in BH. To improve our knowledge on the overall BH process, I propose a different approach to study kinetics of BH by taking samples at different time points from the continuous culture fermenters during the last 3 d of each period and use SAAM II to estimate the BH rates (Appendix B). This method would have the advantage to have continuous infusion of nutrients concomitantly with the estimation of BH rates and control of passage rate and pH. Additionally, bacterial DNA would be used, as in Experiment 3, to associate changes in hydrogenation rate of VA with a shift in bacteria population. More information is needed regarding the magnitude of hydrogenation of VA and trans-10 18:1 to help predict flow of specific FA to the duodenum and identify key points to improve the quality of ruminant products and understanding milk fat depression.
111
THEORIES
Understanding the factors that govern ruminal metabolism of FA of grazing cows could be used to predict and alter the flow of FA to the duodenum, improving milk fat production and milk quality for human consumption. I proposed in Chapter 5 that the low pH associated with a higher percentage of fiber in the rumen during grazing conditions would favor group A bacteria, represented by Butyrivibrio fibrisolvens, over group B. The exact importance of specific bacteria species on ruminal BH is not well defined; just recently, a strain of Butyrivibrio hungateii was reported to completely biohydrogenate 18:2 to 18:0 (van de Vossenberg and Joblin, 2003), making this species belonging to neither group A nor B. Besides pH and percentage of UFA in the diet, there may be other factors affecting ruminal BH. Ultimately, the proportion between groups A and B bacteria will determine the percentage of intermediates of BH flowing to the duodenum. I hypothesize that plant lipases and protozoa have a significant role in the higher percentage of CLA in milk and meat of ruminants under grazing conditions.
Lipases The importance of lipases on the hydrolysis of FA from forages in the rumen is not certain. These enzymes are responsible for hydrolyzing FA from phospholipids, MGDG, and DGDG after harvesting and can stay active in the rumen for up to 6 h (Faruque et al., 1974). The exact effect of lipases on ruminal BH and the amount of free UFA from fresh plants has yet to be assessed. Faruque et al. (1974) reported that plant 112
lipases have a great contribution on the amount of free FA in the rumen. These authors reported a lower percentage of unesterified FA when cows where fed hay compared to fresh forage. Furthermore, the effect of lipases may differ depending on forage species (Lee et al., 2003b). Higher free UFA causes a higher impairment of BH, increasing the amount of intermediates; thus, the higher VA/CLA in the rumen of grazing cows may result from the higher free 18:3 hydrolyzed by bacterial and plant lipases, causing a higher inhibition on the last step of BH. Therefore, I hypothesize that lipases increase the amount of unesterified UFA in the rumen, mainly 18:3; thus, contributing to the higher ratio of VA/CLA in milk fat of grazing cows compared to animals consuming the same amount of FA from oilseed or hay.
Protozoa Although protozoa reach higher numbers in the rumen when animals are grazing, their role on BH is not well defined. The majority of 18:3 in forages is located in the chloroplast walls, and protozoa may be responsible for a decreased BH of 18:3 (Singh and Hawke, 1979) by engulfing chloroplasts (Appendix C) and lipid droplets (Harfoot and Hazlewood, 1997), making them not accessible for BH by ruminal bacteria. Conversely, grazing cows have the highest percentage of VA and CLA in milk fat. Therefore, I hypothesize that protozoa play a significant role in the higher flow of VA to the duodenum of grazing ruminants by two mechanisms:
113
1) Protozoa could be responsible for impairing transfer (cross-feeding) of trans-18:1 FA from the bacterial group A to B during the BH process by engulfing bacteria, causing an increase of trans-18:1 FA in the rumen; and 2) Protozoa may hydrogenate 18:3 to VA, which could be incorporated in their cell wall, and therefore, not be available for further hydrogenation to 18:0.
Either way, the intermediates of BH would also flow to the duodenum associated with protozoa or be recycled in the rumen. Both hypotheses suggest that protozoa would behave as a compartment that delays even more of the hydrogenation of VA. I observed a higher percentage of trans-18:1 FA (Appendix D) in protozoa than bacteria purified from duodenum samples taken from dairy cows fed TMR and fresh forage using a technique developed in our laboratory (Sylvester et al., 2004), suggesting that they are either able to biohydrogenate 18:3 from engulfed chloroplast or incorporate the intermediates of BH (CLA and VA) from engulfed bacteria and FA associated with food particles or both (samples in Appendix D are the same as used in Appendix C). However, whether protozoa play a significant role on VA flow to the duodenum from grazing cows is uncertain.
Fresh Forage Fresh forage may have intrinsic factors affecting ruminal fermentation and favoring growth of fiber-digesting bacteria, and therefore, bacteria species from Group A. Such factors may include high percentage of NDF concomitantly with high soluble sugars and soluble protein (Van Soest, 1994), higher numbers of protozoa, increased 114
feeding frequency, faster passage rates, and low amounts of starch. I postulated that fresh alfalfa maintains the number of cellulolytic (biohydrogenating) species even at lower pH. Because low pH may increase the proportion of intermediates of BH (Qiu et al., 2004) and the activity of cellulolytic bacteria in a low pH medium seems to be less affected with fresh forages (Experiment 4), the diurnal variation in pH of grazing ruminants and the high 18:3 concentration of pasture could increase the flow of VA out of the rumen. The pH necessary to inhibit BH is not precise, but it is most likely around 6.2 (same minimal pH for optimum growth of cellulolytic bacteria; Grant and Weidner, 1992), and the inhibition of BH also depends on the time that the pH is below 6.2 (Calsamiglia et al., 2002). Because low ruminal pH can impair the last step of BH, diets containing high UFA that cause a lower ruminal pH are candidates to cause an increase in the flow of trans 18:1 FA from the rumen. Therefore, grazing cows may experience larger ruminal pH fluctuations with high concentrations of 18:3 in the diet (Kolver and de Veth, 2002), causing the higher percentage of VA and CLA in the milk fat. Furthermore, B. fibrisolvens is more tolerant to low pH than other cellulolytic bacteria (Russell and Dombrowski, 1980) and its number was decreased when steers were changed from a hay to a high grain diet (Klieve et al., 2003).
115
BIBLIOGRAPHY
AOAC. 1990. Official Methods of Analysis. Vol. I. 15th ed., Arlington, VA. Bach, A., I. K. Yoon, M. D. Stern, H. G. Jung, and H. Chester-Jones. 1999. Effects of type of carbohydrate supplementation to lush pasture on microbial fermentation in continuous culture. J. Dairy Sci. 82:153-160. Banni, S., E. Angioni, E. Murru, G. Carta, M. P. Melis, D. Bauman, Y. Dong, and C. Ip. 2001. Vaccenic acid feeding increases tissue levels of conjugated linoleic acid and suppresses development of premalignant lesions in rat mammary gland. Nutr Cancer. 41:91-97. Bauman, D. E., and J. M. Griinari. 2000. Regulation and nutritional manipulation of milk fat. Low-fat milk syndrome. Adv. Exp. Med. Biol. 480:209-216. Baumgard, L. H., B. A. Corl, D. A. Dwyer, A. Saebo, and D. E. Bauman. 2000. Identification of the conjugated linoleic acid isomer that inhibits milk fat synthesis. Am. J. Physiol. Regul. Integr. Comp. Physiol. 278:R179-184. Baumgard, L. H., J. K. Sangster, and D. E. Bauman. 2001. Milk fat synthesis in dairy cows is progressively reduced by increasing supplemental amounts of trans-10, cis12 conjugated linoleic acid (CLA). J. Nutr. 131:1764-1769. Beam, T. M., T. C. Jenkins, P. J. Moate, R. A. Kohn, and D. L. Palmquist. 2000. Effects of amount and source of fat on the rates of lipolysis and biohydrogenation of fatty acids in ruminal contents. J. Dairy Sci. 83:2564-2573. Blankson, H., J. A. Stakkestad, H. Fagertun, E. Thom, J. Wadstein, and O. Gudmundsen. 2000. Conjugated linoleic acid reduces body fat mass in overweight and obese humans. J. Nutr. 130:2943-2948.
116
Bligh, E. G., and W. J. Dyer. 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37:911-917. Body, D. R. 1976. The occurrence of cis-octadec-15-enoic acid as a major biohydrogenation product from methyl linolenate in bovine rumen liquor. Biochem. J. 157:741-744. Boufaied, H., P. Y. Chouinard, G. F. Tremblay, H. V. Petit, R. Michaud, and G. Belanger. 2003a. Fatty acids in forages. I. Factors affecting concentrations. Can. J. Anim. Sci. 83:501-511. Boufaied, H., P. Y. Chouinard, G. F. Tremblay, H. V. Petit, R. Michaud, and G. Belanger. 2003b. Fatty acids in forages. II. In vitro ruminal biohydrogenation of linolenic and linoleic acids from timothy. Can. J. Anim. Sci. 83:513-522. Butler, G. W., and R. W. Bailey. 1973. Chemistry and biochemistry of herbage. Academic Press, London, UK. Calsamiglia, S., A. Ferret, and M. Devant. 2002. Effects of pH and pH fluctuations on microbial fermentation and nutrient flow from a dual-flow continuous culture system. J. Dairy Sci. 85:574-579. Chalupa, W., and A. J. Kutches. 1968. Biohydrogenation of linoleic-1-14C acid by rumen protozoa. J. Dairy Sci. 27:1502-1508. Chalupa, W., G. D. Odell, A. J. Kutches, and R. Lauker. 1967. Changes in rumen chemical characteristics and protozoa populations of animals with depressed milk fat tests. J. Dairy Sci. 50 (Suppl. 1):154 (Abstr.). Charmley, E., D. M. Veira, G. Butler, L. Aroeira, and H. C. V. Cadagnone. 1991. The effect of frequency of feeding and supplementation with sucrose on ruminal fermentation of alfalfa silage given ad libitum or restricted to sheep. Can. J. Anim. Sci. 71:725-737. Chilliard, Y., A. Ferlay, and M. Doreau. 2001. Effect of different types of forages, animal fat or marine oils in cow’s diet on milk fat secretion and composition, especially conjugated linoleic acid (CLA) and polyunsaturated fatty acids. Liv. Prod. Sci. 70:31-48. 117
Chilliard, Y., A. Ferlay, R. M. Mansbridge, and M. Doreau. 2000. Ruminant milk fat plasticity: nutritional control of saturated, polyunsaturated, trans and conjugated fatty acids. Ann. Zootech. 49:181-205. Cobelli, C., and D. M. Foster. 1998. Compartmental models: Theory and practice using SAAM II software system. Page 423 in Mathematical Modeling in Experimental Nutrition. Clifford and Muller, eds. Plenum Press, New York, NY. Croft, K., F. Juttner, and A. J. Slusarenko. 1993. Volatile products of the lipoxygenase pathway evolved from Phaseolus vulgaris (L.) leaves inoculated with Pseudomonas syringae pv phaseolicola. Plant Physiol. 101:13-24. Czerkawski, J. W. 1967. Incubation inside the bovine rumen. Br. J. Nutr. 21:865-878. Czerkwaski, J. W., K. L. Blaxter, and F. W. Wainman. 1966. The metabolism of oleic, linoleic and linolenic acids by sheep with reference to their effects on methane production. Br. J. Nutr. 20:349-362. Dawson, R. M., and N. Hemington. 1974. Digestion of grass lipids and pigments in the sheep rumen. Br. J. Nutr. 32:327-340. Dewhurst, R. J., J. M. Moorby, N. D. Scollan, J. K. S. Tweed, and M. O. Humphreys. 2002. Effects of a stay-green trait on the concentrations and stability of fatty acids in perennial ryegrass. Grass Forage Sci. 57:360-366. Dewhurst, R. J., N. D. Scollan, S. J. Youell, J. K. S. Tweed, and M. O. Humphreys. 2001. Influence of species, cutting date and cutting interval on the fatty acid composition of grasses. Grass Forage Sci. 56:68-74. Dewhurst, R. J., N. D. Scollan, M. R. Lee, H. J. Ougham, and M. O. Humphreys. 2003. Forage breeding and management to increase the beneficial fatty acid content of ruminant products. Proc. Nutr. Soc. 62:329-336. Dhiman, T. R., G. R. Anand, L. D. Satter, and M. W. Pariza. 1999a. Conjugated linoleic acid content of milk from cows fed different diets. J. Dairy Sci. 82:2146-2156.
118
Dhiman, T. R., E. D. Helmink, D. J. McMahon, R. L. Fife, and M. W. Pariza. 1999b. Conjugated linoleic acid content of milk and cheese from cows fed extruded oilseeds. J. Dairy Sci. 82:412-419. Dhiman, T. R., L. D. Satter, M. W. Pariza, M. P. Galli, K. Albright, and M. X. Tolosa. 2000. Conjugated linoleic acid (CLA) content of milk from cows offered diets rich in linoleic and linolenic acid. J. Dairy Sci. 83:1016-1027. Donovan, D. C., D. J. Schingoethe, R. J. Baer, J. Ryali, A. R. Hippen, and S. T. Franklin. 2000. Influence of dietary fish oil on conjugated linoleic acid and other fatty acids in milk fat from lactating dairy cows. J Dairy Sci. 83:2620-2628. Duckett, S. K., D. G. Wagner, L. D. Yates, H. G. Dolezal, and S. G. May. 1993. Effects of time on feed on beef nutrient composition. J. Anim. Sci. 71:2079-2088. Enjalbert, F., P. Eynard, M. C. Nicot, A. Troegeler-Meynadier, C. Bayourthe, and R. Moncoulon. 2003. In vitro versus in situ ruminal biohydrogenation of unsaturated fatty acids from a raw or extruded mixture of ground canola seed/canola meal. J. Dairy Sci. 86:351-359. Erdman, R. 1996. Milk fat depression: Some new insights. Proceeding of Tri-State Dairy Nutrition Conference, Fort Wayne, IN. The Ohio State University, Columbus:pp 116. Faruque, A. J. M. O., B. D. W. Jarvis, and J. C. Hawke. 1974. Studies of rumen metabolism IX: contribution of plant lipases to the release of free fatty acids in the rumen. J. Sci. Food Agric. 25:1313-1328. Fellner, V., F. D. Sauer, and J. K. G. Kramer. 1995. Steady-state rates of linoleic acid biohydrogenation by ruminal bacteria in continuous culture. J. Dairy Sci. 78:18151823. Feussner, I., C. Wasternack, H. Kindl, and H. Kuhn. 1995. Lipoxygenase-catalyzed oxygenation of storage lipids is implicated in lipid mobilization during germination. Proc. Natl. Acad. Sci. 92:11849-11853. Firkins, J. L., and B. S. Oldick. 1997. Molasses, fat blends are highly available energy sources. Pages 14-17 in Feedstuffs. 119
Firkins, J. L., W. P. Weiss, M. L. Eastridge, and B. L. Hull. 1990. Effects of feeding fungal culture extract and animal-vegetable fat on degradation of hemicellulose and on ruminal bacterial growth in heifers. J. Dairy Sci. 73:1812-1822. Folch, J., M. Lees, and G. H. Sloane Stanley. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226:497-509. Franklin, S. T., K. R. Martin, R. J. Baer, D. J. Schingoethe, and A. R. Hippen. 1999. Dietary marine algae (Schizochytrium sp.) increases concentrations of conjugated linoleic, docosahexaenoic and transvaccenic acids in milk of dairy cows. J. Nutr. 129:2048-2054. French, P., C. Stanton, F. Lawless, E. G. O'Riordan, F. J. Monahan, P. J. Caffrey, and A. P. Moloney. 2000. Fatty acid composition, including conjugated linoleic acid, of intramuscular fat from steers offered grazed grass, grass silage, or concentrate-based diets. J. Anim. Sci. 78:2849-2855. Fried, B. 1993. Obtaining and handling biological materials and prefactionating extracts for lipid analyses. Pages 1-10 in Handbook of Chromatography. CRC Press, Boca Raton, FL. Friggens, N. C., J. D. Oldham, R. J. Dewhurst, and G. Horgan. 1998. Proportions of volatile fatty acids in relation to the chemical composition of feeds based on grass silage. J. Dairy Sci. 81:1331-1344. Fuller, M. A., H. Weichert, A. M. Fischer, I. Feussner, and H. D. Grimes. 2001. Activity of soybean lipoxygenase isoforms against esterified fatty acids indicates functional specificity. Arch. Biochem. Biophys. 388:146-154. Galliard, T., D. A. Wardale, and J. A. Mathew. 1974. The enzymic and non-enzymic degradation of colneleic acid, an unsaturated fatty acid ether intermediate in the lipoxygenase pathway of linoleic acid oxidation in potato (Solanum tuberosum) tubers. Biochem. J. 138:23-31. Gardner, H. W. 1991. Recent investigations into the lipoxygenase pathway of plants. Biochim. Biophys. Acta. 1084:221-239.
120
Gerson, T., A. John, and A. S. D. King. 1985. The effects of dietary starch and fiber on the in vitro rates of lipolysis and hydrogenation by sheep digesta. J. Agric. Sci. 105:27-30. Grant, R. J., and S. J. Weidner. 1992. Digestion kinetics of fiber: influence of in vitro buffer pH varied within observed physiological range. J. Dairy Sci. 75:1060-1068. Griinari, J. M., B. A. Corl, S. H. Lacy, P. Y. Chouinard, K. V. Nurmela, and D. E. Bauman. 2000. Conjugated linoleic acid is synthesized endogenously in lactating dairy cows by delta(9)-desaturase. J. Nutr. 130:2285-2291. Griinari, J. M., D. A. Dwyer, M. A. McGuire, D. E. Bauman, D. L. Palmquist, and K. V. V. Nurmela. 1998. Trans-octadecenoic acids and milk fat depression in lactating dairy cows. J. Dairy Sci. 81:1251-1261. Guil-Guerrero, J. L., A. Giménez-Giménez, A. Robles-Medina, M. del Mar RebollosoFuentes, E. Belarbi, L. Esteban-Cerdán, and E. Molina-Grima. 2001. Hexane reduces peroxidation of fatty acids during storage. Eur. J. Lipid Sci. Technol. 103:271-278. Gutierrez, J., P. P. Williams, R. E. Davis, and E. J. Warwick. 1962. Lipid metabolism of rumen ciliates and bacteria. I. Uptake of fatty acids by Isotricha prostoma and Entodinium simplex. Appl. Microbiol. 10:548-551. Ha, Y. L., N. K. Grimm, and M. W. Pariza. 1987. Anticarcinogens from fried ground beef: heat-altered derivatives of linoleic acid. Carcinogenesis. 8:1881-1887. Hall, M. B. 2000. Subject: Neutral detergent-soluble carbohydrates. Nutritional relevance and analysis. http://www.animal.ufl.edu/hall/MBLabMan.htm. Accessed Dec. 5, 2002. Hara, A., and N. S. Radin. 1978. Lipid extraction of tissues with a low-toxic solvent. Anal. Chem. 90:420-426. Harfoot, C. G., and G. P. Hazlewood. 1997. Lipid metabolism in the rumen. Pages 382426 in The Rumen Microbial Ecosystem. 2nd ed. P. N. Hobson, ed. Elsevier, London, UK.
121
Harfoot, C. G., R. C. Noble, and J. H. Moore. 1973. Factors influencing the extent of biohydrogenation of linoleic acid by rumen micro-organisms in vitro. J. Sci. Food Agric. 24:961-970. Hawke, J. C. 1973. Chemistry and biochemistry of herbage. in Lipids. G. W. Butler and R. W. Bailey, eds. Academic Press, London, UK. Hazlewood, G. P., P. Kemp, D. Lander, and R. M. Dawson. 1976. C18 unsaturated fatty acid hydrogenation patterns of some rumen bacteria and their ability to hydrolyse exogenous phospholipid. Br. J. Nutr. 35:293-297. Henderson, C. 1975. The isolation and characterization of strains of lipolytic bacteria from the ovine rumen. J. Appl. Bacteriol. 39:101-109. Hoover, W. H., B. A. Crooker, and C. J. Sniffen. 1976. Effects of differential solid-liquid removal rates on protozoa numbers in continuous cultures of rumen contents. J. Anim. Sci. 43:528-534. Hoover, W. H., and T. K. Miller-Webster. 1998. Role of sugars and starch in ruminal fermentation. Proceedings of Tri-State Dairy Nutrition Conference, Fort Wayne, IN, The Ohio State University, Columbus:pp 1-16. Huhtanen, P., and H. Khalili. 1992. The effect of sucrose supplements on particleassociated carboxymethylcellulase (EC 3.2.1.4) and xylanase (EC 3.2.1.8) activities in cattle given grass-silage-based diet. Br. J. Nutr. 67:245-255. Ip, C., Y. Dong, M. M. Ip, S. Banni, G. Carta, E. Angioni, E. Murru, S. Spada, M. P. Melis, and A. Saebo. 2002. Conjugated linoleic acid isomers and mammary cancer prevention. Nutr. Cancer. 43:52-58. Jahreis, G., J. Fritsche, and H. Steinhart. 1997. Conjugated linoleic acid in milk fat: high variation depending on production system. Nutr. Res. 17:1479-1484. Jeffcoat, R., P. A. Roberts, J. Ormesher, and A. T. James. 1979. Stearolyl-CoA desaturase: a control enzyme in hepatic lipogenesis. Eur. J. Biochem. 101:439-445. Jenkins, T. C. 1993. Lipid metabolism in the rumen. J. Dairy Sci. 76:3851-3863. 122
Jiang, J., L. Bjoerck, R. Fonden, and M. Emanuelson. 1996. Occurrence of conjugated cis-9, trans-11-octadecadienoic acid in bovine milk: effects of feed and dietary regimen. J. Dairy Sci. 79:438-445. Kay, J. K., T. R. Mackle, M. J. Auldist, N. A. Thomson, and D. E. Bauman. 2004. Endogenous synthesis of cis-9, trans-11 conjugated linoleic acid in dairy cows fed fresh pasture. J. Dairy Sci. 87:369-378. Kellens, M. J., H. L. Goderis, and P. P. Tobback. 1986. Biohydrogenation of unsaturated fatty acids by a mixed culture of rumen microorganisms. Biotech. Bioeng. 37:12681276. Kelly, M. L., J. R. Berry, D. A. Dwyer, J. M. Griinari, P. Y. Chouinard, M. E. Van Amburgh, and D. E. Bauman. 1998a. Dietary fatty acid sources affect conjugated linoleic acid concentrations in milk from lactating dairy cows. J. Nutr. 128:881-885. Kelly, M. L., E. S. Kolver, D. E. Bauman, M. E. Van Amburgh, and L. D. Muller. 1998b. Effect of intake of pasture on concentrations of conjugated linoleic acid in milk of lactating cows. J. Dairy Sci. 81:1630-1636. Kelly, M. L., E. S. Kolver, D. E. Bauman, M. E. Van Amburgh, and L. D. Muller. 1998c. Effect of intake of pasture on concentrations of conjugated linoleic acid in milk of lactating cows. J Dairy Sci. 81:1630-1636. Kemp, P., and D. J. Lander. 1976. Inhibition of the biohydrogenation of dietary C18 unsaturated fatty acids by rumen bacteria using some inhibitors of methanogenesis. Proc. Nutr. Soc. 35:31A-32A. Kepler, C. R., K. P. Hirons, J. J. McNeill, and S. B. Tove. 1966. Intermediates and products of the biohydrogenation of linoleic acid by Butyrinvibrio fibrisolvens. J. Biol. Chem. 241:1350-1354. Klieve, A. V., D. Hennessy, D. Ouwerkerk, R. J. Forster, R. I. Mackie, and G. T. Attwood. 2003. Establishing populations of Megasphaera elsdenii YE 34 and Butyrivibrio fibrisolvens YE 44 in the rumen of cattle fed high grain diets. J. Appl. Microbiol. 95:621-630. Klopfenstein, T. J., D. B. Purser, and W. J. Tyznik. 1966. Effects of defaunation on feed digestibility rumen metabolism and blood metabolites. J. Anim. Sci. 25:765-773. 123
Kolver, E. S., and M. J. de Veth. 2002. Prediction of ruminal pH from pasture-based diets. J. Dairy Sci. 85:1255-1266. Kramer, J. K., V. Fellner, M. E. Dugan, F. D. Sauer, M. M. Mossoba, and M. P. Yurawecz. 1997. Evaluating acid and base catalysts in the methylation of milk and rumen fatty acids with special emphasis on conjugated dienes and total trans fatty acids. Lipids. 32:1219-1228. Latham, M. J., J. E. Storry, and M. E. Sharpe. 1972. Effect of low-roughage diets on the microflora and lipid metabolism in the rumen. Appl. Microbiol. 24:971-877. Lawless, F., C. Stanton, P. L. Escop, R. Devery, P. Dillon, and J. J. Murphy. 1999. Influence of breed on bovine milk cis-9, trans-11-conjugated linoleic acid content. Livestock Prod. Sci. 62:43-49. Lee, M. R., L. J. Harris, R. J. Dewhurst, R. J. Merry, and N. D. Scollan. 2003a. The effect of clover silages on long chain fatty acid rumen transformation and digestion in beef steers. Anim. Sci. 76:491-501. Lee, M. R. F., E. M. Martinez, and M. K. Theodorou. 2003b. In vitro rumen plant enzyme mediated lipolysis:grass v. red clover. Conference on Gastrointestinal Function, Chicago, IL:pp 37. Lee, S., and D. B. Min. 1991. Effects, quenching mechanisms, and kinetics of nickel chelates in singlet oxygen oxidation of soybean oil. J. Agric. Food Chem. 39:642646. Lindroth, R. L., and P. A. Koss. 1996. Preservation of salicaceae leaves for phytochemical analyses: further assessment. J. Chem. Ecology. 22:765-771. Loor, J. J., and J. H. Herbein. 1998. Exogenous conjugated linoleic acid isomers reduce bovine milk fat concentration and yield by inhibiting de novo fatty acid synthesis. J Nutr. 128:2411-2419. Loor, J. J., W. H. Hoover, T. K. Miller-Webster, J. H. Herbein, and C. E. Polan. 2003. Biohydrogenation of unsaturated fatty acids in continuous culture fermenters during digestion of orchardgrass or red clover with three levels of ground corn supplementation. J. Anim. Sci. 81:1611-1627. 124
Makoni, N. F., and J. A. Shelford. 1993. Changes in lipids, chlorophyll pigments, hot water insoluble N and pH of alfalfa during ensiling. J. Sci. Food Agric. 63:273-280. Martin, J. C., and K. Valeille. 2002. Conjugated linoleic acids: all the same or to everyone its own function? Reprod. Nutr. Dev. 42:525-536. McDonald, P. 1981. The biochemistry of silage. John Wiley & Sons, Ltd ed. John Wiley and Sons Ltd, Bath, England. Mills, S. C., T. W. Scott, G. R. Russell, and R. M. Smith. 1970. Hydrogenation of C18 unsaturated acids by pure cultures of a rumen micrococcus. Aust. J. Biol. Sci. 23:1109-1113. Min, D. B. 1998. Lipid oxidation of edible oil. Pages 283-296 in Food Lipids. Akoh and Min, eds. Marcel Dekker, Inc., New York, NY. Min, D. B., and H. Lee. 1996. Chemistry of lipid oxidation. Pages 241-268 in Food Lipids and Health. McDonald and Min, eds. Marcel Dekker, Inc., New York, NY. Mir, Z., M. L. Rushfeldt, P. S. Mir, L. J. Paterson, and R. J. Weselake. 2000. Effect of dietary supplementation with either conjugated linoleic acid (CLA) or linoleic acid rich oil on the CLA content of lamb tissues. Small Ruminant Res. 36:25-31. Moate, P. J., W. Chalupa, T. C. Jenkins, and R. C. Boston. 2004. A model to describe ruminal metabolism and intestinal absorption of long chain fatty acids. Anim. Feed Sci. Technol. 112:79-105. Molkentin, J., and D. Precht. 1995. Optimized analysis of trans-octadecenoic acids in edible fats. Chromatographia. 41:267-272. Mosley, E. E., G. L. Powell, M. B. Riley, and T. C. Jenkins. 2002. Microbial biohydrogenation of oleic acid to trans isomers in vitro. J. Lipid Res. 43:290-296. Nelson, G. J. 1991. Isolation and purification of lipids from biological matrices. Pages 20-59 in Analyses of fats, oils and lipoproteins. E. G. Perkins, ed. Americam Oil Chemists' Society, Champain, IL.
125
Noble, R. C., J. H. Moore, and C. G. Harfoot. 1974. Observations on the pattern on biohydrogenation of esterified and unesterified linoleic acid in the rumen. Br. J. Nutr. 31:99-108. Ntambi, J. M. 1999. Regulation of stearoyl-CoA desaturase by polyunsaturated fatty acids and cholesterol. J. Lipid Res. 40:1549-1558. Nunez, A., T. A. Foglia, and G. J. Piazza. 2001. Characterization of lipoxygenase oxidation products by high-performance liquid chromatography with electron impact-mass spectrometric detection. Lipids. 36:851-856. Ørskov, E. R., and J. McDonald. 1979. The estimation of protein degradability in the rumen from incubation measurements weighed according to rate of passage. J. Agric. Sci. (Camb.). 92:499-503. Palmquist, D. L., and T. C. Jenkins. 1980. Fat in lactation rations: review. J. Dairy Sci. 63:1-14. Palmquist, D. L., and T. C. Jenkins. 2003. Challenges with fats and fatty acid methods. J. Anim Sci. 81:3250-3254. Palmquist, D. L., and F. L. Schanbacher. 1991. Dietary fat composition influences fatty acid composition of milk fat globule membrane in lactating cows. Lipids. 26:718. Park, Y., J. M. Storkson, J. M. Ntambi, M. E. Cook, C. J. Sih, and M. W. Pariza. 2000. Inhibition of hepatic stearoyl-CoA desaturase activity by trans-10, cis-12 conjugated linoleic acid and its derivatives. Biochim. Biophys. Acta. 1486:285-292. Parodi, P. W. 1997. Cows' milk fat components as potential anticarcinogenic agents. J. Nutr. 127:1055-1060. Perkins, E. G. 1996. Formation of lipid oxidation products during deep fat frying: Effects on oil quality and their determination. Pages 139-160 in Food Lipids and Health. McDonald and Min, eds. Marcel Dekker, Inc., New York, NY. Piwonka, E. J., and J. L. Firkins. 1993. Effect of glucose on fiber digestion and particleassociated carboxymethylcellulase activity in vitro. J. Dairy Sci. 76:129-139. 126
Piwonka, E. J., and J. L. Firkins. 1996. Effect of glucose fermentation on fiber digestion by ruminal microorganisms in vitro. J. Dairy Sci. 79:2196-2206. Piwonka, E. J., J. L. Firkins, and B. L. Hull. 1994. Digestion in the rumen and total tract of forage-based diets with starch or dextrose supplements fed to Holstein heifers. J. Dairy Sci. 77:1570-1579. Polan, C. E., J. J. McNeill, and S. B. Tove. 1964. Biohydrogenation of unsaturated fatty acids by rumen bacteria. J. Bacteriol. 88:1056-1064. Proell, J. M., E. E. Mosley, G. L. Powell, and T. C. Jenkins. 2002. Isomerization of stable isotopically labeled elaidic acid to cis and trans monoenes by ruminal microbes. J. Lipid Res. 43:2072-2076. Qiu, X., M. L. Eastridge, K. E. Griswold, and J. L. Firkins. 2004. Effects of substrate, passage rate, and pH in continuous culture on flows of conjugated linoleic acid and trans C18:1. J. Dairy Sci. 87:3473-3479. Radin, N. S. 1981. Extraction of tissue lipids with a solvent of low toxicity. Methods Enzymol. 72:5-7. Reis, R. B., and D. K. Combs. 2000. Effects of increasing levels of grain supplementation on rumen environment and lactation performance of dairy cows grazing grasslegume pasture. J. Dairy Sci. 83:2888-2898. Ribeiro, C. V. D. M., and M. L. Eastridge. 2004. Relationship of rate of appearance of vaccenic acid and pH during in vitro biohydrogenation of linolenic acid in alfalfa hay. J. Dairy Sci. 83 (Suppl. 1):37 (Abstr.). Romo, G. A., D. P. Casper, R. A. Erdman, and B. B. Teter. 1996. Abomasal infusion of cis or trans fatty acid isomers and energy metabolism of lactating dairy cows. J. Dairy Sci. 79:2005-2015. Russell, J. B., and R. L. Baldwin. 1978. Substrate preferences in rumen bacteria: evidence of catabolite regulatory mechanisms. Appl. Envir. Microb. 36:319-329. Russell, J. B., and R. L. Baldwin. 1979. Effects of combination of substrates on maximum growth rates of several rumen bacteria. Appl. Envir. Microb. 37:544-549. 127
Russell, J. B., and D. B. Dombrowski. 1980. Effect of pH on the efficiency of growth by pure cultures of rumen bacteria in continuous culture. Appl. Environ. Microbiol. 39:604-610. SAAM. 1997. SAAM II User Guide. SAAM Inst., Seattle, WA. Sackmann, J. R., S. K. Duckett, M. H. Gillis, C. E. Realini, A. H. Parks, and R. B. Eggelston. 2003. Effects of forage and sunflower oil levels on ruminal biohydrogenation of fatty acids and conjugated linoleic acid formation in beef steers fed finishing diets. J. Anim. Sci. 81:3174-3181. SAS. 2004. SAS/STAT User’s Guide. Version 8 ed. SAS Institute Inc, Cary, NC. Singh, S., and J. C. Hawke. 1979. The in vitro lipolysis and biohydrogenation of monogalactosyldiglyceride by whole rumen contents and its fractions. J. Sci. Food Agric. 30:603-612. Slyter, L. L., and P. A. Putnam. 1967. In vivo vs. in vitro continuous culture of ruminal microbial populations. J. Anim. Sci. 26:1421-1427. Solomon, R., L. E. Chase, D. Ben-Ghedalia, and D. E. Bauman. 2000. The effect of nonstructural carbohydrate and addition of full fat extruded soybeans on the concentration of conjugated linoleic acid in the milk fat of dairy cows. J. Dairy Sci. 83:1322-1329. Stanton, C., F. Lawless, G. Kjellmer, D. Harrington, R. Devery, J. F. Connolly, and J. Murphy. 1997. Dietary influences on bovine milk cis-9,trans-11 conjugated linoleic acid. J. Food Sci. 62:1083-1086. Sukhija, P. S., and D. L. Palmquist. 1988. Rapid method for determination of total fatty acid content and composition of feedstuffs and feces. J. Agric. Food. Chem. 36:12021206. Sylvester, J. T., S. K. Karnati, Z. Yu, M. Morrison, and J. L. Firkins. 2004. Development of an assay to quantify rumen ciliate protozoal biomass in cows using real-time PCR. J. Nutr. 134:3378-3384.
128
Timmem, H., and S. Patton. 1988. Milk fat globule: fatty acid composition, size and in vivo regulation of fat liquidity. Lipids. 23:685-689. Troegeler-Meynadier, A., M. C. Nicot, C. Bayourthe, R. Moncoulon, and F. Enjalbert. 2003. Effects of pH and concentrations of linoleic and linolenic acids on extent and intermediates of ruminal biohydrogenation in vitro. J. Dairy Sci. 86:4054-4063. Turpeinen, A. M., M. Mutanen, A. Aro, I. Salminen, S. Basu, D. L. Palmquist, and J. M. Griinari. 2002. Bioconversion of vaccenic acid to conjugated linoleic acid in humans. Am. J. Clin. Nutr. 76:504-510. van de Vossenberg, J. L., and K. N. Joblin. 2003. Biohydrogenation of C18 unsaturated fatty acids to stearic acid by a strain of Butyrivibrio hungatei from the bovine rumen. Lett. Appl. Microbiol. 37:424-428. Van der Veen, J., and H. S. Olcott. 1967. Changes in the lipids of alfalfa prior to and after dehydration. J. Agric. Food Chem. 15:682-684. Van Nevel, C., and D. I. Demeyer. 1995. Lipolysis and biohydrogenation of soybean oil in the rumen in vitro: inhibition by antimicrobials. J. Dairy Sci. 78:2797-2806. Van Nevel, C. J., and D. I. Demeyer. 1996. Influence of pH on lipolysis and biohydrogenation of soybean oil by rumen contents in vitro. Reprod. Nutr. Dev. 36:53-63. Van Soest, P. J. 1994. Nutritional ecology of the ruminant. Cornell University Press, Ithaca, NY. Van Soest, P. J., J. B. Robertson, and B. A. Lewis. 1991. Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J. Dairy Sci. 74:3583-3597. Wales, W. J., E. S. Kolver, P. L. Thorne, and A. R. Egan. 2004. Diurnal variation in ruminal pH on the digestibility of highly digestible perennial ryegrass during continuous culture fermentation. J. Dairy Sci. 87:1864-1871. Ward, P. F., T. W. Scott, and R. M. Dawson. 1964. The hydrogenation of unsaturated fatty acids in the ovine digestive tract. Biochem. J. 92:60-68. 129
Weisbjerg, M. R., T. Hvelplund, and B. M. Bibby. 1998. Hydrolysis and fermentation rate of glucose, sucrose and lactose in the rumen. Acta Agric. Scand. Sect. A, Anim. Sci. 48:12-18. White, R. W., P. Kemp, and M. C. Dawson. 1970. Isolation of a rumen bacterium that hydrogenates oleic acid as well as linolenic acid and linoleic acid. Biochem. J. 116:767-768. Wilde, P. F., and R. M. Dawson. 1966. The biohydrogenation of alpha-linolenic acid and oleic acid by rumen micro-organisms. Biochem. J. 98:469-475. Woolford, M. K. 1984. The Silage Fermentation. Vol. 14. Microbiology Series ed. Maercel Dekker, Inc., New York, NY. Wright, D. E. 1961. Lipase activity of rumen micro-organisms. N. Z. J. Agric. Res. 4:216-223. Wu, Z., and D. L. Palmquist. 1991. Synthesis and biohydrogenation of fatty acids by ruminal microorganisms in vitro. J. Dairy Sci. 74:3035-3046.
130
APPENDICES
APPENDIX A CLUSTER ANALYSES OF 16S RNA EXTRACTED FROM CONTINUOUS CULTURE
6
8
1 0
FA – P4 4% - P4 8% - P2 8% - P1 8% - P4 0% - P4 0% - P2 0% - P3 4% - P3 8% - P3 FA – P2 4% - P2 4% - P1 0% - P1 FA – P3 FA – P1 Ladder Ladder
Figure A.1: Cluster analysis of bacterial profiling using DDGE from samples of Experiment 3. 131
APPENDIX B DIAGRAM OF A MULTI-COMPARTIMENTAL MODEL OF RUMINAL BIOHYDROGENATION KINETICS USING CONTINUOUS CULTURE
Figure B.1: Diagram of a multi-compartmental model from SAAM II applied to samples taken over time using continuous culture. 132
APPENDIX C CONFOCAL MICROSCOPE IMAGES OF CHLOROPLASTS ENGULFED BY RUMINAL PROTOZOA
Figure C.1: Image of a protozoal cell taken from the omasum of a grazing cow. The red spots represent chlorophyll. Picture was made possible by the Ohio Agricultural Research and Development Center, Wooster, OH. 133
APPENDIX D PERCENTAGE OF FATTY ACIDS FROM PROTOZOA SAMPLED FROM THE OMASUM OF A COW FED FRESH FORAGE
Table D.1: Fatty acid profile of protozoa and bacteria collected from the omasum of dairy cows. Diet Grazing cow Protozoa Fatty acids
TMR-fed cow Protozoa
Bacteria
------------------------------- mg/100 mg ---------------------
16:0
42.0
32.9
28.1
18:0
14.5
17.0
41.2
18:1 cis-9
5.6
7.3
12.0
trans-18:1
4.8
7.9
trace
18:2
8.6
8.2
3.9
18:3
9.6
trace
trace
trans-18:1/18:0
0.33
0.46
_
trans-18:1/(18:2 +18:3)
0.26
0.96
_
Total1
3.8
6.9
4.9
UFA2
28.6
23.4
15.9
SFA3
56.5
49.9
69.3
1
Total fatty acids as % DM
2
Unsaturated fatty acids: 18:1 cis-9, trans-18:1, 18:2, and 18:3
3
16:0 plus 18:0 134
APPENDIX E ESTIMATION OF RATE CONSTANTS DURING IN VITRO BIOHYDROGENATION OF 14C-LABELED LINOLEIC ACID
Figure E.1: Diagram of a multi-compartmental model from SAAM II applied to data from in vitro incubation of 14C-labeled linoleic acid. 135
Table E.1: Rate constants and standard error (in parenthesis) of in vitro BH of three different initial amounts of 14C-labeled linoleic acid estimated by SAAM II.
-1
Fractional rates, h k (2,1) k (3,2) k (4,3)
0.32 10.0 (0.633) 6.4 (1.163) 0.33 (0.068)
Linoleic acid, mg/ml 0.65 5.8 (0.365) 2.2 (0.125) 0.12 (0.038)
136
0.95 3.7 (0.163) 0.48 (0.063) 0.04 (0.020)
APPENDIX F IDENTIFICATION OF OCTADECENOIC FATTY ACIDS
The identification of trans/cis 18:1 fatty acid peaks were performed by comparing the peaks of the isothermal region from different samples taken at different time points during in vitro incubation of fresh alfalfa (Figure F.1). It is known that trans-12 and trans-13 18:1 elute between VA and cis-9. The VA peak is very characteristic because it was always the biggest one in all time points after 1 h of incubation. The peak before it was identified as trans-10 and the two peaks after it were identified as trans-12 and trans-13, respectively. During BH of forages, trans-13 is the second major trans-18:1 FA. The identification of trans-15 and cis-15 was possible because these FA are expected to increase during BH of 18:3. I observed an increase of two peaks (8 and 11; Figure F.1) that could correspond to trans- and cis-15 18:1 FA. Major concern was regarding trans15, because it elutes close to cis-9. Any mistake in that region would reflect misidentification of the others 18:1 FA. Because the region where both trans-15 and cis-9 elute had two peaks, where one increased with time (peak 8) while the other one decreased (peak 9), I identified the latest as trans-15 and the other as cis-9. Also, the peak identified as trans-15 had the same amount as the other one identified as cis-15 at all time points (peaks 8 and 11). For this reason, the peak after trans-13 was identified as cis-9 and the peak after that, trans-15. Then I followed the literature to identify the next peaks as cis-11 and cis-12.
137
A
Figure F.1: Temporal change of peak areas and identification of octadecenoic fatty acids during in vitro biohydrogenation. A) 1 h of incubation; peak 3 was higher than peak 4; B) 3 h of incubation; peak 4 is now predominant; C) 4 h of incubation; peak 6 is higher than peak 7; D) 6 h of incubation; E) 9 h of incubation; and F) 12 h of incubation with the peaks identified.
(Continue)
138
Figure F.1: Continue FID1 A, (A204 01\A204 010 5 .D) c o unts 1 600
1 500
B
1 400
1 300
1 200
1 100
32
33
34
35
36
37
38
39
40
min
FID1 A, (A204 01\A2040106 .D) counts 1700
1600
1500
C
1400
1300
1200
1100
33
34
35
36
37
38
39
40
(Continue)
139
min
Figure F.1: Continue FID1 A, (A204 01\A2040107 .D) counts
1900
1800
1700
D
1600
1500
1400
1300
1200
1100
33
34
35
36
37
38
39
40
min
FID1 A, (A204 01\A2040108 .D) counts 2200
2000
1800
E
1600
1400
1200
33
34
35
36
37
38
39
40 min
(Continue)
140
Figure F.1: Continue
19:0
18:0
FID1 A, (A20401\A2040109.D)
2200
F
18:1 t11
counts
2000
1800
33
34
35
36
141
37
18:1 c15
18:1 c12
18:c11
18:1 t15
18:1 t6/8 18:1t9
1200
32
18:1 t12
1400
18:1 c9
18:1 t10
18:1 t13
1600
38
39
40 min
