‘l at.

I. 3.93,. 02 a .

; .uétsfinavyuor
«an .21.;

f... rt. ‘\
I .1211

33.7.:

‘1 16.4%.. r
:Lrfir .fiKILfiHL
$3.. .~.nu.,.:.p.z.u..v
tawhm...

. e

L

l
.7.‘
z

M. ii: . 2f...
1. 14.21.63... I
) .6‘1‘)Piic§.€t’.

 

THESTS
r 2

706 L

This is to certify that the
dissertation entitled
Effects of propionate on
feeding behavior of

lactating dairy cows

presented by
Masahito Oba

has been accepted towards fulfillment
of the requirements for

 

 

Ph.D. degree in Animal Science

M ' rprofessor

Date 3/1§/02

MS U is an Affirmative Action/Equal Opportunity lmlitufl'on 0-12771

 

 

 

LIBRARY
Michigan State
Universlty

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

MAR 0 8 2804

'ndfllL‘l

 

14,1th

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRC/DateDue.p65-p.15

EFFECTS OF PROPIONATE ON
FEEDING BEHAVIOR OF LACTATING DAIRY COWS

By

Masahito Oba

A DISSERTATION

Submitted to
Michigan State University
In partial fulfillment of the requirements
For the degree of

DOCTOR OF PHILOSOPHY
Department of Animal Science

2002

ABSTRACT

EFFECTS OF PROPIONATE ON
FEEDING BEHAVIOR OF LACTATING DAIRY COWS

By

Masahito Oba

Increasing ruminal starch digestion often results in a sharp reduction in DMI of lactating
dairy cows. Feeding ground high moisture corn increased true ruminally degraded
organic matter by 1 kg/d and decreased dry matter intake (DMI) by 1.7 kg/d compared to
dry ground corn when cows were fed a high starch diet. We hypothesized that excess
propionate production in the rumen has hypophagic effects by stimulating oxidative
metabolism in the liver. Intra-ruminal infusion of propionate decreased DMI, energy
intake, meal size, and meal frequency in a dose-dependent manner. Hypophagic effects
of propionate were greater when infused with ammonium compared to sodium although
ammonium did not affect DMI when infused with acetate. Infusion of ammonium
stimulated urea synthesis, and amino acid carbon generated from the urea cycle might
have enhanced oxidative metabolism in the liver. We also hypothesized that hypophagic
effects of propionate infusion are greater for cows fed more fermentable diets because of
greater basal propionate production from ruminal fermentation. Contrary to our
hypothesis, the magnitude of hypophagic effects of propionate was not affected by diets
differing in ruminal starch digestibility of grains or forage to concentrate ratio, indicating
that propionate flux from the rumen does not directly decrease feed intake. Propionate

infusion did not cause hypophagia when it resulted in a large increase in plasma glucose

concentration. However, when propionate infusion resulted in slight increases in plasma
glucose concentration, propionate decreased DMI. A possible explanation for these
observations is that propionate stimulates oxidative metabolism in the liver to a greater
extent when propionate is not extensively utilized for gluconeogenesis. This suggests
that hypophagic effects of propionate are linked to glucose demand of animals and
subsequent gluconeogenesis in the liver. Consistent with this speculation, propionate at
higher rates of infusion decreased DMI to a greater extent in lower producing cows
compared to higher producing cows. It is concluded that hypophagic effects of
propionate in lactating dairy cows are consistent with its effects at stimulating oxidative
metabolism in the liver and that a temporal pattern of oxidative metabolism in the liver

might be related to senses of satiety and hunger.

ACKOWLEDGMENTS

I would like to thank Dr. Allen for his guidance and financial support throughout my
graduate program at Michigan State University. He has guided me through one of the
most fertile and exciting areas of research and helped me realize the fun of science. What
I learned from him is not limited to the biology of dairy cows but includes life
philosophies. I learned that everything needs to be done thoroughly to tell a beautiful
story. I also learned the importance of critical thinking and interactions with colleagues
to develop a variety of research ideas. I have really enjoyed the time I spent with him

discussing or arguing regulation mechanisms of feed intake.

I would also thank Dr. Beede, Dr. Fogwell, Dr. Herdt, and Dr. Romsos for serving as my
graduate committee members and for making valuable advice on my research projects. I
am also grateful for Dr. Ames for performing ruminal and duodenal cannulation surgery
and Dr. Tempelman for assistance in statistical analysis. I would like to thank Dr.
Bucholtz and Dr. VandeHaar for making my graduate life more interesting. I would like
to express my special thanks to Dave Main and Dewey Longuski for their friendship and
support as mentors of American culture and English conversation. I also extend my
appreciation to Steve Mooney, Jackie Yun Ying, Jill Davidson, Bob Kreft, Jim Liesman,

Larry Chapin, and Bob Burnett for their excellent technical assistance.

Finally, I would like to thank Keiko and Maki for sharing the fun of life with me.

iv

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. x
LIST OF FIGURES .............................................................................. xiii
LIST OF ABBREVIATIONS ................................................................... xv
INTRODUCTION .................................................................................. 1
CHAPTER 1
LITERATURE REVIEW
Factors Affecting DMI ..................................................................... 3
Effects of Site of Starch Digestion on DMI ............................................. 5
Effects of Propionate on DMI ............................................................. 9
Mechanisms for hypophagic effect of propionate .................................... 12
Summary .................................................................................... 24
CHAPTER 2

Effects of corn grain conservation method on feeding behavior and productivity of
lactating dairy cows at two dietary starch concentrations

ABSTRACT ................................................................................. 26
INTRODUCTION ......................................................................... 27
MAERIALS AND METHODS .......................................................... 29
Treatment and Cows ............................................................... 29
Data and Sample Collection ...................................................... 32
Sample and Statistical Analysis .................................................. 33
RESULTS AND DISCUSSION ......................................................... 36
DMI and Ruminal Fermentation ................................................. 36

Milk Production and Plasma Metabolites ....................................... 43
CONCLUSION ............................................................................. 51
CAHPATER 3

Dose-response effects of intra-ruminal infusion of propionate on energy intake and
feeding behavior of lactating dairy cows

ABSTRACT ................................................................................. 52
INTRODUCTION ......................................................................... 53
MAERIALS AND METHODS .......................................................... 56
Experiment 1 ........................................................................ 56
Experiment 2 ........................................................................ 61
RESULTS ................................................................................... 63
Experiment 1 ........................................................................ 63
Experiment 2 ........................................................................ 63
DISCUSSION .............................................................................. 67
Feed Intake .......................................................................... 67
Water Intake ........................................................................ 70
Milk Production ..................................................................... 71
CONCLUSION ............................................................................. 72
CHAPTER 4

Effects of intra-ruminal infusion of sodium, potassium, and ammonium on
hypophagic effect of propionate in lactating dairy cows

ABSTRACT ................................................................................. 73
INTRODUCTION ......................................................................... 74
MAERIALS AND METHODS .......................................................... 76

vi

RESULTS AND DISCUSSION ......................................................... 81

DMI and Feeding Behavior ....................................................... 81

Chewing and Drinking Activities ................................................ 85

Milk Production ..................................................................... 86

CONCLUSION ............................................................................. 88
CHAPTER 5

Effects of ammonium infused with propionate or acetate on feeding behavior and
DMI in lactating dairy cows

ABSTRACT ................................................................................. 89
INTRODUCTION ......................................................................... 90
MAERIALS AND METHODS .......................................................... 92
RESULTS AND DISCUSSION ......................................................... 97
DMI ................................................................................ 100
Feeding Behavior ................................................................. 103
Chewing Behavior ................................................................ 105
Milk Production .................................................................... 106
CONCLUSION ........................................................................... 107
CHAPTER 6

Dose-response effects of intra-ruminal infusion of propionate on feeding behavior of
lactating cows fed diets differing in fermentability

ABSTRACT ............................................................................... 108
INTRODUCTION ........................................................................ 109
MAERIALS AND METHODS ......................................................... 112

Experiment 1 ...................................................................... 112

vii

Experiment 2 ...................................................................... 114

Common Infusion Protocol ...................................................... 114
Data and Sample Collection ..................................................... 116
Sample and Statistical Analysis ................................................ 118
RESULTS AND DISCUSSION ........................................................ 122
Experiment 1 ...................................................................... 122
Experiment 2 ...................................................................... 131
Threshold Response .............................................................. 139
Hypophagic Effects of Acetate .................................................. 146
CONCLUSION ........................................................................... 148
CHAPTER 7

Dose-response effects of intra-ruminal infusion of propionate on feeding behavior of
lactating cows in early or mid stage of lactation

ABSTRACT ............................................................................... 149
INTRODUCTION ........................................................................ 150
MAERIALS AND METHODS ......................................................... 152
Experimental Design and Treatments .......................................... 152
Data and Sample Collection ..................................................... 155
Sample and Statistical Analysis ................................................ 157
RESULTS AND DISCUSSION ........................................................ 160
Feeding Behavior and DMI ...................................................... 160
Plasma Metabolites ............................................................... 174
CONCLUSION ........................................................................... 178

viii

CHAPTER 8
IIVIPLICATION S ................................................................................. l 80
APPENDIX

The effect of pulse dose of sodium VF A

ABSTRACT ............................................................................... 1 86
BACKGROUND ......................................................................... l 87
RESULTS ................................................................................. 1 89
DISCUSSION ............................................................................. 189
CONCLUSION ........................................................................... 192
REFERENCES .................................................................................... 193

ix

LIST OF TABLES

CHAPTER 1

Table 1. Animal and dietary characteristics (mean across treatments) for

experiments that reported significant difference in starch NSC digestibility

in the rumen ........................................................................................... 8

CHAPTER 2

TABLE 1. Nutrient composition of corn grains used to formulate experimental
diets ................................................................................................... 30

TABLE 2. Ingredients and nutrient composition of experimental diets (% of
dietary DM) ........................................................................................... 31

TABLE 3. Effects of corn grain conservation method at two dietary starch
contents on feeding behavior ...................................................................... 37

TABLE 4. Effects of corn grain conservation method at two dietary starch
contents on rumen fermentation ................................................................... 40

TABLE 5. Effects of corn grain conservation method at two dietary starch
contents on productivity ............................................................................ 44

TABLE 6. Effects of corn grain conservation method at two dietary starch
contents on plasma metabolite and hormone concentration ................................... 47

CHAPTER 3

TABLE 1. Ingredients and nutrient composition of experimental diets (% of
dietary DM except for DM) ........................................................................ 58

TABLE 2. Dose response effects of intra-ruminal infusion of propionate relative
to acetate on feeding behavior for experiment 1 ................................................. 64

TABLE 3. Dose response effects of intra—ruminal infusion of propionate relative
to acetate on feeding behavior and milk production for experiment 2 ....................... 65

CHAPTER 4

TABLE 1. Ingredients and nutrient composition of experimental diets (% of
dietary DM except for DM) ........................................................................ 78

TABLE 2. Effects of intra-ruminal infusion of propionate salts on feeding
behavior and milk production ...................................................................... 83
CHAPTER 5

TABLE 1. Ingredients and nutrient composition of experimental diets (% of
dietary DM except for DM) ........................................................................ 93

TABLE 2. Effects of intra-ruminal infusion of VFA salts on feeding behavior
and milk production ................................................................................. 98

CHAPTER 6

TABLE 1. Ingredients and nutrient composition of experimental diets in
experiment 1 (% of dietary DM except for DM) ............................................. 113

TABLE 2. Ingredients and nutrient composition of experimental diets in
experiment 2 (% of dietary DM except for DM) .............................................. 115

TABLE 3. Effects of processing method of corn grain on feeding behavior,
ruminal fermentation, blood metabolites, nutrient digestibility, and milk
production ........................................................................................... 123

TABLE 4. Dose—response effects of intra-ruminal infusion of sodium propionate
relative to sodium acetate on feeding behavior and milk production for cows fed
steam flaked corn or dry cracked corn .......................................................... 125

TABLE 5. Dose-response effects of intra-ruminal infusion of sodium propionate
relative to sodium acetate on metabolites in the rumen and the blood for cows fed

steam flaked corn or dry cracked corn .......................................................... 128

TABLE 6. Effects of forage to concentrate ratio on feeding behavior, ruminal
fermentation, blood metabolites, nutrient digestibility, and milk production .............. 132

TABLE 7. Dose-response effects of intra-ruminal infusion of sodium propionate
relative to sodium acetate on feeding behavior and milk production for cows fed

diets differed in forage to concentrate ratio .................................................... 134

TABLE 8. Dose-response effects of intra—ruminal infusion of sodium propionate

xi

relative to sodium acetate on metabolites in the rumen and the blood for cows fed

diets differed in forage to concentrate ratio .................................................... 137
TABLE 9. Characteristics of cows used in experiment 1 and experiment 2 ............... 142
CHAPTER 7

TABLE 1. Ingredients and nutrient composition of experimental diet (% of
dietary DM except for DM) ...................................................................... 154

TABLE 2. Effects of stage of lactation on feeding behavior, ruminal fermentation,
blood metabolites, nutrient digestibility, and milk production ............................... 161

TABLE 3. Dose-response effects of intra-ruminal infusion of sodium propionate
relative to sodium acetate on feeding behavior and milk production for cows in
early and mid lactation ............................................................................. 164

TABLE 4. Dose-response effects of intra-ruminal infusion of sodium propionate

relative to sodium acetate on DMI and interrneal interval for cows in early and

mid lactation ........................................................................................ 166
TABLE 5. Dose-response effects of intra-ruminal infusion of sodium propionate

relative to sodium acetate on metabolites in the rumen and the blood for cows in
early and mid lactation ............................................................................. 172

APPENDIX

TABLE 1. Characteristics of cows that downed by the infusion protocol (DOWN)
and those appeared to be normal (NORM) ..................................................... 191

xii

LIST OF FIGURES

CHAPTER 1
Figure 1. Propionate is utilized for gluconeogenesis ........................................... 15
Figure 2. Propionate stimulates oxidative metabolism by driving the TCA cycle .......... 16

Figure 3. Propionate is oxidized completely to carbon dioxide after entering the
TCA cycle as acetyl CoA ........................................................................... 17
CHAPTER 2

Figure 1. Relationship between true ruminally degraded OM (TRDOM) and
total ruminal VFA concentration (r2 = 0.05, P > 0.21) ......................................... 42

Figure 2. Effect of dietary starch concentration on plasma insulin concentration
relative to feeding time .............................................................................. 48

Figure 3. Effect of dietary starch concentration on plasma glucose concentration

relative to feeding time ............................................................................. 49
CHAPTER 6

Figure 1. Dose-response effects of intra-ruminal propionate infusion on DMI

(kg/ 12h). Pooled means for DMI within an experiment at each infusion rate were
summarized ......................................................................................... 140
Figure 2. Dose-response effects of intra-ruminal propionate infusion on plasma

glucose concentration (mg/d1). Pooled means for plasma glucose concentration
within an experiment at each infusion rate were summarized ............................... 145

CHAPTER 7

Figure 1. Dose-response effect of intra-ruminal propionate infusion on DMI
(kg/ 12h) for cows in early lactation (EL) and mid lactation (ML) .......................... 167

Figure 2. Dose-response effect of intra-ruminal propionate infusion on meal size
(kg) for cows in early lactation (EL) and mid lactation (ML) ............................... 168

Figure 3. Dose-response effect of intra-ruminal propionate infusion on intermeal
interval (min) for cows in early lactation (EL) and mid lactation(ML) ..................... 169

xiii

Figure 4. Dose-response effect of intra-ruminal propionate infusion on plasma
glucose concentration (mg/d1) for cows in early lactation (EL) and mid lactation
(ML) .................................................................................................. 175

Figure 5. Dose-response effect of intra—ruminal propionate infusion on plasma

B-hydroxy butyrate (BHBA; mg/dl) for cows in early lactation (EL) and mid lactation
(ML) .................................................................................................. 176

CHAPTER 8

Figure 1. Relationship between marginal response in DMI (kg) per mmol/min of

propionate infusion and plasma glucose concentration (mg/d1). Marginal response
in DMI = 3.0 - 0.05 x plasma glucose concentration (1'2 = 0.26; P < 0.01) ................ 184

xiv

ADF

BCS

BHBA

BW

CP

DC

DG

DM

DMI

EL

MUN

NEFA

SCM

SF

TRDOM

LIST OF ABBREVIATIONS

Acid Detergent Fiber

Body Condition Score
B-Hydroxy Butyric Acid

Body Weight

Crude Protein

Dry Cracked Corn

Dry Ground Corn

Dry Matter

Dry Matter Intake

Cows in Early Stage of Lactation
High Forage

High Moisture Corn

Low Forage

Metabolizable Energy

Cows in Mid Stage of Lactation
Milk Urea Nitrogen

Neutral Detergent Fiber

Non Esterified Fatty Acids
Solids-Corrected Milk

Steam Flaked Corn

Truly Ruminally Degraded Organic Matter

Volatile Fatty Acids

XV

INTRODUCTION

Maximizing energy intake is an important goal in nutritional management for
high producing dairy herds. As the genetic potential for dairy cows to produce milk
increases, maximum productivity and profitability of high yielding dairy herds have
become more dependent on energy intake. Cows in early lactation often experience
negative energy balance, and energy status at early lactation greatly affects peak milk
yield and persistency of milk production. Because cows in early lactation mobilize body
reserves and sustain high milk yield, they are more susceptible to metabolic diseases such
as ketosis and fatty liver. One approach to increase energy intake is to increase energy
density of diets by feeding more fermentable grains. However, greater starch digestibility
can reduce feed intake (Allen, 2000), and may not necessarily increase energy intake.
Maximizing dry matter intake (DMI) is another important approach to increase energy
intake of lactating dairy cows. It is first necessary to elucidate the mechanisms regulating

voluntary feed intake for lactating dairy cows to maximize energy intake.

Shifting site of starch digestion from the rumen to the intestines by feeding less
fermentable grains often increases DMI of lactating dairy cows because excess
fermentation in the rumen is considered to limit maximum voluntary feed intake in
ruminants (Allen, 2000). Propionate is one of the major metabolic fuels for the ruminant
liver, and Allen (2000) proposed that oxidative metabolism of propionate in the liver is a
primary factor regulating feed intake of ruminants. Many studies on the effects of

propionate infusion on feed intake have been reported in the literature. However, most of

previous studies failed to determine the specific effect of propionate on feed intake
because effects of propionate were confounded with difference in pH or osmolarity of
infusates. Most studies in the literature evaluated the effect of propionate on meal size
only and did not evaluate its effect on intermeal interval although feed intake is regulated
by satiety and hunger and animals can compensate decreased meal size by increasing
meal frequency to maintain feed intake. In addition, some experiments did not discuss
the effect of propionate on energy intake. It is important to determine whether propionate
affects energy intake because our goal in nutritional management of dairy cows is to
maximize energy intake. In addition, the role of propionate in feed intake regulation
might differ depending on diets or energy requirement and physiological status of
animals. Previous research did not evaluate the interaction of propionate with other

factors extensively.

The objectives of this dissertation research were to evaluate effect of site of starch
digestion on feeding behavior of lactating cows, to determine the specific effect of
propionate on energy intake, and to investigate the mechanism by which propionate

affects feed intake.

CHAPTER 1

LITERATURE REVIEW

Factors Affecting DMI

Mechanisms that regulate DMI in lactating dairy cows have been studied
extensively. In general, cows are believed to consume feeds to meet their energy
requirement. When cows are fed high energy diets which are palatable, digestible and
low in fill, feed intake is thought to be regulated by the energy density of the ration
(Mertens, 1994). According to the theory of Conrad et al. (1964), voluntary feed intake
is regulated by either energy density of a diet, or physical fill depending on the
digestibility of the diet consumed. However, this theory over-simplifies the complex
mechanisms that limit DMI of dairy cows. For example, some studies in the literature
(McCarthy et al., 1989; Casper et al., 1990; Overton et al., 1995) showed that less
fermentable grains increase energy intake of lactating cows compared to more
fermentable grains. Cows in those experiments increased DMI more than enough to
compensate for the lower digestibility of less fermentable grains. Greater energy intake
allowed them to increase milk production on less fermentable diets. If animals consume
feeds to meet their energy requirement, we cannot explain these observations. Instead, it
is more logical to think that greater fermentation in the rumen directly or indirectly sends
a satiety signal to regulate DMI. Forbes (1995) suggested that feed intake is regulated by

various factors depending upon metabolic status of animals, and integration of several

stimuli generally contributes to the regulation of voluntary feed intake. We need to

understand what regulates voluntary feed intake of animals more specifically.

Voluntary DMI is regulated by humoral signals coordinating whole body
metabolism. Ingvartsen and Andersen (2000) reviewed the humoral signals that affect
DMI of lactating dairy cows, and suggested hormones such as estrogen, CCK, leptin, and
cytokines might be involved in regulation of feed intake. The transient decrease in DMI
observed in periparturient animals is of interest from the viewpoint of long-term feed
intake regulation. A metabolic characteristic during the periparturient period is elevated
plasma NEFA concentration. Fatty acid metabolism can be linked to the decrease in
DMI during this period because low intake is particularly observed for obese animals
compared to thin animals (Bines and Morant, 1983). Fatty acid oxidation in the liver has
been shown to decrease feed intake in rats (Friedman et al., 1999). Leptin is primarily
produced by adipocytes, and plasma leptin concentration increases as body reserve mass
increases (Maffei et al., 1995). Significant evidence exists for regulation of feed intake

by leptin in rats although little is known about the biology of leptin in ruminant animals.

Dietary factors that affect short-term regulation of feed intake in lactating dairy
cows were recently reviewed (Allen, 2000). In that review, Allen (2000) showed that site
of starch digestion often affects DMI. Starch can be digested either in the rumen or in the
intestines. Starch fermented in the rumen produces volatile fatty acids, and starch
digested in the small intestine is absorbed as glucose. When cows are fed grains that are

rapidly fermented in the rumen, DMI sometimes decreased compared to when cows are

fed grains that are less fermented in the rumen. Greater ruminal fermentation is
characterized by factors such as low ruminal pH, increased osmolarity of ruminal fluid,
and greater fermentation acid production. Choi and Allen (1999) showed that propionate
has greater hypophagic effects in lactating dairy cows compared to acetate, reduced

ruminal pH, or increased osmolarity of ruminal fluid.

Effects of Site of Starch Digestion on DMI

Enhanced ruminal fermentation does not always decrease DMI (Theurer et al.,
1999; Allen, 2000). A recent review by Theurer et a1. ( 1999) reported that steam-flaked
corn and sorghum increased starch digestibility compared to steam rolled corn (6
comparisons) and dry rolled sorghum (24 comparisons), respectively. However, neither
steam-flaked corn compared to steam-rolled corn nor steam-flaked sorghum compared to
dry—rolled sorghum affected DMI. Recent studies that measured starch digestibility in the
rumen using cows with duodenal cannulas were summarized to determine the effect of
ruminal starch digestibility on DMI (Allen, 2000). Although three comparisons
(McCarthy et al., 1989; Overton et al., 1995) showed a significant decrease in DMI for
treatments with greater ruminal starch digestibility, ruminal starch digestibility did not
affect DMI for the others comparisons (Crocker et al., 1998; Knowlton et al., 1998a;
Oliveira et al., 1995; Plascencia and Zinn, 1996; Poore et al., 1993). Other experiments
(Herrera-Saldana and Huber, 1989; Wilkerson et al., 1997; Lycos et al., 1997; Grings et
al., 1992; Callison et al., 2001; Oliveira et al., 1993) comparing different types of grain,

processing methods, or conservation methods that are expected to result in different

ruminal degradability of starch failed to show the negative effects of ruminal starch (or

non-structural carbohydrate) digestion on DMI.

These inconsistent observations might indicate that a threshold exists for the
effect of fermentation acid production in the rumen on DMI. The concept for a threshold
response agreed with observations of Moore et a1. (1992). Moore et a1. (1992) fed
sorghum grains processed as either dry rolled (DR), steam flaked to the flake density of
0.40 kg/L (SF40), or steam flaked to the flake density of 0.27 kg/L (SF27) to dairy cows
at 41.5% of dietary DM. Rate of enzymatic starch hydrolysis measured in vitro was
greater for SF27 compared to SF40 and DR, and for SF40 compared to DR. However,
DMI for DR, SF40, and SF27 was 25.7, 25.4, and 23.8 kg/d (SE = 0.5), respectively.
Milk yield for DR, SF40, and SF27 was 31.0, 33.3, and 31.7 kg/d (SE = 0.6),
respectively. Increasing starch digestibility from DR to SF40 did not affect DM intake
and increased milk yield. However, further increase in starch digestibility from SF40 to
SF27 decreased both DMI and milk yield. Although actual starch digestibility in the
rumen was not measured in this experiment, it is speculated that the amount of
fermentation acid produced in the rumen might not be high enough to affect DMI for

cows fed SF40 but exceeded a threshold to affect DMI for the cows fed SF27

The threshold for ruminal fermentation to affect DMI, if it exists, might be altered
by physiological status of animals, other dietary characteristics, or basal level of
fermentation to which effects of treatment are compared. Some animal and dietary

characteristics from recent experiments that reported significant difference in starch or

NSC digestibility in the rumen using duodenally cannulated cows were summarized for
statistical evaluation (Table 1). Experiments were classified either Y or N; the
experiments that report more fermentable grains decreased DMI were classified as Y and
the experiments that report similar DMI regardless of fermentability of grain were
classified as N. Treatment means were averaged within an experiment to characterize
each study for comparison with others. Although some response variables (e. g. ruminal
pH) were affected by treatments, the difference within an experiment was relatively less
compared to the difference across the experiments. The objective of this statistical
analysis was to determine factors affecting response in DMI to fermentability of grains.

Data were analyzed by one-way ANOVA using JMP (SAS Institute, Cary, NC).

The different response in DMI to fermentability of grains could not be explained
by differences in days in milk, BW, or milk yield of cows across the experiments.
However, ruminal pH and acetate concentration were lower (P < 0.05; 6.17 vs. 6.43 and '
P < 0.01; 58.1 vs. 63.0 mol/100 mole VFA, respectively) and propionate concentration
was higher (P < 0.01; 28.0 vs. 21.8 moles/ 100 moles of VFA) in the experiments for
which more fermentable grains decreased DMI compared to the experiments where more
fermentable grains did not decrease DMI. Dietary crude protein concentration tended to
be lower (P < 0.11; 16.3 vs. 18.2 %) for experiments for which DMI was decreased by
more fermentable grains compared to the experiments that DMI was not affected by
treatments. These results might indicate basal diet characteristics affected how animals
responded to more fermentable grains. Lower ruminal pH and acetate concentration and

higher propionate concentration in experiments for which DMI was decreased by greater

:29 336% 8: En 262w 0383928 208 35 macaw—beau 2S. 2

H55 @3383 £82m ofiflconcom 808 “m5 masonhtomxo 05. >

Branch “oz “m2

038 88:50:00 9 owflom ”0:8 Um
“Banana Lo “5328 05 an x28 5 $95 ”Sun—

 

 

 

 

W: v.§ ode mmé cho WM: Wmm no mmm 2 83m 62?. use 30
W: fimm W5 36 Omuom WE 9mm o2 can 2 Sam :3 Ho SEED
W2 Wow 3% 3:0 mmnmv 0.2 gm mm com 2 moi :3 8 H8380
I: 5.3 W 8 wfio mmumv WON Sam mm com 2 £93 :3 Ho "5:329
fimfi WNN fine 93 bmumv m2 Sam 9: m2 2 82 .:EN 23 «6:883
m2 m2 m2 m2 mcnmm WE W5. 9: MZ z moi :3 Co 8635
.mZ m2 m2 m2 003 v.2 32 wow mo z 39 in 8 Econ—
92 Wmm me 2 .o $an 0.3 o.wm no mmm > ooom do=< use .30
Wo odm 5.3 va mmumv NB WmN vfl wmm > 33 r? 8 8:30
We fiwm fiwm and mmumv WE oém mm mwm s $3 :3 so CEHBUSZ
28>“an Bacofiem 8803.. In came on £25 9:wa Awoc
:08 8525 <m> BEBE Ragga 65 Bo; M52 25 3m

 

See 2: a $3503 82

sebum E ooceobfi Emu—haw? canon“: 35 mucoatomxo 8m @325on 3.83 525 83330820 38% 93 125.2. A 033.

ruminal fermentation suggest that propionate production for the basal diet was close to
the threshold for affecting DMI. If basal ruminal fermentation is close to a threshold to
affect DMI, feeding more fermentable grains in the diet might readily decrease DMI. In
addition, greater crude protein concentration might make animals more tolerant to the
more fermentable diets (maintaining DMI with greater ruminal fermentation) possibly by

increasing milk yield that is allowable by metabolizable protein.

Effects of Propionate on DMI

Hypophagic effects of propionate infusion in ruminants were extensively
documented and summarized by Allen (2000). Some experiments in the literature
reported that propionate infusion did not decrease feed intake (Deetz and Wangsness,
1981; Quigley and Heitmann, 1991; De Jong et al., 1981; Anil etal., 1993). Intrajugular
infusions of propionate did not decrease DMI of sheep while infusion of insulin and
glucagon did (Deetz and Wangsness, 1981). Portal infusion of propionate did not
decrease DMI of sheep (Quigley and Heitmann, 1991). Infusion of VFA mixture either
into the portal or the jugular vein did not affect feed intake of goats (De Jong et al.,
1991). Anil et a1. (1993) showed that infusion of Na propionate at the rate of 44.4
mmol/rnin decreased feed intake of dairy cows in one experiment. However,
intraruminal infusions of Na propionate at rates of 22.2 mmol/min and 44.4 mmol/rnin
had no effect on hay intake in two other experiments (Anil et al., 1993). Lactating cows
used in these experiments differed in age and stage of lactation. Differences in glucose
demand and energy balance of experimental animals probably affected intake response to

propionate infusion in these experiments. It is not likely that a single factor limits

maximum feed intake. A variety of satiety signals can synergistically interact with each
other to regulate maximum feed intake although one factor might become more dominant

than the others.

Mbanya et al. (1993) suggested that DMI can be affected by combined effects of
acid production and distention, not by one of the effects alone. They infused acetate,
propionate, or both, with or without distention of the rumen by a balloon. Combination
of VFA infusion and reticulo-rumen distention significantly depressed DMI while VFA
infusion or distention alone did not. Integration of physical fill and metabolic satiety
signals contributes to the regulation of voluntary feed intake. Similarly, Farningham et
a1. (1993) reported that portal infusion of either sodium propionate at 1.2 mmol/min or
CCK had no effect on food intake of sheep, but together decreased feed intake by 44%.
This reduction in feed intake was similar to when sodium propionate was infused at 2.4
mmol/min. These observations indicated that the response to propionate infusion in feed

intake depends on infusion rate of propionate, and other dietary or physiological factors.

The hypophagic effect of propionate infusion might be altered by fermentability
of the basal diets (Allen, 2000). Leuvenink et a1. (1997) fed sheep with a pelleted grass,
and reported that propionate infusion into the mesenteric vein of mature sheep at a rate of
2 mmol/min decreased intake but infusion at a rate of 1 mmol/rnin had no effect. In the
experiment of Farningham and Whyte (1993), sheep were fed a pelleted diet containing
50% hay, 30% barley, and 10% molasses offered ad lib. Infusion of sodium propionate

decreased feed intake linearly when infused over 3-h periods at rates between 0 and 83

10

mmol/min. The diet fed by Farningham and Whyte (1993) was more fermentable
compared to the diet fed by Leuvenink et a1. (1997). Inconsistent responses to propionate
infusion in DMI may be partially because of differences in fermentability and propionate
production in basal diets. Total amount of propionate that animals absorb and metabolize
is the sum of propionate infused and propionate produced in the rumen from the diet.
Allen (1997) reported that the amount of ruminally fermented organic matter and total
VFA production in the literature ranged from 5.7 to 15.4 kg/d and 42 to 115 moles/d for
lactating dairy cows, respectively. Propionate concentration in the rumen increases from
15 to 45% of total fermentation acids as amount of ruminally fermented OM increases
(Davis, 1967), therefore propionate production is expected to range from 6.3 to 52

moles/d for lactating dairy cows.

Inconsistent hypophagic effects of propionate infusions might be explained by a
concept of threshold response. A threshold for infused propionate to decrease DMI might
be lower for animals fed more fermentable diets. A threshold for total propionate (sum of
infusion and production in the rumen) to decrease DMI might be lower for animals fed
more filling diets (Mbanya et al., 1993) or high fat diets (Farningham et al., 1983). In
addition, energy requirement or physiological status of animals might also affect the
threshold. If propionate regulates maximum DMI, as evidence in the literature suggests,
it is important to know how propionate affects DMI and if a threshold for propionate

exists.

11

Although hypophagic effects of propionate have been investigated extensively,
more research is still needed. Evaluation of feeding behavior would help to understand
the regulation mechanism of DMI by propionate because DMI is a function of both meal
size and intermeal interval, which are determined by satiety and hunger, respectively.
Most experiments in the literature that have evaluated the hypophagic effect of
propionate monitored feed intake over very short periods ranging from 30 minutes to 3
hours, and essentially investigated the effect of propionate on meal size only. The
interaction of hypophagic effect of propionate with other diet characteristics is another
concern because physical fill or dietary fat supplementation sometimes dominate in
regulation of feed intake (Allen, 2000). The majority of previous experiments focused on
the effect of propionate on DMI, not on energy intake. Energy intake should be the
response variable because maximization of energy intake is a major concern for practical
nutritional management of dairy cows with high milk yield, and increasing DMI is an

approach to maximize energy intake.

Mechanisms for Hypophagic Effect of Propionate

Hypophagic effects of propionate might be due to its negative effects on motility
of the reticulo-rumen. McLeay and Pass (1966) showed that infusion of propionate and
butyrate reduced the frequency and amplitude of reticulum and ruminal contractions
compared to iso—osmotic infusion of acetate or N aCl. Mbanya et a1. (1993) showed
hypophagic effects of propionate infusion only for cows challenged by physical fill, but
not for cows infused with propionate only. Thus, greater hypophagic effects of

pr0pionate might be because reduced motility increased physical fill in the rumen.

12

Hypophagic effects of propionate might be explained partially by greater insulin
secretion. Increased serum glucose level and subsequent insulin secretion are thought to
be the major satiety signals for non-ruminants. Infusion of insulin antibodies into the
portal vein of rats increased meal size by 24 to 29%, indicating a role of insulin in feed
intake regulation (Surina-Baumgartner et al., 1995). Similarly, insulin may regulate feed
intake for ruminants because it is reported that insulin infusion decreased feed intake in
sheep (Deetz et al., 1980; Deetz and Wangsness, 1981; Foster et al., 1991). There is
much research that shows that propionate stimulates insulin secretion in ruminants.
Infusion of propionate increased insulin secretion in sheep (Peters et al., 1983) and cows
(Istasse et al., 1987). DeJong (1982) showed that infusion of propionate, n-butyrate, and
n-valerate induced insulin secretion, while acetate is not an effective stimuli for insulin
secretion. Intraruminal infusion of propionate increased insulin secretion while infusion
of acetate did not (Gonda et al., 1997). Elevation of plasma insulin concentration can be
observed under practical feeding situations. Lactating dairy cows responded to a high
starch diet by elevating plasma insulin concentration (Lee et al., 1990), and insulin
concentration increased immediately after feeding cows a high concentrate diet (Vasilatos
and Wangsness, 1980). Infusion of propionate into the portal vein had greater
hypophagic effects compared to the same rate of infusion into the jugular vein (Anil and
Forbes, 1980). This might support the idea that hypophagic effects of propionate are
mediated by insulin secretion because propionate concentrations at the pancreas might be
higher and stimulate more insulin secretion when infused in the portal vein compared to

when infused in the jugular vein (Allen, 2000).

13

It is the liver that senses propionate either directly or indirectly and generates
satiety signals (Anil and Forbes, 1980; Anil and Forbes, 1988). Allen (2000) proposed
that enhanced oxidative metabolism in the liver mediates hypophagic effects of
propionate for ruminants. Although the primary metabolic pathway for propionate is
gluconeogenesis in the liver (Figure l), propionate has two pathways to stimulate
oxidative metabolism in the liver. Propionate is metabolized to oxaloacetate and drives
the TCA cycle in mitochondria stimulating complete oxidation of acetyl CoA (Figure 2).
However, propionate itself is not oxidized in this metabolic pathway. Alternatively,
propionate is metabolized to malate, shuttled out to the cytosol, and metabolized further
to phosphoenol pyruvate. Phosphoenol pyruvate can be metabolized to acetyl CoA,
which can be oxidized completely to carbon dioxide in the TCA cycle (Figure 3).

Propionate oxidized in this pathway would not be utilized for gluconeogenesis.

Role of hepatic oxidative metabolism in feed intake regulation has been
extensively studied for non-ruminants. Langhans et al. (1983, 1984, 1985a, 1985b,
1987a) showed that metabolic fuels that are not utilized by the liver do not have
hypophagic effects while metabolic fuels that are extensively metabolized in the liver
have hypophagic effects. Compared to non-ruminants, ruminant liver has unique
metabolic characteristics. Glucokinase is absent in the ruminant liver (Ballard, 1965) and
hepatic removal of glucose is negligible (Stangassinger and Giesecke, 1986). Similarly,

ruminant liver does not utilize acetate (Reynolds, 1995) because of a lack of acetyl CoA

l4

Glucose

t

f
f

Phosphoenol
pyruvate

Oxaloacetate 5

malate <-:— malate 1‘
‘ Succinyl 00A

4

Propionate

(Cytosol) (Mitochondria)

Figure 1. Propionate is utilized for gluconeogenesis.

15

Acetyl CoA

) Citrate
Oxaloacetate \

m alate TCA cycle (

K I

“Succinyl COA" I It

(Mitochondria) 4 2%:
Propionate

Figure 2. Propionate stimulates oxidative metabolism by driving the TCA cycle.

16

Phosphoenol -> Pyruvate -I-> Acetyl CoA

pyruvate
f a """"""""""""""" )» cmate
5 Oxaloacetate \

Oxaloacetate 5

malate 4__ malate TCA cycle ‘

i ’

‘ Succinyl CoA" K ll

(Cytosol) §(Mltochondrla) 4 2?;

Propionate

 

Figure 3. Propionate is oxidized completely to carbon dioxide after entering the TCA

cycle as acetyl CoA.

l7

synthetase that is required to activate acetate for further metabolism (Ricks and Cook,
1981). Butyrate is another fermentation acid produced in the rumen, but it is extensively
metabolized to B—hydroxybutyrate in the ruminal epithelial tissues, and little is
metabolized in the liver (Demigne et al., 1986). Contrarily, non-esterified fatty acids
(NEFA) are the metabolic fuels extensively oxidized in the liver (Emery et al., 1992).
Ruminant hepatocytes have high activity of propionyl CoA synthetase (Demigne et al.,
1986) and are capable of utilizing propionate extensively. Hypophagic effects of
propionate and fatty acids compared to other metabolic fuels such as glucose, acetate, and
butyrate have been documented extensively for ruminants and have been reviewed
recently (Allen, 2000). The relationship between metabolic fuels and their extent of
utilization by the liver might indicate that propionate has hypophagic effects by

stimulating oxidative metabolism in the liver (Allen, 2000).

Although most experiments in the literature suggest that glucose does not have a
hypophagic effect (Allen, 2000), some reported that feed intake is decreased by
postruminal infusion of glucose (Dhiman et al., 1993; Knowlton et al., 1998b). In the
experiment of Dhiman et a1. (1993), abomasal infusion of propionate (0.75 kg) and
glucose (1.0 kg) decreased DMI for cows fed 98.2% alfalfa silage diet although glucose
infusion did not decrease DMI when 1.2 kg of soy protein was infused with glucose.
Knowlton et al. (1998b) reported that infusion of starch (1.5kg/d) into the rumen or
abomasum decreased DMI to a similar extent. These inconsistent responses to glucose

infusion or post-ruminal starch infusion have not been explained. However, glucose

18

absorbed in the small intestine is metabolized extensively to lactate by gut tissues. These

inconsistent observations might be related to the extent of lactate metabolism in the liver.

Hypophagic effects of lactate have not been consistent for ruminants. Infusion of
lactate into the jugular or mesenteric veins did not decrease feed intake (Baile and
Forbes, 1974). Although Baile and Mayer ( 1969) reported that intra-ruminal infusion of
lactate decreased feed intake, this might be because of metabolism of lactate to VFAs in
the rumen. Lactate concentration in the portal vein can be high as glucose absorption in
the small intestine increases because gut tissues metabolize glucose to lactate. Allen
(2000) suggested that inconsistent hypophagic effects of lactate might depend on the
extent of utilization in the liver because carbon balance and redox state of the liver affect
extent of lactate extraction by the liver (Reynolds, 1995). Effects of lactate on feed

intake regulation need to be investigated further.

Friedman and his colleagues conducted a series of experiment (Tordoff et al.,
1988; Rawson and Frideman, 1994; Rawson et al., 1994a; Rawson et al., 1994b; Koch et
al., 1998), and proposed that hepatic ATP concentration regulates feeding behavior.
Dose of 2, S-anhydro-D-mannitol increases food intake in rats (Tordoff et al., 1988) by
decreasing ATP concentration in the liver (Rawson et al., 1994a). There are two possible
mechanisms for 2, 5-anhydro-D-mannitol to decrease ATP concentration in the liver:
trapping phosphate that would be available for ATP synthesis and decreasing glucose
utilization. Similarly, ethionine decreases hepatic ATP concentration by trapping

adenosine and increases feed intake in rats (Rawson et al., 1994b). Rawson and

19

Friedman (1994) reported that infusion of Na phosphate maintained ATP concentration in
the liver and eliminated the hyperphagic effects of 2, 5-anhydro-D-mannitol. Recently,
Koch et al. (1998) showed that temporal relationships exist between feed intake and

hepatic ATP concentration.

Hepatic ATP concentration is determined by integrated hepatic metabolism.
Friedman and Tordoff (1986) reported that intraperitoneal injection of 2-deoxyglucose
(inhibitor of glucose utilization) or methyl palmoxirate (inhibitor of fatty acid oxidation)
alone did not decrease feed intake significantly, but combined dose decreased feed intake
synergistically. Friedman et al. (1999) evaluated the dose response effect of methyl
palmoxirate, and reported that feed intake in rats increased only for the doses of 5 ppm
and 10 ppm but not for the dose of 1 ppm. The ATP concentration in the liver was not
affected by a lppm close but decreased significantly at 10 ppm. The liver might be able
to maintain hepatic energy status by oxidizing glycogen when challenged with 1 ppm of
methyl palmoxirate because that level of methyl palmoxirate selectively decreased fatty
acid oxidation only in the liver. However, the higher doses of methyl palmoxirate
decreased fatty acid oxidation in the muscle as well as the liver, and increased glucose
demand by peripheral tissues. The rats might not have continued to oxidize glycogen in
the liver because liver glycogen reserves had been depleted. These observations suggest
that inhibition of one type of metabolic fuel is sometimes not sufficient to decrease
hepatic ATP concentration because the liver can utilize other metabolic fuels to some

extent without affecting energy status of the liver.

20

Although extent of oxidative metabolism and energy status in the liver affect feed
intake, it is still not known whether hepatic ATP concentration represents hepatic energy
status. Rawson et al. (1996) reported that 2, 5-anhydro-D-mannitol increased feed intake
for rats fed a high-carbohydrate diet, but not for rats fed a high-fat diet. Rawson et al.
(1996) speculated that the type of diet affects the dominant metabolic fuel in the liver.
Lack of hyperphagic effect of 2, 5-anhydro-D-mannitol for a high fat diet might indicate
that hepatic oxidative metabolism was not altered by inhibition of glucose utilization in
the liver for rats fed a high fat diet. However, the hyperphagic effect of 2, S-anhydro-D-
mannitol was previously attributed to decreased hepatic ATP concentration by trapping
phosphate (Rawson et al., 1994a). If 2, 5-anhydro-D-mannitol increased feed intake by
directly decreasing ATP concentration in the liver via trapping phosphate, it must have
exerted similar hyperphagic effects for a high fat diet in the experiments of Rawson et al.
(1996). Lack of hyperphagic effect of 2, 5-anhydro-D-mannitol for a high fat diet might
indicate that hepatic ATP concentration is not the sole signal that the CNS senses hepatic
energy status by. Reducing equivalent concentration or redox-state in the liver might
send additional satiety signals to the CNS (Langhans et al., 1985a). In a recent
experiment (Ji et al., 2000), 2, 5-anhydro-D-mannitol decreased ATP concentration in the
liver, but did not increase feed intake. They reported that the combined dose of 2, 5-
anhydro-D-mannitol and methyl palmoxirate increased feed intake by inhibition of both
glucose utilization and fatty acid oxidation (Ji et al., 2000). Ji et al. (2000) suggested that
ATP concentration in the liver might not necessarily reflect the availability or turnover
rate of ATP. They proposed the phosphorylation potential calculated as ATP/(ADP x Pi)

is more sensitive index of dynamic cellular energy status.

21

The satiety centers in the hypothalamus may sense signals originating in the liver
via hepatic vagal afferents because hypophagic effect of glycerol, 3-hydroxybutyrate,
malate, lactate and pyruvate were eliminated by selective hepatic vagotomy (Langhans et
al., 1985c). Niijima (1983) reported that discharge rate of hepatic vagal afferents were
reduced by glucose infusion in a dose-dependent manner. Langhans et al. (1985a)
proposed that oxidative metabolism in the liver affects feed intake by hyperpolarizing
membrane potentials of hepatocytes. Langhans and Scharrer (1987b) also found that
ouabain, an inhibitor of the sodium pump, increased feed intake and suggested that
oxidative metabolism within hepatocytes links to sodium pump activity and hepatocyte

membrane potential.

Hyperpolarization of hepatocyte membrane caused by enhanced oxidative
metabolism decreases discharge rate of hepatic vagus possibly because vagal afferents
are in a close association with hepatocytes and some portion of liver parenchyma is
innervated (Langhans, 1996). As described previously, hepatocyte membrane can be
hyperpolarized by enhanced activity of sodium pump that is driven by ATP, but an
alternative mechanism may exist. Langhans (1996) proposed that volumetrically
controlled potassium channels might hyperpolarize hepatocyte membrane by efflux of
potassium. Extraction of metabolic fuels by the liver may increase size of hepatocytes by
drawing water by osmotic pressure, which is independent of oxidative metabolism in the
liver. Hepatic vagus nerve has therrnosensitive fibers and may sense the ATP

concentration in the liver by the rise in liver temperature because oxidative

22

 

phosphorylation and oxidation of metabolic fuels are thermogenic (Langhans, 1996). In
addition, vagal afferents may have mechanoreceptor to sense the change in size of

hepatocytes.

Hepatocytes might release some neuromodulator to vagal afferent terminals. The
ATP synthesized in hepatocytes may be secreted into interstitial fluid (Schlosser et al.,
1996) and send satiety signals directly to the vagal afferents as a neurotransmitter.
Alternatively, the N-methyl-D-aspartate receptor might be involved in transduction of
satiety signals originating from the liver because blockage of this receptor by MK-801,
non-competitive antagonists for N-methyl-D-aspartate receptor, increased meal size of
rats (Burns and Ritter, 1998; Treece et al., 1998; Qian et al., 2000). This receptor exits in
the caudal nucleus of the solitary tract, where vagal sensory fibers terminate (Treece et
al., 1998), and peripherally on hepatic vagal afferents (Qian et al., 2000). However, it is
not known whether the brain senses extent of oxidative metabolism in the liver by this
neural pathway. Mechanisms how hepatocytes communicate with hepatic vagal afferents

remain to be investigated (Langhans, 1996).

23

Summary

Feeding highly fermentable grain sometimes decreases voluntary feed intake in
lactating dairy cows. Excess propionate production in the rumen is hypophagic and the
mechanism might be related to the extent to which propionate stimulates oxidative
metabolism in the liver. Research in the literature suggests that enhanced hepatic
oxidative metabolism decreases feed intake by sending signals to the central nervous
system via hepatic vagal afferents and that hepatic ATP concentration might be an
indicator of hepatic oxidative metabolism. A temporal relationship between hepatic ATP
variation and feeding behavior was observed in rats, but effects of propionate on hepatic
ATP concentration and effects of hepatic oxidative metabolism on discharge rate of vagal
afferents need to be investigated for ruminants. In lactating cows, propionate and NEFA
have greater hypophagic effects compared to other metabolic fuels such as glucose and
acetate. The distinction between the potent hypophagic metabolic fuels and less
hypophagic metabolic fuels is whether or not they are metabolized in the liver.
Propionate and NEFA are extensively metabolized in the liver while glucose and acetate
are not metabolized in ruminant liver. Therefore, extent of oxidative metabolism in the
liver likely plays an important role in regulation of feeding behavior in lactating dairy

COWS.

Hypophagic effects of propionate are documented extensively in the literature.
However, it is not known how propionate exerts hypophagic effects and what factors may
modulate the hypophagic effects of propionate. Feeding more fermentable grains in diets

results in a variable response in feed intake. In addition, the extent of reduction in feed

24

intake by infusion of propionate is variable in the literature. It is speculated that a
threshold exists for propionate to affect feed intake. The threshold response for the effect
of propionate on satiety might be related to the balance among propionate flux to the liver
and rate of propionate utilization for gluconeogenesis by the liver. If a greater amount of
propionate is utilized for gluconeogenesis, less is oxidized in the liver and its hypophagic
effects might be reduced. The flux of other metabolic fuels to the liver might also affect
hypophagic effects of propionate because energy status of the liver might be maintained
by utilizing other dominant metabolic fuels regardless of propionate flux. It is important
to understand the mechanisms for regulation of feed intake in lactating cows fed highly
fermentable diets. Maximizing energy intake in lactating dairy cows improves
productivity of animals and profitability of dairy operations. The role of propionate

metabolism in feeding behavior warrants further investigation.

25

CHAPTER 2

Effects of corn grain conservation method on feeding behavior and

productivity of lactating dairy cows at two dietary starch concentrations

ABSTRACT

Effects of conservation method of corn grain and dietary starch concentration on
DMI and productivity of lactating dairy cows were evaluated. Eight ruminally and
duodenally cannulated Holstein cows (55 i 15.9 days in milk; mean i SD) were used in a
duplicated 4 x 4 Latin square design with a 2 x 2 factorial arrangement of treatments.
Experimental diets contained either ground high moisture corn (HM) or dry ground corn
(DG) at two dietary starch concentrations (32 vs. 21%). Mean particle size and DM
concentration of corn grain were 1863 um and 63.2 %, and 885 um and 89.7%, for HM
and DG, respectively. Dry matter intake was lower for HM compared to DG treatment in
high starch diets (20.8 vs. 22.5 kg/d), but similar for the HM and DG treatments in low
starch diets (19.7 vs. 19.6 kg/d). This reduction in dry matter intake is attributed to
smaller meal size for HM compared to DG in high starch diets (1.9 vs. 2.3 kg of DM for
high starch diets; 2.1 vs. 2.0 kg of DM for low starch diets). Faster starch fermentation
for HM in high starch diets might result in satiety sooner. Milk yield was greater when
cows were fed high starch diets compared to low starch diets (38.6 vs. 33.9 kg/d)

regardless of corn grain treatment. Solids-corrected milk yield was decreased by HM

26

compared to DG in high starch diets (33.2 vs. 35.2 kg/d), but not in low starch diets (32.5
vs. 31.9 kg/d). This is because of lower milk fat and protein concentration for cows fed
HM in high starch diets. Reducing ruminal starch fermentation by substituting DG for
HM can increase productivity of lactating cows fed high starch diets.

(Key words: conservation method of corn, intake, feeding behavior, nutrient utilization)
Abbreviation Key: HM = High moisture corn; DG = Dry ground corn; TRDOM = True

ruminally degraded organic matter; NEFA = Non-Esterified Fatty Acids

INTRODUCTION

An important management goal in dairy nutrition is the maximization of energy
intake. Cows in early lactation often experience negative energy balance, and energy
status greatly affects peak milk yield and persistency of milk production. One approach
to increase energy intake is to increase energy density of diets by feeding more
fermentable grains in diets. Theurer et al. (1999) summarized studies comparing steam-
flaked corn with steam-rolled corn and steam-flaked sorghum with dry-rolled sorghum,
and concluded that steam-flake processing increases milk yield without decreasing DMI.
They attributed the positive production responses to greater starch digestibility in the
rumen and enhanced microbial protein flow to the duodenum. However, greater starch
digestibility results in variable responses in animal productivity. A recent review (Allen,
2000) showed that greater starch digestibility in the rumen is sometimes associated with

sharp reductions in DMI which can decrease energy intake. Inconsistent responses to

27

increased ruminally degraded starch might be because of differences in fermentability of
basal diets and energy requirements of animals across the studies. However, the
interaction between fermentability of grain and dietary starch concentration on

productivity of lactating dairy cows has not been studied extensively.

Corn grain is the major source of dietary starch for lactating dairy cows in the
United States. Ruminal starch digestibility of corn grain can be altered by fineness of
grinding and by conservation method. Ying et al. (2002) reported that ruminal starch
digestibility was reduced over 23% when dry ground corn (DG; mean particle size of 0.8
mm) was substituted for ground high moisture corn (HM; mean particle size of 2.0 mm)
with no difference in total tract starch digestibility. We hypothesized that effects of
starch digestibility of corn grain on productivity of dairy cows differ by concentration of
starch in the diet. Greater ruminally degraded starch from HM is expected to increase
productivity of lactating dairy cows compared to DG when cows are fed low starch diets,
but decrease DMI and productivity when cows are fed high starch diets because of excess
starch fermentation in the rumen. The objective of this experiment was to evaluate
effects of high moisture and dry conservation methods of corn grain on DMI and

productivity for lactating dairy cows fed two levels of dietary starch concentrations.

28

MATERIALS AND METHODS

Treatments and Cows

Experimental procedures were approved by the All University Committee on
Animal Use and Care at Michigan State University. Eight multiparous Holstein cows (55
:t 15.9 DIM; mean i SD) from the Michigan State University Dairy Cattle Teaching and
Research Center were assigned randomly to duplicated 4 x 4 Latin squares balanced for
carry over effects with a 2 x 2 factorial arrangement of treatments. Cows were
cannulated ruminally and duodenally prior to calving, and assigned randomly to
treatment sequence within a square. Treatments were dietary starch concentration (21%
vs. 32%) and conservation method of corn grain (HM vs. DG). Treatment periods were

21 d with the final 10 (1 used to collect samples and data.

One corn hybrid (Pioneer 3730) was grown in 1998, and half of the field was
harvested as HM at a DM concentration of 63.2%. High moisture corn was ground to a
mean particle size of 1863 um and ensiled in a 2.4 m x 9.0 m silage bag (Ag Bagger®,
Ag Bag Corp., Blair, NE). The remaining half of the field was harvested as dry corn.
Dry corn was finely ground to a mean particle size of 885 um. Nutrient composition for
corn grain treatments is shown in Table 1. Experimental diets contained either HM or
DG, corn silage (50% of forage DM), alfalfa silage (50% of forage DM), a premix of
protein supplements (soybean meal, distillers grains, and blood meal), and a premix of
minerals and vitamins (Table 2). All diets were formulated for 18% dietary CP

concentration, and fed as total mixed rations.

29

TABLE 1. Nutrient composition of corn grains used to formulate experimental diets.

 

 

HMl DG2

DM%3 63.2 89.7
~% of DM-

Starch 67.0 70.5
NDF3 7.1 10.3
ADF 2.4 3.0
Crude Protein 9.1 9.1
Ether extract4 6.3 7.3
Ash4 1.6 1.3
Lactate 2.4
Ethanol 0.2
VFA 0.8
Organic acids 3.9
Mean particle size (pm)3 1863 885

 

‘ HM: High moisture corn

2 DG: Dry ground corn

3 Differ significantly between high moisture corn and dry ground corn (P < 0.0001).
4 Differ significantly between high moisture corn and dry ground corn (P < 0.05).

30

8:880:88 8888 oh? 8 8:2 :88 888:8 ”"52 £86885 0

.m 888:» @886 :8 .D 888:, 083.: .< 888:, Wound .8808 8088 008: 08nd .0208 8:60:88 WW: £8 $0.:
0888:: 88088: :81: 0:888: WW: 0885803 88:8 9892 .88 88% b: abodm 8.888 88 8088 0: 8885 n
.808 :83 §md :8 88w 80::8: 0.8.3 .808 80980. @mdh 8:888 88 808.5,.

88 :::o:w ED ”on m

88 08:88 awe ”Em N

.Amod v a: 3802.388 80.8%: 80:2 8808080 80:05:: .3 8:628 Be: 088 8 8:60:88 805:: :8 88$:

 

 

 

32% .9220 :o 8
a mdm scam . Woo a Wm: 8:80 88m :80
a #3 . Wmm 3 W3 a W2 ”52 88cm
0 WV 0 W: 0 Wm a Wm 8880 :08m
WE Wfi 0.x: WM: :0
a S: 0 W3 9 NA fl .0 WA: van—Z 038885
. mm . mm .3 . 8 8w:
. Won 0 Wow 14.2 5 WD 82
_. Wom _. 8.0m s WVN 1 .mm ”52
smgm sogm .mdm 02m 88%
s Wm: .. Wm: .. Wmm Lima 20
anv ”Wm: «Wmm ._Wwv ED
8:888:00 80.5: Z
Wm Em Tm Wm 88 8:088 0: 8885
W3 W2 W3 v.3 0.88 808E
30m oém Wmm Nam 0mm? 8.283.
odm WE Wow Wow own—mm 800
q: 8m 58 2388 :08
W3 W; 88 888 ED
08080.85 :05
On 2: «On— .35
8:80 Bod 8:80 :wE

 

_ .CZD .982: mo :0er 80:: 8888090 mo 828888 808:: :8 0808085 .N max—mash

31

Data and Sample Collection

Throughout the experiment, cows were housed in tie-stalls, and fed once daily
(1400 h) at 110% of expected intake. The amount of feed offered and orts were weighed
for each cow daily during the collection period. Samples of all dietary ingredients (0.5
kg) and orts (12.5%) were collected daily and composited into one sample per cow per
period. Cows were milked twice daily in their stalls during the feeding behavior
monitoring period ((1 16 to d 19), and in a milking parlor the rest of period. Milk yield
was measured daily during the collection period and was averaged over the collection
period. Milk was sampled at every milking on d 12, 16, and 19 of each period and
analyzed for fat, crude protein, and lactose with infrared spectroscopy by Michigan
DHIA (East Lansing). Body weight was measured on two consecutive days immediately
prior to the start of the first period, and on d 19 and d 21 of each period. Empty body
weight was measured after evacuation of ruminal digesta. The BCS was determined
[(Wildman, 1982); five-point scale where 1 = thin to 5 = fat] by three trained
investigators blinded to treatments immediately prior to the start of the first period and on

d 21 of each period.

Feeding behavior and ruminal pH were monitored from d 16 through d 19 (96 h)
of each period by a computerized data acquisition system (Dado and Allen, 1993). Data
of chewing activities, feed disappearance, water consumption, and ruminal pH were
recorded for each cow every 5 sec. Daily means were calculated for number of meal
bouts per day, interval between meals, meal size, eating time, ruminating time, and total

chewing time. These response variables were calculated as daily means, then averaged

32

over the four days for each period. Blood samples and ruminal fluid samples were
collected every 20 min for 24 h by automated sample collection system (Allen et al.,
2000), starting at 1200 h on d 16. Blood was sampled from a jugular vein through a

catheter inserted 1 d prior to sample collection.

Ruminal fluid was centrifuged at 2,000 x g for 15 min immediately after
collection, and supematants were frozen at ~20° C until analysis. Blood samples were
collected into two tubes: one with lithium heparin and the other with potassium oxalate
and sodium fluoride as a glycolytic inhibitor. Both were centrifuged at 2,000 x g for 15
min immediately after sample collection, and plasma was harvested and frozen at -20° C

until analysis.

Sample and Statistical Analysis

Diet ingredients and orts were dried in a 55° C forced-air oven for 72 h and
analyzed for DM concentration. All samples were ground with a Wiley mill (1mm
screen; Authur H. Thomas, Philadelphia, PA). Samples were analyzed for ash, NDF,
ADF, lignin, indigestible NDF, CP, and starch. Ash concentration was determined after 5
h oxidation at 500° C in a muffle furnace. Concentrations of NDF and ADF were
determined [(VanSoest et al., 1991); method A for NDF]. Crude protein was analyzed
according to Hach et al. (1985). Starch was measured by an enzymatic method
(Karkalas, 1985) after samples were gelatinized with sodium hydroxide, and glucose
concentration was measured using a glucose oxidase (Glucose kit #510; Sigma Chemical

Co., St. Louis, MO) and absorbance was determined with micro-plate reader

33

(SpectraMax 190, Molecular Devices Corp., Sunnyvale, CA). Indigestible NDF was
estimated as NDF residue after 120-h in vitro fermentation (Goering and VanSoest,
1990). Concentrations of all nutrients except for DM were expressed as percentages of
DM determined by drying at 105° C in a forced-air oven. Corn grain was dry sieved
(Sieve apertures: 4750, 2360, 1180, 600, 300, 150, 75 um and bottom pan), using a sieve
shaker (Model RX-86, W.S. Tyler Inc., Gastonia, NC) for approximately 20 min until the
bottom pan weight was constant, and mean particle size of corn grain was calculated

(ASAE, 1968). The TRDOM was calculated as described by Oba and Allen (2002a).

Ruminal fluid was analyzed for VFA and lactate concentrations. Samples were
centrifuged at 26,000 x g for 15 min, and supernatant (600 [1].) was mixed with 600 pl.
Ca(OH)2 and 300 uL of CuSO4 containing crotonic acid as an internal marker in 1.7 ml
micro centrifuge tubes. Samples were centrifuged at 12,000 x g for 10 min, and
supernatant (1000 ul) was taken and mixed with 28 pl of H2804 in 1.5 ml micro
centrifuge tubes. Samples were frozen and thawed twice, and centrifuged at 12,000 x g
for 10 min. Supernatant was transferred to HPLC vials. Concentrations of VFA and
lactate of the supernatant were determined by HPLC as described by Dado and Allen
(1995). Rate of VFA production (moles/d) was estimated from the measured true
ruminally degraded organic matter (Oba and Allen, 2002a) and microbial efficiency (Oba

and Allen, 2002b) according to Allen (1997).

Plasma samples were analyzed for concentrations of acetate, glucose, non-

esterified fatty acid (NEFA), insulin, and growth hormone. Plasma was processed to

34

quantify acetate concentration as described for ruminal fluid. Due to greater protein
concentration for plasma samples, the first stage of sample processing was duplicated to
obtain enough supernatant ( 1000 111) to be mixed with 28 ul of H280, in 1.5 ml micro
centrifuge tubes. Plasma growth hormone concentration was determined by
radioimmunoassay (Gaynor et al., 1995). Commercial kits were used to determine
plasma concentration of glucose (Glucose kit #510; Sigma Chemical Co., St. Louis, MO),
NEFA (NEFA C-kit; Wako Chemicals USA, Richmond, VA), and insulin (Coat-A-
Count, Diagnostic Products Corporation, Los Angeles, CA). Frequency and amplitude of

insulin peaks were quantified according to Merriam and Wachter (1982).

All data were analyzed using the fit model procedure of JMP® according to the
following model:

Yijkl = u + Ci + Pj + Tk + emu
where

u = overall mean,

Ci = random effect of cow ( j = 1 to 8),

Pj = fixed effect of period (k = 1 to 4),

Tk = fixed effect of treatment (1 = 1 to 4),

eijld = residual, assumed to be normally distributed.

Period x treatment interaction was originally evaluated, but it was removed from
the statistical model because interaction was not significant for response variables of

primary interest. Orthogonal contrasts were made for the effect of dietary starch

35

concentration, conservation method of corn grain, and interaction of dietary starch
concentration and conservation method. Treatment effects and their interaction were
declared significant at P < 0.05 and P < 0.10, respectively, and tendency for treatment

effects and their interaction were declared at P < 0.10 and P < 0.15, respectively.

RESULTS AND DISCUSSION

DMI and Ruminal Fermentation

Cows fed high starch diets had greater DMI compared to cows fed low starch
diets (P < 0.001; Table 3), and this might be attributed to greater physical fill in the
rumen for low starch diets. Low starch diets contained more forage NDF compared to
high starch diets (25.3 vs. 16.5 % DM), and forage NDF is a primary factor reducing
DMI by physical fill in the rumen (Allen, 2000). Interaction of dietary starch
concentration and conservation method of corn grain was significant for DMI (P < 0.07).
The DG treatment increased DMI by 1.7 kg (22.5 vs. 20.8 kg/d) compared to HM
treatment when fed in high starch diets but had no effect (19.6 vs. 19.7 kg/d) when fed in
low starch diets. Consistent with treatment effects on DMI, meal size was greater for DG
treatment compared to HM treatment (2.3 vs. 1.9 kg) when cows were fed high starch

diets, but not when cows were fed low starch diets (2.0 vs. 2.1 kg).

The number of meal bouts per day was greater for high starch diets compared to

low starch diets (P < 0.04). The HM treatment tended to increase the number of meal

36

.Gmduw ECO .«O @0508 GOflNEOwfiOU Una «Quay—Co £0.85 EHOMU .«O QOUUNHOHGH “Fan

58w 58 do @0508 couatomcoo do 80am ”Boos.

E880 58m; .93on do Hoobm has?

58 28% ED NOON
88 05368 emf ”2:.

 

 

 

 

 

S d 2d 8d V SS dm d we mmmfi m mmovv $de 95 $65 30,—.
mad mud 3d v omE ddowm Ndem 8ch dmmmm Add 9528 wadacgsm
mmd mmd 8d v .35 N32 3de 033 :vE Gd 36:0 madam
A335
mmd Ed Edd v «.8 mach wddh Quad Ndwc 2:: wagono :38.
Ed 5d Sod v 0.2 0wa wade Ndmv fihmv aimed 2:: means—End
wmd Ed Edd v Nd 9th vddm @me ~.mmm 82:5 0:5 madam
odd m d d 5d v mod omd Ed de 3d ”52
mod mod ded v mod med 3d Ed omd 53%
odd Ed mmd dd dd mm md o4 2Q
Amos .5... :32
S .d mmd nmd md m. 2: man: ©de ado GE: 3285 32585
wad omd 5d v a; d.mm mdm :m ddm 3:5 Ewe»: :82
E .d add 3d md md 5d ad v.2 :5 3:09 :32
8d Nd .d Sod v md 0.2 5.2 mdm wdm 8E6: 95
9:5 ”.800 28me mm 0Q 35 «DD 95
and; m :08; 304 £085 nwE

 

.HOTrNSOD wfleoom GO muflvufioo £0.35 EHOE Day as @0508 GOUMEOmflOO :Quw ECU MO EONEW .m mmdmaflh.

37

bouts compared to DG treatment (P < 0.09), and the difference between HM and DG
treatments was numerically greater (interaction P < 0.16) when cows were fed high starch
diets (11.4 vs. 9.9 /d) compared to low starch diets (9.7 vs. 9.5 Id). Although DMI is a
function of meal size and meal frequency, greater DMI for DG treatment in high starch
diets is explained solely by greater meal size because DG treatments did not increase
meal frequency. On the contrary, when high starch diets were fed, cows consuming HM
corn tended to eat more frequently, but were not able to compensate for the decreased

meal size, resulting in an overall depression in DMI.

The reduction in DMI and meal size for HM treatment in high starch diets can be
attributed to greater ruminal fermentation. Because dietary forage NDF concentration
was similar between HM and DG treatment within the same dietary starch concentration,
differences in physical fill were not likely to be responsible for differences in DMI. For
cows fed high starch diets, true ruminally degraded OM (T RDOM) was greater for HM
compared to DG treatment (11.3 vs. 10.3 kg/d; Table 4), and calculated rate of VFA
production was greater for cows fed HM corn compared to DG corn (68.8 vs. 52.0
moles/d). A reduction in DMI with increased ruminal fermentation has been reported
previously. McCarthy et al. (1989) increased starch digestibility in the rumen by
replacing ground shelled corn with steam rolled barley in diets containing more than 40%
starch at DM basis, and reported approximately 3 kg depression in DMI. Overton et al.
(1995) also showed that increased substitution of barley for dry corn linearly decreased

DMI for lactating dairy cows.

38

Lower DMI for HM treatment when cows were fed high starch diets might be
explained by higher propionate production in the rumen. Propionate has greater
hypophagic effect compared to acetate (Farningham and Whyte, 1993; Sheperd and
Combs, 1998). Intra-ruminal infusion of propionate decreased DMI compared to acetate
on both an iso-molar (Farningham and Whyte, 1993) and an iso-energetic basis (Sheperd
and Combs, 1998). Although greater fermentation in the rumen is often associated with
low ruminal pH and greater osmolarity of ruminal fluid, their hypophagic effects have not
been shown. Choi and Allen (1999) showed that ruminal pH per se does not have a
direct effect on DMI and feeding behavior of lactating dairy cows. They also found that
infusion of hyper-osmotic solutions of NaCl decreased meal size, but did not affect DMI

because cows compensated for smaller meal size by increasing meal frequency.

When cows were fed low starch diets, HM treatment did not decrease DMI
compared to DG treatment. Greater starch digestibility in the rumen does not necessarily
decrease DMI (Theurer et al., 1999; Knowlton et al., 1998; Grings et al., 1992; Callison
et al., 2001). Inconsistent effect of ruminal starch digestibility on feed intake might
imply that a threshold exists for propionate to affect DMI. Feeding behavior might be
controlled by a dominant mechanism related to stimulation of tension receptors by
ruminal fill until a mechanism related to propionate begins to dominate on highly

fermentable diets.

High starch diets decreased molar ratio of acetate (P < 0.0001; Table 4), iso-

butyrate (P < 0.001), and iso-valerate (P < 0.03) compared to low starch diets. High

39

€33 :22 ,3 costumed @0808 8838630 2d 8 mad—8863 Smddm 80:4 83 god 582050 35888
83 Aamddm 82?. 83 30V 8838 083.30 388de E3888 28 38.308 8: 88.8 33888 33 Ad\mo_o8v 88 8082608 «d?

8338 083?“. dodfiwod 33888 68H. ”ZOOM?
83% 88 do 8508 8832888 83 8088 893% >386 do 838883 ”H75
83% 88 do @0808 8833388 do 80am ”80nd,.

8088 £635 E386 do 80am 888mm

88 88on E ”DD”
88 28808 cm“: ”SE.

 

 

40

 

 

3d ddd V dd hd dd dd ad in 038:5
odd mdd add 0; md NS v.2 wdd 03:08on
wwd 5d hmd m.m v.8 ddm ddm ddv 888.4.
dwd dd 2 d 5m Name ddm ddm 36 38,—.
AEmo—o8d h8052608 <m>
wad mod ddddd v 2d Nd.m 3d wmd EN 0:8 03.88on 8330<
Ed 2 d ddd v m _ d mo; mm; 3d dad u383>
dmd ddd mdd d_ d dd.m am; we; on; 3803>6£
odd mmd wdd ed 0: find 3: mm“ 3858m—
de dad dddd V dd dm; dag wdg dd; 8885-08
de dwd _ddd v d; NAN 0.3 9mm Sum 880805
3d wed ddddd V dd v.8 WE wfim fiwm 888<
:08 dd_>08v <m>
Ed mud dd d dd dfid mddd m.md_ wddd 98v <m> 38H.
ddd mdd dddd v dd b.» md mdd 0: €th @2059
dmd mmd wdd Edd dhdd dddd HQ d d2 d 8.38; 385
wvd 3d 5d V dd de mad 2 d NH .0 838 >39
mm 3888
“H7: vfioO $635 mm on 2! «DO .2:
83> & £033. 33 £033 swam

 

.ficflwuflogow G085.— GO mufloucoo £0.83 506 03”. um mung—HOE GOUfltUmEOU flaw ECU .«O wuoouwm .V Wimmdah.

starch diets increased molar ratio of propionate (P < 0.001) and valerate (P < 0.01)
concentration compared to low starch diets. Volatile fatty acid profile was not affected
by conservation method of corn grain. Total volatile fatty acid concentration was not
affected by treatments. It is noteworthy that the concentration of total VFA in ruminal
fluid did not reflect the amount of OM truly fermented in the rumen. Although treatment
means for TRDOM varied from 7.7 to 11.3 kg/d in this experiment, no relationship was
observed between TRDOM and total VFA concentration (Figure 1). Total VFA
concentration did not indicate fermentation acid production in the rumen because
treatment effect on rates of absorption and passage compensated for effects of production

rates on VFA concentration.

Daily mean ruminal pH was lower for high starch diets compared to low starch
diets (P < 0.01), but was not affected by corn grain treatment. In this experiment, daily
mean ruminal pH was above 6.1 even for high starch diets regardless of corn grain
treatments, and it was close to the pK, for bicarbonate, the major buffer secreted in saliva.
This indicates that diet fermentability does not negatively affect ruminal pH if buffering
capacity of ruminal digesta exceeds fermentation acid production in the rumen (Allen,
1997). In this experiment, eating time, ruminating time, and total chewing time were
greater for low starch diets compared to high starch diets (P < 0.001). These
observations were consistent with higher ruminal pH for low starch diets because of
greater salivary buffer secretion and less fermentation acid production in the rumen for
low starch diets compared to high starch diets. However, chewing activities were not

affected by com grain treatment. Despite greater fermentation in the rumen, HM

41

140.

130. '
120.
110. o ' .
100. °

00 \O
O o
l l
O
O
O
'a
O
O

 

\l
O

 

5 Ii) 15
True ruminally degraded OM (kg/d)

Total ruminal VFA concentration (mM)

C

Figure 1. Relationship between true ruminally degraded OM (TRDOM) and total

ruminal VFA concentration (P > 0.21, r2 = 0.05).

42

treatment did not decrease ruminal pH possibly because buffering capacity of ruminal
digesta was maintained by sufficient chewing activity and saliva flow. Because ruminal
pH and total VFA concentration were not directly related to fermentation acid production

in the rumen, neither explained hypophagic effects of the HM treatment.

Milk Production and Plasma Metabolites

Milk yield and SCM yield were greater for high starch diets compared to low
starch diets (P < 0.0001 and P < 0.02; Table 5). An interaction of dietary starch
concentration and conservation method of corn grain was detected for SCM yield (P <
0.10). The DG treatment increased SCM yield by 2.0 kg (35.2 vs. 33.2 kg/d) when fed in
high starch diets but decreased SCM by 0.6 kg (31.9 vs. 32.5 kg/d) when fed in low
starch diets. Significant interactions for main effects were also observed for milk fat
concentration (P < 0.06) and milk protein concentration (P < 0.07). The DG treatment
increased milk fat concentration (3.59 vs. 3.05 %) and milk protein concentration (3.02
vs. 2.98 %) compared to HM treatment when cows were fed high starch diets. However,
HM treatment increased milk fat concentration (3.95 vs. 3.73 %) and milk protein
concentration (2.94 vs. 2.87 %) compared with DG treatment when cows were fed low

starch diets.

Cows fed low starch diets benefited more from consuming HM compared to DG

possibly because of greater ruminal fermentation. However, when DG replaced HM in

high starch diets, productivity of lactating cows was enhanced from higher DMI. In

43

830m 800 «0 60608 88320388 83 80800 .633 .9306 do 88680:: F75
83% 8.80 do 60608 88330880 do 80mm— 8.89.

8088 £033 .9806 do 60th 8030mm

8.80 685% in ”OD”

800 08368 swam ”SE.

 

 

 

 

3:0 as :3 v 85 S .o- 8.0- :3 2 .o 630 03:20 300
So 03 :5 v 8.0 3.? 30.0- 35 one 830: 0385 30
«to 8o 53 v 8.0 3+ 3.4 33. 8.4 0862 32
5o Ed :3 v 85 33 3.3 8.3 33 52.8 82
3o 88 :3 v :5 35. 3.3. 3.3 83 8 0:82
8.0 823888 032
3.0 ad 886 v 83 $8 88 3: OS 0882 82
So Ed 58.0 v 85 Rd 3.0 m: E .8 5088 ca:
36 3.0 36 85 Rd 3.8 32 :4 8 82
S .o 9% So no a. d 3 3.33 3.3 «.3 203
35 3.0 NS 3 4.3 83 S3 Em 83.9 200
3d 3; 88¢ V dc 3.3 4.3 $3 3.3 ca:
330: 20;
LE .88 .533 .3 on 2: "on .2:
02.9» nN £0.83 30..— nobfim aw;—

 

5338605 :0 380800 £033 506 9.3 3 60608 8063303800 88m 800 .6 380%de .m mam»;

agreement with our observation, Moore et al. (1992) reported greater milk yield when
starch digestibility was increased by replacing dry-rolled sorghum with steam-flaked
sorghum (flake density of 0.40 kg/L). However, flaked sorghum with a density of 0.27
kg/L (associated with more rapid starch hydrolysis) decreased DMI and milk yield.
Theurer et al. (1999) showed that greater starch digestibility by steam-flake processing of
corn and sorghum grain increased milk yield with similar DMI. Greater starch
digestibility in the rumen might improve milk yield if DMI is maintained and if extra
energy is not directed to body reserves by greater insulin secretion. Experiments in
which a reduction in milk yield was associated with enhanced starch digestibility in the
rumen also reported reduced DMI with greater ruminal starch digestion (MaCarthy et al.,
1989; Overton, 1995). Optimal ruminal starch digestibility is dependent upon starch

concentration and fermentability of diets.

High starch diets decreased milk fat concentration, but increased BW gain (P <
0.01) and body condition score (BCS) gain (P < 0.01) compared to low starch diets.
High starch diets decreased daily mean ruminal pH, but increased daily variance for
ruminal pH. We previously proposed that partitioning of absorbed fuels to milk or body
reserves is influenced by variation in ruminal pH because it determines pattern of supply
of metabolic fuels from the rumen to the blood circulation (Oba and Allen, 2000). Rate
of fermentation acid absorption from the rumen is a function of ruminal pH. Ruminal pH
with less daily fluctuation might result in more consistent supply of metabolic fuels from
the rumen to the blood circulation while a greater fluctuation in ruminal pH might

indicate more pulsatile energy supply. A more pulsatile energy supply may stimulate

45

insulin secretion, increasing energy metabolite utilization in adipose tissues more than
milk fat synthesis (Oba and Allen, 2000). In agreement with this theory, high starch diets
increased insulin concentration (P < 0.01; Table 6). Diurnal pattern for plasma insulin
concentration (Figure 2) shows that high starch diets consistently increased plasma
insulin concentration compared to low starch diets. Greater daily means for plasma
insulin concentration for high starch diets are attributed to a greater baseline of insulin
secretion (P < 0.001) and enhanced amplitude of insulin peaks (P < 0.001) compared to

low starch diets.

A significant interaction of dietary starch concentration and conservation method
of com grain was observed for milk fat concentration and milk fat yield, indicating milk
fat was depressed for HM treatment only for cows fed high starch diets. However, the
reasons for this milk fat depression are not known. Corn grain treatment did not affect
mean plasma insulin concentration although the change in plasma insulin concentration
during a meal was greater for cows fed HM compared to DG (P < 0.03). Corn grain
treatment affected insulin secretion only transiently. Trans-C13,, fatty acids produced in
the rumen can decrease milk fat yield, and increased trans-C13,, fatty acid production in
the rumen was related to low ruminal pH when cows were fed high concentrate diets
(Kalscheur et al., 1997; Kennelly et al., 1999). However, ruminal pH was not different
between HM and DG treatments for cows fed high starch diets. Therefore, the milk fat
depression in this experiment might not be caused by enhanced production of trans—C18,,

fatty acids in the rumen although they were not measured.

46

.Efiw 800 6 60608 0060200000 600 E0600 :08; E0066 6 0060802: F75 ”062w 800
6 60608 0060300000 6 80mm ”:69. 3:00:00 603: @806 6 00005 ”6080mm ”800 60000» ED NOON ”:60 002068 :wE ”2:.

 

 

 

 

mod mod :d Nd dds Wan mdb odb €65 62:02: 060090005

mod odd mud ad ~86 :3 6.: flow 665 6w:0_ 60m

mvd cad Sod v hd in 66 ms fiw 06365:N 60m

5d mad mod ad mdm vdm vdm odm 6: x000 6 302802

Odd Ed Edd V 6d 65 06 dd md 006095

mmd mod 8d wd Wm Wm d.m Wm B08 $666 0w§nu

Ed 3d 5d v v; w.: 5.2 mm _ if R08 6 6:0 2

5d mod Sod v wd md md WE w.m_ R08 6 006063 2

Rd mad 3d 5.5 5.2 6.3 6.8 6.5m 000063 365

Ed de Sod v dd md_ 62 6.2 MS: 308 >609
352$ £35

6d Sd Ed o: mmé we;V EN cm.m 00:06? 365

5d 3d 5d V Hmd Sum mod Odd add .808 36D
GEES 00086: 63000

omd nmd 3d 1mm 83 Ron moi 52 00:3?» >65

mmd Ed 5d 3 5: EN mm _ vm fl 508 >609
@025 mvrimz

owd wmd and vod m _ d w_d 9 d E d 0000:? >65

Ed 66d 3d 3d wad wad wwd 3d 0008 >65
€65 8800.4.

3d Sud cod dd 62 WE WE DE 00563 >65

65d mmd 8d v wd msm odm 0.06 6.6 0008 >65
020.5 3.820

“H7: 4060 «seam mm on 2: «CO .35
0063 m :03: 30,.— 533 awe

 

.0063600000 0:086: 600 0666308 «Emma 00 800600 60003 506 0B“ 3 60608 0060300000 52m 500 6 308$ .6 mamflh

47

25 ‘ —-l—Iigh starch diets

 

Plasma insulin (micro IU/ml)

 

 

5 I I l7

0 4 8 1'2 1'6 2E)
Time relative to feeding (h)

Figure 2. Effect of dietary starch concentration on plasma insulin concentration relative

to feeding time.

48

701 -'-High starch diets

Widow starch diets

 

 
    

I I " ”me iii‘ "iii!

Plasma glucose (mg/d1)

 

 

50 I I I 1 I I
O 4 8 12 16 20

Time relative to feeding (h)

Figure 3. Effect of dietary starch concentration on plasma glucose concentration relative

to feeding time.

49

High starch diets decreased plasma concentrations of growth hormone (P < 0.01)
and NEFA (P < 0.07) compared to low starch diets. Lower plasma acetate concentration
(P < 0.04) for high starch diets might be because of less acetate flux from the rumen and
greater acetate utilization in peripheral tissues from stimulation by insulin compared to
low starch diets. Plasma concentration of glucose was greater for high starch diets
compared to low starch diets (P < 0.01; Figure 3), but it is noteworthy that plasma
glucose concentration decreased after feeding regardless of diet. This reduction in
plasma glucose concentration is partially attributed to an increase in plasma insulin
concentration after feeding. Because insulin decreases gluconeogenesis and increases
glycogen synthesis in the liver, we speculate that absorbed propionate is not directly
metabolized to glucose but transiently utilized for glycogen synthesis after feeding, and

that glucose is released from glycogen storage over time.

50

CONCLUSION

Substitution of DG for HM reduced ruminal fermentation and increased DMI and
SCM yield compared to HM in high starch diets. This is consistent with the theory that
propionate production in the rumen can affect satiety and DMI. Although DG treatment
decreased TRDOM, DG treatment increased productivity of lactating cows fed high
starch diets because of greater DMI. However, cows fed low starch diets increased
productivity when consuming HM compared to DG. Optimal ruminal starch digestibility

is dependent upon the starch concentration and fermentability of diets.

51

CHAPTER 3

Dose-response effects of intra-ruminal infusion of propionate on energy

intake and feeding behavior of lactating dairy cows

ABSTRACT

Dose-response effects of intra-ruminal infusion of propionate on feeding behavior
of lactating dairy cows were evaluated with eight ruminally cannulated Holstein cows
past peak lactation. In experiment 1, treatments were mixtures of propionic acid and
acetic acid infused into the rumen continuously for 14 h at a rate of 16.7 mmol/min.
Treatment solutions contained propionic acid at 8 different concentrations. In experiment
2, treatments were mixtures of sodium propionate and sodium acetate infused into the
rumen continuously for 14 h at a rate of 25 mmol/min. Treatment solutions contained
sodium propionate at 4 different concentrations. Treatment solutions contained acetic
acid and sodium acetate, respectively for experiment 1 and experiment 2, to keep the
osmolarity and pH (but not energy concentration) of infusates constant across the
treatments within an experiment. Experimental diets were formulated to contain 29%
NDF, and dry cracked corn (mean particle size = 3.6 mm) was the major source of starch.
Infusion started 2 h before feeding and ended 12 h after feeding, and feeding behavior
was monitored for 12 h after feeding using a computerized data acquisition system. Total

metabolizable energy (ME) intake was calculated by adding the energy of infusates to

52

dietary energy intake. Experimental designs were an 8 x 8 Latin square for experiment 1
and a duplicated 4 x 4 Latin square for experiment 2. In experiment 1, as infusion rate of
propionate increased, dry matter intake (P < 0.01) and total ME intake (P < 0.05)
decreased linearly. In Experiment 2, as infusion of propionate increased, dry matter
intake (P < 0.0001) and meal size (P = 0.03) decreased linearly, and number of meal
bouts tended to decrease linearly (P = 0.08). Total ME intake also decreased linearly (P
< 0.0001) as proportion of propionate of the VFA infused increased. This indicates that
the reduction in dietary energy intake due to propionate infusion was greater than the
energy supplied from propionate infusions. Our results demonstrate that propionate plays
an important role in feed intake regulation by affecting both satiety and hunger.

(Key words: propionate infusion, feed intake, feeding behavior)

Abbreviation Key: ME = metabolizable energy

INTRODUCTION

Maximizing energy intake is an important goal for nutritional management for
high producing dairy cows. Although feeding more fermentable grains in diets increases
energy density of diets, excess fermentation in the rumen sometimes decreases DMI and
does not necessarily increase energy intake in lactating cows (Allen, 2000). However,
greater ruminal fermentation is more desirable to increase microbial protein production as
well as energy intake unless DMI is decreased. Therefore, it is important to understand

mechanisms that regulate voluntary feed intake when cows are fed high grain diets.

53

Greater ruminal fermentation is characterized by a variety of factors such as low
ruminal pH, increased osmolarity of ruminal fluid, and more fermentation acid
production. Choi and Allen (1999) showed that propionate has greater hypophagic
effects in lactating dairy cows compared to acetate although pH of ruminal fluid per se
did not affect DMI and feeding behavior. In addition they reported that intra-ruminal
infusion of hyper-osmotic solution decreased meal size but did not decrease DMI because
cows increased meal frequency. Hypophagic effects of propionate have been
documented extensively for ruminants (Anil and Forbes, 1980; Elliot et al., 1985;
Farningham and Whyte, 1993; Mbanya et al., 1993; Hurtaud et al., 1993; Wu et al., 1994;

Sheperd and Combs, 1998).

However, some experiments in the literature reported that propionate infusion
did not decrease feed intake (Deetz and Wangsness, 1981; Quigley and Heitmann, 1991;
De Jong et al., 1981; Anil et al., 1993). Inconsistent hypophagic effects of propionate
might be explained by a threshold response of propionate for feed intake regulation.
Infusion of propionate might not affect DMI and feeding behavior unless amount of
propionate exceeds a threshold. This concept agrees with observations that feeding more
fermentable grains does not always decrease DMI in lactating dairy cows (Allen, 2000).
Dose response effects of propionate on feed intake were previously investigated for
lactating dairy cows (Anil et al., 1993) and sheep (Farningham and Whyte, 1993). Feed
intake decreased linearly as infusion rate of propionate increased for experiments of Anil

et al. (1993) and Farningham and Whyte (1993), and a threshold for infused propionate to

54

decrease feed intake was not detected. However, it is difficult to interpret their results
because propionate was infused for only 3 h in those experiments (Anil et al., 1993;
Farningham and Whyte, 1993), so hypophagic effects of propionate were evaluated
essentially on meal size only. Hypophagic effects of propionate should be evaluated by
monitoring feeding behavior for a longer period because cows are able to compensate for
smaller meal size by increasing meal frequency (Choi and Allen, 1999). Evaluation of
feeding behavior would also help to understand the regulatory mechanism of propionate
on DMI because DMI is a function of both meal size and intermeal interval, which are

determined by satiety and hunger, respectively.

Although maximization of energy intake is a primary concern for practical
nutritional management, the majority of previous experiments that studied hypophagic
effects of propionate by infusion focused on effect of propionate on DMI not on energy
intake. Energy intake should be evaluated because animals are supplied extra energy
from infused propionate in those studies. The objectives of this experiment were to
evaluate dose response effects of intra-ruminal infusion of propionate on feeding
behavior and energy intake in lactating dairy cows and to determine if a threshold exists

for effects of propionate infusion on feed intake.

55

MATERIALS AND METHODS

Experimental procedures were approved by All University Committee on Animal

Use and Care at Michigan State University.

Experiment 1

Eight multiparous Holstein cows (113 :1: 26 DIM; mean i SD) cannulated
ruminally for previous experiments were selected from the Michigan State University
Dairy Cattle Teaching and Research Center. Treatments were continuous intra-ruminal
infusion of mixtures of propionic acid and acetic acid at 8 different ratios. Treatment
solutions were prepared by diluting 16.8 moles of VFA (propionic acid and acetic acid at
ratios of 0:7, 1:6, 2:5, 3:4, 4:3, 5:2, 6: 1, and 7:0) to 18 L with de-ionized water. Acetic
acid was added to keep the osmolarity and pH of infusates constant across the treatments
to isolate specific effects of propionate relative to acetate on feeding behavior of dairy
cows. Concentrations of total VFA were 0.93 M across the treatments, and 15 L of each
solution was infused over 14 h. Infusion rate was 17.9 ml/min, which is equivalent to
infusion of 16.7 mmol of VFA/min. The solutions were infused using 4-channel
peristaltic pumps (#78016-30, Cole-Farmer Instrument, IL) and Tygon® tubing (7.5 m x
1.6mm I.D.). Infusion started 2 h before feeding so that treatments could influence
feeding behavior from the first meal immediately after feeding. Treatment periods were

2 d with 14 h of infusion followed by 34 h of recovery.

56

Experimental diets contained dry cracked corn (mean particle size of 3.6 mm),
corn silage, alfalfa silage, a premix of protein supplements (soybean meal, distillers
grains, and blood meal), and a premix of minerals and vitamins (Table 1). Dietary NDF,
CP, and starch concentrations were 29.0, 15.9, and 30.8%, respectively. Dry cracked
corn (mean particle size = 3.6 mm) was the major source of starch to minimize
propionate production from the basal diet. Diet adaptation period was 14 d, and the final
3 d were used for data collection for DMI and milk yield to characterize the cows used in
this experiment. The BW and BCS [(Wildman, 1982); a five-point scale where l = thin
to 5 = fat] were determined on the last day of the diet adaptation period. Means for BW,
BCS, DMI and milk yield were 623 kg, 2.6, 25.4 kg/d, and 36.4 kg/d, respectively. After
14 d of diet adaptation, cows were assigned to an 8 x 8 Latin square balanced for carry

over effects for infusion treatments.

Throughout the experiment, cows were housed in tie-stalls, and fed once daily
(1030 h) at 110% of expected intake. Cows were not allowed access to feed from 0830 h
to 1030 h. The amount of feed offered and 0115 were weighed for each cow daily. On
every infusion day, samples of all dietary ingredients (0.5 kg) were collected, and
treatment solutions were infused from 0830 h to 2230 h. Cows were milked twice daily
in the milking parlor except for the evening milking on infusion day, for which cows
were milked in their stalls. Feeding behavior was monitored from 1030 h to 2230 h on
every infusion day by a computerized data acquisition system (Dado and Allen, 1993).
Data of chewing activities, feed disappearance, and water consumption were recorded for

each cow every 5 sec, and meal bouts, interval between meals, meal size, eating time,

57

TABLE 1. Ingredients and nutrient composition of experimental diets (% of dietary DM
except for DM).

 

Diet Ingredients
Corn silage 27.0
Alfalfa silage 25.4
Dry cracked corn 25.9
Whole linted cottonseed 6.8
Protein mixl 9.9
Vitamin & mineral mix2 5.0

Nutrient Composition

DM 49.5
OM 93.0
Starch 30.8
NDF 29.0
ADF 20.8
CF 15.9
Ether extract 3.8
Forage NDF 21.2
Metabolizable energy (Meal/kg) 3 2.72

 

‘ Protein mix contained 75% soybean meal, 20% distillers grain, and 5% blood meal.

2 Vitamin & mineral mix contains 66.4% dry ground corn, 20.4% dicalcium phosphate,
7.8% salt, 2.4% magnesium oxide, 1.9% trace mineral premix, 0.34% vitamin A, 0.29%
vitamin D, and 0.08% vitamin E.

3 Metabolizable energy was calculated from book values according to NRC (1989)

58

ruminating time, and total chewing time were calculated. Total metabolizable energy
(ME) intake was calculated by adding the ME from infusates to the ME from the diet.
The experimental diet was assumed to contain 2.72 Mcal/kg of ME based on book values
from NRC (1989). Intake of ME was evaluated instead of intake of N EL because the
efficiency of energy conversion from ME to NEL for infused acetate and propionate is not
known and it could be different depending on milk fat concentration. In addition, the
effect of propionate on energy intake was our primary concern, and evaluation of ME
intake is appropriate to accomplish our objective of this experiment. Acetate and
propionate were assumed to contain 0.2094 and 0.3672 Mcal/mol of ME, respectively
(Sheperd and Combs, 1998). Infusates were weighed before and after infusion, and
actual amount of solutions infused into the rumen was calculated. The ME from
infusates was calculated by multiplying ME concentration of infusates by the amount
infused into the rumen for 14 h or 12 h. Total ME intake was calculated using ME intake
from infusion over 14 h and 12 h, but results were similar with same statistical

significance. Therefore, ME intake from 12 h of infusion is discussed.

Diet ingredients were dried in a 55° C forced—air oven for 72 h and analyzed for
DM concentration. All samples were ground with a Wiley mill (1mm screen; Authur H.
Thomas, Philadelphia, PA). Samples were analyzed for ash, NDF, ADF, CP, and starch.
Ash concentration was determined after 5 h oxidation at 500° C in a muffle furnace. The
NDF and ADF concentrations were determined [(VanSoest et al., 1991); method A for
NDF]. Crude protein was analyzed according to Hach et al. (1985). Starch was

measured by an enzymatic method (Karkalas, 1985) after samples were gelatinized with

59

sodium hydroxide, and glucose concentration was measured using a commercial kit
(Glucose kit #510; Sigma Chemical Co., St. Louis, MO). Concentrations of all nutrients
except for DM were expressed as percentages of DM determined from drying at 105° C
in a forced-air oven. Corn grain was dry sieved through 8 sieves (Sieve apertures: 4750,
2360, 1180, 600, 300, 150, 75 um and bottom pan), using a sieve shaker (Model RX-86,
W.S. Tyler Inc., Gastonia, NC) for approximately 20 min until the bottom pan weight

was constant and, mean particle size of corn grain was calculated (ASAE, 1968).

All data for experiment 1 were analyzed using the fit model procedure of JMP®

according to the following model:

Yum = u + Ci + Pj + Lk + Q. + eijklm

where
u = overall mean,
Ci = random effect of cow ( j = 1 to 8),
Pj = fixed effect of period (k = 1 to 8),
L1‘ = linear effect of treatment,
Q] = quadratic effect of treatment, and

e = residual, assumed to be normally distributed.

ijklm

Actual amount of solution infused into the rumen was not affected by treatments, and it

was not included in the statistical model. Linear and quadratic effects of treatments were

60

evaluated. Treatment effects were declared significant at P < 0.05, and tendency for

treatment effects was declared at P < 0.10.

Experiment 2

The cows used for experiment 1 were used for experiment 2 at a later stage of
lactation (159 i 26 DIM; mean :1: SD). Cows were fed the same diet described for
experiment 1, and were assigned to duplicated 4 x 4 Latin squares balanced for carry-
over effects. Treatments were continuous intra-ruminal infusion of mixtures of sodium
propionate and sodium acetate at 4 different ratios. The infusion rate of VFA for
experiment 2 was higher than that for experiment 1 (25.0 vs. 16.7 mmol/rnin). Thus,
acetate and propionate were infused as sodium salts for experiment 2 to avoid risk of
ruminal acidosis, which is different than experiment 1 in which they were infused as
acids. Treatment solutions were prepared by diluting 25.2 moles of VFA salt (sodium
propionic acid and acetic acid at ratios of 0:3, 1:2, 2:1, and 3:0) to 18 L with de-ionized
water. Sodium acetic acid was added to keep the osmolarity of infusates constant across
the treatments to isolate specific effects of propionate relative to acetate on feeding
behavior of dairy cows. Concentration of total VFA was 1.4 M across the treatments, and
15 L of each solution was infused over 14 h. Infusion rate was 17.9 ml/min, which is
equivalent to infusion of 25.0 mmol of VFA/min. Infusion protocol, methods for data
and sample collection, methods for data and sample analysis were as described for
experiment 1. Additionally, milk yield was recorded and milk samples were taken at

both milking on every infusion day for experiment 2. Milk samples were analyzed for

61

fat, true protein, lactose, solids-non-fat, milk urea nitrogen concentration with infrared

spectroscopy by Michigan DHIA (East Lansing).

All data for experiment 2 were analyzed using the fit model procedure of JMP®

according to the following model:

Yijkl = u + Si + C(S)j(,, + Pk + Ll + Qm + CovINF + e”.kl
where

p. = overall mean,

Si = fixed effect of square (i = 1 to 2)

C(S)j(,, = random effect of cow nested in a square (j = 1 to 8),

Pk = fixed effect of period (k = 1 to 4),

LI = linear effect of treatment,

Qm = quadratic effect of treatment,

Cova = effect of actual amount of solution infused into the rumen as covariate,
and

e = residual, assumed to be normally distributed

ijklmn
One pump was used for each square of 4 cows, and random effect of cow was nested in a
square that shared the same infusion pump. Interactions of square x treatment and period
x treatment were originally evaluated, but they were removed from the statistical model
because interactions were not significant for response variables of interest. Actual

amount of solution infused into the rumen was included in the statistical model as a

62

covariate because actual amount of infusates tended to differ by treatments (quadratic
effect of treatments: < 0.09). Linear and quadratic effects of treatments were evaluated.
Treatment effects was declared significant at P < 0.05, and tendency for treatment effects

was declared at P < 0.10.

RESULTS

Experiment 1

As infusion rate of propionate increased, DMI decreased linearly from 15.1
kg/12h for 0% propionate to 13.2 kg/12h for 100% propionate treatment (P < 0.01; Table
2). Intermeal intervals tended to be longer (P < 0.07) and meal size tended to be smaller
(P < 0.09) as close of propionate increased. Total ME intake decreased linearly by
infused propionate from 43.4 Mcal/12h for 0% propionate to 40.2 Mcal/12h for 100%
propionate (P < 0.05). Quadratic effects of treatment were not observed for any response

variable.

Experiment 2

As proportion of propionate in infusate increased, DMI decreased linearly from
15.0 kg/12h for 0% propionate to 8.3 kg/12h for 100% propionate treatment (P < 0.0001;
Table 3). Similarly, propionate decreased meal size linearly from 2.5 kg for 0%
propionate to 1.5 kg/12h for 100% propionate treatment (P < 0.03). Number of meal

bouts over 12h tended to decrease linearly (P < 0.08) with increasing propionate although

63

0606006 6 600.66 0606000 N
060600.: 6 600.66 00005.

 

 

 

 

6.0 Cd vd wd w.v d6 m6 dd w.v 6.v dd 000636.50 669
mvd 66d md 6.3. a.d6 a.66 mam 6.a6 mda 6.66 a.a6 665 606066 625.5
avd 66d 6d ma m6 6 mdd dd_ 6a 6.66 6.: v.6 6 Eds 0606 6666.5
awd 6d 6d ddv mdv 68v mgm a.6v vdv adv a.av EdCr—v 00635 60603
660606 wan—6.5
dad 3d v w; ddm d.wd add add mdd v.md 6.md wdd SEQ 3665
dad mwd v6 and 6mm mam and :v Rm Km and 666665 36.6.
vvd 5d v 6.6 2: v.2 W66 Q66 ddfi wvd v.m_ d.m_ GEO mad—665
and :d d6 wdd ddd :d add Mdd add mad ddd A6d6665 6600636
3.6 2d 6d v.2 add 6.: w.: v.2 nd_ add 6.2 335 3665
dmd 6md ma avd av_ a2 066 m: wv_ m66 wfl £2665 6060mm
066 665060
5d mod 56 ddv adv 6am ddv mvv wdv d.mv vdv EdCfio—Zv 06H
6.0 566d v mod m.v 6v wd md md dd ad md A6dS0025 68:66
awd Ed v ad add w.6m mdm 0.6m d._v wdm mdv adv 2260066 605
06066 .3600 06006060602
6nd add 6d a6 a6 dd 6.6 dd dd md a6 95 we: 006 6002
_ad add m6 mdw flab fiva 62. 6.: 6.3 6.65 6.66 665 603065 6060066
5d wmd md E. 66 dd 66 6d a6 a6 md Edg 8:06 6002
awd Sd v 6d d.m6 6.2 d.m_ 6.2 d66 mvd a.v6 6.66 Edtwo—v 6,5
660606 w0600m
and mvd dd wvH w.v_ fivd m.v_ wvd w.v_ 6.3 0.3 35 60666 0066> 60806.
NO .6 mm dd 76 dum mnv vum md 66 and
0063 & 060m 06060040. ” 0606605

 

.6 60Et0fi0 6.6 660606 m=600a 6 0606000 06 0360—8 0606306 6 63066 6006600065 6 800t0 0066000 06G .d mqufi.

 

 

 

 

«no 00.0 0.0 E 3 00 0.0 0:256 36... 0:06:00
6d.d 6d.d v ad 6.6.6 ad6 md6 ddv 6065 66060606 66606.5
a6d 6d.d v dd 6.66 d6 6 6.66 6.v6 Eds 60606 660606666
ndd 6ddd V dd 6dd hda dda add6 66d6d66 60060606 60603
630606 6660666
dad 66d d6 v.6d d.6d ad 6 d.a6 QED do66665
6d.d 6d.d v 56 6d6 dvd 66d ddd 22665 606.6.
aad vmd m6 dd dd 6.6. dd SEQ ”0665
hvd vmd d6 66 6a va vd6 266665 66006636
dad ndd ad dd6 6d6 a.66 dd6 663666 @6665
ddd 6ddd v d6 d6 6 m6 6 666 dE 6666665 660m
0666 663065
hdd 6dddd v a.6 6.ad d.6m d.6v 6.3 66660065 606.6.
mmd 6dddd v ddd 6.6 6.6 d.v dd 66660026 66.666
6d.d 6dddd v a.6 vdd m. 6 m 6.66 fidv 6566600625 605
006066 36000 06600660606062
661d mdd dd 6.6 6d dd 6d 95 mod 0060 60062
6ad 6d.d 6.a 6da find 6.66 v.65 6:65 606066 60066066
dad ddd 6d 66 d6 v.6. vd A6d6d 0606 60062
6d.d 6dddd V dd md 6.66 06 d.66 6626wa 66665
630606 6600.6
add d6d dd 6.6 6 a.v6 d.66 d6 6 dd 60666 06:60> 603060.
N0 .6 mm cum 6mm N6 To
0:60> 06 0606 B003. 6 0606666066

 

66600606 06666 600 630606 606006 66 0606000 06 0360606 06066606 6 6660066 60066060666 6 0600660 006666006 00066 d “.6626.

.d 6006660960 66

65

£06 98 8:385 88% 835 $35 235—3808 32 ”8:35 m2 ..
A3 .d v a: Hgomfiwmm 83 530825 E2535 x totem m

$58305 do Soto 038930 N
35:53: do Soto .8055

 

 

 

 

mdd 5d v ddd ddd Ed ddd ddd EdSmové .83.: m2

H Amino—2d $.85 #32
who 36 «m as Ed 3N and 8385 355 52
:d :d dd md dd dd dd 23?: coded: «82 VEE
ddd mud ddd vdd ddd mdd ddd Add mZm x52
Sud mmd hdd ndd ddd ddd :uv Add 3803 x52
m d d ddd v ddd ndd mdd dd d S d 3% Edda :52
“dd dvd ddd dd;V ddd a: dd 3% and V=22
dmd ddd 2 ddd mdm ddd d. d m 8E6: 20% x32

deducted x32
NO .1— mm dim 7d d; mud
and? & ovum BS8< H Bacoaoi

 

.6253 m mam/db

66

intermeal interval was not significantly affected by treatment. Total ME intake decreased
linearly by propionate dose from 44.5 Mcal/12h for 0% propionate to 29.1 Meal/12h for
100% propionate (P < 0.0001). Infusion of propionate decreased water intake linearly
from 103.9 L/12h for 0% propionate to 82.6 L/th for 100% propionate treatment. This
reduction of water intake was because of less frequent water intake (P < 0.01) because
water consumed per bout was not affected by treatment. Milk yield, milk fat
concentration and milk lactose concentration were not affected by treatment. However,
milk protein concentration decreased linearly as infusion rate of propionate increased (P

< 0.01). Quadratic effect of treatment was not observed for any response variable.

DISCUSSION

Feed Intake

In both experiments, infusion of propionate decreased DMI in a dose dependent
manner. Treatment effect was attributed to the specific effect of propionate relative to
acetate because infusates were similar in osmolarity and pH. Our observation provides
strong evidence for hypophagic effect of propionate, in agreement with previous studies.
Infusion of propionate into the portal vein of sheep decreased feed intake to a greater
extent compared to infusion of acetate or butyrate (Anil and Forbes, 1980) and compared
to infusion of acetate, mannitol, or saline (Farningham and Whyte, 1993). Infusion of
propionate into the mesenteric vein of steers reduced feed intake while infusion of acetate
did not (Elliot et al., 1985). Although the hypophagic effect of propionate has been

investigated extensively, most experiments in the literature have monitored feed intake

67

over very short periods ranging from 30 minutes to 3 hours, and essentially investigated
the effect of propionate on meal size only. Evaluation of feeding behavior helps to
understand the regulation mechanism of DMI by propionate because DMI is a function of
both meal size and intermeal interval, which are determined by satiety and hunger,
respectively. In the present study, feeding behavior was monitored for 12 h, and effects
of propionate on intermeal interval and meal frequency were evaluated as well as meal
size. Infusion of propionate tended to decrease meal size and increase intermeal interval
in experiment 1, and decreased meal size and tended to decrease meal frequency in
experiment 2. Our observations indicated that propionate decreased feed intake by

affecting both satiety and hunger.

Infusates contained more energy as the proportion of propionate increased
because of the greater energy concentration for propionate compared to acetate, but total
ME intake also decreased linearly for both experiments as propionate increased.
Reduction in ME intake from the diet exceeded that supplied from infusion as proportion
of propionate increased. Previous reports have shown that propionate decreased DMI
compared to iso-caloric infusion of VFA mixture (Hurtaud et al., 1993) or acetate in
lactating dairy cows (Sheperd and Combs, 1998). Wu et al. (1994) reported lower DMI
for cows infused with propionate into the duodenum compared to iso-caloric infusion of
glucose into the rumen. That is consistent with our results because glucose ferments to
acetate and butyrate as well as propionate. These studies along with the present study

suggest that hypophagic effect of propionate cannot be explained simply by the additional

68

energy supplied as propionate. Animals do not consume to meet their energy

requirements per se but have specific mechanisms regulating satiety and hunger.

Increased ruminal fermentation has been related to reduced energy intake for
lactating cows (McCarthy et al., 1989; Overton et al., 1995). McCarthy et al. (1989)
compared ground shelled corn and steam rolled barley in high grain diets containing
more than 45% grain, and found that starch digestibility in the rumen was 77% for cows
fed barley-based diets and 48.5% for cows fed com-based diets. In their experiment,
cows fed ground corn had a DMI of 23.8 kg/d while cows fed steam rolled barley had a
DMI of 20.7 kg/d. Although starch digestibility was lower for corn treatments, amount
of DM and OM digested in the total tract appear to be greater for cows fed ground corn
because of greater DMI. Similarly, Overton et al. (1995) fed steam rolled barley and
ground shelled corn at five different ratios (100:0, 75:25, 50:50, 25:75, and 0:100, for
ground shelled corn starch : steam rolled barley starch) in low forage diets (45% dietary
DM). They found a linear increase in starch digestibility (P < 0.01) and linear decrease
in DMI (P < 0.0001) as the ratio of steam rolled barley increased in the diet. In addition,
amount of DM and OM apparently digested in the total tract linearly decreased as the
fraction of steam rolled barley increased in diets. Excess propionate production in the
rumen might have limited energy intake as well as DMI when cows were fed very

fermentable grains for both experiments.

Quadratic effects were not significant for any response variable in either

experiment, providing no evidence for a threshold response to infused propionate in

69

feeding behavior and energy intake in this study. Additionally, the breakpoint for
response in DMI to treatments, estimated as -b/2a for the regression equation of ax2+bx+c
(SAS, 1990), was not identified within the range of rate for propionate infusion for both
experiments, indicating a linear hypophagic effect of propionate only. In agreement with
our results, Anil et al. (1993) and Farningham and Whyte (1993) reported that infusion of
propionate linearly decreased feed intake in a dose dependent manner without a
threshold. Leuvenink et al. (1997) showed that propionate infusion into the mesenteric
vein of mature sheep at a rate of 1 mmol/min did not decrease feed intake, but the
infusion at a rate of 2 mmol/min significantly decreased feed intake. However, their data
cannot be used to support a threshold response of propionate in feed intake because the
effect of propionate was evaluated at only two levels of infusion and feed intake was
numerically decreased for the lower dose although it was not statistically different from

control.

Water Intake

Water intake was nearly twice as high for experiment 2 compared to experiment 1
(93.6 vs. 48.7 L/12h). This difference is attributed to infusion of acids in experiment 1
and sodium salts in experiment 2. Murphy (1983) suggested that a gram of sodium intake
increase water intake by 0.05 L. Using this relationship, infusion of 483 g of sodium
over 14 h in experiment 2 would be expected to explain 24.2 L of the 44.9 L difference in
water intake between experiment 1 and experiment 2. Infusion of sodium might have
increased osmolarity of ruminal fluid drawing water into the rumen from the blood,

resulting in thirst and increasing water intake. In addition, greater water intake might be

70

because of increased urine volume to excrete excess sodium. Although infusion
treatment did not affect water intake in experiment 1, water intake decreased linearly as
propionate infusion increased for experiment 2. This might be explained by the greater
effect of propionate treatment on feed intake in experiment 2 compared to experiment 1
because infusates were iso-osmotic within in each experiment. Both feed intake and salt

intake stimulate water intake in lactating dairy cows (Murphy et al., 1983).

Milk Production

In experiment 2, milk yield was not significantly affected by infusion of
propionate despite the linear reduction in energy intake. This is probably because the
short duration (14 h) of the infusions was not adequate to affect milk yield significantly.
However, total ME intake decreased by 35% for 100% propionate treatment compared to
0% propionate treatment. If energy intake was affected by propionate for a longer period,
milk yield would be expected to decrease. Previous experiments that reported lower
digested OM intake for more fermentable diets also reported reductions in milk yield
(McCarthy et al., 1989; Overton et al., 1995). Oba and Allen (2001) showed that DMI
and fat corrected milk yield were decreased by feeding high moisture corn compared to
dry ground corn for cows fed high grain diets. Reduction in milk protein concentration
for cows infused with more propionate might be explained by lower microbial protein
production. Infused acetate and propionate provided additional energy for animals but
not for microorganisms in the rumen. Therefore, infused propionate reduced fermentable
energy to synthesize microbial protein by decreasing DMI. Although propionate can

increase availability of amino acids for the mammary gland by sparing them from being

71

utilized for gluconeogenesis, the linear reduction in milk protein concentration by
increased propionate infusion indicates that milk protein synthesis is limited by microbial
protein production to a greater extent (Wu et al., 1994). In addition, when metabolizable
protein limits maximum milk yield, infusion of glucogenic energy as propionate without
additional amino acid supply may have diluted milk protein by relatively greater lactose

synthesis and milk yield.

CONCLUSION

Intra-ruminal infusion of propionate decreased DMI and ME intake in a dose-
dependent manner. This indicates that the reduction in dietary energy intake from
propionate infusion was greater than the energy supplied from propionate infusion and
that excess propionate production in the rumen can decrease energy intake in lactating
dairy cows consuming highly fermentable diets. However, quadratic effects of
propionate infusion were not significant for DMI and ME intake, providing no evidence
for a threshold response to infused propionate in feeding behavior and energy intake. As
proportion of propionate in infused VFA increased, meal size tended to decrease and
intermeal interval increased in experiment 1, and meal size decreased and meal frequency
tended to decrease in experiment 2. These observations suggest that propionate plays an

important role in feed intake regulation by affecting both satiety and hunger.

72

CHAPTER 4

Effects of intra-ruminal infusion of sodium, potassium and ammonium

on hypophagic effect of propionate in lactating dairy cows

ABSTRACT

The objective of this experiment was to evaluate effects of salt type on
hypophagic effects of intra-ruminal infusion of propionate in lactating dairy cows. Our
working hypothesis is that oxidative metabolism of propionate causes satiety by
increasing hepatic ATP concentration and decreasing the discharge rate of the hepatic
vagus. We hypothesized that infusion of ammonium would reduce the hypophagic
effects of propionate because of increased utilization of ATP for urea synthesis. We also
hypothesized that infusion of potassium would decrease hypophagic effects of propionate
compared to sodium by increasing the discharge rate of the hepatic vagus. Eight
ruminally cannulated Holstein cows in mid-lactation were used in a duplicated 4 x 4
Latin square design, and treatments were intra-ruminal infusion of propionic acid,
ammonium propionate, sodium propionate, and potassium propionate. Treatment
solutions were 0.93 M for propionate among treatments and 0.67 M for salts among the
treatments except for control (propionic acid). Treatment solutions were infused over 14

h starting 2 h before feeding at 17.9 ml/min, which is equivalent to 16.7 and 11.9

73

mmol/min for propionate and salts, respectively. Infusion of ammonium propionate
decreased DMI compared to sodium propionate and potassium propionate (P < 0.04; 11.0
vs. 14.0 and 13.9 kg/ 12 h), and DMI was not different between sodium propionate and
potassium propionate infusions. Contrary to our hypothesis, ammonium infusion did not
reduce hypophagic effects of propionate possibly because the urea cycle indirectly
stimulated oxidative metabolism in the liver by generating oxidizable carbon from amino
acid catabolism.

(Key words: propionate infusion, ammonium, urea synthesis, feed intake, feeding
behavior)

Abbreviation Key: MUN = Milk Urea Nitrogen

INTRODUCTION

Increasing ruminal fermentation is desirable to maximize microbial protein
production and energy intake in high producing dairy cows. However, greater ruminal
fermentation sometimes decreases feed intake and excess propionate production in the
rumen is considered to have a direct hypophagic effect (Allen, 2000). Choi and Allen
(1999) showed that meal size, intermeal interval, and DMI were not affected by infusion
of acids compared to salts, indicating that ruminal pH per se does not have direct
hypophagic effects. In that experiment, infusion of acetate resulted in similar meal size
as iso-osmotic infusions of N aCl. It was concluded that acetate affects satiety by

mechanisms related to osmotic effects in the rumen. However, infusion of propionate

74

decreased meal size and DMI, and increased intermeal interval compared to infusion of
acetate, providing the evidence for a specific hypophagic effect of propionate. Anil and
Forbes (1980, 1988) suggested that propionate decreases feed intake by a mechanism via
the hepatic vagus but the mechanism for regulation of feed intake by propionate has not

been elucidated.

Allen (2000) proposed that propionate decreases feed intake of ruminants by
stimulating oxidative metabolism in the liver. Oxidative metabolism in the liver has been
shown to affect satiety in rats (Langhans et al., 1983, 1984, 1985a) and a temporal
relationship between feeding behavior and hepatic ATP concentration has been
demonstrated (Koch et al., 1998). Langhans et al. (1985a) proposed that oxidative
metabolism in the liver affects feed intake by hyperpolarizing cell membrane potentials.
The sodium pump inhibitor, ouabain, increased feed intake in rats when injected
intraperitoneally (Langhans and Scharrer, 1987b). Satiety signals originating in the liver
are mediated by hepatic vagal afferents (Langhans et al., 1985c; Anil and Forbes, 1988),
and Niijima (1983) reported that discharge rates of hepatic vagal afferents were reduced
by glucose infusion in a dose-dependent manner in guinea pigs. These observations
suggest that oxidative metabolism within hepatocytes generates ATP, increases sodium
pump activity, hyperpolarizes hepatocyte membrane potential, and decreases the

discharge rate of hepatic vagal afferents, resulting in satiety.

Our long-term goal is to alleviate hypophagic effects of propionate by a diet

formulation that modulates a metabolic or neural pathway by which propionate causes

75

satiety. If propionate metabolism in the liver generates satiety signals by increasing
hepatic ATP concentration, hypopagic effects of propionate can be alleviated by reducing
hepatic ATP concentration. We hypothesized that infusion of ammonium reduces the
hypophagic effects of propionate because urea synthesis consumes ATP in the liver. In
addition, if decreased discharge rate of hepatic vagal afferents sends a satiety signal to the
brain, increasing their discharge rate can alleviate hypophagic effects of propionate. We
hypothesized that infusion of potassium increases feed intake by increasing discharge rate
of vagal affemets compared to infusion of sodium. We expected infusion of potassium to
decrease the potassium gradient across the membrane of vagal afferents, reducing
potassium efflux, depolarizing resting transmembrane potential of vagal affemets, and

increasing their discharge rate.

The objective of this experiment was to evaluate effects of salt type on

hypophagic effects of intra-rurninal infusion of propionate in lactating dairy cows.

MATERIALS AND METHODS

Experimental procedures were approved by All University Committee on Animal
Use and Care at Michigan State University. Eight multiparous Holstein cows (143 i 26
DIM; Mean i SD) cannulated ruminally for previous experiments were selected from the
Michigan State University Dairy Cattle Teaching and Research Center. Experimental

diets contained dry cracked corn (mean particle size of 3.6 mm), corn silage, alfalfa

76

silage, a premix of protein supplements (soybean meal, distillers grains, and blood meal),
and a premix of minerals and vitamins (Table 1). Dietary NDF, CP, and starch
concentrations were 29.0, 15.9, and 30.8%, respectively. Means for BW, BCS
[(Wildman, 1982); a five-point scale where 1 = thin to 5 = fat], DMI and milk yield
before experiment were 623 kg, 2.6, 25.4 kg/d, and 36.4 kg/d, respectively. Cows were

assigned to duplicated 4 x 4 Latin squares balanced for carry over effects.

Treatments were continuous intra-ruminal infusion of mixtures of propionic acid
and bicarbonate salts (control, NH3, Na, or K). Treatment solutions were prepared by
diluting 16.8 moles of propionic acid and 12.0 moles of bicarbonate salts (NH3, Na, or K)
to 18 L with de-ionized water. Bicarbonate salts were used instead of hydroxide salts to
maximize the purity of treatment salts; the purity of commercially available potassium
hydroxide was approximately 85% whereas the purity of potassium bicarbonate was
99.9%. Control treatment (Acid) was propionic acid without any bicarbonate salts. The
concentration of propionate was 0.93 M across treatments, and concentration of salts was
0.67 M across the treatments except for control. Solutions were infused at a rate of 15 L
over 14 h. This infusion rate of 17.9 ml/min is equivalent to 16.7 and 11.9 mmol/min for
propionate and salts, respectively. This rate of propionate infusion was shown to
decrease DMI significantly in a previous experiment (Chapter 3), and was selected to
evaluate how salt type affects hypophagic effects of propionate in this experiment. The
concentration of salts was less than propionate to avoid the risk of ammonia toxicity for
ammonium treatment; total amount of ammonium infused into the rumen was equivalent

to 875 g of CP over 14 h. Solutions were infused using 4-channel peristaltic pumps

77

TABLE 1. Ingredients and nutrient composition of experimental diets (% of dietary DM
except for DM).

 

Diet Ingredients
Corn silage 27.0
Alfalfa silage 25.4
Dry cracked com 25 .9
Whole linted cottonseed 6.8
Protein mixl 9.9
Vitamin & mineral mix2 5.0

Nutrient Composition

DM 49.5
CM 93.0
Starch 30.8
NDF 29.0
ADF 20.8
CF 15.9
Ether extract 3.8
Forage NDF 21.2
Metabolizable energy (Meal/kg) 3 2.72

 

1 Protein mix contained 75% soybean meal, 20% distillers grain, and 5% blood meal.

2Vitamin & mineral mix contains 66.4% dry ground corn, 20.4% dicalcium phosphate,
7.8% salt, 2.4% magnesium oxide, 1.9% trace mineral premix, 0.34% vitamin A, 0.29%
vitamin D, and 0.08% vitamin E.

3 Metabolizable energy was calculated from book values according to NRC (1989).

78

(#78016-30, Cole—Farmer Instrument, IL) and Tygon® tubing (7.5 m x 1.6mm I.D.).

Treatment periods were 2 d with 14 h of infusion followed by 34 h of recovery.

Throughout the experiment, cows were housed in tie-stalls, and fed once daily
(1030 h) at 110% of expected intake. Cows were not allowed access to feeds between
0830 h to 1030 h. The amount of feed offered and orts were weighed for each cow daily.
On every infusion day, samples of all dietary ingredients (0.5 kg) were collected, and
cows were infused treatment solutions from 0830 h to 2230 h. Cows were milked twice
daily in the milking parlor except for the evening milking on infusion days, for which
cows were milked in their stalls. Feeding behavior was monitored from 1030 h to 2230 h
on each infusion day by a computerized data acquisition system (Dado and Allen, 1993).
Data of chewing activities, feed disappearance, and water consumption were recorded for
each cow every 5 sec, and meal bouts, interval between meals, meal size, eating time,
ruminating time, and total chewing time were calculated. Milk yield was recorded and
milk samples were taken at both milking on each infusion day for experiment 2. Milk
samples were analyzed for fat, true protein, lactose, solids-non-fat, milk urea nitrogen

(MUN) concentration with infrared spectroscopy by Michigan DHIA (East Lansing).

Diet ingredients were dried in a 55° C forced-air oven for 72 h and analyzed for
DM concentration. All samples were ground with a Wiley mill (1mm screen; Authur H.
Thomas, Philadelphia, PA), and analyzed for ash, NDF, ADF, CP, and starch. Ash
concentration was determined after 5 h oxidation at 500° C in a muffle furnace. The

NDF and ADF concentrations were determined [(VanSoest et al., 1991); method A for

79

NDF]. Crude protein was analyzed according to Hach et al. (1985). Starch was
measured by an enzymatic method (Karkalas, 1985) after samples were gelatinized with
sodium hydroxide, and glucose concentration was measured using a commercial kit
(Glucose kit #510; Sigma Chemical Co., St. Louis, MO). Concentrations of all nutrients
except for DM were expressed as percentages of DM determined by drying at 105° C in a
forced-air oven. Corn grain was dry sieved through 8 sieves (Sieve apertures: 4750,
2360, 1180, 600, 300, 150, 75 pm and bottom pan), using a sieve shaker (Model RX-86,
W.S. Tyler Inc., Gastonia, NC) for approximately 20 min until the bottom pan weight

was constant, and mean particle size of corn grain was calculated (ASAE, 1968).

All data were analyzed using the fit model procedure of JMP® according to the

following model:

Ya,“ = u + Si + C(S)jm + Pk + T1 + CovINF + em,n
where
u = overall mean,
Si = fixed effect of square (i = 1 to 2)
C(S)j(,, = random effect of cow nested in a square 0 = l to 8),
Pk = fixed effect of period (k = 1 to 4),
T1 = fixed effect of treatment (1 = 1 to 4),
Cov,NF = effect of actual amount of solution infused into the rumen, and

eijkl = residual, assumed to be normally distributed

80

One pump was used for each square of 4 cows, and random effect of cow was nested in a
square that shared the same infusion pump. Interactions of square x treatment and period
x treatment were evaluated, but they were removed from the statistical model because
interactions were not significant for response variables of interest. Volume of solution
infused into the rumen was included in the statistical model as a covariate. Orthogonal
contrasts were made for the effect of acid (acid vs. NH3, Na, and K), the effect of
ammonium (NH3 vs. Na and K), and the effect of cation type (Na vs. K). Treatment
effects was declared significant at P < 0.05, and tendency for treatment effects was

declared at P < 0.10.

RESULTS AND DISCUSSION

Volume of solution infused into the rumen was slightly greater for propionate salt
compared to propionic acid treatment and for sodium and potassium propionate
compared to ammonium propionate treatment (P < 0.01; Table 2). This was unexpected
because a 4—channel peristaltic pump was used for each square with identical tubing and
fittings, but a difference in viscosity of infusates might have affected flow rate.
Therefore, volume of solution infused into the rumen was included in the statistical

model as a covariate.

DMI and Feeding Behavior
Dry matter intake, number of meal bouts, intermeal interval, and meal size were

similar between infusion of propionic acid and propionate salts, indicating pH per se does

81

not have a direct effect on feeding behavior of lactating dairy cows. These results agree
with previous observations by Choi and Allen (1999). However, infusion of ammonium
propionate decreased DMI compared to infusion of sodium propionate and potassium
propionate (P < 0.04; Table 2). Contrary to our hypothesis, infusion of ammonium did
not decrease, but increased the hypophagic effect of propionate. This reduction in DMI
was not likely from toxic effects of ammonia because MUN concentrations for the
ammonium propionate treatment were within the range of normal values for lactating
cows (Butler, 1996). Ammonium infusion increased intermeal interval (P < 0.01) and
decreased number of meal bouts (P < 0.02) compared to sodium propionate and
potassium propionate treatments, but did not affect meal size. These observations
indicate that the reduction in feed intake from ammonium infusion was from delaying the
sense of hunger after a previous meal. No difference in DMI and feeding behavior was
observed between infusion of sodium and potassium propionate. We infused 230 g of
sodium or 391 g of potassium over 14 h, which are far greater than animal requirements
(NRC, 2001). Any possible effects of potassium ions on alleviation of hypophagic

effects of propionate are not likely from dietary manipulation.

Prior to the experiment, we hypothesized that infusion of ammonium propionate
would increase DMI compared to other propionate salts because cows infused with
ammonium need to synthesize additional urea, which consumes 3 moles of ATP per mole
of urea synthesis. Greater MUN values for ammonium propionate treatment compared to
other treatments (P < 0.001) support that additional urea synthesis occurred for cows

infused with ammonium propionate. Observed hypophagic effect of ammonium may be

82

 

ddd Ed ddd dd dd dd dd _d 953V ONE xctfl

 

 

 

ddd 5d V ddd v.2 hdv ddd ddd ddd 3:5 «BEBE wad—5.5
ddd _dd.d v ddd d4 ddd v.3 dd dd Eda 330d wad—5.5
ddd Sod v Sod v dd WE. ddd d5» dddV Addtd 835 83>?
.8333 mafia—ta
ddd ddd fiddd v d; ddd ddd ddd ddd :20 $383
ddd d_ .d d dd dd ddd ~dd ddd _dd EdCEEV 38,—.
dvd ddd dddd V dd dd_ 5: ddd ddd QED wfififiv
vdd ddd Edd v d d dt dd“ dd_ ddd 33:35 deducing—m
Vdd ddd ddd dd d._ _ d.: E: d; d QED $355
ddd ddd ddd E ddd >2 Vdd ddd Edifiav madam
can wEBonU
ddd ddd ddd dd dd d; dd dd 95 dd ads :32
dvd _d.d V dd ddd ddd ddd ddE ddd 3:5 BEBE REESE
ddd ddd ddd dd dd dd dd dd Eds 3:8 :32
3d Vdd ddd dd ddd 0.3 d.: d.: Edtmoc H35
8323 madden—
ddd ddd V dd v dd dd fl dd d dd d d4; Ad 9533 0823 383‘
MU am _< mm Esmmmfiom Eidom 8:25:82 Eo<
025, & $58389

 

.2283on #28 EB 8323 wEdoud so 33. Bacoaoa do command Ecgaébfi do 88am .d mqmfifi.

83

30% 98 08835 God 0035 $0000 030309308 .300 ”8:35 m2 n
Add .d v at agodawmm 003 0030885 00253: x noted 0

Sammfiom .m> 8:63 0833 00303800 m
Esamfiom new Sadom .m> 8300802 5253 08:09:00 N

Simmfiom Ea .Eadom .8:EoEE< .0> 20¢. 000303 :85d800.

 

 

 

 

00.0 00.0 000 00.0 5.0 00.0 00.0 00.0 205850 .8025 m2
H 2500—): 35:0 052
8.0 00.0 00.0 0.0 0.00 SN 02.. EN 60820 $85 002
00.0 30.0 v 50.0 v 0.0 3 E 5.2 0.0 02080 Smog: 02: 03
8.0 v 05.0 00.0 2.0 80 $0 00.0 3.0 $0 .020 002
00.0 v R0 00.0 S .0 NB. 5.0 00.0 00.0 38 0832 0%
00.0 t .0 $0 8.0 00.0 00.m 00.m 03 £00 .5380 002
$0 000 «0.0 00.0 3.0 mm... 00.0 mm... 3% E 0002
2.0 00.0 8.0 v 0.~ a? m. 0 m 0. 0 m 5.00 830 2% é:

codosdoa 052
mU mm _< mm 823305 Eamfiom 858082 Eo<
0%; K maven—deb.

 

.5803 d mam/2.

84

explained by increased oxidation of amino acids in the liver. One of the two amino
groups of urea is from ammonia but the other is from amino acids via aspartate; urea
production in the liver is associated with (IL-amino nitrogen removal in the liver
(Reynolds, 1992; Parker etal., 1995). Infusion of NH4C1 into the mesenteric vein of
sheep increased oxidation of leucine in splanchnic tissues (Lobley et al., 1995).
Therefore, urea synthesis might increase net ATP production in the liver by increasing
hepatic amino acid oxidation because each turn of TCA cycle generates 12 ATP while

each turn of urea cycle consumes 3 ATP.

Chewing and Drinking Activities

Infusion of ammonium propionate decreased eating time compared to infusion of
sodium and potassium propionate (P < 0.03), which is in agreement with lower DMI for
the ammonium treatment. Ruminating time (P < 0.001) was lower for infusion of
propionate salts compared to propionic acid. Ruminating time per kg of DMI was also
decreased by infusion of propionate salts compared to propionic acid (P < 0.001), but
reduction in rumination per kg of DMI tended to be greater for sodium and potassium
propionate compared to ammonium propionate treatment (P < 0.06). However,
ruminating time per kg of DMI was similar for sodium and potassium treatments. These
observations might be attributed to the expected greater osmolarity of ruminal fluid for
propionate salt treatments compared to propionic acid treatment and for sodium and
potassium treatments compared to ammonium treatment because greater osmotic pressure
in ruminal fluid is related to decreased rumination (W elch, 1982). Iso-osmotic infusion

of ammonium did not have as great of an effect at reducing ruminating time as infusion

85

of sodium and potassium possibly because of potential incorporation of ammonia into
microbial protein, and osmolarity of rumen fluid was increased to a less extent compared

to sodium and potassium treatments.

Consistent with treatment effects on ruminating time, infusion of propionate salts
increased water intake compared to infusion of propionic acid. Sodium and potassium
treatments increased frequency of drinking compared to ammonium treatment, indicated
by a greater number of drinking bouts (P < 0.001) and decreased interval between bouts
(P < 0.01). Drink size per bout tended to be greater for infusion of propionate salts
compared to propionic acid (P < 0.09). Drinking behavior was similar for sodium and
potassium treatments. Increased water intake from the infusions is probably because
greater osmotic pressure in ruminal fluid results in translocation of water into the rumen
from the blood resulting in thirst and from greater urine volume (not measured) to excrete

sodium and potassium.

Milk Production

Infusion of propionate salts decreased milk yield drastically compared to infusion
of propionic acid (P < 0.01). Reduction in milk yield is partially attributed to lower DMI
for ammonium treatment but not for sodium and potassium treatments because DMI was
numerically greater for sodium and potassium treatments compared to the acid treatment.
Milk energy output per ME intake tended to be lower for sodium and potassium
treatments compared to ammonium treatment (P < 0.08) which was similar to the acid

treatment. Infused solutions were iso-energetic across treatments, and the difference in

86

energy intake cannot explain our observation. Reduction in milk yield for sodium and
potassium treatments can be attributed to greater energy expenditures to maintain
homeostasis of blood for osmolarity and ion balance but not attributed to a specific effect
of sodium or potassium ion because milk energy output per Mcal of ME intake was

similar for sodium and potassium treatments.

87

CONCLUSION

Contrary to our hypothesis, hypophagic effects of propionate were greater when
infused as ammonium salt compared to sodium and potassium salts. Infusion of
ammonium decreased DMI by decreasing meal frequency without affecting meal size,
indicating that ammonium delayed the sense of hunger. Urea synthesis requires N from
amino acids, and deamination of amino acids might stimulate oxidative metabolism in the
liver by increasing amino acid carbon available for oxidation. No difference in DMI and
feeding behavior was observed between infusion of sodium and potassium propionate at
concentrations greater than practical for diet formulation, indicating that any possible
effects of sodium or potassium ions on feeding behavior are not likely from dietary

manipulation.

88

CHAPTER 5

Hypophagic effects of ammonium were greater when infused with

propionate compared to acetate in lactating dairy cows

ABSTRACT

The objective of this experiment was to determine if hypophagic effects of
ammonium are greater when infused with propionate compared to acetate in lactating
dairy cows. Urea synthesis generates amino acid carbon available for gluconeogenesis or
oxidation in the liver. We hypothesized that ammonium infusion stimulates oxidative
metabolism in the liver causing greater hypophagia when infused with propionate
compared to acetate because propionate is a primary substrate for gluconeogenesis while
acetate is not metabolized in the liver. Eight ruminally cannulated Holstein cows in mid-
lactation were used in a duplicated 4 x 4 Latin square design with a 2 x 2 factorial
arrangement of treatments. Factors evaluated were type of VFA (acetate vs. propionate)
and type of salt (sodium vs. ammonium). Concentration of VFA salts in infusates was
0.93 M across treatments, and infusion rate of 17.9 ml/min is equivalent to 16.7 mmol of
VFA salts/min. Treatment solutions were infused continuously into the rumen starting 2
h before feeding and ending 12 h after feeding. Dry matter intake was lower for
propionate compared to acetate treatment (P < 0.0001) and for ammonium compared to

sodium treatment (P < 0.001). Hypophagic effects of ammonium were significantly

89

greater for cows infused with propionate (4.3 vs. 12.1 kg/ 12 h) compared to acetate (13.5
vs. 15.3 kg/ 12 h; interaction P < 0.01). This interaction is attributed to greater reduction
in meal frequency for ammonium treatment compared to sodium treatment when infused
with propionate (3.9 vs. 7.2 / 12 h) compared to when infused with acetate (6.6 vs. 7.0 / 12
h), indicating that infusion of ammonium propionate delayed the sense of hunger. Meal
size was decreased by infusion of propionate compared to acetate (P < 0.01), but was not
affected by salt type. Greater hypophagic effects of ammonium propionate might be
because the urea cycle generates substrate for oxidation in the liver, increasing hepatic
ATP concentration.

(Key words: propionate infusion, ammonium, urea synthesis, feed intake, feeding
behavior)

Abbreviation Key: MUN = Milk Urea Nitrogen

INTRODUCTION

Greater ruminal fermentation sometimes decreases feed intake and excess
propionate production in the rumen might have a direct hypophagic effect by stimulating
oxidative metabolism in the liver (Allen, 2000). Oxidative metabolism in the liver has
been shown to affect satiety in rats (Langhans et al., 1983, 1984, 1985a) and a temporal
relationship between feeding behavior and hepatic ATP concentration has been
demonstrated (Koch et al., 1998). Langhans et al. (1985a) proposed that metabolic fuels

that are extensively metabolized in the liver have hypophagic effects. In previous

90

experiment (Chapter 4), infusion of ammonium decreased DMI compared to sodium and
potassium propionate by decreasing meal frequency. Hypophagic effects of ammonium
infusion were not likely from ammonia toxicity because observed milk urea nitrogen
(MUN) concentration for the ammonium propionate treatment was within the range of

normal variation for lactating cows (Butler, 1996).

Hypophagic effect of ammonium may be explained by increased oxidative
metabolism in the liver. One of the two amino groups of urea is from ammonia but the
other is from amino acids via aspartate; urea production in the liver is associated with on—
amino nitrogen removal in the liver (Reynolds, 1992; Parker et al., 1995). Infusion of
NH4C1 into the mesenteric vein of sheep increased oxidation of leucine by splanchnic
tissues (Lobley et al., 1995). Therefore, urea synthesis might increase net ATP
production in the liver by increasing hepatic amino acid oxidation despite utilization of
ATP by the urea cycle. Although carbon from some amino acids can be utilized for
gluconeogenesis and consume ATP in the liver, they might be oxidized to a greater extent
when glucose demand of peripheral tissues is low or when the liver has plenty of other
substrates for gluconeogenesis. If enhanced oxidative metabolism in the liver decreases
DMI, hypophagic effects of ammonium are expected to be greater when infused with
propionate compared to acetate because propionate is the primary substrate for
gluconeogenesis while acetate is not metabolized in the liver (Ricks and Cook, 1981).
When ammonium is infused with acetate, carbon from amino acids would be utilized for

gluconeogenesis to a greater extent. We hypothesized that hypophagic effects of

91

ammonium infusion vary depending on its stimulatory effect on oxidative metabolism in

the liver and increase when infused with propionate compared to acetate.

The objective of this experiment was to determine if hypophagic effects of
ammonium are greater when infusion with propionate compared to acetate in lactating

dairy cows.

MATERIALS AND METHODS

Experimental procedures were approved by All University Committee on Animal
Use and Care at Michigan State University. Eight multiparous Holstein cows (151 i 26
DIM; mean i SD) cannulated ruminally for previous experiments were selected from the
Michigan State University Dairy Cattle Teaching and Research Center. Experimental
diets contained dry cracked corn (mean particle size of 3.6 mm), corn silage, alfalfa
silage, a premix of protein supplements (soybean meal, distillers grains, and blood meal),
and a premix of minerals and vitamins (Table 1). Dietary NDF, CP, and starch
concentrations were 29.0, 15 .9, and 30.8%, respectively. Means for BW, BCS
[(Wildman, 1982); a five-point scale where 1 = thin to 5 = fat], DMI and milk yield

before experiment were 623 kg, 2.6, 25.4 kg/d, and 36.4 kg/d, respectively.

92

TABLE 1. Ingredients and nutrient composition of experimental diets (% of dietary DM
except for DM).

 

Diet Ingredients
Corn silage 27.0
Alfalfa silage 25.4
Dry cracked corn 25.9
Whole linted cottonseed 6.8
Protein mixl 9.9
Vitamin & mineral mix2 5.0

Nutrient Composition

DM 49.5
OM 93.0
Starch 30.8
NDF 29.0
ADF 20.8
CP 15.9
Ether extract 3.8
Forage NDF 21.2
Metabolizable energy (Meal/kg) 3 2.7 2

 

' Protein mix contained 75% soybean meal, 20% distillers grain, and 5% blood meal.
2Vitamin & mineral mix contains 66.4% dry ground corn, 20.4% dicalcium phosphate,
7.8% salt, 2.4% magnesium oxide, 1.9% trace mineral premix, 0.34% vitamin A, 0.29%
vitamin D, and 0.08% vitamin E.
3 Metabolizable energy was calculated from book values according to NRC (1989).

93

Cows were assigned to duplicated 4 x 4 Latin squares balanced for carry-over
effects with a 2 x 2 factorial arrangement of treatments. Factors evaluated were type of
VFA (acetate vs. propionate) and type of salt (sodium vs. ammonium). Treatments were

continuous intra-ruminal infusion of sodium acetate, ammonium acetate, sodium
propionate, or ammonium propionate. Treatment solutions were prepared by diluting
16.8 moles of VFA salts to 18 L with de-ionized water. Concentration of VFA salts were
0.93 M across treatments, and 15 L of each solution was infused over 14 h. Infusion rate
of 17.9 ml/min is equivalent to 16.7 mmol of VFA salts/min. Solutions were infused
using 4-channel peristaltic pumps (#78016-30, Cole-Farmer Instrument, IL) and Tygon®
tubing (7.5 m x 1.6mm I.D.). Treatment periods were 2 d with 14 h of infusion followed
by 34 h of recovery. We thought that the infusion rate of ammonium in a previous
experiment (11.9 mmol/min; Chapter 4) could be increased without risk of ammonia
toxicity based on MUN data from that experiment. In addition, we expected to see a
greater treatment effect on feeding behavior and DMI for this experiment compared to
the experiment in Chapter 4 by increasing the infusion rate for salts. Total amount of
ammonium infused into the rumen in this experiment was equivalent to 1227g of CP over

14h.

Total metabolizable energy (ME) intake was calculated by adding ME from
infusates to ME of the diet because energy concentration of infusates differed. The
experimental diet was assumed to contain 2.72 Meal/kg of ME based on book values
from NRC (1989). Acetate and propionate were assumed to contain 0.2094 and 0.3672

Mcal/mol of ME, respectively (Sheperd and Combs, 1998). Infusates were weighed

94

before and after infusion, and actual amount of solutions infused into the rumen was
calculated. The ME from infusates was calculated by multiplying ME concentration of
infusates by the amount that actually infused into the rumen for 14 h or 12 h. Total ME
intake was calculated using ME intake from infusion of 14 h and 12 h, but results were
similar with same statistical significance. Therefore, ME intake for the 12 h period for

which feeding behavior was monitored is discussed.

Throughout the experiment, cows were housed in tie-stalls, and fed once daily
(1030 h) at 110% of expected intake. Cows were not allowed access to feed between
0830 h to 1030 h. The amount of feed offered and orts were weighed for each cow daily.
On every infusion day, samples of all dietary ingredients (0.5 kg) were collected, and
cows were infused treatment solutions from 0830 h to 2230 h. Cows were milked twice
daily in the milking parlor except for the evening milking on infusion days, for which
cows were milked in their stalls. Feeding behavior was monitored from 1030 h to 2230 h
on each infusion day by a computerized data acquisition system (Dado and Allen, 1993).
Data of chewing activities, feed disappearance, and water consumption were recorded for
each cow every 5 sec, and meal bouts, interval between meals, meal size, eating time,
ruminating time, and total chewing time were calculated. Milk yield was recorded and
milk samples were taken at both milking on each infusion day. Milk samples were
analyzed for fat, true protein, lactose, solids—non-fat, milk urea nitrogen (MUN)

concentration with infrared spectroscopy by Michigan DHIA (East Lansing).

95

Diet ingredients were dried in a 55° C forced-air oven for 72 h and analyzed for
DM concentration. All samples were ground with a Wiley mill (1mm screen; Authur H.
Thomas, Philadelphia, PA), and analyzed for ash, NDF, ADF, CP, and starch. Ash
concentration was determined after 5 h oxidation at 500° C in a muffle furnace. The
NDF and ADF concentrations were determined [(VanSoest et al., 1991); method A for
NDF]. Crude protein was analyzed according to Hach et al. (1985). Starch was
measured by an enzymatic method (Karkalas, 1985) after samples were gelatinized with
sodium hydroxide, and glucose concentration was measured using a commercial kit
(Glucose kit #510; Sigma Chemical Co., St. Louis, MO). Concentrations of all nutrients
except for DM were expressed as percentages of DM determined by drying at 105° C in a
forced-air oven. Corn grain was dry sieved through 8 sieves (Sieve apertures: 4750,
2360, 1180, 600, 300, 150, 75 um and bottom pan), using a sieve shaker (Model RX-86,
W.S. Tyler Inc., Gastonia, NC) for approximately 20 min until the bottom pan weight

was constant, and mean particle size of corn grain was calculated (ASAE, 1968).

All data except for MUN were analyzed using the fit model procedure of JMP®

according to the following model:

Yijkl = u + Si + C(S)jm + Pk + T1 + CovINF + eijklm
where

u = overall mean,

Si = fixed effect of square (i = 1 to 2)

C(S)jm = random effect of cow nested in a square (j = 1 to 8),

96

Pk = fixed effect of period (k = 1 to 4),
T1 = fixed effect of treatment (1 = 1 to 4),
CovINF = effect of actual amount of solution infused into the rumen, and

eijkl = residual, assumed to be normally distributed

One pump was used for a square of 4 cows, and random effect of cow was nested in a
square that shared the same infusion pump. Interactions of square x treatment and period
x treatment were originally evaluated, but they were removed from the statistical model
because interactions were not significant for response variables of interest. Volume of
solution infused into the rumen was included in the statistical model as covariate.
Orthogonal contrasts were made for the effect of VFA type (acetate vs. propionate), salt
type (sodium vs. ammonium), and interaction of VFA and salt. Treatment effects and
their interaction were declared significant at P < 0.05 and P < 0.10, respectively, and
tendency for treatment effects and their interaction were declared at P < 0.10 and P <

0.15, respectively.

RESULTS AND DISCUSSION

Volume of solution infused into the rumen with sodium treatment was greater
than ammonium treatment (P < 0.001; Table 2). This was unexpected because a 4-

channel peristaltic pump was used for each square with identical tubing and fittings, but a

97

 

 

 

 

2d 5d v Ed od :4 ms fim H6 Csofiv .33 x55
mmd mdd mdd v 3: ddd mdm Q: m.mm 9:5 LSCSE waver—Q
Ndd Nod 8d v ad ad WS 5.: “.2 ENS amazon 9.3:th
Sd v 58.0 V dd v dd wdm fiww vdm odw Emir—V 83E .83?
8323 mew—5.5
mmd wdd vmd Nd wém d.mm mSN Ndm QED EBB:
3d v and Sod v am dmd gm mom 2m ENQEEV :33.
2 d 3d 3d a; mdfi 2: ad d dd 935 @355
5d v wmd “ddd V mm mm v3 2N Ema ENCEEV wanna—«85M
mod 3d 3d wd m.m_ QQ 0: fl: QED $355
5d V dd v Sod v om mm mm d On H N: ENCEEV madam
0:5 wagonu
5d v Sod v Sood v Wm 02 mfim fiwm NS ENSSSC =38.
ddd odd fidddd V dd wen we 9N Wm 2393—): .8833
3d v Sod v Sodd v Wm w.: w.mm Wmm 5:4 flags—2V 85
835 3.85 03328302
Nbd mmd 5d v md m; 54 fim Wm 95 mod 36 ~32
mud 3d mvd dei mod mdb wdw 0.3 8:5 EESE 32585
5d V dd v mod wd dd NB dd d5 ENS anon :32
8d V Sodv dddd v m; mé fig fimfi mwfi Emtwod H35
8323 wEdoom
Ed Edd v Sud dd mm“ m.m d m4; v.2 3d @835 2:29 88%
MHZH anew _<m> mm Esaog 833m 6:80:82 shadow
Bacoaoi 3303..
2:? a $5830;

 

.5326on 658 new 5333 wfinoom :0 £3. <m> do command EEEEAEE .«0 303mm .N mafia.

98

806 98 3:335 8on 8:35 $85 0332382: :38 “835 m2 a
Add d v a: “Swami—ma 33 830335 5253: x cocoa ..

69$ :3 new 09$ <m> 5923 88885 do muootm m
8:805 .m> 838m ”25 :3 90 88am?
03:28.5 .m> 2803. ”on? <m> do 38:?

 

 

 

 

mod 35 :3 v 2 .o a: 86 So So 33:83: .335 m2

H 2533: 33:0 #52
Mad :3 v :3 v S 3: mam a. a nmm @386 E55 5:2
”8 Se v m; 2 EN 4.» EN 2 gamed Emcee 8:. a3
8.0 t .o 2.0 t .o 8.x «3 on.” 22 GE "Em é:
98 So was :3 NS. $4 $4 m? 33 3.902 é:
2 .o 35 Rd 85 2 .m mom 2 .m 2 .m es 53% £3
:5 as mg 86 a? mnm 9% 3m GE 5 s52
3o 86 :3 v E 4.8 3m 2m 3% 63V 22% 53

8:26on M32
Ea 2% 33 mm 62885. 828m 82888< 828m
oumcoaoi 3.309%
on?» a 8:253;

 

.8253 N mdmaed.

99

difference in viscosity of infusates might have affected flow rate. Therefore, volume of

solution infused into the rumen was included in the statistical model as a covariate.

DMI

Interactions of main effects were observed for DMI (P < 0.01). Infusion of
ammonium propionate decreased DMI by 65% compared to sodium propionate (4.3 vs.
12.1 kg/12h), but reduction in DMI for ammonium acetate compared to sodium acetate
treatment was only 14% ( 13.1 vs. 15.3 kg/12h). Although propionate treatments contain
more energy in infusates compared to acetate treatments, total ME intake was decreased
by infusion of propionate compared to acetate. In addition, the total ME intake reduction
by ammonium infusion was much greater when infused with propionate compared to

when infused with acetate.

The hypophagia caused by infusion of ammonium propionate cannot be attributed
to ammonium toxicity in this experiment because the same amount of ammonium did not
decrease DMI to a similar extent when infused with acetate. It is more logical to
speculate that propionate exacerbated the hypophagic effects of ammonium by
stimulating oxidative metabolism in the liver. As discussed previously, urea synthesis
requires amino acid catabolism for a source of N and increase availability of amino acid
carbon to be either oxidized or utilized for gluconeogenesis in the liver. However,
maximum rate of gluconeogenesis at any point in time is affected by enzyme activity
regulated by hormones such as insulin and glucagon. When ammonium acetate was

infused, carbon from amino acids that provided N for urea synthesis might have been

100

utilized for gluconeogenesis to a greater extent compared to oxidation in the liver because
of a relative lack of glucose precursors. However, when ammonium propionate was
infused, amino acid carbon was probably oxidized extensively in the liver because
propionate also serves as substrate for gluconeogenesis in the ruminant liver.
Concentration of ATP in the liver is expected to be greatest for ammonium propionate
treatment compared to other treatments. Our working hypothesis that ATP production in
the liver affects DMI and feeding behavior may explain the strong interaction between

ammonia and propionate observed for DMI in this experiment.

Our speculation is also supported by observations of Dhiman et al., (1993).
Dhiman et al. (1993) reported abomasal infusion of glucose at 1 kg/d decreased DMI by
18% compared to control for cows fed alfalfa silage at 98% of dietary DM (18.5% dietary
CP) and that blood urea nitrogen concentration was greater than 30 mg/dl regardless of
treatments. It is contrary to observation of Frobish and Davis (1977) in which abomasal
infusion of glucose at 2.15 kg/d did not decrease DMI for cows fed a diet containing 60%
concentrate. Although dietary CP and blood urea nitrogen concentrations were not
reported by Frobish and Davis (1977), ammonium flux to the liver is expected to be
significantly lower for their experiment because of greater fermentability of the diet
compared to that in the experiment of Dhiman et al. (1993). Inconsistent hypophagic
effects of glucose can be attributed to the interaction of glucose infusion with ammonium
flux to the liver because glucose per se does not have hypophagic effects in ruminants

(Allen, 2000). Enhanced urea synthesis might have stimulated oxidative metabolism to a

101

greater extent with abomasal infusion of glucose because glucose infusion is expected to

decrease gluconeogenesis in the liver.

Dhiman et al. (1993) also reported that infusion of glucose with soy protein did
not decrease DMI and this observation appeared to be inconsistent with the discussion
above. However, lack of hypophagic effect of glucose when infused with soy protein can
be explained by the difference in glucose demand of the mammary gland and subsequent
gluconeogenesis in the liver. Infusion of glucose with soy protein increased milk yield
by 6.1 kg/d compared to infusion of glucose alone. Milk production might have been
limited by availability of essential amino acids when cows were infused with glucose
only, but not when infused with glucose and soy protein. This difference in milk
production increases glucose demand by 435 g/d according to the calculation method of
Amaral-Phillips et al. (1993), and this is equivalent to 44% of the infused glucose. Thus,
the gluconeogenesis is expected to be greater for cows infused with glucose and soy
protein compared to cows infused with glucose only, decreasing oxidative metabolism of

amino acid carbon generated by the urea cycle.

The explanation for the interaction of main effects observed in our experiment
might be challenged because propionate can decrease hepatic capacity to detoxify
ammonia by inhibiting the synthesis of N-acetyl glutamate, the activator of carbamoyl
phosphate synthetase (Choung and Chamberlain, 1995). Choung and Chamberlain
(1995) showed that intra-ruminal infusion of propionate and urea markedly increased

plasma ammonia concentration compared to infusion of urea alone in dairy cows.

102

Mutsvangwa et al. (1997) evaluated the effect of propionate on urea production with
isolated hepatocytes prepared from sheep, and found that urea production was decreased
by half when hepatocytes were incubated with 1.25 mM of propionate and 1.25 mM of
NH4Cl compared to when they were incubated with 1.25 mM of N H4Cl alone. However,
in our experiments, the ammonium propionate treatment increased MUN to the same
extent as the ammonium acetate treatment. Because urea equilibrates within body fluids,
MUN is highly correlated with blood urea nitrogen and is an indicator of urea production
by the liver (Bulter et al., 1996). Our data show that infusion of ammonium increased
MUN synthesis similarly for each VFA type infused, indicating that propionate did not
inhibit urea production in the liver substantially. Furthermore, although hepatocytes
decreased urea production in vitro when incubated with propionate (Mutsvangwa et al.,
1997), urea production increased four fold when hepatocytes were incubated with 1.25
mM of propionate and 1.25 mM of NH4C1 compared to control (no NH4C1), indicating

that propionate did not inhibit the urea cycle completely.

Feeding Behavior

Infusion of ammonium decreased meal frequency and this effect was greater
when infused with propionate compared to acetate. Ammonium propionate decreased
number of meal bouts by 46% compared to sodium propionate treatment (3.9 vs. 7.2
/12h), but ammonium acetate treatment decreased number of meal bouts by only 6%
compared to sodium acetate treatment (6.6 vs. 7.0 /12h). However, meal size was not
affected by ammonium treatment. This is consistent with observations from the previous

experiment (Chapter 4), in which infusion of ammonium propionate decreased meal

103

frequency without affecting meal size compared to sodium and potassium propionate.
These observations indicate that ammonium infusion does not result in satiety sooner, but
that hunger is delayed. Our data did not provide conclusive evidence that ammonium
exerts its hypophagic effect by oxidative metabolism of carbon from amino acids in the
liver because ATP concentration in the liver was not determined in this experiment.
However, sustained satiety or delayed hunger observed for ammonium treatments is

consistent with our theory.

A possible explanation for the lack of ammonium effect on meal size is that urea
synthesis occurs over time. The liver is a heterogeneous organ varying in enzyme
activity between periportal and perivenous regions. Urea synthesis occurs in periportal
hepatocytes and at physiological portal concentrations of ammonia, about two thirds of
ammonia are incorporated in urea while glutamine synthesis in perivenous hepatocytes
scavenges the remaining ammonia (Haussinger et al., 1992). Glutarnine synthesis
minimizes ammonia escaping hepatic detoxification when ammonium flux to the liver
exceeds rate of urea synthesis; Rodriguez et al. (1997) showed that a diurnal variation in
ruminal ammonia concentration is 10 times greater than that for plasma urea nitrogen
concentration. Ammonia incorporated in glutamine is available for later urea synthesis
because of significant activity of glutaminase in periportal hepatocytes (Haussinger et al.,
1992). Nitrogen in carbamoyl phosphate comes from either ammonia absorbed from the
gut or ammonia from glutamine. Generation of ammonia from glutamine in periportal
hepatocytes allows urea synthesis to continue over time after meals. Ammonium

infusion decreased DMI by delaying the sense of hunger without affecting satiety

104

possibly because urea synthesis occurs over time after meals providing carbon from
amino acids for oxidative metabolism in the liver for an extended period. Total urea
synthesis per day, estimated from MUN concentration, was similar for ammonium
acetate and ammonium propionate treatments. Thus, decreased rate of urea synthesis by
propionate (Choung and Chamberlain, 1995) following meals probably extended urea
synthesis for propionate compared to acetate treatment over a longer period of time

increasing intermeal interval.

Chewing Behavior

Eating time was decreased by infusion of propionate (P < 0.001) and ammonium
(P < 0.01) compared to acetate and sodium, respectively. In addition, reduction in eating
time was greater when ammonium was infused with propionate compared to acetate.
Ruminating time was shorter for ammonium treatment compared to sodium treatment
when infused with propionate (75 vs. 124 min), and this might be explained by decreased
stimuli for chewing by the decreased digesta mass in the rumen from lower DMI.
However, sodium treatment decreased ruminating time compared to ammonium
treatment when infused with acetate (137 vs. 211 min), and ruminating time per kg DMI
was lower for sodium treatment compared to ammonium treatment (P < 0.04). This
might be attributed to the expected greater osmolarity of ruminal fluid for sodium
treatment compared to ammonium treatment because of ammonium incorporation into
microbial N. Greater osmotic pressure in ruminal fluid can decrease rumination (W elch,
1982). Water intake (P < 0.0001) and number of drinking bouts (P < 0.02) were greater

for sodium treatments compared to ammonium treatments consistent with treatment

105

effects on ruminating time per kg DMI. An interaction of main effects was significant for
water intake (P < 0.01), indicating that water intake was lower for ammonium propionate

compared to ammonium acetate treatment while VFA type did not affect water intake for

sodium treatments. This interaction can be attributed to lower DMI for ammonium

propionate treatment compared to the others.

Milk Production

An interaction of main effects was observed for milk yield (P < 0.06) and milk fat
concentration (P < 0.01). Ammonium treatment decreased milk yield compared to
sodium treatment to a greater extent when infused with propionate (27.4 vs. 32.9 kg/d)
compared to when infused with acetate (32.1 vs. 34.0 kg/d). Ammonium treatment
decreased milk fat concentration (3.42 vs. 3.99 %) compared to sodium treatment when
infused with acetate, but infusion of ammonium increased milk fat concentration (3.79
vs. 3.53 %) compared to sodium treatment when infused with propionate. Milk energy
output was lower for propionate compared to acetate treatment (P < 0.01) and for
ammonium compared to sodium treatment (P < 0.01), and these reductions in milk
energy output can be attributed to lower ME intake. Milk lactose concentration was
greater for sodium treatment compared to ammonium treatment (P < 0.05) regardless of
VFA treatments. Milk lactose concentration is usually constant because milk fluid is
synthesized by osmotic pressure of lactose in the Golgi apparatus, and osmolarity of milk
is similar to that of blood (Halt, 1983). Increased milk lactose concentration might

indicate that osmolarity of blood was increased (W heelock et al., 1965) for sodium

106

treatments compared to ammonium treatments although blood osmolarity was not

measured for this experiment.

CONCLUSION

Infusion of ammonium decreased DMI compared to sodium to a greater extent
when infused with propionate compared to when it was infused with acetate. Reduction
in DMI from ammonium infusion is attributed to decreased meal frequency not smaller
meal size, indicating that ammonium delayed the sense of hunger. Urea synthesis
requires N from amino acids, and deamination of amino acids increases carbon available
for oxidation or gluconeogenesis in the liver. When ammonium was infused with
propionate, increased oxidative metabolism and production of ATP in the liver are
expected because rate of substrate supply for gluconeogenesis is more likely to exceed its
rate of utilization. These results are consistent with the hypothesis that ATP production
in the liver can affect feeding behavior and DMI. However, hepatic ATP concentration
was not analyzed, and further research is needed to investigate the mechanism of feed

intake regulation by propionate.

107

CHAPTER 6

Dose-response effects of intra-ruminal infusion of propionate on feeding

behavior of lactating cows fed diets differing in fermentability

ABSTRACT

Two experiments were conducted to evaluate how dose-response effects of intra-
ruminal infusion of propionate on feeding behavior and DMI are altered by diets differing
in fermentability. Twelve ruminally—cannulated Holstein cows (99 i 25 and 53 i- 21 days
in milk, respectively for experiment 1 and 2; mean 1 SD) were used in each experiment.
Cows were fed diets containing either steam flaked corn or dry cracked com (30% of
dietary DM) in experiment 1, and cows were fed diets differing in forage to concentrate
ratio (66:34 vs. 36:64) in experiment 2. For both experiments, experimental design was a
crossover for dietary treatment, and a 6 x 6 Latin square for infusion treatment within a
diet for each period. Infusion treatments were mixtures of sodium propionate and sodium
acetate, at ratios of 0:5, 1:4, 2:3, 3:2, 4:1 and 5:0, infused into the rumen continuously for
18 h starting 6 h before feeding at a rate of 23.1 mmol of sodium VFA/min. Propionate
infusion decreased DMI for all dietary treatments. Although a difference in basal
propionate production between diets was expected within each experiment, an interaction
of main effects was not observed for DMI in both experiments. This indicates that

propionate flux through the rumen per se does not generate satiety signals. Cows used in

108

experiment 1 decreased DMI and increased plasma glucose concentration linearly as
propionate infusion increased. However, cows used in experiment 2 did not decrease
DMI by lower rates of propionate infusion which were much more effective at increasing
plasma glucose concentration (quadratic effect P < 0.01). It is speculated that propionate
had less hypophagic effects when infused propionate was extensively utilized for
gluconeogenesis but decreased DMI when the marginal effect of infused propionate on
plasma glucose concentration decreased because propionate is oxidized in the liver unless
it is utilized for gluconeogenesis. Propionate decreases feed intake in lactating dairy
cows possibly by stimulating oxidative metabolism in the liver.

(Key words: propionate infusion, threshold response, diet fermentability, fill)
Abbreviation Key: SF = Steam Flaked Corn; DC = Dry Cracked Corn; HF = high

forage diet; LF = low forage diet

INTRODUCTION

Feeding fermentable grains in diets sometimes decreases DMI in lactating dairy
cows (Allen, 2000), and hypophagic effects of propionate have been reported (Choi and
Allen, 1999; Chapter 3). However, some experiments in the literature have reported no
effects of propionate infusion on feed intake (Deetz and Wangsness, 1981; Quigley and
Heitmann, 1991; De Jong et al., 1981; Anil et al., 1993). Inconsistent hypophagic effects
of propionate might be explained by a threshold response of propionate in feed intake

regulation; infusion of propionate might not affect feeding behavior and DMI unless the

109

amount of infused propionate exceeds a threshold. Dose response effects of propionate
on feed intake were previously investigated for lactating dairy cows (Anil et al., 1993;
Chapter 3) and sheep (Farningham and Whyte, 1993), and infusion of propionate linearly
decreased feed intake as infusion rate of propionate increased. A threshold response in

DMI was not observed for those experiments.

Fermentability of experimental diets may affect the threshold for infused
propionate to decrease DMI. In a review of the literature (Allen, 1997), amount of
ruminally fermented organic matter and total VFA production were reported to range
from 5.7 to 15.4 kg/d and 42 to 115 moles/d for lactating dairy cows, respectively.
Because propionate concentration in the rumen can increase from 15 to 45% of total
fermentation acids as amount of ruminally fermented OM increases (Davis, 1967),
propionate production can range from 6.3 to 52 moles/d. Lack of a threshold response
for infused propionate on DMI in the experiment reported by Farningham and Whyte
(1993) might be because sheep were fed a very fermentable diet ad libitum containing
50% hay, 30% barley, and 10% molasses, and propionate production from diets might
have already exceeded the threshold. However, Leuvenink et al. (1997) fed sheep a
pelleted grass, and reported that propionate infusion into the mesenteric vein of mature
sheep at a rate of 2 mmol/min decreased intake but the infusion at a rate of 1 mmol/min
had no effect. Fermentability of diets can be altered by feeding grains differing in
fermentability in the rumen or by feeding diets differing in forage to concentrate ratio.
We hypothesized that fermentability of diets affects animal responses to intra-ruminal

infusion of propionate in feeding behavior and DMI and that cows fed more fermentable

110

diets decreases DMI at lower doses of propionate compared to cows fed less fermentable

diets.

In a previous study (Chapter 3), in which dose-response effect of propionate on
DMI was evaluated, a threshold response was not observed although we tried to minimize
fermentability of the experimental diet by feeding dry cracked corn that is poorly
fermented in the rumen. In that study, propionate was dosed at 8 different rates of
infusion from 0 to 16.7 mmol/min in experiment 1 or at 4 different rates of infusion from
0 to 25.0 mmol/min in experiment 2. Marginal reduction in DMI was numerically greater
as infusion rate of propionate increased in both experiments, but quadratic effects of
propionate infusion were not detected. Range of propionate infusion might not have been
great enough for experiment 1 (Chapter 3) and increments of treatments might not have
been sufficient for experiment 2 (Chapter 3) to detect the quadratic effects of propionate
infusion on DMI. We expected to observe a threshold response to infused propionate on
DMI using an appropriate experimental design with a wider range and sufficient

treatment increments of infusion rates.

The objective of this experiment was to evaluate how dose-response effects of

intra-ruminal infusion of propionate on feeding behavior and DMI are altered by diets

differing in fermentability.

111

MATERIALS AND METHODS

Experimental procedures were approved by All University Committee on Animal

Use and Care at Michigan State University.

Experiment 1

Twelve multiparous Holstein cows (99 i 25 DIM; mean i SD) cannulated
ruminally for previous experiments were selected from the Michigan State University
Dairy Cattle Teaching and Research Center. Experimental design was a crossover for
dietary treatments and a 6 x 6 Latin square within a diet for each period. Experimental
diets contained either steam flaked corn (SF) or dry cracked corn (DC) at 30% of dietary
DM. Both corn grains were obtained from Pennfield Feeds (Lancaster, PA). Flake
density of SF was 0.36 kg/L, and the mean particle size of DC was 3.7 mm. Both
experimental diets contained corn silage (50% of forage DM), alfalfa silage (50% of
forage DM), cottonseeds, a premix of protein supplements (soybean meal, distillers
grains, and blood meal), and a premix of minerals and vitamins (Table 1). Dietary NDF
and CP concentrations were approximately 27.8 and 16.7%, respectively for both diets,
and fed as total mixed rations. Throughout the experiment, cows were housed in tie-
stalls, and fed once daily at 110% of expected intake. Periods were 34 din length, and
each period consisted of 20 d for diet adaptation, 3 d for data and sample collection to
determine effects of dietary treatments, and 11 d for data and sample collection to

determine effects of infusion treatments.

112

TABLE 1. Ingredients and nutrient composition of experimental diets in experiment 1 (%
of dietary DM except for DM).

 

SFl DC2
Diet Ingredients
Steam flaked corn 29.7
Dry cracked corn 29.6
Corn silage 23.5 23.6
Alfalfa silage 22.4 22.4
Whole linted cottonseed 6.7 6.7
Protein mix3 12.6 12.7
Vitamin & mineral mix4 5.0 5.1
Nutrient Composition
DM 50.5 50.7
OM 94.3 94.1
Starch 32.1 29.9
NDF 27.6 27.9
ADF 19.5 19.5
CF 16.5 16.9
Ether extract 4.7 5.4
Forage NDF 19.9 20.0

 

lDiet containing steam flaked corn

2 Diet containing dry cracked corn

3 Protein mix contained 75 % soybean meal, 20% distillers grain, and 5% blood meal.

4 Vitamin & mineral mix contains 66.4% dry ground corn, 20.4% dicalcium phosphate,
7.8% salt, 2.4% magnesium oxide, 1.9% trace mineral premix, 0.34% vitamin A, 0.29%
vitamin D, and 0.08% vitamin E.

113

Experiment 2

Twelve multiparous Holstein cows (53 i 21 DIM; mean i SD) cannulated
ruminally for previous experiments were selected from the Michigan State University
Dairy Cattle Teaching and Research Center. Experimental design was a crossover for
dietary treatments and a 6 x 6 Latin square within a diet for each period. Experimental
diets differed in forage to concentrate ratio; 66:34 for high forage diet (HF; Table 2) and
36:64 for low forage diet (LF). Both diets contained corn silage, alfalfa silage,
cottonseeds, a premix of protein supplements (soybean meal, distillers grains, and blood
meal), and a premix of minerals and vitamins). The primary difference in diets was
substitution of corn silage and distillers grain in the HF diet for dry ground corn in the LF
diets. Distillers grain was used to compensate the lower CP concentration for corn silage
compared to dry ground corn so that diet contained similar CP concentration with similar
amino acid profiles. Dietary NDF and starch concentrations were 34.0 and 21.3 % for
HF and 25.2 and 35.1 % for LP, respectively. Throughout the experiment, cows were
housed in tie-stalls, and diets were fed as total mixed rations once daily at 110% of
expected intake. Periods were 35 d in length, and each period consisted of 21 d for a diet
adaptation, 3 d for data and sample collection to determine effects of dietary treatments,

and 11 d for data and sample collection to determine effect of infusion treatments.

Common Infusion Protocol
Infusion treatments were continuous intra-ruminal infusion of mixtures of sodium
propionate and sodium acetate at 6 different ratios. Cows were assigned to a 6 x 6 Latin

square balanced for carry—over effects. Treatment solutions were prepared by diluting

114

TABLE 2. Ingredients and nutrient composition of experimental diets in experiment 2 (%
of dietary DM except for DM).

 

 

HFl LF2
Diet Ingredients
Corn silage 53.5 24.4
Alfalfa silage 12.1 12.0
Dry ground corn 34.3
Protein mix3 18.0 17.7
Distillers grain 4.7
Whole linted cottonseed 6.7 6.6
Vitamin & mineral mix“ 5.0 5.0
Nutrient Composition
DM 44.1 56.0
OM 94.0 94.8
Starch 21.3 35.1
NDF 34.0 25.2
ADF 23.0 16.8
CP 17.6 17.0
Ether extract 6.0 4.9
Forage NDF 26.3 15.4
‘ High forage diet
2Low forage diet

3 Protein mix contained 75% soybean meal, 20% distillers grain, and 5% blood meal.
“Vitamin & mineral mix contains 66.4% dry ground corn, 20.4% dicalcium phosphate,
7.8% salt, 2.4% magnesium oxide, 1.9% trace mineral premix, 0.34% vitamin A, 0.29%
vitamin D, and 0.08% vitamin E.

115

28.1 moles of sodium VFA (sodium propionate and sodium acetate at ratios of 0:5, 1:4,
2:3, 3:2, 4: 1, and 5:0) to 18 L with de-ionized water. Sodium acetate was added to keep
the osmolarity and pH of infusates constant across the treatments to isolate the specific
effects of propionate relative to acetate on feeding behavior of dairy cows. Concentration
of total VFA was 1.56 M across treatments, and 16 L of each solution was infused over
18 h beginning 6 h prior to feeding. Infusion rate was 14.8 ml/min, which is equivalent
to infusion of 23.1 mmol of VFA/min. The solutions were infused using 4-channel
peristaltic pumps (#78016-30, Cole-Farmer Instrument, IL) and Tygon® tubing (7.5 m x
1.6mm I.D.). Infusion started 6 h prior to feeding so that VFA concentrations in the
rumen reached steady state (assuming absorption rate and passage rate of 20 %/h and 15
%lh, respectively) by feeding time, at which monitoring for feeding behavior began.
Sub-periods for infusion treatment were 2 d with 18 h of infusion followed by 30 h of

recovery.

Data and Sample Collection

The amount of feed offered and orts were weighed for each cow daily during the
collection period. Samples of all dietary ingredients (0.5 kg) were collected daily during
each 3 d collection period and on feeding behavior monitoring days during each infusion
period ((1 1, 3, 5, 7, 9, and 11) and composited to one sample per diet period. Samples of
orts (12.5%) were collected daily during the 3d-collection period and composited into
one sample per cow per period. Body weight and BCS were measured [(Wildman,
1982); five-point scale where 1 = thin to 5 = fat] on d 23 of each period. Cows were

milked twice daily in the milking parlor except for the evening milking for days in which

116

feeding behavior was monitored ((1 1, 3, 5, 7, 9, and 11 of each infusion period) when
cows were milked in their stall. Milk yield was measured daily during the 3 d collection
period and was averaged to determine effects of dietary treatments. Milk yield was also
measured on feeding behavior monitoring days to determine effects of infusion
treatments. Milk was sampled at every milking during the 3 d collection period and d 1,
3, 5, 7, 9, and 11 of the infusion period, and analyzed for fat, true protein, lactose, solids-

non-fat with infrared spectroscopy by Michigan DHIA (East Lansing).

Samples of feces, ruminal fluid, and blood were collected every 9 h during the 3 d
collection period. Ruminal fluid samples were collected from 5 different sites in the
rumen and squeezed through a nylon screen, and pH was determined immediately after
collection. Samples were frozen at -20° C until further analysis. Blood samples were
collected from coccygeal vessels into two VacutainerTM tubes (Becton Dickinson,
Franklin Lakes, NJ): one with sodium heparin and the other with potassium oxalate and
sodium floride as a glycolytic inhibitor. Both were centrifuged at 2,000 x g for 15
minutes immediately after sample collection, and plasma was harvested and frozen at -

20° C until analysis.

On feeding behavior monitoring days ((1 1, 3, 5, 7, 9, and 11 of infusion period),
infusion started at 0800 h, 6 h prior to feeding, and continued for 18 h. Cows were not
allowed access to feed between 1000 h to 1400 h to minimize the confounding effects of
ruminal fermentation from previous feeding. Feeding behavior was monitored for 12 h

(1400 h to 0200 h) by a computerized data acquisition system (Dado and Allen, 1993).

117

Data of chewing activities, feed disappearance, and water consumption were recorded for
each cow every 5 sec, and meal bouts, interval between meals, meal size, eating time,
ruminating time, and total chewing time were calculated. At the end of the feeding
behavior monitoring period (0200 h), samples of ruminal fluid were collected from 5
different sites in the rumen for each cow. Blood samples were collected from coccygeal
vessels, centrifuged at 2,000 x g for 15 minutes, and plasma was harvested and frozen at -

20° C until analysis.

Sample and Statistical Analysis

Diet ingredients, orts, and fecal samples were dried in a 55 °C forced-air oven for
72 h and analyzed for DM concentration. All samples were ground with a Wiley mill
(1mm screen; Authur H. Thomas, Philadelphia, PA). Samples were analyzed for ash,
NDF, ADF, CP, and starch. Ash concentration was determined after 5 h oxidation at
500° C in a muffle furnace. Concentrations of N DF and ADF were determined
[(VanSoest et al., 1991); method A for NDF]. Crude protein was analyzed according to
Hach et al. (1985). Starch was measured by an enzymatic method (Karkalas, 1985) after
samples were gelatinized with sodium hydroxide, and glucose concentration was
measured using a commercial kit (Glucose kit #510; Sigma Chemical Co., St. Louis,
MO). Indigestible NDF was estimated as NDF residue after 120 h in vitro fermentation
(Goering and Van Soest, 1970) and used as an internal marker to calculate apparent total
tract digestibility (Cochran et al., 1986). Concentrations of all nutrients except for DM
were expressed as percentages of DM determined from drying at 105° C in a forced-air

oven. Corn grain was dry sieved through 8 sieves (Sieve apertures: 4750, 2360, 1180,

118

600, 300, 150, 75 ,um and bottom pan), using a sieve shaker (Model RX-86, W.S. Tyler
Inc., Gastonia, NC) for approximately 20 min until the bottom pan weight was constant,

and mean particle size of corn grain was calculated (ASAE, 1968).

Ruminal fluid samples were analyzed for VFA concentrations according to the
method described previously (Chapter 2). Plasma samples were processed to determine
concentrations of acetate and propionate in the similar manner as described for ruminal
fluid (Chapter 2). Commercial kits were used to determine plasma concentration of
glucose (Glucose kit #510; Sigma Chemical Co., St. Louis, MO) and insulin (Coat-A-

Count, Diagnostic Products Corporation, Los Angeles, CA).

For both experiments, all data for dietary effects from 3 d of collection period

were analyzed using the fit model procedure of JMP® according to the following model:

Yijklm = u + Si + C(S),0, + Pk + Dl + eijklm
where
[.1 = overall mean,
Si = fixed effect of diet sequences (i = 1 to 2)
C(S),0, = random effect of cow nested in a diet sequence (j = l to 12),
Pk = fixed effect of periods (k = 1 to 2),
Dl = fixed effect of diets (l = 1 to 2), and

e = residual, assumed to be normally distributed.

ijklm

119

Orthogonal contrasts were made for the effect of diets (SF vs. DC or HF vs. LF).
Treatment effects were declared significant at P < 0.05, and tendency for treatment

effects was declared at P < 0.10.

All data for infusion effects from 11 d of infusion period were analyzed using the
fit model procedure of JMP® according to the following model:
Yijklmnop = u + Si + C(S),0, + Pk + SP(P).(,) + Dm + Ln + Qo + DLmn + DQmo + eijklmnop
where

p. = overall mean,

Si = fixed effect of diet sequences (i = 1 to 2)

C(S),(j, = random effect of cows nested in a diet sequence (j = 1 to 12),

Pk = fixed effect of periods (k = 1 to 2),

SP(P),(k, = fixed effect of sub—periods nested in a period (1 = 1 to 6),

Dm = fixed effect of diets (l = 1 to 2),

Ln = linear effect of infusion

Qo = quadratic effect of infusion

DLmn = interaction between effect of diet and linear effect of infusion

DQmo = interaction between effect of diet and quadratic effect of infusion, and

e = residual, assumed to be normally distributed.

i jklmnop

Orthogonal contrasts were made for the effect of diet (SF vs. DC or HF vs. LF),

linear effect of infusion, quadratic effect of infusion, interaction between effect of diet

120

and linear effect of infusion, and interaction between effect of diet and quadratic effect of
infusion. Main treatment effects were declared significant at P < 0.05, and tendency for
treatment effects was declared at P < 0.10. Interaction effects were declared significant
at P < 0.10. Total expected number of observations (n) was 144, but 7 observations for
experiment 1 and 3 observations for experiment 2 were missing. Our infusion protocol
caused temporary metabolic alkalosis, and some cows experienced hypocalcernia and
hypokalemia. Administration of infusates was cancelled when cows had adverse effects

of previous infusion treatment.

Data from 3 d of collection period before infusion were analyzed to characterize
the animals used in experiment 1 and experiment 2, using ANOVA procedure of JMP®

according to the following model:

Y,,.=].L+E,+eij
where
u = overall mean,
Ei = fixed effect of experiment (i = 1 to 2; experiment 1 or experiment 2), and

e = residual, assumed to be normally distributed.

ii

The effect of experiment was declared significant at P < 0.05, and tendency was

declared at P < 0.10.

121

RESULTS AND DISCUSSION

Experiment 1

Dry matter intake was not affected by dietary treatment (19.1 vs. 19.3 kg/ 12 h,
respectively for SF and DC; Table 3), but milk yield was greater for cows fed SF
compared to cows fed DC (37.4 vs. 35.8 kg/d; P < 0.01). Cows fed SF had higher
apparent total tract digestibility of starch (95.2 vs. 77.3 %; P < 0.001), but lower total
tract NDF digestibility (16.7 vs. 28.3 %; P < 0.001) compared to cows fed DC. Ruminal
propionate concentration was greater (27.3 vs. 22.6 %; P < 0.001) and ruminal acetate
concentration was less (55.8 vs. 61.1 %; P < 0.001) for SF compared to DC treatment
although ruminal pH was not affected by treatments. Plasma insulin concentration was
greater for SF compared to DC treatment (12.3 vs. 9.3 uIU/ml; P < 0.001) while plasma
glucose concentration was not affected by dietary treatment. Cows fed SF increased milk
protein (3.07 vs. 2.91%; P < 0.001) and milk lactose concentration (4.71 vs. 4.64%; P <
0.01), but decreased milk fat concentration (3.25 vs. 3.64%; P < 0.05) compared to cows
fed DC. These observations are consistent with expected greater ruminal fermentation
and greater ruminal propionate production for the SF diet compared to the DC diet

although they were not directly measured.

Dry matter intake, meal size, and total ME intake were decreased linearly by
increasing rate of intra-ruminal infusion of propionate (P < 0.001; Table 4). Interactions
between diet and infusion effects were not significant for feeding behavior and DMI,

indicating that diet did not affect the response to propionate infusion contrary to our

122

 

 

.3. 9o mm as A323 .535
«no to 0% Sn 22mg 3820
moo So 86 go 3.5 «388$
86 mod 32 2: 9:5 282

«Ema—m

.2... 86 m2 8a .2352.
So :6 m3 EN oi cases;
.3... 86 w? w: E 93%?3
: .o to 8.: ed as ashram
r... 85 «3 mod 3% 225353
.2... no 92 EN GE 92288;
.2... to Se man 22 383.
m3 2 $2 38 9:5 <m> Bob
5. 8.0 2 .0 Se ma
cosficonbom HERE

2 .o 3 Se 9% ESQ 835 835
Ed m cm... 84 2223 Boy
cod m MNN ca 2355 8:283.
r. m SN 5: £235 88m
0:5 wEBonU

m3 2 we. we. 28%;: 23:: m:
56 3 mm mm 95 was as :82
£3 3. a; cam :55 .325 38:25
and 3 no we ENS 9:8 as:
$8 to 32 $2 £38: :26

8323 wagon

 

02w>i Mm NUQ Em

 

.888on see Ea 255586
52.33: $320332: UGO—D .GOUNEOEDM Edge Gog—Eon wcmfioom GO Edam ECU we ©0508 mammmuoOuQ mo muootm .m m—Agrfi

123

:56 V R .13..

5.0 V R **

mod v a: ...

88:: have: 03:33:88 mm: m

ova: 832:8: 9 8884:. v

.82: .8 3:33:03: 88:: so 3:3 Good UM Z 8 32:88: 38328 :85 3:26 03323802 m
58 vow—:8: :6 8:58.80 :05 N

”F8: 3me :88: 8:58.80 8R?

 

8.: 8.: 8.: 8.: mom
8.: : 8: 8: 8: 5:83 :8:
:: .: 8.: 8.: 8.: 22:83: nE3: H 8:83: 888 U83:
8.: 2 3.: 3.: 8:83: :88 Va3:
1...... 8.: :3: 8.: :8 ”:2: 83:
.2. 8.: 8.: :2. :8 3.28: :53:
1...... 8.: 8.: 8.: :8 528: Va3:
._. N: .: 8.: ma: :8 a: 83:
t. 3: :8. 4.8 88: 22: u:3:
538:8: x52

.. N: :8 :8 mm
8.: w: :8 m8 .8
.23.. MA mfih ”no £0.35
.1... 2 3: S: .52
._.._. :.: mm: 4:: 3:0
.2. :.: 3.: :8 35

:8 5588:: 88 82 88:8.

 

eggi um NUQ Em

 

8.68: m 59...:

124

 

 

 

 

 

ddd .33.. dvd v 3: ww md dd dd dd 5 mm md dd dd: :d: Edtac 8:88 88:?
ddd a... ddd V: dd: .mdd Cd mdd End ddd dd: ddd dmd wdd mdd :dd 88:.
d: .d ddd wdd d db dd dd md md :Vd dd dw md: dd dd mm 8:98:83:
mmd .12.. Ed w ::: dd: :m: :m: dm: 3: dd: dd: dd: dd: dd: dd: madam:
28:88
08:: $8525
b: d is. .2. 5.: mdd :dm ddm ddm v.:m m. : m ddd mdm ddd mdm :dm mdm :88.
ddd 5.... ddd ddd dd 5d dd Ev :8 dd dd dd dd dd dd Wm @8838:
d: d is. i. 5.: mdd ddd ddd ddd mdd ddd mdd ddd ddd :.:m :.:m ddd .85
A5582:
8::8: m2
:md .12.. :dd dd d: d.: d: :d dd dd Na: d.: :d :d dd dd 3.4:: was 8:: :83:
mmd ddd mdd dd :.wd :d: dd: ddd :dw m.:d d.:d :dd :.:d mdd mdw ddd 2:8:
388: 3258::
wed ddd mmd wd md dd 5; md dd :d dd dd wd dd dd dd Ed:\: :82: :83:
:dd 8.... 3d dd 5d d. :: 5:: :d: dd: dd: dd: 5:: dd: 5m: m.m: we: Edtwvd ::>:n:
8:289: deduce”:
wmd ddd w:.d dd :d: dd: md: dd: dd: dd: md: md: dd: dd: dd: md: 3: 88:8
28:9, :::8<
.0 .1: b ::: :: :: :8 :N : :2 :w :: :8 :N .:
magi Mm NOD Em

 

.88 88:8 .3: 8 :80 8:3,: :82: :8: :38 8.: 88:88:: :58
:8: 83:52: deduce: 8 888: 828: 8 2622 8:28:88 83:8: :0 888:: 8582-88: 8 88:8 o80d:o:-omofl .v mAm<H

125

:5de *3"
~O.OV& “.3."
modvm *

muomb Uflw moummfiuwfim 80H.“ OMNHGM %MHOGO o—DNNSODNHOE Racy .HmH—Z a

Eggoama

.Bacoioa cam 83an 8a 3:582 mnemd Ea vmcmd .«o bmmcow $85 8853“ 3:335 80¢ 8:35 $.85 oEwfiBpfio—Z m
.808 Mo 5:353me 32% no woman :88 DMZ 8 $6803 @82330 33 66 Bob 83:: 385 03323302 h
Samoaoa Ho .8835 Ragaébfi go “onto 383:0 o

3329on Mo :28qu Eageébfi mo Soto 30:5 n

56 Mo 5th V

CE moummafi 5 38:8 82235 m
58 @8680 5 $8388 65 N
58 Exam Edam wamcgcoo BR?

 

 

 

 

23:82?ng

3d * cod Sd mad 56 Ed nwd mwd and owd whd mod $6 93 :16 ”3533:
3.85 x32

omd mad * M: Qmm QVN oém canm cém QMN fimm wém Nam Ndm mdm mdm 9583: $55 M52
mmd wad mod 26 wow wow 3.x 50.x 3.x vfiw 3; 53w wed hww oww cod 3% mZm 5:2
omd mod one mod wwé mmé mwé Ré mbé mg» mwé 5% mud 56 Rd E: 3% 3802 x52
mad *3 * 86 SN :.~ 8N S.m mod cod 8d :.m S.m 3 .m 24m mmd $3 58on V52
mod omd mmd mad nod 86 56 w“ .v wmé 2 .v mud mw.m mud E .v 3d mod GE HE v52
omd owd in fio fimm Ném fimm fimm mam mdm Nmm 63H fiom Nwm Ymm tmm 8E5 203 M52
couosnoa x32

00 “A «G GS ow on ow ON 0 OS ow ow ow om mo
023i mm «DO 7mm

 

.8253 v mama?

126

hypothesis. Milk yield was not affected by infusion treatments although milk protein
concentration decreased linearly as rate of propionate infusion increased (P < 0.001).
This is consistent with previous observations (Chapter 3), and might be because
decreased DMI compromised microbial protein production in the rumen while additional
energy supplied in infusates sustained milk yield. Infusion of glucogenic energy as
propionate without additional amino acid supply may have diluted milk protein by
relatively greater lactose synthesis and milk yield when metabolizable protein limited
maximum milk yield. A significant interaction was observed between dietary treatment
and linear effect of infusion for ruminal propionate concentration at the end of infusion.
As rate of propionate infusion increased, ruminal propionate concentration increased
linearly (P < 0.001) for both diets. However, ruminal propionate concentration was
greater for SF compared to DC treatment at lower rates of propionate infusion while it
was similar for both treatments at higher rates of propionate infusion. This might be
because DMI decreased to a greater extent and ruminal fermentation of diets contributed
to ruminal propionate concentration to a lesser extent at higher rates of propionate
infusion. However, similar propionate concentration in ruminal fluid does not
necessarily indicate similar propionate production in the rumen because concentration is

determined by the rate of removal (absorption and passage) as well as rate of production.

Propionate infusion linearly increased plasma concentration of propionate,

glucose, and insulin, but decreased plasma acetate concentration (P < 0.001; Table 5).

Interaction for main effects was significant for plasma concentrations of propionate and

127

 

 

 

 

.1... .2... N3 cod 8; m3 3: a3 «3 and. N3 n? me: :.~ w? m? 8.362382.
.3 .0 *** * 00.0 0v; ~VA TVA 0N4 :4 :4 fin; mm; 0v; 0%; VNA w—A A050 8823/
M00 *** 5.0.0 50.0 w0.~ mm; 50; in; 0m; 0?; GVA hm; :ué 0*!— RNA VNA A050 832w>ém~
Ed .2... and no we 3 3 2 2 3w 3 we as E 2. as $00353
5Y0 3.3." mmd 00.0 No.0 ~00 05.0 00.0 00.0 00.0 wbd mw.0 00.0 N00 00.0 00.0 A950 Baubsntofl
“.3." 3% * w.0 00v 0.wm NAM m.0N NAN N00 0m? Ywm fimm m.wN ~.VN 00— EA§V 03:20on
t. I... ... he «.3 new 2m was QB was new mom new 08 one use a§Z§8<
00.0 w0.0 a .0 m NS be m: a: a: mm~ 0N~ mg mm“ mg mg Nag A290 <nm> Hach.
socssqbfl 35830

00 m..— vQ 000 0m 00 0v ON 0 00_ 0m 00 0v 0N m0

2527K mm NUQ Em

 

.58 @8330 be 8 58 @830 :8on 3.0 $58 .80 0003 05 was :25:
05 E moEonSoE no 0388 8:68 9 03:38 Bacofioa 8:68 Mo 83:05 HEEE$bE 00 38.09 omaommEémoQ .m mama?

128

80 0 v 5 832080 00 80.00 008930 05 Soho 36 00 509285 “52.0:me 8

80 .0 v “0 03:20on 00 88.3 Eve: 0%.. Soto 86 .3 800885 885:me an

"00.0 V R *3“
~0.0V& ”.3."
00.0VK *

08:20on 00 8635 Ragaébfi Mo 8000 008030 e
8820qu .8 H8635 3:0:8355 00 Some .8054 n

as 00 seem .

3% @8335 5 38:8 8320on m
88 08380 S wcmfificoo 85 N
88 083.0 Emma @8580 85.

 

 

 

 

000 "5.3.. :0 N00 wad m0; m0; 004 #0; m0; ~04 v0.0 00.0 84 >04 00.0 €65 ImU
wm0 Vm .0 * 00 N00 méo mayo 000 méa mdo Nana v00 v.8 Sum 0.8 5mm A280 .6
0V0 * * 5&0 000 wmd find “Wm end wim w0.m 0N.m wim 0v.m NYm mvd Dim 02:0 +v~
mm0 Vm0 mm .0 v0 02 03 a: 02 03 02 a: 03 03 02 a: 03 on 02.5 +«Z
ww0 0N0 mmd ~00 vmfi mmfl fins fins Vmfi mmfi umfi mmfi mms mmfi mmfi mms in
000 .12.. mN0 0.0 N00 m.: 0.: v.0 0.5. m6 0.m 0 m4: 0: v.9 50 0.0 on 39330.: 5:55
30 *3. I. w0 m.m0 0:00 m.~0 v. 00 w00 5.0m 000 500 0.m0 000 000 0.00 QEmEV 8820
* * * * * 0m0 N00 80 0N0 mm0 NNO ~N0 0N0 3H0 wm0 NNO NNO 30 50 on .SAEEV 8220on
* .13.. 30 20 :0 00.0 2.0 3; mm; SN $0 0w0 mo; 92 EA 000 02:0 888<
«835

$0 a: .b 000 0w 00 0v 0m 0 00— 0m 00 0v 0m m0

ofigi Mm NUQ Tam

 

.5380 m mam/a.

129

insulin; cows fed SF increased plasma concentrations of propionate and insulin to a
greater extent at higher rates of propionate infusion compared to cows fed DC. Greater
response to propionate infusion in plasma insulin concentration for SF compared to DC
treatment might be due to greater propionate absorption for SF because propionate
stimulates insulin secretion in ruminants (De Jong, 1982; Istasse et al., 1987). Propionate
infusion at 100% of VFA increased plasma propionate concentration by 62% compared
to propionate infusion at 0% of VFA for cows fed SF diet (0.21 and 0.34 mM,
respectively for O and 100% propionate) and for cows fed DC diet by 35% (0.20 and 0.27
mM, respectively for 0 and 100% propionate). These observations indicate that some
propionate escapes hepatic metabolism and reaches peripheral tissues although infused

propionate is extensively metabolized in the liver.

In experiment 1, no interactions between diet and infusions were detected for
feeding behavior or DMI, and we speculate that cows fed SF were more tolerant to
hypophagic effects of propionate. Rate of glucose clearance from blood circulation was
expected to be greater for SF treatment because of increased milk yield and greater
plasma insulin concentration. However, plasma glucose concentration was not affected
by dietary treatments during 3 d collection period before infusion, and was greater for SF
compared to DC treatment during the infusion period. This suggests that rate of
gluconeogenesis was greater for cows fed SF compared to cows fed DC possibly because
greater starch digestion in the rumen increased availability of glucose precursors.
Although propionate flux to the liver is expected to be greater for SF compared to DC

treatment, greater rate of gluconeogensis might decrease the relative proportion of

130

infused propionate oxidized in the liver for the SF treatment. Our results suggest that the
propionate flux to the liver does not relate directly to generation of satiety signals, but
implies that hypophagic effects of propionate are altered by how propionate is utilized in
the liver. Propionate may not exert hypophagic effects if it is utilized for
gluconeogenesis, but decreases feed intake by stimulating oxidative metabolism in the

liver.

Experiment 2

No effect of diet was observed on DMI and milk yield (Table 6). Although cows
fed HF had greater apparent total tract digestibility of NDF (25.6 vs. 18.5 %; P < 0.05),
CF (65.5 vs. 61.2 %; P < 0.01), and EE (84.7 vs. 67.9 %; P < 0.001) compared to cows
fed LF, apparent total tract OM digestibility was not affected by dietary treatments. This
is because HF treatment contained less starch, which is a highly digestible fraction of the
diet, compared to LP treatment. The LP treatment increased ruminal propionate
concentration (28.3 vs. 22.5%; P < 0.01) but decreased acetate concentration (55.0 vs.
60.2%; P < 0.001), and ruminal pH (5.87 vs. 6.04; P < 0.01) compared to HF. Plasma
acetate concentration was greater for HF compared to LP treatment (1.23 vs. 0.94 leI; P
< 0.01), but plasma concentrations of glucose and insulin were not affected by dietary
treatments. The HF treatment increased milk fat concentration compared to LP (3.67 vs.
3.14%; P < 0.01). These observations are consistent with our expectation of greater
ruminal fermentability and ruminal propionate production for LF compared to HF diets,

although they were not directly measured.

131

 

a .0 b0 v.2 0.0 95210 5:55

 

N8 2 0% in 23.5 38:6
m2 :3 m3 :6 9:5 058305
.2. 3o 88 m: 22,5 8902
«Emma
.. 2 .o SN 23 .88 .3.
._. Ed o2 N: 00 use;
2 .o 2 .o a: 53 $3 subarea
2 .o to 8.: ed 0% 288m
.. mod 33 mg 03 9353-0:
.2. 3 3a SN 00 22285
.2. ad own «.8 £8 338<
:3 3 E: 3.: A25 55 Bop
.3. mod :3 8.8 :8
:omuficogom 3580M
23 3. 3.8 38 £23 9.8: as?
“I. N; 00m 0.3V 0320.25 030%
.2. : 2: SN ESEE 358253.
.. w m: E adage macaw
080 $0325
38 2 3m 92 .. 22:85 0x55 82
mad S ta 3 95 we: 3; :82
m3 3 wow nfi :55 3.5:: Rogues
ad No 2 3 ES 3:8 :32
3o ed m: S; ESE: EB
8333 wfivoom

0%: mm N”: E...

 

.couoscoa 0:: 05
5250830 Bogs: £20382: 0003 $338258 12:82 .8333 wanton.“ so 008 886850 8 awake mo $8.0m .0 mqmflw

132

 

~86 V R “In."

_0.0 V R “.3."

mod v m *

8135 E35 2.333808 ”52 m

0:8 8229on 9 888< v

.306 Mo bfinumomfi 388 no woman :88 OM Z 9 @6808 833:8 $3 $85 oEwunonSoE m

 

86 omfifi 33 N

86 owfifi :wi.
cod mod oo.~ 9mm mum
So v a 0% 9: £303 38
2 .o wed $6 and ENSaoEv “Hm—E ” 6:33: $85 M52
$6 a; 0.8 fimm 8:85: >328 5:2
mod mod 3w f .w GE mZm x52
mod vod and and GE 8803 x22
2 .0 mod 2mm wed GE :8on “ES
in Q6 Ed $.m GE 5 i=2
So 2 fix a? 63V 22» Va:
cowosuoa x22
*1 NA 3.0 new mm
in ad NS fine no
mmd o4 .023 wda seam
* ON nwfi 9mm ”52
:3 ad ode how 20
wad ad Yam 3% 20

35V basswowmu 8a: 138 80.8%?

 

02w>i Mm Nu: RE

 

.5283 c mam/E.

133

 

 

omd vmd amd m ma aw mm va a aw ma aw oa ca aa aw 22%: 8682063

* *1 ** afi o2 am d o: mmm vmm 03 ca mam EN mmm mmm 3N Bob

wmd wdd is." 9 mo 6m 60 ma R. me E. m2 Q _ m3 ma 9 _ @6363.

* is.“ * 2 ma 2: N: Ed D: 62 m: mg a3 62 69 m2 wcuam
2225

0:5 $6525

** is." awd dd. dém 0.3 mam w.mm mém NAM mew fiam o. H m fimm 9mm adm gob

mod "is." Nod mod fie 9m fim m6 dd Wm fie Wm fim 0.6 d6 Wm $8635

in 3% awd fim a.S ddm New m. d m cdm ham 3: c. d N mdm fiwm 9am Yam $05

336366

86:: m2

in *3" 3d Nd v; m._ fim m.m ad a; m4 mg a; vd a; a; 95 wxv 0N6 :32
Ed 3d Rd d.a Nda wda ddw vda fima wda m2 Nda vda wda vdw Kama 3:5

BEBE 62585
and * mod nd ad md md Nd ed m6 dd Nd m6 ad dd ad Eng 9:09 602
"In *** wed ad ms 5w 6.3 W? «.2 62 as ca 0: ”2 ON“ 6.: £2va :20
8333 wa68m

avd vmd 3d Nd adfi QB QB a.2 adfi ad H a.m~ add awfi 0.2 awfi add Add 633E
0823 652

 

00 n4 VD do“ ow do ow ON C o2 ow do ow om no

 

 

02m>i mm mm:— 1m

 

.062 8360280 2 owfioth E 620:6 $06 60m $58 H8 58689 v68
65 83202 @608 no 8835 8:68 8 9,628 Encoaoa 8:68 6 .8835 Ragaébfi .6 $086 omcoamouémoa .a mamas—L

134

~86VR "I;
~O.OV& **
mode *

flow—u fig mom—away: 80¢ ou—du—Gm hwuoauo O—DNNSODNHDE WHO“ Ag a

Eggumme

.8225“on ES 880% “8 68:82 thmd Ea vmomd 90 36:3 35:0 @0833 333%: Bob 8.35 EH28 uEmEonSoE w
.306 mo 3:328ch 388 no 683 388 DMZ 9 38808 35:23 33 56 69a 835 >328 03323582 F
Bacoaoa mo comma”: Rafi—PTabfi mo Soto 383:0 o
8328a .8 count: EEEEAEE mo Soto 30:5 n

86 mo “outm v

Aux; 82835 5 38:8 220%on m

86 owfia Ben—N

 

 

 

 

BE awash swmf
ENSwQEVLmE
* * 3.6 M _.c we; Ed cog mud Owd co; 84 Ed mud mud mud mud A3336
36:0 v52
hmd * mmd NA 9mm NNN 9mm mém m.mm 3mm fimm fimm w.mm Dam QVN QNN 258—): 35:0 5:2
006 in nod mod cod bow wow end mud ww.w 3w cow 00.x Nod $.w mod $3 mZm x52
cod "In." * wed mod 26 mod 00% mod 3.? wed cod mag” 86 mod owé A05 8803 x32
.13.” “3.3." Q5 86 and cod and w>.m 3N cod mod mad %N Rd R.N $.m 33 5803 x52
2.9 *5.” “I" 36 3mm Ca Nvd mvd and Nw.m 86. mod 33m and :6 mmé GE “3 x32
Ed mud * NA m.mm cém Wmm mdm Nfim 3mm Wmm m.mm 7mm Wmm mém odm A39: 22% V52
.8326on x32
00 “A ,9 OS ow cc ow cm 0 CE ow cc ow om mo
oBgi mm mm: .nE

 

55:03 h mqm<rw

135

Quadratic effects of infusion were significant for DMI, meal size, and total ME
intake (P < 0.01; Table 7). This observation indicates that a threshold for the effect of
infused propionate on DMI exists and that infused propionate did not decrease DMI at
lower rates of propionate infusion but decreased DMI after the threshold was reached.
However, interactions between diet and infusion effects were not detected for feeding
behavior and DMI, indicating that diet did not affect the response to propionate infusion,
which is contrary to our hypothesis. We speculate that physical fill is a more dominant
mechanism regulating feed intake for cows fed HF while satiety related to propionate ‘
metabolism is more dominant feed intake regulator for cows fed LF. A potential
hypophagic effect of fill for HF treatment and that of greater basal propionate production
for LP treatment might have had similar effects on DMI because integration of both
physical fill and metabolic satiety signals contributes to the regulation of voluntary feed
intake (Forbes, 1995). Mbanya et al. (1993) infused acetate, propionate, or both, with or
without distention of the rumen by a balloon. Combination of VFA infusion and
distention significantly depressed DMI while VFA infusion or distention alone did not.

Their observation indicates that the threshold for infused prOpionate to decrease DMI can

be altered by dietary fill.

Intra-ruminal infusion of propionate linearly increased milk lactose concentration
(P < 0.001), but decreased concentrations of milk fat (P < 0.001), protein (P < 0.001),
and SNF (P < 0.01). Propionate infusion linearly increased ruminal propionate
concentration (P < 0.001; Table 8) and decreased ruminal acetate concentration (P <

0.001). Propionate infusion linearly increased plasma concentration of propionate (P <

136

 

"In... .13.. Gd omd we.“ no; 92 mod SM mud mad om; mm." VNN cod and eozfimuxx

 

 

 

.2... .2... e8 So on; R; R; a: o? woo m2 :2 a: m2 mi 86 @3353,
ad .2. So 85 w? a; Sc 02 e2 a: 02 $4 m2 m3 9: mm.“ Aesvseearoa
:8 r... ad to no we 3 es 3 m.» we 3 i as 2 SW @3353
33 93 93 36 o; mg mg med «3 m3 So Ed «3 $5 m3 :5 @Eaafiea
3o 5.... So 3 a? 0.? 3m 3m v.8 «.2 2:. gm v.2 3h 3m _.2 $338305
85 3.... N3 2 3:. me. 2m 2% one has m: we. 3n Se 2% Q? @3832
one 3o :3 e E m: N: me: o: ma 5 as 9; e2 5 m2 382% 38.
2286258 .2593
so 3 b 8a 3 8 ow om o 9: ow 8 9. om mo
2%: mm N”: h:

 

.038 2320280 8 ommufi E 35th 306 we.“ $58 .5“ 803 05 can 58.:
05 E 85382: no 3308 8:68 9 3:22 898%on 8:68 mo 53.35 Ecaaébfi mo magneto omaommouomon .w mama?

137

Q: d v A: Bacofloa do Soho 53:: 98 “onto 86 do 530885 “50$:me so

~OOdV l *3.
_O.OV& "In
nodvm *

Bacoaoa do :06an B:§E-mb£ .«0 Soto 038930 e
Bacofioa do .5685 Ragaébfi .8 Soto 32:4 m

as do seem .

GE 8:335 E 3038 82285 m

86 0me8 254N

 

 

 

 

86 owfie awn?

Ed “.22. * Nod 3d 3d 3d No; No; No; mad do; 34 mo; 84 we; €65 tau
* mmd mwd 5d ddd 03 v.3 9mm fimo WE ego $5 ado ado 5.5 Q; $65-5
mad in. de mod mod. 56 24m afim vwd N~.m 3N ofim mfim mmd RM mm.m 9:50—
mmd * * 5d #2 EH N2 _m_ d2 d3 #3 #2 of d2 9: of €65 RZ
:d * mud 5d mms ems mmfi mmfi wms mms wms mmfi mms mmfi mms ms .5 In
odd *1. mmd o._ NS NdH O: N: md_ as 2: wd 0.: odd nd ms 285:: .535
in 32.. :d d._ N60 «.3 ddd 5am hfim mean 0.6 NS mdm ode vdm 06m CUES 3820
5d $2.. awd mod 2 d E d d dd dd d 2 d 8d 3 d m d d m _ d d dd d do A: d 92:: 82235
.12.. .13.. mod Nd hd ad d._ v; n.“ ad wd ad mg .3 mm 1m 328v Bfioo<
«83$

6 “A b d9 ow do dv om d d9 ow dd dv om no

osfi>i mm an: Rm

 

.6253 w mqmaen—t

138

0.001) and insulin (P < 0.001), but decreased quadratically plasma acetate concentration
(P < 0.001). Linear reduction in plasma acetate concentration was expected because the
proportion of acetate in the VFA infused decreases as the proportion of propionate in the
VFA infused increases. However, a significant quadratic effect of infusion on plasma
acetate concentration indicates that acetate concentration was decreased by propionate
infusion to a greater extent at lower rates of infusion compared to higher rates of infusion.
Similarly, quadratic effect of infusion treatment was significant for plasma glucose
concentration (P < 0.01). This indicates that propionate infusion increased glucose
concentration to a greater extent at lower dose of propionate infusion, but further increase
in rates of propionate infusion increased plasma glucose concentration to a lesser extent.
Interactions between diet and infusion effects were not observed for any ruminal

fermentation and plasma metabolite response variables.

Threshold Response

In experiment 1, a threshold for effect of propionate on DMI did not exist and
infused propionate decreased DMI linearly (Figure 1). However, in experiment 2, a
threshold for propionate to affect DMI was observed; infused propionate did not decrease
DMI at lower rates of propionate infusion and decreased DMI linearly after a threshold
was reached (Figure 1). Inconsistent threshold responses observed between experiments
1 and 2 cannot be attributed to differences in dietary characteristics between experiments
1 and 2. Both experiments were designed to evaluate how dose-response effects of intra-
ruminal infusion of propionate are affected by different fermentability of diets within

each experiment. Apparent total tract starch digestibility was different by 17.9

139

 

Experiment 1
DMI (kg/12h)

15
14
13 a
12 —
11 e

w- -- -Experiment 2

 

 

 

 

 

 

 

 

 

 

 

0 20 40 60 80 100

Propionate (% of infusates)

Figure l. Dose-response effects of intra-ruminal propionate infusion on DMI (kg/12h).

Pooled means for DMI within an experiment at each infusion rate were used.

140

percentage units in experiment 1 (95.2 vs. 77.3 %), and dietary NDF concentration was
different by 8.8 percentage units for experiment 2 (34.0 vs. 25.2 %DM). If diet affects
DMI response to propionate infusion, dietary difference within each experiment should
be great enough to detect the interaction between diet and infusion effects. Therefore, the
threshold response observed in experiment 2 is more likely attributed to animal factors

that differed from experiment 1 because different cows were used in each experiment.

Data from the 3 d collection period prior to infusion were analyzed to
characterize the animals used in experiment 1 and experiment 2 (Table 9). Cows used in
experiment 1 had greater DMI (19.2 vs. 15.3 kg/12h; P < 0.001) and meal size (3.3 vs.
2.2 kg; P < 0.01) compared to cows used in experiment 2 although milk yield was not
different. Milk energy output per ME intake was greater for cows used in experiment 2
compared to cows used in experiment 1 (0.68 vs. 0.58 Meal/ME intake). This indicates
that efficiency of energy utilization for milk production was greater for cows used in
experiment 2 compared to cows used in experiment 1. In addition, cows used in
experiment 1 had greater BW than cows used in experiment 2 (698 vs. 620 kg; P < 0.01).
Plasma concentrations of propionate (0.21 vs. 0.12 mM; P < 0.001) and glucose (55.8 vs.
51.1 mg/dl; P < 0.001) were greater for cows in experiment 1 compared to cows in
experiment 2. The difference in plasma glucose concentration between cows used in
each experiment might help to explain inconsistent threshold responses to propionate
infusion in DMI and feeding behavior observed between the experiments because

propionate is a primary substrate for gluconeogenesis in ruminants. We speculate that

141

 

 

:3 3 we me £533 £85
3.... mo in mam 23mg 3820
r... 8.0 NS So 33 22295
:3 36 8.2 E .2 A25 382
«Ema—m
Mad 2 3 mg mg secs .3.
e3 26 SN 2: as 382;
r... cod m: a: St savage
mod do 34 02 GE 2853
._. mod mac and €er 28325:
m; we v.8 cam 33 £885
and mo Em 3m 3% 2302
8d 3 we: ES 225 $5 :89
r. 3.0 can 2.8 :8
Gomaficogoh Esgzm
as 3 «:8 3e ESE 835 as?
86 : cow was 323.5 Eek
a .o w an an 32:55 88:83.
moo S a: E ESEE weuem
0:5 masses—U
r. E mom a? m ENSSE 335 m:
.2. do 2 on 95 was 8a :32
one 3 3» 9% 225 stag R853
2 .o no 2 8 ans 3:8 we:
t... ed 62 we 22$ :5
8323 wamdoom
t... 3 3w 23 was a was
023i mm N EoEw—omxm fl Eofitomxm

 

._N EoEtono d5 _ Eofitomxo 5 new: $58 mo 3553220 .a mam/2.

142

25.0 V R “I;

ficd V R **

mod v m *

00238 $2020 058202808 22 n

028 0280808 02 02800».~ ..

.8022 20 5252832 22800 :0 2082 Good 02 Z 02 82220000 20880200 83 $2020 038202802 m
80820920 .20 828802 08 8 028 8 $an N

 

.222 8 0820008

<>OZ< 82: 2052280 22002282 280 8082098 2 8223 300 2822328 202 20888 0.8 20202 2020020022 8222 8809.0?
mod 2nd mmd mod mum
r. om 08 80 80 £023 >80
* mod wed mmd 3222025 222 N 220022 28020 222
$0 2 new 3m 3288 82.0 82
3. 8d mm.w mm.w SE "Em 222
Ed mod 25.2 02.2 €er 802002 022
r... 3.0 23 8a 80 5222 82
and Q d dim 9% 2.8V :2 0222
8.0 M: 3m 0.8 682 202 82
80200208 022
i. 2.2 m2» wdn mm
*5." 5O .v.m© w.wm AU
13.. 0.2 23 New 208%
owd m2 fimm Wmm mDZ
odd wd mdc Nwm 2O
* nd fiwm 2.2 2Q

82 £32302 08 82 058%

 

0228i mm N 80820me 2 8082092

 

22.803 m magi

143

cows with low plasma glucose concentration are more tolerant to hypophagic effects of

propionate.

Response of plasma glucose concentration to propionate infusion was different
between cows used in each experiment; propionate infusion increased plasma glucose
concentration linearly for cows in experiment 1, but quadratically for cows in experiment
2. In experiment 2, the marginal increase in plasma glucose concentration by propionate
infusion was greater at lower rates of propionate infusion compared to higher
rates of propionate infusion (Figure 2). Propionate infusion did not affect DMI while
propionate infusion increased plasma glucose to a greater extent, but decreased DMI
when propionate infusion increased plasma glucose to lesser extent. Infusion of
propionate did not exert hypophagic effects while infused propionate was extensively
utilized for glucose synthesis. In experiment 2, increases in plasma glucose concentration
at lower rates of propionate infusion are attributed to a greater rate of gluconeogenesis.
Although plasma glucose concentration is determined by rates of glucose supply from the
liver and glucose utilization by tissues, milk yield and plasma insulin concentration
increased at lower rates of propionate infusion. Therefore, greater plasma glucose
concentration cannot be attributed to decreased glucose clearance from the blood

circulation.

Marginal effect of infused propionate on plasma glucose concentration decreased
at higher rates of propionate infusion in experiment 2, indicating that infused propionate

might be extensively oxidized in the liver once glucose demand of body tissues is

144

 

Plasma glucose Experiment 1

 

 

 

 

 

 

 

 

0 (mg/dl) m - ~Experiment 2
68 TH _ _ _ _ _ __ _ _ _.~ _.0 _ 0 __0 ___.._M,__0__ ______.________._Hz_+_z._0_r
66 WWWWW W W W—WWWWW—W
64 W ,

62 _

60

58 a , ..
5 6 2”" 2‘ ’24?" "‘25:" ”y“ 2 ' i '— ’ _ “A A; *— # ‘~’ ##—
54 , g- _ »— 4— H- 0— “—m ___-_
52 W WW WW , W W WWWW W W W WW W W-WW
sol—w---_»—-T—» r I if-

0 20 40 60 80 100

Propionate (% of infusates)

Figure 2. Dose-response effects of intra-ruminal propionate infusion on plasma
glucose concentration (mg/d1). Pooled means for plasma glucose concentration within an

experiment at each infusion rate were summarized.

145

satisfied. Enhanced oxidative metabolism in the liver might have caused reduction in
DMI at higher rates of propionate infusion in experiment 2. Although propionate
infusion increased plasma glucose concentration linearly in experiment 1, the response in
glucose concentration to propionate infusion was less than that observed at lower rates of
propionate infusion in experiment 2. Therefore, infused propionate might have been
more extensively oxidized in the liver even at the lower rates of propionate infusion in

experiment 1 compared to experiment 2, resulting in a linear decrease in DMI.

DMI, meal size, and plasma glucose concentration during 3 d collection period
were lower, but efficiency of energy utilization for milk production was greater for cows
used in experiment 2 compared to cows in experiment 1. In addition, cows used in
experiment 2 did not decrease DMI at lower rates of propionate infusion. These
observations suggest that something other than propionate dominated regulation of feed
intake for cows used in experiment 2. Although distention in the rumen from physical fill
is another important factor that can regulate feed intake, it is not possible to identify if
ruminal distention limited maximum DMI for cows used in experiment 2 at lower rates of

propionate infusion from the data obtained in this experiment.

Hypophagic Effects of Acetate
Infusion of acetate decreased DMI compared to 3 d collection period in the
present studies. Dry matter intake during 3 d collection period before infusion was 19.1,

19.3, 14.7, and 15.8 kg/12h, respectively for SF, DC, HF, and LF treatments while it was

146

14.8, 13.2, 11.6, and 11.7 kg/12h, respectively for SF, DC, HF, and LF treatments during
infusion of 100% acetate. Several studies in the literature reported hypophagic effect of
acetate (Ulyatt, 1964; Eagan, 1966; Baile, and Pfander, 1966; Orskov et al., 1969;
Bhattacharya and Alulu, 1975; Choi and Allen, 1999). Reduction in DMI by acetate
infusion cannot be explained by extra energy supplied from acetate because total ME
intake was lower when 100% acetate was infused compared to 3 d collection period.
However, the liver does not utilize acetate, and the mechanism for hypophagic effects of
acetate is not well understood. Hypophagic effects of acetate can be partially attributed
to reduction in meal size due to increased osmolarity of ruminal fluid (Grovum, 1995;
Choi and Allen, 1999), but Choi and Allen (1999) reported that acetate infusion
decreased DMI by increasing intermeal interval compared to iso-osmotic infusion of

saline.

Acetate might also exert hypophagic effects by sparing glucose utilization in
tissues and decreasing glucose demand. Plasma glucose concentrations for 3 d collection
period before infusion were 56.1, 55.5, 51.4, and 50.8 mg/dl, respectively for SF, DC,
HF, and LF treatments. However, plasma glucose concentrations for the 100% acetate
infusion (treatment 0) were 61.9, 59.7, 54.6, and 54.2 mg/dl, respectively for SF, DC, HF,
and LF treatments. Plasma glucose concentration increased by infusion of 100% acetate
solution for all dietary treatments although acetate is not a precursor for gluconeogenesis.
This might be because acetate spared glucose in body tissues, decreasing rate of glucose
clearance from the blood, increasing plasma glucose concentration, and increasing

oxidative metabolism of organic acids in the liver.

147

CONCLUSION

Hypophagic effects of propionate were not affected by fermentability of dietary
starch sources in experiment 1 or by forage to concentrate ratio in experiment 2. Our
results indicate that propionate flux through the rumen per se did not generate satiety
signals but that propionate metabolized in the liver might have affected satiety. A
quadratic effect of propionate infusion on DMI was observed in experiment 2 but not in
experiment 1 regardless of dietary treatments. Cows in experiment 2 did not decrease
DMI at lower rates of propionate infusion while cows used in experiment 1 decreased
DMI linearly. Different response to propionate infusion might be explained by the rate at
which propionate is used for gluconeogenesis. Propionate infusion increased plasma
glucose concentration to a greater extent at lower rates of infusion for cows used in
experiment 2. Propionate may not exert hypophagic effects while infused propionate is
extensively utilized for glucose synthesis, but decrease feed intake when marginal effect
of propionate infusion on plasma glucose concentration becomes lower. We speculate
that propionate decreases feed intake in lactating dairy cows by stimulating oxidative

metabolism in the liver.

148

CHAPTER 7

Dose-response effects of intra-ruminal infusion of propionate on feeding

behavior of lactating cows in early or mid lactation

ABSTRACT

The objective of this experiment was to evaluate if dose-response effects of intra-
ruminal infusion of propionate on feeding behavior and DMI differ by stage of lactation.
Five cows in early lactation (EL) and five cows in mid lactation (ML) were used in a
duplicated 6 x 6 Latin square design (9 i 6 and 192 i 17 days in milk, respectively for
EL and ML; mean i SD). All cows were ruminally cannulated prior to the experiment.
The experimental diet was formulated to contain 30% NDF, and dry cracked corn (mean
particle size = 3.6 mm) was the major source of starch. Treatments were mixtures of
sodium propionate and sodium acetate, at ratios of 0:5, 1:4, 2:3, 3:2, 4:1 and 5:0, infused
into the rumen continuously for 18 h starting 6 h before feeding at a rate of 21.7 mmol of
sodium VFA/min. We hypothesized hypophagic effects of propionate infusion were
greater for EL compared to ML because of greater plasma concentration of non-esterified
fatty acids (275 vs. 76 uMeq/ml; P < 0.001) and expected greater basal oxidative
metabolism in the liver for EL compared to ML. Propionate infusion decreased DMI for

EL and ML, but quadratic effect of propionate infusion was observed for ML but not EL

149

(interaction P < 0.10), indicating greater marginal reduction in DMI at higher doses of
propionate for ML compared to BL. Propionate infusion decreased meal size similarly
for both stages of lactation, but linearly increased intermeal interval for ML but not EL.
We speculate that greater milk yield for EL compared to ML (42.0 vs. 30.8 kg/d P <
0.001) increased glucose demand by the mammary gland and decreased the proportion of
infused propionate oxidized in the liver for EL compared to ML. Future research needs
to evaluate independent effects of glucose demand and rates of basal oxidative
metabolism in the liver on hypophagic effects of propionate.

(Key words: propionate infusion, threshold response, stage of lactation, NEFA)
Abbreviation Key: EL = Cows in early lactation; ML = Cows in mid lactation; NEFA =

Non-esterified fatty acids; BHBA = B~hydroxy butyrate)

INTRODUCTION

Maximizing energy intake is extremely important for nutritional management of
cows in early lactation not only for maximizing milk production but also preventing
metabolic disorders such as ketosis and hepatic lipidosis. However, mechanisms
regulating voluntary feed intake are not well understood for cows in early lactation (EL).
Physical fill can be the most dominant mechanism limiting DMI for high yielding cows
around peak lactation (Allen, 2000), but it may not contribute in EL (Ingvartsen and
Andersen, 2000). Low DMI in EL might be related to elevated concentration of plasma

non-esterified fatty acids (NEFA; Ingvartsen and Andersen, 2000). The NEFA are

150

metabolic fuels extensively utilized by the liver, and extent of uptake by the liver is
determined by plasma concentration of NEFA (Emery et al., 1992). Fatty acid oxidation

in the liver provides a satiety signal in rats (Langhans et al., 1987a; Friedman et al., 1999)

Propionate has hypophagic effects in ruminants (Allen, 2000). Propionate and
NEFA are the primary metabolic fuels extensively utilized by the ruminant liver
(Demigne et al., 1986; Emery et al., 1992). Glucose, acetate, and butyrate are the other
major metabolic fuels for ruminants but have no, or inconsistent hypophagic effects
(Allen, 2000), and they are not extensively utilized in the liver (Ballard, 1965; Ricks and
Cook, 1981; Demigne et al., 1986). Langhans et al. (1985a) proposed that metabolic
fuels that are extensively metabolized in the liver have hypophagic effects. Koch et al.
(1998) showed that temporal relationships exist between feed intake and hepatic ATP
concentration in rats, supporting the hypothesis that oxidative metabolism in the liver is
involved in feed intake regulation. If propionate decreases feed intake by stimulating
oxidative metabolism in the liver (Allen, 2000), hypophagic effect of propionate is
expected to be greater for EL compared to cows in mid lactation (ML) because of greater

basal oxidative metabolism in EL from mobilized NEFA.

Although hypophagic effects of propionate are well documented, some
experiments in the literature report infusion of propionate does not affect feed intake
(Deetz and Wangsness, 1981; Quigley and Heitmann, 1991; De Jong et al., 1981; Anil et
al., 1993). Anil et al. (1993) reported that infusion of sodium propionate decreased feed

intake linearly in one experiment, but not in two other experiments. They used cows

151

differing in age and stage of lactation in each experiment. Inconsistent hypophagic
effects of propionate infusion in the literature might be because of unidentified
differences in animal characteristics. Cows differing in physiological state might respond
differently to intra-ruminal infusion of propionate by changing the threshold and

marginal effects of propionate on DMI.

The objective of this experiment was to determine if dose-response effects of

intra-ruminal infusion of propionate on feeding behavior differ by stage of lactation.

MATERIALS AND METHODS

Experimental Design and Treatments

Experimental procedures were approved by All University Committee on Animal
Use and Care at Michigan State University. Six multiparous Holstein cows in early
lactation and six multiparous Holstein cows in mid lactation were used in this
experiment. However, one cow in early lactation and another cow in mid lactation had
adverse reactions to infusions (temporary metabolic alkalosis, hypocalcemia and
hypokalemia) and were removed from the experiment. Days in milk were 9 :t 6 and 192
i 17 (Mean i SD) for five cows in early lactation and for five cows in mid lactation,
respectively. Cows in early lactation were ruminally cannulated at least 30 d prior to
calving, and cows in mid lactation ruminally cannulated for previous experiments were

selected from Michigan State University Dairy Cattle Teaching and Research Center.

152

Experimental diets contained dry cracked com, corn silage, alfalfa silage, cottonseeds, a
premix of protein supplements (soybean meal, distillers grains, and blood meal), and a
premix of minerals and vitamins (Table 1). Dietary NDF, CP, and starch concentrations
were 30.0, 18.1, and 27.4%, respectively. Dry cracked corn (mean particle size = 3.6
mm) was used as the major source of starch to limit propionate production from the basal
diet. Throughout the experiment, cows were housed in tie-stalls and fed a total mixed

ration once daily at 110% of expected intake.

The experimental period was 28 d consisting of 14 d for diet adaptation, 3 d for
data and sample collection to characterize cows at each stage of lactation, and 11 d for
data and sample collection to determine the effect of infusion treatments. Treatments
were continuous intra-ruminal infusion of mixtures of sodium propionate and sodium
acetate at 6 different ratios. Cows were assigned to duplicated incomplete 6 x 6 Latin
squares balanced for carry-over effects: one square for cows in early lactation and the
other square for cows in mid lactation. Treatment solutions were prepared by diluting
28.1 moles of sodium VFA (sodium propionate and sodium acetate at the ratio of 0:5,
1:4, 2:3, 3:2, 4:1, and 5:0) to 18 L with de-ionized water. Sodium acetate was added to
keep the osmolarity and pH of infusates constant across the treatments to isolate specific
effect of propionate relative to acetate on feeding behavior of dairy cows. Concentrations
of total VFA were 1.56 M across the treatments, and 15 L of each solution was infused
over 18 h starting 6 h prior to feeding. Infusion rate was 13.9 ml/min, which is

equivalent to infusion of 21.7 mmol of VFA/min. Solutions were infused using

153

TABLE 1. Ingredients and nutrient composition of experimental diet (% of dietary DM
except for DM).

Diet Ingredients
Corn silage 24.3
Alfalfa silage 25.9
Dry cracked corn 24.8
Whole linted cottonseed 6.4
Protein mixl 13.9
Vitamin & mineral mix2 4.7

Nutrient Composition

DM 48.4
OM 93.9
Starch 27.4
NDF 30.0
ADF 20.7
CF 18.1
Ether extract 5.2

Forage NDF 22.3

 

lProtein mix contained 75% soybean meal, 20% distillers grain, and 5% blood meal.

2 Vitamin & mineral mix contains 66.4% dry ground corn, 20.4% dicalcium phosphate,
7.8% salt, 2.4% magnesium oxide, 1.9% trace mineral premix, 0.34% vitamin A, 0.29%
vitamin D, and 0.08% vitamin E.

154

4-channel peristaltic pumps (#78016—30, Cole-Farmer Instrument, IL) and Tygon®
tubing (7.5 m x 1.6mm I.D.). Infusion started 6 h prior to feeding so that VFA
concentrations in the rumen reached steady state (assuming absorption and passage rates
of 20%/h and 15%lh, respectively) by feeding time when feeding behavior monitoring
started. Sub-periods for infusion treatment were 2 d with 18 h of infusion followed by 30

h of recovery.

Data and Sample Collection

The amount of feed offered and orts were weighed for each cow daily during the
collection period. Samples of all dietary ingredients (0.5 kg) were collected daily during
the 3 d collection period and on the feeding behavior monitoring days during the infusion
period ((1 1, 3, 5, 7, 9, and l 1) and compositied to one sample per period. Samples of orts
(12.5 %) were collected daily during the 3 d collection period ((1 15 to 17) and composited
into one sample per cow per period. Body weight and BCS were measured [(Wildman,
1982); five-point scale where 1 = thin to 5 = fat] on d 17 of each period. Cows were
milked twice daily in the milking parlor except for the evening milking on feeding
behavior monitoring days ((1 1, 3, 5, 7, 9, and 11 of infusion period) when cows were
milked in their stalls. Milk yield was measured daily during the 3 d collection period and
averaged to characterize cows in each stage of lactation. Milk yield was measured on
feeding behavior monitoring days to determine effects of infusion treatments. Milk was
sampled at every milking during the 3 d collection period and on d l, 3, 5, 7, 9, and 11 of
the infusion period. Milk samples were analyzed for fat, true protein, lactose, solids-non-

fat with infrared spectroscopy by Michigan DHIA (East Lansing).

155

Samples of feces, ruminal fluid, and blood were collected every 9 h on d 15 to 17.
Ruminal fluid samples were collected from 5 different sites in the rumen, squeezed
through a nylon screen, and pH was detemlined immediately after collection. Samples
were frozen at -20° C until further analysis. Blood samples were collected from
coccygeal vessels into two VacutainerTM tubes (Becton Dickinson, Franklin Lakes, NJ):
one with sodium heparin and the other with potassium oxalate and sodium floride as a
glycolytic inhibitor. Both were centrifuged at 2,000 x g for 15 minutes immediately after

sample collection, and plasma was harvested and frozen at -20° C until analysis.

On feeding behavior monitoring days ((1 l, 3, 5, 7, 9, and 11 of infusion period),
infusion started at 0800 h, 6 h prior to feeding, and continued for 18 h. Cows were not
allowed access to feed between 1000 h to 1400 h to minimize confounding effects of
ruminal fermentation from the previous feeding. Feeding behavior was monitored for 12
h (1400 h to 0200 h) by a computerized data acquisition system (Dado and Allen, 1993).
Data of chewing activities, feed disappearance, and water consumption were recorded for
each cow every 5 sec, and meal bouts, interval between meals, meal size, eating time,
ruminating time, and total chewing time were calculated. At the end of feeding behavior
monitoring period (0200 h), samples of ruminal fluid were collected from 5 different sites
in the rumen for each cow. Blood samples were collected from coccygeal vessels,
centrifuged at 2,000 x g for 15 minutes, and plasma was harvested and frozen at -20° C

until analysis.

156

Sample and Statistical Analysis

Diet ingredients, orts, and fecal samples were dried in a 55°C forced-air oven for
72 h and analyzed for DM concentration. All samples were ground with a Wiley mill
(1mm screen; Authur H. Thomas, Philadelphia, PA). Samples were analyzed for ash,
NDF, ADF, CP, and starch. Ash concentration was determined after 5 h oxidation at
500° C in a muffle furnace. Concentrations of NDF and ADF were determined
[(VanSoest et al., 1991); method A for NDF]. Crude protein was analde according to
Hach et al. (1985). Starch was measured by an enzymatic method (Karkalas, 1985) after
samples were gelatinized with sodium hydroxide, and glucose concentration was
measured using a commercial kit (Glucose kit #510; Sigma Chemical Co., St. Louis,
MO). Indigestible NDF was estimated as NDF residue after 120 h in vitro fermentation
(Goering and Van Soest, 1970) and used as an internal marker to calculate apparent total
tract digestibility (Cochran et al., 1986). Concentrations of all nutrients except for DM
were expressed as percentages of DM determined from drying at 105° C in a forced-air
oven. Corn grain was dry sieved through 8 sieves (Sieve apertures: 4750, 2360, 1180,
600, 300, 150, 75 um and bottom pan), using a sieve shaker (Model RX-86, W.S. Tyler
Inc., Gastonia, NC) for approximately 20 min until the bottom pan weight was constant,

and mean particle size of corn grain was calculated (ASAE, 1968).

Ruminal fluid samples were analyzed for VFA concentrations according to the
method described previously (Chapter 2). Plasma samples were processed to determine
concentrations of acetate and propionate in a similar manner as described for ruminal

fluid (Chapter 2). Commercial kits were used to determine plasma concentration of

157

glucose (Glucose kit #510; Sigma Chemical Co., St. Louis, MO), insulin (Coat-A-Count,
Diagnostic Products Corporation, Los Angeles, CA), NEFA (NEFA C-kit; Wako
Chemicals USA, Richmond, VA), and B-hydroxy butyrate (BHBA; kit #310-A; Sigma

Chemical Co., St. Louis, MO).

All data from the 3 d collection period were analyzed to characterize cows at each
stage of lactation using the ANOVA procedure of JMP® according to the following

model:

Y”: n + Si + en
where
t1 = overall mean,
Si = fixed effect of lactation stage (i = 1 to 2)

eijklnrl = residual, assumed to be normally distributed.

Treatment effects were declared significant at P < 0.05, and tendency for treatment

effects was declared at P < 0.10.

All data for infusion effects from 11 d of infusion period were analyzed using the

fit model procedure of JMP® according to the following model:

Yijklmn = l1 + Si + C(S)i(j) + Pk + Ll + Qm + SL1! + SQim + eijklmn

where

158

11 = overall mean,

Si = fixed effect of lactation stage (i = 1 to 2)

C(S),0, = random effect of cows nested in a lactation stage (i = 1 to 10),

Pk = fixed effect of periods (k = 1 to 2),

L| = linear effect of infusion

Qm = quadratic effect of infusion

SLil = interaction between lactation stage and linear effect of infusion

SQ“, = interaction between lactation stage and quadratic effect of infusion, and

eijklmn = residual, assumed to be normally distributed.

Orthogonal contrasts were made for the effects of lactation stage (EL vs. ML), linear
effect of infusion, quadratic effect of infusion, interaction between effect of lactation
stage and linear effect of infusion, and interaction between effect of lactation stage and
quadratic effect of infusion. Main treatment effects were declared significant at P < 0.05,
and tendency for treatment effects was declared at P < 0.10. Interaction effects were

declared significant at P < 0.10.

159

RESULTS AND DISCUSSION

DMI and Feeding Behavior

During the 3 d collection period before infusion, DMI tended to be lower for EL
compared to ML (16.2 vs. 18.3 kg/12h; P < 0.09; Table 2) although milk yield was
greater for EL compared to ML (42.0 vs. 30.8 kg/d; P < 0.001). Milk energy output per
ME intake was greater for EL compared to ML (0.88 vs. 0.54 Mcal/ME intake; P <
0.001), indicating greater efficiency of dietary energy utilization for milk production for
EL compared to ML. An interaction between effect of lactation stage and quadratic
effect of infusion was significant on DMI (P < 0.10; Table 3). Propionate infusion
decreased DMI linearly for EL (P < 0.001) but quadratically for ML (P = 0.03; Table 4;
Figure 1). This indicates that there was less marginal reduction in DMI at lower doses of
propionate but greater marginal reduction in DMI at higher doses of propionate for ML
compared to BL. Propionate infusion linearly decreased meal size for EL and ML to a
similar extent (Figure 2). However, an interaction between effect of lactation stage and
linear effect of infusion was significant for intermeal interval (P < 0.10). Propionate
infusion linearly increased intermeal interval for ML (P < 0.01) but not EL (Figure 3).
The greater marginal reduction in DMI at higher rates of propionate infusion for ML
compared to BL is because propionate infusion decreased DMI by both decreasing meal
size and increasing intermeal interval in ML while decreasing meal size only for EL.

Propionate infusion delayed the sense of hunger for ML but not EL.

160

 

wmd 56 EN mud 00200 0240.

 

 

... 3 0 00.0 0: 0.00 00005
00.0 00.0 00.0 00.0 000 0083.80
40.0 4.0 0.: 0.00 000 32000
2.0 8.0 00.0 00.0 000 08000.80
000 0.0 0.00 0.00 000 208280
00.0 0.0 0.00 30 0.00 2880.
00.0 0 000 000 2250 <0> 300.
0 0.0 00.0 00.0 3.0 :0

2523:0503 35:5.“

000 00 0.00 0.00 2050 335 .203

00.0 00 00.. 000. 30:85 000.0

00.0 0 000 400 30:85 00020000

00.0 00 000 000 200205 0:08

082 223020

:0 0.0 0.: 0.00 2005820 8005 m:

0_ .0 0.0 0.0 0.0 98 00 000 82

00.0 0.0 0.00 0.00 0000 2325 8825

00.0 0.0 0.0 0.0 203 0500 82

00.0 0.0 0.00 0.00 30:00 020

0030202 w8200m

02030 00 0002 200

80200205

228 200 52320082 8028: 0002020008 2003 8020808002 288:: 0030202 $22000 00 00200002 20 0w000 20 00002.2 .N ”21:73.

161

 

 

00.0 0.0 0.00 0.00 mm
00.0 0.0 0.00 0.00 00
00.0 00 0.00 0.00 0080
40.0 0._ 0.00 0.00 ”82
00.0 2 0.40 0.00 20
04.0 00 0.00 0.40 48
302 52320082 0002 2082 80000904
00 0.0 0.0 0.0 000080 <88
*0... 00 00 000 20500210 <82
.2... 0.0 0.: 0.0 0:083 5305
r. 0.0 0.00 0.00 0208 00820
04.0 00.0 00.0 00.0 0200 008280
40.0 00.0 00.0 00.0 2280 23800
08003

020: 00 0002 2200

 

.8080 0 0:05.

162

28.0 V Q t:

5.0 v k 0....

mod v m *

02008 $0000 022002020008 22 0
0200 000002208 00 000000.24~ 4

.0002 20 202520022 20800 00 20002 Good UM Z 00 828000 2000200200 003 30000 03002020002 0

000080 00 00000 002 0
00200002 .20 0m000 >209

 

 

:0 00.0 00.0 00.0 000
00.0 00 200 000 200 00203 080
r... 000 40.0 00.0 02020820 W82 0 008.20 0080 002
r... 0.0 0.20 0.00 608220 0080 002
._. 020 00.0 00.0 0.00 ”020 022
00.0 00.0 00.4 00.4 000 00900 002
r. 00.0 00.0 00.0 2000 0280 002
00.0 02 .0 00.0 00.4 0.00 00 002
r; 0.2 0.00 0.04 00000 220 002
00200208 022

003-0 00 0022 2000

 

02.8000 N mamaofi.

163

 

Om ASN #:v

$4.0 02.. 20.0 N24 m.22 N20. 222 0.0.0 n.2, 020 NE. 002 0.2. Emu Numb m.mw 002082 00203

mmd .13.. mud cm 002 mom 20m new mom 20m 22m 0mm 20m mom mom 02 m 208.21

222.0 wwd and m 2 32 Om 2 S2 22 2 002 w2 2 22 32 m2 2N2 m~2 wm2 82002832

mad 1.... and 02 N0 02 $2 22 mm2 202 242 2 $2 22 0242 N22 N02 w020m2
328.482

082 823020

2.0 *0... mmd 22 00mm 0.5 2mm 0.2m w.mm 9mm Wmm Wmm NSN 2.0m 0.3 0.2.0. 00 208,2.

mod *1. 20.0 2.0 26 ed wé N24 Em Nd 5m Nd 2.24 ~24 5m NM 00020022

m 2 .o 03.3.. mmd 02 24.0.2 5mm Vow 2.0m 2.0N 40mm 0.02 mdm 2.3 2mm Nam m.wm 00 020202

A2N220022

002082 m2

0. 03.2.. 00.0 md o.2 2.2 w.2 w.2 2.2 w.2 m2 2.2 w.2 2.2 2d 0.2 A2Qw0220~202002

3.0 022.0 2.0 22 tom 002 0.00 2.00. 0.240 N62 2.20 0.0.x W20 0.00 82 0.0.0 020282
202082 20080082

 

 

 

00.0 00.0 00.0 0.0 0.0 0.0 2.0 0.0 4.0 0.0 0.0 0.0 0.0 0.0 2.0 0.0 20020080 282
02.0 I... 00.0 0.0 2.0 4.0 0.02 0.: 0.02 0.02 4.0 0.0 0.02 0.: 0.: 0.02 8200280020
00302202 000220002
04.0 04.0 00.0 0.0 0.42 0.42 0.02 0.42 0.42 0.42 0.42 0.42 0.42 0.42 0.42 0.42 2000000
0830; ~030<
00 m2 .0 002 00 00 04 00 0 002 00 00 04 00 M0
022-0 00 N002 .000

 

00202002 228 200 022000 02 0300 00.2 0020020002 02228
200 0030202 8.22002 00 0002000 802200 02 0>20200 0200022002 80200 .20 00200202 200288028 20 0000.20 00000200000002 .m mama;

164

02000220002 20 200220 02002000 200 02020 00202002 20 002000082 800228w2m 00

0000022002 20 200220 00002 200 0w000 00202002 20 002000082 800222022 00

2OOdVK 02.3..
2odvn~ 0....
mch0~ ...

002082 220000 0220020020208 ”222 0

50.200900

0200022002 200 0202000 002 2082002 thmd 200 voomd 20 520002 30000 2080000 002000202 80.22 002082 220000 022002020202 0
02022 20 2222220022 20800 00 20002 283 022 02 202200000 2020200200 003 2022 8002 002082 220000 022002020202 0
0200022002 20 00200202 200288-028 20 200220 02002000 0
02000202002 20 00200202 20028000082 20 200220 000022 0
00200002 20 02020 20 20022.2 4
GE 002000202 02 80800 0200022002 0

 

 

 

 

00202002 20 02020 2220
00202002 20 0w000 220.2.
0020020820202
0.020 ...._.... ...... 00.0 no. 2 mod 22.0 mod and mod 02 .2 02 .2 mo. 2 m2 .2 no.2 mad H0220022
30000 0222
mwd mad ...... 24.2 2.0a 0.2 9mm 0.02 0.0m 2.- 0.0m 0.0m 2.0N mdm mdm 2.0m 2220022 $0000 0222
... mad .1... N2 .0 00.0 20.0 00.0 00.0 00.0 02.0 2400 20:0 mad 000 0.0.0 22.0 802 "220 02222
00.0 02.0 mmd mod mo.m 00.24 00.0 20.0 mad 00.24 00:4 00.0 240.24 00.24 00.24 00.24 20002 0002002 02222
00.0 00.0 ...... 02 .0 mod 09m 000 m2.m ~00 mm.m 200 mm.m mmd 22.~ cod 200 0002 020208 02222
00.0 mmd 00.0 and No.24 5.0 no.0 00.0 m0.m 224.4 22.4 2.0 0.24.24 0.0.0 2024 240.24 20002 202 02222
... 024.0 .13.. 24.2 0.0m m.wm 0.0m 0.0m 0.0m m.wm odm 2.0m 24.00” mdv 0.0m 00m 2202 2202.0 02222
002002002 02222
.0 .0 .0 002 00 00 04 00 0 002 00 00 04 00 m0
2.20.7.2 00 .002 .0000

 

02.8002 m m2m<H

165

 

286 V R .13..

2o.o V R ......

no.0 V & ...

0200020020 20 00200202 202.282-0002 20 200220 0220220000 n
0200020020 20 00200202 200280202202 20 200.220 3005 N
€02 8280202 :2 2002000 020002005.

 

 

 

«no .2. no «.02. 0.8. 02 0.2 oi «.8 8.3.0... E... a. 360
v0.0 8.0 02 2.2 3.. n; E 0.8. E 82.32.. 0...... =2 :60
02225 2030202 2000:0202
.. .2... E E 3.. E. n: 0.2 0.2 80.28. 28 a. $60
3.0 .3... 00 3. 3 no. 0.: 0: n: 82.82.. be... a. :60
330.0 220

.0 .0 mm 8. cm 8 9. cm .0

002.? m

 

002202002 228 200 .2230 :2 0300 202 20>.20202

200820202 2:0 235 :0 028000 80208 02 032202 0200020020 80228 .20 0029.202 200288-825 20 $002.20 000000020032 .2. m4m<H

166

 

DMI (kg/12h)

 

Propionate (% of infusates)

Figure l. Dose-response effect of intra-ruminal propionate infusion on DMI (kg/12h) for

cows in early lactation (EL) and mid lactation (ML).

167

 

 

Meal size

 

 

 

 

 

 

 

0 20 40 60 80 100

Propionate (% of infusates)

Figure 2. Dose-response effect of intra-ruminal propionate infusion on meal size (kg) for

cows in early lactation (EL) and mid lactation (ML).

168

 

Intermeal
interval (min) ”" "' 'ML

 

 

 

 

0 20 40 60 80 100

Propionate (% of infusates)

Figure 3. Dose-response effect of intra-ruminal propionate infusion on intermeal interval

(min) for cows in early lactation (EL) and mid lactation (ML).

169

The different response in intermeal interval to propionate infusion might be
related to the fate and rate of propionate utilization in the liver because post-meal
propionate flux to the liver is expected to be similar for EL and ML at respective infusion
treatments. The same diet was fed to both groups of cows and propionate infusion
decreased meal size to a similar extent for EL and ML. Ruminal propionate
concentration and apparent total tract OM digestibility were not affected by stage of
lactation during the 3 d collection period before infusions. We speculate that glucose
demand of body tissues affects propionate metabolism in the liver; infused propionate is
used for gluconeogenesis to a greater extent and less propionate is oxidized in the liver as
glucose demand of body tissues increases. Propionate infusion did not affect intermeal
interval for EL possibly because greater glucose demand from greater milk yield
stimulated gluconeogenesis and decreased the relative proportion of infused propionate
used for oxidative metabolism in the liver. However, infused propionate is available as
TCA cycle intermediates for a longer period of time if glucose demand in peripheral
tissues is relatively low, increasing oxidation of fuels in the liver and delaying the sense
of hunger. Intermeal interval increased linearly by propionate infusion for NIL possibly
because of lower glucose demand related to lower milk yield and higher DMI compared

to EL.

Contrary to our hypothesis, propionate infusion decreased meal size for EL and
ML to a similar extent. We hypothesized that lower rates of propionate infusion would
have a greater effect on meal size for EL compared to ML because basal oxidative

metabolism in the liver is expected to be greater for EL compared to ML. In support of

170

this, plasma NEFA concentration was greater for EL compared to ML (275 vs. 76
ueq/ml; P < 0.001), and a greater extent of NEFA uptake by the liver was expected for
EL (Emery et al., 1992). However, the similar response to propionate infusion in meal
size for EL and ML indicates that greater plasma NEFA concentration did not decrease
meal size synergistically when propionate was infused. This might be because uptake of
NEFA by the liver is not directly related to rate of oxidation. Although extent of NEFA
uptake by the liver is proportional to its concentration in plasma (Emery et al., 1992),
NEFA can be used to form triglycerides that are stored in the liver or exported into the
blood as VLDL. In addition, infused propionate can decrease mobilization of NEFA by
increasing insulin secretion. In this experiment, propionate infusion linearly increased
insulin secretion (P < 0.05; Table 4) and decreased plasma NEFA concentration (P <
0.05) for EL. In addition, propionate may have inhibited fatty acid oxidation in the liver

by decreasing activity of fatty acyl-CoA dehydrogenase (Emery et al., 1992).

Another explanation for the similar effect of propionate infusion on meal size for
both stages of lactation might be the greater glucose demand for EL compared to ML,
which might have reduced the hypophagic effects of infused propionate that were
expected to be greater for EL because of greater basal oxidative metabolism of NEFA in
the liver. The relative proportion of infused propionate oxidized in the liver might have
been lower for EL compared to ML because greater milk production and glucose demand
in the mammary gland enhanced propionate utilization for gluconeogenesis in the liver.

It is difficult to isolate specific hypophagic effects of mobilized NEFA from effects of

171

 

 

 

 

 

.3... .2... 00.0 0.0 5.. 00.. 3.. 03 Nam .0... 8.. 00. 20.0 03 00.0 00.0 $00.20....
.. .2... 00.0 00.0 2... mm. R2 00. 20. m... 0.... 00.. 00. mm. 00.. ~00 0.0.9.205
$0 .2... 2.0 00.0 00.. 00.. 00.. 0.... mm. mm. 00.. an. 0.... 00. 0a.. 0... 2.0.0.8225...
P0 .. 00.0 0.0 0.0 3 E 0.0 0.0 m... 0.0 2.0 E 0.0 0.0 0... 0020.80.00
00.0 .2... ~00 3.0 00.0 00.0 .00 0.0 00.0 0.0 5.0 :2 3.0 00.0 3.0 2.0 00.2000982
000 .2... 00.0 0.. 0.00 0.0 0.0m 0.00 mg 0.0. 0...... 0.? 0.00 0.2 0.0. 0.: 02.002.25.00...
90 .2... 00.0 0.. 0.0.. 0.? 000 3.0 0.00 0.00 S... 0.00 0% 0.0 0.00 ~00 02.00288...
000 3.0 00.0 m a. 0: mm. 0N. mm. £2 .0. 0m. 2. 8. 0m. mm. 2250<0>§£
:Oflfiflogom EGEDM
.0 .0 .m 00. 00 00 0.. 0m 0 00. 00 00 0.. om .0
8.9-.. mm .22 .030

 

002202002 2028 000 .2230 02 0300 20.2 200022 0.22 2000 00:20..
022 02 00222020208 :0 0202000 8022000 02 032202 0200020020 8022000 .20 00200202 2002.02-00.22 20 02002.20 000000020009 .2 mAm<H

172

0200020020 20 80.2.20 022022000 2000 0200 002202002 20 0022080202 2000020222 00
0200020020 .20 200220 200.222 2000 00020 00202002 20 0020020202 2000220222 .0

28.0 V R 0......

20.0 v 02 .3.

20.0 V R ...

0200020000 20 00200202 2002.08-02.02 .20 .0000 020.205 0
0200020020 20 00200202 2002.08-02.02 .20 200.220 805.2 0
002202002 .20 02020 .20 200.25 ..

€02 002000.202 :2 2002.200 0200020020 0

002202002 .20 02020 2232.

002202002 .20 00.0.0 023m.

 

22.0 .3... 22.0 20.0 00.2 00.2 20. 2 20. 2 20. 2 00.2 00.0 20.2 00. 2 20. 2 00.2 2 2.2 A200 100
20.0 N20 22.0 2.0 2.00 2.00 0.00 2.20 2.20 2.20 2.20 2.20 0.20 0.~0 2.00 2.20 2280 .20
02.0 .3... 22.0 22.0 N2 .2 20.2 20.2 22.2 22.2 20.2 N22 20.2 02.2 20.2 02.2 20.2 32.5 +v2
20.0 00.0 2 2.0 0.0 002 202 202 002 002 002 00 2 002 002 002 022 002 A2205 +072
22.0 .3. 00.0 20.0 22.0 22.0 02.0 02.0 22.0 02.0 02.0 22.0 22.0 22.0 22.0 22.0 2.20

.3... .3... 20.0 02 0.2 0.2 0.2 2.0 2.2 0.2 2.2 2.2 2.2 2.2 2.2. 0.22 0040A22\wEV<m2.2m
20.0 .. .1. N2 02 20 22 222 00 02 002 022 202 00m 002 00m 00 3500212 <n2m272

 

 

 

00.0 .. 00.0 0.. ..0. 00 0.0. ..0. 0.0 00 0.0 0.0 ..0 ~.0 0.0 0... 005.0003...
.. .1... .. 0.. 0.00 0.00 0.0 0.0 ~...0 ~..0 0.00 0.~0 ...~0 ~00 0.00 0.00 000.000.88.00
0.0 ...... 00.0 ~00 ~00 0~0 0~0 -0 2.0 :0 .00 00.0 -0 0~0 0~0 0.0 .250280...e..
.. .3... ~00 00 0.0 0.0 0.. 0.. 0.. ..~ 0.0 0.. ~.. 0.. 0.~ 0.0 0312.... 0.0.8....
080.05
.0 m. .0 00. 00 00 0.. ..~ 0 00. 00 00 0.. 0~ .0
02.3-.. 00 .02.. .00....

 

0.008. 0 0.1.3....

173

greater milk production and glucose demand in peripheral tissues and to evaluate if
infused propionate exerts more hypophagic effects for EL compared to ML. Future
research needs to examine independent effects of different demand for gluconeogenesis

and different basal oxidative metabolism in the liver on hypophagic effects of propionate.

Plasma Metabolites

Plasma concentrations of glucose (59.0 vs. 52.8 mg/dl; P < 0.01) and insulin (11.9
vs. 5.7 ulU/ml; P < 0.001) were greater for ML compared to BL during the 3 d collection
period before infusion. Propionate infusion increased plasma glucose concentration
linearly for ML but quadratically for EL; the effect of propionate infusion on plasma
glucose concentration was greater for EL especially at lower rates of propionate infusion
(Figure 4). In Chapter 6, it was discussed that hypophagic effects of propionate are
related to the marginal effect of propionate infusion on plasma glucose concentration.
However, in this experiment, propionate infusion decreased DMI linearly for EL even at
lower rates of propionate infusion that increased plasma glucose concentration to a
greater extent. This observation appears inconsistent with the discussion in Chapter 6.
However, infused propionate at lower rates of infusion might have stimulated oxidative
metabolism in the liver by increasing TCA cycle intermediates and oxidation of acetyl
CoA. Plasma BHBA concentration was greatly reduced from 15.0 mg/dl with infusion of
0% propionate solution to 3.5 mg/dl with infusion of 40% propionate solution for EL
(Figure 5). This reduction in plasma BHBA concentration might indicate that acetyl CoA
in the liver was completely oxidized as the proportion of propionate in infusates

increased to 40%. Although infused propionate increased plasma glucose concentration,

174

 

Plasma glucose

(mg/d1) .... " ”ML

 

 

 

 

 

 

 

0 20 40 60 80 100

Propionate (% of infusates)
Figure 4. Dose-response effect of intra-ruminal propionate infusion on plasma glucose

concentration (mg/d1) for cows in early lactation (EL) and mid lactation (ML).

175

 

Plasma BHBA

 

 

 

 

 

 

Propionate (% of infusates)

Figure 5. Dose-response effect of intra-ruminal propionate infusion on plasma B—hydroxy

butyrate (BHBA; mg/dl) for cows in early lactation (EL) and mid lactation (ML).

176

it probably also increased oxidation of acetyl CoA in the liver, decreasing DMI.
Discussion in Chapter 6 indicated that gluconeogenesis is the major metabolic pathway
for propionate in the liver, but accumulation of acetyl CoA in the liver might increase the
relative proportion of infused propionate used for driving TCA cycles. Accumulation of
acetyl CoA stimulates pyruvate carboxylase and increases oxaloacetate formation from
pyruvate (Ballard et al., 1969). Therefore, infused propionate might be sufficient to
stimulate gluconeogenesis and oxidative metabolism in the liver simultaneously.
Although BHBA reduction at lower rates of propionate infusion indicates enhanced
oxidative metabolism in the liver, it does not exclusively support this explanation.
Propionate may decrease ketogenesis by increasing insulin secretion and decreasing
lipolysis in the adipose tissues or by decreasing B—oxidation of NEFA (Shaw and Engel,

1985).

Plasma glucose concentration for ML was increased by infusion of 100% acetate
solution (0% propionate treatment) compared to that of the 3 d collection period before
infusions (61.2 vs. 59.0 mg/dl), which is in agreement with the observation in Chapter 6.
This observation suggests that acetate spared glucose in some tissues for ML and
decreased clearance rate of glucose from the blood circulation. However, plasma glucose
concentration was not affected by infusion of the 100% acetate solution for EL
(53.0 vs. 52.8 mg/dl), and plasma acetate concentration increased from 1.3 mM for the 3
d collection period before infusion to 6.3 mM with infusion of the 100% acetate solution.
The different response to acetate infusion for EL and ML suggests that acetate spares

glucose utilization to a lesser extent for EL compared to ML. Although the mechanism

177

 

for lower utilization of acetate in BL is not known, activity of enzymes needed to utilize
acetate as a metabolic fuel or for fatty acid synthesis might be lower in EL. Activity of
acetyl CoA carboxlylase that catalyzes the limiting step for de novo fatty acid synthesis
might not be sufficient due to low concentration of plasma insulin in EL (Vernon et al.,
1991). In addition, Guesnet et al. (1991) reported that insulin stimulated incorporation of
acetate into fatty acids to a lesser extent for ewes in early lactation compared to non-

lactating ewes in mid—stage of pregnancy.

CONCLUSION

Propionate infusion decreased DMI for both EL and ML, but a quadratic effect of
propionate infusion was observed only for ML indicating a greater marginal reduction in
DMI at higher rates of propionate infusion for ML compared to BL. The different
response in DMI between EL and ML can be attributed to effects of propionate on
intermeal interval. Propionate infusion linearly increased intermeal interval for ML but
not EL while propionate infusion decreased meal size similarly for EL and ML. Greater
milk production relative to ME intake for EL might increase demand for gluconeogenesis
and decrease the proportion of infused propionate oxidized in the liver at higher rates of
propionate infusion compared to ML. Although plasma NEFA concentration was greater
for EL compared to ML, the hypophagic effects of propionate were not greater for EL at
higher rates of infusion. This might be attributed to confounding effects of greater

glucose demand and subsequent greater gluconeogenesis in EL. However, a sharp

178

reduction in plasma BHBA concentration and linear reduction in DMI were observed for
EL at lower rates of propionate infusion, suggesting that propionate might have caused
hypophagia by stimulating oxidation of acetyl CoA in the liver despite greater glucose
demand. Future research needs to examine independent effects of different glucose
demand and different rates of basal oxidative metabolism in the liver on hypophagic

effects of propionate.

179

CHAPTER 8

IMPLICATIONS

Shifting primary site of starch digestion from the rumen to intestines by replacing
dry ground corn for ground high moisture corn increased DMI of lactating dairy cows fed
a high starch diet (Chapter 2). Excess propionate production in the rumen was
considered to limit maximum voluntary feed intake in lactating dairy cows. Lactating
dairy cows linearly decreased DMI, meal size and meal frequency as the rate of intra-
ruminal infusion of propionate increased (Chapter 3). Infusion of propionate decreased
DMI and total metabolizable energy (ME) intake. This means that hypophagic effects of
propionate are not explained by additional energy supplied as propionate, but that feed
intake was regulated specifically by propionate. Hepatic ATP concentration has been
suggested as a regulator of satiety and hunger in rats. Because propionate is one of the
major metabolic fuels for the ruminant liver, the extent and rate of propionate production
in the rumen may influence hepatic ATP concentration and feeding behavior. Although
hepatic ATP was not directly measured in experiments for this dissertation, the series of
experiments showed that oxidative metabolism in the liver is involved in regulatory

mechanism of feeding behavior by affecting satiety and hunger in lactating dairy cows.

Infusion of ammonium decreased DMI (Chapter 4), and hypophagic effects of
ammonium are facilitated by intra-ruminal infusion of propionate but not by acetate

(Chapter 5). It is speculated that infusion of ammonium stimulated urea synthesis that

180

requires N from amino acids and that deamination of amino acids increases carbon
available for oxidation or gluconeogenesis in the liver. When ammonium was infused
with acetate, DMI was similar compared to infusion of sodium acetate possibly because
the liver does not utilize acetate, and carbon from amino acids were utilized for
gluconeogenesis without extensively stimulating oxidative metabolism. However, when
ammonium was infused with propionate, oxidative metabolism in the liver might have
increased because the liver utilizes propionate extensively. Propionate and amino acid
carbon generated from urea synthesis are substrates for gluconeogenesis. Substrate
supplied above that needed to meet glucose demand of body tissues increases oxidation
in the liver. Our observations in this experiment support the theory that oxidative

metabolism in the liver affects DMI and feeding behavior.

Two experiments in Chapter 6 were designed to determine if a threshold for
infused propionate to affect DMI exists and if it exists how the threshold is affected by
fermentability of diets. The extent of hypophagic effects of propionate on feeding
behavior and DMI were not affected by fermentability of dietary starch or by forage to
concentrate ratio, although response in feeding behavior to propionate infusion differed
between cows used in each trial. Cows used in trial 2 did not decrease DMI at lower
doses of propionate while cows used in trial 1 decreased DMI linearly. Different
responses to propionate infusion might be explained by the marginal effect of propionate
on plasma glucose concentration; propionate infusion at lower doses increased plasma
glucose concentration to a greater extent for cows used in trial 2 compared to cows in

trial 1. These observations suggest that propionate does not exert hypophagic effects if

181

infused propionate is extensively used for glucose synthesis, but propionate infusion
decreases feed intake when marginal effects of propionate infusion on plasma glucose
concentration decrease. Cows in early lactation decreased DMI linearly at lower doses of
propionate while infused propionate greatly increased plasma glucose concentrations
(Chapter 7). Although this appears to be inconsistent with observations described in
Chapter 6, propionate infusion decreased plasma concentration of B-hydroxy butyrate.
Infusion of propionate might have stimulated oxidative metabolism of acetyl CoA in the

liver for cows in early lactation, and resulted in a linear reduction in DMI.

Experiments in chapter 6 and 7 showed that propionate flux to the liver does
not directly decrease feed intake but hepatic oxidative metabolism stimulated by
infusion of propionate has direct hypophagic effects. Treatment means from the
experiments described in Chapter 6 and Chapter 7 were plotted to determine the
relationship between marginal reduction in DMI (kg) per mmol/min of propionate
infusion and plasma glucose concentration (Figure 1). As plasma glucose concentration
increases, DMI decreased to a greater extent by propionate infusion. It is probably
because glucose demand of body tissues is satisfied at greater concentration of plasma
glucose and infused propionate is not extensively utilized for gluconeogenesis but
oxidized in the liver. Our observations imply that the threshold for propionate to affect
DMI becomes greater for cows with lower plasma glucose concentration and that plasma
glucose concentration might be useful to predict DMI responses to diet change. Cows
with low plasma glucose concentration might be more tolerant to hypophagic effects of

propionate, thus it is likely to increase their productivity by feeding more fermentable

182

diets. However, cows with high plasma glucose concentration might have greater risk to
decrease DMI by a similar diet change. This knowledge can be utilized to make

management decisions on how to group cows within a herd.

Propionate infusion decreased DMI linearly for cows in early lactation (EL), but
quadratic effect of propionate infusion was observed for cows in mid lactation (ML),
indicating greater marginal reduction in DMI for ML compared to BL at higher rates of
propionate infusion (Chapter 7). Propionate infusion linearly increased intermeal interval
for ML, but not EL. Greater milk production for EL might have increased
gluconeogenesis and decreased the proportion of infused propionate oxidized in the liver

at higher rates of propionate infusion compared to ML.

The series of experiments in this dissertation showed that the effects of propionate
on DMI in lactating dairy cows is consistent with its effects at stimulating oxidative
metabolism in the liver. This knowledge can be utilized to improve profitability of dairy
operations. Maximizing energy intake is an important goal in nutritional management for
high producing dairy herds, and future research should focus on how to increase energy
intake by optimum diet formulation. Replacing high moisture corn with dry ground corn
is a possible approach to decrease rate of propionate production in the rumen (Allen,
2000). It is also necessary to investigate how splanchnic tissues metabolize glucose that
is absorbed in the small intestine. If oxidative metabolism in the liver affects DMI and

feeding behavior as proposed, it is important to understand how to formulate diets that

183

 

 

 

_f§0.4-
E ' .

EE 0.2' o

5%

$05 0 I

:6

ch

cum-0.2-

8:2

kc

Fig-0.4- . .

DD

35-0.3

233 ' -
520.8 . . . . .

56 58 60 62 64 66

Plasma glucose (mg/d1)

Figure 1. Relationship between marginal response in DMI (kg) per mmol/min of
propionate infusion and plasma glucose concentration (mg/d1). Marginal response in

DMI = 3.0 - 0.05 x plasma glucose concentration (r2 = 0.26; P < 0.01).

184

will minimize oxidative metabolism of propionate. Interactions with other dietary
components such as lipid and protein need to be investigated further because both fatty
acids and amino acids are metabolic fuels utilized extensively in the ruminant liver. In
addition, if hypophagic effects of propionate are altered by glucose demand of body
tissues, the potential for dietary treatments to alter the metabolic fate of propionate in the
liver (i.e. increasing gluconeogenesis and decreasing oxidative metabolism) should be
investigated. These are only a few examples of possible research that can be done in the
future. Elucidating regulation mechanisms for feed intake is a fertile and exciting area of

research.

185

APPENDIX

The Effect of Pulse Dose of Sodium VFA

ABSTRACT

Dose—response effects of intra-ruminal infusion of propionate on feeding behavior
of cows that differed in stage of lactation were to be evaluated, but the experiment was
aborted due to the severe adverse effects of the infusion protocol. Treatment solutions
were prepared by diluting 33.8 moles of sodium VFA (sodium propionate and sodium
acetate at ratios of 0:5, 1:4, 2:3, 3:2, 4:1, and 5:0) to 18 L with de-ionized water.
Concentrations of total VFA were 1.88 M across the treatments. A priming dose of 2.29
L of each solution was administered ruminally 2 h prior to feeding and 13.4 L of each
solution was infused over 14 h starting immediately after the priming dose. Infusion rate
was 16 ml/min, which is equivalent to infusion of 30.0 mmol of VFA/min. Infusion
started 2 h prior to feeding with priming dose so that VFA concentrations in the rumen
reached steady state concentrations prior to feeding. However, S cows out of 12 were
unable to stand or walk properly 6 h after the end of infusion. Common symptoms were
dehydration, hypocalcemia, hypokalemia, and metabolic alkalosis. It is speculated that
the priming dose of VFA salt increased osmolarity of ruminal fluid causing systemic
dehydration. Decreased gastric emptying from hyper-osmotic duodenal digesta probably

increased the severity of dehydration by limiting water intake because of ruminal

186

 

distention. In addition, infusion of sodium with VFA caused metabolic alkalosis and
decreased serum concentration of potassium. Metabolic alkalosis might also induce
conformational change in PTH receptors, developing hypocalcemia. It is concluded that
systemic dehydration was caused by a priming dose of hyper-osmotic solutions, and
infusion of sodium VFA solution affected acid-base balance in serum, resulting in

metabolic alkalosis, hypokalemia, and hypocalcernina.

BACKGROUND

Six multiparous Holstein cows in early lactation (11 :l: 6 DIM; mean i SD) and
six multiparous Holstein cows in mid lactation (100 :l: 17 DIM; mean i SD) were used
for this experiment. Cows in early lactation (ruminally cannulated at least 30 (1 prior to
calving), and cows in mid lactation (ruminally cannulated for previous experiments) were
selected from the herd of Michigan State University Dairy Cattle Teaching and Research
Center. Mean BW and BCS for the cows were 666 kg and 2.8, respectively.
Experimental diet and research protocols were similar to those described in Chapter 7
except for the infusion protocol. The experiment described in Chapter 7 originally
planned to use total of 24 cows (12 cows in early lactation and 12 cows in mid lactation)
with two blocks in time using 12 cows each because an automated system for feeding
behavior monitoring was available for 12 stalls. The experiment described in this

appendix was the first part of the experiment that evaluated dose-response effects of

187

intra—ruminal infusion of propionate on feeding behavior of lactating cows in early or mid

lactation

Experimental periods were 28 (1 consisting of 14 d for a diet adaptation, 3 d for
data and sample collection to determine effects of stage of lactation, and 11 d for data and
sample collection to determine effect of infusion treatments. Infusion treatments were
continuous intra-ruminal infusion of mixtures of sodium propionate and sodium acetate at
6 different ratios. Treatment solutions were prepared by diluting 33.8 moles of sodium
VFA (sodium propionate and sodium acetate at ratios of 0:5, 1:4, 2:3, 3:2, 4: 1, and 5:0) to
18 L with de-ionized water. Sodium acetate was added to keep the osmolarity and pH of
infusates constant across the treatments to isolate specific effect of propionate relative to
acetate on feeding behavior of dairy cows. Concentrations of total VFA were 1.88 M
across the treatments, and 2.29 L of each solution was prime-dosed 2 h prior to feeding
and 13.4 L of each solution was infused over 14 h starting immediately after
administration of the priming dose. The priming dose was administered to shorten the
infusion period required to reach steady state concentrations of VFA prior to feeding; it
would take 6 h of infusion to reach steady state concentrations of VFA by continuous
infusion without the priming dose, assuming absorption and passage rates of 20%/h and
15 %lh, respectively. Infusion rate was 16.0 ml/min, which is equivalent to infusion of
30.0 mmol of VFA/min. The solutions were infused using peristaltic pumps (#78016-30,
Cole-Farmer Instrument, IL) and Tygon® tubing (7.5 m x 1.6mm I.D.). At the end of
infusion, blood samples were collected from coccygeal vessels, and serum was harvested

and analyzed for the concentrations of Na and K. Water intake and DMI were recorded.

188

RESULTS

After the first infusion period, S cows (4 cows in early lactation and 1 cow in mid
lactation) were unable to stand or walk properly within 6 h following cessation of
infusion. Common symptoms were dehydration, hypocalcemia, hypokalemia, and
metabolic alkalosis. Two cows showed abnormally high blood PCO2 (70.2 and 92.0
mmHg) and died within two days while three cows recovered within a week. Necropsy
revealed that abomasums were impacted with extremely dry digesta although epithelial

cells were not damaged.

DISCUSSION

Continuous infusion without a priming dose did not cause adverse effects in
previous experiments (Chapter 3 and 4). In addition, no relationship was observed
between type of VFA infused and occurrence of metabolic disorders in this experiment.
Therefore, the priming dose of hyper-osmotic solutions likely caused the metabolic
disorders observed. Dry matter intake, water intake, and serum concentrations of sodium
and potassium for cows downed by infusion treatment (DOWN) were compared to those
of cows that appeared to be normal (NORM) to identify the factors characterizing the
adverse effect of the infusion protocol. Serum concentrations of sodium and potassium

were determined by atomic absorption spectrometry after digestion with hydrogen

189

peroxide. Dry matter intake was lower for DOWN compared to NORM (4.0 vs. 10.1
kg/12h; Table 1). Although infusion of sodium VFA increased water intake, water intake
was increased less by DOWN than by NORM. Compared to the average for 3 (1 prior to
the infusion period, infusion of sodium VFA increased water intake by 24% for DOWN

(111.3 vs. 88.2 U12h), but by 53% for NORM (126.0 vs. 85.2 U12h).

These observations lead us to speculate that a priming dose of hyper-osmotic
solution of sodium VFA resulted in systemic dehydration great enough to affect
subsequent drinking behavior. Although cows are tolerant to sodium chloride at
maximum concentration of 9.0% of dietary DM (Meyer et al., 1955), the total amount of
sodium infused over 14 h was expected to be less than 4% of dietary DM in this
experiment. A pulse dose may result in greater adverse effects compared to continuous
infusion or feeding because cows are given less time to respond to infusion of hyper-
osmotic solutions. The priming dose of VFA salt increased osmolarity of ruminal fluid,
but cows might not have been able to drink enough water to adjust for hyper-osmolarity
of the ruminal fluid, and water flux from blood to the rumen caused systemic dehydration
for all cows. Hyper-osmolarity in duodenal digesta decreases gastric emptying
(Ruckebusch, 1993) and might further worsen dehydration status by limiting water intake
because of ruminal distention. Necropsy revealed abomasums impacted by dry digesta
for 2 DOWN cows, indicating that the decrease in gastric emptying might have been

more severe for DOWN.

190

TABLE 1. Characteristics of cows that downed by the infusion protocol
(DOWN) and those appeared to be normal (NORM).

 

 

NORM DOWN

Days in milk 97.3 43.8
BW 664 670
BCS 2.79 2.80
DMI (kg/12h)

3d-collection period before infusion 13.1 11.6

Infusion day 10.1 4.0
Water intake (Ll 12h)

3d-collection period before infusion 85.2 88.2

Infusion day 126.0 1 11.3
Serum Na (mM)

3d-collection period before infusion 129.6 124.6

Infusion day 136.2 138.1
Serum K (mM)

3d-collection period before infusion 4.59 4.75

Infusion day 3.55 3.10

 

191

Infusion of sodium with VFA can cause metabolic alkalosis, and decrease serum
concentration of potassium by activating a potassium/ H+ pump: intracellular protons are
exchanged with potassium in blood to maintain blood pH (NRC, 2001). Although
infusion of sodium VFA decreased serum concentration of potassium for all cows,
potassium concentration to greater extent for DOWN compared to NORM. Compared to
the average of 3 d prior to the infusion period, infusion of sodium VFA decreased serum
potassium concentration by 35% for DOWN (4.75 vs. 3.10 mM), but only by 23% for
NORM (4.59 vs. 3.55 mM). DOWN cows experienced more severe hypokalemia
possibly because of less water intake and DMI compared to NORM. Metabolic alkalosis
might also induce conformational change in PTH receptors (Goff et al. 1991), and some
cows could not stand or walk properly due to hypocalcemia (less than 0.7 mM of ionized
Ca concentration). In addition, cows were observed to slow down respiration rate to
correct metabolic alkalosis. Two cows increased PCO2 drastically (possibly resulting in

permanent cell damage) and died.

CONCLUSION

A priming dose of hyper-osmotic solutions resulted in dehydration. In addition,
infusion of sodium VFA solution affected acid-base balance in serum, resulting in
metabolic alkalosis, hypokalemia, and hypocalcemina. A priming dose of hyper-osmotic

solution should not be administered for future experiments.

192

REFERENCES

Allen, M. S. 1997. Relationship between fermentation acid production in the rumen and
the requirement for physical effective fiber. J. Dairy Sci. 80: 1447-1462.

Allen, M. S. 2000. Effects of diet on short-term regulation of feed intake by lactating
dairy cattle. J. Dairy Sci. 83: 1598-1624.

Allen, M. S., M. Oba, and C. S. Mooney. 2000. Automated system for collection of
ruminal fluid and blood of ruminants. J. Dairy Sci. 83 (Suppl. l):288.

Aldrich, M. M., L. D. Muller, G. A. Varga, and L. C. Griel, Jr. 1993. Nonstructural
carbohydrate and protein effects on ruminal fermentation, nutrient flow, and performance
of dairy cows. J. Dairy Sci. 76: 1091-1 105.

Amaral-Phillips, D. M., A. D. McGilliard, G. L. Lindberg, J. J. Veenhuizen, and J. W.
Young. 1993. Effects of decreased availability of glucose for dairy cows. J. Dairy Sci.
76:752-761.

Anil, M. H., and J. M. Forbes. 1980. Feeding in sheep during intraportal infusions of
short-chain fatty acids and the effect of liver denervation. J. Physiol. 298:407-414.

Anil, M. H., and J. M. Forbes. 1988. The roles of hepatic nerves in the reduction of food
intake as a consequence of intraportal sodium propionate administration in sheep. Q. J.
Exp. Physiol. 73:539-546.

Anil, M. H., J. N. Mbanya, H. W. Symonds, and J. M. Forbes. 1993. Responses in the
voluntary intake of hay or silage by lactating cows to intraruminal infusions of sodium

acetate or sodium propionate, the tonicity of rumen fluid or rumen distension. Br. J.
Nutri. 69:699-712.

A.S.A.E. 1968. Method of determining and expressing fineness of feed material by
sieving. ASAE standard S319.

Baile, C. A., and W. H. Pfander. 1966. A possible chemosensitive regulatory
mechanism of ovine feed intake. Am. J. Physiol. 210: 1243-1248.

Baile, C. A., and J. Mayer. 1969. Depression of feed intake of goats by metabolites
injected during meals. Am. J. Physiol. 217:1830-1836.

Baile, C. A., and J. M. Forbes. 1974. Control of feed intake and regulation of energy
balance in ruminants. Physiol. Rev. 54:160-214.

193

 

 

 

Ballard, F. J. 1965. Glucose utilization in mammalian liver. Comp. Biochem. Physiol.
14:437-443.

Ballard, F. J ., R. W. Hanson, D. S. Kronfeld, and F. Raggi. 1969. Metabolic changes in
liver associated with spontaneous ketosis and starvation in cows. J. Nutr. 95: 160-173.

Bhattacharya, A. N., and M. Alulu. 1975. Appetite and insulin-metabolite harmony in
portal blood of sheep fed high or low roughage diet with or withour intraruminal infusion
of VFA. J. Anim. Sci. 41:225-233.

Bines, J. A., and S. V. Morant. 1983. The effect of body condition on metabolic changes
associated with intake of food by the cow. Br. J. Nutr. 50:81-89.

Burns, G. A., and R. C. Ritter. 1998. Visceral afferent participation in delayed satiation
following NMDA receptor blockade. Physiol. Behav. 65:361-366.

Butler, W. R., J. J. Calaman, and S. W. Beam. 1996. Plasma and milk urea nitrogen in
relation to pregnancy rate in lactating dairy cattle. J. Anim. Sci. 74:858-865.

Callison, S. L., J. L. Firkins, M. L. Eastridge, and B. L. Hull. 2001. Site of nutrient
digestion by dairy cows fed corn of different particle sizes or steam-rolled. J. Dairy Sci.
84: 1458-1467.

Casper, D. P., D. J. Schingoethe, and W. A. Eisenbeisz. 1990. Response of early
lactation dairy cows fed diets varying in source of non-structural carbohydrate and crude
protein. J. Dairy Sci. 73: 1039-1050.

Choi, B. R., and M. S. Allen. 1999. Intake regulation by volatile fatty acids and physical
fill. S. Afr. J. Anim. Sci. 29(ISRP):40—41(Abstr.).

Choung, J. J ., and D. G. Chamberlain. 1995. Effects of intraruminal infusion of
propionate on the concentrations of ammonia and insulin in peripheral blood of cows
receiving an intraruminal infusion of urea. J. Dairy Res. 62:549-557.

Cochran, R. C., D. C. Adams, J. D. Wallace, and M. L. Galyean. 1986. Predicting
digestibility of different diets with internal markers: evaluation of four potential markers.
J. Anim. Sci. 63: 1476-1483.

Conrad, H. R., A. D. Pratt, and J. W. Hibbs. 1964. Regulation of feed intake in dairy
cows. 1. Change in importance of physical and physiological factors with increasing
digestibility. J. Dairy Sci. 47:54.

Crocker, L. M., E. J. Depeter, J. G. Fade], H. Perez-Monti, S. J. Taylor, J. A. Wyckoff,

and R. A. Zinn. 1998. Influence of processed corn grain in diets of dairy cows on
digestion of nutrients and milk composition. J. Dairy Sci. 81:2394-2407.

194

Dado, R. G., and M. S. Allen. 1993. Continuous computer acquisition of feed and water
intake, chewing reticular motility, and ruminal pH of cattle. J. Dairy Sci. 76:1589-1600.

Dado, R. G., and M. S. Allen. 1995. Intake limitations, feeding behavior, and rumen
function of cows challenged with rumen fill from dietary fiber or inert bulk. J. Dairy Sci.
78: 1 18-133.

Davis, C. L. 1967. Acetate production in the rumen of cows fed either control or low-
fiber, high-grain diets. J. Dairy Sci. 50: 1621-1625.

Deetz, L. E., and P. J. Wangsness. 1980. Effect of intrajugular administration of insulin
on feed intake, plasma glucose and plasma insulin of sheep. J. Nutr. 110: 1976-1982.

Deetz, L. E., and P. J. Wangsness. 1981. Influence of intrajugular administration of
insulin, glucagon and propionate on voluntary feed intake of sheep. J. Anim. Sci. 53:
427—433.

Deetz, L. E., P. J. Wangsness, Kavanaugh, J. F., and L. C. Griel, Jr. 1980. Effect of
intraportal and continuous intrajugular administration of insulin of feeding in sheep. J.
Nutr. 110:1983-1991.

DeJong, A. 1982. Patterns of plasma concentrations of insulin and glucagon after
intravascular and intraruminal administration of volatile fatty acids in the goat. J.
Endocrinol. 92:357-370.

DeJong, A., A. B. Steffens, and L. De Ruiter. 1981. Effects of portal volatile fatty acid
infusions on meal patterns and blood composition in goats. Physiol. Behav. 27:683-689.

Demigné, C., C. Yacoub, C. Rémésy, and P. Fafournoux. 1986. Propionate and butyrate
metabolism in rat or sheep hepatocytes. Biochim. Biophys. Acta 875:535-542.

Dhiman, T. R., C. Cadomiga, L. D. Satter. 1993. Protein and energy supplementation of
high alfalfa silage diets during early lactation. J. Dairy Sci. 76:1945-1959.

Egan, A. R. 1966. Nutritional status and intake regulation in sheep V. Effects of
intraruminal infusions of volatile fatty acids upon voluntary intake of roughage by sheep.

Aust. J. Agric. Res. 17:741-755.

Elliot, J. M., H. W. Symonds, and B. Pike. 1985. Effect on feed intake of infusing sodium
propionate or sodium acetate into a mesenteric vein of cattle. J. Dairy Sci. 68: 1 165-1 170.

Emery, R. S., J. M. Liesman, and T. H. Herdt. 1992. Metabolism of long chain fatty
acids by ruminant liver. J. Nutr. 122:832-837.

195

 

Farningham, D.A.H., and C. C. Whyte. 1993. The role of propionate and acetate in the
control of food intake in sheep. Br. J. Nutr. 70:37-46.

Farningham, D.A.H, J. G. Mercer, and C. B. Lawrence. 1993. Satiety signals in sheep:
Involvement of CCK, propionate, and vagal CCK binding sites. Physiol. Behav. 54:437-
442.

Forbes, J. M. 1995. Voluntary Food Intake and Diet Selection in Farm Animals. CAB
International, Oxon, UK.

Foster, L. A., N. K. Ames, and R. S. Emery. 1991. Food intake and serum insulin
response to intraventricular infusions of insulin and IGF-1. Physiol. Behav. 50:745-749.

Friedman, M. I., and M. G. Tordoff. 1986. Fatty acid oxidation and glucose utilization
interact to control food intake in rats. Am. J. Physiol. 251 :R840-R845.

Friedman, M. I., R. B. Harris, H. Ji, 1. Ramirez, and M. G. Tordoff. 1999. Fatty acid
oxidation affects food intake by altering hepatic energy status. Am. J. Physiol.
276:R1046-R1053.

Frobish, R. A., and C. L. Davis. 1977. Effects of abomasal infusions of glucose and
propionate on milk yield and composition. J. Dairy Sci. 60:204-209.

Gaynor, P. J ., K. J. Lookingland, and H. A. Tucker. 1995. 5-hydroxytryptaminergic
receptor-stimulated growth hormone secretion occurs independently of changes in
peripheral somatostatin concentration. Proc. Soc. Exp. Biol. Med. 209:79-85.

Goering, H. K., and P. J. Van Soest. 1970. Forage Fiber Analysis (Apparatus, Reagents,
Procedures, and Some Applications). Agric. Handbook No. 379. ARS-USDA,
Washington, DC.

Goff, J. P, R. L. Horst, F. J. Mueller, J. K. Miller, G. A. Kiess, and H. H. Dowlen. 1991.
Addition of chloride to a prepartal diet high in cations increases 1, 25-dihydroxyvitamin
D response to hypocalcemia preventing milk fever. J. Dairy Sci. 74:3863-3871.

Gonda, H. L., J. E. Lindberg, and K. Holtenius. 1997. Plasma levels of energy
metabolites and pancreatic hormones in relation to the level of intake and intraruminal

infusions of volatile fatty acids in fed wether sheep. Comp. Biochem. Physiol. 116A:65-
73.

Grings, E. E., R. E. Roffler, and D. P. Deitelhoff. 1992. Evaluation of corn and barley as

energy sources for cows in early lactation fed alfalfa-based diets. J. Dairy Sci. 75: 193-
200.

196

 

Grovum, W. L. 1995. Mechanisms explaining the effects of short chain fatty acids on
feed intake in ruminants-osmotic pressure, insulin and glucagon. Pages 173-197 in
Ruminant Physiology: Digestion, Metabolism, Growth and Reproduction, W. v.
Englehardt, S. Leonhard-Marek, G. Breves, D. Geisecke, ed. Ferdinand Enke Verlag,
Stuttgart, Germany.

Guesnet, P. M., M. J. Massoud, and Y. Demame. 1991. Regulation of adipose tissue
metabolism during pregnancy and lactation in the ewe: The role of insulin. J. Anim. Sci.
69:2057-2065.

Hach, C. C., B. K. Bowden, A. B. Lopelove, and S. V. Brayton. 1987. More powerful
peroxide Kjeldahl digestion method. J. AOAC 70:783-787.

Haussinger, D., W. H. Lamers, A.F.M. Moorman. 1992. Hepatocyte heterogeneity in the
metabolism of amino acids and ammonia. Enzyme 46:72-93.

Herrera-Saldana, R. E., and J. T. Huber. 1989. Influence of varying protein and starch
degradabilities on performance of lactating dairy cows. J. Dairy Sci. 72: 1477-1483.

Holt, C. 1983. Swelling of Golgi vesicles in mammary secretory cells and its relation to
the yield and quantitative composition of milk. J. Theor. Biol. 101:247-261.

Hurtaud, C., Rulquin, H., and R. Verite. 1993. Effect of infused volatile fatty acids and

caseinate on milk composition and coagulation in dairy cows. J. Dairy Sci. 76:3011-
3020.

Ingvartsen, K. L., and J. B. Andersen. 2000. Integration of metabolism and intake
regulation: A review focusing on periparturient animals. J. Dairy Sci. 83:1573-1597.

Istasse, L., F.D. Deb. Hovel], N.A. Macleod, and ER. Orskov. 1987. The effects of
continuous or intermittent infusion of propionic acid on plasma insulin and milk yield in
dairy cows nourished by intragastric infusion of nutrients. Livest. Prod. Sci. 16:201.

Ji, H., G. Graczyk-Milbrandt, and M. I. Friedman. 2000. Metabolic inhibitors
synerfistically decrease hepatic energy status and increase food intake. Am. J. Physiol.
278:R1579-R1582.

Kalscheur, K. F., B. B. Teter, L. S. Piperova, and R. A. Erdman. 1997. Effect of dietary
forage concentration and buffer addition on duodenal flow of trans-Cm.1 fatty acids and

milk fat production in dairy cows. J. Dairy Sci. 80:2104-2114.

Karkalas, J. 1985. An improved enzymatic method for the determination of native and
modified starch. J. Sci. Food Agric. 36: 1019-1027.

197

Kennelly, J. J ., B. Robinsion, and G. R. Khorasani. 1999. Influence of carbohydrate
source and buffer on rumen fermentation characteristics, milk yield, and milk
composition in early—lactation Holstein cows. J. Dairy Sci. 82:2486-2496.

Knowlton, K. F., B. P. Glenn, and R. A. Erdman. 1998a. Performance, ruminal
fermentation, and site of starch digestion in early lactation cows fed corn grain harvested
and processed differently. J. Dairy Sci. 81:1972-1984.

Knowlton, K. F., T. E. Dawson, B. P. Glenn, G. B. Huntington, and R. A. Erdman.
1998b. Glucose metabolism and milk yield of cows infused abomasally or ruminally
with starch. J. Dairy Sci. 81:3248-3258.

Koch, J. E., H. Ji, M. D. Osbakken, and M. D. Friedman. 1998. Temporal relationships
between eating behavior and liver adenine nucleotides in rats treated with 2,5-anhydro-D
mannitol. Am. J. Physiol. 274:R610-R617.

Langhans, W. 1996. Role of the liver in the metabolic control of eating: what we know-
and what we do not know. Neurosci. Biobehav. Rev. 20: 145-153.

Langhans, W., and E. Scharrer. 1987a. Evidence for a vagally mediated satiety signal
derived from hepatic fatty acid oxidation. J. Auton. Nerv. System 18:13-18.

Langhans, W., and E. Scharrer. 1987b. Evidence for a role of the sodium pump of
hepatocytes in the control of food intake. J. Auton. Nerv. System 20: 199-205.

Langhans, W., F. Wiesenreitter, and E. Scharrer. 1983. Different effects of

subcutaneous D,L-3-hydroxybutyrate and acetoacetate injections on food intake in rats.
Physiol. Behav. 31:483-486.

Langhans, W., U. Damaske, and E. Scharrer. 1984. Subcutaneous glycerol injection
fails to reduce food intake in rats fed a high protein diet. Physiol. Behav. 32:785-789.

Langhans, W., U. Damaske, and E. Scharrer. 1985a. Different metabolites might reduce
food intake by the mitochondrial generation of reducing equivalents. Appetite 6: 143-
152.

Langhans, W., G. Egli, and E. Scharrer. 1985b. Regulation of food intake by hepatic
oxidative metabolism. Brain Res. Bull. 15:425-428.

Langhans, W., G. Egli, and E. Scharrer. 1985c. Selective hepatic vagotomy eliminates the
hypophagic effect of different metabolites. J. Auton. Nerv. Sys. 13:255-262.

Lee, J.A., J .D. Oldham, W. Harseign, and RC. Gamsworthy. 1990. The effects of

patterns of rumen fermentation on the response by dairy cows to protein concentration.
B. J. Nutri. 63:177.

198

Leuvenink H.G.D., E.J.B. Bleumer, L.J.G.M. Bongers, J. VanBruchem, and D.
VanDerHeide. 1997. Effect of short-term propionate infusion on feed intake and blood
parameters in sheep. Am. J. Physiol. E997-E1001.

Lobley, G. E., A. Connell, M. A. Lomax, D. S. Brown, E. Milne, A. G. Calder, and
D.A.H.F. Farningham. 1995. Hepatic detoxification of ammonia in the ovine liver:
possible consequences for amino acid catabolism. Br. J. Nutr. 73:667-685.

Lycos, T., G. A. Varga, and D. Casper. 1997. Varying degradation rates of total
nonstructural carbohydrates: Effects on ruminal fermentation, blood metabolites, and

milk production and composition in high producing Holstein cows. J. Dairy Sci.
80:3341-3355.

Maffei, M., J. Halaas, E. Ravussin, R. E. Pratley, G. H. Lee, Y. Zhang, H. Fei, S. Kim, R.
Lallone, S. Ranganathan, P. A. Kern, and J. M. Friedman. 1995. Leptin levels in human

and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced
subjects. Nature Med. 1:1155-1161.

Mbanya, J. N, M. H. Anil, and J. M. Forbes. 1993. The voluntary Intake of hay and silage
by lactating cows in response to ruminal infusion of acetate or propionate, or both, with
or without distension of the rumen by a balloon. Br. J. Nutr. 69:713-720.

McCarthy Jr., R. D., T. H. Klusmeyer, J. L., J. L. Vicini, J. H. Clark, and D. R. Nelson.
1989. Effects of source of protein and carbohydrate on ruminal fermentation and passage
of nutrients to the small intestine of lactating cows. J. Dairy Sci. 72:2002-2016.

McLeay, L. M., and M. A. Pass. 1996. Inhibitory effects of short intravascular infusions
of propionate on reticulo-rumen motility in the sheep. Comp. Biochem. Physiol.
115A263-71.

Merriam, G. R., and K. W. Wachter. 1982. Algorithms for the study of episodic
hormone secretion. Am. J. Phsyiol. 243:E310-E318.

Mertens, D. R. 1994. Regulation of forage intake. In: G. C. Fahey, Jr., M. Collins, D. R..
Mertens, and L. E. Moser (Eds). Forage Quality, Evaluation, and Utilization. American

Society of Agronomy, Crop Science Society of America, Soil Science Society of
America, Madison, WI.

Meyer, J. H., W. C. Weir, N. R. Ittner, and J. D. Smith. 1955. The influence of high
sodium chloride intakes by fattening sheep and cattle. J. Anim. Sci. 14:412-418.

Moore, J. A., M. H. Poore, T. P. Eck, R. S. Swingle, J. T. Huber, and M. J. Arana. 1992.

Sorghum grain processing and buffer addition for early lactation cows. J. Dairy Sci.
75:3465-3472.

199

 

Mutsvangwa, T., J. G. Buchanan-Smith, and B. W. McBride. 1997. Effects of ruminally
degradable nitrogen intake and in vitro addition of ammonia and propionate on the

metabolic fate of L-[l-‘4C] alanine and L-[l-‘5N] alanine in isolated sheep hepatocytes. J.
Anim. Sci. 75:1149-1159.

National Research Council. 1989. Nutrient Requirements of Dairy Cattle. 6th rev. ed.
Natl. Acad. Sci., Washington DC.

Niijima, A. 1983. Glucose-sensitive afferent nerve fibers in the liver and their role in
food intake and blood glucose regulation. J. Auton. Nerv. Sys. 9:207—220.

Oba, M., and M. S. Allen. 2000. Effect of brown midrib 3 mutation in corn silage on
productivity of dairy cows fed two levels of dietary NDF: 1. Feeding behavior and
nutrient utilization. J. Dairy Sci. 83:1333-1341.

Oba, M. and M. S. Allen. 2000. Effects of conservation method of corn grain and

dietary starch concentration on DMI and productivity of lactating dairy cows. J. Dairy
Sci. 83(Suppl 1): 248

Oba, M., and M. S. Allen. 2002a. Effects of corn grain conservation method on ruminal
digestion kinetics of lactating dairy cows at two dietary starch concentrations. J. Dairy
Sci. 85: submitted.

Oba, M., and M. S. Allen. 2002b. Effects of diet fermentability on efficiency of
microbial nitrogen production. J. Dairy Sci. 85: submitted.

Oliveira, J. S., J. T. Huber, D. Ben-Ghedalia, R. S. Swingle, C. B. Theurer, and M.
Pessarakli. 1993. Influence of sorghum grain processing on performance of lactating
dairy cows. J. Dairy Sci. 76:575-581.

Oliveira, J. S., J. T. Huber, J. M. Simas, C. B. Theurer, and R. S. Swingle. 1995.Effect of
sorghum grain processing on site and extent of digestion of starch in lactating dairy cows.
J. Dairy Sci. 78:1318-1327.

Orskov, E. R., W. P. Flatt, P. W. Moe, and A. W. Munson. 1969. The influence of
ruminal infusion of volatile fatty acids on milk yield and composition and on energy
utilization by lactating dairy cows. Br. J. Nutr. 23:443-453.

Overton, T. R., M. R. Cameron, J. P. Elliott, and J. H. Clark, and D. R. Nelson. 1995.
Ruminal fermentation and passage of nutrients to the duodenum of lactating cows fed
mixtures of corn and barley. J. Dairy Sci. 78:1981-1998.

Parker, D. S., M. A. Lomax, C. J. Sea], and J. C. Wilton. 1995. Metabolic implications
of ammonia production in the ruiminant. Proc. Nutr. Soc. 54:549-563.

200

Peters, J. P., E. N. Bergman, and J. M. Elliot. 1983. Changes of glucose, insulin and
glucagon associated with propionate infusion and vitamin B-12 status in sheep. J. Nutr.
113:1229-1240.

Plascencia, A., and R. A. Zinn. 1996. Influence of flake density on the feeding value of
steam-processed corn in diets for lactating dairy cows. J. Anim. Sci. 74:310-316

Poore, M. H., J. A. Moore, T. P. Eck, R. S. Swingle, and C. B. Theurer. 1993. Effect of
fiber source and ruminal starch degradability on site and extent of digestion in dairy
cows. J. Dairy Sci. 76:2244-2253.

Qian, M., S. W. Zhu, A. E. Johnson, P. Sodersten. 2000. Evidence that MK-801
stimulates intraoral intake by acting on hepatic afferents. Neuroreport 11:2617-2620.

Quigley, J. D., and R. N. Heitmann. 1991. Effects of propionate infusion and dietary
energy on dry matter intake in sheep. J. Anim. Sci. 69: 1 178-1 187.

Rawson, N. E., and M. I. Friedman. 1994. Phosphate-loading prevents the decrease in
ATP and increase in food intake produced by 2,5-anhydro-D-mannitol. Am. J. Physiol.
2662R1792-Rl796.

Rawson, N. E., H. Blum, M. D. Osbakken, and M. I. Friedman. 1994a. Hepatic
phosphate trapping decreased ATP and increased feeding after 2,5-anhydro-D-mannitol.
Am. J. Physiol. 266:R112-R117.

Rawson, N. E., P. M. Ulrich, and M. I. Friedman. 1994b. Ethionine, an amino acid
analogue, stimulates eating in rats. Am. J. Physiol. 267:R612-R615.

Rawson, N. E., P. M. Ulrich, and M. I. Friedman. 1996. Fatty acid oxidation modulates
the eating response to the fructose analogue 2,5-anhydro-D-mannitol. Am. J. Physiol.
271:R144-R-148.

Reynolds, C. K. 1992. Metabolism of nitrogenous compounds by ruminant liver. J.
Nutr. 122:850—854.

Reynolds, C. K. 1995. Quantitiative aspects of liver metabolism in ruminants. Pages 351-
372 in Ruminant Physiology: Digestion, Metabolism, Growth and Reproduction, W. v.
Englehardt, S. Leonhard-Marek, G. Breves, D. Geisecke, ed. Ferdinand Enke Verlag,
Stuttgart, Germany.

Ricks, C. A., and R. M. Cook. 1981. Regulation of fatty acid uptake by mitochondrial
acyl CoA synthetases of bovine liver. J. Dairy Sci. 64:2324-2335.

201

Rodriguez, L. A., C. C. Stallings, J. H. Herbein, and M. L. McGilliard. 1997. Diurnal
variation in milk and plasma urea nitrogen in Holstein and Jersey cows in response to
degradable dietary protein and added fat. J. Dairy Sci. 80:3368-3376.

Ruckebusch, Y. 1993. Motility of the gastro-intestinal tract. In The ruminant Animal
Digestive Physiology and Nutrition. Ed. D. C. Church. Waveland Press, Inc. Prospect
Hights, IL.

SAS. 1990. Segmented Model. pp.1162. in SAS/STAT” User's Guide, Version 6. SAS
Institute Inc., Cary, NC.

Schlosser, S. F., A. D. Burgstahler, and M. H. Nathanson. 1996. Isolated rat hepatocytes

can signal to other hepatocytes and bile duct cells by release of nucleotides. Proc. Natl.
Acad. Sci. 93:9948-9953.

Shaw, L., and P. C. Engel. 1985. The suicide inactivation of ox liver short-chain acyl-
CoA dehydrogenase by propionyl-CoA. Biochem. J. 230:723-731.

Sheperd, A. C., and D. K. Combs. 1998. Long-term effects of acetate and propionate on
voluntary feed intake by midlactation cows. J. Dairy Sci. 81:2240-2250.

Stangassinger, M., and D. Giesecke. 1986. Splanchnic metabolism of glucose and related
energy substrates. Pages 347-366 in Control of Digestion and Metabolism in Ruminants.
L. P. Milligan, W. L. Grovum, and A. Dobson, ed., Prentice Hall, Englewood Cliffs, NJ.

Surina-Baumgartner, D. M., W. Langhans, and N. Geary. 1995. Hepatic portal infusion
antibody infusion increases, but insulin does not alter, spontaneous meal size in rats. Am.
J. Physiol. 269:R978-R982.

Theurer, C. B., J. T. Huber, A. Delgado-Elorduy, and R. Wanderley. 1999. Invited
review: summary of steam-flaking corn or sorghum grain for lactating dairy cows. J.
Dairy Sci. 82:1950-1959.

Tordoff, M. G., R. Rafka, M. J. DiNovi, and M. I. Friedman. 1988. 2, S-anhydro-D-
mannitol: a fructose analogue that increases food intake in rats. Am. J. Physiol.
254:R150—R153.

Treece, B. R., M. Covasa, R. c. Ritter, and G. A. Burns. 1998. Delay in meal
termination follows blockade of N-methyl-D-aspartate receptors in the dorsal hindbrain.

Brain Res. 810:34-40.

Ulyatt, M. J. 1964. The effects of intra-ruminal infusions of volatile fatty acids on food
intake of sheep. N. Z. J. Agric. Res. 8:397-408.

202

Vandermeerschen-Doize, F., and R. Raquay. 1984. Effects of continuous long-term
intravenous infusion of long-chain fatty acids on feeding behavior and blood components
of adult sheep. Appetite 5: 137-146.

Vasilatos, R., and P. J. Wangsness. 1980. Changes in concentrations of insulin, growth
hormone and metabolites in plasma with spontaneous feeding in lactating dairy cows. J.
Nutr. 110:1479-1487.

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.

Vernon, R. G., M. C. Barber, and E. Finley. 1991. Modulation of the activity of acetyl-
CoA carboxylase and other lipogenic enzymes by growth hormone, insulin and

dexamethasone in sheep adipose tissue and relationship to adaptations to lactation.
Biochem. J. 274:543-548.

Welch, J. G. 1982. Rumination, particle size and passage from the rumen. J. Anim. Sci.
54:885-894.

Wheelock, J. V., J. A. F. Rook, and F. H. Dodd. 1965. The relationship in the cow
between the osmotic pressure of milk and of blood. J. Dairy Res. 32:79-88.

Wildman, E. E., G. M. Jones, P. E. Wagner, R. L. Boman, H. F. Troutt Jr., and T. N.
Lesch. 1982. A dairy cow body condition scoring system and its relationship to selected
production characteristics. J. Dairy Sci. 65:495-501.

Wilkerson, V. A., and B. P. Glenn. 1997. Energy balance in early lactation Holstein
cows fed corn grains harvested and processed differently. J. Dairy Sci. 80:2487-2496.

Wu, 2., F. T. Sleiman, C. B. Theurer, F. Santos, J. M. Simas, M. Francolin, and J. T.
Huber. 1994. Effect of isocaloric infusion of glucose in the rumen or propionate in the
duodenum. J. Dairy Sci. 77: 1556-1562.

Ying, Y., M. S. Allen, N. K. Ames, and M. H. VandeHaar. 2002. Effects of fineness of
grinding and conservation method of corn grain on productivity of Holstein dairy cows
before and after calving: 1.Nutrient digestibility and feeding behavior. J. Dairy Sci 85:
submitted.

203