PROPIONATE CONTROL OF FEED INTAKE: INTERACTION WITH LIPOLYTIC
STATE
By
Sarah Stocks

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Animal Science – Doctor of Philosophy
2013

ABSTRACT
PROPIONATE CONTROL OF FEED INTAKE: INTERACTION WITH LIPOLYTIC
STATE
By
Sarah Stocks
The concept that feed intake is controlled by hepatic oxidation of fuels has developed
over the last several decades, mostly with laboratory species. Propionate has been
implicated in the control of feed intake in ruminants, and the mechanism is consistent
with simulation of hepatic oxidation. Previous research in our laboratory has
demonstrated that intraruminal propionate infusion, compared with iso-osmotic acetate
infusion, is more hypophagic in early lactation than in mid-lactation dairy cows. Three
experiments were conducted to investigate the mechanisms related to the control of feed
intake in early lactation dairy cows. The first experiment was a randomized block design
with 30 lactating Holstein cows from 3 – 40 days in milk (DIM) to evaluate the extent of
hypophagia from intraruminal propionate infusion for cows in early lactation.
Intraruminal propionate infusion, compared with iso-osmotic acetate infusion, depressed
dry matter intake (DMI) 20%. There was an interaction between treatment and DMI such
that propionate was increasingly hypophagic for cows with greater hepatic acetyl CoA
concentration. The second experiment evaluated longer-term infusions over 3 d to
identify adaptive mechanisms using 12 lactating Holstein cows from 2 – 13 DIM in a
cross-over design experiment. Propionate infusion decreased DMI over the 3-d infusion
period, however a treatment by period interaction was detected. During period 1,
propionate infusion, compared to acetate infusion, decreased DMI by 18% by decreasing

meal size by 21%. Similar to the results of the first experiment, propionate was more
hypophagic than acetate for cows with greater hepatic acetyl CoA concentration.
However, there was no treatment by day interaction for DMI suggesting that cows did not
adapt to intraruminal propionate over the 3 d infusion period. The third experiment was
designed to evaluate the effects of intravenous lipid infusion on feed intake response to
intraruminal propionic acid infusion by lactating cows. Eight lactating Holstein cows
past-peak lactation were used in a replicated 4x4 Latin square experiment with a 2x2
factorial arrangement of treatments. Lipid infusion, compared to a saline control, did not
affect plasma non-esterified fatty acid (NEFA) and β-hydroxybutyrate (BHBA)
concentrations, but did increase hepatic acetyl CoA concentration. Intraruminal
propionate infusion, compared to a sham control, decreased DMI 15%, but lipid did not
affect DMI. There was no interaction between lipid and propionate for DMI, likely
because lipid infusion failed to mimic the metabolite concentrations of cows in a lipolytic
state. Collectively, these experiments provide strong evidence that propionate is more
hypophagic for cows in a lipolytic state. Additional research is needed to evaluate the
longer-term effects of propionate and its application to diet formulation for cows in the
postpartum period	
  
	
  

ACKNOWLEDGEMENTS

First, I would like to thank Dr. Michael Allen for his patience and persistence with me
during my time at Michigan State University. He never hesitated to offer me advice, support, or
prodding, and he seemed to always understand how to challenge me to achieve more even when
I wasn’t so sure myself. I will always be grateful for his guidance. I would also like to thank my
graduate committee members, Dr. Dale Romsos, Dr. Thomas Herdt, Dr. Richard Ehrhardt, and
Dr. Adam Lock for their time and mentoring. A special thank you goes out to Dr. Adam Lock
for serving as my committee chairman. I really appreciate all of Dr. Lock’s counsel; he really
seemed to have a knack for being there when I needed to talk.
I was incredibly lucky to have the opportunity to work with the Allen Laboratory
technicians Jackie Ying, Dave Main, and Dewey Longuski while at MSU. They were there with
me through all of the challenges and highlights during my experiments; it really wouldn’t have
been the same without them. I also had the opportunity to work with the most wonderful group
of graduate and undergraduate students during my time here. Without them, I would never have
remained sane, and I would never have been able to complete all of my infusion studies. There
is nothing like bonding over a card game in the dairy barn at 3 AM in freezing temperatures to
really get to know people.
I have to say a big thank you to everyone at the Dairy Cattle Teaching and Research
Center, particularly Bob Kreft and Rob West. I know that I pushed the envelope on a regular
basis, but I appreciate the support and flexibility that allowed me to conduct some really exciting
research.

iv

Finally, thank you to my family for your continued love and encouragement. Bryan
could not possibly have been more understanding and supportive of me while I attended MSU,
and I am grateful every day that he is a part of my life. My parents taught me that I could
achieve anything I set my mind on, and they continue to challenge me to reach for more. I am
incredibly blessed to have such wonderful role models to look up to. And lastly, I would like to
thank my brother, sister (in-law), and nephews for always being there to put a smile on my face.

v

TABLE OF CONTENTS

LIST OF TABLES ………………………………………………………………………ix
LIST OF FIGURES ………………………………………………………………………xi
KET TO ABBREVIATIONS

………………………………………………………xii

CHAPTER 1
LITERATURE REVIEW
Metabolism and dry matter intake during the transition period
………………………1
Hepatic oxidation theory and control of feed intake ………………………………………2
Hepatic oxidation in dairy cows
………………………………………………………4
Propionate
………………………………………………………………………5
Starch ………………………………………………………………………………6
Fatty acids
………………………………………………………………………6
Glycerol
………………………………………………………………………8
Control of feed intake by other mechanisms ………………………………………………8
Central nervous system
………………………………………………………8
Hypothalamic neuropeptides ………………………………………………………10
Hormonal Mechanisms
………………………………………………………………11
Insulin ………………………………………………………………………………11
Leptin ………………………………………………………………………………12
Cholecystokinin
………………………………………………………………12
Ghrelin
………………………………………………………………………13
Gastrointestinal tract ………………………………………………………………………13
Fill
………………………………………………………………………………13
Enterocyte signaling ………………………………………………………………14
Specific mechanisms for control of feed intake in early lactation ………………………15
High-energy prepartum diets ………………………………………………………16
High-fill, low-energy prepartum diets
………………………………………17
Postpartum feeding strategies ………………………………………………………18
Conclusions ………………………………………………………………………………19
REFERENCES
………………………………………………………………………20
CHAPTER 2
HYPOPHAGIC EFFECTS OF PROPIONATE INCREASE WITH ELEVATED
HEPATIC ACETYL COA CONCENTRATION FOR COWS IN THE EARLY POSTPARTUM PERIOD
ABSTRACT ………………………………………………………………………………29
INTRODUCTION ………………………………………………………………………30
MATERIALS AND METHODS
………………………………………………………32
Animals, housing, and diets ………………………………………………………32
Experimental design and treatments ………………………………………………33

vi

Data and sample collection ………………………………………………………34
Analysis of samples ………………………………………………………………35
Statistical analysis
………………………………………………………………36
RESULTS
………………………………………………………………………………38
Variation among animals during the covariate period
………………………38
Feed intake, total ME intake, and feeding behavior ………………………………40
Ruminal pH, concentration and profile of VFA, and ammonia-N concentration
41
Plasma metabolites and hormones before and after infusion ………………………43
Interactions between treatments and indicators of metabolic status ………………45
DISCUSSION
………………………………………………………………………47
CONCLUSIONS
………………………………………………………………………51
REFERENCES
………………………………………………………………………52
CHAPTER 3
HYPOPHAGIC EFFECTS OF PROPIONIC ACID ARE NOT ATTENUATED DURING
A THREE-DAY INFUSION IN THE EARLY POSTPARTUM PERIOD IN HOLSTEIN
COWS
ABSTRACT ………………………………………………………………………………57
INTRODUCTION ………………………………………………………………………59
MATERIALS AND METHODS
………………………………………………………61
Animals, housing, and diets ………………………………………………………61
Experimental design and treatments ………………………………………………62
Covariate sample and data collection ………………………………………………63
Infusion period data and sample collection ………………………………………64
Analysis of samples ………………………………………………………………64
RNA extraction and real-time quantitative PCR
………………………………65
Statistical analysis
………………………………………………………………65
RESULTS
………………………………………………………………………………67
Feed intake and feeding behavior
………………………………………………67
Milk production and energy balance ………………………………………………67
Interaction between treatment and day of infusion ………………………………69
Interaction between treatment and hepatic acetyl CoA on intake and feeding
behavior
………………………………………………………………………72
Ruminal pH, VFA concentration and profile, and ammonia-N
………………74
DISCUSSION ………………………………………………………………………………75
CONCLUSIONS
………………………………………………………………………79
APPENDIX ………………………………………………………………………………80
REFERENCES
………………………………………………………………………87
CHAPTER 4
EFFECTS OF LIPID AND PROPIONIC ACID INFUSIONS ON FEED INTAKE OF
LACTATING DAIRY COWS
ABSTRACT ………………………………………………………………………………91
INTRODUCTION ………………………………………………………………………93
MATERIALS AND METHODS
………………………………………………………95
Animals, housing, and diets ………………………………………………………95

vii

Experimental design and treatments ………………………………………………96
Data and sample collection ………………………………………………………96
Analysis of samples ………………………………………………………………97
Statistical analysis
………………………………………………………………98
RESULTS
………………………………………………………………………………100
Plasma metabolite and hormone and liver metabolite concentrations
………100
Milk production and feed intake
………………………………………………103
Ruminal pH and concentration and profile of VFA ………………………………106
DISCUSSION ………………………………………………………………………………108
CONCLUSIONS
………………………………………………………………………113
REFERENCES
………………………………………………………………………114
CHAPTER 5
EVALUATION OF THE EFFECT OF INTRAJUGULAR PROPIONATE RELATIVE
TO GLYCEROL INFUSION ON FEED INTAKE IN EARLY LACTATION DAIRY
COWS: A PRELIMINARY STUDY
INTRODUCTION ………………………………………………………………………118
MATERIALS AND METHODS
………………………………………………………121
Animals, housing, and diets ………………………………………………………121
Experimental design and treatments
………………………………………122
Data and sample collection ………………………………………………………123
RESULTS AND DISCUSSION
………………………………………………………124
CONCLUSIONS
………………………………………………………………………129
REFERENCES
………………………………………………………………………130
CHAPTER 6
PRACTICAL IMPLICATIONS AND FUTURE RESEARCH ………………………133

viii

LIST OF TABLES

Table 2.1. Ingredients and nutrient composition of experimental diet (% of dietary
DM except for DM).
……………………………………………………33
Table 2.2. Characteristics of cows during the covariate period.

……………………38

Table 2.3. Effects of intraruminal infusion of sodium propionate relative to sodium
acetate on feeding behavior and energy intake for cows in early lactation..…41
Table 2.4. Effect of intraruminal infusion of sodium propionate relative to sodium
acetate on rumen VFA and ammonia-N concentration changes throughout
the 18 h infusion period (reported as post-infusion – pre-infusion).
……42
Table 2.5. Effect of intraruminal infusion of sodium propionate relative to sodium
acetate on plasma metabolite and hormone concentration changes
throughout the 18 h infusion period (reported as post-infusion –
pre-infusion).
……………………………………………………………44
Table 3.1. Ingredients and nutrient composition of experimental diet (% of dietary
DM except for DM).
……………………………………………………62
Table 3.2. Effects of intraruminal infusion of propionic acid relative to acetic acid
on feeding behavior and intake for cows in early lactation (overall effects
of treatment).
……………………………………………………………68
Table 3.3. Effects of intraruminal infusion of propionic acid relative to acetic acid
on feeding behavior and intake for cows in early lactation during period 1
and period 2.
……………………………………………………………68
Table 3.4. Effect of intraruminal infusion of propionic acid relative to acetic acid on
feeding behavior for cows in early lactation during period 1 only.
……69
Table 3.5. Effect of intraruminal infusion of propionic acid relative to acetic acid on
plasma metabolite and hormone concentrations and hepatic acetyl CoA
during period 1. ……………………………………………………………72
Table 3.6. Effect of intraruminal infusion of propionic acid relative to acetic acid on
feeding behavior for cows in early lactation during period 1. ……………73

Table 3.7. Primer sequences and accession numbers for genes analyzed
with real-time quantitative PCR. ……………………………………………82

ix

Table 3.8. Effect of intraruminal infusion of propionic acid, relative to
acetic acid, on hepatic gene expression during period 1. Cov = covariate
period, D1 = day 1 relative to the start of the infusion, D3 = day 3 relative
to the start of the infusion.
……………………………………….……84
Table 4.1. Ingredients and nutrient composition of experimental diet (% of dietary
DM except for DM).
……………………………………………………95
Table 4.2. Plasma metabolites and hormones and hepatic acetyl CoA
concentrations at the termination of infusion for cows post-peak lactation.....101
Table 4.3. Changes in plasma metabolites and hormones and hepatic acetyl CoA
from the covariate day to post-infusion for cows post-peak lactation………..102
Table 4.4. Effects of treatment on milk yield and components for cows post-peak
lactation. ……………………………………………………………………104
Table 4.5. Effects of treatment on feeding behavior for cows post-peak lactation……...105
Table 4.6. Effects of treatment on rumen fermentation for cows post-peak lactation…...107
Table 5.1. Ingredients and nutrient composition of experimental diet (% of dietary
DM except for DM). Estimated diet composition and nutrient
composition.
……………………………………………………………122
Table 5.2. Temperature, pulse, and respiration recordings for cows 4575 and 4499
collected during the infusion experiment.……………………………...…….125
Table 5.3. Daily feed intake and milk production for cow 4575 and 4499.
Prop = propionic acid infusion; Glyc = glycerol infusion……………………127

x

LIST OF FIGURES

Figure 2.1. Relationship between liver acetyl CoA concentration during the covariate
day and plasma BHBA concentration during the covariate day (r2 = 0.24,
P = 0.006 quadratic).
……………………………………………………45
Figure 2.2. Interaction (P= 0.05) between treatment and liver acetyl CoA
concentration for DMI during the 12 h infusion period in period 1 for
cows infused with acetic acid (dotted line) or propionic acid (solid line)
intraruminally. Equation: DMI = 10.29 + 1.43*A – 0.0026*ACoA +0.033
*(ACoA – 29.43)*A where A = acetic acid treatment and A
CoA = concentration of liver acetyl CoA in the covariate period.
……46
Figure 3.1. The interaction (P = 0.10) between treatment infusion and meal
frequency in infusion period 1 (days 3 – 5) for cows infused
intraruminally with either propionic or acetic acid.
……………………70
Figure 3.2. Interaction (P = 0.001) between treatment and day for plasma BHBA
concentration from the covariate day to infusion period 1 (days 3 – 5) for
cows infused intraruminally with either propionic or acetic acid.
……71
Figure 3.3. Interaction (P= 0.07) between treatment and liver acetyl CoA
concentration for DMI during the 3 d infusion in period 1 (days 3 – 5) for
cows infused intraruminally with propionic acid (solid line) or acetic acid
(dotted line).
……………………………………………………………74
Figure 5.1. Feed intake (as fed) for cows receiving propionic acid or glycerol
infusion in a switchback design experiment in early lactation. The ration
DM was ~43%. Panel A = Cow 4575; panel B = Cow 4499.
PROP = propionic acid, GLYC = glycerol.
……………………………128

xi

KEY TO ABBREVIATIONS

AgRP = agouti related peptide
BHBA = beta hydroxybutyric acid
CCK = cholecystokinin
CO = control
CP = crude protein
CPT = carnitine palmitoyltransferase
DIM = days in milk
DMI = dry matter intake
FA = fatty acid
LCFA = long-chain fatty acid
LI = lipid infusion
NDF = neutral detergent fiber
NEFA = non-esterified fatty acid
NPY = neuropeptide Y
PI = propionate infusion
POMC = proopiomelanocortin
SI = saline infusion
TCA = tri-carboxylic acid
TMR = total mixed ration
Trt = treatment
VFA = volatile fatty acid

xii

CHAPTER 1

LITERATURE REVIEW

Metabolism and dry matter intake during the transition period
During the transition period, from ~3 wk pre- to ~3 wk post-parturition, dairy cows
experience dramatic changes in metabolism to support fetal growth, parturition, and lactogenesis
(Bell, 1995). Fetal nutrient demands peak during the last trimester of pregnancy and shifts in
tissue sensitivity and hormone concentration occur to support the fetus. Mammary development
is maximal during this time when fetal nutrient demands are increasing. Glucose accounts for
50-75% of the total fuel oxidized by the fetus and amino acids account for the remainder of
energy available to the fetus (Bauman and Currie, 1980), while the fetus utilizes negligible
energy from FA. Gluconeogenesis increases and glucose uptake by peripheral tissues decreases
in late-pregnant ewes, sparing glucose for the fetus (Bell, 1995). Bell (1995) estimated that
glucose demand increases 2.7 fold, amino acid demand increases 2 fold, and fatty acid (FA)
demand increases 4.5 fold for the lactating mammary gland within days of calving compared
with fetal demands at 250 d in gestation. During late gestation, insulin concentration decreases
and adipose tissue sensitivity to insulin also decreases (Bell, 1995), resulting in a decrease in
adipose tissue lipogenesis and an increase in lipolysis. This increases the pool of circulating
non-esterified fatty acids (NEFA), and the liver extracts NEFA from the blood in proportion to
its concentration in the blood (Emery et al., 1992). NEFA can then be re-esterified and stored,
exported as very low density lipoprotein, or oxidized to acetyl CoA, which, in turn, is oxidized in
the tri-carboxylic acid (TCA) cycle or converted to ketone bodies and exported from the liver

1

(Emery et al., 1992). These shifts in nutrient partitioning provide energy for the cow, but also
might depress dry matter intake (DMI).
DMI gradually decreases in the weeks preceding parturition (Douglas et al., 2006; Gulay
et al., 2004; and Rabelo et al., 2003). Rabelo et al. (2003) reported that DMI was reduced for
cows fed high or low energy density diets during the last 3 wk prior to calving but the magnitude
of the reduction was greatest for cows fed a higher energy diet. Douglas et al. (2006) fed cows
either ad libitum or restricted diets and reported that ad libitum fed cows had a 47% total
reduction in DMI prior to calving while cows fed a restricted diet maintained DMI over the same
time period. Gulay et al. (2004) reported that DMI decreased during the last 2 wk prior to
calving in concert with decreased plasma insulin and increased plasma NEFA concentration.
Changes in DMI have been proposed to be a result of changes in metabolism discussed above,
including decreased insulin concentration and sensitivity resulting in increased plasma NEFA
concentration and subsequent oxidation in the liver, generating ATP, increasing energy charge,
and reducing intake (Allen et al., 2005).
Hepatic oxidation theory and control of feed intake
Mayer (1953) first proposed that glucose concentration in the blood could regulate feed
intake. Later, Russek (1963) reported that intraperitoneal injections of glucose stimulated
anorexia within 1.5 min, leading to his proposal that feeding behavior is controlled by nutrient
supply to the liver. He suggested that a hepatic receptor could inform the central nervous system
about the intracellular concentration of glucose, leading to cessation of feeding. Niijima (1969)
demonstrated that the firing rate of the hepatic vagal afferent nerves was related to blood glucose
concentration and that the firing rate decreased with intraportal glucose infusion (Niijima, 1981).

2

Additionally, decreasing glucose oxidation using pharmacological agents stimulated feeding in
rats (Del Prete et al., 2004).
Oxidation of fatty acids in the liver may also be involved in the control of intake.
Scharrer and Langhans (1986) demonstrated inhibition of acyl-CoA dehydrogenase, which
catalyzes the initial step in FA β-oxidation, by mercaptoacetate stimulates feeding in rats.
Inhibiting carnitine palmitoyltransferase (CPT) 1 using methyl palmoxirate stimulated feeding in
rats fed a diet high in long-chain fatty acids (LCFA), but not when rats received a diet high in
medium-chain FA (Friedman et al., 1986). CPT 1 is an enzyme responsible for transporting
LCFA into the mitochondria, but is not involved in the transport of medium-chain FA into the
mitochondria, supporting the lack of a response of CPT 1 inhibition on intake for the mediumchain FA fed rats.
Consistent with the results of the studies above, inhibiting both glycolysis (with 2deoxyglucose) and FA oxidation (with methyl palmoxirate) resulted in additive effects on food
intake in rats (Friedman and Tordoff, 1986). Additionally, inhibiting glycolysis (with 2deoxyglucose) and lipolysis (with niacin) combined resulted in increased food intake in rats
(Friedman et al., 1986). Collectively, this work suggests that there is link between oxidation of
glucose and/or FA in the liver and the control of food intake in non-ruminants.
Langhans et al. (1985a) conducted a series of experiments in rats injecting several
metabolites including lactate, glycerol, malate, and their oxidation products pyruvate,
dihydroxyacetone, or oxaloacetate. Subcutaneous injection of pyruvate, lactate, glycerol, and
malate, but not dihydroxyacetone or oxaloacetate, resulted in reduced food intake in rats fed high
carbohydrate diets. As a result, the authors concluded that the generation of reducing equivalents
by hepatic oxidation of those fuels contributed to the control of food intake in rats. In a similar

3

experiment, Langhans et al. (1985b) injected glycerol, malate, D-3-hydroxybutyrate, and lactate,
or their oxidation products dihydroxyacetone, oxaloacetate, acetoacetone, and pyruvate
subcutaneously in rats, and also preformed selective hepatic vagotomy to evaluate the role of the
liver in control of feed intake by oxidation of fuels. Again, glycerol, malate, and D-3hydroxybutyrate injection reduced food intake while injection of their oxidation products did not,
supporting the hypothesis that generation of reducing equivalents through oxidation of fuels
controls intake. Lactate and pyruvate both decreased food intake when injected into rats that
were fed high carbohydrate diets, but not for rats fed high fat diets, which might have been a
result of a reduction in pyruvate dehydrogenase activity during fat feeding (Begum et al., 1982;
Begum et al., 1983). Additionally, glycerol, malate, D-3-hydroxybutyrate, lactate, and pyruvate
injection failed to illicit a response when the rats were vagotomized (Langhans et al., 1985b),
demonstrating that there is a direct connection between the liver and the brain that is involved in
the control of food intake.
Hepatic oxidation in dairy cows
Hepatic oxidation of fuels may control intake in ruminants as well, but the fuels oxidized
are different than those oxidized by non-ruminants. As reviewed by Allen (2000), glucose
infusions into a variety of ruminants failed to affect feed intake despite evidence in nonruminants discussed above demonstrating consistent hypophagic effects of glucose, likely
because glucose is not oxidized in the ruminant liver. Allen (2000) supported this with the
following evidence: glucose is not readily extracted from the blood by the ruminant liver
(Stangassinger and Giesecke, 1986) because the activity of glucokinase is extremely low
(Ballard, 1965), and as a result, uptake of glucose by the liver is low. Acetate is also not
metabolized to any appreciable extent in the ruminant liver; hepatic uptake of acetate is low

4

(Reynolds, 1995) due to low activity of acetyl CoA synthetase in the liver (Ricks and Cook,
1981). Several fuels are metabolized extensively in the ruminant liver, including FA, amino
acids, lactate, glycerol, and propionate (Allen et al., 2005), and may be involved in the control of
feed intake.
Propionate
Allen (2000) suggested that of the fuels oxidized in the ruminant liver, propionate is the
most likely fuel to be implicated in the control of feed intake in dairy cows because its flux to the
liver increases greatly during meals (Benson et al., 2002) and hepatic extraction of propionate is
~70% (Reynolds et al., 2003). Sheperd and Combs (1998) reported that intraruminal propionate
infusion decreased DMI in lactating dairy cows in midlactation compared to isoenergetic
intraruminal acetate infusion. In a dose response experiment conducted in our laboratory, the
hypophagic effect of propionate was greater as the percent of propionate in the infusion was
increased, and propionate was more hypophagic for cows in early lactation compared to cows in
mid-lactation (Oba and Allen, 2003a). In that experiment, propionate reduced DMI by reducing
meal size while not affecting intermeal interval, a finding that indicates satiety. Propionate is
increasingly hypophagic for cows with elevated plasma glucose concentration (Oba and Allen,
2003b), likely because gluconeogenesis may be down regulated and oxidation of propionate is
faster within meals (Allen et al., 2009). Anil and Forbes (1980) demonstrated that hepatic
vagotomy eliminated the hypophagic effect of propionate in sheep, consistent with results of
hepatic vagotomy in non-ruminants utilizing other fuels. During in vitro incubations of ovine
hypopthalamic tissue, propionate failed to increase the concentration of mRNA coding for
neuropeptide Y (NPY), agouti related peptide (AgRP), and proopiomelanocortin (POMC;
Relling et al., 2012). NPY and AgRP are neuropeptides that stimulate feed intake (Leibowitz,

5

1991; Wagner et al., 2004) and POMC is a neuropeptide that has been shown to depress feed
intake (Millington, 2007). The failure of propionate to affect the expression of these
neuropeptides in the work of Relling et al. (2012) supports the hypothesis that the hypophagic
effect of propionate is mediated through peripheral signals rather than a central effect.
Starch
Propionate is the predominant volatile fatty acid (VFA) produced during the fermentation
of starch in the rumen (Davis, 1967). Ruminal fermentability of starch can vary due to
differences in the type of grain, conservation method, and processing method (Huntington,
1997). The flux of propionate to the liver increases following feeding (Benson et al., 2002), and
is greater with highly fermentable starch sources. Allen (2000) reviewed the effect of
fermentable grains on DMI and reported ~13% reduction in intake when more fermentable grains
replaced less fermentable grains in the diets of lactating dairy cows. Consistent with this effect,
Oba and Allen (2003c) reported that diets with greater ruminal starch fermentability decreased
meal size 17% and DMI 8% in lactating cows. This response was likely because of an increase
in the rate of production of VFA, particularly propionate, in the rumen with the more
fermentable starch diet. Similarly, Dann et al. (1999) reported that steam flaked corn (a more
ruminally fermentable starch source) tended to reduce DMI compared with cracked corn when
fed to cows during the postpartum period. The results of these experiments provide evidence
that propionate, from diets high in ruminally fermentable starch, is involved in the control of
feed intake.
Fatty acids
While FA oxidation is implicated in the control of feed intake in non-ruminants as
discussed above, limited work in ruminants exists to establish a relationship between hepatic FA

6

oxidation and control of feed intake. Supplemental fat sources can be free FA or triglycerides,
and can vary in chain length and saturation, which might explain the variable effect of fat
supplementation on DMI as discussed by Allen (2000). Stimulating FA oxidation in ruminants
by supplementation of carnitine, which transports long-chain FA into the mitochondria for
oxidation, in transition cows resulted in decreased DMI (Carlson et al., 2007). However,
inhibiting β-oxidation in ruminants using mercaptoacetate failed to increase DMI in heifers (Choi
et al., 1997). Allen et al. (2009) proposed that this was likely because activation of
metcarptoacetate to mecaptoacetyl CoA by acetyl CoA synthetase is likely necessary, and
activity of acetyl CoA synthetase is low in the ruminant liver. Unsaturated FA are more rapidly
oxidized than saturated FA (Leyton et al., 1987) and have been shown to cause greater reduction
in feed intake in dairy cows when fed (Harvatine and Allen, 2006) or abomasally infused
(Drackley et al., 1992).
In the liver, FA must enter the mitochondria to be oxidized. FA longer than 12 carbons
must be transported across the outer and inner mitochondrial membranes via CPT 1, and CPT 2,
respectively. FA of 12 carbons or less can freely pass into the mitochondria without need for
CPT 1 or CPT 2. Because there is no control point for oxidation of these medium and short
chain FA, they can be readily and rapidly oxidized resulting in depression in feed intake.
Consistent with this, Hristov et al. (2011) reported that feeding lauric acid (C12:0) decreased
DMI 25.7% and Dohme et al. (2004) reported an 18% reduction in DMI for cows fed a diet
supplemented with lauric acid. Additionally Hollmann and Beede (2012) reported that feeding
2% or 4% coconut oil (mostly medium chain FA) decreased DMI by up to 40% compared with
cows fed a control diet with no added fat. These results are consistent with effects of FA on
satiety through hepatic oxidation.

7

Glycerol
Glycerol is converted to glyceraldehyde-3-phosphate, which can either be used as a
gluconeogenic substrate, or be oxidized in the TCA cycle, dependent on the metabolic state of
the animal. However, glycerol is rapidly converted to propionate in the rumen (Rémond, 1993).
Glycerol is typically fed at ≤ 5% of the diet and often provided as a top dress in research studies
with low inclusion levels, and effects of glycerol on DMI are negligible (Defrain et al., 2004;
Chung et al. 2007). However at higher feeding rates, DMI may be reduced. Donkin et al. (2009)
reported that a diet containing 15% glycerol (substituted for ground corn) depressed DMI during
the first week of an 8-wk study, but did not affect DMI during the remaining weeks. Carvahlo et
al. (2011) included 11.5% or 10.8% glycerol (substituted for high moisture corn) in the pre- and
postpartum total mixed ration (TMR), respectively, from 28 days prior to calving through 56
DIM and reported that while intake was not different, there was a treatment by time interaction
prepartum; glycerol decreased intake for the first 10 days of the experiment. Glycerol supplied
as a drench can increase plasma glucose concentration (Goff and Horst, 2001). Absorbed
glycerol would not be expected to decrease DMI because glycerol is converted to
dihydroxyacetone phosphate, bypassing a rate-limiting step for gluconeogenesis and stimulating
glucose production rather than being oxidized in the TCA cycle. The effect of glycerol on feed
intake is consistent with the control of feed intake by hepatic oxidation; at high dietary addition
rates, glycerol is fermented to propionate causing hypophagia, and when provided as a drench, it
escapes ruminal fermentation and is used as a glucose precursor with little hypophagic effect.
Control of feed intake by other mechanisms
Central nervous system

8

Metabolites, neuropeptides, and hormones may all act on the central nervous system to
control feed intake, but research in dairy cows is limited. While propionate and long-chain FA
can cross the blood-brain barrier, their direct effects on intake are likely mediated through
hepatic mechanisms as discussed above; in the case of both metabolites, hepatic vagotomy
eliminated their effects on intake. Additionally, Oba and Allen (2003a) reported that propionate
is more hypophagic for cows in early lactation with higher plasma beta hydoxybutyric acid
(BHBA) concentration than for cows in mid-lactation. If propionate stimulated satiety through a
centrally mediated mechanism, the hypophagic effect of propionate likely would have been
reduced for the cows in early lactation with high plasma ketone concentration in that experiment
because BHBA can inhibit propionate transport across the blood-brain barrier as reviewed by
Allen et al. (2009). Ketone bodies such as BHBA can cross the blood brain barrier and BHBA
concentration in the cerebrospinal fluid is highly correlated to plasma BHBA concentration (R2 =
0.73; Sato et al., 2002). Laeger et al. (2012) reported that 50% feed restriction in cows tended to
increase plasma BHBA concentration (by almost 2-fold), but concentration of BHBA in the
cerebrospinal fluid was not affected. These authors speculated that BHBA might be utilized as a
fuel for the brain during feed restriction, preventing an increase in cerebrospinal BHBA
concentration and potentially inhibiting feed intake.
Glucose may be involved in feed intake control via a central mechanism. Injection of
glucose into the cerebrospinal fluid in the third brain ventricle in sheep decreased DMI (Seoane
and Baile, 1972), and Laeger et al. (2012) reported that feed restriction in dairy cows tended to
decrease glucose in the cerebrospinal fluid, consistent with a possible central role for glucose in
control of feed intake in ruminants. Laeger et al. (2012) also reported the potential of three
amino acids (serine, threonine, and tyrosine) to be hypophagic as their concentration in the

9

cerebrospinal fluid is decreased in feed restricted cows. Della-Fera and Baile (1980) reported
that continuous lateral ventricular infusion of cholecystokinin (CCK) decreased feed intake in
sheep, however, in vitro incubations of sheep hypothalamus tissue with CCK did not affect
concentration of neuropeptides involved in feed intake control (Relling et al., 2012). While
several metabolites and hormones may be involved in centrally mediated control of feed intake,
it is important to consider that alternative mechanisms, such as hepatic oxidation of fuels, may be
involved, as discussed for propionate and FA. Additional research is necessary to evaluate
control of feed intake by central mechanisms involving in dairy cows.
Hypothalamic neuropeptides
Signals relating to energy status from peripheral tissues are integrated in brain feeding
centers and involved in control of feed intake. As reviewed by Sartin et al. (2011), several
neuropeptides, including NPY and AgRP, are involved in transmitting peripheral signals to the
hypothalamus to control intake. NPY is produced in arcuate nucleus at the base of the
hypothalamus and involved in maintenance of energy balance and regulation of feed intake in
sheep and cattle as reviewed by Sartin et al. (2011). When administered to the hypothalamus,
NPY can cause hyperphagia in rats resulting in increased food intake and daily gain (Stanley et
al., 1986). Additionally, negative energy balance can increase NPY mRNA or peptide
concentration in some regions of the brain (Ahima et al., 1996; Kalra et al., 1991; Sahu et al.,
1988). AgRP is a neuropeptide that is synthesized in the hypothalamus and stimulates feed
intake in sheep (Wagner et al., 2004). POMC is a neuropeptide that is involved in decreasing
feed intake; mouse POMC knock-out models have an obese phenotype (Millington, 2007).
Hypothalamic NPY and AgRP mRNA expression is reduced in ad libitum fed sheep compared
with restricted fed sheep (Relling et al., 2010), consistent with the proposed mechanism for

10

intake control by neuropeptides discussed above. In sheep, fasting elevated plasma NEFA
concentration and increased gene expression of hypothalamic NPY and AgRP (Adam et al.,
2002). However, the effect might have been indirect because while feed restriction increased
NEFA concentration in blood, NEFA concentration in cerebrospinal fluid did not increase
(Laeger et al., 2012). The increase in NPY and AgRP gene expression appears to be a
mechanism to stimulate intake. No effects of feed restriction on POMC mRNA in the
hypothalamus have been shown in sheep or growing wethers (Adam et al., 2002; Relling et al.,
2010). Additional research evaluating the effects of peripheral signals on neuropeptide
concentration in the brain and the control of feed intake in ruminants is necessary to identify the
interactions between neuropeptides and other metabolic, hormonal, or physical mechanisms
involved in controlling feed intake.
Hormonal mechanisms
Insulin Insulin has been implicated in the control of feed intake in ruminants. In early
lactation cows subjected to a 48 h hyperinsulemic euglycemic clamp, DMI was reduced by 33%
compared with control (Leury et al., 2003). Larsen and Kristensen (2009) infused 1500 g/d of
glucose into dairy cows for the first 29 d postpartum resulting in a 3-fold increase in plasma
insulin concentration (P = 0.07) and a more limited rate of DMI increase compared with control
cows. Research in our laboratory showed a negative correlation between DMI of midlactation
cows fed a highly fermentable diet and plasma insulin concentration prior to the start of the
experiment (r2 = 0.28, P < 0.01, n = 29; Bradford and Allen, 2007a). Additionally, cows with a
greater insulin secretory response during a glucose tolerance test were not as susceptible to feed
intake suppression when fed a highly fermentable diet (r2 = 0.40, P < 0.01, n = 23 Bradford and
Allen, 2007a). It is possible that chronically elevated insulin may be involved in the control of

11

intake by reducing hepatic gluconeogenesis from propionate and stimulating hepatic oxidation of
propionate (Allen et al., 2005).
Leptin Leptin is secreted by white adipose tissue and reduces intake through action on the
central nervous system. In dairy cows, leptin concentration is positively correlated with BCS
(Ehrhardt et al., 2000), consistent with relationships between leptin and body fatness in humans
(Caro et al., 1996). Leptin has been shown to decrease intake in mice, rats, humans, and pigs
(Invartsen and Andersen, 2000). In periparturient dairy cows, plasma leptin concentration is
positively correlated with plasma insulin and glucose, and negatively correlated with plasma
NEFA (Block et al., 2001), suggesting a relationship between energy metabolism and leptin.
These authors speculated that the reduction in plasma leptin concentration during the early
postpartum period could result in a faster increase in DMI following parturition.
Cholecystokinin Choi and Palmquist (1996) were the first to report that feeding dairy
cows diets with high fat concentrations resulted in increased plasma concentration of
cholecystokinin (CCK), a peptide hormone that stimulates satiety via directly binding to CCK
receptors and by stimulating the vagal nerve. Relling and Reynolds (2007) reported that feeding
mono- or poly-unsaturated FA increased plasma CCK compared to saturated FA, and that the
depression in DMI associated with feeding rumen inert unsaturated FA was related to plasma
CCK concentration. Similarly, work in our laboratory demonstrated that feeding unsaturated FA
suppress feed intake in part due to an increase in plasma CCK concentration (Bradford et al.,
2008). Abomasal infusion of unsaturated FA or triglyceride failed to affect plasma CCK,
however unsaturated fat (FA or triglyceride) reduced DMI (Litherland et al., 2005; Benson et al.,
2001; Benson and Reynolds, 2001). Bradford et al. (2008) suggested 2 possible explanations for
the discrepancy in plasma CCK concentration between studies with fat feeding vs. abomasal fat

12

infusion: 1) abomasal infusion eliminates the oral-sensing responses to fat that may stimulate
CCK release, and 2) abomasal infusion prevents ruminal bio-hydrogenation of FA resulting in
decreased delivery of trans-C18:1 FA to the abomasum, which might limit CCK release.
Previous work in our laboratory showed that inhibiting CCK by intravenous injection MK-329
(devazepide) temporarily reversed the hypophagia caused by high-fat feeding in heifers (Choi et
al., 2000). The results of these experiments support a role for CCK in the control of feed intake
in dairy cows.
Ghrelin Ghrelin is a 28 amino acid peptide that is synthesized in abomasal and ruminal
tissue in cows. In ruminants, ghrelin has been shown to increase prior to conditioned meals
(Hayashida et al., 2001; Sugino et al., 2002) and with fasting (Wertz-Lutz et al., 2006),
suggesting that ghrelin in involved in stimulating intake in ruminants. Previous work in our
laboratory showed that plasma ghrelin concentration was elevated prior to the primary meal of
the day for cows in negative energy balance (Bradford and Allen, 2008). While pulse injections
of recombinant bovine ghrelin in steers did not affect plasma concentrations of insulin, NEFA, or
glucose, time spent eating was increased and DMI tended to be increased (Wertz-Lutz et al.,
2006). Collectively, these results suggest that ghrelin is involved in stimulating feed intake in
dairy cows.
Gastrointestinal tract
Fill Ruminal distension is a physical mechanism that can control feed intake in dairy
cows, particularly for cows near peak lactation or fed high forage diets as reviewed by Allen
(2000). Research in our laboratory (Dado and Allen, 1995) evaluated differences in DMI in
response to the filling effects of rumen-inert bulk and high or low dietary NDF. Inert bulk
coupled with low NDF did not affect DMI, but DMI was reduced for cows with the rumen-inert

13

bulk fed the high NDF diet, confirming that rumen fill can limit DMI under certain dietary
conditions. Consistent with those results, previous research in our laboratory showed that cows
fed diets based on brown midrib corn silage (which has lower lignin and higher digestibility than
conventional corn silage) had higher DMI regardless of dietary NDF concentration, likely due to
decreased rumen fill (Oba and Allen, 2000). In a 32 cow crossover design experiment conducted
in our laboratory, cows were fed diets with high or low forage to concentrate ratio (67:33 or
44:56, respectively); DMI was reduced for cows fed the high forage ration and time eating and
ruminating per kg forage NDF intake was reduced compared to cows fed the low forage diet
(Voelker et al., 2002). This likely resulted in higher rumen fill for the high forage diet compared
to the low forage diet (Voelker et al., 2002). However, as reviewed by Ingvartsen and Andersen
(2000), it is unlikely that ruminal fill is the dominant mechanism controlling feed intake through
the periparturient period, as DMI slowly increases after parturition. If intake prepartum was
limited by decreased capacity due to the fetus and associated membranes and fluid, intake should
increase rapidly when that pressure is relieved, but rather a gradual increase in DMI is observed
postpartum.
Enterocyte signaling Langhans (2008) proposed that energy intake is controlled by a
signal of FA oxidation occurring in the enterocyte, discounting control of feed intake based on
hepatic oxidation of fuels. Part of this speculation is due to work demonstrating that there is
limited innervation of hepatocytes by hepatic vagal nerves and the common hepatic vagal nerve
also innervates the duodenum (Berthoud, 2004). However, Allen and Bradford (2009) refuted
Langhans’s claims and provided strong evidence for the role of the liver in the control of feed
intake. Briefly, despite the failure of mercaptoacetate to stimulate feed intake in several recent
experiments in Langhan’s laboratory, mercaptoacetate may inhibit intake via a β-adrenergic

14

effect, potentially confounding the effect of mercaptoacetate in experiments that involve intake
restriction as a method to increase FA oxidation (Allen and Bradford, 2009).
Specific mechanisms for control of feed intake in early lactation
Dairy cows in early lactation are in a lipolytic state as a result of changes in nutrient
demands and hormonal profile due to parturition and lactogenesis as discussed above. As a
result, considerable effort is made to understand the best nutritional management strategies
available to maintain healthy productive dairy cows through the transition period. In order to
accomplish this, it is necessary to evaluate factors controlling feed intake in early lactation.
Allen et al. (2005) suggested that hepatic oxidation of fuels might interact with lipolytic state
resulting in reduced DMI during the immediate postpartum period.
Oba and Allen (2003a) conducted a dose response infusion study evaluating feeding
behavior and DMI in dairy cows in early and mid-lactation. Propionate linearly decreased DMI
and meal size as the percentage of propionate in the infusion increased for cows in early lactation
without affecting intermeal interval. In that experiment, propionate was more hypophagic for
cows in a lipolytic state, possibly due to stimulation of hepatic oxidation of acetyl CoA (Allen et
al., 2005). It is possible that stimulating lipolysis may result in reduced DMI. Bradford and
Allen (2007b) increased glucose demand using phlorizin, which was expected to delay hepatic
propionate oxidation within meals, which could increase DMI by increasing meal size.
However, the resulting decrease in plasma glucose concentration likely stimulated lipolysis
decreasing DMI for the phlorizin treated cows. If increased lipolysis is implicated in decreasing
DMI (Oba and Allen, 2003a; Bradford and Allen, 2007b), it follows that reducing lipolysis may
stimulate intake, particularly in early lactation cows. Consistent with this idea, thiazolidinedione
(TZD), a synthetic PPARγ ligand that can potentiate insulin action (Houseknecht et al., 2002),

15

has been injected into prepartum dairy cows and resulted in decreased plasma NEFA
concentration and increased postpartum DMI (Smith et al., 2009). These results suggest that
increased lipolysis can lead to reduced DMI postpartum, and limiting lipolysis can potentially
increase DMI during the critical postpartum period. Feeding dairy cows to reduce lipolysis
during the peripartum period might be beneficial for stimulating DMI during the critical
postpartum period, and several possible strategies will be discussed below.
High-energy prepartum diets
Feeding strategies for late dry period and into early lactation have been evaluated in
efforts to improve the metabolic status of cows during the peripartum period. Quite often, the
energy densities of the pre- and postpartum diets are increased while NDF is decreased and
effects on intake and metabolic parameters are evaluated, but the results of these strategies are
not consistent. Doepel et al. (2002) altered energy concentration in diets fed to dairy cows
during the late prepartum period by increasing the amount of barley and barley silage in the diet.
In that experiment, the high-energy diets decreased periparturient plasma NEFA concentration
and increased intake postpartum, but did not affect milk production. Cows fed a high energy
density diet through the late dry period had higher DMI compared to cows fed a low energy
density diet (Rabelo et al., 2003) and cows fed the high-energy diet had higher concentrations of
glucose and insulin and lower concentration of NEFA prepartum (Rabelo et al., 2005). The low
energy diet prepartum tended to increase plasma NEFA concentration postpartum, but
concentrations of other liver or plasma metabolites immediately following calving were not
affected (Rabelo et al., 2005). Dann et al. (1999) fed dry cracked or steam flaked corn during the
pre- and postpartum periods to increase carbohydrate availability, hypothesizing that steam
flaked corn would provide more glucose precursors to the cow to improve metabolic status and

16

lactation performance. Feeding steam flaked corn resulted in decreased plasma NEFA
concentration pre- and postpartum and increased milk production, but tended to decrease feed
intake postpartum. The goal of a high-energy prepartum diet is to create a more positive
metabolic profile including increased plasma insulin and decreased plasma NEFA
concentrations. If that is achieved, higher plasma insulin could limit lipolysis through
parturition, and potentially result in increased DMI postpartum. However, as described above,
the postpartum metabolic state is not consistently altered by a high-energy prepartum feeding
strategy. This may be due to the more dramatic reduction in intake in the last week prior to
calving for cows fed low NDF diets reported in a meta-analysis (Hayrili et al., 2002).
High-fill, low-energy prepartum diets
Other researchers have speculated that maintaining DMI through the late dry period may
have positive effects on metabolism and postpartum milk production. While Hayirli et al. (2002)
reported that cows fed high NDF diets had lower prepartum DMI, the magnitude of the reduction
in intake during the last week prior to parturition for cows fed higher NDF diets was not as
dramatic as for cows fed lower NDF (typically higher energy) diets. Several studies have
reported positive effects of limiting consumption of energy during the dry period with regard to
metabolism, DMI, and health postpartum (Grum et al., 1996; Dann et al., 2006; Douglas et al.,
2006; Janovick et al., 2011). For example, cows restricted to 80% of National Research Council
(NRC) requirements maintained feed intake compared with a significant reduction in feed intake
for cows fed ad libitum prepartum (Douglas et al., 2006). In this experiment, cows fed the
restricted diet had higher plasma NEFA concentration prepartum, but lower plasma NEFA and
BHBA concentrations postpartum and higher feed intake and milk production postpartum.
Consistent with these results, cows fed high-fill, controlled energy diets maintained prepartum

17

feed intake and had lower postpartum plasma NEFA and BHBA concentrations, which was
similar to cows in the same experiment fed a diet restricted to 80% of NRC requirements
(Janovick et al., 2011). Maintaining DMI prepartum may be beneficial for dairy cows as they
transition to lactating cow diets.
Postpartum feeding strategies
The experiments discussed above highlight the potential for feeding strategies during the
prepartum period to reduce lipolysis and stimulate DMI postpartum, which may limit the
incidence of postpartum metabolic disorders. Higher feed intake should result in higher plasma
insulin concentrations, limiting lipolysis. As discussed above, increased hepatic oxidation of FA
has been implicated in the control of feed intake (Scharrer and Langhans, 1986; Friedman et al.,
1986), therefore, by limiting the pool of FA available for oxidation, intake might be maintained
at a higher level.
During the postpartum period, feeding strategies to maintain rumen fill while providing
maximal glucose precursors for milk production should be implemented. As discussed above,
cows are in a lipolytic state postpartum and careful management of the feeding program through
the late dry period can affect the metabolic state of the cow postpartum. Milk yield and nutrient
demand for milk production increase dramatically at calving (Bauman and Currie, 1980) and
feeding strategies postpartum should be designed to meet these needs. Feeding highly
fermentable diets postpartum increases ruminal propionate production (Davis, 1967) and
propionate is the major gluconeogenic precursor (Reynolds et al., 2003); however, propionate
may interact with lipolysis to reduce feed intake of cows during the postpartum period.
Limiting the fermentability of the diet postpartum may be advantageous for several
reasons. First, lipolysis is high in early lactation and propionate may increase oxidation of acetyl

18

CoA in the TCA cycle as reviewed by Allen (2000). Inhibiting β-oxidation of FA in nonruminants has been shown to stimulate intake (Scharrer and Langhans, 1986; Friedman et al.,
1986), which suggests that feeding strategies to avoid stimulation of hepatic oxidation may
reduce hypophagia postpartum. Second, as reviewed by Shaver (1997), the risk of displaced
abomasum is highest for cows during the postpartum period and feeding strategies to maintain
rumen fill and limit acid production in the rumen are essential in maintaining cow health and
productivity postpartum. Providing postpartum diets with dry ground corn rather than high
moisture corn may allow higher dietary starch concentrations, increasing intake of glucose
precursors, which is likely beneficial for cow health and productivity as reviewed by Allen et al.
(2009). Dry ground corn is not extensively fermented in the rumen but is digestible in the small
intestine and provides glucose for milk production while not reducing intake. As intake and milk
production increase postpartum, cows become increasingly limited by fill and cows should be
fed a more fermentable diet to maximize milk production.
Conclusions
This review provides evidence that hepatic oxidation of fuels is involved in the control of
feed intake in the postpartum period. There is supporting data to suggest that there is an
interaction between propionate and the lipolytic state in control of feed intake in the postpartum
dairy cow. However, additional research is necessary to elucidate mechanisms involved during
the postpartum period. The overall hypothesis of this dissertation is that propionate controls
DMI by stimulating oxidation of acetyl CoA in the liver. To address this, we conducted three
experiments to investigate the hypophagic effect of propionate for cows in a lipolytic state and
the results are reported herein.

19

REFERENCES

20

REFERENCES

Adam, C. L., Z. A. Archer, P. A. Findlay, L. Thomas, and M. Marie. 2002. Hypothalamic gene
expression in sheep for cocaine- and amphetamine-regulated transcript, proopiomelanocortin, neuropeptide Y, Agouti-related peptide and leptin receptor and
responses to negative energy balance. Neuroendocrinology. 75:250-256.
Ahima, R. S., D. Prabakaran, C. Mantzoros, D. Qu, B. B. Lowell, E. Maratos-Flier, and J. S.
Flier. 1996. Role of leptin in the neuroendocrine response to fasting. Nature. 382:250-252.
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., and B. J. Bradford. 2009. Control of eating by hepatic oxidation of fatty acids. A
note of caution. Appetite. 53:272-273.
Allen, M.S., B. J. Bradford, and K. J. Harvatine. 2005. The cow as a model to study food intake
regulation. Annu. Rev. Nutr. 25:523-547.
Allen, M. S., B. J. Bradford, and M. Oba. 2009. BOARD-INVITED REVIEW: The hepatic
oxidation theory of the control of feed intake and its application to ruminants. J. Anim. Sci.
87:3317-3334.
Anil, M. H., and J. M. Forbes. 1980. Feeding sheep during intraportal infusions of short-chain
fatty acids and the effect of liver denervation. J. Physiol. 298:407-414.
Ballard, F. J. 1965. Glucose utilization in mammalian liver. Comp. Biochem. Physiol. 14:437443.
Bauman, D. E., and W. B. Currie. 1980. Partitioning of nutrients during pregnancy and lactation:
a review of mechanisms involving homeostasis and homeorhesis. J. Dairy Sci. 63:15141529.
Begum, N., H. M. Tepperman, and J. Tepperman. 1982. Effects od high carbohydrate and high
fat diets on rat adipose tissue pyruvate dehydrogenase responses to concanavalin A and
spermine. Endocrinology. 111:1491-1497.
Begum, N., H. M. Tepperman, and J. Tepperman. 1983. Effects of high fat and high
carbohydrate diets on liver pyruvate dehydrogenase and its activation by a chemical
mediator released from insulin-treated liver particulate fraction: Effect of neuraminidase
treatment on the chemical mediator activity. Endocrinology. 112:50-59.
Bell, A. W. 1995. Regulation of organic nutrient metabolism during transition from late
pregnancy to early lactation. J. Anim. Sci. 73:2804-2819.

21

Benson, J. A., C. K. Reynolds, D. J. Humphries, S. M. Rutter, and D. E. Beever. 2001. Effects of
abomasal infusion of long-chain fatty acids on intake, feeding behavior and milk
production in dairy cows. J. Dairy Sci. 84:1182-1191.
Benson, J. A., and C. K. Reynolds. 2001. Effects of abomasal infusion of long-chain fatty acids
on splanchnic metabolism of pancreatic and gut hormones in lactating dairy cows. J. Dairy
Sci. 84:1488-1500.
Benson, J. A., C. K. Reynolds, P. C. Aikman, B. Lupoli, and D. E. Beever. 2002. Effects of
abomasal vegetable oil infusion on splanchnic nutrient metabolism in lactating dairy cows.
J. Dairy Sci. 85:1804-1814.
Berthoud, H. R. 2004. Anatomy and function of sensory hepatic nerves. Anat. Rec. 280A:827835.
Block, S. S., W. R. Butler, R. A. Ehrhardt, A. W. Bell, M. E. Van Amburgh, and Y. R. Boisclair.
2001. Decreased concentration of plasma leptin in periparturient dairy cows is caused by
negative energy balance. J. Endocrinol. 171:339-348.
Bradford, B. J., and M. S. Allen. 2007a. Depression in feed intake by a highly fermentable diet is
related to plasma insulin concentration and insulin response to glucose infusion. J Dairy
Sci. 90:3838-3845.
Bradford, B. J., and M. S. Allen. 2007b. Phlorizin administration does not attenuate hypophagia
induced by intraruminal propionate infusion in lactating dairy cattle. J. Nutr. 137:326-330.
Bradford, B. J., and M. S. Allen. 2008. Negative energy balance increases periprandial ghrelin
and growth hormone concentrations in lactating dairy cows. Domest. ANim. Endocrinol.
34:196-203.
Bradford, B. J., K. J. Harvatine, and M. S. Allen. 2008. Dietary unsaturated fatty acids increase
plasma glucagon-like peptide-1 and cholecystokinin and may decrease premeal ghrelin in
lactating dairy cows. J. Dairy Sci. 91:1443-1450.
Carlson, D. B., N. B. Litherland, H. M. Dann, J. C. Woodworth, and J. K. Drackley. 2007.
Dietary L-carnitine alters periparturient nutrient metabolism and lactation in multiparous
cows. J. Dairy Sci. 90:3422-3441.
Caro, J. F., M. K. Sinhu, J. W. Kolaczynski, P. L. Zhang, and R. V. Considine. 1996. Leptin: the
tale of an obesity gene. Diabetes. 45:1455-1462.
Carvalho, E. R., N. S. Schmelz-Roberts, H. M. White, P. H. Doane, and S. S. Donkin. 2011.
Replacing corn with glycerol in diets for transition dairy cows. J. Dairy Sci. 94:908-916.

22

Choi, B. -R., and D. L. Palmquist. 1996. High fat diets increase plasma cholecystokinin and
pancreatic polypeptide, and decrease plasma insulin and feed intake in lactating cows. J.
Nutr. 126:2913-2919.
Choi, B. -R., D. L. Palmquist, and M. S. Allen. 1997. Sodium mercaptoacetate is not a useful
probe to study the role of fat in regulation of feed intake in dairy cattle. J. Nutr. 127:171176.
Choi, B. -R., D. L. Palmquist, and M. S. Allen. 2000. Cholecystokinin mediates depression of
feed intake in dairy cattle fed high fat diets. Domest. Anim. Endocrinol. 19:159-175.
Chung, Y. –H., D. E. Rico, C. M. Martinez, T. W. Cassidy, V. Noirot, A. Ames, and G. A.
Varga. 2007. Effects of feeding dry glycerin to early postpartum Holstein dairy cows on
lactational performance and metabolic profiles. J. Dairy Sci. 90:5682-5691.
Conn, A. R., D. I. Fell, and R. D. Steel. 1983. Characterization of alpha-keto acid transport
across the blood-brain barrier in rats. Am. J. Physiol. Endocrinol. Metab. 245:E253-E260.
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. 78:118-133.
Dann, H. M., G. A. Varga, and D. E. Putnam. 1999. Improving energy supply to late gestation
and early postpartum dairy cows. J. Dairy Sci. 82:1765-1778.
Dann, H. M., N. B. Litherland, J. P. Underwood, M. Bionaz, A. D’Angelo, J. W. McFadden, and
J. K. Drackley. 2006. Diets during far-off and close-up dry periods affect periparturient
metabolism in multiparous cows. J. Dairy Sci. 89:3563-3577.
Davis, C. L. 1967. Acetate production in the rumen of cows fed either control or low-fiber, highgrain diets. J. Dairy Sci. 50:1621–1625.
DeFrain, J. M., A. R. Hippen, K. F. Kalscheur, and P. W. Jardon. 2004. Feeding glycerol to
transition dairy cows: effects on blood metabolites and lactation performance. J. Dairy Sci.
87:4195-4206.
Della-Fera, M. A., and C. A. Baile. 1980. CCK-octapeptide injected in CSF decreases meal size
and daily foor intake in sheep. Peptides. 1:51-54.
Del Prete, E., T. A. Lutz, and E. Scharrer 2004. Inhibition of glucose oxidation by alpha-cyano4-hydroxycinnamic acid stimulates feeing in rats. Physiol. Behav. 80:489-498.
Doepel, L., H. Lapierre, and J. J. Kennelly. 2002. Peripartum performance and metabolism of
dairy cows in response to prepartum energy and protein intake. J. Dairy Sci. 85:2315-2334.

23

Dohme, F., A. Machmüller, F. Sutter, and M. Kreuzer. 2004. Digestive and metabolic utilization
of lauric, myristic, and stearic acid in cows, and associated effects on milk fat quality.
Arch. Anim. Nutr. 58:99-116.
Donkin, S. S., S. L. Koser, H. M. White, P. H. Doane, and M. J. Cecava. 2009. Feeding value of
glycerol as a replacement for corn grain in rations fed to lactating dairy cows. J. Dairy Sci.
92:5111-5119.
Douglas, G. N., T. R. Overton, H. G. Bateman II, H. M. Dann, and J. K. Drackley. 2006.
Prepartal plane of nutrition, regardless of dietary energy source, affects periparturient
metabolism and dry matter intake in Holstein cows. J. Dairy Sci. 89:2141-2157.
Drackley, J. K., T. H. Klusmeyer, A. M. Trusk, and J. H. Clark. 1992. Infusion of long-chain
fatty acids varying in saturation and chain length into the abomasum of lactating dairy
cows. J. Dairy Sci. 75:1517-1526.
Ehrhardt, R. A., R. M. Slepetis, J. Siegal-Willott, M. E. Van Amburg, A. W. Bell, and Y. R.
Boisclair. 2000. Development of a specific radioimmunoassay to measure physiological
changes of circulating leptin in dairy cattle and sheep. J. Endocrinol. 166:519-528.
Emery, R. S., J. S. Liesman, and T. H. Herdt. 1992. Metabolism of long chain fatty acids by
ruminant liver. J. Nutr. 122:832-837.
Friedman, M. I., M. G. Tordoff, and I. Ramirez. 1986. Integrated metabolic control of food
intake. Brain Res. Bull. 17:855-859.
Friedman, M. I., and M. G. Tordoff. 1986. Fatty acid oxidation and glucose utilization interact to
control food intake in rats. Am. J. Physiol. 51:R840-R845.
Goff, J.P., and R.L. Horst. 2001. Oral glycerol as an aid in the treatment of ketosis/fatty liver
complex. J. Dairy Sci. 84(Suppl. 1):153.(Abstr.).
Grum, D. E., J. K. Drackley, R. S. Younker, D. W. LaCount, and J. J. Veenhuizen. 1996.
Nutrition during the dry period and hepatic lipid metabolism of periparturient dairy cows. J.
Dairy Sci. 79:1850-1864.
Grummer, R. R. 1995. Impact of changes in organic nutrient metabolism on feeding the
transition dairy cow. J. Anim. Sci. 73:2820-2833.
Gulay, M. S., M. J. Hayen, M. Liboni, T. I. Belloso, C. J. Wilcox, and H. H. Head. 2004. Low
doses of bovine somatotropin during the transition period and early lactation improves milk
yield, efficiency of production, and other physiological responses of Holstein cows. J.
Dairy Sci. 87:948-960.
Harvatine, K. J., and M. S. Allen. 2006. Effects of Fatty acid supplements on feed intake, and
feeding and chewing behavior of lactating dairy cows. J. Dairy Sci. 89:1104-1112.

24

Hayashida, T., K. Murakami, K. Mogi, M. Nishihara, M. Nakazato, M. S. Mondal, Y. Horii, M.
Kojima, K. Kangawa, and N. Murakami. 2001. Ghrelin in domestic animals: distribution in
the stomach and its possible role. Domest. Anim. Endocrinol. 21:17-24.
Hayirli, A., R. R. Gummer, E. V. Nordheim, and P. M. Crump. 2002. Animal and dietary factors
affecting feed intake during the prefresh transition period in Holsteins. J. Dairy Sci.
85:3430-3443.
Hollmann, M., and D. K. Beede. 2012. Comparison of effects of dietary coconut oil and animal
fat blend on lactational performance of Holstein cows fed a high-starch diet. J. Dairy Sci.
95:1484-1499.
Houseknecht, K. L., B. M. Cole, and P. J. Steele. 2002. Peroxisome proliferator-activated
receptor gamma (PPARgamma) and its ligands: a review. Domest. Anim. Endocrinol. 22:123.
Hristov, A. N., C. Lee, T. Cassidy, M. Long, K. Heyler, B. Corl, and R. Forster. 2011. Effects of
lauric and myristic acids on ruminal fermentation, production, and milk fatty acid
composition in lactating dairy cows. J. Dairy Sci. 94:382-395.
Huntington, G. B. 1997. Starch utilization by ruminants: from basics to the bunk. J. Anim. Sci.
75:852-867.
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.
Janovick, N. A., Y. R. Boisclair, and J. K. Drackley. 2011. Prepartum dietary energy intake
affects metabolism and health during the periparturient period in primiparous and
multiparous Holstein cows. J. Dairy Sci. 94:1385-1400.
Kalra, S. P., M. G. Dube, A. Sahu, C. P. Phelps, and P. S. Kalra. 1991. Neuropeptide Y secretion
increases in the paraventricular nucleus in association with increased appetite for food.
Proc. Natl. Acad. Sci. U. S. A. 88:10931-10935.
Laeger, T., S. Görs, C. C. Metges, and B. Kuhla. 2012. Effect of feed restriction on metabolited
in cerebrospinal fluid and plasma of dairy cows. J. Dairy Sci. 95:1198-1208.
Langhans, W. 2008. Fatty acid oxidation in the energostatic control of eating – a new idea.
Appetite. 51:446-451.
Langhans, W., U. Damaske, and E. Scharrer. 1985. Different metabolites might reduce food
intake by the mitochondrial generation of reducing equivalents. Appetite. 6:143-152.
Langhans, W., G. Egli, and E. Scharrer. 1985. Regulation of food intake by hepatic oxidative
metabolism. Brain Res. Bull. 15:425-428.

25

Larsen, M., and N. B. Kristensen. 2009. Effect of abomasal glucose infusion on splanchnic and
whole body glucose metabolism in periparturient dairy cows. J. Dairy Sci. 92:1071-1083.
Leibowitz, S. F. 1991. Brain neuropeptide Y: an integrator of endocrine, metabolic and
behavioral processes. Brain Res. Bull. 27: 333– 337.
Leury, B. J., L. H. Baumgard, S. S. Block, S. Nthabisheng, R. A. Ehrhardt, R. P. Rhoads, D. E.
Bauman, A. W. Bell, and Y. R. Boisclair. 2003. Effect of insulin and growth hormone on
plasma leptin in periparturient dairy cows. Am. J. Physiol. Regul. Integr. Comp. Physiol.
285:R1107-15.
Leyton, J., P. J. Druru, and M. A. Crawford. 1987. Differential oxidation of saturated and
unsaturated fatty acids in vivo in the rat. Br. J. Nutr. 57:383-393.
Litherland, N. B., S. Thire, A. D. Beaulieu, C. K. Reynolds, J. A. Benson, and J. K. Drackley.
2005. Dry matter intake is decreased more by abomasal infusion of unsaturated free fatty
acids than by unsaturated triglycerides. J. Dairy Sci. 88:632-643.
Mayer, J. 1953. Glucostatic mechanism of regulation of food intake. N. Engl. J. Med. 249:13-16.
Millington, G. W. 2007. The role of proopiomelanocortin (POMC) neurones in feeding
behaviour. Nutr. Metab. 4:18.
Niijima, A. 1969. Afferent impulse discharges from glucoreceptors in the liver of the guinea pig.
Ann. N. Y. Acad. Sci. 157:690-700.
Niijima, A. 1981. Neurophysiological evidence for hepatic glucose-sensitive afferents.
Commentary on "The current status of hepatic theory of food intake control". Appetite.
2:151-152.
Oba, M., and M. S. Allen. 2000. Effects of brown midrib 3 mutation in corn cilage on
productivity of dairy cows fed two concentrations of dietary neutral detergent fiber: 1.
Feeding behavior and nutrient utilization. J. Dairy Sci. 88:1333-1341.
Oba, M., and M. S. Allen. 2003a. Dose-response effects of intraruminal infusion of propionate
on feeding behavior of lactating cows in early or midlactation. J. Dairy Sci. 86:2922-2931.
Oba, M., and M. S. Allen. 2003b. Extent of Hypophagia Caused by Propionate Infusion Is
Related to Plasma Glucose Concentration in Lactating Dairy Cows. J. Nutr. 133:11051112.
Oba, M., and M. S. Allen. 2003c. Effects of corn grain conservation method on feeding behavior
and productivity of lactating cows at two dietary starch concentrations. J. Dairy Sci.
86:174-183.

26

Rabelo, E., R. L. Rezende, S. J. Bertics, and R. R. Grummer. 2003. Effects of transition diets
varying in dietary energy density on lactation performance and ruminal parameters of dairy
cows. J. Dairy Sci. 86:916-925.
Rabelo, E., R. L. Rezende, S. J. Bertics, and R. R. Grummer. 2005. Effects of pre- and postfresh
transition diets varying in dietary energy density on metabolic status of periparturient dairy
cows. J. Dairy Sci. 88:4375-4383.
Relling, A. E., and C. K. Reynolds. 2007. Feeding rumen-inert fats differing in their degree of
saturation decreases intake and increases plasma concentrations of gut peptides in lactating
dairy cows. J. Dairy Sci. 90:1506-1515.
Relling, A. E., J. L. Pate, C. K. Reynolds, and S. C. Loerch. 2010. Effect of feed restriction and
supplemental dietary fat on gut peptide and hypothalamic neuropeptide messenger
ribonucleic acid concentrations in growing wethers. J. Anim. Sci. 88:737-748.
Relling, A. E., K. Lee, S. C. Loerch, C. K. Reynolds. 2012. Effects of glucose, propionate and
splanchnic hormones on neuropeptide mRNA concentrations in the ovine hypothalamus.
Journal of Animal Physiology and Animal Nutrition. 96:656-662.
Rémond, B., E. Souday, and J. P. Jouany. 1993. In vitro and in vivo fermentation of glycerol be
rumen microbes. Anim. Feed Sci. Technol. 41:121-132.
Reynolds, C. K. 1995. Quantitative aspects of liver metabolism in ruminants. Ruminant
Physiology: Digestion, Metabolism, Growth and Reproduction, W. v. Englehardt, S.
Leonhard-Marek, G. Breves, D. Geisecke, ed. Ferdinand Enke Verlag, Stuttgart,
Germany.p 351-372.
Reynolds, C. K., P. C. Aikman, B. Lupoli, D. J. Humphries, and D. E. Beever. 2003. Splanchnic
metabolism of dairy cows during the transition from late gestation through early lactation.
J. Dairy Sci. 86:1201-1217.
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.
Russek, M. 1963. Participation of hepatic glucoreceptors in the control of intake of food. Nature.
197:79-80.
Sahu, A., P. S. Kalra, and S. P Kalra. 1988. Food deprivation and ingestion induce reciprocal
changes in neuropeptide Y concentrations in the paraventricular nucleus. Peptides. 9:83-86.
Sartin, J. L., B. K. Whitlock, and J. A. Daniel. 2011. TRIENNIAL GROWTH SYMPOSIUM:
Neural regulation of feed intake: Modification by hormones, fasting, and disease. J. Anim.
Sci. 89:1991-2003.
Sato, H., T. Kurosawa, and S. Oikawa. 2002. Acetic acid and 3-hydroxybutyric acid levels in

27

cerebrospinal fluid of cattle. Anim. Sci. J. 73:137-141.
Scharrer, E., and W. Langhans. 1986. Control of food intake by fatty acid oxidation. J. Am.
Physiol. 250:R1003-R1006.
Seoane, J. R., and C. A. Baile. 1972. Effects of intraventricular (3 ventricle) injections of 2deoxy-D-glucose, glucose, and xylose on feeding behavior of sheep. Physiol. Behav.
9:423-428.
Shaver, R. D. 1997. Nutritional risk factors in the etiology of left displaced abomasum in dairy
cows: a review. J. Dairy Sci. 80:2449-2453.
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.
Smith, K. L., W. R. Butler, and T. R. Overton. 2009. Effects of 2,4-thiazolidinedione on
metabolism and performance in transition dairy cows. J. Dairy Sci. 92:3623-3622.
Stangassinger, M., and D. Giesecke. 1986. Splanchnic metabolism of glucose and related energy
substrated. Pages 347-366 in Control of Digestion and Metabolism in Ruminants. Proc. 6th
Int. Symp. Rumin. Physiol. L. P. Milligan, W. L. Grovum, and A. Dobson, ed. Prentice
Hall, Englewood Cliffs, NJ.
Stanley, B. G., S. E. Kyrkouli, S. Lampert, and S. F. Leibowitz. 1986. Neuropeptide Y
chronically injected into the hypothalamus: a powerful neurochemical inducer of
hyperphagia and obesity. Peptides. 7:1189-1192.
Sugino, T., Y. Hasegawa, Y. Kikkawa, J. Yamaura, M. Yamagishi, Y. Kurose, M. Kojima, K.
Kangawa, and Y. Terashima. 2002. A transient ghrelin surge occurs just before feeding in a
scheduled meal-fed sheep. Biochem. Biophys. Res. Commun. 295:255-250.
Voelker, J. A., G. M. Burato, and M. S. Allen. 2002. Effects of pretrial milk yield on responses
of feed intake, digestion, and production to dietary forage concentration. J. Dairy Sci.
85:2650-2661.
Wagner, C. G., C. D. McMahon, D. L. Marks, J. A. Daniel, B. Steele, J. L. Sartin. 2004. A role
for agouti-related protein in appetite regulation in a species with continuous nutrient
delivery. Neuroendocrinology. 80:210-218.
Wertz-Lutz, A. E., T. J. Knight, R. H. Pritchard, J. A. Daniel, J. A. Clapper, A. J. Smart, A.
Trenkle, and D. C. Beitz. 2006. Circulating ghrelin concentrations fluctuate relative to
nutritional status and influence feeding behavior in cattle. J. Anim. Sci. 84:3285-3300.

28

CHAPTER 2

HYPOPHAGIC EFFECTS OF PROPIONATE INCREASE WITH ELEVATED
HEPATIC ACETYL COA CONCENTRATION FOR COWS IN THE EARLY
POSTPARTUM PERIOD

ABSTRACT
Thirty multiparous lactating dairy cows were used in a randomized block design experiment to
evaluate factors related to the degree of hypophagia from intraruminal infusion of propionate.
Cows between 3 and 40 days postpartum at the start of the experiment were blocked by calving
date and randomly assigned to treatment. Treatments were 1.0 mol/L propionic acid or 1.0
mol/L acetic acid adjusted to pH 6 with sodium hydroxide and infused at 0.5 mol volatile fatty
acid/h from 6 h before feeding until 12 h after feeding. Propionate infusion decreased dry matter
intake 20.0% (P < 0.001), total metabolizable energy intake 22.5% (P < 0.001), and plasma βhydroxybutyrate concentration 45.7% (P < 0.001) compared with acetate. Effects of treatment
on DMI were related to concentration of acetyl CoA in the liver; hypophagic effects of
propionate compared with acetate increased as liver acetyl CoA concentration increased
(interaction P = 0.05). These results show that the hypophagic effects of propionate are greater
for cows with elevated concentration of acetyl CoA in the liver.

29

INTRODUCTION

Research with rodents suggests that meals can be terminated by a signal carried from the
liver to the brain via vagal afferents, which is affected by hepatic oxidation of fuels and
generation of ATP (Friedman, 1995; Langhans and Scharrer, 1992). Allen et al. (2005)
suggested that, of fuels metabolized by the ruminant liver, propionate is likely a primary satiety
signal because 1) its flux to the liver increases greatly during meals (Benson et al., 2002), 2)
hepatic extraction of propionate from the portal vein exceeds 70% (Reynolds et al., 2003), and 3)
hypophagic effects of propionate are eliminated by hepatic vagotomy (Anil and Forbes, 1980).
Propionate might stimulate satiety by its oxidation via conversion to acetyl CoA if its uptake by
the liver exceeds gluconeogenic flux, or by stimulation of oxidation by anapleurosis of an
existing pool of acetyl CoA derived from other fuels (Allen et al., 2009).
Increasing ruminal starch fermentation increases propionate flux to the liver by
increasing production of VFA as well as propionate as a fraction of total VFA (Davis, 1967).
Increased ruminal starch fermentation reduced feed intake of lactating cows in several
experiments reported in the literature as reviewed by Allen (2000). An experiment from our
laboratory demonstrated that a more rapidly fermented starch source nearly doubled the
fractional rate of starch digestion in the rumen and reduced feed intake 8% because of a 17%
reduction in meal size compared with a less fermentable starch source (Oba and Allen, 2003a).
The reduction in meal size might be because hepatic oxidation was stimulated by propionate
(Allen, 2000).
Hypophagic effects of propionate are greater for cows in the immediate postpartum
period compared with mid-lactation (Oba and Allen, 2003b). Beginning pre-partum and for

30

several weeks postpartum, cows are in a lipolytic state when energy requirements for milk
production increase at a greater rate than energy consumed. Hyperlipidemia in the periparturient
period is initially caused by a reduction in plasma insulin concentration combined with a
reduction in insulin sensitivity of adipose tissues (Bell, 1995). Uptake of NEFA by the liver
increases greatly (Reynolds et al., 2003), resulting in increased FA oxidation, buildup of acetyl
CoA, and hepatic export of ketones. Hepatic oxidation of FA and generation of ATP might
suppress feed intake in the periparturient period, and hypophagic effects of propionate may be
enhanced because propionate uptake by the liver stimulates oxidation of the existing pool of
acetyl CoA (Allen et al., 2009).
The objective of this experiment was to evaluate the relationship between hypophagia
from intraruminal infusion of propionate relative to acetate and characteristics of cows in the
postpartum period that are in a lipolytic state. We hypothesized that hypophagic effects of
propionate increase when hepatic acetyl CoA concentration is elevated. Propionate is expected to
stimulate oxidation of acetyl CoA in the TCA cycle, causing satiety sooner and decreasing meal
size for cows with elevated acetyl CoA concentration in the liver. Understanding mechanisms
controlling energy intake during the peripartum period cows is vital to maintaining healthy,
productive cows.

31

MATERIALS AND METHODS

Animals, housing, and diets The Institutional Animal Care and Use Committee at
Michigan State University approved all experimental procedures for this experiment. Thirty
lactating Holstein cows were ruminally cannulated at least 45 d before calving. Cows were
housed in individual tie stalls for the duration of the experiment. Cows were fed once daily
(1200 h) at 115% of expected intake and received a common experimental diet from parturition
through the end of the experiment. The experimental diet (Table 2.1) was composed of corn
silage, alfalfa silage, alfalfa hay, ground corn, soybean meal, soy hulls, and a vitamin and
mineral mix and formulated to meet requirements for absorbed protein, minerals and vitamins
(NRC, 2001).

32

Table 2.1. Ingredients and nutrient composition of experimental diet (% of dietary DM except
for DM).
Diet ingredient
Corn silage
Alfalfa silage
Alfalfa hay
Ground corn
Soybean meal
Soy hulls
1
Vitamin and mineral mix
Nutrient composition
DM
OM
Starch
NDF
ADF
CP
Ether extract

38.7
30.2
6.0
10.9
6.9
4.2
4.0
51.0
92.2
19.2
37.1
27.5
15.9
3.7

1

Vitamin and mineral mix contained 24.8% ground corn grain, 21.5% dehydrated cane molasses,
11.2% limestone, 9.6% blood meal, 9.0% sodium bicarbonate, 6.6% dicalcium phosphate, 4.2%
Reashure choline, 3.1% magnesium sulfate, 2.8% salt, 2.0% animal fat, 1.5% niacin, 1.3% trace
mineral mix, 0.95% biotin, 0.70% Yeast Plus, 0.54% vitamin ADE premix, 0.32% selenium
yeast, and 0.09% Rumensin 90.

Experimental design and treatments The experiment was a randomized block design
with a covariate period. Cows were between 3 and 40 days postpartum at the start of the
experiment and were assigned to block by calving date, then randomly assigned to treatment
within a block. The experiment was conducted with 5 blocks of cows containing from 4 to 8
cows each within the same calendar year. The length of the experiment was 3 d for each block
of cows including a covariate day to establish baseline values for all measurements on d 1. Day
2 of the experiment was a rest day with no treatment or sampling. Treatments were propionic or
acetic acids (1 mol/L, adjusted to pH 6.0±0.1 with sodium hydroxide) continuously infused into
the rumen at 0.5 mol VFA/h from 0600 h to 2400 h (9 moles/18 h infusion) on d 3 of the
experiment. Solutions were infused at 500 mL/h using peristaltic pumps (#78016-30, Cole33

Parmer Instrument, IL) with Tygon tubing (1.6 mm i.d.) from individual containers that were
manually refilled with 500 mL of treatments hourly to insure accurate infusion rates per hour.
Infusions began 6 h prior to feeding to reach a steady state VFA concentration in the rumen
before starting feeding behavior monitoring.
Data and sample collection Cows were blocked from feed from 1000 h to 1200 h daily to
allow for weighing of orts, collection of orts samples, and offering feed. Samples of all diet
ingredients (0.5 kg), the TMR (0.5 kg), and orts (12.5% of the remaining feed) were collected
daily and composited into one sample per cow per block for analysis. Body weight and body
condition score were recorded on d 1 of the experiment. Body condition was scored by 3 trained
investigators on a 5-point scale, where 1 = thin and 5 = fat, as described by Wildman et al.
(1982). Cows were milked twice daily at 0400 and 1700 h in the milking parlor with the
exception of the covariate day and infusion day when cows were milked in their stalls. Milk
samples were collected from each milking during the covariate day and analyzed for fat, true
protein, lactose, and solids non-fat by Michigan Dairy Herd Information Association (AOAC,
1997).
Blood, rumen, and fecal samples were collected every 6 h for 24 h (n = 4) during the
covariate day. Rumen fluid samples were collected from 5 different sites in the rumen, squeezed
through a nylon screen, and pH determined immediately. Samples were then frozen at -20°C for
later analysis of VFA and ammonia-N concentrations. Fecal samples were collected and frozen
at -20°C for later analysis to determine diet digestibility for use in calculating metabolizable
energy intake. Blood samples were collected via coccygeal venipuncture into two Vacutainer®
tubes: one with potassium EDTA and one with potassium oxalate and sodium fluoride (as a
glycolytic inhibitor). Blood samples were cooled on ice until centrifuged at 2000 × g for 20 min

34

(within 12 minutes of sample collection); plasma was then harvested and stored at -20°C for later
analysis of metabolites and hormones. A 1 ml aliquot of plasma from each potassium EDTA
tube was stored with 0.05 M benzamidine (final concentration) to prevent enzymatic degradation
of glucagon. Liver tissue was collected by needle biopsy (Bradford and Allen, 2005) once prior
to feeding at the end of the covariate day and stored at -80°C until analysis for acetyl CoA
concentration.
Feeding behavior was monitored for 12 h (1200 h to 2400 h) on the covariate and the
infusion day (d 3). Feeding behavior data (chewing, feed disappearance, and water intake) were
recorded via computer every 5 s, which allowed for calculation of meal size, intermeal interval,
water intake, and eating, ruminating, and total chewing time (Dado and Allen, 1993). Blood and
rumen samples were collected immediately prior to the start of each infusion period (0530 h) and
immediately after the infusion period ended (0015 h). The samples were processed and stored as
previously described.
Analysis of samples Feed, 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 (1-mm
screen; Arthur H Thomas, Philadelphia, PA) and analyzed for ash, NDF, indigestible NDF, CP,
starch, and ether extract. Ash concentration was determined after 5 h of oxidation at 500°C.
Concentration of NDF was analyzed according to Van Soest et al. (1991; method A for NDF)
with the inclusion of amylase and Na sulfite. Indigestible NDF was estimated as NDF residue
after 240 h in vitro fermentation (Goering and Van Soest, 1970); flasks were re-inoculated at 120
h to ensure a viable microbial population. Ruminal fluid for the in vitro incubations was
collected from a non-pregnant dry cow fed dry hay only. Indigestible NDF was used as an
internal marker to estimate total tract nutrient digestibility (Cochran et al., 1986). Crude protein

35

was determined according to Hach et al. (1987). Starch was analyzed using an enzymatic
method (Karkalas, 1985) after samples were gelatinized with sodium hydroxide. Glucose was
measured with a glucose oxidase method (Sigma Chemical Co., St. Louis, MO). Ether extract
was determined using a modified Soxhlet apparatus (AOAC, 1990). All nutrients are expressed
as percentages of DM determined by drying at 105°C in a forced air oven for more than 8 h.
Plasma samples were analyzed using commercial kits for concentration of NEFA (NEFA
HR kit, Wako Chemicals USA, Richmond, VA), BHBA (kit #2240, Stanbio Laboratory, Boerne,
TX), insulin (Coat-A-Count, Siemens Healthcare Diagnostics, Deerfield, IL), and glucagon (kit
#GL-32K, Millipore, Billerica, MA). Plasma glucose concentration was analyzed using a
glucose oxidase method (Sigma Chemical Co., St. Louis, MO). Plasma and rumen fluid VFA
concentrations were determined by HPLC according to Oba and Allen (2003b). Rumen
ammonia-N concentration was determined by colorimetric assay using the method of Broderick
and Kang (1980). Liver acetyl CoA was analyzed using the method of King and Reiss (1985)
with modifications. Sixty to 80 mg wet tissue weight of liver was homogenized in 6 µl of 100
mM sodium phosphate buffer (pH = 3.0) and 240 µl of 3.6% perchloric acid solution. The
homogenized solution was centrifuged at 10,000 x g and the supernatant transferred into an
HPLC vial for analysis. Aliquots (50 µl) were sampled with a Waters 717 plus Autosampler
(Milford, MA), injected onto a Phenomenex SYNERGI 4 micron Hydro RP80A (150 x 4.6 mm;
4 micron; Torrance, CA) column maintained at 30° C, and absorbance was measured at 254 nm
by a Waters 486 Tunable Absorbance detector.
Statistical analysis Data from the covariate period was summarized using the
Distributions procedure of JMP (Version 8.0.2, 2009, SAS Institute, Cary, NC). Linear and
quadratic relationships between DMI and covariates of interest were evaluated by regression

36

with JMP. Feeding behavior, intake, and rumen fermentation characteristics data were analyzed
using the Fit Model procedure of JMP with the following model:
Yij = µ + Ti + CovDMI + CovAcCoA + TiCovAcCoA + eij
Where µ = overall mean, Ti = fixed effect of treatment (i = 1 to 2), CovDMI = effect of covariate
DMI, CovACoA = effect of covariate acetyl CoA concentration in the liver, TiCov = interaction
between treatment and the covariate acetyl CoA concentration in the liver, eij = residual
(normally distributed). Plasma metabolite and hormone data were analyzed using the same
model with CovDMI removed because plasma responses were evaluated as change over the 18 h
infusion rather than during the 12 h period for intake and feeding behavior. Treatment effects
were declared significant at P < 0.05 and tendencies for treatment effects at P < 0.10.
Interactions were declared significant at P < 0.10. Covariate interactions were removed from the
model if P > 0.20.

37

RESULTS

Variation among animals during the covariate period Results of data collected during
the covariate day to provide baseline values for variables of interest are summarized in Table 2.2.
As expected, feed intake and plasma metabolite concentrations were highly variable among cows
because of the range in milk yield and day PP. Dry matter intake ranged from 8.0 to 19.5 kg/12
h and ME intake ranged from 50.1 to 143.5 MJ/12 h, while plasma NEFA concentration ranged
from 103 to 1330 µEq/L and plasma BHBA ranged from 5.52 to 19.20 mg/dl. Liver acetyl CoA
concentration was highly variable, ranging from 5.1 to 106.5 nmol/g wet tissue weight and was
positively correlated with pre-infusion plasma BHBA concentration (r = 0.37, quadratic, P<
0.01).

Table 2.2. Characteristics of cows during the covariate period.
Mean
Feed intake
DMI (kg/d)
DMI (kg/12 h)
1
ME intake (MJ/12 h)
Feeding behavior
Meal bouts (/12 h)
Meal length (min/meal)
Intermeal interval (min)
Meal size (kg DM)
Chewing time
Eating (min/12 h)
Ruminating (min/12 h)
Total (min/12 h)
Water intake (L/12 h)

Minimum

Maximum

SE

19.0
13.2
100.3

14.4
8.0
50.1

24.6
19.5
143.5

0.3
0.4
2.8

6.2
40.2
83
2.3

4
19.6
38
0.9

12
73.7
142
4.2

0.2
1.7
3.0
0.1

186
206
393
57.0

121
48
238
26.6

258
299
486
79.1

4.1
8.0
9.1
1.6

38

Table 2.2 (cont’d)
Mean
Minimum Maximum
SE
Ruminal fermentation
pH
6.37
5.97
7.23
0.03
Total VFA (mM)
118.1
81.1
157.5
2.12
VFA composition (% of total)
Acetate
63.1
54.1
68.9
0.46
Propionate
21.4
18.0
29.4
0.30
iso-Butyrate
0.8
0.4
1.2
0.03
Butyrate
11.6
8.1
15.0
0.24
iso-Valerate
1.2
0.6
1.7
0.03
Valerate
1.5
1.1
2.1
0.03
Ammonia-N (mM)
6.17
1.3
17.1
0.45
Liver acetyl CoA (nmol/g wet tissue)
29.9
5.10
106.5
2.76
Plasma metabolites and hormones
Acetate (mM)
1.14
0.48
2.42
0.05
Propionate (mM)
0.32
0.16
0.47
0.09
Glucose (mg/dL)
52.4
38.4
64.1
0.90
5.82
1.34
17.85
0.475
Insulin (µIU/mL)
580
103
1330
31
NEFA (µEq/L)
BHBA (mg/dL)
10.57
5.52
19.20
0.437
Glucagon (pg/mL)
137
101
183
2
Apparent total tract digestibility (%)
DM
46.7
23.1
59.2
1.1
OM
54.5
39.9
61.1
0.5
NDF
34.7
13.3
41.8
0.7
Starch
91.8
87.2
97.0
0.4
CP
48.7
19.7
63.9
1.2
Ether extract
76.7
39.0
91.5
1.5
Milk production
Milk yield (kg/d)
38.1
23.4
49.5
0.74
Milk fat (%)
5.29
3.13
8.06
0.18
Milk protein (%)
3.17
2.47
4.26
0.05
Milk lactose (%)
4.66
4.18
5.13
0.03
Milk SNF (%)
8.55
7.57
9.65
0.06
2
Milk energy (MJ/d)
136.3
92.6
192.2
25.8
Body weight (kg)
668
513
760
7.52
Body condition score
2.6
1.5
4.1
0.07
Days postpartum d 1
18.6
3
40
1.40
1
Metabolizable energy intake from diet was calculated according to Oba and Allen (2003b).
2

Milk energy was calculated according to NRC (2001).

39

Feed intake, total ME intake, and feeding behavior Propionate infusion decreased DMI
20% compared with acetate (12.8 vs. 16.0 kg/12 h; P < 0.001; Table 2.3) because of a nonsignificant decrease in meal size and meal frequency. Although propionate has greater ME
concentration than acetate such that ME of propionate infused was greater than acetate infused
(9.22 vs. 5.26 MJ/12 h; Table 2.3), propionate decreased total ME intake (infusion + diet) 22.5%
compared with acetate (120 vs. 93 MJ/12 h; P < 0.001; Table 3). Treatment did not affect total
daily eating time (P = 0.23; Table 2.3), however, propionate tended to decrease total daily
ruminating time (170 vs. 204 min/d; P = 0.06; Table 2.3), and decreased total daily chewing time
(350 vs. 399 min/d; P = 0.03; Table 2.3). Propionate infusion decreased water intake 15.3%
compared with acetate infusion (61.8 vs. 73.0 l/12 h; P = 0.003; Table 2.3).

40

Table 2.3. Effects of intraruminal infusion of sodium propionate relative to sodium acetate on
feeding behavior and energy intake for cows in early lactation.

Acetate
Feeding behavior
DMI (kg/12 h)
Meal size (kg)
Meal length (min/meal)
Meal frequency (#/12 h)
Intermeal interval (min)
2
ME intake (MJ/12 h)
Diet
3

Infusion
Total
Chewing activity (min/12 h)
Eating
Ruminating
Total
Water Intake (L/12 h)
Milk (kg/d)

Propionate

SE

Trt

P
Cov
CoA

16.0
2.47
39.0
6.59
77.4

12.8
2.09
35.8
6.35
83.8

0.47
0.170
2.34
0.435
6.18

<0.001
0.12
0.34
0.70
0.47

0.53
0.33
0.49
0.69
0.33

0.04
0.82
0.43
0.59
0.32

115

84

5.1

<0.001

0.17

0.32

5.26
120

9.22
93

NA
5.1

NA
<0.001

NA
0.17

NA
0.32

Trt ×
1
CovCoA

195
180
8.4
0.23
0.82
0.19
204
170
12.6
0.06
0.54
0.97
399
350
15.6
0.03
0.71
0.49
73.0
61.8
2.38
0.003
0.40
0.06
40.4
39.7
1.68
0.75
0.11
0.24
1
P-values for treatment (Trt), covariate acetyl CoA (Cov CoA), and treatment by covariate
acetyl CoA interaction (Trt Cov × CoA)
2
Metabolizable energy intake from the diet was calculated according to Oba and Allen
(2003b).
3
Metabolizable energy intake from the infusion was based on energy density of 0.876 and
1.536 MJ/mol for acetate and propionate (Oba and Allen, 2003b), respectively, and was the
same for all cows.

Ruminal pH, concentration and profile of VFA, and ammonia-N concentration
Propionate infusion, compared with acetate infusion, did not affect ruminal pH (P = 0.80, Table
2.4), total VFA concentration (P = 0.50, Table 2.4), or ammonia-N concentration (P = 0.57,
Table 2.4). Propionate infusion increased ruminal propionate as a percent of total VFA (28.4%
vs. 21.2%, P < 0.001, Table 2.4), while acetate infusion increased ruminal acetate as a percent of
total VFA (64.6% vs. 57.1%, P < 0.001, Table 2.4).

41

Table 2.4. Effect of intraruminal infusion of sodium propionate relative to sodium acetate on
rumen VFA and ammonia-N concentration changes throughout the 18 h infusion period
(reported as post-infusion – pre-infusion).
1

Infusion
Item

Acetate

Propionate

SE

Trt

P
Cov

Trt ×
Cov

Post infusion
0.04
Rumen pH
6.25
6.28
0.072
0.80
0.19
0.03
Total VFA (mM)
130
126
3.4
0.50
0.49
VFA composition (%)
Acetate
64.6
57.1
1.14
<0.001
0.02
0.97
0.02
Propionate
21.2
28.4
0.97
<0.001
0.03
0.02
iso-Butyrate
0.70
0.71
0.032
0.84
0.73
11.2
10.7
0.31
0.26
<0.001
0.23
Butyrate
1.14
1.21
0.065
0.46
<0.001
0.56
iso-Valerate
0.01
Valerate
1.45
1.52
0.052
0.33
0.18
<0.01
Ammonia-N (mM)
6.89
5.98
1.120
0.57
0.32
Post - pre infusion
0.02
Rumen pH
-0.043
-0.017
0.071
0.80
0.19
<0.01
Total VFA (mM)
4.86
1.60
3.377
0.50
0.49
VFA composition (%)
0.96
Acetate
1.13
-6.32
1.14
<0.001
0.97
0.29
7.47
0.97
<0.001
0.76
0.03
Propionate
-0.06
-0.05
0.032
0.84
<0.001
0.73
iso-Butyrate
-0.63
-1.13
0.31
0.26
0.22
0.23
Butyrate
-0.06
0.009
0.065
0.46
0.99
0.56
iso-Valerate
-0.02
0.06
0.052
0.34
<0.01
0.18
Valerate
0.24
-0.67
1.120
0.57
0.11
0.32
Ammonia-N (mM)
1
P-values for treatment (Trt), covariate (Cov), and treatment by covariate interaction (Trt ×
Cov)

42

Plasma metabolites and hormones before and after infusion Propionate infusion tended
to increase plasma propionate concentration compared with acetate infusion (P = 0.09; Table
2.5) and acetate infusion increased plasma acetate concentration compared with propionate
infusion (P < 0.001; Table 2.5). Propionate infusion increased plasma glucose concentration
114.8% (61.1 vs. 53.2 mg/dl; P < 0.001; Table 2.5) and reduced plasma BHBA concentration
45.7% compared with acetate infusion (6.09 vs. 13.32 mg/dl; P < 0.01; Table 2.5). Covariate
plasma BHBA concentration was positively correlated to covariate hepatic concentration of
acetyl CoA (R2 = 0.24, P < 0.01 quadratic; Figure 2.1). There were no differences in postinfusion plasma insulin (P = 0.54), glucagon (P = 0.30), and NEFA (P = 0.45) concentrations
(Table 2.5).

43

Table 2.5. Effect of intraruminal infusion of sodium propionate relative to sodium acetate on
plasma metabolite and hormone concentration changes throughout the 18 h infusion period
(reported as post-infusion – pre-infusion).
1

Acetate

Propionate

SE

Trt

P
Cov

Trt ×
Cov

Post infusion
<0.001
Glucose (mg/dL)
53.2
61.1
1.44
<0.001
0.40
<0.01
Insulin (µIU/mL)
6.85
7.78
1.060
0.54
0.06
<0.001
Glucagon (pg/mL)
147
141
4.1
0.30
0.60
<0.001
365
320
41.5
0.45
0.16
NEFA (µEq/L)
<0.001
BHBA (mg/dL)
13.32
6.09
0.788 <0.001
<0.01
<0.001 <0.001
Acetate (mM)
1.96
0.94
0.098 <0.001
0.005
Propionate (mM)
0.24
0.28
0.018
0.09
0.80
Post - pre infusion
0.98
Glucose (mg/dL)
-2.52
5.23
1.423 <0.001
0.40
0.60
Insulin (µIU/mL)
1.84
2.76
1.060
0.54
0.06
0.98
Glucagon (pg/mL)
10.2
4.0
4.14
0.30
0.60
<0.001
-209
-254
41.5
0.45
0.16
NEFA (µEq/L)
0.11
BHBA (mg/dL)
2.73
-4.49
0.788 <0.001
<0.01
0.87
Acetate (mM)
0.792
-0.222
0.098 <0.001
<0.001
<0.001
Propionate (mM)
-0.032
0.013
0.018
0.09
0.80
1
P-values for treatment (Trt), covariate (Cov), and treatment by covariate interaction (Trt ×
Cov)

44

Figure 2.1. Relationship between liver acetyl CoA concentration during the covariate day and
plasma BHBA concentration during the covariate day (r2 = 0.24, P = 0.006 quadratic).

Interactions between treatments and indicators of metabolic status Interactions
between treatment and indicators of metabolic status measured during the covariate day
(e.g. plasma BHBA and NEFA concentrations, liver acetyl CoA concentration) were
evaluated for DMI response. There was an interaction (P = 0.05) between treatment and
liver acetyl CoA on DMI (Figure 2.2); hypophagia from propionate infusion compared
with acetate infusion increased as hepatic acetyl CoA concentration increased. No
interactions were detected between treatment and any other metabolite or hormone
measured during the covariate day (Table 2.2).

45

Figure 2.2. Interaction (P= 0.05) between treatment and liver acetyl CoA concentration for DMI
during the 12 h infusion period in period 1 for cows infused with acetic acid (dotted line) or
propionic acid (solid line) intraruminally. Equation: DMI = 10.29 + 1.43*A – 0.0026*ACoA
+0.033*(ACoA – 29.43)*A where A = acetic acid treatment and ACoA = concentration of liver
acetyl CoA in the covariate period.

46

DISCUSSION

As expected, ruminal propionate and acetate infusion increased concentrations of
propionate and acetate in rumen fluid and plasma, respectively, consistent with a previous
experiment from our laboratory at the same infusion rate (Oba and Allen, 2003b). The reduction
in DMI by propionate compared with acetate was expected based upon previous experiments
(Allen et al., 2005; Allen, 2000). We previously proposed that the hypophagic effects of
propionate compared with acetate are related to their effects on hepatic oxidation (Allen, 2000).
Propionate can be oxidized by conversion to acetyl CoA (Knapp et al., 1992) and stimulate
hepatic oxidation by anapleurosis, whereas hepatic uptake of acetate from the blood is negligible
(Reynolds, 1995) because activity of acetyl CoA synthetase is low in ruminant liver (Ricks and
Cook, 1981). Acetate was used as the control in this experiment to account for mechanisms
affecting feed intake that relate to osmotic effects of intraruminal infusions rather than those
related to the specific effects of fuels per se (Choi and Allen, 1999). Total ME intake from the
diet and the infusion was calculated to account for the higher ME of propionate relative to
acetate. The 22.5% reduction in total ME intake in the present study was similar to previous
experiments at a similar infusion rate of propionate (Oba an Allen, 2003bc). Previous work in
our laboratory showed that propionate linearly decreased ME intake in early lactation dairy cows
in a dose-dependent manner as the fraction of propionate in the infusate compared with acetate
increased (Oba and Allen, 2003b). The slight reduction in water intake was likely a result of the
effect of treatment on DMI because DMI and water intake are positively correlated (Woodford et
al., 1984) and DMI accounts for the majority of the variation in water intake (Murphy et al.,
1983).

47

Effects of treatment on plasma metabolite and hormone concentrations were as expected
and in agreement with previous experiments (Oba and Allen, 2003bc). Propionate infusion
increased plasma glucose concentration compared with acetate infusion, consistent with previous
results (Oba and Allen, 2003bc), but did not affect plasma insulin concentration. Effects of
propionate infusions on plasma insulin concentration have been variable with no effect of
treatment observed in some experiments (Allen et al, 2009).
Propionate infusion decreased plasma BHBA concentration, but did not affect plasma
NEFA concentration, consistent with our previous results at the same infusion rate in dairy cows
in the early postpartum period (Oba and Allen, 2003b). The reduction in plasma BHBA
concentration without a reduction in plasma NEFA concentration might have been from 1)
decreasing β-oxidation by reducing the transport of long-chain FA (LCFA) into the mitochondria
by propionate either by increasing esterification of LCFA in the cytosol (Jesse et al., 1986) or
inhibiting CPT1 through conversion of propionate to methylmalonyl CoA (Brindle et al., 1985),
2) reducing ketogenesis from propionate in ruminant liver (Faulkner and Pollock, 1986) by
decreasing the activity of HMG-CoA synthase (Lowe and Tubbs, 1985), or 3) increasing
oxidation of acetyl CoA in the TCA cycle. While all of the above are possible explanations for
the effect of propionate on plasma BHBA concentration, increased oxidation of acetyl CoA is
consistent with the hypophagic effect of propionate by stimulation of hepatic oxidation (Allen et
al., 2009). Hepatic oxidation of acetyl CoA would increase as propionate supply to the liver
stimulates oxidation of acetyl CoA in the TCA cycle and decreases its export from the liver by
inhibiting ketogenesis (Allen et al., 2009). This is expected to increase the rate at which energy
charge ([ATP]+½ [ADP])/([ATP]+[ADP]+[AMP]) increases within meals, decreasing meal size,
consistent with our results. While propionate might also decrease transport of LCFA into the

48

mitochondria inhibiting β-oxidation, this is inconsistent with effects of inhibiting or enhancing βoxidation on feed intake as reviewed by Allen et al. (2009). Inhibiting transport of LCFA into
the mitochondria by blocking CPT1 with methyl palmoxirate (Friedman et al., 1986) stimulated
eating in rats, while stimulating transport of LCFA into the mitochondria by supplementation of
carnitine to transition dairy cows decreased feed intake during the first 2 wk of lactation (Carlson
et al., 2007).
The hypophagic effects of propionate, but not acetate, were increasingly pronounced at
elevated hepatic acetyl CoA concentrations (Figure 2.2). Propionate likely stimulated oxidation
of hepatic acetyl CoA, reducing the pool of acetyl CoA available for export as ketones,
depressing DM and ME intake. There was a positive relationship between liver acetyl CoA and
DMI for acetate infusion, possibly because acetyl CoA oxidation in the TCA cycle was
increasingly limited for cows with greater intake, while propionate infusion may have enhanced
oxidation of acetyl CoA in the TCA cycle as discussed above.
It is unlikely that the effect of propionate infusion on DMI was a result of the observed
increase in plasma glucose because glucose infusions generally do not decrease energy intake in
ruminants (Allen et al., 2009; Al-Trad et al., 2009). Larsen and Kristensen (2009) reported that
abomasal glucose infusion in postpartum dairy cows (0.35 mol/h from 1-29 DIM) prevented the
increase in DMI and milk yield observed with control cows receiving no infusion. However, in
that experiment, there was no effect of glucose infusion on plasma glucose concentration and the
glucose infusion tended to increase plasma insulin concentration approximately three-fold (P =
0.07). In another experiment, Leury et al. (2003) reported that DMI was reduced for early
lactation cows that were subjected to a hyperinsulemic euglycemic clamp. The consistently high
plasma insulin concentration in both of these experiments likely decreased the rate of

49

gluconeogenesis, resulting in faster oxidation of fuels in the liver within meals, limiting meal
size and DMI. Consistent with this, we reported a negative relationship between DMI of midlactation cows on a highly fermentable diet and mean daily plasma insulin concentration
measured prior to treatment (r2 = 0.28, P < 0.01, n = 29; Bradford and Allen, 2007). In the
present experiment, propionate treatment increased plasma glucose concentration but did not
affect plasma insulin concentration. Therefore, it is more likely that the effect of propionate on
DMI was through stimulation of oxidation of acetyl CoA in the liver.
Langhans et al. (2011) contended that enterocytes could act as a fuel gauge to signal
satiety and that rat enterocytes are sufficiently innervated to send signals about energy status to
the brain. While propionate is oxidized by enterocytes, acetate is as well (Oba et al., 2004). In
addition, flow of acetate from the rumen is greater than propionate because production of acetate
in the rumen is greater (Sutton et al., 2003) and it is absorbed more slowly than propionate
(Djikstra et al., 1993). Because of this, and because of the interaction observed between
treatment and acetyl CoA concentration reported here, it is more likely that the hypophagic
effects of propionate are from hepatic oxidation.

50

CONCLUSIONS

Feed intake of cows in the postpartum period is likely suppressed by hepatic oxidation of
NEFA mobilized from body reserves by homeorhetic mechanisms related to reduced insulin
concentration and sensitivity of adipose tissues prior to parturition. Beta-oxidation of NEFA in
the liver increases the pool of acetyl CoA and propionate uptake by the liver during meals might
stimulate its oxidation, causing satiety. Diets containing highly fermentable starch might depress
energy intake of cows in the postpartum period over the short-term because of rapid propionate
production and absorption during meals. Feeding studies evaluating the effect of starch
fermentability on energy intake during the postpartum period and adaptive responses to longerterm infusions of propionate are needed to better understand how to increase energy intake in the
postpartum period.

51

REFERENCES

52

REFERENCES

Allen, M. S. 2000. Effects of diet on short-term regulation of feed intake by lactating dairy cattle.
J. Dairy Sci. 83:598-624.
Allen, M.S., B. J. Bradford, and K. J. Harvatine. 2005. The cow as a model to study food intake
regulation. Annu. Rev. Nutr. 25:523-547.
Allen, M. S., B. J. Bradford, and M. Oba. 2009. BOARD-INVITED REVIEW: The hepatic
oxidation theory of the control of feed intake and its application to ruminants. J. Anim. Sci.
87:3317-3334.
Al-Trad, B., K. Reisberg, T. Wittek, G. B. Penner, A. Alkaassem, G. Gäbel, M. Fürll,
J.R. Aschenbach. 2009. Increasing intravenous infusions of glucose improve body
condition but not lactation performance in midlactation dairy cows. J. Dairy Sci. 92:56455658.
Anil, M. H., and J. M. Forbes. 1980. Feeding sheep during intraportal infusions of short-chain
fatty acids and the effect of liver denervation. J. Physiol. 298:407-414.
th

AOAC. 1990. Official Methods of Analysis. 15 ed. Association of Analytical Chemists.
Gaithersburg, MD.
th

AOAC. 1997. Official Methods of Analysis. 16 ed. Association of Analytical Chemists.
Gaithersburg, MD.
Bell, A. W. 1995. Regulation of organic nutrient metabolism during transition from late
pregnancy to early lactation. J. Anim. Sci. 73:2804-2819.
Benson, J. A., C. K. Reynolds, P. C. Aikman, B. Lupoli, and D. E. Beever. 2002. Effects of
abomasal vegetable oil infusion on splanchnic nutrient metabolism in lactating dairy cows.
J. Dairy Sci. 85:1804-1814.
Bradford, B. J., and M. S. Allen. 2005. Phlorizin administration increases hepatic gluconeogenic
enzyme mRNA abundance but not feed intake in late-lactation dairy cows. J. Nutr.
35:2206-2211.
Bradford, B. J., and M. S. Allen. 2007. Depression in feed intake by a highly fermentable diet is
related to plasma insulin concentration and insulin response to glucose infusion. J Dairy
Sci. 90:3838-3845.
Brindle, N. P. J., V. A. Zammit, and C. I. Pogson. 1985. Regulation of carnitine
palmitoyltransferase activity by malonyl-CoA in mitochondria of sheep liver, a tissue with
low capacity for fatty acid synthesis. Biochem. J. 232:177-182.
53

Broderick, G. A., and J. H. Kang. 1980. Automated simultaneous determination of ammonia and
total amino acids in rumen fluid and in vitro media. J. Dairy Sci. 63:64–75.
Carlson, D. B., N. B. Litherland, H. M. Dann, J. C. Woodworth, and J. K. Drackley. 2007.
Dietary L-carnitine alters periparturient nutrient metabolism and lactation in multiparous
cows. J. Dairy Sci. 90:3422-3441.
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.).
Cochran, R. C., D. C. Adams, J. D. Wallace, and M. L. Galyean. 1986. Predicting the
digestibility of different diets with internal markers: evaluation of four potential markers. J.
Anim. Sci. 63:1476-1483.
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.
Davis, C. L. 1967. Acetate production in the rumen of cows fed either control or low-fiber, highgrain diets. J. Dairy Sci. 50:1621–1625.
Dijkstra, J., H. Boer, J. Van Bruchem, M. Bruining, and S. Tamminga. 1993. Absorption of
volatile fatty acids from the rumen of lactating dairy cows as influenced by volatile fatty
acid concentration, pH and rumen liquid volume. Br. J. Nutr. 69: 385-396.
Faulkner, A., and H. T. Pollock. 1986. Propionate metabolism and its regulation by fatty acids in
ovine hepatocytes. Comp. Biochem. Physiol. B. 84:559-563.
Friedman, M. I. 1995. Control of energy intake by energy metabolism. Am. J. Cli. Nutr.
62:1096S-1000S.
Friedman, M. I., M. G. Tordoff, and I. Ramirez. 1986. Integrated metabolic control of food
intake. Brain Res. Bull. 17:855-859.
Goering, H. K., and P. J. Van Soest. 1970. Forage Fiber Analysis (Apparatus, Reagents,
Procedures, and Some Applications). Agricultural Handbook no. 379. ARS-USDA,
Washington, DC.
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.
Jesse, B. W., R. S. Emery, and J. W. Thomas. 1986. Control of bovine hepatic fatty acid
oxidation. J. Dairy Sci. 69:2290-2297.
Karkalas, J. 1985. An improved enzymatic method for the determination of native and modified
starch. J. Sci. Food Agric. 36:1019–1027.

54

King, T. M., and P. D. Reiss. 1985. Separation and measurement of short-chain coenzyme A
compounds in rat liver by reversed- phase high-performance liquid chromatography. Anal.
Biochem. 146:173-179.
Knapp, J. R., H. C. Freetly, B. L. Reis, C. C. Calvert, and R. L. Baldwin. 1992. Effects of
somatotropin and substrates on patterns of liver metabolism in lactating dairy cattle. J.
Dairy Sci. 75:1025-1035.
Langhans, W., and E. Scharrer. 1992. Metabolic control of eating. World Rev. Nutr. Diet. 70:167.
Langhans W., C. Leitner, and M. Arnold 2011. Dietary fat sensing via fatty acid oxidation in
enterocytes: possible role in the control of eating. Am. J. Physiol. 300:R554-R565.
Larsen, M., and N. B. Kristensen. 2009. Effect of abomasal glucose infusion on splanchnic and
whole body glucose metabolism in periparturient dairy cows. J. Dairy Sci. 92:1071-1083.
Leury, B. J., L. H. Baumgard, S. S. Block, S. Nthabisheng, R. A. Ehrhardt, R. P. Rhoads, D. E.
Bauman, A. W. Bell, and Y. R. Boisclair. 2003. Effect of insulin and growth hormone on
plasma leptin in periparturient dairy cows. Am. J. Physiol. Regul. Integr. Comp. Physiol.
285:R1107-15.
Lowe, D. M., and P. K. Tubbs. 1985. Succinylation and inactivation of 3-hydroxyl-3methylglutaryl-CoA synthase by succinyl-CoA and its possible relevance to the control of
ketogenesis. Biochem. J. 232:37-42.
Murphy, M.R., C. L. Davis, and G. C. McCoy. 1983. Factors affecting water consumption by
Holstein cows in early lactation. J. Dairy Sci. 66:35-38.
th

NRC. 2001. Nutrient requirements of dairy cattle. 7 rev. ed. Washington: National Academy of
Science.
Oba, M., and M. S. Allen. 2003a. Effects of corn grain conservation method on feeding behavior
and productivity of lactating dairy cows at two dietary starch concentrations. J. Dairy Sci.
86:174-183.
Oba, M., and M. S. Allen. 2003b. Dose-response effects of intraruminal infusion of propionate
on feeding behavior of lactating cows in early or midlactation. J. Dairy Sci. 86:2922-2931.
Oba, M., and M. S. Allen. 2003c. Extent of hypophagia caused by propionate infusion is related
to plasma glucose concentration in lactating dairy cows. J. Nutr. 133:1105-1112.
Oba, M., R. L. Baldwin VI, and B. J. Bequette. 2004. Oxidation of glucose, glutamate, and
glutamine by isolated ovine enterocytes in vitro is decreased by the presence of other
metabolic fuels. J. Anim. Sci. 82:479-486.

55

Reynolds, C. K. 1995. Quantitative aspects of liver metabolism in ruminants. Ruminant
Physiology: Digestion, Metabolism, Growth and Reproduction, W. v. Englehardt, S.
Leonhard-Marek, G. Breves, D. Geisecke, ed. Ferdinand Enke Verlag, Stuttgart,
Germany.p 351-372.
Reynolds, C. K., P. C. Aikman, B. Lupoli, D. J. Humphries, and D. E. Beever. 2003. Splanchnic
metabolism of dairy cows during the transition from late gestation through early lactation.
J. Dairy Sci. 86:1201-1217.
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.
Sutton, J. D., M. S. Dhanoa, S. V. Morant, J. France, D. J. Napper, and E. Schuller. 2003. Rates
of production of acetate, propionate, and butyrate in the rumen of lactating dairy cows
given normal and low-roughage diets. J. Dairy Sci. 86:3620-3633.
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.
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.
Woodford, S. T., M. R. Murphy, and C. L. Davis. 1984. Water dynamics of dairy cattle as
affected by initiation of lactation and feed intake. J. Dairy Sci. 67:2336-2343.

56

CHAPTER 3

HYPOPHAGIC EFFECTS OF PROPIONIC ACID ARE NOT ATTENUATED DURING
A THREE-DAY INFUSION IN THE EARLY POSTPARTUM PERIOD IN HOLSTEIN
COWS

ABSTRACT

We previously showed that propionic acid was more hypophagic than acetic acid when infused
intraruminally in cows in the postpartum period and that the degree of hypophagia from shortterm propionic acid infusion (18 h) was related to the acetyl CoA concentration in the liver. The
objective of this experiment was to evaluate adaptation over time with longer-term infusions
over three days. Twelve multiparous cows (2-13 d postpartum) were blocked by calving date
and assigned randomly to treatment sequence in a crossover design experiment with a covariate
period. Treatments were 1.0 M propionic acid or 1.0 M acetic acid, infused intraruminally at 0.5
mol VFA/h beginning 6 h prior to feeding and continuing for 78 h with 3 d between infusions.
Propionic acid decreased DMI relative to acetic acid (15.9 vs. 17.0 kg/d; P < 0.05). However, a
period by treatment interaction was detected for DMI (P = 0.07). During period 1, propionic
acid decreased DMI relative to acetic acid (14.3 vs. 17.5 kg/d; P = 0.006) because of a reduction
in meal size (1.30 vs. 1.65 kg; P = 0.05) with no effect on intermeal interval (P = 0.72).
Propionic acid decreased DMI over the first 4 h following feeding (5.86 vs. 8.23 kg; P = 0.006)
but did not affect DMI 4 h to 24 h after feeding (P = 0.22). The depression in DMI in period 1
was positively related to hepatic acetyl CoA concentration during the covariate period.

57

Propionic acid was increasingly more hypophagic than acetic acid as hepatic acetyl CoA
concentration is elevated (interaction P = 0.07 for daily DMI and P = 0.007 for DMI during the
first 4 h following feeding). There was no treatment by day interaction for DMI (P = 0.47)
suggesting little or no measurable adaptation to treatment over the 3-d infusion period. These
results suggest that hypophagia from propionic acid is enhanced when hepatic acetyl CoA
concentrations are elevated, such as when cows are in a lipolytic state.

58

INTRODUCTION

In the weeks following parturition, cows are in negative energy balance as a result of
increased energy demand at the onset of lactation and suppressed feed intake in the peripartum
period (Ingvartsen and Andersen, 2000; Gulay et al., 2004). Whole body glucose demand more
than doubles following calving (Bell, 1995), and propionate accounts for up to 60% of hepatic
glucose release during this time (Reynolds et al., 2003). In order to support milk production and
improve energy balance after parturition, starch sources are included in diets to increase energy
and provide glucose precursors (Allen et al., 2009). Ruminal starch fermentability varies from
50 to 94% depending on processing method and grain type (Huntington, 1997), but there is
limited research examining the effects of feeding a highly fermentable starch source after
calving. Dann et al. (1999) reported that steam flaked corn (a more ruminally fermentable starch
source) tended to reduce DMI compared with cracked corn when fed to cows in the postpartum
period. Greater ruminal fermentability of starch increases propionic acid production in the
rumen (Oba and Allen, 2003a), and the flux of propionic acid to the liver increases rapidly
following feeding (Benson et al., 2002).
Intraruminal propionic acid infusion has been shown to be hypophagic in ruminants and
this effect is greater for cows early lactation compared with mid-lactation (Oba and Allen,
2003b). Recent work in our laboratory has shown that hypophagic effects of propionic acid
infusions are more pronounced when cows are in a lipolytic state (Stocks and Allen, 2012).
Elevated plasma NEFA concentration during lipolysis increases concentration of hepatic acetyl
CoA, which can be oxidized in the TCA cycle or exported as ketones. Intraruminal infusion of
propionic acid reduces plasma BHBA concentration without reducing plasma NEFA

59

concentration (Stocks and Allen, 2012), suggesting increased oxidation of liver acetyl CoA,
rather than a reduction in lipolysis leading to the reduction in plasma BHBA concentration. We
proposed that propionic acid stimulates oxidation of hepatic acetyl CoA in the TCA cycle
generating ATP and sending a satiety signal to brain feeding centers (Allen et. al, 2009).
Despite consistent hypophagic effects of propionate during short-term infusions in our
laboratory, the potential exists for adaptation over long-term infusions, which might affect
feeding responses. Possible adaptations to propionic acid include altered concentrations of
hormones or metabolites in plasma, or alterations in gene expression. Therefore, the objective of
this experiment was to determine feeding response to longer-term (3 d) propionic acid infusion.
We hypothesized that intraruminal infusion of propionic acid, compared with acetic acid, would
decrease feed intake, but that the effects would be attenuated over time.

60

MATERIALS AND METHODS

Animals, housing, and diets The Institutional Animal Care and Use Committee at
Michigan State University approved all experimental procedures for this experiment. Twelve
multiparous, lactating Holstein cows were ruminally cannulated at least 45 d before calving and
cows were housed in individual tie stalls for the 12-d duration of the experiment. Cows were fed
once daily (1200 h) at 115% of expected intake and received a common experimental diet from
parturition through the end of the experiment. The experimental diet (Table 3.1) was composed
of corn silage, alfalfa silage, alfalfa hay, ground corn, soybean meal, soy hulls, and a vitamin and
mineral mix and formulated to meet requirements for absorbed protein, minerals and vitamins
(NRC, 2001).

61

Table 3.1. Ingredients and nutrient composition of experimental diet (% of dietary DM except
for DM).
Diet ingredient
Corn silage
Alfalfa silage
Alfalfa hay
Ground corn
Soybean meal
1
Vitamin and mineral mix
Nutrient composition
DM
OM
Starch
NDF
ADF
CP
Ether extract

%
43.0
29.4
6.4
10.0
7.2
4.0
47.6
93.0
19.1
38.6
28.8
14.4
3.3

1

Vitamin and mineral mix contained 24.8% ground corn grain, 21.5% dehydrated cane molasses,
11.2% limestone, 9.6% blood meal, 9.0% sodium bicarbonate, 6.6% dicalcium phosphate, 4.2%
Reashure choline, 3.1% magnesium sulfate, 2.8% salt, 2.0% animal fat, 1.5% niacin, 1.3% trace
mineral mix, 0.95% biotin, 0.70% Yeast Plus, 0.54% vitamin ADE premix, 0.32% selenium
yeast, and 0.09% Rumensin 90.

Experimental design and treatments The experiment was a crossover design with two
pretreatment covariate days used to establish baseline measurements. Cows were between 2 and
13 d postpartum at the start of the experiment and were assigned to block by calving date, then
randomly assigned to treatment within a block. The experiment was 12 d long and conducted
with 3 blocks of cows containing 4 cows each. Jugular catheters were inserted according to the
procedure of Bradford and Allen (2006) 2 d prior to the start of the experiment and were
maintained through the end of the experiment. The two covariate days, d 1 and d 8, were one
day prior to the start of each infusion period. Following each covariate day, there was an 18 h
rest period prior to the initiation of the infusions. Treatments were infused on d 3 – 5 (period 1)
and d 10 – 12 (period 2) of the experiment. Treatments were propionic or acetic acids (1 mol/L)

62

infused continuously into the rumen at 0.5 mol VFA/h for 78 h (39 moles/78 h infusion).
Solutions were infused at 500 mL/h using peristaltic pumps (#78016-30, Cole-Parmer
Instrument, IL) with Tygon tubing (1.6 mm i.d.) from individual containers that were manually
replenished hourly to ensure an accurate hourly infusion rate. Infusions began 6 h prior to
feeding on the first day of each infusion period to reach a steady state VFA concentration in the
rumen before starting feeding behavior monitoring.
Cows were blocked from feed from 1000 h to 1200 h daily to weigh and collect orts. At
1200 h, fresh feed was offered. Samples of all diet ingredients (0.5 kg), the TMR (0.5 kg), and
orts (12.5% of the remaining feed) were collected daily and composited into one sample per cow
per block for analysis. Body weight and body condition score were recorded on d 1 of the
experiment. Body condition was scored by three trained investigators using a 5-point scale,
where 1 = thin and 5 = fat, as described by Wildman et al. (1982). Cows were milked twice
daily at 0500 and 1700 h in their stalls during covariate and infusion days and in the milking
parlor on remaining days.
Covariate sample and data collection Feeding behavior was monitored for 24 h on each
of the covariate days. Feeding behavior data (chewing, feed disappearance, and water intake)
were recorded via computer every 5 s, which allowed calculation of meal size, intermeal interval,
water intake, and eating, ruminating, and total chewing time as described by Dado and Allen
(1993). Blood, rumen, and fecal samples were collected every 4 h for 24 h (n = 6) during each
covariate day. Rumen fluid samples were collected from five different sites in the rumen,
squeezed through a nylon screen, and pH was determined immediately. Samples were stored
frozen at -20°C for later analysis of VFA and ammonia-N concentrations. Blood samples were
collected via jugular catheter and transferred into two Vacutainer tubes: one with potassium

63

EDTA and one with potassium oxalate and sodium fluoride (as a glycolytic inhibitor). Blood
samples were cooled on ice until centrifugation at 2000 × g for 20 min (within 15 min of sample
collection); plasma was harvested and frozen at -20°C for analysis of metabolites and hormones.
A 1 mL aliquot of plasma from each potassium EDTA tube was stored with 0.05 M benzamidine
(final concentration) to prevent enzymatic degradation of glucagon. Liver tissue was collected
by needle biopsy (Bradford and Allen, 2005) once at the end of each covariate day and stored at 80°C until analysis for acetyl CoA concentration.
Infusion period data and sample collection Feeding behavior data was collected as
described above for 72 h starting at feeding on the first day of each infusion period. Blood and
rumen fluid samples were collected prior to the start of each infusion period, after 30 and 54 h
(end of d 1 and d 2) of infusion, and immediately after each infusion period ended. Additional
blood samples were collected every 4 h on the first and third day each infusion period (1200,
1600, 2000, 2400, 0400, and 0800 h). Liver biopsies were collected at 1200 h after 1 d and 3 d
of infusion during each period. The samples were processed and stored as described above.
Analysis of samples Feed, 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 in a Wiley mill (1-mm
screen; Arthur H Thomas, Philadelphia, PA) and analyzed for ash, NDF, indigestible NDF, CP,
starch, and ether extract. Ash concentration was determined after 5 h of oxidation at 500°C.
Concentration of NDF was analyzed according to Mertens (2002). Indigestible NDF was used as
an internal marker to estimate total tract nutrient digestibility (Cochran et al., 1986). Indigestible
NDF was estimated as NDF residue after 240 h in vitro fermentation (Goering and Van Soest,
1970); flasks were re-inoculated at 120 h to ensure a viable microbial population. Ruminal fluid
for the in vitro incubation was collected from a non-pregnant dry cow fed dry hay only. Crude

64

protein was determined according to Hach et al. (1987). Starch was analyzed using an enzymatic
method (Karkalas, 1985) after samples were gelatinized with sodium hydroxide. Glucose was
measured using a glucose oxidase method (PGO Enzyme product No. P7119, Sigma Chemical
Co., St. Louis, MO). Ether extract was determined using a modified Soxhlet apparatus (AOAC,
1990). All nutrients are expressed as percentages of DM determined by drying at 105°C in a
forced air oven for more than 8 h.
Plasma samples were analyzed using commercial kits for concentration of NEFA (NEFA
HR kit, Wako Chemicals USA, Richmond, VA), BHBA (kit #2240, Stanbio Laboratory, Boerne,
TX), insulin (Coat-A-Count, Siemens Healthcare Diagnostics, Deerfield, IL), and glucagon (kit
#GL-32K, Millipore, Billerica, MA). Plasma glucose concentration was analyzed using a
glucose oxidase method (PGO Enzyme product No. P7119, Sigma Chemical Co., St. Louis, MO).
Plasma and rumen fluid VFA concentrations were determined by HPLC according to Oba and
Allen (2003b). Rumen ammonia-N concentration was determined by colorimetric assay using
the method of Broderick and Kang (1980). Liver acetyl CoA was analyzed by HPLC using a
method previously described (Stocks and Allen, 2012).
RNA extraction and real-time quantitative PCR Expression of genes related to glucose,
lipid, and propionate metabolism were analyzed to evaluate responses to treatment infusion of
propionic or acetic acid. Methods, primers, and results are reported in Appendix A at the end of
this chapter.
Statistical analysis Feeding behavior, intake, metabolite and hormone responses, and
rumen fermentation characteristics data were analyzed using Proc Mixed in SAS (version 9.2)
with the following model:
Yijklm = µ + Bi + C(Bi)j + P(Bi) k + Dl + Tm + PkTm + DlTm + Cov + TmCov + eijklm

65

where Yijklm = the dependent response variable of interest, µ = overall mean, Bi = fixed effect of
block (i = 1 to 3), C(Bi)j = random effect of cow within block (j = 4), P(Bi) k = fixed effect of
period within block (k = 1 to 2), Dl = fixed effect of day of the treatment infusion (l = 1 to 3), Tm
= fixed effect of treatment (m = 1 to 2), PkTm = period x treatment, Dl Tm = day x treatment,
Cov = effect of covariate, TmCov = treatment x covariate, and eijklm = residual, normally
distributed. Treatment effects were declared significant at P ≤ 0.05 and tendencies for treatment
effects at P ≤ 0.10. Interactions were declared significant at P ≤ 0.10. Covariate interactions
were removed from the model if P > 0.20. All data is expressed as least squared means and
standard error of the means, unless otherwise specified.

66

RESULTS

Feed intake and feeding behavior Propionic acid infusion, relative to acetic acid
infusion, reduced DMI (P = 0.05; Table 3.2), despite a treatment by period interaction (P = 0.07;
Table 3.2). During period 1 (Table 3.3), propionic acid infusion reduced DMI by 18.3% relative
to acetic acid infusion (14.3 vs. 17.5 kg/d; P = 0.006) but did not affect DMI (P = 0.52) during
period 2. Although, propionic acid infusion did not affect meal size, meal frequency, or
intermeal interval, there was a treatment by period interaction for meal size (P = 0.02; Table
3.2). The reduction in DMI in period 1 was because propionic acid infusion decreased meal size
by 20.7% (1.30 vs. 1.65 kg; P = 0.05; Table 3.3), while meal size in period 2 was not affected by
treatment (P = 0.26; Table 3.3). The treatment by period interaction might have been from a
carryover effect of treatment, therefore, the remainder of the results will pertain to the treatment
responses from period 1 only. Propionic acid reduced feed intake during the first 4 h after
feeding (P = 0.006; Table 3.4) but not during the remaining 20 h of the day (P = 0.22; Table 3.4)
and propionic acid reduced water intake 25.3% relative to acetic acid (P = 0.01; Table 3.4).
Milk production and energy balance Propionic acid increased milk production compared
with acetic acid (40.7 vs. 39.6; P = 0.05). However, there was no difference in milk production
during period 1 or period 2 (P > 0.33). All cows were in negative energy balance during the first
covariate period. The range in energy balance was from -21 to -158 MJ/d and the mean was -74
MJ/d.

67

Table 3.2. Effects of intraruminal infusion of propionic acid relative to acetic acid on feeding
behavior and intake for cows in early lactation (overall effects of treatment).
P
Per

Acetic acid

1

Propionic
SE
Trt
Trt × Per
acid
0.06
DMI (kg/d)
17.0
15.9
0.70
0.05
0.07
0.59
Meal size (kg)
1.55
1.47
0.144 0.49
0.02
0.53
Meal frequency (#/d)
11.3
11.4
0.68
0.86
0.70
0.76
Intermeal interval (min)
83.7
89.0
7.06
0.37
0.42
1
P-values for treatment (Trt), period (Per), and treatment by covariate interaction (Trt × Per)

Table 3.3. Effects of intraruminal infusion of propionic acid relative to acetic acid on feeding
behavior and intake for cows in early lactation during period 1 and period 2.
1

Acetic
acid
DMI (kg/d)
Period 1
Period 2
Meal size (kg)
Period 1
Period 2
1
P-values for treatment (Trt)

Propionic
acid

SE

P
Trt

17.5
16.6

14.3
17.4

0.70
1.11

0.006
0.52

1.65
1.46

1.30
1.64

0.12
0.19

0.05
0.26

68

Table 3.4. Effect of intraruminal infusion of propionic acid relative to acetic acid on feeding
behavior for cows in early lactation during period 1 only.
1

Acetic acid
DMI (kg/d)
ME intake (MJ/d)
Total ME intake (MJ/d)
Meal size (kg)
Meal length (min)
Meal frequency (#)
Intermeal interval (min)
DMI 0-4 h (kg)
DMI 4-20 h (kg)
Water intake (L)
1

17.5
111
121
1.65
32.3
10.9
87.3
8.23
9.25
67.5

Propionic
acid
14.3
92
110
1.30
28.5
11.5
83.9
5.86
8.41
50.4

SE

Trt

P
Day

0.70
6.41
6.41
0.12
1.84
0.67
6.85
0.75
0.73
3.81

<0.01
0.03
0.15
0.05
0.17
0.54
0.72
<0.01
0.22
0.01

0.53
0.15
0.15
0.18
0.46
0.13
0.95
0.04
0.64
0.26

Trt × day
0.47
0.56
0.56
0.24
0.15
0.10
0.46
0.30
0.41
0.24

P-values for treatment (Trt), day (Day), and treatment by day interaction (Trt × day)

Interaction between treatment and day of infusion There was no interaction between
infusion treatment and day of the infusion for DMI, total ME intake, meal size, intermeal
interval, DMI from 0 – 4 h and 4 – 24 h, or water intake (P > 0.24; Table 3.4). However, there
was a treatment by day interaction for meal frequency (P = 0.10); propionic acid increased meal
frequency on d 2 and d 3 relative to d 1 (Figure 3.1), while acetic acid did not.

69

Figure 3.1. The interaction (P = 0.10) between treatment infusion and meal frequency in
infusion period 1 (days 3 – 5) for cows infused intraruminally with either propionic or acetic
acid.

No significant treatment-by-day interactions were detected for plasma and liver
metabolites and hormones during the infusion period (data not shown). However, there was a
significant interaction between the covariate day (d 0) and the infusion period for plasma BHBA
concentration (P = 0.001; Figure 3.2). Propionic acid reduced plasma BHBA concentration from
the covariate day to d 1 and the reduction in plasma BHBA concentration was sustained
throughout the infusion period, while the opposite relationship was observed for acetic acid.
Additionally, there were no effects of propionic acid on plasma glucose, insulin, glucagon,
NEFA, acetic acid, or propionic acid concentrations (P > 0.11; Table 3.5). Propionic acid

70

reduced plasma BHBA by 72.9% (5.1 vs. 18.8 mg/dL; P = 0.007; Table 3.5) and reduced hepatic
acetyl CoA by 62.4% (2.85 vs. 7.58 nmol/g; P = 0.04; Table 3.5) compared with acetic acid.

Figure 3.2. Interaction (P = 0.001) between treatment and day for plasma BHBA concentration
from the covariate day to infusion period 1 (days 3 – 5) for cows infused intraruminally with
either propionic or acetic acid.

71

Table 3.5. Effect of intraruminal infusion of propionic acid relative to acetic acid on plasma
metabolite and hormone concentrations and hepatic acetyl CoA during period 1.
1

Glucose (mg/dL)
Insulin (µIU/mL)
Glucagon (pg/mL)
NEFA (µEq/L)
BHBA (mg/dL)
Acetic acid (mM)
Propionic acid (mM)
Hepatic acetyl CoA (nmol/g)
1
P-values for treatment (Trt)

Acetic
acid
48.1
2.36
176
910
18.8
1.06
0.38
7.58

Propionic
acid
50.8
3.99
168
747
5.1
0.69
0.38
2.85

SE

P
Trt

2.16
1.39
40.1
181.6
2.82
0.15
0.04
1.92

0.19
0.18
0.73
0.42
<0.01
0.11
0.87
0.04

Interaction between treatment and hepatic acetyl CoA on intake and feeding behavior Infusion
treatment interacted with covariate hepatic acetyl CoA concentration (P = 0.07; Table 3.6);
propionic acid had a more pronounced hypophagic effect with greater hepatic acetyl CoA
concentration (Figure 3.3). The significance of the interaction was greater for DMI during the
first 4 h after feeding (P = 0.007; Table 6). No interactions were observed between treatment
and hepatic covariate acetyl CoA concentration for other intake or feeding behavior responses
(Table 3.6).

72

Table 3.6. Effect of intraruminal infusion of propionic acid relative to acetic acid on feeding
behavior for cows in early lactation during period 1.
P

DMI (kg/d)
ME intake (MJ/d)
Total ME intake
(MJ/d)
Meal size (kg)
Meal length (min)
Meal frequency
(#)
Intermeal interval
(min)
DMI 0-4 h (kg)
DMI 4-20 h (kg)
Water intake (L)

Trt
0.72
0.83
0.72

Cov
AcCoA
0.30
0.33
0.33

Trt ×
1
CovAcCoA
0.07
0.16
0.16

0.12
1.85
0.68

0.05
0.20
0.61

0.40
0.88
0.79

0.32
0.98
0.96

83.9

6.85

0.72

0.27

0.55

5.37
8.41
50.4

0.45
0.73
3.90

0.82
0.22
0.01

0.08
0.59
0.25

0.007
0.83
0.26

Acetic
acid
17.5
111
121

Propionic
acid
13.9
88
106

SE
0.58
5.8
5.8

1.65
32.4
10.9

1.30
28.8
11.4

87.3
8.24
9.25
67.5

1

P-values for treatment (Trt), covariate acetyl CoA (Cov AcCoA), and treatment by covariate
acetyl CoA interaction (Trt × CovAcCoA)

73

Figure 3.3. Interaction (P = 0.07) between treatment and liver acetyl CoA concentration for DMI
during the 3 d infusion in period 1 (days 3 – 5) for cows infused intraruminally with propionic
acid (solid line) or acetic acid (dotted line).

Ruminal pH, VFA concentration and profile, and ammonia-N There were no treatment
effects on any ruminal parameter measured. Mean rumen pH was 6.59 and mean rumen
ammonia-N concentration was 8.22 mM. The total rumen VFA concentrations averaged 115.4
mM for total VFA, 75.6 mM for acetate, and 22.5 mM for propionate. There was a numeric
increase in rumen propionic acid concentration for the propionic acid infusion and a numeric
increase in rumen acetic acid concentration for the acetic acid infusion.

74

DISCUSSION

Intraruminal propionic acid infusion decreased DMI relative to acetic acid as previously
reported in the literature as reviewed by Allen et al. (2009). However, the significant treatment
by period interaction suggested a potential carryover effect of the propionic acid treatment and
led us to analyze each period independently. The 18.3% reduction in DMI by propionic acid
infusion during period 1 of this experiment was expected and consistent with previous results for
cows at a similar stage of lactation (Oba and Allen, 2003b; Stocks and Allen, 2012). The
ruminant liver rapidly extracts absorbed propionic acid (Benson et al., 2002) and up to 70% of
propionic acid in the portal vein is extracted on the first pass (Reynolds et al., 2003). In contrast,
acetic acid is not readily extracted from the blood by the liver (Reynolds et al., 1995) because
activity of acetyl CoA synthetase is low (Ricks and Cook, 1981). Therefore, propionic acid can
stimulate oxidation of hepatic acetyl CoA (Allen et al., 2009), which may stimulate satiety, while
acetic acid does not. Propionic acid infusion decreased water intake relative to acetic acid
infusion consistent with previous work in our laboratory (Stocks and Allen, 2012). Water intake
is positively correlated to DMI (Woodford et al., 1984), and most variation in water intake is due
to variation in DMI (Murphy et al., 1983). Therefore the reduction in water intake by propionic
acid (Table 3.4) is likely explained by the reduction in DMI, which is consistent with the
treatment responses and with previous results (Stocks and Allen, 2012).
The current experiment and previous experiments in our laboratory (Oba and Allen, 2003b;
Stocks and Allen, 2012) have shown that propionic acid is hypophagic for cows in early
lactation, particularly when cows are in a lipolytic state. Short-term infusions of propionic acid
have caused hypophagia when cows are in a lipolytic state (Oba and Allen, 2003b; Stocks and
75

Allen, 2012), but it is necessary to explore the longer-term impact of propionic acid infusion on
feed intake to determine if cows adapt to the increased propionic acid load. This experiment was
designed to evaluate the adaptive response to propionic acid infusion over the 3-d infusion
period. We anticipated that propionic acid would reduce DMI on the first day of the infusion
period but the depression would be alleviated as the infusion progressed and hepatic acetyl CoA
concentration decreased. The lack of a significant treatment by day interaction for DMI (Table
3.4) does not support our hypothesis, and propionic acid infusion resulted in a sustained
reduction in DMI over the course of the 3-d infusion period. However, propionic acid infusion,
relative to acetic acid infusion, decreased, but did not deplete, plasma BHBA (Figure 3.2) and
hepatic acetyl CoA concentrations (data not shown). While there was no treatment by day
interaction for plasma BHBA concentration for the 3-d infusion period (data not shown), we did
observe an interaction between treatment and day for the covariate day (d 0) and the 3 infusion
days for plasma BHBA concentration. The 41.4% reduction in plasma BHBA concentration
from the covariate day to d 1 of the infusion period (after 24 h of treatment infusion) is consistent
with previous results (Oba and Allen, 2003b; Stocks and Allen, 2012), and this reduction was
sustained for the duration of the infusion period (Figure 3.2). The reduction in plasma BHBA
concentration could have been a result of 1) decreased β-oxidation of NEFA, 2) reduced
ketogenesis, or 3) increased oxidation of acetyl CoA (Stocks and Allen, 2012). The reduction in
plasma BHBA and hepatic acetyl CoA concentrations are likely a result of propionic acid
stimulating increased oxidation of acetyl CoA in the TCA cycle (reviewed by Allen et al., 2009)
and decreasing ketogenesis (Faulkner and Pollock, 1986) by reducing the activity of 3-hydroxy3methyl-glutaryl (HMG)-CoA synthase (Lowe and Tubbs, 1985). This would be consistent with
the reduced meal size, and resulting in decrease in DMI, as observed.
76

Propionic acid infusion decreased meal size but not meal frequency, resulting in a
reduction in DMI (Table 3.4) consistent with our previous results (Oba and Allen, 2003b; Stocks
and Allen, 2012). We also observed that the hypophagic effect of propionic acid was within the
first 4 h following feeding (Table 3.4) which could be a result of the larger meal size during this
time (data not shown). The first 2 meals, which generally occur within the first 4 h of feeding,
tend to be the largest meals observed for cows throughout the day (Dado and Allen, 1994), and
subsequent meals are smaller. The production of propionic acid in the rumen, and subsequent
flux to the liver would be largest within the time course of the larger meals. Additionally,
plasma NEFA concentration, and likely hepatic acetyl CoA concentration, is highest at feeding
and reaches a nadir by 4 h post-feeding (Allen et al., 2005). Increased flux of propionic acid to
the liver would stimulate oxidation of hepatic acetyl CoA, increasing energy charge within the
meal, leading to a reduction in meal size and DMI during the first 4 h following feeding, as
observed. However, because plasma NEFA concentration is reduced within 4 h of feeding and
remains low through most of the day and only begins to increase within hours of the next
conditioned meal at feeding, it is likely that propionic acid would have less effect on DMI from 4
– 20 h post-feeding, consistent with our results.
The interaction between treatment and hepatic acetyl CoA concentration for DMI (Figure
3.3) was consistent with previous research in our laboratory (Stocks and Allen, 2012). Propionic
acid was more hypophagic when hepatic acetyl CoA concentration was high when cows are in a
lipolytic state (e.g. early lactation). Hypophagia from propionic acid was sustained over the 3-d
infusion period despite the continued reduction, but not depletion, of hepatic acetyl CoA
concentration. We cannot discern whether hypophagia from propionic acid occurs simply in the

77

presence of hepatic acetyl CoA or rather if it is due to the rate of reduction of the pool of hepatic
acetyl CoA.
Additionally, liver samples were collected only once per day during this experiment, so
we are unable to determine the temporal pattern of accretion or depletion of hepatic acetyl CoA
throughout the course of a day and how this may affect feed intake and feeding behavior. The
largest meals generally occur within the first 4 h following feeding and that the greatest
hypophagic effects of propionic acid infusion occur during this same time period, so it is possible
that either the pool of acetyl CoA is largest during this time or that the rate of oxidation of acetyl
CoA is highest during this time, stimulating satiety. We speculate that acetyl CoA accumulates
in the liver in the hours prior to feeding, reaching its peak right at the beginning of the
conditioned meal at feeding, and then is subsequently oxidized during large meals immediately
following feeding. The idea of a cyclic pattern of accumulation and depletion of acetyl CoA in
the ruminant liver is consistent with the reduction in feed intake observed during the first 4 h
following feeding in the current study with no difference in intake during the remaining 20 h of
the day. While data on diurnal variation of hepatic acetyl CoA concentration is not available, it
is likely consistent with the diurnal variation in plasma NEFA concentration, which is highest
immediately prior to feeding (Allen et al., 2005).

78

CONCLUSIONS

Feed intake in the periparturient period is likely suppressed by oxidation of NEFA
mobilized from adipose tissue. Diet composition likely interacts with lipolytic state because of
differences in supply of propionate to the liver within the timeframe of meals. Propionate supply
to the liver likely stimulated oxidation of acetyl CoA, resulting in increased energy charge which
stimulated satiety. Greater flux of propionate to the liver within the timeframe of meals with
higher starch fermentability likely decreases meal size and DMI. While attenuation of propionic
acid stimulated hypophagia may happen over the longer-term as hepatic acetyl CoA pool is
depleted, our data do not support adaptation to elevated propionic acid load over three days.
More research is needed to determine the diurnal variation in hepatic acetyl CoA as this could
lend additional information as to the within day control of feed intake. Feeding studies
evaluating cow response to highly fermentable diets and longer-term infusion studies would also
generate valuable information for use in determining how to feed and manage cows during the
early postpartum period.

79

APPENDIX

80

SUPPLEMENTAL MATERIAL REFERENCED IN CHAPTER 3

RNA Extraction and Real-Time Quantitative PCR Total RNA was extracted from 20
mg of hepatic tissue using the PerfectPure RNA Fibrous Tissue Kit (5 Prime, Inc., Gaithersburg,
MD) and genomic DNA contamination was removed using the RNeasy MinElute Cleanup kit
(Qiagen, Inc., Valencia, CA). RNA concentration was determined using a NanoDrop 1000
Spectrophotometer (Thermo Scientific, Wilmington, DE) and quality was analyzed by automated
capillary gel electrophoresis using an Agilent 2100 Bioanalyzer (Agilent Technologies, Inc.,
Santa Clara, CA).
Gene expression of pre-determined genes was determined using real-time quantitative
PCR (qPCR) using a 7900 Fast-Real Time PCR System (life Technologies Corporation,
Carlsbad, CA). Primers were designed using Primer Express 3.0 (Life Technologies Corp.) and
synthesized commercially (Sigma-Aldrich, St. Louis, MO). Primer sequences and accession
numbers are summarized in Supplemental Table 3.7. cDNA was synthesized from 500 ng total
RNA as a template using the High-Capacity cDNA Reverse Transcription Kit (Life Technologies
Corp.), and amplification efficiency was validated for endogenous control and target genes with
pooled cDNA from all samples using a 5-fold dilution series. Relative gene expression was
determined in triplicate using 5 μL Power SYBR Green (Life Technologies Corp.), 3 μL of a
forward and reverse primer mix (1 μM), and 2 μL of sample (2.5 ng cDNA/μL). The relative
gene expression was determined using the 2

-ΔΔCT

method (Livak and Schmittgen, 2001) with

HPRT or ACTB as the endogenous control gene. Gene expression is reported as the fold
difference of a gene compared to the expression during the covariate period (or d 1), which was
normalized to 1. Results of gene expression analysis are reported in Supplemental Table 3.8.
81

Table 3.7. Primer sequences and accession numbers for genes analyzed with real-time
quantitative PCR.
Gene1
ACCS3
ACTB2
AMPK
CPT1A
CS
G6PC
GPAM
HMGCL
HMGCS1
HMGCS2
HNF4A
HPRT12
MCEE
MUT
PC
PCCA
PCCB
PDK4
PEPCK
PPARA

Accession
number
NM_001102137

Forward Sequence (5’ – 3’)
Reverse Sequence (5’ – 3’)
TTGTGTTCCCGTTCTTTCTGAA
AACCCCGTCGTTCCAGATG
NM_173979
TCACGGAGCGTGGCTACAG
TTGATGTCACGGACGATTTCC
NM_001205605.1 ACAGAGATCGGGCTCAGTTAG
GAGCCTCAGCATCTGAATCATTC
FJ415874.1
ACCATGCGCTACTCCCTGAA
TGCTCGGCGAACATCCA
NM_001044721
ACTAATGCATGTAGTGTGGGTTAGGT
AAGAGCCAGATTCCCACTCTGA
NM_183364
CCATATCCGAAACCAATCAAGA
GAACAGGCAAGGAGGAGGAGTT
NM_001012282
GAAGTCATGAGGGTGACGAGAAA
TGCCACAACCTGAGCTTACACT
NM_001075132
CTGCCTGACCTGAGTATGAGTGA
TGGGTCCAGGTGGAAGGA
DAA17883
AAGCTCCGAGAGGATACTCATCA
TCGAAGAGGGAATCTATGGAACTC
AAI12667
CCTGCTGCAATCACTGTCATG
TCTGTCCCGCCACCTCTTC
NM_001015557
GGCGTCCCGCCAGACT
GCACAGGACCCGATGACATC
NM_001034035
TGGCGTCCCAGTGAAATCA
CAGCTGGCCACAGAACAAGA
NM_001045995
GGGCTGGCCCTGTATGG
GCACTGCGACTGCAACATG
NM_173939
CTCGCTGCTGGGCACAA
AAGGGCGTTGAGTTCCTTGA
NM_177946
AGGCAAGACGCTGCACATC
GCCCGCCCGGTTGA
NM_001083509
GGTCAGGAAATCTGTGTGATTGAA
GTCTTCCCGGCCGTCATA
BC109784
CGAATCTGCTGTGACCTGGAT
TCCTCCAAGGGCGCTGTAC
NM_001101883
GAGGTGGTGTTCCCCTGAGA
TTGGTGCAGTGGAGTACGTGTAA
NM_174737
CAGCCAAGCTGCCCAAGA
CCGGCCTTGTCCTTTCG
NM_001034036
CAGCGCCGAGGAGTCATC
TGTCCCCGCAGATCCTACAC
82

Table 3.7 (cont’d)
Gene1
Accession
number
SLC37A4

Forward Sequence (5’ – 3’)
Reverse Sequence (5’ – 3’)

NM_001193045

TTGTCATGCCGTCGTTGGT
GGTGATGAGCCCCAAGTCA
SREBP1
AB355703
GCGGCGAGAAGCGTACAG
ACGATCTTGTCATTGATGGAAGAG
1
ACCS3 = acyl-CoA synthetase short-chain; ACTB = actin, beta; AMPK = AMP kinase;
CPT1A = carnitine palmitoyltransferase 1A; CS = citrate synthase; G6PC = Glucose-6phosphatase; GPAM = Glycerol-3-phosphate acyltrasferase; HMGCL = HMG-CoA lyase;
HMGCS1 = HMG-CoA synthase 1; HMGCS2 = HMG-CoA synthase 2; HNF4A = Hepatocyte
nuclear factor 4, alpha; HPRT1 = hypoxanthine phosphoribosyltransferase 1; MCEE =
Methylmalonyl CoA epimerase; MUT = Methylmalonyl CoA mutase; PC = Pyruvate
carboxylase; PCCA = Propionyl-CoA carboxylase alpha; PCCB = Propionyl-CoA carboxylase
beta; PDK4 = Pyruvate dehydrogenase kinase; PEPCK = Phosphoenolpyruvate carboxykinase 1;
PPARA = Peroxisome proliferator activated receptor alpha; SLC37A4 = Solute carrier family 37
A4; SREBP1 = Sterol regulatory element-binding protein 1
2
Endogenous control gene

83

Table 3.8. Effect of intraruminal infusion of propionic acid, relative to acetic acid, on hepatic
gene expression during period 1. Cov = covariate period, D1 = day 1 relative to the start of the
infusion, D3 = day 3 relative to the start of the infusion.

Gene

Acetic acid

1
2

ACCS3 D1 – Cov

3

ACCS3 D3 – Cov
4

ACCS3 D3 - D1
AMPK D1 - Cov
AMPK D3 - Cov
AMPK D3 - D1
CPT1A D1 - Cov
CPT1A D3 - Cov
CPT1A D3 - D1
CS D1 - Cov
CS D3 - Cov
CS D3 - D1
G6PC D1 - Cov
G6PC D3 - Cov
G6PC D3 - D1
GPAM D1 - Cov
GPAM D3 - Cov
GPAM D3 - D1
HMGCL D1 - Cov
HMGCL D3 - Cov
HMGCL D3 - D1
HMGCS1 D1 - Cov
HMGCS1 D3 - Cov
HMGCS1 D3 - D1
HMGCS2 D1 - Cov
HMGCS2 D3 - Cov
HMGCS2 D3 - D1
HNF4A D1 - Cov
HNF4A D3 - Cov
HNF4A D3 - D1
MCEE D1 - Cov
MCEE D3 - Cov
MCEE D3 - D1
MUT D1 - Cov
MUT D3 - Cov
MUT D3 - D1

Propionic acid

SE

P
Trt

0.84

0.96

0.223

0.71

0.63

0.69

0.317

0.83

0.63
0.66
0.56
0.69
0.72
0.70
0.78
1.16
1.21
1.04
0.97
0.64
0.64
0.72
0.71
0.99
0.73
0.44
0.56
0.90
0.54
0.80
0.66
0.31
0.54
0.82
0.68
0.86
0.77
0.59
0.76
0.78
0.53
0.69

1.03
1.38
0.88
1.32
0.97
0.53
1.20
1.02
0.95
1.07
0.84
0.70
0.79
0.99
1.02
1.09
1.04
0.55
0.75
0.75
0.72
0.99
0.96
0.45
0.60
1.46
1.34
1.28
0.76
0.61
0.92
0.84
0.73
1.03

0.275
0.375
0.347
0.332
0.258
0.323
0.305
0.403
0.549
0.250
0.156
0.166
0.161
0.141
0.194
0.242
0.201
0.214
0.234
0.275
0.170
0.204
0.221
0.059
0.158
0.296
0.348
0.272
0.115
0.104
0.187
0.172
0.095
0.184

0.20
0.20
0.53
0.10
0.52
0.59
0.35
0.78
0.38
0.93
0.60
0.81
0.41
0.22
0.29
0.78
0.32
0.59
0.53
0.70
0.44
0.55
0.36
0.37
0.79
0.17
0.22
0.28
0.95
0.88
0.53
0.80
0.18
0.19

84

Table 3.8 (cont’d)
P
Trt

1

Acetic acid Propionic acid
SE
Gene
PC D1 - Cov
1.08
0.91
0.145
0.45
PC D3 - Cov
0.95
0.93
0.224
0.94
PC D3 - D1
0.84
1.11
0.133
0.21
PCCA D1 - Cov
0.76
0.98
0.143
0.28
PCCA D3 - Cov
0.61
0.86
0.115
0.18
PCCA D3 - D1
0.79
0.98
0.175
0.44
PCCB D1 - Cov
0.65
0.95
0.122
0.13
PCCB D3 - Cov
0.42
0.72
0.060
0.01
PCCB D3 - D1
0.63
0.97
0.152
0.16
PEPCK D1 - Cov
0.80
1.01
0.203
0.42
PEPCK D3 - Cov
0.49
0.77
0.134
0.19
PEPCK D3 - D1
0.69
0.79
0.128
0.60
PDK4 D1 - Cov
0.96
0.68
0.190
0.32
PDK4 D3 - Cov
0.91
0.85
0.162
0.79
PDK4 D3 - D1
1.06
1.49
0.243
0.21
PPARA D1 - Cov
0.98
0.82
0.119
0.38
PPARA D3 - Cov
0.70
1.05
0.219
0.31
PPARA D3 - D1
0.73
1.13
0.156
0.12
SLC37A4 D1 - Cov
0.95
1.19
0.295
0.59
SLC37A4 D3 - Cov
0.48
0.76
0.335
0.05
SLC37A4 D3 - D1
0.38
1.26
0.320
0.08
SREBP1 D1 - Cov
0.82
1.07
0.249
0.50
SREBP1 D3 - Cov
0.70
1.19
0.220
0.16
SREBP1 D3 - D1
1.02
1.44
0.462
0.53
1
ACCS3 = acyl-CoA synthetase short-chain; ACTB = actin, beta; AMPK = AMP kinase;
CPT1A = carnitine palmitoyltransferase 1A; CS = citrate synthase; G6PC = Glucose-6phosphatase; GPAM = Glycerol-3-phosphate acyltrasferase; HMGCL = HMG-CoA lyase;
HMGCS1 = HMG-CoA synthase 1; HMGCS2 = HMG-CoA synthase 2; HNF4A = Hepatocyte
nuclear factor 4, alpha; HPRT1 = hypoxanthine phosphoribosyltransferase 1; MCEE =
Methylmalonyl CoA epimerase; MUT = Methylmalonyl CoA mutase; PC = Pyruvate
carboxylase; PCCA = Propionyl-CoA carboxylase alpha; PCCB = Propionyl-CoA carboxylase
beta; PDK4 = Pyruvate dehydrogenase kinase; PEPCK = Phosphoenolpyruvate carboxykinase 1;
PPARA = Peroxisome proliferator activated receptor alpha; SLC37A4 = Solute carrier family 37
A4; SREBP1 = Sterol regulatory element-binding protein 1
2
D1 – Cov = comparison of gene expression on day 1 compared to the covariate day of the
infusion of propionic acid or acetic acid.
3
D3 – Cov = comparison of gene expression on day 3 compared to the covariate day of the
infusion of propionic acid or acetic acid.
85

Table 3.8 (cont’d)
4

D3 – D1 = comparison of gene expression on day 3 compared to day 1 of the infusion of
propionic acid or acetic acid.

86

REFERENCES

87

REFERENCES

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., B. J. Bradford, and K. J. Harvatine. 2005. The cow as a model to study food intake
regulation. Annu. Rev. Nutr. 25:523-547.
Allen, M. S., B. J. Bradford, and M. Oba. 2009. BOARD-INVITED REVIEW: The hepatic
oxidation theory of the control of feed intake and its application to ruminants. J. Anim. Sci.
87:3317-3334.
th

AOAC. 1990. Official Methods of Analysis. 15 ed. Association of Analytical Chemists.
Gaithersburg, MD.
Bell, A. W. 1995. Regulation of organic nutrient metabolism during the transition from late
pregnancy to early lactation. J. Anim. Sci. 73:2804-2819.
Benson, J. A., C. K. Reynolds, P. C. Aikman, B. Lupoli, and D. E. Beever. 2002. Effects of
abomasal vegetable oil infusion on splanchnic nutrient metabolism in lactating dairy cows.
J. Dairy Sci. 85:1804-1814.
Bradford, B. J., and M. S. Allen. 2005. Phlorizin administration increases hepatic gluconeogenic
enzyme mRNA abundance but not feed intake in late-lactation dairy cows. J. Nutr.
35:2206-2211.
Bradford, B. J., A. D. Gour, A. S. Nash, and M. S. Allen. 2006. Propionic acid challenge test
have limited value for investigating bovine metabolism. J. Nutr. 136:1915-1920.
Broderick, G. A., and J. H. Kang. 1980. Automated simultaneous determination of ammonia and
total amino acids in rumen fluid and in vitro media. J. Dairy Sci. 63:64–75.
Cochran, R. C., D. C. Adams, J. D. Wallace, and M. L. Galyean. 1986. Predicting the
digestibility of different diets with internal markers: evaluation of four potential markers. J.
Anim. Sci. 63:1476-1483.
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. 1994. Variation in and relationships among feeding, chewing, and
drinking variable for lactating dairy cows. J. Dairy Sci. 77:132-144.
Dann, H. M., G. A. Varga, and D. E. Putnam. 1999. Improving energy supply to late gestation
and early postpartum dairy cows. J. Dairy Sci. 82:1765-1778.
88

Faulkner, A., and H. T. Pollock. 1986. Propionic acid metabolism and its regulation by fatty
acids in ovine hepatocytes. Comp. Biochem. Physiol. B 84:559-563.
Goering, H. K., and P. J. Van Soest. 1970. Forage Fiber Analysis (Apparatus, Reagents,
Procedures, and Some Applications). Agricultural Handbook no. 379. ARS-USDA,
Washington, DC.
Gulay, M. S., M. J. Hayen, M. Liboni, T. I. Belloso, C. J. Wilcox, and H. H. Head. 2004. Low
doses of bovine somatotropin during the transition period and early lactation improves milk
yield, efficiency of production, and other physiological responses of Holstein cows. J.
Dairy Sci. 87:948-960.
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.
Huntington, G. B. 1997. Starch utilization by ruminants: from basics to the bunk. J. Anim. Sci.
75:852-867.
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.
Karkalas, J. 1985. An improved enzymatic method for the determination of native and modified
starch. J. Sci. Food Agric. 36:1019–1027.
Livak, K. J., and T. D. Schmittgen. 2001. Analysis of Relative Gene Expression Data Using
-ΔΔCT
Real-Time Quantitative PCR and the 2
Method. Methods. 25:402-408.
Lowe, D. M., and P. K. Tubbs. 1985. Succinylation and inactivation of 3-hydroxyl-3methylglutaryl-CoA synthase by succinyl-CoA and its possible relevance to the control of
ketogenesis. Biochem. J. 232:37-42.
Mertens, D. R. 2002. Gravimetric determination of amylase-treated neutral detergent fiber in
feeds using refluxing in beakers or crucibles: collaborative study. J. AOAC Int. 85:1217–
1240.
Murphy, M.R., C. L. Davis, and G. C. McCoy. 1983. Factors affecting water consumption by
Holstein cows in early lactation. J. Dairy Sci. 66:35-38.
th

NRC. 2001. Nutrient requirements of dairy cattle. 7 rev. ed. Washington: National Academy of
Science.
Oba, M., and M. S. Allen. 2003a. Effects of corn grain conservation method on feeding behavior
and productivity of lactating cows at two dietary starch concentrations. J. Dairy Sci.
86:174-183.
89

Oba, M., and M. S. Allen. 2003b. Dose-response effects of intraruminal infusion of propionic
acid on feeding behavior of lactating cows in early or midlactation. J. Dairy Sci. 86:29222931.
Reynolds, C. K.. 1995. Quantitative aspects of liver metabolism in ruminants. Ruminant
Physiology: Digestion, Metabolism, Growth and Reproduction, W. v. Englehardt, S.
Leonhard-Marek, G. Breves, D. Geisecke, ed. Ferdinand Enke Verlag, Stuttgart, Germany.
p 351-372.
Reynolds, C. K., P. C. Aikman, B. Lupoli, D. J. Humphries, and D. E. Beever. 2003. Splanchnic
metabolism of dairy cows during the transition from late gestation through early lactation.
J. Dairy Sci. 86:1201-1217.
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.
Stocks, S. E. and M. S. Allen. 2012. Hypophagic effects of propionic acid increase with elevated
hepatic acetyl coenzyme A concentration for cows in the early postpartum period. J. Dairy
Sci. 95:3259-3268.
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.
Woodford, S. T., M. R. Murphy, and C. L. Davis. 1984. Water dynamics of dairy cattle as
affected by initiation of lactation and feed intake. J. Dairy Sci. 67:2336-2343.

90

CHAPTER 4

EFFECTS OF LIPID AND PROPIONIC ACID INFUSIONS ON FEED INTAKE OF
LACTATING DAIRY COWS

ABSTRACT

Propionic acid is more hypophagic for cows in the postpartum period with elevated hepatic
acetyl CoA concentration. The objective of this experiment was to evaluate the interaction of
hepatic acetyl CoA concentration, which is elevated by intravenous lipid infusion, and
intraruminal propionic acid infusion on feed intake and feeding behavior responses of lactating
cows. Eight multiparous, ruminally cannulated, Holstein dairy cows past peak lactation were
used in a replicated 4x4 Latin square experiment with a 2x2 factorial arrangement of treatments.
Treatments were propionic acid (PI) infused intraruminally at 0.5 mol/h for 18 h starting 6 h
prior to feeding or sham control (CO), and intravenous jugular infusion of lipid (LI, Intralipid
20%) or saline (SI, 0.9% NaCl) infused at 250 mL/h for 12 h prior to the start of the feeding
behavior period then 500 mL/h during the 12 h feeding behavior period. Changes in plasma
concentrations of metabolites and hormones and hepatic concentration of acetyl CoA from
before infusion until the end of infusion were evaluated. There was a tendency for an interaction
between PI and LI for the change in plasma NEFA concentration from the covariate day to the
end of the infusion period (P = 0.12). Infusion of PI decreased DMI 15% (16.1 vs. 19.0 kg/12 h,
P = 0.02) compared with CO but lipid infusion did not affect DMI over the 12 h infusion period.
Infusion of PI tended to decrease hepatic acetyl CoA concentration from the covariate day to the
91

end of the infusion compared with CO (-5.7 vs. -2.4 nmol/g wet tissue, post-infusion – covariate
day; P = 0.09), consistent with PI decreasing DMI by stimulating oxidation of acetyl CoA. LI
increased hepatic acetyl CoA concentration (P = 0.04). Contrary to our expectations, LI did not
increase concentrations of NEFA (P = 0.13) or BHBA (P = 0.26) in plasma, and did not increase
milk fat yield (P = 0.67), suggesting that the infused lipid was stored or oxidized by extrahepatic
tissues. As a result, there was no interaction between PI and LI for DMI. While the effect of PI
on DMI was consistent with our previous results, this lipid infusion model using cows post-peak
lactation was not useful to simulate the lipolytic state of cows in the postpartum period.

92

INTRODUCTION

Dairy cows undergo substantial metabolic adaptations as they transition from late
gestation to the immediate postpartum period. Energy demands for lactation are ~3 fold greater
than for late gestation (as reviewed by Bell, 1995), and feed intake is often suppressed
postpartum resulting in negative energy balance (Doepel et al., 2002). Plasma insulin and
glucose concentrations are low and tissue sensitivity to insulin is reduced during the immediate
postpartum period (Bell and Bauman, 1997), elevating plasma NEFA and BHBA concentrations
postpartum (Doepel et al., 2002). In a retrospective study, Seifi et al. (2011) indicated that
elevated serum BHBA (> 12.4 mg/dL) and NEFA (> 1000 μEq/L) concentration postpartum
increased the risk of ketosis by 4.7 and 6.3 times, respectively. In addition, cows with serum
BHBA concentration > 10.4 mg/dL were 13.6 times more likely to have a DA, and cows with
NEFA concentration > 1000 μEq/L were 3.6 times more likely to be culled in the first 60 days
postpartum. Furthermore, Roberts et al. (2012) reported that elevated plasma BHBA and NEFA
increased the risk of culling in the first 60 DIM by 1.8 to 2 fold. Therefore, careful management
of the nutritional program through this time is necessary to meet the nutrient demands of
postpartum dairy cows to limit the extent and duration of lipolysis.
During the postpartum period, starch concentration in the diet is increased to provide
glucose precursors to support milk production. As reviewed by Firkins et al. (2001), ruminal
starch fermentability ranges from ~45 – 87% depending on the type and processing of corn, and
higher ruminal starch fermentability increases propionate production in the rumen as a fraction
of total VFA (Davis, 1967). Propionate accounts for about 60% of the total hepatic glucose
production for cows in early lactation (Reynolds, et al., 2003). However, previous research in
93

our laboratory has demonstrated that propionic acid is hypophagic during the immediate
postpartum period (Oba and Allen, 2003a; Stocks and Allen, 2012; Stocks and Allen, accepted)
and the extent of hypophagia is greater when cows are in a lipolytic state (Stocks and Allen,
2012; Stocks and Allen, accepted). Our laboratory has also shown that increasing glucose
demand with pholizin, which was expected to delay hepatic propionate oxidation within meals
increasing meal size and DMI, resulted in reduced DMI, presumably by stimulating lipolysis
(Bradford and Allen, 2007). This is consistent with the results of our previous experiments that
showed a greater extent of hypophagia from propionate for cows in a lipolytic state.
Despite consistent evidence that propionic acid is more hypophagic when cows are in a
lipolytic state, experiments to test cause and effect between lipolytic state and propionic acid
infusion on DMI are necessary. Experimental elevation of plasma NEFA concentration
combined with infusion of propionate to test interactions among treatments for effects on feeding
behavior will allow interactions detected to be specifically attributed to increased oxidation of
acetyl CoA. Lipid infusion is expected to increase NEFA supply to the liver and increase flux of
carbon through acetyl CoA. . The objective of this experiment was to evaluate the interaction of
hepatic acetyl CoA concentration, which is elevated by intravenous lipid infusion, and
intraruminal propionic acid infusion on feed intake and feeding behavior responses of lactating
cows. We hypothesized that the hypophagic effects of propionic acid will be enhanced for cows
receiving the lipid infusion.

94

MATERIALS AND METHODS

Animals, housing, and diets The Institutional Animal Care and Use Committee at
Michigan State University approved all experimental procedures for this experiment. Eight
lactating, ruminally cannulated Holstein cows from 81 – 252 DIM, mean BW of 718 kg (+/- 11
kg) and mean BCS of 2.65 (+/- 0.54), were housed in individual tie-stalls for the duration of the
experiment. One cow was removed from the experiment because feed intake and milk yield
were reduced by more than 75% after initiation of lipid infusion. Cows were fed at 115% of
expected intake and received a common experimental diet. The experimental diet (Table 4.1)
included corn silage, alfalfa silage, soybean meal, ground corn, and a mineral and vitamin mix
and was formulated to meet requirements for absorbed protein, minerals, and vitamins (NRC,
2001).

Table 4.1. Ingredients and nutrient composition of experimental diet (% of dietary DM except
for DM).
Diet ingredient
Corn silage
Alfalfa silage
Ground corn
Soybean meal
1
Vitamin and mineral mix
Nutrient composition
DM
OM
Starch
NDF
ADF
CP

%
50.7
28.9
9.7
6.16
4.6
51.6
92.4
21.1
36.3
23.9
13.7

1

Vitamin and mineral mix contained 60.4% ground corn grain, 18.8% limestone, 8.8% sodium
bicarbonate, 4.0% urea, 3.1% magnesium sulfate, 2.9% salt, 0.63% trace mineral mix, 0.63%
biotin, 0.63% vitamin ADE premix, and 0.20% selenium yeast.
95

Experimental design and treatments The experimental design was a duplicated 4 x 4
Latin square with a 2 x 2 factorial arrangement of treatments. Cows were randomly assigned to
block and treatment sequence, and squares were balanced for carry-over effects. The experiment
was 24 d long, and included an 8-d diet adaptation period, and four infusion periods, each with 3
d for determination of digestibility, followed by the 24 h infusion period. On d 8, cows were
fitted with bi-lateral jugular catheters according to Bradford et al. (2006) and catheters were
maintained for the duration of the experiment. The treatments were propionic acid (1 mol/L; PI)
infused continuously into the rumen at 500 mL/h for 18 h or sham control (CO) and intravenous
drip infusion of 20% Intralipid® (LI; Baxter U. S., Deerfield, IL) or physiological saline (SI)
infused at 250 mL/h for 12 h then 500 mL/h for the next 12 h. Propionic acid infusion was 18 h
to allow the treatment to approach steady state in the rumen prior to feeding. Propionic acid was
infused using peristaltic pumps (#78016-30, Cole-Parmer Instrument, Vernon Hills, IL) with
Tygon tubing (1.6 mm i.d.). The lipid and saline control infusions were initiated at a lower rate
to allow for gradual increase in blood lipid concentration to avoid complications. Treatments
were infused on d 11, 15, 19, and 23.
Data and sample collection Cows were blocked from feed from 1000 to 1200 h daily to
weigh of orts, collect orts samples, and offer feed. Samples of all diet ingredients (0.5 kg), the
TMR (0.5 kg), and orts (12.5% of remaining feed) were collected for the 3 d prior to each
infusion day and on the infusion day and composited by infusion period. Body weight and BCS
were recorded on d 1 of the experiment. Body condition was scored by 3 trained investigators on
a 5-point scale, where 1 = thin and 5 = fat, as described by Wildman et al. (1982). Cows were
milked twice daily at 0500 and 1700 h in their stalls on infusion days and in the milking parlor
on all other days. Milk samples were collected on the day prior to the infusion and on the
96

infusion day and were analyzed for fat, true protein, lactose, and SNF by Michigan DHIA
(AOAC, 1997).
Fecal samples were collected every 9 h over the 3 d prior to each infusion (n = 5) and
frozen at -20°C for later analysis to determine diet digestibility for use in calculating ME intake.
Rumen fluid, liver, and blood samples were collected 1 d prior to each infusion (d 10, 14, 18, and
22) at 1000 h and on infusion days (d 11, 15, 19, and 23) at 1000 h and 2400 h. Rumen fluid
samples were collected from 5 different sites in the rumen, squeezed through a nylon screen, and
pH was determined immediately. Samples were stored at -20°C for later analysis for VFA
concentrations. Liver tissue was collected by needle biopsy (Bradford and Allen, 2005) and
stored at -80°C until analysis for acetyl CoA concentration. Blood samples were collected by
coccygeal veinipuncture into 2 Vacutainer® tubes (Becton Dickinson, Franklin Lakes, NJ): one
tube contained potassium EDTA and one tube contained potassium oxalate and sodium fluoride
(as a glycolytic inhibitor). Blood samples were cooled on ice until centrifuged at 2,000 x g for
20 min (within 15 min of sample collection); plasma was harvested and frozen at -20°C for later
analysis of metabolites and hormones. A 1-mL aliquot of plasma from each potassium EDTA
tube was stored with 0.05 M benzamidine (final concentration) to prevent enzymatic degradation
of glucagon.
Feeding behavior was monitored for 12 h (1200 to 2400 h) on infusion days. Feeding
behavior data (chewing, feed disappearance, and water intake) were recorded via computer every
5 s, which allowed for calculation of meal size, intermeal interval, water intake, and eating,
ruminating, and total chewing times (Dado and Allen, 1993).
Analysis of samples Feed, 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 in a Wiley mill (1-mm
97

screen; Arthur H Thomas, Philadelphia, PA) and analyzed for ash, NDF, indigestible NDF, CP,
starch, and ether extract. Ash concentration was determined after 5 h of oxidation at 500°C.
Concentration of NDF was analyzed according to Mertens (2002). Indigestible NDF was used as
an internal marker to estimate total tract nutrient digestibility (Cochran et al., 1986). Indigestible
NDF was estimated as NDF residue after 240 h in vitro fermentation (Goering and Van Soest,
1970); flasks were re-inoculated at 120 h to ensure a viable microbial population. Rumen fluid
for the in vitro incubation was collected from a non-pregnant dry cow fed dry hay only. Crude
protein was determined according to Hach et al. (1987). Starch was analyzed using an enzymatic
method (Karkalas, 1985) after samples were gelatinized with sodium hydroxide. Glucose was
measured using a glucose oxidase method (PGO Enzyme product No. P7119, Sigma Chemical
Co., St. Louis, MO). Ether extract was determined using a modified Soxhlet apparatus (AOAC,
1990). All nutrients are expressed as percentages of DM determined by drying at 105°C in a
forced air oven for more than 8 h.
Plasma samples were analyzed using commercial kits for concentration of NEFA (NEFA
HR kit, Wako Chemicals USA, Richmond, VA), BHBA (kit #2240, Stanbio Laboratory, Boerne,
TX), insulin (Coat-A-Count, Siemens Healthcare Diagnostics, Deerfield, IL), and glucagon (kit
#GL-32K, Millipore, Billerica, MA). Plasma glucose concentration was analyzed using the
glucose oxidase method described above. Plasma and rumen fluid VFA concentrations were
determined by HPLC according to Oba and Allen (2003b). Liver acetyl CoA was analyzed by
HPLC using a method previously described (Stocks and Allen, 2012).
Statistical analysis Feeding behavior, intake, metabolite and hormone responses, and
rumen fermentation characteristics were analyzed using the Fit Model procedure of JMP (version
10.0.0, 2012, SAS Institure Inc., Cary, NC) with the following model:
98

Yijkl = µ + Si + C(Si)j + P k + Tl + PkTl + Cov + TlCov + eijkl
where Yijkl = the dependent response variable of interest, µ = overall mean, Si = fixed effect of
square (i = 1 to 2), C(Si)j = random effect of cow within square (j = 4), Pk = fixed effect of
period (k = 1 to 2), Tl = fixed effect of treatment (l = 1 to 4), PkTl = period x treatment, Cov =
effect of covariate, TlCov = treatment x covariate, and eijkl = residual, normally distributed.
Treatment effects were declared significant at P ≤ 0.05 and tendencies for treatment effects at P
≤ 0.10. Interactions were declared significant at P ≤ 0.10. Covariate interactions were removed
from the model if P > 0.20. All data is expressed as least squared means and standard error of
the means, unless otherwise specified.

99

RESULTS

Plasma metabolite and hormone and liver metabolite concentrations PI, compared with CO,
did not affect concentration of metabolites or hormones at the termination of infusion (P > 0.15;
Table 4.2). Intrajugular LI decreased plasma insulin concentration (P = 0.04; Table 4.2) and
increased plasma glucagon (P = 0.05) and liver acetyl CoA (P = 0.04) concentrations compared
with SI. LI did not affect plasma NEFA (P = 0.13; Table 4.2) or BHBA (P = 0.26; Table 4.2)
concentrations compared with SI. No interactions were detected between PI and LI for any
metabolite or hormone measured (P > 0.35; Table 4.2).
Differences between the post-infusion and covariate plasma and liver concentrations of
metabolites and hormones are reported in Table 4.3. LI increased plasma insulin concentration
(P = 0.04) and insulin:glucagon (P = 0.02) to a lesser extent than SI, while decreasing plasma
glucose (P = 0.04) concentration more than SI. PI suppressed the increase in plasma glucagon
compared with CO (P = 0.02). PI tended to decrease hepatic acetyl CoA concentration from the
covariate day to the end of the infusion compared with CO (-5.7 vs. -2.4 nmol/g wet tissue; P =
0.09). There was a tendency for an interaction between PI and LI for the change in plasma
NEFA concentration from the covariate day to the end of the infusion period (P = 0.12), and was
a result of the dramatic reduction in plasma NEFA by PI for the saline and propionic acid
treatment (data not shown). No other interactions between PI and LI for any change in plasma or
liver metabolites or plasma hormones were detected (P > 0.17).

100

Table 4.2. Plasma metabolites and hormones and hepatic acetyl CoA concentrations at the termination of infusion for cows post-peak
lactation.
Saline
Control

Propionate

1

Lipid
Control

Plasma Post Infusion
10.14
11.87
Insulin μIU/mL
7.82
173
78
NEFA μEq/L
204
5.26
4.01
BHBA mg/dL
8.15
117
113
Glucagon pg/mL
127
59.1
60.9
Glucose mg/dL
57.2
0.193
0.777
Insulin:glucagon
0.238
7.01
6.90
Post Liver Acetyl CoA nmol/g
8.47
1
Int (PI × LI) = interaction between propionic acid and lipid infusion

101

Propionate

SE

Propionate

P
Lipid

5.86
204
5.60
119
59.3
0.059
11.02

2.34
48.8
2.00
7.06
1.72
0.363
1.22

0.95
0.35
0.33
0.15
0.23
0.51
0.35

0.04
0.13
0.26
0.05
0.28
0.32
0.04

Int (PI
× LI)
0.35
0.35
0.74
0.64
0.92
0.39
0.31

Table 4.3. Changes in plasma metabolites and hormones and hepatic acetyl CoA from the covariate day to post-infusion for cows
post-peak lactation.
Saline
Control

Propionate

1

Lipid
Control

Plasma Post - Cov Infusion
5.38
8.01
Insulin μIU/mL
4.06
-14.62
-185.74
NEFA μEq/L
-103.32
2.96
2.51
5.31
BHBA mg/dL
7.12
-4.19
Glucagon pg/mL
15.92
-0.34
2.31
Glucose mg/dL
-2.34
Insulin:glucagon
0.0423
0.0742
0.0235
-1.59
-7.36
Post - Cov Acetyl CoA nmol/g
-3.16
1
Int (PI × LI) = interaction between propionic acid and lipid infusion

102

Propionate

SE

Propionate

P
Lipid

2.09
-51.79
2.84
2.88
-1.71
0.0168
-4.10

1.81
72.63
1.84
5.77
1.35
0.014
2.82

0.84
0.40
0.45
0.02
0.24
0.38
0.09

0.04
0.75
0.48
0.11
0.04
0.02
0.65

Int (PI × LI)
0.17
0.12
0.60
0.92
0.46
0.19
0.21

Milk production and feed intake Milk production and component yields on the covariate and
infusion days are reported in Table 4.4. Treatment infusions did not affect yields of milk or milk
components (P > 0.12). There was a tendency for an interaction between PI and LI for milk
protein and lactose yields (P = 0.14 and P = 0.11, respectively), which are a result of reduced
yields of both for the SI-PI treatment. No additional interactions were detected between PI and
LI on yields of milk or milk components (P > 0.29).
Feed intake and feeding behavior are reported in Table 4.5. Intraruminal PI decreased
daily DMI (P = 0.01), DMI over the 12 h infusion period (P = 0.02), and tended to decrease
intake over the first 4 h following feeding (P = 0.07), but did not affect DMI during the 12 h
following the termination of the infusion (P = 0.40) compared with CO. PI decreased water
intake (P = 0.01) compared with CO. Intrajugular LI increased meal frequency (P = 0.03),
reduced meal length (P = 0.05), and tended to reduce meal size (P = 0.06) and intermeal interval
(P = 0.08) compared with SI. There was a tendency for an interaction between LI and PI for
meal frequency and intermeal interval (P = 0.12 for both). The LI-PI treatment had the greatest
meal frequency and lowest intermeal interval while SI-PI treatment had the lowest meal bouts
and highest intermeal interval. There were no other interactions between LI and PI on any feed
intake or feeding behavior response measured (P > 0.30).

103

Table 4.4. Effects of treatment on milk yield and components for cows post-peak lactation.
Saline

1

Lipid

Control
Covariate Day
Milk yield kg
Milk fat yield kg
Milk protein yield kg
Milk lactose yield kg
3.5% FCM yield kg

Propionate

Control

Propionate

SE

Propionate

P
Lipid

33.1
1.37
1.07
1.59
36.4

29.6
1.31
0.97
1.43
34.0

33.7
1.36
1.12
1.61
36.4

30.5
1.36
0.99
1.46
35.2

5.29
0.22
0.13
0.27
5.71

0.28
0.87
0.25
0.29
0.60

0.79
0.89
0.69
0.86
0.85

0.96
0.99
0.89
1.0
0.86

31.3
1.21
1.01
1.51
33.1

4.28
0.17
0.11
0.20
4.34

0.36
0.19
0.16
0.12
0.16

0.84
0.67
0.26
0.27
0.81

0.29
0.85
0.14
0.11
0.53

Infusion Day
32.8
28.7
Milk yield kg
31.0
1.37
1.23
Milk fat yield kg
1.31
1.03
0.83
Milk protein yield kg
1.01
1.56
1.22
Milk lactose yield kg
1.50
36.3
32.3
3.5% FCM yield kg
34.6
1
Int (PI × LI) = interaction between propionic acid and lipid infusion

104

Int (PI × LI)

Table 4.5. Effects of treatment on feeding behavior for cows post-peak lactation.
Saline

1

Lipid

Control
Feeding behavior
DMI (kg/d)
DMI (kg 12 P to 12 A)
DMI (kg 12 P to 4 P)
DMI (kg 12 A to 12 P)
DMI 1st meal (kg)
Meal bouts (#/12 h)
Meal length (min/meal)
Meal size (kg/meal)
Intermeal interval (min)
Ruminating bouts (#/12 h)
Chew time (min/12 h)
2
ME intake (MJ/d)
Diet
3

Propionate

Control

Propionate

SE

Propionate

P
Lipid

28.5
20.0
12.2
8.3
6.42
5.51
42.0
3.98
104.7
5.77
364.9

23.9
16.3
10.2
7.8
6.81
4.26
44.0
4.23
147.8
5.08
342.8

25.9
17.9
10.9
8.1
5.81
5.89
37.7
3.35
100.4
5.96
402.2

23.4
15.9
10.1
7.4
4.97
6.05
30.8
2.74
93.8
6.37
360.5

1.25
1.45
1.48
0.75
1.35
0.47
5.46
0.66
17.0
0.82
24.6

0.01
0.02
0.07
0.40
0.79
0.23
0.56
0.72
0.25
0.84
0.13

0.24
0.30
0.35
0.68
0.16
0.03
0.05
0.06
0.08
0.28
0.17

0.41
0.44
0.41
0.93
0.46
0.12
0.30
0.42
0.12
0.43
0.63

184
0.17

153
-0.61

172

148

14.7

34.3
9.22
195
42.3

1.51
< 0.001
15.0
6.08

0.05
0.14

0.51
< 0.001

0.79
0.14

< 0.001
0.28
0.01

0.46
0.04
0.87

0.46
0.86
0.93

Lipid
39.5
0
9.22
Propionate
0
178
164
Total
213
Water Intake (L/12 h)
53.3
42.0
54.3
1
Int (P × L) = interaction between propionic acid and lipid infusion.
2

Int (PI ×
LI)

Metabolizable energy intake from the diet was calculated according to Oba and Allen (2003a)

3

Metabolizalbe energy intake from the propionate infusion was based on energy density of 1.536 MJ/mol for propionate (Oba and
Allen, 2003a), respectively, and was the same for all cows. ME from lipid was 4.2 MJ/h of lipid infusion during the feeding behavior
period.

105

Ruminal pH and concentration and profile of VFA Rumen pH and concentration and profile of
VFA are reported in Table 4.6. Ruminal pH was not affected by treatment infusions, and there
was no interaction between PI and LI on rumen pH. Intraruminal PI increased both the
concentration and percent of propionic acid (P < 0.001 for both) and decreased both the
concentration of acetic (P = 0.002) and butyric (P = 0.05) acid and the percent of acetic (P <
0.001) and butyric (P = 0.01) acid in the rumen. Intrajugular LI tended to increase the total VFA
(P = 0.06), and increased acetic acid (P = 0.04), butyric acid (P = 0.02), and iso-valeric acid (P
= 0.03) concentrations in the rumen. LI also increased butyric acid (P = 0.04) and decreased
propionic acid (P = 0.05) as a percent of total VFA. There were no interactions between rumen
PI and LI on rumen VFA concentration (P > 0.29), but there was an interaction between PI and
LI percent iso-butyric acid (P = 0.08) in the rumen.

106

Table 4.6. Effects of treatment on rumen fermentation for cows post-peak lactation.
Saline
Control

Propionate

1

Lipid
Control

Propionate

Rumen VFA (mM)
144.5
133.5
Total
158.7
147.3
92.6
75.7
Acetic acid
100.8
85.6
18.3
16.1
Butyric acid
22.6
18.3
1.23
1.16
Iso-butyric acid
1.44
1.22
1.95
1.87
Iso-valeric acid
2.37
2.11
28.3
36.3
Propionic acid
28.7
37.4
Rumen VFA (%)
64.1
56.6
Acetic acid
63.6
58.4
12.6
11.90
Butyric acid
14.2
12.2
0.81
0.89
Iso-butyric acid
0.90
0.82
1.26
1.37
1.47
1.41
Iso-valeric acid
19.6
27.5
Propionic acid
18.1
25.4
5.78
6.01
Rumen pH
5.90
6.00
1
Int (PI × LI)= interaction between propionic acid and lipid infusion

107

SE

Propionate

P
Lipid

7.74
4.28
1.87
0.098
0.227
2.14

0.12
0.001
0.02
0.14
0.32
0.0001

0.06
0.04
0.02
0.16
0.03
0.62

0.97
0.83
0.38
0.43
0.35
0.83

1.06
0.79
0.063
0.118
0.95
0.16

<0.0001
0.002
0.91
0.74
<0.0001
0.30

0.39
0.02
0.87
0.15
0.02
0.75

0.11
0.10
0.14
0.30
0.69
0.68

Int (PI × LI)

DISCUSSION

The objective of this experiment was to infuse lipid to simulate the metabolic profile of a
cow in early lactation by increasing NEFA supply to, and oxidation in, the liver, as previously
observed (Mahek et al., 2005; Pires et al., 2007). In the present study, intrajugular infusion of
Intralipid® failed to achieve the desired results. Previous reports using lipid infusions at similar
rates to those in the current experiment have resulted in metabolic profiles that are consistent
with our expected response of increased plasma NEFA concentration. Pires et al. (2007) infused
a 20% tallow emulsion at a rate of ~72 g TG/h in non-lactating, non-gestating Holstein cows and
reported a 3.7 fold increase in plasma NEFA concentration (P < 0.001) within 2 h, reaching a
peak by 8 h of infusion. These researchers also reported an increase in plasma glucose and
serum insulin concentrations within 8 h of infusion. Mashek et al. (2005) infused 16 g TG/h of
20% lipid emulsions (based on either tallow, linseed, or fish oil) into non-lactating, nongestating, fasted Holstein cows and reported that regardless of lipid source, plasma NEFA was
increased, with tallow increasing plasma NEFA to the greatest extent. Chelikani et al. (2003)
infused lipid at a rate of ~69 g TG/h for 6 h into cows in late lactation and reported an eight-fold
increase in plasma NEFA concentration within the first 3 h of infusion with no difference in
plasma glucose or insulin concentrations. Bareille and Faverdin (1996) infused ~131 g TG/h for
4 h into cows ~60 and 92 DIM and reported increases in plasma NEFA, glucose, and insulin
concentrations. The variation plasma hormone and metabolite responses could be the result of
differences in amount of triglyceride infused or stage of lactation between the studies.
Despite an infusion rate that was within the range of TG infused in previous reports (92.6
g TG/h), LI infusion did not result in a significant increase in plasma NEFA concentration,
108

however it was numerically higher (Table 4.2). This could be a result of increased storage of
lipid, increased oxidation of lipid, or transfer of lipid to milk fat. In the current experiment, LI
was infused at half the final rate for the first 12 h of infusion (46.3 g TG/h) to allow for
adaptation to avoid adverse reactions, such as increased body temperature or respiration rate.
During this time, LI increased plasma NEFA concentration from 281.5 μEq/L on the covariate
day to 384.5 μEq/L at the start of feeding behavior. In the experiments previously mentioned,
plasma NEFA concentration was elevated within the first 3 – 8 h of initiating the lipid infusion.
It is possible that there were metabolic adaptations to LI infusion during the initial 12 h
triglyceride infusion that resulted in less dramatic responses to LI during the 12 h feeding
behavior monitoring period. The slight increase in hepatic acetyl CoA concentration for LI
suggests that hepatic lipid oxidation was increased to a limited extent.
While we did not observe an increase in milk fat yield (Table 4.4), we did not measure
milk FA profile, and it is possible that there was an increase in long-chain FA yield at the
expense of de novo FA (< C16) yield. Chelikani et al. (2003) reported that lipid infusion did not
increase milk fat yield in late lactation, consistent with our results. Barielle and Faverdin (1996)
reported that lipid infusion tended to increase milk fat yield, but also resulted in a 60% increase
in C18:1 and C18:2 percent while decreasing the percent of FA < C16. The shift in FA profile
observed by Barielle and Faverdin (1996) may explain our lack of response in milk fat yield.
The lack of treatment effects of PI infusion on plasma and liver metabolite and plasma
hormone concentrations (Table 4.2) is inconsistent with results of previous experiments using
similar infusion rates. Intraruminal propionate infusions have increased plasma glucose (Oba
and Allen, 2003a, c; Sheperd and Combs, 1998) and insulin (Oba and Allen, 2003a, c)
concentrations, and decreased plasma BHBA (Oba and Allen, 2003a; Sheperd and Combs, 1998)
109

and NEFA (Oba and Allen, 2003a) concentrations. However, Sheperd and Combs (1998) did
not detect an effect of propionate infusion on plasma insulin or NEFA concentrations, consistent
with the results of the current experiment. Cows in these experiments were all in mid- to latelactation, and differences in plasma hormone and metabolite responses could be a result of
differences in DMI, milk production, or lipolytic state between the experiments.
There are limited data reporting feeding behavior for cows fed fat or infused with lipid.
Heinrichs et al. (1982) reported that feeding a higher fat concentrate mix resulted in decreased
meal size at the conditioned meal (first meal following feeding) and that meal length of the
conditioned meal was decreased. The researchers speculated that this might have been because
of reduced palatability or because satiety was reached sooner with the high fat feed. However,
higher dietary fat diets have been shown to stimulate gut peptides, which may lead to decreased
DMI (Choi et al., 2000; Relling and Reynolds, 2007; Bradford et al., 2008). While not
significant, in the current experiment LI numerically reduced the size of the first meal by 1.23 kg
(5.39 vs. 6.62 kg), and tended to reduce meal size overall (Table 5), consistent with the reduction
in the size of the conditioned meal reported by Heinrichs et al. (1982). Effects of LI on meal size
in the current experiment might have been a result of increased oxidation of FA from LI, which
would generate ATP and stimulate satiety (Allen, 2000), as intravenous LI would not stimulate
gut peptides or have an effect on palatability of the diet. While other researchers have reported
reduced DMI during intravenous lipid infusion, feeding behavior has not been previously
reported. Chelikani et al. (2003) infused a lipid emulsion at ~69 g TG/h for 6 h resulting in a
14% reduction in DMI. Bareille and Faverdin (1996) reported that infusion of a 20% lipid
emulsion at ~104 g TG/h at the start of the main morning meal resulted in a 5% reduction in
DMI over the 4 h infusion period.
110

The hypophagic effect of PI in this experiment (Table 4.5) is consistent with previous
research utilizing intraruminal propionate infusions. Sheperd and Combs (1998) infused
propionate intraruminally at 14.1 mmol/min and reported a 5.6% reduction in DMI compared
with an iso-energetic acetate infusion in mid-lactation cows. In a dose response study in our
laboratory using cows averaging 192 DIM, propionate infusion at a rate similar to that of the
current experiment (8.7 mmol/min) decreased DMI 18.5%, with an increased intermeal interval
and decreased meal size compared with iso-osmotic acetate infusion (Oba and Allen, 2003a). PI
decreased DMI 15% in the present experiment with cows ranging from 81 – 252 DIM and a
propionic acid infusion rate of 8.3 mmol/min. Previous research in our laboratory (Oba and
Allen, 2003c) showed that propionate infusion reduced DMI 7.4% for cows in mid-lactation, and
the marginal response in DMI reduction was positively related to plasma glucose concentration.
We speculated that the reduction in intake was consistent with increased propionate oxidation
because gluconeogenic flux was likely decreased as plasma glucose concentration increased. In
the current experiment, glucose concentration was high (Table 4.2), which may have resulted in
increased hepatic oxidation of propionate, resulting in a reduction in DMI consistent with our
previous results (Oba and Allen, 2003c).
We have previously reported that intraruminal propionate infusion is hypophagic in cows
in a lipolytic state (Oba and Allen, 2003a; Stocks and Allen, 2012; Stocks and Allen, accepted).
This is likely a result of propionate stimulating oxidation of hepatic acetyl CoA, resulting in a
reduction of the concentration of hepatic acetyl CoA as previously observed (Stocks and Allen,
accepted). In the current experiment, LI increased the concentration of hepatic acetyl CoA at the
termination of the infusion (Table 4.2), and PI only tended (P = 0.09) to decrease hepatic acetyl
CoA concentration compared with CO over the course of the infusion (Table 4.3). This is
111

inconsistent with our previous results in early lactation (Stocks and Allen, accepted) and might
explain the lack of an interaction between LI and PI for most measured responses, particularly
for DMI.

112

CONCLUSIONS

Experimentally increasing plasma NEFA concentration through intravenous lipid
infusion to simulate the metabolic state of an early lactation cow was expected to allow for
investigation of cause and effect between lipolytic state and propionate infusion on DMI.
However, this model failed to mimic a lipolytic state previously reported by others. Additional
research investigating the interaction between propionate and lipid infusion, as well as between
diet fermentability and lipolytic state, are necessary for determining feeding strategies for cows
in the immediate postpartum period.

113

REFERENCES

114

REFERENCES

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

AOAC. 1990. Official Methods of Analysis. 15 ed. Association of Analytical Chemists.
Gaithersburg, MD.
th

AOAC. 1997. Official Methods of Analysis. 16 ed. Association of Analytical Chemists.
Gaithersburg, MD.
Bareille, N., and P. Faverdin. 1996. Lipid metabolism and intake behavior of dairy cows: effects
of intravenous lipid and β-adrenergic supplementation. J. Dairy Sci. 79:1209-1220.
Bell, A. W. 1995. Regulation of organic nutrient metabolism during transition from late
pregnancy to early lactation. J. Anim. Sci. 73:2804-2819.
Bell, A. W., and D. E. Bauman. 1997. Adaptations of glucose metabolism during pregnancy and
lactation. J. Mammary Gland Biol. Neoplasia. 2:265-78.
Bradford, B. J., and M. S. Allen. 2005. Phlorizin administration increases hepatic gluconeogenic
enzyme mRNA abundance but not feed intake in late-lactation dairy cows. J. Nutr.
35:2206-2211.
Bradford, B. J., A. D. Gour, A. S. Nash, and M. S. Allen. 2006. Propionic acid challenge test
have limited value for investigating bovine metabolism. J. Nutr. 136:1915-1920.
Bradford, B. J., and M. S., Allen. 2007. Phlorizin administration does not attenuate hypophagia
induced by intraruminal propionate infusion in lactating dairy cattle. J. Nutr. 137:326-330.
Bradford, B. J., K. J. Harvatine, and M. S. Allen. 2008. Dietary unsaturated fatty acids increase
plasma glucagon-like peptide-1 and cholecystokinin and may decrease premeal ghrelin in
lactating dairy cows. J. Dairy Sci. 91:1443–1450.
Chelikani, P. K., D. H. Keisler, and J. J. Kennelly. 2003. Response of plasma leptin
concentration to jugular infusion of glucose or lipid is dependent on the stage of lactation
of Holstein cows. J. Nutr. 133:4163-4171.
Choi, B. R., D. L. Palmquist, and M. S. Allen. 2000. Cholecystokinin mediates depression of
feed intake in dairy cattle fed high fat diets. Domest. Anim. Endocrinol. 19:159–175.
Cochran, R. C., D. C. Adams, J. D. Wallace, and M. L. Galyean. 1986. Predicting the
digestibility of different diets with internal markers: evaluation of four potential markers. J.
Anim. Sci. 63:1476-1483.
115

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.
Davis, C. L. 1967. Acetate production in the rumen of cows fed either control or low-fiber, highgrain diets. J. Dairy Sci. 50:1621–1625.
Doepel, L., H. Lapierre, and J. J. Kennelly. 2002. Peripartum performance and metabolism of
dairy cows in response to prepartum energy and protein intake. J. Dairy Sci. 85:2315-2334.
Firkins, J. L., M. L. Eastridge, N. R. St-Pierre, and S. M. Noftsger. 2001. Effects of grain
variability and processing on starch utilization by lactating dairy cattle. J. Anim. Sci.
79:E218-E238.
Goering, H. K., and P. J. Van Soest. 1970. Forage Fiber Analysis (Apparatus, Reagents,
Procedures, and Some Applications). Agricultural Handbook no. 379. ARS-USDA,
Washington, DC.
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.
Heinrichs, A. J., D. L. Palmquist, and H. R. Conrad. 1982. Feed intake patterns of cows fed high
fat grain mixtures. J. Dairy Sci. 65:1325-1328.
Karkalas, J. 1985. An improved enzymatic method for the determination of native and modified
starch. J. Sci. Food Agric. 36:1019–1027.
Mashek, D. G., S. J. Bertics, and R. R. Grummer. 2005. Effects of intravenous triacylglycerol
emulsion on hepatic metabolism and blood metabolites in fasted cows. J. Dairy. Sci.
88:100-109.
Mertens, D. R. 2002. Gravimetric determination of amylase-treated neutral detergent fiber in
feeds using refluxing in beakers or crucibles: collaborative study. J. AOAC Int. 85:1217–
1240.
th

NRC. 2001. Nutrient requirements of dairy cattle. 7 rev. ed. Washington: National Academy of
Science.
Oba, M., and M. S. Allen. 2003a. Dose-response effects of intraruminal infusion of propionic
acid on feeding behavior of lactating cows in early or midlactation. J. Dairy Sci. 86:29222931.
Oba, M., and M. S. Allen. 2003b. Effects of corn grain conservation method on feeding behavior
and productivity of lactating dairy cows at two dietary starch concentrations. J. Dairy Sci.
86:174-183.
116

Oba, M., and M. S. Allen. 2003c. Extent of hypophagia caused by propionate infusion is related
to plasma glucose concentration in lactating dairy cows. J. Nutr. 1105-1112.
Pires, J. A. A., A. H. Souza, and R. R. Grummer. 2007. Induction of hyperlipidemia by
intravenous infusion of tallow emulsion causes insulin resistance in Holstein cows. J. Dairy
Sci. 90:2735-2744.
Relling, A. E., and C. K. Reynolds. 2007. Feeding rumen-inert fats differing in their degree of
saturation decreases intake and increases plasma concentrations of gut peptides in lactating
dairy cows. J. Dairy Sci. 90:1506–1515.
Reynolds, C. K., P. C. Aikman, B. Lupoli, D. J. Humphries, and D. E. Beever. 2003. Splanchnic
metabolism of dairy cows during the transition from late gestation through early lactation.
J. Dairy Sci. 86:1201-1217.
Roberts, T., N. Chapinal, S. J. LeBlanc, D. F. Kelton, J. Dubuc, and T. F. Duffield. 2012.
Metabolic parameters in transition cows as indicators for early lactation culling. J. Dairy
Sci. 95:3057-3063.
Seifi, H. A, S. J. LeBlanc, K. E. Leslie, and T. F. Duffield. 2011. Metabolic predictors of postpartum disease and culling risk in dairy cattle. The Veterinary Journal. 188:216-220.
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.
Stocks, S. E., and M. S. Allen. 2012. Hypophagic effects of propionic acid increase with elevated
hepatic acetyl coenzyme A concentration for cows in the early postpartum period. J. Dairy
Sci. 95:3259-3268.
Stocks, S. E., and M. S. Allen. 2013. Hypophagic effects of propionic acid are not attenuated
during a three-day infusion in the early postpartum period in Holstein cows. J. Dairy Sci.
Accepted.
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.

117

CHAPTER 5

EVALUATION OF THE EFFECT OF MESENTERIC PROPIONATE OR GLYCEROL
INFUSION ON FEED INTAKE IN EARLY LACTATION DAIRY COWS: A
PRELIMINARY STUDY

INTRODUCTION
Energy intake of dairy cows during the late prepartum and early postpartum period is
suppressed, resulting in negative energy balance (Gulay et al., 2004; Hayirli et al., 2002).
Plasma NEFA concentration is elevated prepartum as a result of decreased plasma insulin
concentration and reduced adipose tissue sensitivity to insulin, which spares glucose for the
developing fetus (Bell, 1995). Low plasma insulin concentration and insulin resistance persists
in the early postpartum period increasing the risk of hepatic lipidosis, ketosis, and suppressed
feed intake.
Allen (2000) proposed that suppression of energy intake for cows in the periparturient
period is caused by hepatic oxidation of NEFA and that highly fermentable diets may exacerbate
intake depression. Oba and Allen (2003a) demonstrated that highly fermentable starch sources
could decrease DMI by decreasing meal size (an indication of satiety) despite a numerical
decrease in the intermeal interval. Additionally, intraruminal propionate infusion, relative to isoosmotic intraruminal acetate infusion, decreased energy intake in early lactation cows by
decreasing meal size but not intermeal interval (Oba and Allen, 2003b; Stocks and Allen, 2012;
Stocks and Allen, accepted). Cows in a greater lipolytic state, such as those in the immediate
postpartum period, may experience a greater reduction in intake due to increased oxidation of
118

acetyl CoA in the TCA cycle caused by propionate, stimulating a satiety signal and terminating
the meal (Stocks and Allen, 2012).
In early lactation, uptake of glycerol by the liver, and subsequent use of glycerol as a
precursor for glucose synthesis, is high relative to late lactation or prepartum (Reynolds et al.,
2003). However, the fate of glycerol metabolism depends on the delivery method. Glycerol is
rapidly fermented in the rumen when added to feed (Rémond et al., 1993), but it is likely that
glycerol can escape fermentation if it is administered as an oral drench. Linke et al. (2004)
reported that oral glycerol drenching increased plasma glucose concentration while feeding
glycerol did not. Goff and Horst (2001) also reported that oral drenches of glycerol were
effective at increasing plasma glucose concentration. Glycerol that escapes ruminal fermentation
is converted to dihydroxyacetone phosphate, bypassing a rate-limiting step for gluconeogenesis
and stimulating glucose production. Glycerol can also enter the TCA cycle and stimulate
oxidation of fuels, but this is unlikely in early lactation due to the increased glucose demand
(Bell, 1995).
We have demonstrated previously that intraruminal propionate infusion (relative to isoosmotic acetate infusion) decreased DMI and that hypophagia was greater when hepatic acetyl
CoA concentration was highest (Stocks and Allen, 2012; Stocks and Allen, accepted). In early
lactation, it is likely that glycerol is converted to glucose without stimulating oxidation of acetyl
CoA in the TCA cycle, and therefore is not expected to decrease DMI by stimulation of hepatic
oxidation. The objectives of this preliminary experiment were 1) to evaluate the surgical
mesenteric catheterization procedure and subsequent cow recovery period in the days prior to
parturition and 2) to determine the effect of iso-energetic glycerol and propionate infusion on

119

energy intake in the immediate postpartum period. We hypothesized that propionate will be more
hypophagic than glycerol when infused on an equal energy basis.

120

MATERIALS AND METHODS
Animals, housing, and diets The Institutional Animal Care and Use Committee at
Michigan State University approved all experimental procedures for this experiment. Two nonlactating, pregnant Holstein cows (#4575 and #4499) had a mesenteric vein catheter installed
according to the method of Huntington et al. (1989) 10-14 days prior to expected calving. The
catheters were polypropylene (0.24 cm o.d. × 0.17 cm i.d. tubing, MRE 095; Braintree Scientific,
Braintree, MA) and were approximately 182 cm long. Following catheterization, cow health and
the surgical site were monitored daily. Catheter patency was also checked daily. Cows were
housed in individual tie-stalls for the duration of the experiment. Cows were fed a common dry
cow diet prior to calving and a common early lactation diet postpartum once daily at 115% of
expected intake. The dry cow diet contained corn silage, grass hay, ground corn, soybean meal,
SoyChlor® (West Central Cooperative, Ralston, IA), and a vitamin and mineral mix and was
formulated to meet NRC requirements (2001). The early lactation cow diet is reported in Table
5.1.

121

Table 5.1. Ingredients and nutrient composition of experimental diet (% of dietary DM except
for DM). Estimated diet composition and nutrient composition.
Diet ingredient
Corn silage
Alfalfa silage
Ground corn
Soybean meal
1
Vitamin and mineral mix
Limestone
Sodium bicarbonate
Nutrient composition
DM
OM
Starch
NDF
CP
Ether extract

%
60.3
23.6
1.9
6.9
0.93
0.45
0.49
43.3
91.2
23.0
33.0
16.5
2.94

1

Vitamin and mineral mix contained 34.1% ground corn grain, 25.6% salt, 21.8% calcium
carbonate, 9.1% Biophos (21% P), 3.9% magnesium oxide, 2% soybean oil, 1% vitamin A
(30,000 IU/gm), 0.63% manganese sulfate (31%), 0.56% zinc sulfate (35.5% zinc), 0.48%
ferrous sulfate (30%), 0.30% vitamin E (50% absorbate), 0.20% copper sulfate (25.2%), 0.15%
selenium (1%), 0.14% vitamin D3 (16 million IU/lb), 0.02% iodine (50 gm/lb), and 0.01% cobalt
carbonate (46%).
Experimental design and treatments The experiment was a switchback design with three
2 d infusion periods and no rest days between infusion periods. The experiment started at 2 DIM
and each cow was assigned one of two treatment sequences. Cow 4575 started the experiment
on May 24, 2012 and completed the experiment on May 30, 2012. Cow 4499 started the
experiment on May 25, 2012 and completed the experiment on May 31, 2012. The treatments
were iso-energetic mesenteric vein infusions of propionic acid (1 mol/L) or glycerol (1.15 mol/L)
continuously infused at 500 mL/h using Baxter Flo-Gard 6201 pumps (Baxter, Deerfield, IL)
providing 0.768 MJ/h. The gross energy of the glycerol (Glycerin, 99.7% USP Kosher;
ChemWorld, Roswell, GA) was ~18.1 MJ/kg (Kerr et al., 2009) and the gross energy of the
propionic acid (KemProp, 99.5%, Kemin Industries, Des Moines, IA) was ~20.8 MJ/kg (Oba and
122

Allen, 2003b). Treatment sequences were propionic acid-glycerol-propionic acid for cow 4575
and glycerol-propionic acid-glycerol for cow 4499.
Data and sample collection Cows were blocked from feed from 1100 h to 1200 h to
weigh orts and offer feed. Feed intake was calculated daily from feed offered and orts. Cows
were milked twice per day in the milking parlor at 0600 h and 1700 h and milk weights were
recorded. Infusions were paused twice each day for approximately 35 minutes at each milking
time to allow the cows to go to the parlor for milking. Temperature, pulse, and respiration were
recorded at least 4 times for the first 6 h of infusion on the first day of the experiment and then
once every ~6 hours during the remainder of the infusion.

123

RESULTS AND DISCUSSION

The surgical procedure for mesenteric catheter insertion was evaluated to determine if
there were any deleterious effects of catheter insertion surgery within 2 wk of expected calving.
Mesenteric catheter insertion within 2 wk of expected calving did not result in any adverse
effects on the cow, calving, or calf health.
Treatment infusions had no adverse effects on body temperature, pulse, or respiration
(Table 5.2). Milk production is reported in Table 5.3 and was similar for both treatments. Feed
intake is reported in Table 5.3 and Figure 5.1. Propionic acid infusion reduced feed intake by
10.8% compared to glycerol infusion. The reduction in feed intake with propionic acid infusion
is consistent with previous experiments with both mesenteric and ruminal propionic acid
infusion. Previous experiments in our laboratory have reported that intraruminal infusion of
propionic acid during the early postpartum period decreased DMI by up to a 20% (Oba and
Allen, 2003b; Stocks and Allen, 2012; Stocks and Allen accepted). Additionally, Elliot et al.
(1985) reported that mesenteric infusion of propionate in steers reduced intake while acetate
infusion did not. Glycerol was not expected to affect intake as it would likely have been used as
a glucogenic precursor (Reynolds et al., 2003) and would likely not have been oxidized in the
TCA cycle due to high glucose demands postpartum.

124

Table 5.2. Temperature, pulse, and respiration recordings for cows 4575 and 4499 collected
during the infusion experiment.

Cow
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575
4575

Date
5/24/12
5/24/12
5/24/12
5/24/12
5/24/12
5/24/12
5/24/12
5/24/12
5/25/12
5/25/12
5/25/12
5/25/12
5/25/12
5/25/12
5/25/12
5/25/12
5/25/12
5/25/12
5/26/12
5/26/12
5/26/12
5/26/12
5/26/12
5/26/12
5/27/12
5/27/12
5/27/12
5/27/12
5/27/12
5/28/12
5/28/12
5/28/12
5/28/12
5/29/12
5/29/12
5/29/12
5/29/12
5/30/12

Time
11:00 AM
1:16 PM
2:10 PM
3:17 PM
4:30 PM
5:30 PM
7:10 PM
9:40 PM
12:00 AM
5:30 AM
7:35 AM
10:50 AM
12:55 PM
2:00 PM
3:00 PM
4:15 PM
5:50 PM
9:00 PM
12:00 AM
5:30 AM
11:00 AM
12:00 PM
4:30 PM
11:00 PM
5:30 AM
10:50 AM
12:40 PM
4:30 PM
11:00 PM
6:00 AM
10:50 AM
4:30 PM
11:45 PM
5:45 AM
11:00 AM
5:05 PM
11:00 PM
7:00 AM

Temperature
(°F)
101.5
101.7
101.7
102.3
101.7
102.0
101.6
101.8
101.7
101.3
101.8
101.2
100.6
101.5
101.2
101.3
101.3
101.4
101.4
101.0
101.4
100.3
101.8
101.1
101.2
100.9
100.8
101.9
101.5
101.3
101.4
102.3
102.2
101.0
101.2
101.9
101.4
101.5
125

Pulse
(Beats/min)
80
80
80
80
84
84
80
NR1
80
72
80
80
84
72
92
84
80
80
NR
72
76
80
88
84
80
88
84
88
88
84
76
84
80
76
72
88
80
75

Respiration
(Breaths/min)
44
44
44
44
48
40
40
NR
48
36
44
44
36
36
44
40
48
40
40
36
36
36
36
36
28
44
32
44
32
32
36
48
36
40
36
36
32
21

Table 5.2 (cont’d)
Cow
4575
4575
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499
4499

Date
5/30/12
5/30/12
5/24/12
5/24/12
5/24/12
5/24/12
5/24/12
5/24/12
5/25/12
5/25/12
5/25/12
5/25/12
5/25/12
5/25/12
5/25/12
5/25/12
5/26/12
5/26/12
5/26/12
5/26/12
5/26/12
5/26/12
5/27/12
5/27/12
5/27/12
5/27/12
5/27/12
5/28/12
5/28/12
5/28/12
5/28/12
5/29/12
5/29/12
5/29/12
5/29/12
5/30/12
5/30/12
5/30/12
5/30/12
5/31/12

Time
11:18 AM
5:24 PM
11:00 AM
2:20 PM
4:30 PM
5:30 PM
7:10 PM
9:40 PM
5:30 AM
10:50 AM
12:55 PM
2:00 PM
3:00 PM
4:15 PM
5:50 PM
9:00 PM
12:00 AM
5:30 AM
11:00 AM
12:00 PM
4:30 PM
11:00 PM
5:30 AM
10:50 AM
12:40 PM
4:30 PM
11:00 PM
6:00 AM
10:50 AM
4:30 PM
11:45 PM
5:45 AM
11:00 AM
5:00 PM
11:00 PM
7:00 AM
11:19 AM
4:37 PM
11:37 PM
6:30 AM

Temperature
(°F)
101.7
102.8
101.4
101.8
102.2
102.7
102.8
103.1
101.3
101.0
101.2
101.5
101.4
101.3
101.8
101.8
101.3
100.5
100.8
101.2
101.4
101.0
101.0
100.5
101.2
101.0
101.3
101.0
101.0
101.4
101.1
100.5
100.4
102.0
101.1
100.6
101.1
101.2
101.7
100.6
126

Pulse
(Beats/min)
76
80
92
100
92
88
88
NR
88
88
92
80
80
88
NR
80
80
80
NR
80
84
84
84
84
76
80
88
84
88
92
88
76
76
92
80
87
92
84
96
87

Respiration
(Breaths/min)
32
28
48
56
48
48
48
44
40
36
40
48
48
40
48
40
40
36
36
40
36
40
36
40
36
36
36
36
44
48
48
32
32
48
36
27
48
36
48
27

Table 5.2 (cont’d)
Temperature
Pulse
Respiration
Cow
Date
Time
(°F)
(Beats/min) (Breaths/min)
4499 5/31/12 11:09 AM
101.1
88
32
1
NR = Not recorded to avoid disturbing the cow if she was laying down.

Table 5.3. Daily feed intake and milk production for cow 4575 and 4499. Prop = propionic acid
infusion; Glyc = glycerol infusion.
Cow
4575
4575
4575
4575
4575
4575
4499
4499
4499
4499
4499
4499

Date
5/24/12
5/25/12
5/26/12
5/27/12
5/28/12
5/29/12
5/25/12
5/26/12
5/27/12
5/28/12
5/29/12
5/30/12

Treatment
Prop
Prop
Glyc
Glyc
Prop
Prop
Glyc
Glyc
Prop
Prop
Glyc
Glyc

Feed intake, kg
28.1
24.5
31.3
33.6
30.8
30.8
25.4
32.7
27.2
24.5
29.0
29.5

127

Milk, kg
39.1
42.8
45.1
43.1
47.9
48.4
38.4
39.2
39.4
42.3
44.6
46.6

Figure 5.1. Feed intake (as fed) for cows receiving propionic acid or glycerol infusion in a
switchback design experiment in early lactation. The ration DM was ~43%. Panel A = Cow
4575; panel B = Cow 4499. PROP = propionic acid, GLYC = glycerol.
A.

B.

128

CONCLUSIONS
Both the catheterization surgery and recovery period proceeded without any negative
consequences, and cows calved with normal, healthy calves. In addition, no negative health
effects were observed as a result of the infusion. Propionic acid infusion decreased feed intake
(as fed) compared with glycerol consistent with our hypothesis. Additional research using
mesenteric propionic acid and glycerol infusions will be necessary to evaluate treatment
responses and to make interpretations about the role of propionic acid in the control of feed
intake in early lactation.

129

REFERENCES

130

REFERENCES

Allen, M. S. 2000. Effects of diet on short-term regulation of feed intake by lactating dairy cattle.
J. Dairy Sci. 83:1598-1624.
Bell, A. W. 1995. Regulation of organic nutrient metabolism during transition from late
pregnancy to early lactation. J. Anim. Sci. 73:2804-2819.
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. 68:1165-1170.
Goff, J.P., and R.L. Horst. 2001. Oral glycerol as an aid in the treatment of ketosis/fatty liver
complex. J. Dairy Sci. 84(Suppl. 1):153.(Abstr.).
Gulay, M. S., M. J. Hayen, M. Liboni, T. I. Belloso, C. J. Wilcox, and H. H. Head. 2004. Low
doses of bovine somatotropin during the transition period and early lactation improves milk
yield, efficiency of production, and other physiological responses of Holstein cows. J.
Dairy Sci. 87:948-960.
Hayirli, A., R. R. Grummer, E. V. Nordheim, and P. M Crump. 2002. Animal and dietary factors
affecting feed intake during the prefresh transition period in Holsteins. J. Dairy Sci.
85:3430-3443.
Huntington, G. B., C. K. Reynolds, and B. H. Stroud. 1989. Techniques for measuring blood
flow in splanchnic tissues of cattle. J. Dairy Sci. 72:1583-1595.
Kerr, B. J., T. E. Weber, W. A. Dozier III, and M. T. Kidd. 2009. Digestible and metabolizable
energy content of crude glycerin originating from different sources in nursery pigs. J.
Anim. Sci. 87:4042-4049.
Linke, P.L., J.M. DeFrain, A.R. Hippen, and P.W. Jardon. 2004. Ruminal and plasma responses
in dairy cows to drenching or feeding glycerol. J. Dairy Sci. 87(Suppl. 1):343. (Abstr.).
th

NRC. 2001. Nutrient requirements of dairy cattle. 7 rev. ed. Washington: National Academy of
Science.
Oba, M., and M. S. Allen. 2003a. Effects of corn grain conservation method on feeding behavior
and productivity of lactating dairy cows at two dietary starch concentrations. J. Dairy Sci.
86:174-183.
Oba, M., and M. S. Allen. 2003b. Dose-response effects of intraruminal infusion of propionate
on feeding behavior of lactating cows in early or midlactation. J. Dairy Sci. 86:2922-2931.

131

Rémond, B., E. Souday, and J. P. Jouany. 1993. In vitro and in vivo fermentation of glycerol be
rumen microbes. Anim. Feed Sci. Technol. 41:121-132.
Reynolds, C. K., P. C. Aikman, B. Lupoli, D. J. Humphries, and D. E. Beever. 2003. Splanchnic
metabolism of dairy cows during the transition from late gestation through early lactation.
J. Dairy Sci. 86:1201-1217.
Stocks, S. E., and M. S. Allen. 2012. Hypophagic effects of propionic acid increase with elevated
hepatic acetyl coenzyme A concentration for cows in the early postpartum period. J. Dairy
Sci. 95:3259-3268.
Stocks, S. E., and M. S. Allen. 2013. Hypophagic effects of propionic acid are not attenuated
during a three-day infusion in the early postpartum period in Holstein cows. J. Dairy Sci.
Accepted.

132

CHAPTER 6
PRACTICAL IMPLICATIONS AND FUTURE RESEARCH

The research presented in this dissertation provides support for the interaction between
propionate and lipolytic state in the control of feed intake in dairy cows. Propionate infusion
consistently reduced DMI, and was increasingly hypophagic in cows with elevated hepatic acetyl
CoA concentration. We showed that propionate was hypophagic for dairy cows across a wide
range of days in milk during experiment 1, and over longer-term infusions during experiment 2.
However, there are several questions that remain unanswered and are essential in reaching a
more thorough understanding of the mechanisms involved in the control of feed intake in early
lactation.
In the 3 d infusion experiment (experiment 2), we expected that the effect of propionate
on DMI would be attenuated over time. However, we observed a sustained reduction in DMI.
Conducting a longer-term infusion experiment might provide more time for metabolic adaptation
to propionate infusion and result in increased feed intake at the end of the infusion period. These
adaptations might include increased plasma glucose and insulin concentrations, decreased
plasma NEFA and BHBA concentrations, and decreased hepatic acetyl CoA concentration.
These changes occur over the first few weeks following parturition regardless of treatment
intervention in healthy cows. Propionate infusion may increase the rate of adaptation by
decreasing lipolysis or may decrease the rate of adaptation by suppressing of feed intake.
Additionally, if propionate infusion altered gene expression by upregulating gluconeogenic
enzymes or decreasing FA oxidation, we would expect feed intake to be stimulated. We
analyzed gene expression in the second experiment, but did not detect changes in most measured
133

enzymes. However, 3 d may not have been long enough to alter gene expression, and it is
possible that we did not have adequate power to detect differences in gene expression in that
experiment. Conducting longer-term infusion studies with more cows would allow us to
investigate possible changes in gene expression.
During the second experiment, it is also important to note that we observed that
propionate infusion reduced DMI during the first 4 h post-feeding, but did not affect DMI from 4
– 24 h. This reduction in DMI occurred at the same time that plasma NEFA concentration was
likely highest based on the typical diurnal variation in plasma NEFA concentration. In our
experiment, we collected liver samples to analyze for hepatic acetyl CoA once per day. It would
be useful to know diurnal variation in hepatic acetyl COA concentrations for cows in the
postpartum period. This would allow us to relate hepatic acetyl CoA concentration 4 h following
feeding to DMI during the same period. This is important because we showed that propionate is
most hypophagic for cows with elevated hepatic acetyl CoA. Hepatic acetyl CoA may be
elevated prior to the conditioned meal, and then subsequently be oxidized during the large meals
that occur in the first 4 h following feeding, with little effect of propionate for the remaining
hours in the day.
In experiment 3, lipid infusion failed to increase plasma NEFA and BHBA and hepatic
acetyl CoA. It would be useful to repeat this experiment with a few adaptations. First, the lipid
infusion rate was likely sufficient to elicit a response, however because we infused lipid for 12 h
prior to feeding behavior monitoring, we may have missed the time where lipid infusion had the
most pronounced effect on plasma NEFA concentration. Our decision to infuse lipid for 12 h
prior to feeding was to avoid potential negative health effects of the lipid infusion, such as
increased body temperature or respiration rate. As such, if this experiment were to be repeated,
134

the lipid infusion should be initiated concurrently with daily feed offering. Additionally, cows in
our experiment were lactating, and it is possible that a portion of the lipid infused was
incorporated into milk fat, while decreasing de novo FA synthesis. Repeating this experiment
with non-lactating dairy cows would reduce the potential sinks for infused FA. Medium-chain
FA can be rapidly oxidized in the liver because they do not require CPT 1 or CPT 2 to be
transported into the mitochondria for oxidation. Because of this, I would expect that if the third
experiment were repeated with infusion of medium-chain FA, the rapid oxidation of the lipid
infusion would stimulate hypophagia, particularly when infused concomitantly with propionate,
and might be a viable option.
The fourth experiment in this dissertation was a pilot study with two objectives, 1) to
evaluate the mesenteric catheterization surgery and subsequent cow recovery, and 2) to evaluate
the effect of mesenteric propionate and glycerol infusion on feed intake in early lactation dairy
cows. The main experiment was proposed to evaluate feed intake and feeding behavior in cows
in the postpartum period with iso-energetic propionate or glycerol mesenteric infusions. The
pilot study demonstrated that the surgical catheterization procedure could be successful, although
subsequent attempts at mesenteric vein catheterization surgeries were not all successful. No
differences in surgical technique or time of surgery relative to expected parturition were noted in
the main experiment surgical attempts. We propose that surgical insertion of ruminal vein
catheters may be a possible alternative to mesenteric vein catheterization. Other possible options
include conducting the catheterization surgery under general anesthesia rather than with the cow
standing under local anesthesia or conducting the surgery approximately 30 – 45 days prior to
expected calving.

135

The research in this dissertation provides strong evidence that propionate interacts with
lipolytic state to control feed intake in early lactation. However, feeding studies during the
postpartum period to evaluate feed intake responses to diets that vary in starch concentration and
fermentability are necessary. There is limited research evaluating either starch fermentability or
starch concentration in diets fed to early lactation cows, and almost no research evaluating both
in the same experiment. Conducting experiments to evaluate intake responses to diets with high
and low starch concentration and fermentability will provide much needed practical information
about the most advantageous feeding strategies for cows in a lipolytic state during the
postpartum period.

136