This item is only available as the following downloads:
MECHANISMS FOR MYOCARDIAL OXIDATIVE STRESS IN OBESITY
HEATHER KETELAAR VINCENT
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
Heather Ketelaar Vincent
This dissertation is dedicated to my husband, Kevin, and my "little buddy" Ian.
You are my inspiration and the lights in my life.
More than anything, I wanted to leave the Center for Exercise Science with the
sense that I had contributed somehow toward making where I studied and worked a better
place for my having been there. I hope that the spirit and forays of this project into a
different area have helped to do just that. I can be sure, however, that I have become a
better person for having been here. I owe this is great part to my friend and mentor, Dr.
Scott Powers. There was no better education than the understanding, trust, and faith that
came with his friendship during my stay here. He has helped me in many ways evolve
into who I have become, and he has clarified the direction in which my efforts have led
me. I can only hope that this is just the beginning of a lifelong friendship. As a person
and scientist, he is a hard example to follow. I will do my best to represent him well in
Sincere appreciation to a fellow student who spent tireless hours with me in the
lab, designing equipment and performing surgeries and assays: Amie Dirks. If I can repay
you for your help, I will be here for you. Sincere thanks to three very special loyal
friends who were there for me every time I needed them Rachel Cutler, Shannon
Lennon and Darby Stewart. Without any compensation, you spent many hours taking
care of Ian and helping us during many situations to make life a bit easier. I cannot thank
To those faculty to whom I am indebted for the use of your equipment and
facilities Dr. Charles Wood, Dr. Randy Braith, Dr. Dodd, and Dr. Stephen Borst your
patience and support helped me to collect invaluable data for this work.
Most importantly, to my husband, Kevin, and my son, lan, I love you more than
anything in this whole world. It has been a whirlwind of a tour through our lives during
these past couple years lots of decisions, anxiety, and trying experiences. Above all else,
you are the bright lights in my life. You are my son, my sunshine, and my husband, my
soul mate. May this be just the beginning of a beautiful, healthy and loving family!
TABLE OF CONTENTS
1 INTRODUCTION ..................
Specific Aims ...............
Hypothesis Justification. ............ ....
Significance . . .
2 LITERATURE REVIEW ...............
Introduction . . .
The Obesity Syndrome ...... ...
Obesity and Myocardial Overload .. .....
Myocardial Production of Reactive Oxygen Species. ..
Major ROS and Sources ofROS .. ....
Myocardial Antioxidant Defense .............
Enzymatic Defense .. .... ...
G lutathione . . .
Dietary Antioxidants and Myocardial Protection. .
Oxidative Injury to Myocardial Tissue ..........
Lipid Hydroperoxides and Malondialdehyde ..
Oxidative Damage to Membranes ... .. ...
Obesity and Oxidative Stress .... ... ..
Increased Myocardial Work Rate. .
Compromised Antioxidant Defense. ...
Myocardial Fat Composition and Oxidizability. .
The Genetic Contribution to Obesity .
Preliminary Experiments . .
Susceptibility to an Oxidative Challenge in vitro.
Primary Antioxidant Defense .
Tertiary Antioxidant Defense .. .. ...
. .. 3
. . 9
. . 16
. . 17
. . 1 7
. 2 1
. 2 8
. . 32
. . 33
. .. ix
Lipid Content of the Myocardium . ..33
Unanswered Issues ........ .. 34
3 METHODS ................................. 36
Animals. ... .. ..... ... .... .. ..... 36
Experimental Design and Diet .. .. ..... 36
Animal Model Justification .. ... 38
Assessment of Systemic Changes With Obesity .... ...... 38
Resting Oxygen Consumption (V02) .. .. ........ 40
Heart Rate and Blood Pressure .. .. 40
Blood Glucose and Insulin Concentrations. .. ..... 41
Heart Weight .. .................. ... 42
Adiposity. ............... ................. 42
Heart Tissue Composition. .. .............. 42
Lipid Content of the Myocardium... . ........ 42
Water Content and Dry Weight ......... 43
Radical Production by the Myocardium ... ... 43
Isolated Papillary Muscle Experiments .. 44
Assessment of Myocardial Antioxidant Status .. 45
Oxidative and Antioxidant Enzyme Activity .. .. .. 46
Tissue Thiol Measurements. ......... . 46
Biochemical Indicators of Oxidative Stress .. ... 46
Lipid Peroxidation Measurements ... ...... ... 47
Oxidative Challenges in vitro. . ..... ...... 47
Xanthine-Xanthine Oxidase System (Superoxide Generator) 48
Hydrogen Peroxide System .... .. ... 48
Ferric Chloride System (Hydroxyl Generator) .. .... 48
AAPH System (Peroxyl generator in the Lipid Phase) .. 49
Statistical Analysis ................ .......... .49
4 RESULTS ............. .... ..... .... ..... 50
Diet and Antioxidant Consumption ....... . 50
Body Weight Changes With Feeding.. .. ...... .... 53
Morphological Characteristics . ...... 53
Physiological Characteristics . ...... 56
Heart Rates, Blood Pressures, and Heart Work ... 56
Oxygen Consumption and Body Mass Index (BMI) .. .. ... 60
Blood Glucose and Insulin Concentrations .. .. ... 62
Heart Tissue Characteristics ....... .. .... 62
02" Production: Cytochrome C Assay .. .
Oxidative and Antioxidant Enzyme Activities .
Tissue Thiols .. .. .. ........
Basal Lipid Peroxidation .........
Oxidative Challenges in vitro . .
Correlations Between Lipid Hydroperoxides and
Physiologic Measures ..... .. ..
Stepwise Regression Model for Myocardial Lipid
Peroxidation. ... .... .. .
5 DISCUSSION .
. . .. 63
. . 65
Overview of Principal Findings. .. ... .. .
Lipid Peroxidation in Myocardial Tissue of Obese Animals .
Potential Pathways for Obesity-Induced Oxidative Stress .
Elevated Heart Work ..... .... .
Compromised Antioxidant Defense. .. .... ....
Elevated Lipid Content .... . .
Superoxide Radical Production by Isolated Papillary Muscles ..
Major Conclusions .. .
Physiological Significance ........ . .
Limitations to the Experiment and Future Directions. .. .
A SAMPLE SIZE ESTIMATION. .
B DIETARY AND VITAMIN MIXES FOR
EXPERIMENTAL DIETS .. .. .
C RESIDUAL PLOTS FOR THE REGRESSION EQUATION
IN TABLE 12 . ... . .
. . 96
BIOGRAPHICAL SKETCH ...................
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
MECHANISMS FOR MYOCARDIAL OXIDATIVE STRESS IN OBESITY
Heather Ketelaar Vincent
Chairman: Scotty K. Powers
Major Department: Exercise and Sport Sciences
Obesity is associated with increased myocardial oxidative stress, yet the
mechanisms) responsible for this damage are unknown. We hypothesized that elevated
heart work, an increased rate of superoxide (02 ) production, increased myocardial lipid
content, and insufficient antioxidant defenses contribute to oxidative stress in obesity. To
test this hypothesis, Zucker rats (7 weeks old) were fed experimental diets for 9 weeks to
promote obesity by high-fat intake or lack of expression of the leptin receptor. Lean
control rats (CON, Fa/?) were fed either a control diet (10% fat) or a high-fat diet (FAT,
45% fat), while obese rats (OB, fa/fa) were fed the control diet. Oxidative stress was
assessed by measurement of hydroperoxides (PEROX) and thiobarbituric reactive acid
Compared to CON, the FAT and OB had similar elevations in PEROX and
TBARS (+21% and +33%, respectively, p<0.05). Small but significant differences
(p<0.05) in resting heart work (heart rate X systolic blood pressure) existed between the
FAT and OB compared to CON. Activities ofantioxidant enzymes CuZn-superoxide
dismutase and catalase and endogenous glutathione levels were elevated (32%, 15%, and
18%, respectively, p<0.05) in OB compared to CON. Myocardial lipid content was
increased similarly among all FAT and OB animals (p<0.05) compared to CON. The rate
of 02 formation by isolated papillary muscles in vitro did not differ among the
experimental groups (p<0.05). Regression analysis revealed that the largest contributor to
oxidative damage was myocardial lipid content (R2=0.76, p<0.05). These data indicate
that myocardial oxidative injury is not closely linked with elevated heart work,
insufficient antioxidant defenses or a greater rate of 02 production. In contrast,
myocardial lipid content is a key contributor to obesity-related myocardial oxidative
Obesity is a serious clinical disorder affecting millions of Americans (33%), and the
incidence is steadily increasing per year (7, 26). An alarming trend is the increase in the
percentage of adolescents and young adults who are becoming obese (estimates ranging
from 35 to 40%), with an ever-rising percentage of obese youths becoming severely obese
(51). Obesity is an independent risk factor for cardiovascular disease and increased
mortality (52). The underlying mechanisms for this increased morbidity and mortality are
unknown. Information is scarce regarding the myocardial alterations that occur at the
cellular level that make the obese individual more prone to myocardial injury or
irreversible damage leading to death. Hence, research in this area is clearly warranted
A preliminary biochemical investigation in our laboratory indicates that obesity is
associated with increased myocardial lipid peroxidation and susceptibility to oxidative
damage in vitro (111). The potential consequences of oxidative damage can be severe;
studies examining oxidative stress and the myocardium have shown that cellular membrane
integrity can be lost (97, 114), and lipids and proteins are transiently or irreversibly
altered, resulting in myocardial contractile dysfunction (40). These factors can ultimately
lead to cardiac arrhythmias, poor contractility, infarction, cardiac failure, or sudden death
Genetics and environment have both been identified as major contributors in the
etiology of obesity (11). The individual effects of these factors on the predisposition to
oxidative stress in the myocardium is unknown. The current animal models used in the
study of the systemic effects of obesity include the widely accepted genetically inbred fatty
Zucker rat (fa/fa) and the overfed, overweight rat. Although both models increase
myocardial work (e.g., elevated systolic blood pressure), the acquired and genetic obesity
models have different influences on myocardial work and possibly the level or type of
oxidative stress. The significance of this issue is such that in human obesity, there are both
heritable and environmental factors that are involved in the pathogenesis of obesity and its
association with cardiovascular disease (11). Recent evidence suggests that oxidative
stress is involved in the cellular damage incurred by cardiovascular disorders such as
coronary artery disease, hypertension, atherosclerosis, and vasospastic angina (27).
Considering that obesity (a) is often accompanied by these cardiovascular disorders, and
(b) is associated with myocardial oxidative stress, it is important to examine oxidative
cellular injury in both genetic and acquired obesity models.
There are several potential mechanisms to explain the increase in myocardial lipid
peroxidation associated with obesity: (1) increased myocardial work and oxygen flux
through the mitochondrial respiratory chain (97); (2) a decreased myocardial antioxidant
defense (23,24); (3) increased fat deposition within myocardial tissue (63); and (4)
increased rates of radical formation (12). Although there is indirect evidence to support
that all these mechanisms contribute to myocardial oxidative damage in the obese, direct
evidence is lacking. In addition, it in unclear how each of these factors are differentially
influenced by the fa/fa genotype and diet. Therefore, this investigation will examine the
relationships between oxidative stress and the cellular characteristics of the myocardium
from animals of two obesity models: high-fat fed animals and animals possessing the leptin
receptor defect (fa/fa). We attempted to determine the influence of the leptin receptor
defect (fa/fa) and high-fat feeding on myocardial lipid peroxidation.
Obesity is associated with increased myocardial lipid oxidative damage in the obese
fatty Zucker rat (111). It is unknown whether these obese animals and lean animals that
are fed a high-fat diet are at the same risk for myocardial lipid peroxidation. Several lines
of evidence suggest that there are many potential mechanisms that could promote lipid
peroxidation in the myocardium in either the fa/fa rat or the high-fat fed animal. This
investigation will compare the cellular antioxidant characteristics and in vitro responses of
myocardial tissue from genetically lean (Fa/?), genetically obese (fa/fa), and high-fat fed
rats (Fa/?). Therefore, the specific aims of this project are as follows:
Specific Aim 1: To determine if high-fat fed rats (Fa/?, 44 9% dietary fat) and the obese
rat (fa/fa, 10% dietary fat) experience the same levels of oxidative injury (i.e., lipid
peroxidation) in the myocardium
Hypothesis 1: We hypothesize that the high-fat fed and obese animals (fa/fa) will
have similar levels of lipid peroxidation. Further, we hypothesize further that the
degree of lipid peroxidation will be independent of the fa/fa genotype and
dependent upon the degree of adiposity.
Specific Aim 2: To systematically examine several factors which could contribute to
elevated myocardial oxidative stress (i e., lipid peroxidation) in obesity. These include: (1)
increased heart work (rate pressure product) due to obesity; (2) insufficient intracellular
primary antioxidants such as antioxidant enzymes and glutathione; (3) increased fat
deposition within myocardial tissue; and (4) increased rate of radical formation
superoxidee) by isolated papillary muscles form the heart.
We will test the following hypotheses:
Hypothesis 2a: Obese animals will have a higher double product (i.e., heart rate X
systolic blood pressure) compared to lean animals.
Hypothesis 2b: Glutathione levels and antioxidant enzyme activities will be
reduced in hearts of obese animals compared to lean animals.
Hypothesis 2c: Myocardial tissue obtained from the left ventricles of obese
animals will contain more lipid compared to myocardial tissue obtained by lean
Hypothesis 2d: Contracting papillary muscles from obese animals will produce
superoxide anions at a greater rate compared to lean animals.
We hypothesize that the degree of myocardial lipid peroxidation will not differ
between high-fat fed rats (Fa/?) and the obese fa/fa rats. The factors which can increase
oxidative stress on the heart are similar between both of these animal models. Specifically,
obesity-induced hypertension has been documented in both overfed animals fed high-fat
diets (14, 54, 55) and in obese Zucker rats (18). Hypertension forces the myocardium to
work at greater workloads independent of genotype (14, 61). The myocardial rate
pressure product and oxygen uptake increase in obesity. These processes can lead to
excessive 02 production. In addition, the presence of excessive fat deposition within the
myocardium in genetically obese or overfed, overweight animals serves as an enlarged
target for lipid peroxidation (63). These factors appear to be independent of genetics.
Hence, we speculate that lipid peroxidation levels will be the same in both groups of obese
Our second series of hypotheses relate to factors contributing to increased
myocardial lipid peroxidation in obese animals. First, hypothesis 2a states that obese
animals will have a higher double product compared to lean animals. As previously
mentioned, obesity places a substantial mechanical load on the heart which increases
myocardial oxygen consumption, as evidenced by the increased rate pressure product (2,
14). Second, hypothesis 2b states that glutathione and antioxidant enzyme levels will be
reduced in hearts of obese animals compared to lean animals. Obesity in humans and
animals is associated with lowered serum or tissue vitamin E, 0-carotene, and/or tissue
glutathione (19, 24, 79, 91), and some reports indicate reduced antioxidant enzyme
activities. Hypothesis 3b postulates that the left ventricles of obese (fa/fa) animals will
contain more lipid compared to ventricular tissue of lean animals. Increased deposition of
(polyunsaturated or saturated) fats within tissues is common to obesity and increases the
risk for lipid peroxidation by increasing the oxidation target number (7, 62, 63, 78).
Lastly, hypothesis 4b states that contracting papillary muscles from obese animals will
produce superoxide anions at a greater rate compared to lean animals. It has been
suggested that there is a greater mitochondrial lipid oxidation rate in tissue of obese
animals (12) providing indirect evidence for the notion that the elevated oxygen flux
through the mitochondria may be a primary source of ROS production within myocytes
from obese animals (12).
Obesity is an increasingly prevalent metabolic disorder affecting not only the U.S.
population but also that of the developing world (26). It is associated with many
comorbidities and it complicates health conditions in patients with various cancers,
diabetes, and cardiovascular disease (7, 25). Obesity is linked with a high morbidity and
mortality rate, particularly among cardiovascular patients (52). More frequent have the
incidences of fatal and non-fatal heart disease, arrhythmias and sudden death become in
obese persons during recent years (26). The economic burden that obesity and its related
disorders place on the U.S. health care system is enormous. The direct cost of obesity
alone was recently estimated at $45.8 billion (52). However, this estimate is low due to
the fact that many of the disorders, such as heart disease, that arise from obesity are
classified as illnesses separate from obesity; an additional $ 23 billion are used to treat
these disorders that are associated with obesity.
The prevalence of obesity in the adult population is rapidly increasing by 0.6% per
year for men and 1% per year for women (currently, -33 % of the adult population is
severely obese)(7, 26). More alarming is the fact that the percentage of children that are
becoming obese is rising rapidly (51). These statistics suggest that obesity is a disorder
that will continue to affect our nation's population and drain our health care resources
The U.S. will continue to suffer economically as a result of this disease. Hence, it is
essential that we expand our knowledge about the obesity syndrome.
Obesity is a prevalent disorder in the U.S. population and is also becoming an
epidemic in European countries, Canada and the third world (7). According to the recent
phase of the National Health and Nutrition Examination Survey (NHANES III, 1988-
1991), approximately 33% of the U.S. population is obese, 8% more than when the last
phase of NHANES II was completed (1976-1980) (60).
Obesity is defined as the accumulation of excess fat, such that the body mass index
(BMI, m-na. heIhi'i is greater than 30 kg/m2 (52). This overfatness is associated with a
number of comorbidities, including many forms of heart disease. Fat distribution also
represents risk for heart disease; deposition of fat in the abdomen indicates a greater risk
for coronary heart disease mortality (7).
Obesity can develop as a consequence of environment, such that overeating or
consumption of a high-fat diet induces fat accretion that may exacerbate weight gain.
Genetics can significantly contribute to 30-70% of the cases of obesity by influencing fat
accretion throughout life and passing on genes that predispose the offspring to obesity-
related complications (10, 52). While it is well established that obesity is associated is an
increased risk of heart disease, recent evidence also indicates that obesity is also associated
with an increase in oxidative damage to the myocardium (111). The purpose of this review
is to discuss possible physiological and biochemical links between obesity and myocardial
The Obesity Syndrome
Obesity is characterized by a complex pathophysiology that can impose potentially
harmful consequences on the cardiovascular system. Obesity is often accompanied by
increased plasma volume and hypertension, poor glycemic control, hyperlipidemia and
increased adrenergic drive with reduced adrenergic sensitivity at the tissue level (15, 88).
Further, the myocardium is often hypertrophied and may have fatty infiltration (2).
Obesity can have deleterious cardiovascular effects, but could also initiate several
cellular pathways that may promote myocardial oxidative damage. Specifically, the
workload on the heart is increased by obesity (15), this would increase the oxygen flux
through the mitochondrial respiratory chain (12). Further, the myocardial antioxidant
defense may be insufficient to protect against damage by reactive oxygen species (ROS)
(23, 24). Finally, there may be fatty infiltration and increased polyunsaturated fat
deposition within myocardial tissue that increases the risk for oxidation by ROS (63).
Obesity and Mvocardial Overload
There are several ways in which obesity contributes to an excessive workload on
the heart. Figure 1 summarizes the potential mechanisms involved in obesity-induced
myocardial overload. Excessive weight gain is associated with an increased plasma volume
and subsequent blood volume expansion (4). In addition, the hypertension so often
associated with obesity can be induced by insulin-mediated mechanisms. Hyperinsulinemia
causes sodium and water retention by acting directly on the renal tubules and on the renin-
angiotensin-aldosterone system, and it can also promote arterial smooth muscle
proliferation (88), all of which increase blood pressure and end diastolic volume. Increased
volume enhances diastolic filling and places a stretch overload on the heart (increased
preload). The wall stress in the ventricles increases. According to LaPlace's law (wall
stress = pressure X radius/ 2 X wall thickness), the bigger the left ventricle and/or the
greater the pressure developed by the left ventricle, the greater the wall stress. An
increase in wall stress increases myocardial 02 uptake as more ATP must be used to
generate greater tension to contract against this wall stress (80). In response to this wall
stress, eccentric and/or concentric cardiac hypertrophy will occur. Although gross
mechanical function may appear normal, there are several subtle mechanical alterations
that affect cardiac performance. Specifically, systolic function is compromised such that
the rate of shortening velocity is reduced (80). Also, diastolic dysfunction occurs.
Relaxation time is delayed and peak filling rates are reduced (2). In some specific cases,
vascular resistance may be reduced to counteract the increased blood volume. This may
override the renin-angiotensin effects. Diastolic filling is compromised and stroke volume
Excessive Adipose Accumulation
t Circulating Blood Volume ->
t LV Stroke Volume
T HR X SV = 1 Rate Pressure Product
LV Wall Stress Eccentric Hy
1' LV Wall Stress -* Eccentric Hy
LV Systolic and Diastolic Dysfunction
LV Systolic and Diastolic Dysfunction
where LV = left ventricle, HR = heart rate, SV = left ventricular stroke volume
Figure 1. A schematic representation of the pathogenesis of heart complications induced
Therefore, the heart rate must increase in turn to maintain cardiac output (CO) (2, 80). In
all scenarios, the heart is working at higher work rates in obesity.
Hypertension is present in approximately 60% of obese individuals, with 10% of
those cases being classified as severe (94). The myocardium of hypertensive obese
individuals works against a greater systemic resistance created by elevated blood pressure
(increased afterload). Specifically, blood pressure is increased on average 6 mmHg systolic
and 4 mmHg diastolic for each 10% gain in body fat, with a greater effect observed in
those genetically susceptible to obesity (52).
Systolic dysfunction may manifest as depression of left ventricular peak rate of
contractility. Furthermore, obesity-induced hypertrophied hearts are susceptible to
potentially fatal arrhythmias, or heart failure (14).
In summary, the neuroendocrine and mechanical alterations that occur with obesity
are associated with an increase in the stress placed on the heart. Blood pressure and heart
rate are typically elevated, and preload and afterload are increased. The myocardium
hypertrophies to counteract this overload stress. Alterations in heart performance include
depressed rate of contractility, relaxation, and manifestation of arrhythmias or failure.
Mvocardial Production of Reactive Oxygen Species
A radical is a molecule or a molecular fragment containing an unpaired electron. In
general, approximately 2-5% of the oxygen consumed used during oxidative metabolism is
transformed to radicals or other reactive oxygen species (ROS) (48, 49). ROS are
considered essential in cellular homeostasis and when present in small amounts have been
shown to enhance contractile processes (86). ROS are often scavenged by naturally
occurring protective antioxidant defenses such as enzymes, vitamins and other molecules
within the myocyte so that an antioxidant-prooxidant balance is achieved. Under
pathological conditions or conditions in which electron flux through the electron transport
chain increases (such as exercise or increased contractile activity), ROS production
increases and causes a prooxidant state within the cell (114). When the cellular antioxidant
defenses are overpowered, this causes major disruptions to muscle contractile function,
cellular homeostasis and subsequent damage termed oxidativee stress" (49, 86). This
section will highlight the potential sources of ROS in cells and the types of molecules
Maior ROS and Sources of ROS
The major ROS include superoxide radicals (02 ), hydrogen peroxide (H202), and
the hydroxyl radical (OH.); other contributing ROS are the nitrogen containing species
such as peroxynitrite (ONOOH) and nitric oxide (NO). Molecular oxygen itself is a
diradical, though not highly reactive. The major potential sources of myocardial ROS in
obesity are the mitochondrial respiratory chain, xanthine oxidase activity, the neutrophil
oxidative burst, nitric oxide synthesis, catecholamine oxidation (49, 112). In the healthy
myocardium, a small fraction of the electrons flowing through the electron transport chain
leak from the reaction paths and collide with 02 to form superoxide radicals (48). This
anion is produced at greater rates under conditions that stress the respiratory chain,
including greater myocyte contraction rates (49). Superoxide can react with another 02
radical and two hydrogens to form H202 ; direct dismutation of 02 via superoxide
dismutase can produce H202 in the mitochondrion (Asayama et al. 1991, Girotti 1998).
Though H202 is not a radical, it is a reactive oxygen specie and can cause serious cellular
damage if it accumulates (48). Homolytic fission of the 0-0 bond in H202 produces two
hydroxyl radicals (OH), and can be achieved by an iron salt and H202 by Fenton
chemistry. The OH molecules react with an extremely high rate constant especially with
nearby biological membranes (34, 67). Therefore, OH reacts very close to its point of
formation in a phenomenon known as "site specific" reactivity (34).
Mainly located in the vessel walls of most tissues, xanthine dehydrogenase
catalyzes the oxidation of hypoxanthine to xanthine, and xanthine to uric acid. Xanthine
oxidase uses 02 as the electron acceptor and produces 02" while catalyzing the oxidation
of hypoxanthine to uric acid (114)
Another major source of ROS in cardiovascular disease is the reaction of O2 and
NO that occurs within the milieu of neutrophilic and macrophage oxidative bursts. Upon
phagocytic cell infiltration at the site of injury, these cells release a mixture of these two
ROS that form the strong oxidant peroxynitrate (ONOOH). Also, activated neutrophils
release myeloperoxidase-generated hypochlorous acid (HOCI) which is a strong oxidant
that can cause serious cellular damage (102, 112). It is unclear whether obesity per se
initiates myocardial infiltration by neutrophils, but studies have implicated phagocytic
infiltration ofatherogenic plaques and ischemic/necrotic cardiac tissue as complications
often associated with obesity (102).
While there are many targets for oxidative damage in the myocyte, these processes
elicit significant damage to the myocyte membranes (22). Membrane damage results in
alterations which could lead to ion imbalance and subsequent myocardial dysfunction (63).
It is well documented that the oxidative stress resulting from events such as ischemia-
reperfusion cause serious depression in myocyte function, such as inability to develop
maximal force, reduction in the rate of force development, and reduced segment
shortening (72). Therefore, the myocardium has several defense mechanisms to counteract
these potential effects.
Myocardial Antioxidant Defense
The endogenous antioxidant defense within the myocyte includes both enzymatic
and non-enzymatic components. The defense system exists as a multilevel system, such
that the major protection against acute oxidative stress is provided by glutathione, primary
antioxidant enzymes and dietary antioxidants (20). The secondary defense includes repair
enzymes, such as lipolytic and proteolytic enzymes, proteases and phospholipases that
repair cellular damage following oxidative stress.
The myocardial antioxidant enzymatic defense system includes superoxide
dismutase (SOD) which scavenges 02O, catalase (CAT) which scavenges H202 and
glutathione peroxidase (GPX) which reduces lipid peroxides and neutralizes H202. These
enzymes exist in differing concentrations in varying compartments within the myocyte,
thus providing a strategic defense against ROS generated in vivo. Specifically, the
manganese-dependent isoform of SOD (Mn-SOD) is found in the mitochondria, the
copper/zinc-dependent SOD isoform (CuZn-SOD) is found within the cytosol (Ji 1995).
GPX is found within the mitochondrial matrix and the cytosol, with most of the enzyme
located within the cytosol with a ratio of 2:1 (49). GPX is activated by H202 at lower
concentrations (K, = 1 iM) whereas the other H202 scavenger, CAT, is activated by
higher concentrations of the same substrate (K, = 1 mM). This overlap of substrate for
these two enzymes appears physiologically relevant in that if the oxidative stress of H202
exceeds the capability of GPX, CAT is present to provide protection. CAT, though found
throughout the cell, can be found primarily within peroxisomes. Activity of this enzyme
depends upon the binding of Fe3* to the active site (114)
Glutathione (GSH, y-glutamylcysteinylglycine), the predominant thiol in cells, has
direct antioxidant activity and is also involved in recycling other dietary antioxidants (49).
For example, GSH may neutralize hydroxyl radicals and singlet oxygen by abstracting an
electron and/or donating a proton, and can maintain tissue antioxidant vitamins in the
reduced state (74, 114). GSH is a substrate for GPX, which removes lipid peroxides and
H202. GSH is synthesized primarily by the liver and is transported to extrahepatic tissues
via the circulation. GSH is imported by the tissues and its constituent amino acids are
imported by membrane-bound enzymes (39, 74). Together, these enzymes and thiol
compounds serve as part of the primary defense against ROS and radicals.
Dietary Antioxidants and Myocardial Protection
Exogenous antioxidants have been extensively used to reduce the injury
associated with oxidative stress in muscle tissue Antioxidants are general scavengers of
ROS and radical species, and they can work alone or in combination with each other to
reduce the reactivity of the radicals.
Vitamin E belongs to a class oftocopherol phenolic molecules (at-tocopherol
being the most potent) and can covert Oz2, OH and peroxyl radicals to less reactive forms
(17). It is lipid soluble, and its trimethylhydroquinone head serves to break the chain
reaction of lipid peroxidation that occurs in cell membranes during oxidative stress (48,
71, 106). Once oxidized in this process, the vitamin E radical can be recycled once again
to its native state by other antioxidants such as vitamin C and GSH. Thus, vitamin E acts
synergistically with vitamin C and glutathione during periods of oxidative stress (49, 114).
Vitamin C is water soluble and can directly scavenge 02 OH and peroxyl radicals
in the cytosol and plasma. In addition, vitamin C can reduce the vitamin E radical back to
its original state. Oxidized vitamin C can be reduced again to its native form by electron
donors such as glutathione or dihydrolipoic acid Thus, vitamin C is extremely important
in the restoration of vitamin E in the lipid components of the cell, and in the scavenging of
radicals in the aqueous phase. Further, it is linked with reduced protein glycosylation, a
radical-generating process (21).
3-carotene is also a lipid soluble molecule found in membranes that can scavenge
singlet 02 and 02 Similar to vitamin E, this molecule can provide great protection against
lipid peroxidation and has a postulated role in the reduced uptake of oxidized low density
lipoproteins in the cardiac endothelium. Both benefits are associated with reduced
atherosclerotic disease in the heart. P-carotene appears to be most effective at low doses
Alpha-lipoic acid is an endogenous thiol containing compound that is a potent
antioxidant against all major ROS. It is found in low concentrations in the aqueous
compartments of myocytes, and is largely bound to enzyme complexes, rendering it
unavailable for scavenging ROS. As an exogenous, unbound supplement, lipoic acid may
be effective in recycling vitamin C and serving as a protective thiol-containing molecule
that can aid in reaction of oxidized molecules (53).
Oxidative Injury to Mvocardial Tissue
Although oxidative damage can be incurred on cellular proteins and carbohydrates
(114), much of the recent work has focused on the oxidative injury to cellular membrane
lipids (62, 85). Damaged lipids, as will be described in the next sections, can alter cellular
homeostasis, deteriorate contractile function or cause cell death by apoptosis (34, 105).
Hence, this section will focus on oxidative injury to lipids.
Lipid peroxidation is a common type of damage observed in tissue following
exposure to ROS. Lipid peroxidation is the destruction of polyunsaturated fats (PUFA) in
membranes and is initiated when a ROS is able to abstract an allylic H atom (1 electron
reduction) from a methylene group of a PUFA molecule. This forms a reactive lipid
peroxyl radical, ROO (where R = lipid chain, 00 denotes peroxyl group) which can react
with an adjacent PUFA, triggering exacerbating rounds of free-radical mediated lipid
peroxidation (28) This process initiates a cascade of lipid peroxidation that amplifies the
destructive effects of the initial peroxidation insult. Alternatively, two electron reduction
reactions with lipid hydroperoxides (ROOH) can lead to formation of redox-inert alcohols,
a process which serves as a secondary or reparativee" level of cytoprotection.
Recent work has shown that basal levels of lipid peroxidation by-products are
elevated within the myocardium and liver tissue of obese rats (fa/fa genotype) compared
to lean control animals (Fa/- genotype)(57, 111). The significance of lipid peroxidation
within membranes is that the membrane fluidity is decreased and permeability is increased.
Furthermore, consumption of high fat diets or diets high in unsaturated fats (especially
containing > n-3 fatty acids) exacerbates the susceptibility to free radical-mediated
Several by-products of lipid peroxidation such as lipid hydroperoxides,
malondialdehyde (MDA), conjugated dienes and 4-hydroxynonenal (4HNE) are measured
to determine the extent of oxidative damage incurred on the tissue (38). The most
common measures used to determine the degree of oxidative injury in lipids are MDA and
lipid hydroperoxides (47).
Lipid Hvdroperoxides and Malondialdehyde
An outline of the postulated pathways by which hydroperoxides and MDA are
produced is shown in Figure 2. The importance of measuring these two by-products is that
each by-product occurs at different places within the pathway. Decomposition of PUFA
by 02 directly results in primary hydroperoxide products. Further breakdown of
hydroperoxide products generates endoperoxide radicals. These radicals, when exposed to
heat or acids, produce secondary MDA products (47). Because of considerable
disagreement within the literature with regard to the "optimal" lipid peroxidation measure
and the inability to specifically detect from where the MDA was derived, it is
recommended at least two different measures are used. Therefore, we will measure both
MDA and lipid hydroperoxides to evaluate the level of lipid peroxidation in the
myocardium of both lean and obese animals.
Oxidative Damage to Membranes
Lipid peroxidation reduces membrane integrity (97). A loss of membrane integrity
in cardiac myocytes can lead to arrhythmias, myocyte contractile dysfunction and cell
death (105). In addition, changes in membrane lipid content and permeability can alter
enzymatic membrane processes (i.e., ATPase activity)(46).
The impact ofperoxidative processes is enhanced when dietary consumption of
PUFA increases; reports using overfeeding models indicate that cell membrane
composition of liver and aortic tissue generally reflects dietary consumption of specific
lipid groups, though not absolutely (46, 59, 62, 107). A consistent finding is that
peroxidative injury increases as a function of dietary consumption of PUFA (46, 78, 107).
The lipid radicals that form as a result of peroxidation are believed to interfere with
essential membrane function of protein channels embedded within the membrane and
maintenance of ion gradients between cellular compartments (97). Lipid peroxidation of
Figure 2. The chemical pathway summary of lipid peroxidation products
maliondialdehyde and lipid hydroperoxides as described previously (47).
PUFA also leads to the formation of 4-hydroxynenal (4HNE), a toxic aldehyde among
others, that potentiates cytotoxic oxidation processes (98).
Furthermore, lipid peroxidation is associated with decreased membrane integrity,
and may in fact assist in "labeling" that affected cell as a target for neutrophilic attack (48,
58). Recent evidence also suggests that greater formation ofhydroperoxides can initiate
events that lead to apoptosis (34). Although it is beneficial to increase the fluidity of
cellular membranes to an optimal physiological level to enable recycling of receptors or
improve membrane-bound enzyme movement, fluidity that compromises cellular integrity
is dangerous for all cellular functions.
Obesity and Oxidative Stress
Evidence from our laboratory suggests that obesity is associated with an increase
in myocardial oxidative damage as evidenced by increased levels of by-products of lipid
peroxidation (111). Damaged lipids can alter cellular homeostasis, deteriorate contractile
function or cause cell death (105) Substantial indirect evidence suggests that there are
several potential factors that may contribute to myocardial lipid peroxidation in the obese.
Increased Myocardial Work Rate
Obesity is characterized by an increased mechanical load on the heart due to
increased fat, total body mass, and peripheral resistance (7). Hyperinsulinemia appears to
elevate blood pressure by activating the renin-angiotensin system or renal tubules directing
to retain sodium, by stimulating arterial smooth muscle hypertrophy, or by altering the
ionic (Ca2*) efflux from the smooth muscle cells (88). Blood pressure is elevated, and
stroke work is increased (2). Very recent evidence has shown that leptin, the adipocyte-
derived hormone, can act (1) through the central nervous system to either increase the
sympathetic nerve activity and vascular resistance in the kidney, (2) a decrease in renal
plasma flow, and/or (3) an increase in heart rate (43, 70) Together, these factors appear to
increase the rate pressure product and myocardial oxygen consumption (15).
Theoretically, the oxygen flux through the mitochondrial respiratory chain would be
increased in response to this workload. Animal studies indicate that blood pressures are
elevated in obese Zucker rats or rabbits fed either Purina chow diets or high fat diets (14,
18, 54, 55, 64). Human studies also report elevated blood pressures and/or heart rates in
obese humans (2). This increased work rate of the heart would increase myocardial
oxygen consumption, and subsequently, oxygen flux through the mitochondrial electron
Studies have reported elevated mitochondrial oxidative capacities in hearts of
obese mice (12) and mitochondrial lipid oxidation providing indirect evidence for the
notion that oxygen flux through the mitochondria may be a primary source ofROS
production (76). Some investigators report that maximal state 3 respiration (i.e., ADP-
stimulated) is greater in cardiac mitochondria from older obese (ob/ob) mice when
compared to lean mice (12). These data have been interpreted as an indication that obese
animals have tighter mitochondrial coupling, and therefore a greater rate of 02
consumption (76). As mentioned earlier, it is well established that increased 02 uptake and
electron flux through the electron transport chain leads to increased 02 formation (114).
Compromised Antioxidant Defense
Myocardial oxidative damage may be the result of an insufficient cellular
antioxidant defense (48). Reductions in antioxidant enzymes activities (SOD, GPX and
CAT) or reductions in the level of dietary antioxidants such as vitamin E, P-carotene and
vitamin C have been associated with increased lipid peroxidation and cellular damage (23,
24, 114). Lowered glutathione levels also contribute to lipid peroxidation (98). It is
possible that any of these mechanisms alone, or in combination, could contribute to the
elevated lipid peroxidation in obese animals.
Our initial experiment indicated that heart homogenates from genetically obese
animals were more susceptible to oxidative damage. Specifically, an iron-mediated
oxidative challenge in vitro resulted in greater production of thiobarbituric acid reactive
substances (TBARS) in heart homogenates of obese Zucker animals (111). These findings
may suggest that the antioxidant defense may not have been sufficient to protect against
the oxidative stress incurred by obesity.
Obesity lowered plasma levels of antioxidants such as vitamin E and p-carotene in
obese children and adults (24, 75, 79). In addition, this lowered antioxidant status was
associated with increased plasma levels of PUFA in obese children (23). The relationship
between myocardial lipid content and antioxidant status in various obesity models is not
Tissue antioxidant enzyme activities are also altered in various obesity models. Our
initial study indicated that there were no differences in the activities of the primary
antioxidant enzymes within the heart, CAT, GPX and CuZn-SOD (111). However, the
activity of the Mn-SOD isoform was elevated in hearts of these obese Zuckers. Other
studies report reduced myocardial CAT and GPX activity (units/g heart tissue) in gold-
thioglucose-induced obese mice (12). In contrast, investigators treating lean rats with
obesity-inducing high fat diets using sources such as PUFA (corn or fish oils) or lard have
shown increased liver GPX activity with no change in SOD (107), or unaltered hepatic
antioxidant enzyme activities (45). Lean Wistar rats were fed a variety of n-3 fatty acid
high fat diets, and myocardial total and Mn-SOD and GPX values were found to be lower
than compared to rat chow fed controls; urinary and tissue TBARS values were also
greater compared to those of chow-fed controls (62). Clearly, these findings regarding the
effects of obesity on antioxidant enzyme activities are divergent. To provide clarity to the
question whether myocardial antioxidant enzyme activities are sufficient to protect against
the stress of obesity, this study will measure SOD, GPX and CAT in both high-fat fed and
fa/fa obese animals.
GSH levels appear to be altered in obesity. Glutathione is depleted (reduced by
45%) in the livers of overfed, overweight animals, and synthesis rates are reduced by 40%
(93). These data are corroborated by findings which indicated that liver glutathione stores
are depressed in overfed, obese mice (92). Furthermore, these animals were more
susceptible to allyl alcohol-induced injury and necrosis. Interestingly, the obese Zucker rat
resists drug-induced hepatotoxicity, and this has been correlated with this genotype's
higher glutathione levels compared to their lean counterparts (9). Our initial study
indicated that the myocardial non-protein thiol level (90% glutathione) was higher in
obese Zucker rats compared to lean, though this value was not found to be significant.
These contrasting findings do not elucidate the relationship between GSH and obesity-
induced myocardial lipid peroxidation. Hence, this is justification to assess the thiol
content (GSH) of the myocardial tissue of genetic and overfed obese animals to determine
whether alterations in GSH content can contribute to the elevated lipid peroxidation in
In summary, there is no definitive evidence to support that the antioxidant defense
is compromised in the myocardium of obese animals. This forms the rationale to compare
the antioxidant enzyme activities and GSH content of myocardial tissue from two models
of obesity in this experiment.
Myocardial Fat Composition and Oxidizability
Findings from other studies report that significant correlations exist between the
type of tissue lipid and degree of lipid oxidizability in vivo or in vitro (109). Increased
myocardial PUFA increases the risk for oxidative attack. Specifically, greater serum
formation of TBARS occurs in the presence of triglycerides compared to high-density
lipoprotein (HDL) cholesterol (85). Lipoprotein oxidizability is also influenced by
lipoprotein composition, such that PUFA are more readily oxidized than monounsaturated
fats (107, 113). Whether the enhanced oxidizability of the tissue lipids in obesity is due to
increased oxidant challenge, decreased antioxidant defense, or altered phospholipid
composition requires further investigation (109).
Fatty infiltration of the myocardium increases the risk for lipid oxidation by
providing more fat substrate targets for oxidation (63). In addition, dietary consumption
of PUFAs or saturated fats can affect the fatty acid composition of the
phosphatidylcholine or ethanolamine molecules in total heart membrane phospholipids (62,
63). Considering the PUFA target for oxidative damage can be enhanced by dietary
consumption of fats, this may in part explain the greater susceptibility to myocyte damage
in obesity (57).
Both human and animal investigations have reported reduced antioxidant levels in
tissue and plasma of obese subjects (19, 24, 91). Swine fed a diet enriched in fish oils or
other 18:3 oils showed mild symptoms of vitamin E deficiency; following supplementation
with vitamin E and selenium, these symptoms disappeared (19, 91).
It is unknown whether the high-fat diet typically consumed by obese humans which
consists of-45% fats (approximately 18% saturated fats, 16% monounsaturated fats and
6% PUFAs) (27) renders the myocardium susceptible to lipid peroxidation in vivo or in
vitro. It is also unclear how this type of diet affects antioxidant enzyme activities and GSH
levels in myocardial tissue. This experiment will address this issue by feeding lean animals
the typical fattening diet, and assessing lipid peroxidation and susceptibility to oxidative
damage in vitro.
The Genetic Contribution to Obesity
The obesity phenotype cannot be simply reduced to Mendelian segregation
patterns largely for two reasons. First, body fat content and excess body fat results from
an intricate network of interactive causes that may be related to specific DNA sequencing
but are also a function of behavior and lifestyle (11). In addition, obesity is also a
heterogeneous phenotype, and there is growing evidence that a certain phenotype is
modified by other causal factors. Considering that human obesity classification schemes
consist of a minimum of four different phenotypes, each with different cases, it is difficult
to identify the etiology.
Any contribution of a genetic effect on caloric intake in human appears to be
minimal. Familial correlations computed in different relative types from the Quebec Family
Study revealed that there were no significant heredity effects on caloric intake (10, 11).
However, when intakes of carbohydrate, fat and protein were expressed in a percent of
total energy intake, the contribution of genetic factors increased (11). These results
suggest that macronutrient selection may be under genetic control, and could indicate
susceptibility of some individuals to be in positive energy balance over a long period of
time. It is also possible that differential genetic expression of specific (neuro) peptides that
modulate or inhibit food intake are the underlying basis for individual variations in energy
and macronutrient intakes (11). For example, over expression of stimulatory
neuropeptides such as norepinephrine, neuropeptide Y, galanin and endogenous opiates
increase food intake, and elevated levels ofserotonin, histamine, neurotensin,
cholescytokinin, glucagon, insulin and corticotropic releasing hormone suppress food
Among the animal studies that employed high-fat or calorically dense diets to
fatten animals and simulate diet induced hyperphagia, there is much heterogeneity in
dietary treatments. The result is a wide range of physiological responses differential energy
intakes based on fat or sugar content of the diet and animal model (14, 42, 55, 62, 95).
Recently, investigators have designed specialty diets to identify animals that are
"susceptible" to obesity (64). Outbred Sprague-Dawley rats show differential responses to
a condensed milk diet such that half of the rats became obese, and some gained moderate
fat weight (64). Other studies show that genetically inbred Zucker rats possessing the
heterozygous Fa/? genotype also respond to fattening diets within 3-4 weeks (33). The
responsive nature of these susceptible animals has been attributed to a number of factors
including alterations in expression of satiety factors (leptin) and lipogenic enzymes,
improved energy efficiency (improved ability to utilize less energy per unit food intake,
fewer energetic futile cycles), and increased sensitivity to insulin that would promote more
rapid fat deposition in response to calorically dense diets (11).
It is unknown whether the leptin receptor defect in the fa/fa obese Zucker rat
increases the risk for myocardial oxidative stress. Intuitively, oxidative stress would be
elevated in any myocyte with elevated rates of oxidative phosphorylation (i.e., myocytes
that function against a greater workload) regardless of the single genetic defect. However,
there may be a greater risk for oxidative stress in human or animal subjects with high
heritability for obesity. It is well established that familial predisposition, or genetics, is a
primary risk factor for cardiovascular disease, and that heart disease is complicated by
obesity (7). Considering the important and significant role of the genetic influence upon
the development of obesity, it is also essential to examine the influence of the fa/fa
genotype in our investigation.
Our laboratory has recently performed initial experiments investigating the
relationship between myocardial oxidative stress and obesity (111). Using the fatty
Zucker rat animal model, hearts from 12 month old lean (590 + 60 g, Fa/?) and obese (881
+ 56 g, fa/fa) males were analyzed for a series of antioxidant and lipid peroxidative
characteristics. The data revealed that the hearts from the obese animals had higher basal
levels of lipid hydroperoxides and thiobarbituric reactive acid substances. We performed
experiments to compare the primary enzymatic, non-enzymatic and tertiary antioxidant
defense within the heart.
Susceptibility to an Oxidative Challenge In Vitro
Production of lipid peroxidation was exacerbated in the homgenized heart tissue of
obese animals compared to the lean following an iron-induced oxidative challenge in vitro.
These data indicate that the myocardial tissue from obese animals had higher basal lipid
peroxidation, but also a higher level of hydroperoxides per unit lipid following the same
exposure to oxidative reagents (ferric chloride) in an in vitro bath. We attributed this
susceptibility to oxidative damage to either: (1) a compromised antioxidant defense within
the myocardium, or (2) an increased lipid substrate availability within the myocardium of
Primary Antioxidant Defense
The left ventricular myocardial samples from the obese animals did not have
higher antioxidant enzyme activity levels compared the lean animals. Specifically, catalase
(CAT), glutathione peroxidase (GPX), total superoxide dismutase (SOD) and the copper-
zinc dependent SOD activities were not higher than those in the myocardium of the lean
animals. However, the manganese-dependent isoform of SOD (Mn-SOD) in the obese
animals showed significantly higher activity compared to the Mn-SOD activity in the lean
rats. The myocardial non-protein thiol content (representative of glutathione) was not
significantly different between the groups (14.1 + 1.1 lean, 16.2 + 2.0 obese).
Tertiary Antioxidant Defense
Evidence exists to support the notion that heat shock proteins (HSP)s of the 72 kD
family provide protection against oxidative stress in vitro and in vivo (20, 66). In our
preliminary study, we measured the relative contents of the constituitive isoform of HSP
73 kD, and the inducible HSP 72 kD. Our data revealed that the relative contents of both
of these HSPs were not significantly different between the lean and the obese groups.
Lipid Content of the Mvocardium
The lipid content of the hearts from obese animals was nearly double that of the
lean animals (59% versus 33% of the total wet mass). Others have suggested that elevated
lipid peroxidation may be the result of increased lipid substrate availability (34, 63), and
our data appeared to be in agreement with this postulate.
Our data suggest that elevated lipid substrate and an insufficiently up-regulated
antioxidant defense are important issues in obesity induced myocardial oxidative damage.
However, there are other interpretations of these data that require investigation. First, the
elevated Mn-SOD activity suggests that either oxidative respiration is occurring at a
higher rate in the obese animals compared to the lean (i.e., greater heart work), or there is
an increased rate of superoxide production at the site of the mitochondrion to cause up-
regulation of the enzyme. Second, the antioxidant defense also includes dietary
antioxidants which are important in preventing lipid peroxidation, inlcuding vitamin E and
P-carotene. These dietary antioxidant characteristics of the myocardial tissue of these fatty
Zucker animals have not previously been measured. It is possible that obesity elevates the
level of oxidative stress such that the tissue vitamin E and 3-carotene levels may not be
sufficient to scavenge the ROS generated in the myocardial tissue. Third, it is possible that
this elevated lipid peroxidation may be in part due to the leptin receptor defect of the
Zucker strain. All the perturbations that are associated with the genetic defect of the
Zucker strain have not yet been elucidated, and it is important to address whether the fa/fa
genotype or diet-induced obesity in heterozygotes (Fa/?) is more important in contributing
to myocardial oxidative stress.
Male lean and obese fatty Zucker rats (7 weeks of age) were used in this
experiment. This age of rat was chosen due to the inability to separate animals by body
weight between the lean and obese (fa/fa) groups (13, 18). Males were used to prevent
any possible antioxidant-protective effect of estrogen in females (114). Sample sizes are
based on a statistical power analyses performed using data from our preliminary
experiment (Appendix A). Animals were individually housed, maintained on a 12:12 hour
light:dark cycle. Following a one-week adjustment period, the experimental diets were
Experimental Design and Diet
The control diet contained the National Research Council's recommended daily
nutrient intake for rats (77, 84) High sugar or high fat diets have been widely used to
fatten animals within 3-4 weeks, with results being more pronounced at 8-10 weeks (33),
and have been characterized as highly palatable to rats (68, 89, 95). This study will employ
a high-fat, high-refined sugar diet in an effort to most closely mimic the diet consumed by
obese Americans (27, 83). This type of diet was also chosen in the effort to induce
voluntary hyperphagia. Data from theFramingham and Lipid Research Clinics (LRC)
projects (27, 83) and other cohort studies (6, 35, 110) indicate that the typical diet
Table 1. Schematic of Experimental Design using the fatty Zucker Rat Model
LEAN (Fa/?) Control Diet n=15
LEAN (Fa/?) High-Fat Diet n=15
OBESE (fa/fa) Control Diet n=15
consumed by obese humans is comprised of-15% protein, 39-50% total fats (-18%
saturated fat of animal origin, 6% polyunsaturated fats, 16% monounsaturated fats with
unsaturated/saturated ratios ranging from 0.24-0.4) and -37-42% carbohydrates (15-20%
from added or simple sugars). To determine whether obesity elicits an alteration in the
antioxidant status of obese animals, dietary antioxidant concentrations were the same per
kg food, and the daily antioxidant intakes with feeding were retrospectively calculated.
Therefore, in this study, lean animals (Fa/?) were fed a calorically dense diet for a period
of nine weeks, as this time period appears to be adequate to induce significant obesity in
high-fat fed groups (33, 107). The diets for all three groups were prepared by Research
Diets Inc. (New Brunswick, NJ). The macronutrient percentages of the diets are contained
in Table 2. Specific macronutrient and micronutrient composition of the diets are in Table
3. Animals were allowed to eat food and water ad libitum, and caloric intake was
monitored daily (food volume in g, and total calories consumed.
Animal Model Justification
This study used an animal model comparable to human obesity, the fatty Zucker
rat. This animal was chosen because: (1) the invasive nature of these experiments
prevents the use of human subjects; (2) the fatty Zucker rat is a genetic model of obesity
possessing similar symptomology as humans including hypertrophy and hypertension (13);
(3) its widespread acceptance as a model for investigating human obesity; and (4) fatty
Zucker animals demonstrate depressed ventricular function similar to that observed in
human cases of obesity, such that responsiveness to 0-adrenergic stimulation is reduced
and the pressure developing ability of the ventricle is reduced (lower peak systolic stress at
any given volume)(l5, 82).
The overfed, overweight rat model has been cited as the best animal model to use
in any pharmacokinetic research (16) and studies of human hypertension.
Therefore, both the genetic (fa/fa lepitn receptor defect) and overfed (high-fat),
overweight animal model will be used in this investigation.
Assessment of Systemic Changes With Obesity
Several important physiological variables were measured before, during, and
following the dietary treatment period. These measures included resting body weight,
Table 2. Description of the control and high-fat, high-carbohydrate diet constituents fed to
lean and obese Zucker rats.
Control Diet (D12450) Fattening Diet (D12451)
Protein 15% Protein 15%
Carbohydrate* 75% Carbohydrate** 45.1%
Fat 10% Fat 44.9%
Total 100% Total 100%
Largely complex sugars (2/3 cornstarch) **Added -10% refined sugars (2/3 sucrose)
Table 3. Macronutrient and micronutrient composition of the diets fed to the experimental
groups of lean and obese fatty Zucker rats.
Ingredient g/kg diet Ingredient g/kg diet
Casein, 80 mesh 150,00 Casein 150.00
L-Cysteine 3.00 L-Cysteine 3.00
Cornstarch 250.00 Comstarch 72.80
Sucrose 350.00 Sucrose 172.80
Soybean Oil 25.00 Soybean Oil 25.00
Lard 20.00 Lard 177.50
S10026 10.00 S10026 10.00
L-Cysteine 3.00 L-Cysteine 3.00
Dicalcium Phosphate 13.00 Dicalcium phosphate 13.00
Calcium Carbonate 5.50 Calcium carbonate 5.50
Potassium Citrate, 1 H20 16.50 Potassium Citrate, 1 H20 16.50
Vitamin Mix (V10001) 10.00* Vitamin Mix (VI0001) 10.00*
Choline bitartate 2.00 Choline bitartate 2.00
*See Appendix B for details of Vitamin Mix constituents.
(Soybean oil is 14% saturated, 23% monounsaturated, 51% linoeic acid and 7% linolenic
oxygen consumption, heart rate, blood pressure, and blood glucose and insulin levels.
Experimental details for each measure follow.
Resting Oxygen Consumption (VO)
Body masses were recorded at the beginning of the study and weekly thereafter
until sacrifice. Resting oxygen consumption (VO2) of each animal was assessed at the
conclusion of the feeding treatment to determine differences between groups (32). Oxygen
consumption was measured by open-circuit spirometry using a specially constructed,
sealed metabolic chamber (5 X 6 X 5 cm, Truemax gas analyzing system). Animals
remained in the chamber with oxygen consumption measured upon equilibration of the gas
in the chamber (-40 min). Flow rates for gas sampling were set at 0.3 L/min., and resting
VO2 was estimated using the following formula: (flow rate)(% 02 difference between the
ambient air and the chamber)/ body mass (kg) = V02 in ml/kg/min.
Heart Rate and Blood Pressure
Systolic blood pressure (BP) and heart rate (HR) in awake, conscious animals
were assessed in all animals. A tail pressure cuff system (Kent Scientific, model #s
BP 1001, BP1004) was used to determine systolic blood pressure. HR was determined
using this same apparatus by allowing the piezoelectric transducer to detect the pulsations
of blood flow within the proximal region of the lateral tail vein. The analogue signal was
directed through a pre-amplifier and A/D converter to a pen chart recorder (Grass
instruments). One lean and one obese animal were tested simultaneously to reduce any
experimental variations between testing sessions. Prior to any data collection, animals
were placed into the warming restrainer on three different occasions for a period of 30 min
to acclimate them to the procedure and reduce any inflation of the true BP. BP and HR
measures were collected prior to the feeding treatment, once each week, and at the
conclusion of the feeding period.
Blood Glucose and Insulin Concentrations
Immediately prior to sacrifice, a -5mL blood sample was obtained from cardiac
puncture. Fresh blood samples were collected using EDTA treated vacutainer tubes.
Plasma was separated by centrifugation and immediately frozen for later analysis of blood
glucose and insulin levels. Blood glucose was assessed using an enzymatic, colorimetric
technique (101). Insulin levels were assessed by a radioimmunoassay technique described
previously (5)(commercial kit, LINCO Research 125I label). Glucose samples were
performed in duplicate, and insulin samples were performed in quadruplate. The average
of two hematocrits values were recorded as the sample score.
Immediately following sacrifice, the hearts were rapidly excised and placed in
aerated ice-cold modified Kreb's solution to remove the remaining blood. Following
excision of the papillary muscles, the hearts were blotted and weighed immediately.
A BMI equivalent for rats, the adiposity index, was performed in all animals using
a well-documented method (65). Briefly, the length of the rat from the tip of the nose to
the anus was measured The body mass and the length of the animal was calculated using
the formula: Adiposity Index = the cube root of the body mass (g)/ length (mm) X 104.
The adiposity index was determined immediately prior to sacrifice.
Heart Tissue Composition
To determine whether the dietary treatment affected heart tissue composition, the
fat, water composition and dry weight of the hearts of lean and obese animals were
determined as follows.
Lipid Content of the Myocardium
A modified version of an earlier extraction technique was used to isolate
myocardial lipid (30). Briefly, heart samples were homogenized in a methanol: chloroform
mix (2:1 v/v) for 2 min at room temp. Samples were centrifuged for 4 min at 400 X g.
Supernatants were decanted, and the pellet was resuspended and re-extracted with
methanol:chloroform:0.2 N HCI (2:1:0.8 v/v). The two phases were separated by another
2 min centrifugation at 400 X g. The supernatants were pooled and the phases were
separated by a third 4 min centrifugation at 400 X g. The lower chloroform phase was
removed and neutralized by drop-wise addition of methanolic NH4OH. Samples were
concentrated under a stream of nitrogen. The weight of the residue was recorded as the
amount of lipid mass per unit heart weight.
Water Content and Dry Weight
To determine the myocardial water content and dry weight, a piece of ventricular
tissue was cut and placed in a pre-massed tube. The sample's weight was recorded. The
sample was freeze-dried in an evaporator at a negative pressure of 10' mmHg, and re-
weighed to obtain the sample's dry weight (tissue protein/lipid). The water weight of the
sample (tissue water) was calculated by: wet weight- post-drying weight.
Radical Production by the Myocardium
To determine whether obesity affects the respiration rates or the radical production
(specifically 02') of myocardial tissue papillary muscles were isolated and stimulated in
vitro, and an indirect assessment ofROS production was performed.
Isolated Papillary Muscle Experiments
To determine whether hearts from obese animals generate ROS at greater rates,
papillary muscles were isolated from hearts from all animals as described previously (15).
In brief, animals were anesthetized with an intraperitoneal injection of sodium
pentobarbital (50 mg/kg). After reaching a plane of surgical anesthesia, the chest of the
animal was opened, exposing the contracting heart. The heart was subjected to
cardioplegic arrest by infusion of ice cold modified Kreb's Hensleit buffer (all in mmol/L:
NaCI 115.0, NaHCO3 20.0, KCI 4.0, K2HPO4 0.9, MgSO4 1.1, CaCI 2.5, and glucose
11.0) by a syringe through a cut in the aortic root. Ice cold buffer was applied to the heart
topically to assist in cooling of the organ. After flushing the heart with buffer, the heart
was removed from the animal and placed into an iced tissue bath with modified Kreb's
buffer bubbled with 95% 02/5% CO2. Under a magnifying glass, a papillary muscle was
rapidly excised and tied with sutures on both ends. The muscle was transferred to a small
tissue chamber containing warm Kreb's buffer (37C, 30 mi) bubbled with the same gas
mixture. One suture end was fixed to the chamber, and the other end fixed to a force
transducer (Grass Instruments, Model #FTIO). A specially designed pair of platinum
electrodes provided field stimulation to the muscle at the following parameters: 100V,
50Hz, 2 ms duration and 3 pulse per sec (to simulate 180 bpm). Following a 15 min
equilibration period, the muscle was stimulated for a 30 minute time period. To determine
whether 02 anions are formed at greater rates in response to electrical stimulation, the
bathing medium also contained 10"' M cytochrome C (C-2506, Sigma Chemical).
Cytochrome C is reduced as a function of superoxide production. The tissue bath was
wrapped in foil and the experiments were conducted in a dimmed laboratory to reduce
photoreduction in ambient light. Following the 15 min stimulation treatment, the bathing
medium was collected and analyzed spectrophotometrically for the reduction of
cytochrome C as previously described (87). The magnitude of absorbance change at 550
nm reflected the amount of 02 in the bathing medium (82). To confirm that the assay was
detecting 02 production, hypoxanthine (5 X 10" M; H-9377 Sigma Chemical) was added
to xanthine oxidase (0.02 U/ml; X-4500 Sigma Chemical) to produce 02 that reduced
cytochrome C. Alternatively, native SOD (103 U/ml; S-7008 Sigma Chemical) was added
to the medium to test inhibition of the radical.
Assessment of Mvocardial Antioxidant Status
To determine whether obesity or the overfeeding treatment affected oxidative and
antioxidant capacity, left ventricular samples from all groups were assessed for oxidative
and antioxidant enzyme activities,
Oxidative and Antioxidant Enzyme Activity
Citrate synthase (CS; EC 220.127.116.11) activity was used as a marker for oxidative
capacity using a method previously described (99). Superoxide dismutase (SOD; EC
18.104.22.168), selenium glutathione peroxidase (GPX; EC 22.214.171.124) and catalase (CAT; EC
126.96.36.199) activities were used as markers for antioxidant capacity using previously
described procedures (1, 29, 81). All assays were performed in duplicate and on the same
day to reduce interassay variation. Activities were normalized to protein in the sample
using previously described spectrophotometric dye binding methods (36, 111).
Tissue Thiol Measurements
Tissue thiols are molecules that contain sulfhydryl groups. They are important in
the regulation of both cellular redox status and antioxidant capacity (37). Therefore, total,
protein and non-protein thiols from the left ventricle were assayed from all experimental
animals. Thiol content was determined spectrophotometrically using a previously
described DTNB-based technique (50). Since glutathione is the dominant non-protein thiol
in the cell, this measure was used as a marker of tissue glutathione levels (49).
Biochemical Indicators of Oxidative Stress
To determine the amount of radical-mediated oxidative damage in the heart, left
ventricular levels of two by-products of lipid peroxidation were measured.
Lipid Peroxidation Measurements
Malondialdehyde levels were determined spectrophotometrically using the
thiobarbituric acid-reactive substances (TBARS) method previously described (108). The
agent 1,1,3,3-tetraethoxypropane was used as the standard for this assay. Samples were
performed in duplicate.
Lipid hydroperoxides were quantified using the ferrous oxidation/xylenol orange
technique previously reported (44). Cumene hydroperoxide was used as the standard for
this assay. In our laboratory, the coefficients of variation for the TBARS and lipid
hydroperoxide assays are -3 and 4 percent, respectively. All samples were performed in
Oxidative Challenges in vitro
To investigate the relationship between obesity and myocardial antioxidant
capacity, heart homogenates from animals in all groups were subjected to a series of
several different ROS-generating systems. A section of the left ventricle from each heart
was homogenized in 0.9% saline at a concentration of 10:1 in nitrogen gassed 50 mM
potassium phosphate buffer at pH 7.4 according to a previous method (40). Aliquots of
the homogenates were incubated at a concentration of 10 mg protein/mi in the presence of
an ROS generating system. Following each challenge, the homogenates were analyzed for
lipid peroxidation using the previously mentioned technique (108).
Xanthine-Xanthine Oxidase System (Superoxide Generator)
Superoxide radicals were generated by the reactions involved in a xanthine-
xanthine oxidase system similar to an earlier method (87). One ml of 1 mM xanthine and
0.1 IU xanthine oxidase were added to a I ml aliquot of heart homogenate and incubated
at 37C for 15 min.
Hydrogen Peroxide System
One ml of hydrogen peroxide (100 mM) was added directly to a one ml aliquot of
heart homogenate and incubated at 370C for 15 min according to a previous method (96).
Ferric Chloride System (Hydroxvl Generator)
Hydroxyl radicals were produced in the heart homogenates by adding 0.1 mM
ferric chloride (FeCI3) and 1 mM ADP. The choice of these particular concentrations is
based on previous (3) who found that this concentration of iron-ADP induced free-radical
mediated arrhythmias in the isolated perfused rat heart and that it was possible to prevent
these arrhythmias by perfusing the heart with SOD (3).
AAPH System (Peroxvl Generator in the Lipid Phase)
Peroxyl radicals were generated in the aqueous phase of homogenate by thermal
decomposition of2,2'azobis(2-amidinopropane)-dihydrochloride, (AAPH). Iml of AAPH
solution and Iml of heart homogenate will be mixed in and incubated at 37C for 2 hrs.
Following incubation of the heart homogenates in each system, 200 mM butylated
hydroxytoluene (BHT) was added to stop the oxidative reaction. TBARS formation and
lipid hydroperoxide concentration were then analyzed as previously described (108).
All dependent measures (antioxidant and biochemical parameters) were subjected
to a one-way analysis of variance (ANOVA). Significance was established at p<0.05. In
the case of significant differences, Scheffe post-hoc analysis was performed to determine
where differences existed. Bivariate correlations were performed between TBARS and
hydroperoxide levels and systemic, dietary and biochemical measures to determine any
relationships between lipid peroxidation and these variables. Furthermore, a stepwise
(forward) regression was performed on select variables to determine which variables
contribute most to lipid peroxidation in both models of obesity.
Due to the variance between the adiposity levels attained within experimental
groups, each of the three groups of animals were separated into two groups based on
adiposity: low BMI and high BMI. BMIs that were above the group average were defined
as "high" and BMIs that were below the group average were defined as "low". Therefore,
the following annotation will be used throughout the remainder of the manuscript:
CONTROLS: Low BMI= C-L-BMI
High BMI = C-H-BMI
HIGH-FAT FED: Low BMI = F-L-BMI
High BMI = F-H-BMI
OBESE: Low BMI = O-L-BMI
High BMI = O-H-BMI
Diet and Antioxidant Consumption
The weekly food intake of all groups during the nine weeks of feeding is shown in
Figure 3. The total caloric intake and dietary consumption of vitamins A and E are
contained in Table 4. Although the fat-fed groups F-L-BMI and F-H-BMI consumed less
total food (p<0.05), the caloric intake of these groups was higher compared to C-L-BMI
and C-H-BMI (p>0.05). This is due to the fact that the food density of the high-fat diet
- w ** ** **
** ** *
1 2 3 4 5 6 7 8 9
Figure 3. Food intake of the six experimental groups.
Values are means + SE. *Denotes greater than control
p<0.05, and ** denotes less than controls at p<0.05.
was 4.74 kcal/gm compared to the diet consumed by the two lean groups (3.84 kcal/gm).
The O-L-BMI and O-H-BMI animals consumed more total food and more calories than all
other control and fat-fed groups (p<0.05). Vitamin E and A intake were a direct function
of total food consumed (500 IU/ gm food vitamin E, 500,000 IU/ gm vitamin A).
Therefore, groups F-L-BMI and F-H-BMI consumed less of vitamin E and A compared to
the other four groups. In contrast, the obese groups O-L-BMI and O-H-BMI consumed
the most of these two vitamins of all six groups (p<0.05).
Table 4. Total diet consumption, caloric intake and antioxidant intake of lean control,
high-fat fed and obese Zucker rats during a 9-week feeding period. Values are means +
SEM. ** p<0.05 greater than groups 1-4, t p<0.05 less than all control and obese groups.
Group Diet Intake Calories Vitamin E (IU) Vitamin A (IU)
(g) (kcal) (IU, x 10') (IU, x 108)
C-L-BMI 1204+34 4922+85 6.02+0.17 6.02+0.17
C-H-BMI 1199+29 5013+83 6.0+0.14 6.0+0.14
F-L-BMI 1043 + 18t 5661 + 135* 5.21 + 0.09 5 21 + 0.09t
F-H-BMI 1062+ 18t 5682+ 163* 5.31+0.08t 5.31 +0.08t
O-L-BMI 1525+43** 7199+201** 7.62+0.21** 7.62+ 0.21**
O-H-BMI 1553 + 56** 7330 + 263** 776 + 0.28** 7.76 + 0.28**
where Vitamin E = vitamin E acetate, Vitamin A = Vitamin A Palmitate
Body Weight Changes With Feeding
Nine weeks of feeding the lean and obese Zucker rats resulted in a distinct
separation between the body weights of the groups. Figure 4 contains the body weights
data before, during and after the feeding treatment. The O-L-BMI and O-H-BMI were
heavier than the other four groups at the start of the study, and gained weight rapidly
during the first 4 weeks and were significantly heavier (p<0.05) than the other four groups
at all time points during the feeding period. By week 6, the F-H-BMI became significantly
(p<0.05) heavier than the control lean groups C-L-BMI and C-H-BMI and the fat-fed
group, F-L-BMI. These body weight separations remained present throughout the
remainder of the feeding period.
The morphological characteristics of the six groups are summarized in Table 5a.
The O-L-BMI and O-H-BMI groups were characterized by a significantly (p<0.05) larger
liver weights compared to all other groups. The heart weights were not significantly
different between groups (p>0.05), but the heart weight/body weight ratio was
significantly lower (p<0.05) in the groups O-L-BMI and O-H-BMI compared to all other
groups. The diaphragm weight of the F-H-BMI was significantly greater than the weights
of all other groups (p<0.05). The locomotor muscle weights are shown in Table 5b. The
data indicate that the locomotor muscle weights of the soleus, gastrocnemius,
m 300 I C-L-BMI
Pre 1 2 3 4 5 6 7 8 Post
Figure 4 Body weight ranges before, during and following the nine week
feeding period. Values are reans +SE denotes greater than control at
Table 5a. Comparison of morphological characteristics between the six groups of Zucker
rats following 9 weeks of feeding. Values are means + SE., p<-0.05 compared to all
other groups, **p<0.05 greater than all lean control and fat-fed groups, t p<0.05 less than
all lean control and fat-fed groups.
Heart Weight Heart/ Body Weight
(g) Ratio (g/kg)
1 06+0.03 2.91+0.12
1.08+ 0.08 2.59+ 0.05
1.10+0.08 2.21 + 0.06
1.12+0.05 2.53 +0.05+
8.92 + 0.40
9.99 + 1.23
10.54 + 0.44
22.11 + 1.51**
0.86 + 0.02
0.96 + 0.21
0.93 + 0.07
0.76 + 0.09
Table 5b. Comparison of locomotor muscle weights between the six groups of Zucker
rats following 9 weeks of feeding. Values are means + SE. p<0.05 greater than all other
groups, t p<0.05 smaller than all other groups. All values are expressed in g.
Soleus Gastrocnemius Plantaris Tibialis Anterior
C-L-BMI 0.169+0.006 1.42+0.19 0.33+0.02 0.711+0.02
C-H-BMI 0.151 + 0.014 1.39+0.14 0.31 +0.03 0.647+0.06
F-L-BMI 0.179+ 0.008 1.45+0.05 0.36+ 0.01* 0.726+0.02
F-H-BMI 0.184+0.006* 1.63+0.21* 0.37+0.01* 0.755+0.02*
O-L-BMI 0.112+0.003t 1.03+0.03t 0.21 +0.01 0.415 0.01t
O-H-BMI 0.119+ 0.003t 1.05 +0.03t 0.22+0.01t 0.432 0.01t
plantaris and the tibialis angterior muscles of the obese groups O-L-BMI and O-H-BMI
were significantly smaller (p<0.05) than those of their lean and high-fat fed counterparts in
all other groups. The muscle weights of group F-H-BMI were significantly greater
(p<0.05) compared to all other groups.
The physiological characteristics of the animals were recorded before, during and
following the feeding protocol. These measures included resting heart rates and systolic
blood pressures (double product ofHR x BP = heart work), oxygen consumption, BMI
and blood glucose and insulin concentrations.
Heart Rates. Blood Pressures, and Heart Work
The resting heart rates of the six groups did not differ (p>0.05) at any time point
during the study (Figure 5a). However, the systolic blood pressures of the obese animals
of groups O-L-BMI and O-H-BMI were significantly higher than those of all other groups
during weeks 1-2 (Figure 5b). By week three, the fat-fed groups F-L-BMI and F-H-BMI
exhibited a significantly higher (p<0.05) systolic blood pressure compared to groups C-L-
BMI and C-H-BMI. This difference between groups persisted throughout the remaining
weeks of the feeding period.
Pre 1 2 3 4 5
6 7 8 Post
Figure 5. Cardiovascular functional measures.
5a) Resting heart rates of all experimental
groups. Values are means + SE.
100 I I I I ,
Ib ^ s U
Figure 5. Cardiovascular functional measures.
5b) Systolic blood pressure in all experimental groups,
Values are means + SE. ** denotes different from fat-
fed and controls, *denotes different from controls.
Pre 2 3 4 6 7 8 Post45000
Figure 5. Cardiovascular functional measures.
5c) Myocardial work among the six groups. Values
are means + SE. Denotes greater than control
40000 groups at p<0.05, greater than fat-fed and
controls at p<0.05.F-L-BMI
Pre 1 2 3 4 5 6 7 8 Post
Figure 5. Cardiovascular functional measures.
5c) Myocardial work among the six groups. Values
are means + SE. Denotes greater than control
groups at p
The rate pressure products of the heart rates X systolic blood pressures (defined as
heart work) for each group during weekly assessments are shown in Figure 5c. The data
indicate that the heart work generated by the obese groups O-L-BMI and O-H-BMI was
significantly greater (p<0.05) than that of all other groups during weeks 1-7. By week 5,
the heart work generated by the fat-fed groups F-L-BMI and F-H-BMI was significantly
greater than that of lean control groups C-L-BMI and C-H-BMI. This difference between
the six groups existed throughout the remainder of the study.
Oxygen Consumption and Body Mass Index (BMI)
Resting oxygen consumption (V02) and BMI values are shown in Table 6. The C-
H-BMI group demonstrated a significantly higher (p<0.05) resting oxygen consumption
value compared to all other groups, whereas, groups O-L-BMI and O-H-BMI showed the
lowest Vo2 values compared to the four remaining groups (p<0.05).
The BMI values following nine weeks of feeding resulted in significant differences
(p<0.05) between the groups. The BMI values for O-H-BMI was greater (p<0.05) than all
other groups, whereas the BMI values for groups F-H-BMI and O-L-BMI were greater
(p<0.05) than groups C-L-BMI, C-H-BMI and F-L-BMI. C-L-BMI had the lowest
(p<0.05) BMI value of all six experimental groups.
Table 6. Resting oxygen consumption and body mass index values (BMI) for all
experimental groups of Zucker rats following 9 weeks of dietary treatment. Values are
means + SE. *p<0.05 lower than groups C-L-BMI, C-H-BMI, and F-L-BMI **p<0.05
greater than groups C-L-BMI and C-H-BMI, ***p<0.05 compared to all 5 other groups.
Group V02 (ml*kg*min) BMI (g/mm4)
C-L-BMI 29.62 + 2.88 276.51 + 5.70
C-H-BMI 35.46 + 2.42*** 308.25 + 6.50
F-L-BMI 29.85+ 1.78 30710 + 3.84**
F-H-BMI 28.1 + 2.36* 320.24 + 2.78**
O-L-BMI 24.11 + 2.52* 323.95 + 4.03**
O-H-BMI 22.18+ 1.30* 358.18+ 17.5**
Table 7. Blood glucose and insulin concentrations in all six experimental groups.
Hematocrit values are also provided. Values are means + SE. *p<0.05 greater than groups
C-L-BMI, C-H-BMI and F-L-BMI, **p<0.05 greater than all other 5 groups.
140.7 + 13
162.5 + 12
176.7 + 14
180.6 + 9
182.7 + 7*
3.44 + 0.23
3.30 + 0.25
13 67 + 0.65**
17.29 + 1.67**
39.5 + 0.86
40.2 + 0.36
41.3 + 0.99
37.7 + 1.89
37.4 + 1.38
Blood Glucose and Insulin Concentrations
The blood glucose and insulin concentrations are shown in Table 7. The data
indicate that groups F-H-BMI and O-H-BMI had higher (p<0.05) glucose levels than
groups C-L-BMI, C-H-BMI and F-L-BMI, and O-L-BMI had the highest concentration
(p<0.05) of blood glucose compared to all other groups. Insulin concentrations increased
as a function of adiposity. The O-L-BMI and O-H-BMI groups had significantly greater
(p<0.05) insulin concentrations compared to all other groups. There were no significant
differences (p>0.05) in the hematocrits between the six groups.
Heart Tissue Characteristics
Heart tissue characteristics are shown in Table 8. Although trends existed, the
water content and dry weights of the heart samples were not significantly different
between groups. In contrast, the myocardial lipid content values for F-L-BMI, F-H-BMI,
O-L-BMI and O-H-BMI were significantly greater (p<0.05) compared to the two groups
of lean control animals. However, there were no significant differences in lipid content
among the four fat-fed and obese groups.
Table 8. Myocardial tissue water content, dry weight and lipid content (mg/g) of the six
experimental groups. Values are means + SE. *p<0.05 greater compared to C-L-BMI and
Group Water (%) Dry Mass (%) Lipid (mg/g)
C-L-BMI 79.72+ 0.67 20.27+ 0.67 21.55+5.06
C-H-BMI 79.21 +0.53 20.78+0.53 18.67+ 2.47
F-L-BMI 79.13 + 0.26 20.86 + 0.26 39.76 + 4.71*
F-H-BMI 78.57+0.19 21.42+0.19 42.17+5.75*
O-L-BMI 78.3 + 0.97 21.68+0.97 35.89+ 5.41*
O-H-BMI 77.02 + 0.92 22.98 + 0.92 42.93 + 6.17*
0' Production: Cvtochrome C Assay
The production of 02 production by isolated papillary muscles in vitro, was
determined using a cytochrome C assay. The results of this assay are shown in Figure 6.
Electrolysis and bubbling of the O2/CO2 gas mixture resulted in minimal 02.- production as
evidenced by the low absorbance. High absorption values were recorded following
incubation of the tissue bathing medium with hypoxanthine and xanthine oxidase,
indicating the viability of the assay. Purified SOD was added to the hypoxanthine/xanthine
oxidase bathing medium and inhibited 02.- production as shown by the reduced
absorbance value. Further, SOD was also added to the bathing medium containing
contracting papillary muscle. The low absorbance value indicates that SOD did inhibit 02.-
reduction of cytochrome C. However, there were not detectable differences (p>0.05) in
--o c,. A o o 'o
.. I I 6 I
Figure Cytodirorr Creduction in response to expeinental
conditions and contracting isolated papillay muscle fiomthe six
Values are reans + SE
bathing medium absorbance values between the muscle preparations from all six
experimental groups, indicating no differences in 02 production. To demonstrate viability
of these muscle preparations, tetanic muscle forces were recorded. The average
contractile forces for all groups ranged from 4-6 mN/mm2.
Oxidative and Antioxidant Enzyme Activities
The oxidative (CS) and antioxidant enzyme activities (CAT and GPX) for left
ventricular samples are shown in Table 9a. Although a trend existed, there were no
significant differences in CS or GPX activity. However, the O-H-BMI group has higher
CAT activity compared to all other groups. Interestingly, the only significant difference in
the SOD activities was found with the CuZn-SOD isoform. Specifically, the CuZn-SOD
activity was greater in the O-L-BMI and O-H-BMI groups compared to the lean controls,
L-L-BMI and L-H-BMI. There were no other significant changes in SOD activities with
either model of obesity.
When antioxidant enzyme activities were normalized to the amount of myocardial
lipid, the antioxidant enzyme profile was quite different. The results of this analysis are
shown in Table 9c. In all cases, the lean, control groups (C-L-BMI and C-H-BMI) had
greater enzyme activities compared with all the high-fat fed and obese groups.
Table 9a. Oxidative and antioxidant enzyme activities of left ventricular samples from all
six experimental groups. Values are means + SE. CS and GPX units are in Pmol*mg
protein*min; CAT units are in U/gww. *p<0.05 greater compared to C-L-BMI and F-L-
82.6 + 1.1
128.3 + 9.9
148.2 + 10.8
133.5 + 13.2
Table 9b. SOD activities of left ventricular samples from all six experimental groups.
Values are means SE. SOD is expressed as Units*min*mg protein. *greater than all fat-
fed and control groups at p<0.05.
101 4 +6.2
26 8 + 1.7
Table 9c. Antioxidant enzyme activities of left ventricular samples from all six
experimental groups. Values are means + SE. CS and GPX units are in pImol*mg
lipid*min; CAT units are in U/mg lipid. *p<0.05 greater compared to all high-fat fed and
Group GPX CAT T-SOD Mn-SOD CuZn-SOD
C-L-BMI 8.1 1.7* 3.9 1.1* 6.6 1.4* 4.9 1.1* 1.70.32
C-H-BMI 7.60.73* 3.70.50* 6.3 1.1* 4.3 0.70* 1.90.45*
F-L-BMI 3.2 0.72 1.6 0.37 2.6 0.32 2.1 0.49 0.7 0.12
F-H-BMI 3.2 0.54 1,7 + 0.24 2.8 0.45 1.7 0.23 1.1 +0.29
O-L-BMI 4.4 0.79 1.9 0.22 3.7 0.44 2.2 0.22 1.5 0.23
O-H-BMI 2.8 0.43 1.7 0.49 2.8 0.35 1.7 0.28 1.10.13
Myocardial thiol status was measured at the completion of the 9-week feeding
period (Figure 7). No significant differences existed (p>0.05) between groups in total thiol
content. Compared to C-L-BMI and C-H-BMI, F-L-BMI and F-H-BMI had significantly
lower protein thiols. F-H-BMI and O-H-BMI had greater levels (p<0.05) of non-protein
bound thiols compared to all other groups, suggesting that glutathione levels were
elevated in these groups.
Total Thiols Protein Thiols Non-protein Thiols
Figure 7. Myocardial thiol fractions. Values are means
+ SE. Denotes greater than all other groups at p<0.05,
** less than control groups at p<0.05.
Basal Lipid Peroxidation
Left ventricular hydroperoxide content is shown in Figure 8. The data indicate that
the F-L-BMI, F-H-BMI, O-L-BMI and O-H-BMI groups had significantly (p<0.05)
higher tissue hydroperoxide levels compared to C-L-BMI and C-H-BMI. No differences
existed between F-L-BMI, F-H-BMI, O-L-BMI and O-H-BMI.
Oxidative Challenges in Vitro
The results for the oxidative challenges are shown in Figure 9a and 9b. When
expressed per mg of protein (Figure 9a), the TBARS concentration was significantly
greater (p<0.05) in F-L-BMI, F-H-BMI, and O-H-BMI compared to all other groups at
the basal level. Following the FeCI3 challenge, the H202 challenge, and the xanthine/
xanthine oxidase challenge, the TBARS concentration was greater in F-H-BMI, O-L-BMI
and O-H-BMI groups. There were no differences (p>0.05) between groups following the
AAPH challenge. When expressed per mg lipid (Figure 9b), TBARS levels did not differ
between any group with the exception of the AAPH challenge. Specifically, the TBARS
levels of groups F-HBMI, O-LBMI and O-HBMI were significantly lower (p<0.05) than
those in C-LBMI and C-HBMI, suggesting that lipid peroxidation is dependent on the
amount of lipid substrate present for oxidation.
C-L-BMI C-H-BMI F-L-BMI F-H-BMI O-L-BMI O-H-BMI
Figure 8. Myocardial lipid hydroperoxide content in
all six experimental groups. Values are means + SE.
*Denotes greater than control groups at p<0.05.
1.2 0 Basal
Fie 9a. Myocardial TARS content.
C-L-BMI C-H-BMI F-L-BMI F-H-BMI O-L-BMI O-H-BMI
Figure 9a. Myocardial TBARS content.
Myocardial TBARS/ mg protein Values
are means + SE. *greater than controls and
F-L-BMI at p<0.05.
CL-BM C-BI F-BM F--VM OL.-l O4BM
Figure 9b. Myocadial TARS content.
Myocardial TBARS/ ng lipid Values are rrans
+ SE *less than controls at pO.5
Correlations Between Lipid Hvdroperoxides and Physiologic Measures
Pearson correlations were performed between lipid hydroperoxides and the
proposed radical-generating biochemical and systemic measures to determine which
measures best correlate with myocardial lipid peroxidation (Table 11). Lipid
hydroperoxide content is correlated with (systolic blood pressure and) heart work and
myocardial lipid content. Considering that these same variables were also significantly
correlated with myocardial TBARS content (data not shown), we suggest that heart work
and blood pressure may be good non-invasive predictors of myocardial oxidative stress.
Stepwise Regression Model for Mvocardial Lipid Peroxidation
A stepwise regression was performed to determine those variables which are most
closely related to the magnitude of myocardial lipid peroxidation (hydroperoxides/ mg
lipid). The results are found in Table 12. These data suggest that lipid content contributes
most to the myocardial lipid peroxidation observed in both obesity models (regression
equation: y=-0.0096X + 0.731). In addition to lipid content, heart work, cytochrome C
reduction (evidence of excessive superoxide production), CuZn-SOD and CAT activities
contributed to the regression model, although their individual contributions were not
significant. To illustrate the relationship between lipid hydroperoxides/ mg lipid and
myocardial lipid content, a scatter plot was created (Figure 10). To test for adequacy of
Table 11. Correlations (r) between basal lipid hydroperoxides/ mg lipid and selected
systemic measures in all groups and separated by obesity model. r values are shown with
p-values in parentheses. denotes p<0.05
MEASURE ALL GROUPS HIGH-FAT FED fa/fa
BMI 0.282 (0.064) 0.210 (0.472) 0.290 (0.295)
HEART WORK 0354 (0 034)* 0.217 (0.521) 0.587 (0.035)*
LIPID CONTENT 0.867 (0.0001)* 0.915 (0.0001)* 0.784 (0.001)*
VITAMIN E 0.008 (0.957) 0.538 (0.047)* 0.136 (0.628)
CS 0.313 (0.046)* 0.391 (0.167) 0.466 (0.080)*
TOTAL SOD -0.068 (0.674) -0.419 (0.136) -0.083 (0.767)
Mn-SOD 0.168 (0.293) 0.098 (0.737) 0.057 (0.840)
CuZn-SOD 0.170 (0.288) -0.322 (0.261) -0.140 (0.619)
GPX -0.098 (0.542) -0.052 (0.860) -0.192 (0.493)
CAT -0.109 (0.510) -0.267 (0.356) -0.560 (0.037)*
CYTOCHROME C 0.144 (0.524) 0.156 (0.738) 0.282 (0.499)
Table 12. Stepwise regression analysis (forward) for obesity-induced myocardial lipid
peroxidation (hydroperoxides mg/lipid). Each step is additive, representing 5 different
equations. denotes a significant contribution to the model
Step Variable r R2 P-value
1 Lipid Content 0.87 0.75* 0002*
2 Heart Work 0.35 0.76 0.434
3 Cyt C Reduction 0.14 0.78 0.968
4 CuZn-SOD 0.17 0.80 0.653
5 CAT 0.11 0.84 0.237
y = -0.0096x + 0.731
20 40 60 8(
Myocardial Lipid (mglg)
Figure 10 Scatter plot of the relationship between myocardial
lipid and hydroperoxide content (mg lipid).
the model fit, the normal probability plots of the residuals, the standardized residuals and
the coefficient of determination (R2) indicate the model is adequate to predict myocardial
lipid peroxidation. All residual plots are found in Appendix C
Overview of Principal Findings
This study investigated the mechanisms underlying myocardial oxidative stress in
the Zucker rat. In our preliminary experiment, we found that obese Zuckers had elevated
myocardial lipid peroxidation; to determine whether this was due to obesity per se or the
leptin receptor defect (fa/fa), we examined lipid peroxidation in high-fat fed (Fa/?) and
obese (fa/fa) animals. This study tested two hypotheses. First, we postulated that the
high-fat fed (Fa/?) animals and the obese (fa/fa) animals would have similar levels of
myocardial lipid peroxidation and that lipid peroxidation is highly correlated with the level
of adiposity. Second, we postulated that several factors could contribute to myocardial
oxidative stress in obesity, including: a) higher myocardial work, b) a compromised
antioxidant defense, c) higher myocardial lipid content, and d) increased superoxide anion
These data partially support hypothesis #1, that the two obesity models are
associated with elevated levels of lipid peroxidation. Indeed, the level of myocardial lipid
peroxidation, (hydroperoxides) was significantly correlated (p<0.05) with the level of
Hypothesis #2 was only partly supported by these data. Specifically, heart work
(systolic blood pressure X heart rate) and lipid content of the myocardium from the high
fat-fed (Fa/?) and obese (fa/fa) groups were greater (p<0.05) compared to the lean control
groups. Glutathione content, however, was increased in the F-H-BMI sand O-H-BMI
groups compared to lean controls. Furthermore, antioxidant enzyme activities (CAT,
CuZn-SOD) were elevated in the O-L-BMI and O-H-BMI groups compared to all other
groups. Lastly, there were no differences existed in the superoxide production by isolated
contracting papillary muscles in vitro. The following sections will discuss each of these
Lipid Peroxidation in Mvocardial Tissue of Obese Animals
A previous descriptive study from our laboratory indicated that genetically obese
Zucker animals (fa/fa) contained higher myocardial levels of TBARS and lipid
hydroperoxides compared to their age-matched lean counterparts (111). The current data
are the first to comprehensively investigate the potential sources of this lipid damage, and
possible mechanisms for increased susceptibility to myocardial oxidative stress using two
different models of obesity. Other reports indicate that lipid peroxidation is elevated in
other tissues such as liver and plasma in humans and obese animals (23, 24, 62).
Furthermore, diets enriched in fats (either saturated or unsaturated) are also associated
with increased lipid peroxidation in several tissues such as the myocardium, aorta, liver
and plasma (62, 69, 107). Collectively, these data indicate that high-fat feeding or
expression of the fa/fa gene is indeed associated with increased oxidative stress in various
We attempted to address the issue of whether lipid peroxidation was due to high-
fat feeding or the leptin receptor defect by using a unique approach We investigated
animals that were developing obesity naturally as a consequence of their geneotype (fa/fa),
and we used cohorts of their lean counterparts (Fa/?) as the lean controls and the high-fat
fed groups. Over the course of the nine-week feeding period, individual animals developed
adiposity at different rates. By separating animals based on adiposity and genotype at the
conclusion of the study, we were able to examine the effects of the high-fat diet and the
leptin receptor defect on myocardial lipid peroxidation.
Basal lipid peroxidation products, (i.e., TBARS and hydroperoxides), were
elevated in the high-fat fed (F-H-BMI) and obese (O-L-BMI, O-H-BMI) groups. Post-
oxidative challenge TBARS values were elevated in F-H-BMI, O-L-BMI and O-H-BMI
groups. Interestingly, when expressed as TBARS/ mg tissue lipid, these differences
between groups disappeared, indicating the importance of increased lipid substrate target
for oxidation in the myocardium (this issue is discussed further in a subsequent section).
Considering the finding that hydroperoxide levels did not differ (p>0.05) between
the high-fat fed and obese groups, this suggests that the leptin receptor defect (fa/fa) is not
responsible for myocardial oxidative stress in obesity. In the following sections, we discuss
the contribution of four potential major pathways that could contribute to the elevated
lipid peroxidation in obesity
Potential Pathways for Obesity-Induced Oxidative Stress
We examined several systemic variables and biochemical parameters which could
contribute to myocardial oxidative stress in obesity. These measures included heart work
(double product: heart rate X systolic blood pressure), glutathione and antioxidant enzyme
activities, lipid content of the myocardial tissue, and superoxide anion production by
isolated papillary muscles. Each of these potential pathways and their relationship to
myocardial lipid peroxidation is examined in the following paragraphs.
Elevated Heart Work
The heart work estimate (rate pressure product = heart rate X systolic blood
pressure) was highest in the genetically obese groups (O-L-BMI, O-H-BMI), high in the
high-fat fed groups (F-L-BMI, F-H-BMI), and lowest in the controls (C-L-BMI, C-H-
BMI). Also, increased afterload (hypertension) was present in both models of obesity, and
the heart/body weight ratios of the O-L-BMI and O-H-BMI groups were lower than all
other groups. Although heart work was significantly correlated with both measures of
lipid peroxidation (hydroperoxides/ mg protein r=0.35, p=0.034, and TBARS levels
r=0.39, p=0.02), the regression analysis suggests that the overall contribution to the to
obesity-induced lipid peroxidation is not large (Table 12).
It is well established that in skeletal and heart muscle that elevated muscle work
(such as exercise, or some mechanical overload) is associated with increased free radical
production (49). Exercise or overload-induced increased oxygen consumption increases
the electron flux through the mitochondria in proportion to the overload, thus increasing
the risk for electron leakage in the electron transport chain. The result is excessive
production of superoxide anions or hydrogen peroxide (114). Excessive radical formation
can trigger a cascade of reactions that result in lipid peroxidation (34).
What is an explanation for the failure of the increasing increments in heart work to
proportionately increase lipid peroxidation between the high-fat fed and fa/fa animals?
There are two possibilities. First, it is possible that the obese animals were responding to
the (hypertension-induced) elevated heart work by improving elements of the myocardial
antioxidant defense. Even small elevations in the antioxidant capacity may provide enough
protection in the heart to remove reactive oxygen species generated by the hypertension-
induced elevations in heart work. Second, we monitored the physiologic variables such as
blood pressure in these animals at rest. It is possible that at rest, the workload on the heart
was not the level required to result in a significant increase in oxidant protection.
In summary, these data do not support the hypothesis that elevated heart work is a
major contributor to elevated lipid peroxidation in young adult high-fat fed animals (Fa/?)
and the obese fa/fa animals.
Compromised Antioxidant Defense
We hypothesized that a compromised myocardial antioxidant defense in obesity
was a potential mechanism to explain the elevated myocardial lipid peroxidation. Previous
investigations have reported obesity-related lower plasma or tissue levels of antioxidants
and/or increased susceptibility to oxidative challenges in vitro (24, 92, 93, 111).
This experiment demonstrated that glutathione levels (GSH, estimated by the non-
protein thiol fraction) were increased in hearts of the F-H-BMI and O-H-BMI groups
compared to the lean control groups. Furthemore, myocardial CAT activity was elevated
in the O-H-BMI group, and CuZn-SOD activity was elevated in both the O-L-BMI and 0-
H-BMI groups compared to the lean control groups. Hence, the antioxidant capacity of
the myocardium from obese animals does not appear to be depressed.
The elevations in the myocardial antioxidant defense in obesity suggest that these
hearts were exposed to greater radical production compared to the lean controls. It is well
established that the antioxidant enzyme activities and GSH concentrations increase in
response to free radical formation in an effort to protect the myocyte against subsequent
oxidative damage (49). The increased activities of SOD and CAT suggest that there was
excessive production of their substrates, superoxide and hydroperoxides respectively
(104), within the ventricular tissue of the O-L-BMI and O-H-BMI groups.
The adaptation of the primary antioxidant defense in the hearts of high-fat fed
animals (F-L-BMI, F-H-BMI) appeared to be incomplete, as indicated by the failure of any
antioxidant enzyme activity to increase with the high-fat diet. The failure of these groups
to demonstrate significant antioxidant enzyme up-regulation may be related to the
elevations in GSH in the F-H-BMI group. It is possible that the higher GSH content was
sufficient to suppress the signals necessary for antioxidant enzyme up-regulation in the
hearts of the fat-fed animals. For example, McDuffee et al. (1997) reported that oxidized
proteins containing non-native disulfide bonds are products ofoxidative stress, and can act
as signals for up-regulation of the tertiary antioxidant defense. It is possible then, that the
elevations in GSH in the F-H-BMI group could sufficiently reduce the oxidation of
proteins and lipids alike that may serve as stimuli for antioxidant enzyme up-regulation.
The oxidative challenge results (Figures 9a-b) indicated that the F-H-BMI, O-L-
BMI and O-H-BMI had similar levels of TBARS/mg protein following exposure to H202,
02 and OH radicals in vitro. The O-L-BMI and O-H-BMI groups had lower (p<0.05)
TBARS levels following the xanthine/ xanthine oxidase challenge superoxidee generating
system) compared to the other challenges (Figure 9a). This could suggest that the
elevations in CuZn-SOD in these groups were sufficient to protect against lipid
peroxidation following exposure to exogenous superoxide. Furthermore, the H202
generating system produced more TBARS in the F-H-BMI, O-L-BMI and O-H-BMI
groups compared to the remaining groups. This finding may be a consequence of the lack
of adaptation of GPX, as H202 is a substrate for GPX (114).
In summary, the current data do not support the hypothesis that the antioxidant
defense was insufficient; rather, antioxidant enzyme activities and GSH were up-regulated
in response to obesity. Furthermore, the data indicate that the antioxidant adaptations
were not a function of the leptin receptor defect (fa/fa), but a response to obesity per se.
Elevated Lipid Content
We also tested the hypothesis that the myocardial tissue from high-fat fed and fa/fa
animals contains higher lipid levels compared to their lean counterparts. Our data support
this hypothesis. Indeed, myocardial tissue obtained from the left ventricles of all high-fat
fed and fa/fa animals did indeed contain more lipid (p<0.05) compared to myocardial
tissue obtained from both groups of control animals (Table 8). Obesity due to both high-
fat feeding and development of obesity in the fa/fa genotype appeared to promote similar
deposition of fat into the myocardial tissue in the two obesity models (Table 8).
Several investigators have reported that lipid peroxidation is elevated in tissues
from fatty Zucker rats (fa/fa genotype) and in high-fat fed animals (62, 69, 107). We
previously reported that 12 month old fa/fa Zucker rats had a -30% greater myocardial
lipid content compared to their lean (Fa/?) counterparts (111). A potential mechanism for
increased lipid damage in obesity is that increased lipid substrate within the myocardium
can function as a larger target for oxidation by free radicals (63). Increasing the number of
lipid molecules within the cardiovascular system (within the cardiac cells and embedded
within the coronary vasculature intimal layers) may amplify lipid peroxidation injury (34).
Because we measured lipid peroxidation products from left ventrcular tissue
homogenates, it is likely that the lipid peroxidation reflects the combination of the
peroxidation products of both the myocytes and the vasculature. In the high-fat feeding
model, we surmise that the elevated lipid content is due to fat deposition and storage
within the myocytes (46, 111) and fat deposition onto the coronary endothelium (69, 107).
Previous experiments report that lipid peroxidation products are found within
atherosclerotic plaques from cardiovascular and/or obese patients, and within cardiac
tissue from high-fat fed animals (90). Both TBARS and lipid hydroperoxides/ mg lipid
were correlated with tissue lipid content (r=0.431 and r=0.760, p<0.05, respectively).
Lastly, the series of oxidative challenges revealed that the degree of lipid
peroxidation following the H202, FeC13 and xanthine/xanthine oxidase challenges was
related to amount of myocardial lipid. Based on these data, and the high correlation
between myocardial lipid content and lipid peroxidation products, it seems likely that
myocardial lipid content is an important contributor to myocardial lipid peroxidation in
both high-fat fed animals and fa/fa animals. Further, it appears that lipid peroxidation is a
function of obesity per se (consumption and deposition), and not a function of the leptin
Superoxide Radical Production by Isolated Papillary Muscles
We hypothesized that contracting papillary muscles from obese animals will
produce superoxide anions at a greater rate compared to lean animals. Our data reveal that
superoxide formation by isolated papillary muscles in vitro does not differ across our
groups. There are two possibilities to explain this result.
First, it is possible that the antioxidant defense within the papillary muscles of the
heart adapted in response to the obesity overload in the obese models. Given that the
antioxidant defense increased in proportion to the overload, if excessive formation of
superoxide did occur in isolated papillary muscles during the stimulation protocol in vitro,
the majority of the radicals may have been scavenged and dismutated by endogenous SOD
before diffusion into the tissue bath. This is a likely possibility considering CuZn-SOD was
elevated in the ventricular tissue of the O-L-BMI and O-H-BMI groups, and these groups
exhibited a protection against the superoxide challenge in vitro.
Second, our findings do not preclude the possibility that ventricular tissue does
produce superoxide at a greater rate. Although isolated papillary muscles are widely used
as a model for cardiac contractility and performance (15, 56), there still may be enough of
a metabolic difference between the papillary muscle and ventricle that precludes significant
differences from being detected using our in vitro technique. Simple experiments
comparing 02' formation in the working isolated whole heart and the papillary muscle
could be performed to determine if this is the case.
In summary, these data suggest that a greater rate of superoxide production is not
a major contributor to the elevated lipid peroxidation in ventricular tissue in the high-fat
fed or fa/fa animals.
Obesity that results from high-fat feeding and the leptin receptor defect (fa/fa) is
associated with elevated levels of lipid peroxidation. The level of myocardial lipid
peroxidation (hydroperoxides) was significantly correlated with the level of adiposity
(BMI) and lipid content, regardless of genotype (fa/fa or Fa/?). In contrast, elevated heart
work (systolic blood pressure X heart rate), insufficient antioxidant defenses, and
increased rate ofsuperoxide formation were not significant contributors to obesity-
induced myocardial lipid peroxidation. Hence, it seems likely that myocardial lipid
peroxidation is primarily due to obesity per se and not the leptin receptor defect (fa/fa
Chronically elevated levels of lipid peroxidation by-products could indicate that the
myocardium is less able to combat oxidative species and is more likely to sustain oxidative
injury (28, 41). For example, oxidative tissue injury was elevated in the myocardial tissue
of the high-fat fed and obese animals following several oxidative challenges in vitro. We
speculate that the heart tissues from these groups are less able to defend against oxidative
species generated in physiological scenarios such as ischemia-reperfusion and acute
exhaustive exercise. Furthermore, the elevated lipid by-products also trigger signal
transduction pathways that lead to apoptotic death or chemotaxis of tissue-devouring
macrophages (8, 34). The net result is increased risk for tissue damage during the
physiological stress and a reduced ability to repair itself and restore normal contractile
Limitations to the Experiment and Future Directions
A limitation to this experiment was the difficulty in attaining the same degree of
adiposity between the high-fat fed (Fa/?) and obese (fa/fa) groups. The high-fat fed
animals controlled their dietary intake based on the caloric density of the food, such that
they consumed smaller volumes of the richer diet compared to the control diet fed lean
control animals. This delayed the accruement of body fat in the high-fat fed animals
(Figure 2). Even at the conclusion of the study, there were a few obesity "resistant"
animals that did not gain significantly higher fat than the C-H-BMI animals. This made it
difficult to determine the effects of diet-induced obesity and genetic obesity on myocardial
lipid peroxidation. Separation of the groups based on low and high BMI was required to
fully examine the relationships between systemic and biochemical variables and lipid
A second limitation of this study was lack of a definite conclusion we could reach
with regard to lipid peroxidation and the rate of ventricular superoxide formation based on
the indirect measurement from isolated papillary muscle. It is currently unknown whether
radical formation and detection in vitro differ between papillary muscle and ventricle
tissue. To our knowledge, this study was the first to employ this new superoxide anion
detection technique in the isolated papillary muscle.
There are several possibilities for subsequent experiments. First, it is unknown
whether obesity increases lipid peroxidation in response to an oxidative challenge (i.e., an
acute bout of exercise) or an ischemic challenge in vivo. The data that could be obtained
from these experiments could provide information about the possible functional
consequences of obesity on the ability of the heart to withstand and ischemic-reperfusion
injury. Second, it is unclear how aging affects the myocardial lipid peroxidation profile in
obese animals. Our first experiment indicted that 12 month old animals had higher levels of
lipid peroxidation despite the normalization to myocardial lipid content, perhaps
suggesting that aging can increase the rate of radical formation, reduce the antioxidant
defense or increase the lipid deposition within the myocardium. These research questions
are testable and warrant further investigation. Lastly, it is unknown whether antioxidant
supplementation can reduce the myocardial lipid peroxidation in obese animals. Simple
feeding experiments can be conducted to determine the potential effects of various
antioxidants on lipid peroxidation and heart performance characteristics either in vitro or
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EAC21WV42_XIB5IL INGEST_TIME 2013-09-28T02:33:53Z PACKAGE AA00014296_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC