THE ASSOCIATION OF SERUM BIOMARKERS WITH CARDIAC HEALTH IN CAPTIVE GORILLAS
A thesis submitted to Kent State University in partial fulfillment of the requirements for the degree of Master of Arts
by
Eric E. Henthorn
August, 2010
Thesis written by Eric E. Henthorn B.S., Kent State University, 2004 M.A., Kent State University, 2010
Approved by:
______Dr. Mary Ann Raghanti Advisor
______Dr. Richard Meindl Chair, Department of Anthropology
______Dr. Timothy Moerland Dean, College of Arts and Sciences
ii
TABLE OF CONTENTS
AKNOWLEDGEMENTS...... viii
ABSTRACT……………………………………………………………………….1
Chapter I. INTRODUCTION ...... 3
Gorilla Status and Conservation ...... 4 Heart Disease ...... 6 G. g. gorilla Heart Disease ...... 6 The Diet/Activity Level Connection ...... 7 A Natural Gorilla Diet ...... 8 Captivity and Captive Gorilla Diets ...... 8 Adipose, Obesity and Thrifty Genotypes...... 10 Metabolic Syndrome ...... 13 Glucose and Insulin...... 14 Indices of Insulin Resistance and Sensitivity ...... 15 Leptin ...... 16 Cholesterol ...... 17 Oxidized LDL ...... 18 Prolactin ...... 19 Ferritin...... 20 Research Questions and Predictions ...... 21
II. MATERIALS AND METHODS ...... 23
Echocardiograph Measures ...... 23 Enzyme Immunoassays (EIAs) ...... 24 Insulin Sensitivity ...... 25 Statistical Analyses ...... 25
III. RESULTS ...... 27
EIA Validation ...... 27 Biomarker and Echocardiograph Values ...... 27 Statistical Analyses ...... 30 PCA 1 ...... 30
iii
PCA 2 ...... 32 PCA 3 ...... 34 PCA 4 ...... 36 Correlation Analyses ...... 38 Correlation Analysis 1 ...... 38 Correlation Analysis 2 ...... 46 Correlation Analysis 3 ...... 50 Correlation Analysis 4 ...... 54
IV. DISCUSSION ...... 58
Principal findings ...... 58 Implications...... 59 Great Ape Heart Disease ...... 64 Confounding Factors ...... 66 Assessment ...... 66 Study Limitations ...... 67 Future Directions ...... 67
V. CONCLUSIONS ...... 69
LITERATURE CITED ...... 70
APPENDIX A...... 83
APPENDIX B...... 84
iv
LIST OF FIGURES
1. PCA 1 scree plot ...... 31
2. PCA 2 scree plot ...... 33
3. PCA 3 scree plot ...... 35
4. PCA 4 scree plot ...... 37
5. LogEF against age ...... 39
6. LogCholesterol against age ...... 40
7. LogQUICKI against age ...... 40
8. LogLeptin against age ...... 41
9. LogFerritin against age ...... 41
10. LogLeptin and LogCholesterol ...... 42
11. LogLeptin and LogQUICKI ...... 42
12. LogLeptin and LogFerritin ...... 43
13. LogEF and LogQUICKI controlling for age ...... 43
14. LogQUICKI and LogLeptin controlling for age ...... 44
15. LogQUICKI and LogFerritin controlling for age ...... 44
16. LogQUICKI and LogCholesterol controlling for age ...... 45
17. LogProlactin and LogCholesterol controlling for age ...... 45
18. LogCholesterol and LogFerritin controlling for age ...... 46
19. LogIVS and age ...... 45
v
20. LogIVS and LogFerritin ...... 45
21. LogIVS and LogCholesterol ...... 46
22. LogIVS and LogQUICKI controlling for age ...... 46
23. LogIVS and LogProlactin controlling for age ...... 47
24. LogLVPW and age ...... 49
25. LogLVPW and LogFerritin controlling for age ...... 49
26. LogLVPW and LogProlactin controlling for age ...... 50
27. LogFerritin and LogoxLDL controlling for age ...... 50
28. LogLVIDd and age ...... 53
29. LogLVIDd and LogProlactin controlling for age ...... 53
30. LogLVIDd and LogQUICKI controlling for age ...... 54
vi
LIST OF TABLES
1. Descriptive statistics of biomarkers for gorillas with heart disease ...... 28
2. Descriptive statistics of biomarkers for gorillas without heart disease ...... 28
3. Descriptive statistics of echocardiograph measures for gorillas with heart disease ...... 29
4. Descriptive statistics of echocardiograph measures for gorillas without heart disease ...... 29
5. PCA 1 solution ...... 31
6. PCA 2 solution ...... 33
7. PCA 3 solution ...... 35
8. PCA 4 solution ...... 37
9. Correlation analysis 1 ...... 39
10. Correlation analysis 2 ...... 47
11. Correlation analysis 3 ...... 51
12. Correlation analysis 4 ...... 55
vii
ACKNOWLEDGMENTS
This project would have not been possible without the personal and professional support of many individuals. I am a fortunate to have made so many close relationships both outside and inside the Kent State Anthropology Department. I wish to express my deepest gratitude to you all that have helped along this journey.
A big thank you goes to my patient advisor, Dr. Mary Ann Raghanti. From the moment I walked into your office you made me feel welcomed and supported in pursuing a topic that I had very little background in. I will never forget the calming chit chat you gave to me when I was convinced that I completely fouled up the samples. “No, Eric, the samples are fine, you are fine. Get a grip.” I am glad I had your strength to grip on to.
You are a friend and it is my deepest wish to continue our collaboration in future projects. Oh, by the way, this is for the sake of the children.
I would be remiss if I did not make a special thank you to Dr. Marilyn Norconk. If it were not for you I would not be here. You lit a fire in my belly and nurtured a desire to pursue a career in academia. As always, I hope and expect to work with you in the years to come. To Dr. Richard Meindl, a statistician’s statistician; I have garnered the upmost respect for you and hope to collaborate on future endeavors. Don’t worry, the ‘bubba vote’ will only make our jobs more interesting. Also, a special thank you is due to Dr.
Pam Dennis. Thank you for reminding me the world does not revolve around ferritin. It would be my privilege to work with you in the future as this project undoubtedly grows.
viii
To my family, friends and colleagues, past and present; there simply is not enough beer in this world to say ‘thank you’ enough. I would have never been able to accomplish this without my father’s support. You do inspire. How could I forget Hootie,
Phil Peachock and family, Richard and Janice John, Jo Ann Cawley, and of course “The
Square”. Keep your eyes on the prize! Thank you!
ix
ABSTRACT
Heart Disease has been cited as the foremost cause of captive adult male gorilla
deaths. Given that captivity does not offer gorillas the same level of physical activity and
diet as their wild counterparts, one hypothesis for captive gorilla heart disease etiology is that akin to metabolic syndrome-related heart disease in humans. To address this; hormonal data on leptin, insulin, glucose, cholesterol, oxidized low-density lipoprotein, prolactin, and ferritin serum concentrations were collected with concurrent echocardiograph data [ejection fraction (EF), interventricular septum thickness (IVS), left ventricle posterior wall thickness (LVPW), and left ventricle internal diameter at diastole
(LVIDd)] to assess any association between these biomarkers and cardiac health when
controlling for age. The Quantitative Insulin Sensitivity Check Index was used as a proxy
measurement of insulin sensitivity. Diagnoses were provided for 30 individual male
gorillas with 11 diagnosed as without heart disease and 19 diagnosed with heart disease.
Results demonstrate that any association between these hormones and echocardiograph
parameters is strongly mediated by age. Unpredictably, results also indicate insulin
sensitivity is negatively associated with EF. Insulin values probably underscore that
relationship. This would suggest that EF may be preserved by the upregulation of glucose
metabolism over beta-oxidation in cardiac cells. Although not significant, a negative
associational trend between prolactin and LVIDd might also indicate another
compensatory pathway against further cardiac dilation. If so, cardiac disease progression
1 2
in this sample may be well underway despite diagnoses and age. Given the complexity of both, the relationships between these hormones and echocardiographic data, and among the hormones themselves, further analysis is needed to assess the relationships between captive gorilla hormonal profiles and cardiac health along age graded classes.
Determinations of baseline insulin sensitivity also need further refinement to provide better context when exploring captive gorilla cardiac disease.
CHAPTER 1
INTRODUCTION
Heart disease is often cited as the foremost cause of captive male gorilla
morbidity and mortality at Association of Zoos and Aquariums (AZA) institutions
(Cousins, 1983; Meehan and Lowenstein, 1994). One hypothesis accounting for the
incidence of captive gorilla cardiac disease is that their disease profile may result from
etiologies similar to what is found in western human societies (Cousins, 1983; Popovich
and Dierenfeld, 1997; Ryder, 2005). Obesity has become epidemic in western societies
and is associated with a highly sedentary lifestyle coupled with a high-calorie diet
(Blundell and Cooling, 2000; Hill and Peters, 2002; Cefalu, 2008). This has resulted in the prevalence of metabolic syndrome and impaired insulin sensitivity, both factors that contribute to heart disease as predicated by abnormal levels of cholesterol, oxidized low density lipoproteins (oxLDL), leptin, insulin, glucose, prolactin, and ferritin (Djousse and
Gaziano, 2009; Gerstein et al, 1999; Group TDS, 2004; Holvoet et al, 2001; Tsutsui et al,
2002; Judd, 2005; Söderberg et al, 1999; Sierra-Johnson et al, 2007; Singh, 2007; Limas et al, 2002; Halle et al, 1997; Jehn et al, 2004). As indicators of inflammation, these biomarkers have been associated with the progression of cardiac dysfunction in humans.
However, little is known about how each of these biomarkers associates with gorilla cardiac health. By analyzing individual gorilla serum biomarker profiles in relation to cardiac echocardiograph data taken at the time of serum sampling, this study provides a
3 4
baseline for the characterization and understanding of captive gorilla cardiac health.
These baseline data will aid in the prevention and treatment of gorilla cardiovascular
disease, an important consideration in any ex situ conservation effort.
Gorilla Status and Conservation
Following Groves’ (2003) taxonomy gorillas are divided into two species, the eastern gorilla (Gorilla beringei) and the western gorilla (Gorilla gorilla). Each of these species has two representative subspecies. The eastern gorilla includes the eastern lowland gorilla (Gorilla beringei graueri) and the mountain gorilla (Gorilla beringei beringei), while western gorillas are comprised of both the Cross River gorilla (Gorilla gorilla diehli) and the western lowland gorilla (Gorilla gorilla gorilla). As their colloquial names suggest, the eastern and western gorillas are endemic to the eastern and western forested regions of equatorial Africa. There are an estimated 17,000 G. b. graueri, 650-700 G. b. beringei, 150-200 G. g. diehli, and 110,000-150,000 G. g. gorilla left in the wild (Plumbtree et al, 2003).
All great ape populations have been declining in diversity (Benefit & McCrossin,
1995) and in numbers since the late Miocene (Fleagle, 1999) as climate-induced habitat changes sequestered species into a limited range of Miocene-like forest refugia (Clifford et al, 2003; Ballisari, 2008). Yet, historical reductions of extant Gorilla populations have been documented since the first published field research began with George Schaller in
1963 and the pioneering habituation of G. g. beringei by Dian Fossey during the 1970’s-
1980’s (Stewart et al, 2001). This decline is primarily due to zoonotic disease exchange
5
and anthropogenic effects. Ebola alone was responsible for a 95% decrease in the
number of G. g. gorilla at Minkebe, Gabon during the 1990’s (Mehlman, 2008). Human-
induced habitat destruction, encroachment, anthropozoonotic disease exchange, along
with the bushmeat and pet trade, threaten existing populations (Plumbtree et al, 2003) to
the point of local and global extinction (Mehlman, 2008). Consequently, as of 2008, all
gorilla species are listed as endangered according to the International Union for
Conservation of Nature, with both western lowland subspecies G. g. diehli and G. g.
gorilla listed as critically endangered (Robbins & Williamson, 2008; Walsh et al, 2008).
Accordingly, international trade of these animals is prohibited according to international
agreement (CITES: www.cites.org).
Ex situ conservation programs at zoos and animal parks contribute to gorilla
protection by supplying general funding (Stoinski et al, 2008) and the necessary
conditions for experimental research in nutritional and behavioral health (Feistner and
Price, 2002; Kawata, 2008). Specifically, the protection of approximately 360 captive
western lowland gorillas (G. g. gorilla) in 52 North American zoos that are accredited by
the AZA requires the monitoring of G. g. gorilla behavioral, social, nutritional, and physical well-being (Stoinski et al, 2008). To that end, the Gorilla Cardiac Health Project
(GCHP) began in 2006 to monitor captive G. g. gorilla heart disease. Heart disease is now the leading cause of death in captive adult male gorillas (Schmidt et al, 2006),
accounting for 41% of mortality among male gorillas housed at AZA institutions
(Meehan and Lowenstine, 1994
6
Heart Disease
The cardiovascular system is comprised of the heart and a closed system of
arteries, veins, and capillaries (Judd, 2005). Heart disease refers to many types of
pathological conditions that can affect different levels of the system. However, it also
disrupts other organ systems including, but not limited to, the renal, endocrine, and
nervous systems as well as adipose tissue (Judd, 2005). The symptoms and causes of
cardiac disease vary according an animal’s age, diet, activity level, genetic history,
metabolism, and environmental stressors (Judd, 2005; Rector et al, 2007; Lindop, 2007;
Ding and Kullo, 2009; Masson et al, 2005; Robert, 2005; Brydon et al, 2006). Identifying
the particular agent(s) responsible for the onset and progression of disease is critical in
treating the condition and reestablishing a homeostatic balance (Judd, 2005).
G. g. gorilla Heart Disease
Deaths attributed to heart disease were reported during the early years of maintaining gorillas in captivity (Ratcliffe, 1965; Bourne and Cohen, 1975; Cousins,
1983; Schulman et al, 1995; Coe and Lee, 1996). However, mortality rates due to
bacterial, viral, or parasitic infections were comparably higher (Cousins, 1983). For example, Cousins (1983) identified heart disease as the cause of death in only 6.2% of his sample of 226 gorilla necropsies. The higher rate of deaths due to infectious agents or
gastrointestinal disorders was likely due to poor sanitary conditions during his sampling
period. Yet, in reference to captive gorilla heart disease, Cousins (1983) prophetically
warned that “zoo diets and housing are far from ideal” (p.5).
7
By the 1990’s, zoo conditions had improved and deaths from infectious disease
decreased, but reports of captive male gorilla heart disease increased (Schulman et al,
1995; Trupkiewicz et al, 1995; Scott et al, 1995). This pattern parallels the historical trend of increased rates of obesity-related heart disease and decreasing rates of infectious disease in humans living in western societies over the past 100 years (Morris, 2007). For
the most part, cardiac-related deaths in gorillas were sudden and without ante-mortem diagnoses (Patterson and Matevia, 2001; Scott et al, 1995). As a result, details regarding the progression of heart disease in gorillas remain unknown. Also, little to no information on serum biomarker fluctuations prior to or during the progression of heart disease has been published (Baitchman et al, 2006). While reports of serum cholesterol values for captive and free ranging G. g. gorilla do suggest that gorillas are hypercholesterolemic compared to humans (Schmidt et al, 2006: 292; Baitchman et al, 2006), no data were available for cardiac health status. To date, we do not have hormonal data that identifies a captive gorilla heart disease etiology; we do not know what a normal gorilla hormonal profile looks like, nor have we ascertained the underlying genetics that contribute to their heart disease progression.
The Diet/Activity Level Connection
Diet and activity level are major factors that contribute to heart disease progression in humans (Judd, 2005). Therefore, it is important to consider both the diet and level of activity of captive versus wild gorillas.
8
A Natural Gorilla Diet
As obligate forest dwellers, western lowland gorillas are selective herbivores with
diets fluctuating between degrees of folivory and frugivory. Fruit is relatively abundant in
lowland gorilla habitat and gorillas are known to consume 148 different food species
(Rogers et al, 2004). Analysis of fecal samples at Mondika, Central African Republic
showed an average consumption of 3.5 fruit species per day and 10 fruit species per
month (Doran et al, 2002). To obtain this dietary breadth, western lowland gorillas must
increase their daily path length (Doran-Sheehy et al, 2004).
While fruit is undoubtedly preferred by western lowland gorillas, they still
consume a highly fibrous diet compared to both humans and chimpanzees. Herbaceous
material provides the bulk of that fiber. In fact, Rogers et al (2004) reported that 100% of
all gorilla fecal samples in their study contained leaves, stems, pith, shoots, roots, and
bark. The consumption of plant material is facilitated by their specialized digestive tract.
Digestion is aided by their enlarged colons and cecums that contain cellulose-digesting
ciliates (Remis, 2003; Popovich and Dierenfeld, 1997). This aids in the absorption of
nutrients from low-quality foods by increasing the passage rates through the intestinal
tract (Remis and Dierenfeld, 2004).
Captivity and Captive Gorilla Diets
When gorillas were first introduced to Western zoos during the mid 19th century
(Bourne and Cohen, 1975), little was known of gorilla diet, natural habitat, and basic health requirements. As a result, great ape mortality was high (Coe and Lee, 1996). For
9
example, the first gorilla acquired in 1887 by London Zoo was fed “…two sausages and a
pint of beer in the morning, followed later in the day by cheese sandwiches, boiled
potatoes and mutton, and more beer” as related by Wilfred Blunt (Kawata, 2008). Not surprisingly, the male gorilla died after a few weeks (Kawata, 2008).
By the time of the first successful captive gorilla birth in 1956 at the Columbus
Zoo (Maple and Hoff, 1982; Hoff et al, 1998), zoo diets and hygiene had improved. AZA
institutions began to use more naturally designed enclosures during the early 1970’s (Coe
and Lee, 1996) and environmental enrichment included the introduction of browse and
cultivated fruits and vegetables (Kawata, 2008). A new emphasis on the nutritional
requirements of all zoo animals was being met by cost effective, commercially available
foods (Dierenfeld, 1997). Commercial foods are now uniformly used in captive gorilla
diets. Popovich and Dierenfeld (1997) found this to be true for 73% of surveyed zoos and
accounted for 14.7% of each adult male gorilla meal, 11% of each adult female gorilla
meal, and 11% of each juvenile gorilla meal (Popovich and Dierenfeld, 1997). The
survey also found six zoos feeding gorillas meat once daily, 19 zoos feeding eggs three
times per day, 18 zoos feeding milk ten times per day, and 21 zoos feeding gorillas
yogurt four times per day (Popovich and Dierenfeld, 1997).
Like most hominids, gorillas have a sweet tooth. Remis (2003) tested taste
preference in captive gorillas and found that they prefer domestic fruits to vegetable
foods (Remis, 2003). This preference was highly correlated to sugar content and
incremental additions of tannins did not appear to influence their preference for sweets.
In the same manner, Simmen and Charlot (2003) tested the taste threshold for fructose in
10
captive Pan, Pongo and Gorilla and found gorillas to have the highest threshold tolerance
for sugar.
Despite growing interest in enrichment, captive gorillas have an absolute lower
level of physical activity than their free-ranging counterparts (Schulman et al, 1995).
Furthermore, all gorillas may not exhibit the same level of ‘fidgetiness’ as other hominoids. Research into human obesity has shown a relationship between overweight individuals and a lower amount of non-exercise activity thermogenesis (NEAT)
(Koeppen and Stanton, 2008). Consequently, NEAT accounts for 20%- 30% of total average daily expenditure of a 2,300 kcal diet in humans (Koeppen and Stanton, 2008).
Primatologists have often commented on the slow lethargic nature of gorilla interaction with the environment (Patterson & Matevia, 2001). Thus, gorillas become prone to increased obesity when an unnaturally high calorie diet is offered without an increase in either voluntary or involuntary physical activity to offset it.
Adipose, Obesity and Thrifty Genotypes
Adipose
Adipose refers to a non-contiguous distribution of subcutaneous and visceral fat deposits within the mammalian body (Cinti, 2007). These fat stores can be accessed quickly during times of fight or flight as they transfer energy-loaded lipoproteins to the liver (Björntorp, 2000). Previous understanding of adipose was limited to its energy storage potential. However, adipose is now understood to be the largest endocrine organ in the body (Cinti, 2007) and its functional units are the fat cells called adipocytes
11
(Jackson and Ahima, 2007). Adipocytes are involved in multiple physiological functions
via the secretion of adipokines (Ferrante, 2007). Adipokines circulate through the body
and relay information about satiety and energy, immune function, metabolism, and
thermoregulation (Ferrante, 2007).
Obesity
Obesity is a pathological state characterized by an increase in both size and
number of adipocytes. This is associated with an increase in stored triglyceride and
cholesterol in fat tissue. The proliferation of adipocytes leads to inflammation as
adipocytes release more adipokines in direct proportion to the amount of adipose tissue
(Wells, 2006). In turn, adipokines stimulate immune cells, including macrophages and helper T cells and they secrete other more deleterious cytokines (Stehno-Bittel, 2008).
Obesity disrupts the normal adipokine signaling and can be best described as a ‘low grade inflammatory state’ (Ferrante, 2007; Wells, 2006; Stehno-Bittel, 2008).
The increased storage of cholesterol and triglycerides in adipocytes parallels an increased secretion of long chain fatty acids by adipocytes (Cefalu, 2008). Cardiac muscle cells alternate between glucose and long chained fatty acids for energy (Schwenk et al, 2008). Obesity increases the accumulation of long chain fatty acids and the cardiomyocytes utilize these rather than glucose (Schwenk et al, 2008). This indirectly increases circulating glucose and may lead to decreased insulin sensitivity (Schwenk et al, 2008).
12
Disruption in the role of insulin to regulate glucose uptake is further compromised as increased fatty acids derived from adipose are introduced to the liver. Metabolism of these fatty acids induces more endogenous glucose production and the individual becomes hyperglycemic (Cefalu, 2008). Furthermore, increased fatty acids are taken up by skeletal muscle and the liver, both of which would otherwise assist in clearing glucose from circulation (Sikaris, 2004). This further decreases insulin sensitivity. Anecdotally, both male and female gorillas appear to show more obesity in captivity than in the wild
(Hatt and Liesegang, 2002; Zihlman and McFarland, 2000).
Thrifty genotypes
Fat deposition offers animals a means of staving off starvation when food resources are scarce. Consequently, there is a question of whether a “thrifty genotype”, promoting adipose accumulation during periods of food abundance offers a selective advantage against starvation during lean times. Neel (1962) developed this hypothesis to explain why type II diabetes mellitus is prevalent among human populations. Neel proposed that the current pattern of type II diabetes resulted from the genetic legacy of hunters and gatherers fending off starvation by decreasing insulin sensitivity (i.e., insulin resistance). A temporary diabetic state would favor increased adipose storage of energy during times of food abundance (Neel, 1962). That energy store would later be advantageous during periods of famine. However, western cultures have an excess and continuous supply of food due to a high-calorie diet and sedentary lifestyle. Neel’s
13
putative temporary diabetic state has become chronic, pathological, and epidemic
(Stehno-Bittel, 2008; Hill and Peters, 2002)
While controversy surrounds the applicability of the thrifty genotype hypothesis
(Bindon and Baker, 1997; Bellisari, 2008), it does provide a reasonable model for the occurrence of obesity found in captive G. g. gorillas given that the diet offered in captivity has not been adapted for and these gorillas are more sedentary.
Metabolic Syndrome
Dyslipidemia, hyperinsulinemia, and obesity are associated with metabolic syndrome in humans (Singh et al, 2007). Specific diagnosis in humans is met by the following American Heart Association/National Heart, Lung, and Blood Institute criteria: a waist measuring greater or equal to 40” and 35” in men and women, respectively; triglyceride levels greater than or equal to 150 mg/dl; high density lipoprotein levels below 40 mg/dl in men and 50 mg/dl in women; hypertension; and fasting glucose levels of 100 mg/dl or above (Grundy et al, 2005). Collectively, these increase the risk of coronary artery disease and type II diabetes mellitus (Singh et al, 2007), and underlying these criteria is insulin resistance; when insulin levels remain elevated postprandially and during a fasted state (Tentolouris et al, 2008).
A characteristic of metabolic syndrome is the creation of reactive oxygen species that increase oxidative stress (Ando and Fujita, 2009). Additionally, those with metabolic syndrome are at greater risk for myocardial abnormalities when controlling for age and gender (Grandi et al, 2006). Several biomarkers have been used to assess metabolic
14
syndrome. These include: insulin, glucose, leptin, cholesterol, oxLDL, prolactin, and
ferritin. Below is a review of each.
Glucose and Insulin
Food consumption and deposited fat are mechanisms to obtain and regulate
energy-providing glucose. While there are many hormones that regulate the release of
glucose, insulin is the primary hormone that regulates glucose storage (Watson et al,
2007). Insulin, produced by pancreatic β-cells, directs the clearance of circulating glucose into skeletal muscle, cardiac muscle, and adipose tissue (Watson et al, 2007) and the synthesis and release of insulin is stimulated by higher blood glucose levels following a meal (Watson et al, 2007). Postprandially, insulin directly inhibits hepatic glucose production by inhibiting gluconeogenic enzymes (Courtney and Olefsky, 2007). This allows the circulating levels of glucose to be controlled.
In type II diabetes mellitus, the regulation of glucose is disrupted by 1) excessive caloric intake can allow triglycerides to accumulate in skeletal muscle which impairs the ability of muscle tissue to store glucose. Individuals with type II diabetes show a 50% reduction in muscle glycogen synthesis compared to non-diabetic individuals (Courtney and Olefsky, 2007); 2) increased visceral adiposity interferes with insulin regulation of hepatic glucose production. Insulin resistance follows as increased free fatty acids and triglycerides are loaded into hepatic tissue and impair insulin signaling (Courtney and
Olefsky, 2007); and 3) excessive glucose intake may lead to a progressive overproduction of insulin leading to eventual β-cell exhaustion (Kahn, 2003).
15
Early studies have demonstrated that circulating levels of insulin are lowered post-heart attack via the suppressive action of the sympathetic nervous system on pancreatic β-cells (Taylor, 1971). This induces hyperglycemia. Conversely, heart cells must switch to the more efficient glucose metabolism following heart attack related hypoxia (Taylor, 1971). Chronic low insulin sensitivity, characteristic of obesity and metabolic syndrome, impairs this substrate switch during heart disease progression as both conditions combine to increase circulating free fatty acids (Witteles and Fowler,
2008). Thus, insulin resistance been implicated in the progression of several types of heart disease (Doehner et al, 2005; Swan et al, 1997; Bajraktari et al, 2006).
Indices of Insulin Resistance and Sensitivity
To determine insulin resistance in vivo, a host of techniques have been developed.
These include the measurement of insulin-mediated metabolism in response to steady state glucose infusion and frequent glucose tolerance tests in addition to using a ratio of fasting glucose to insulin levels (Quon, 2001). The glucose to insulin ratio provides a good proxy when insulin and glucose levels are normal; however, this is less precise measurement of insulin sensitivity when fasting glucose values are abnormal (Quon,
2001).
A sensitive indicator of insulin resistance is provided by the Quantitative Insulin
Sensitivity Check Index (QUICKI) (Katz et al, 2000). This simple calculation uses fasting insulin and glucose values and QUICKI is read as a reciprocal of insulin resistance. Therefore, this index measures the degree of insulin sensitivity. Lower
16
QUICK scores reflect low sensitivity to insulin. QUICKI is calculated according to the original Katz et al (2000) model.
QUICKI = 1/ [log (I0) + log (G0)]
Where: I0 = fasting insulin and G0 = fasting glucose
Leptin
The hormone leptin was identified in 1994 by Zhang and colleagues as the product of the ob gene (Zhang et al, 1994). The authors found that mutations in ob, and subsequent ob/ob homozygous mice, yielded obese phenotypes and type II diabetes. By manipulating leptin concentrations, subsequent researchers demonstrated dramatic weight loss and restored insulin sensitivity in mice (Martin, 2000). This reaffirmed an earlier hypothesis of a hormonal regulating ‘satiety factor’ as predicted by the ‘lipostatic theory’ of food intake (Kennedy, 1953). It also promised a new means of treatment for metabolic syndrome.
Leptin is a soluble, hydrophobic protein-based hormone produced primarily in mature adipocytes (Martin, 2000). Leptin receptors are expressed within the feeding center of the hypothalamus (Dias et al, 2006) and are critical in the modulation of appetite, energy expenditure, and body weight (Houseknecht & Spurlock, 2003). Leptin communicates with the satiety center about the amount of stored triglycerides in adipose tissue, establishing a negative feedback loop where increased fat deposition and overall leptin output should cause a lowered desire for food intake under a normal physiological state (Ahima & Osei, 2004). Increased amounts of adipose should secrete greater
17
amounts of leptin to signal satiation. In turn, a signal of satiation should lower the amount
of deposited adipose (Houseknecht & Spurlock, 2003).
Leptin resistance is related to increased levels of leptin in serum and is a chronic
condition that has been associated with cardiovascular disease (Soderberg et al, 1999;
Sierra-Johnson et al, 2007; Wallander et al, 2008) and often accompanies hyperinsulinaemia (Kellerer et al, 2001). Both the up-regulation of leptin gene expression and increased leptin secretion by adipocytes are directly enhanced with increased circulating insulin (Margetic et al, 2002). Conversely, normal levels of leptin appear to suppress insulin secretion and gene expression (Seufert et al, 1999). Obesity, coupled with insulin resistance, disrupts this regulation (Prolo et al, 1998). Leptin levels have not yet been published for captive or wild gorillas.
Cholesterol
As a biological lipid, cholesterol is critical for the formation of cellular membranes, steroid hormones, and bile acid synthesis (Medicine, 2005). Despite this importance, exogenous cholesterol is not a daily requirement in humans over the age of two (Djousse & Gazianno, 2009). Our cells synthesize cholesterol and consumption of foods naturally high in cholesterol adds to circulating cholesterol levels (Medicine,
2005).
Hypercholesterolemia can be directly assessed by looking at the levels of total cholesterol. The National Institutes of Health provide a guide to determine disease risk associated with total cholesterol in humans. Total cholesterol above 240 mg/dl is
18
considered high risk for heart disease, while total cholesterol below 200 mg/dl is considered desirable (Judd, 2005). Gorillas appear to be hypercholesterolemic compared to humans, with total cholesterol levels above 240 mg/dl (Schmidt et al, 2006; Baitchman et al, 2006).
Oxidized LDL
Oxidation of low density lipoprotein (oxLDL) occurs as free radicals bombard the lipoprotein. The oxLDL is then received with greater affinity along the endothelium and vascular smooth muscles (see review in Negre-Salvayre et al, 2006). In response, localized cells mount an immune response that includes the aggregation of macrophage cells (Horkko et al, 2000). Holvoet et al (2001) found an association between high levels of oxLDL and metabolic syndrome, suggesting oxLDL can be generated via dyslipidemic and hyperinsulinemic mechanisms. This increases the likelihood of heart disease (Shen,
2006; Mertens & Holvoet, 2001; Boullier et al, 2001). Furthermore, patients with dilated cardiomyopathy have elevated oxLDL concentrations (Tsutamoto et al, 2001) and cardiac remodeling is also associated with increased oxLDL. This includes the contractile function of cardiomyocytes during systole and diastole (Rietzschel et al, 2008).
Therefore, oxLDL is both an indicator of atherogenesis and cardiomyopathy. In addition, levels of oxLDL have been found to correlate positively with leptin (Beltowski, 2006).
19
Prolactin
Prolactin is a polypeptide hormone primarily produced in the anterior pituitary
(Hu et al, 1998; Holt, 2008). Prolactin secretion is regulated by dopamine which inhibits prolactin release (Holt, 2008). Greater amounts of dopamine decrease prolactin secretion and greater amounts of secreted prolactin decrease dopamine concentrations.
In addition to its roles in lactation and reproduction, prolactin also plays a role in
the stress response (Ben-Jonathan and Hanasko, 2001) and immune regulation (Hu et al,
1998). At physiological levels, prolactin reduces fat metabolism which is probably
necessary in maintaining sufficient adiposity during pregnancy. However, high prolactin
levels (hyperprolactinemia) may lead to an increase in inflammation (Cejkova et al, in
press). Prolactin can also act as a co-activator of platelet aggregation during developing atherosclerotic plaque (Wallaschofski et al, 2004; Raaz et al, 2006).
Surprisingly, Limas et al (2002) found that prolactin levels were positively
associated with a greater ejection fraction, and ventricle dilation was more pronounced in
those with hyperprolactinemia. Those results indicate that prolactin levels may increase
as a post-infarction protective response. However, this increase leads to the activation of
T-lymphocytes, as cells involved in inflammatory respond to tissue damage. Thus,
paradoxically, hyperprolactinemia induces platelet aggregation in blood vessels
(Wallaschofski et al, 2004) while also having positive tropic effect on the heart after an
ischemic event, due to this very same inflammatory induction (Limas et al, 2002).
Prolactin also plays a role in insulin regulation and resistance, although its effects are not clearly defined. Prolactin was reported to promote pancreatic β-cell proliferation
20
(Holstad and Sandler, 1999) and insulin secretion (Cejkova et al, in press), and has also
has been implicated in insulin resistance (Serri et al., 2006).
Ferritin
Iron is crucial for the transport of oxygen throughout the body by erythrocytes
(Orino and Watanabe, 2008). It also plays a role in metabolism (Orino & Watanbe,
2008). However, excessive iron (iron overloading) can lead to cell damage through its effects that result in reactive oxygen species (Rajpathak et al, 2009) increasing the oxidative stress associated with type II diabetes (Rajpathak et al, 2009), insulin resistance
(Guenno et al, 2007), metabolic syndrome (Jehn et al, 2004), and increase the level of oxLDL (Brouwers et al, 2004).
Ferritin is an iron binding protein and is often used as a proxy to measure iron loading (Weinberg, 1999). Serum ferritin levels correlate with body iron stores
(Weinberg, 1999) and a reduction in serum ferritin should correspond to a reduction in stored iron and thus a diminished risk for generating reactive oxygen species (Welch et al, 2001). There is evidence that indicates that individuals that shed more blood, either through phlebotomy or through menstruation, have a decreased risk for heart disease
(Sullivan, 1999) and have improved insulin sensitivity (Fernandez-Real et al, 2002;
Guenno et al, 2007).
Increased serum ferritin is an indicator of inflammation (Orino and Watanabe,
2008). Insulin stimulates iron uptake in adipocytes (Fernandez-Real et al, 2002).
Correspondingly, increased iron loading in fat cells increases free fatty acid release
21
(Green et al, 2006). Increased free fatty acids are taken up in hepatocytes where they oxidated and stimulate the production of glucose (Clark and Newgard, 2007). During this, free iron increases in the β-cells (Fernandez-Real et al, 2002). This disrupts insulin secretion which cannot control increased liver glucose production.
Research Questions and Predictions
The aim of this project was to determine whether the aforementioned biomarkers were also associated with four echocardiographic parameters of cardiac health in captive male gorillas. The measures of cardiac health were determined in terms of: efficiency, as measured by ejection fraction (EF); thickening, as measured by both interventricular septum thickness (IVS) and left ventricle posterior wall thickness (LVPW); and a measure of dilation given by the left ventricle internal diameter at diastole (LVIDd). The specific hypotheses are listed below.
H1) QUICKI scores will correlate positively with EF and negatively correlate with IVS, LVPW and LVIDd.
H2) Serum leptin concentrations will associate negatively with EF and positively with IVS, LVPW, and LVIDd.
H3) Serum prolactin will positively correlate with EF, IVS, LVPW, and LVIDd.
H4) Levels of total cholesterol will negatively correlate with EF and positively with IVS, LVPW, and LVIDd.
H5) Levels of serum oxLDL will negatively correlate with EF and positively with IVS, LVPW, and LVIDd.
22
H6) Levels of serum ferritin will negatively correlate with EF and positively with
IVS, LVPW, and LVIDd. .
H7) Among the serum biomarkers, QUICKI will negatively correlate with leptin, prolactin, total cholesterol, oxLDL, and ferritin. Leptin will positively correlate with, prolactin, total cholesterol, oxLDL, and ferritin. Prolactin will positively correlate with total cholesterol, oxLDL, and ferritin. Total cholesterol will positively correlate with oxLDL and ferritin. Finally, oxLDL will positively correlate with ferritin.
Chapter Two
Materials and Methods
Concurrent echocardiograph data and serum samples were collected during annual health examinations at AZA institutions across North America. The study sample included 30 male gorillas, ages 3 – 48 years (mean = 20.53 years, SD=12.39) (See
Appendices A and B). Due to the absence of cardiac disease and other confounding health problems, females were excluded from this analysis.
Echocardiograph Measures
Echocardiograph data provides an independent method to determine gorilla cardiac health status (Junge et al, 1998) and have proven invaluable in determining cardiac disease progression and treatment in humans (Judd, 2005; Lohr, 2005). One diagnostic variable is how efficiently the heart pumps blood out of the left ventricle
during diastole, i.e., ejection fraction (EF) (Lohr, 2005). An additional variable important in assessing cardiac health is cardiac thickening. Increased thickening of the cardiac walls decreases compliancy and increases the workload for the heart (Lohr, 2005). Key measures of thickening include the interventricular septum thickness (IVS) and left ventricle posterior wall thickness (LVPW). Correspondingly, left ventricle internal diameter at diastole (LVIDd) provides a measure of dilation of the left ventricle (Lohr,
2005).
23 24
Enzyme immunoassays (EIAs)
Commercially available kits were use to measure the concentrations of leptin, prolactin, oxidized LDL, ferritin, and insulin. Because the kits were designed to measure concentrations in human serum, each assay was validated for use with gorilla serum using parallelism and recovery tests. For parallelism, serially diluted pooled samples were assayed and compared with the standard curve to ensure that the sample curve was parallel to the standard curve. For recovery, known concentrations (low and high) of hormone were added to a sample of pooled serum and analyzed to evaluate potential interference with the assay. All samples were assayed in duplicate and read using a spectrophotometer (Dynex Technologies Revelation, 4.25).
Analysis for leptin was performed using a mouse monoclonal anti-human serum enzyme-linked assay (ELISA) kit (Mercodia AB, Uppsala, Sweden) with individual gorilla samples diluted at 50 μl (serum) to 450 μl of sample buffer. Prolactin concentrations were measured with a mouse monoclonal anti-human sandwich method
ELISA kit (BIO-CLIN #1203, St. Louis, MO) with individual serum samples diluted at a ratio of 50 μl serum to 50 μl of sample buffer. Serum oxLDL was assessed by a monoclonal competitive ELISA (Mercodia AB, Uppsala, Sweden) with individual serum samples diluted at a ratio of 100 μl:1000 μl. Serum ferritin was measured by using a human sandwich method ELISA kit (Calbiotech Inc. Spring Valley, CA) with 20 μl of individual serum samples diluted with 20 μl of assay buffer. Insulin was analyzed by a human insulin sandwich method ELISA kit (Mercodia, Uppsala, Sweden) with undiluted
(neat) serum samples.
25
In vitro chemical analysis on glucose and total cholesterol was performed using a
calibrated IDEXX VetTest analyzer (Indexx Laboratories Inc. Westbrook, ME)
Insulin Sensitivity
The QUICKI method was used as a proxy to assess insulin sensitivity in gorillas.
In humans, insulin resistance and subsequent insulin sensitivity is best quantified using a
hyperinsulinemic euglycemic glucose clamp technique (Katz et al, 2000). Unfortunately,
this technique, along with the frequent samples necessary for a glucose tolerance test, are
impractical for assessing gorilla insulin resistance as both tests would require sedation for
long periods of time with steady state insulin and glucose infusions. QUICKI is
calculated from fasting insulin and glucose values with the equation given below:
(QUICKI = 1/ [log (I0) + log (G0)]
Where: I0 = fasting insulin and G0 = fasting glucose
Therefore, QUICKI may be currently the only way to establish gorilla insulin sensitivity parameters. Low QUICKI scores denote poor insulin sensitivity, while higher QUICKI scores indicate better insulin sensitivity.
Statistical Analyses
All statistical analyses were performed on SPPS version16.0 software for
Windows (SPSS, Chicago, IL). Initial tests of normality revealed a positively skewed distribution for all variables except age. A simple log transformation was applied (X’ =
Log10 (X); Zarr, 1984) to both dependent (EF, IVS, LVPW, and LVIDd) and independent
variables (all biomarkers), but not for age. Separate unrotated principal components
26
analyses (PCA) based on an inter-correlation matrix were used for each echocardiograph measurement (PCA 1-4) with all biomarkers. Zero-order correlation analyses
(Correlation Analyses 1-4) of the variables loading the highest and lowest on each component derived from the four individual PCAs were used to further evaluate these relationships. Partial correlation analyses, controlling for age, were conducted in the same manner. Tests were one-tailed, as per hypotheses, and alpha was set at 0.05.
While the research question of this study is concerned with how inflammatory
biomarkers associate with these specific echocardiographic measures, diagnoses were
provided by Haley Murphy, DV, Ilana Kutinsky, MD. These diagnoses were given as
simple dichotomous, “without heart disease” or “with heart disease”.
Chapter Three
Results
EIA Validation
The serial dilution curve for each assay was parallel to its standard curve, with no difference between slopes [leptin (t(9) = -0.10, p = 0.93); oxLDL (t(9) = -1.97, p = 0.09),
insulin (t(9) = 1.43 p = 0.20); prolactin (t(8) = 1.95 p = 0.10); ferritin (t(10)= 0.58, p =
0.58)]. Recovery tests did not indicate significant interference. Recovery for leptin was
91% for high and 95% for low; oxLDL 101.4% high, 100.2% low; insulin 101.8% high,
98.2% low; prolactin 93% high, 87.5% low; ferritin 97.4% high and 96.7% low. The
mean intra-assay coefficients of variation were: leptin 3.57%; prolactin 4.87%; oxLDL
2.83%; ferritin 5.87%; and insulin 4.00%.
Biomarker and Echocardiograph Values
Values for individual gorilla biomarkers along with QUICKI scores are provided
in Appendix A. Descriptive statistics of biomarkers and QUICKI are given in Table 1 for
gorillas diagnosed with heart disease. Table 2 contains the same for gorillas diagnosed
without heart disease. Descriptive statistics for EF, IVS, LVPW, and LVIDd are given in
Table 3 for gorillas with heart disease. Table 4 provides the same echocardiograph
measures for those gorillas without heart disease. All individual gorilla echocardiograph
measures are provided in Appendix B.
27 28
Table 1. Descriptive statistics for all biomarker variables and QUICKI, including insulin and glucose values used in the calculation of QUICKI for those gorillas diagnosed with heart disease.
N Minimum Maximum Mean Std. Deviation
Glucose (mg/dl) 19 56.00 126.00 85.26 18.31
Insulin (mU/l) 19 0.42 15.03 5.40 4.49
QUICKI 19 0.33 0.67 0.43 0.11
Leptin (ng/dl) 19 0.67 13.75 6.34 4.49
Cholesterol (ng/dl) 19 187.00 338.00 249.53 42.23 oxLDL (U/l) 19 2.25 7.58 4.02 1.27
Prolactin (ng/ml) 18 22.43 76.63 36.60 14.36
Ferritin (ng/dl) 18 403.94 1753.93 731.12 307.94
Table 2. Descriptive statistics for all biomarker variables and QUICKI, including insulin and glucose values used in the calculation of QUICKI for those gorillas without heart disease
N Minimum Maximum Mean Std. Deviation
Glucose (mg/dl) 11 44.00 135.00 83.09 25.06
Insulin (mU/l) 11 .47 6.61 3.04 2.22
QUICKI 11 .35 .76 0.47 0.13
Leptin (ng/dl) 11 .86 10.58 2.86 3.14
Cholesterol (ng/dl) 11 186.00 367.00 284.82 62.84 oxLDL (U/l) 11 2.79 4.91 3.73 0.79
Prolactin (ng/ml) 11 23.19 104.53 46.79 23.09
Ferritin (ng/dl) 11 333.05 1018.08 606.61 219.83
29
Table 3. Descriptive statistics of each echocardiograph measure for those gorillas with heart disease.
N Minimum Maximum Mean Std. Deviation
EF (%) 19 30 80 50.45 12.66
IVS (cm) 19 1.0 2.9 1.80 0.50
LVPW (cm) 16 .9 2.6 1.76 0.48
LVIDd (cm) 19 4.4 7.5 5.93 0.86
Table 4. Descriptive statistics of each echocardiograph measure for those gorillas without heart disease.
N Minimum Maximum Mean Std. Deviation
EF (%) 11 55 89 68.36 9.81
IVS (cm) 11 .6 1.5 1.18 0.29
LVPW (cm) 10 .6 1.5 1.11 0.28
LVIDd (cm) 11 3.0 7.5 4.77 1.50
30
Statistical Analyses
PCA 1
PCA 1 included LogEF and the six biomarkers (n=29). Results from this analysis yielded three components with eigenvalues exceeding one, explaining a cumulative
68.94% of the variance. Component one explained 29.68%, component two explained
21.38%, and component three explained 17.88% of the variance respectively (Table 5).
Visual inspection of the scree plot showed a continuous slope, indicating that more information may be contained in components with eigenvalues less than 1 (Figure 1).
LogLeptin and LogFerritin loaded positively on component one. Loading heavily in the negative direction on component one was LogCholesterol. Component one can be described as ‘high leptin and ferritin and low cholesterol”. On component two, a strong positive loading included LogEF while a heavy negative loading included LogQUICKI.
Component two can be described as ‘high cardiac efficiency with low insulin sensitivity’.
Component three contained a strong negative loading for LogProlactin with no variable loading substantially positive. Component three can be described as ‘low prolactin’.
31
These six variables were included in correlation analysis one. Given the weak loadings of
LogoxLDL, it was excluded from correlation analysis one.
Table 5. Unrotated principle components solution (PCA 1), n = 29 1 2 3
LogEF -.284 .723 -.096
LogLeptin .836 .320 .267
LogoxLDL .208 -.090 .440
LogCholesterol -.697 .163 .573
LogQUICKI -.381 -.763 -.294
LogProlactin -.046 .435 -.742
LogFerritin .788 -.255 -.112
Eigenvalue 2.078 1.496 1.252
% variation 29.68 21.38 17.88 Total: 68.94
Figure 1. Scree plot for PCA 1.
32
PCA 2
PCA 2 included LogIVS and the six biomarkers (n=29). Results from this analysis yielded three components with eigenvalues exceeding one, explaining a cumulative
69.94% of the variance. Component one explained 32.81%, component two explained
19.22%, and component three explained 17.91% of the variance (Table 6). Visual inspection of the scree plot showed clear breaks after components one, two, and three
(Figure 2). Strong positive loadings on component one included LogLeptin and
LogFerritin. A heavy negative loading on component one was associated with
LogCholesterol. Again, component one is best described as ”high leptin and ferritin and low cholesterol”. Component two included a strong positive loading for LogQUICKI.
Here, component two may be described as ”good insulin sensitivity”. Component three contained a heavy negative loading for LogProlactin. LogIVS did not appear to load well
33
with any of these biomarkers. These six variables were included in correlation analyses two. Given the weak loadings for LogoxLDL, it was excluded from correlation analysis two.
Table 6: Unrotated principle components solution (PCA 2), n = 29 1 2 3
LogIVS .633 .450 .121
LogLeptin .803 -.381 .225
LogoxLDL .182 .079 .397
LogCholesterol -.680 -.128 .592
LogQUICKI -.334 .841 -.239
LogProlactin -.112 -.468 -.781
LogFerritin .794 .223 -.120
Eigenvalue 2.297 1.345 1.254
% variation 32.81 19.22 17.91 Total: 69.94
34
Figure 2. Scree plot for PCA 2
PCA 3
PCA 3 included LogLVPW and the six biomarkers (n=25). Results from this analysis yielded four components with eigenvalues exceeding one, explaining a cumulative 83.33% of the variance. Component one explained 31.77%, component two explained 19.35%, component three explained 17.84%, and component four explained
14.72% of the variance (Table 7). The scree plot showed clear breaks after components one, two, three, and four (Figure 3). Strong positive loadings on component one included the variables LogLeptin, LogFerritin, and LogLVPW . The strong negative loading on component one was associated with LogCholesterol. Component one can be described as
‘high leptin, ferritin, and LVPW with low cholesterol’. Component two contained a strong negative loading for LogProlactin. Component three contained a strong negative loading for LogCholesterol and a strong positive loading for LogQUICKI. Component
35
describes ‘good insulin sensitivity with low cholesterol’. Component four had a strong positive loading for LogoxLDL. All variables were included in correlation analyses 3.
Table 7. Unrotated principle components solution (PCA 3), n = 25 1 2 3 4
LogLVPW .622 .460 -.071 .183
LogLeptin .791 -.207 -.394 -.075
LogoxLDL -.013 .470 -.141 .816
LogCholesterol -.717 .112 -.571 -.095
LogQUICKI -.335 .597 .653 -.206
LogProlactin -.062 -.696 .502 .457
LogFerritin .762 .159 .256 -.200
Eigenvalue 2.224 1.355 1.249 1.006
% variation 31.77 19.35 17.84 14.72 Total: 83.33
36
Figure 3. Scree plot for PCA 3
PCA 4
PCA 4 included LogLVIDd and the six biomarkers (n=29). Results from this analysis yielded three components with eigenvalues exceeding one, explaining a cumulative 67.56% of the variance. Component one explained 30.06%, component two explained 19.44%, and component three explained 18.06% of the variance (Table 8).
Visual inspection of the scree plot showed clear breaks after components one, two, and three (Figure 4). Positive loadings on component one included LogLeptin and
LogFerritin. A negative loading on component one was associated with LogCholesterol.
Component one again describes ‘high leptin and ferritin with low cholesterol’.
Component two included a positive loading for LogLVIDd and a negative loading for
LogProlactin and may be described as ‘increased cardiac dilation with low prolactin’.
Component three loaded positive for LogQUICKI and negative for LogCholesterol.
37
Component three is described as ‘good insulin sensitivity and low cholesterol’. These six variables were included in correlation analyses four. Given the weak loadings for
LogoxLDL, it was excluded from correlation analysis four.
Table 8. Unrotated principle components solution (PCA 4), n = 29 1 2 3
LogLVIDd .350 .638 .024
LogLeptin .868 -.093 -.321
LogoxLDL .180 .053 .362
LogCholesterol -.675 .099 -.625
LogQUICKI -.417 .531 .629
LogProlactin -.077 -.801 .407
LogFerritin .748 .099 .278
Eigenvalue 2.104 1.361 1.264
% variation 30.06 19.44 18.06 Total: 67.56
38
Figure 4. Scree plot for PCA 4
Correlation Analyses
Correlation Analysis 1
Zero-order and partial correlations coefficients for LogEF and the biomarkers selected from PCA 1 are provided in Table 9. Age was significantly correlated with
LogEF (r = -.363) (Figure 5), LogCholesterol (r = -.431) (Figure 6), LogQUICKI (r = -
.413) (Figure 7), LogLeptin (r = .519) (Figure 8), and LogFerritin (r = .671) (Figure 9).
However, no significant correlations were found between LogEF serum biomarkers.
Among the biomarkers, LogLeptin was significantly correlated with
LogCholesterol (r = -.408) (Figure 10), LogQUICKI (r = -.558) (Figure 11), and
LogFerritin (r = .519) (Figure 12).
39
When controlling for age, LogEF and LogQUICKI had a significant negative correlation (r = -.407) (Figure 13). No other significant associations emerged between EF and the serum biomarkers.
Among the biomarkers in the partial correlation, LogQUICKI was negatively correlated with LogLeptin (r = -.442) (Figure 14) and positively correlated with
LogFerritin(r = .327) (Figure 15). Although not significant, LogCholesterol demonstrated the predicted negative trend with LogQUICKI (r = -.304, n = 29, p = .058) (Figure 16) and an unpredicted negative trend with LogProlactin (r = -.292, n = 29, p = .066) (Figure
17) and LogFerritin (r = -.315, n = 29, p = .051) (Figure 18).
LogLeptin demonstrated a notable split between those gorillas diagnosed with heart disease and those without heart disease (Figures 8 through 12).
Table 9. Zero-order and Partial correlations for logEF and logLeptin, logCholesterol, logQUICKI, logProlactin, logFerritin, and age. Control Variables LogEF LogLeptin LogCholesterol LogQUICKI LogProlactin LogFerritin Age nonea LogEF 1 LogLeptin .007 1 LogCholesterol .199 -.408* 1 LogQUICKI -.192 -.558** -.072 1 LogProlactin .225 -.168 -.262 -.158 1 LogFerritin -.265 .466** -.500** -.056 -.094 1 Age at sample -.363* .519** -.431** -.413* -.004 .671** 1
Age LogEF 1
LogLeptin .245 1
LogCholesterol .050 -.239 1
LogQUICKI -.403* -.442** -.304 1
LogProlactin C .240 -.194 -.292 -.175 1
LogFerritin Correlation -.031 .185 -.315 .327* -.123 1 a. Cells contain zero-order (Pearson) correlations. * sig. at p < .05, ** sig. at p < .01
40
Figure 5. Bivariate plot of LogEF against age with coded diagnosis.
Figure 6. LogCholesterol and age with coded diagnosis
41
Figure 7. LogQUICKI and age with coded diagnosis.
Figure 8. LogLeptin and age
42
Figure 9. LogFerritin and age.
Figure 10. LogLeptin and LogCholesterol.
43
Figure 11. LogLeptin and LogQUICKI.
Figure 12. LogLeptin and LogFerritin
44
Figure 13. bivariate plot of LogEF against LogQUICKI controlling for age (residuals).
Figure 14. Bivariate plot of LogQUICKI and LogLeptin controlling for age (residuals).
45
Figure 15. Bivariate plot of LogQUICKI and LogFerritin controlling for age (residuals).
Table 16. LogQUICKI and LogCholesterol controlling for age (residuals)
46
Table 17. LogProlactin and LogCholesterol controlling for age (residuals)
Table 18. LogCholesterol and LogFerritin controlling for age (residuals)
47
Correlation Analysis 2
Age was significantly correlated with LogIVS (r = 0.476) (Figure 19) and with each of the serum biomarkers except prolactin (Table 10). LogIVS was positively correlated with LogFerritin (r = 0.439) (Figure20), but negatively correlated with
LogCholesterol (r = -.333) (Figure 21). Similar to correlation analysis 1, there were several correlations among the serum biomarkers (see Table 9)
When age was controlled for, LogIVS did not significantly correlate with any of the serum biomarkers. However, LogIVS had a positive trend with LogQUICKI (r =
.252, n = 29, p = .098) (Figure 22) and a negative trend with LogProlactin (r = -.276, n =
29, p = .077) (Figure 23). As in correlation analysis 1, LogQUICKI was negatively correlated with LogLeptin and positively correlated with LogFerritin (see Table 9).
Table 10. Zero-order and partial correlations for logIVS and logLeptin, logCholesterol, logQUICKI, logProlactin, logFerritin, and age
Control Variables LogIVS LogLeptin LogCholesterol LogQUICKI LogProlactin LogFerritin Age nonea LogIVS 1
LogLeptin .255 1
LogCholesterol -.333* -.408* 1
LogQUICKI .005 -.558** -.072 1
LogProlactin -.245 -.168 -.262 -.158 1
LogFerritin .439** .466** -.500** -.056 -.094 1
48
Age at sample .476** .519** -.431** -.413* -.004 .671** 1
Age LogIVS 1
LogLeptin .011 1
LogCholesterol -.161 -.239 1
LogQUICKI .252 -.442** -.304 1
LogProlactin -.276 -.194 -.292 -.175 1
LogFerritin .183 .185 -.315 .327* -.123 1 a. Cells contain zero-order (Pearson) correlations. * sig. at p < .05, ** sig. at p < .01
Figure 19. Bivariate plot showing LogIVS and age.
Figure 20. LogIVS and LogFerritin.
49
Figure 21. LogIVS and LogCholesterol
50
Figure 22. Bivariate plot for LogIVS and LogQUICKI controlling for age (residuals).
Figure 23. LogIVS and LogProlactin controlling for age (residuals).
51
Correlation Analysis 3
Age was significantly correlated with LogLVPW (r = 0.544, n = 25) (Figure 24) and each of the serum biomarkers except LogProlactin (Table 11). LogLVPW was significantly correlated with LogFerritn ( r = .389, n = 25) (Figure 25). As with the previous correlation analyses, there were several significant correlations among the serum biomarkers (see Table 10). Notably, LogoxLDL was not significantly correlated with any other biomarker.
When controlling for age, LogLVPW was not significantly associated with any of the serum biomarkers (see Table 9). However, there was a trend towards a significant negative correlation between LogLVPW and LogProlactin (r = -.281, n = 25, p = .09)
(Figure 26). There were correlations among the biomarkers in accordance with correlation analyses 1 and 2, with one additional finding: LogFerritin was negatively associated with LogoxLDL (r = -.413, n = 25, p < .05) (Figure 27).
Table 11. Zero-order and partial correlations for logLVPW and logLeptin, logoxLDL, logCholesterol, logQUICKI, logProlactin, logFerritin, and age
Control Variables logLVPW LogLeptin LogoxLDL LogCholesterol LogQUICKI LogProlactin LogFerritin Age at
sample nonea LogLVPW 1
LogLeptin .250 1
LogoxLDL .185 -.037 1
52
LogCholesterol -.314 -.436* .048 1
LogQUICKI -.091 -.530** .060 -.085 1
LogProlactin -.224 -.185 -.062 -.290 -.181 1
LogFerritin .389* .409* -.067 -.502** -.007 -.078 1
Age at sample .544** .518** .336* -.435* -.415* .022 .665** 1
Age at LogLVPW 1 sample LogLeptin -.044 1
LogoxLDL .003 -.263 1
LogCholesterol -.102 -.273 .229 1
LogQUICKI .176 -.405* .233 -.324 1
LogProlactin -.281 -.230 -.074 -.312 -.189 1
LogFerritin .043 .101 -.414* -.316 .396* -.124 1
a. Cells contain zero-order (Pearson) correlations. * sig. at p < .05, ** sig. at p < .01 Figure 24. LogLVPW and age.
Figure 25. LogLVPW and LogFerritin without controlling for age
53
Figure 26. LogLVPW and LogProlactin controlling for age (residuals).
Figure 27. LogFerritin and LogoxLDL controlling for age (residuals).
54
Correlation Analysis 4
LogLVIDd was significantly correlated with age (r = .355) (Table 12 and Figure
28). Among biomarkers, the same correlations emerged as in the previous correlation analyses (see Tables 9-11).
When controlling for age, there were no significant correlations between
LogLVIDd and any biomarkers (see Table 12). However, LogProlactin demonstrated the strongest negative trend with LogLVIDd than any other echocardiograph measure (r = -
.304, n = 29, p = .058) (Figure 29) and a positive trend was observed between LogLVIDd and LogQUICKI (Figure 30). As with the earlier correlation analyses, there were several significant correlations among biomarkers (see Tables 9-11).
55
Table 12. Zero-order and partial correlations for LogLVIDd and LogLeptin, LogCholesterol, LogQUICKI, LogProlactin, LogFerritin, and age.
Control Variables LogLVIDd LogLeptin LogCholesterol LogQUICKI LogProlactin LogFerritin Age nonea LogLVIDd 1
LogLeptin .179 1
LogCholesterol -.195 -.408* 1
LogQUICKI .045 -.558** -.072 1
LogProlactin -.286 -.168 -.262 -.158 1
LogFerritin .126 .466** -.500** -.056 -.094 1
Age at sample .355* .519** -.431** -.413* -.004 .671** 1
Age LogLVIDd 1
LogLeptin -.007 1
LogCholesterol -.050 -.239 1
LogQUICKI .225 -.442** -.304 1
LogProlactin -.304 -.194 -.292 -.175 1
LogFerritin -.163 .185 -.315 .327* -.123 1
56
Control Variables LogLVIDd LogLeptin LogCholesterol LogQUICKI LogProlactin LogFerritin Age nonea LogLVIDd 1
LogLeptin .179 1
LogCholesterol -.195 -.408* 1
LogQUICKI .045 -.558** -.072 1
LogProlactin -.286 -.168 -.262 -.158 1
LogFerritin .126 .466** -.500** -.056 -.094 1
Age at sample .355* .519** -.431** -.413* -.004 .671** 1
Age LogLVIDd 1
LogLeptin -.007 1
LogCholesterol -.050 -.239 1
LogQUICKI .225 -.442** -.304 1
LogProlactin -.304 -.194 -.292 -.175 1
LogFerritin -.163 .185 -.315 .327* -.123 1 a.Cells contain zero-order (Pearson) correlations. * sig. at p < .05, ** sig. at p < .01
Figure 28. LogLVIDd and age
57
Figure 29. LogLVIDd and LogProlactin controlling for age (residuals).
Figure 30. LogLVIDd and LogQUICKI controlling for age
58
Chapter Four
Discussion
The goal of this study was to identify associations between biomarkers of inflammation, characteristic of metabolic syndrome, and echocardiographic measures of cardiac health in a sample of captive western lowland gorillas. The choice of the serum biomarkers in this study was based on their previously described associations with metabolic syndrome and cardiac disease in humans (Djousse and Gaziano, 2009; Gerstein et al, 1999; Group TDS, 2004; Holvoet et al, 2001; Tsutsui et al, 2002; Judd, 2005;
Söderberg et al, 1999; Sierra-Johnson et al, 2007; Singh, 2007; Limas et al, 2002; Halle et al, 1997; Jehn et al, 2006).
Principal Findings
There were similar patterns in how each variable loaded with each component among the four PCAs, suggesting that the serum biomarkers have a common relationship to each echocardiograph measure. However, this would be expected as each of the four echocardiograph measures are measures of cardiac health. Unexpectedly, the associations between each biomarker and echocardiographic measure were strongly mediated by the effects of age as indicated by the subsequent correlation analyses. Age also mediated most of the associations among the biomarkers.
58 59
This study found a significant negative correlation between EF and insulin sensitivity only when controlling for age. When age is not controlled for, IVS and cholesterol share a significant negative association, while ferritin and IVS share a positive association. In turn, LVPW thickness is positively correlated with ferritin.
These associations, however, do not hold when controlling for age. No significant relationships were found between LVIDd and the serum biomarkers.
Age demonstrated positive associations with leptin, ferritin, and oxLDL and negative associations with cholesterol and QUICKI. Age showed a significant negative association with EF. However, it had significant positive associations with,
IVS, LVPW, and LVIDd.
Among biomarkers, without controlling for age, significant negative associations were observed between leptin and cholesterol, leptin and QUICKI, and ferritin and cholesterol; a positive association was observed between ferritin and leptin. This pattern was evident in all correlation analyses. When age was controlled for, QUICKI demonstrated a positive association with ferritin and a negative association with leptin. These were also present in all correlation analyses. Only correlation analysis 3 included oxLDL. When age was controlled for oxLDL demonstrated a significant negative association with ferritin.
Implications
While the findings of this study are contrary to the predicted hypotheses, they may be suggestive of underlying mechanisms in the development of diabetic-related cardiomyopathy and diastolic dysfunction in this sample. Indeed, the negative
60
association between insulin sensitivity (QUICKI) and EF in this study implies that
high fasting insulin levels are associated with higher ejection fractions. It has been
demonstrated in humans and rabbits that exogenous administration of insulin
promotes a compensatory pathway against sudden heart failure by preserving EF after
a hypoxic insult common during ischemic events (Sasso et al, 2000; van der Horst et
al, 2003; Wong et al, in press). Correspondingly, Di Bello et al (2006) found EF to be
higher in subjects with insulin resistance. Endogenously, heart cells may accomplish
EF preservation by switching from fatty acid oxidation to glucose metabolism
(Schwenk et al, 2008) which allows for more ATP production per molecule of oxygen
consumed (Witteles and Fowler, 2008). This increase in metabolic efficiency allows
the heart to function in low oxygen conditions. However, the elevated levels of
insulin that accompany this phenomenon also accompany the formation of reactive
oxygen species that contribute to progressive cardiac stiffening, fibrosis, and eventual
impaired left ventricular performance (Khavandi et al, 2009). Underscoring the
progressive nature of cardiac dysfunction, Niethammer and colleagues (2008) found
that patients with preserved ejection fraction heart disease also had intermediate
levels of pro-inflammatory markers when compared to healthy controls and subjects with reduced ejection fraction heart disease.
The findings from this study are congruent with similar studies that have failed to demonstrate a direct link between leptin and cardiac disease in humans
(Couillard et al, 1998; Brennan et al, 2007). However, leptin has also been implicated in the preservation of cardiac contractility following ischemia (McGaffin et al, 2008)
61
and has been associated with IVS thickness in hypertensive men (Paolisso et al,
1999). In the present study, leptin showed a negative association with QUICKI
(Figure 14) implying a positive association with insulin. If preserved EF, associated
with low insulin sensitivity, is indicative of greater glucose metabolism in this sample, then leptin may be simultaneously increasing glucose production in the liver while increasing pancreatic insulin secretion (Ruige et al, 2006). However, leptin- induced lipolysis of fatty acids is expected to trigger better peripheral insulin sensitivity and cardiac fatty acid metabolism (Ruige et al, 2006). The results from this study do not support this as leptin shows a negative association with insulin sensitivity.
In humans, obesity is known to increase circulating leptin concentrations
(Prolo et al, 1998) which reduces proper insulin secretion (Seufert et al, 1999). The relationship between QUICKI and leptin in this sample is inconsistent with this interaction. Furthermore, prolactin is known to induce pancreatic β-cell insulin secretion (Holstad and Sandler, 1999). Yet, in this study, prolactin does not increase with leptin as is typical for obese humans (Ben-Jonathan & Hnasko, 2001). Instead, this study showed the relationship between prolactin and leptin to have a trend towards a negative association, albeit not significantly.
It is plausible that increased prolactin levels may have a protective effect on cardiac function by preserving higher EF (Limas et al, 2002). Prolactin and EF do associate together in a positive association, whether controlling for age or not.
However, this relationship was weak and not significant.
62
Despite the preservation of ejection fraction, an uptake imbalance between glucose and fatty acids signals the progression of left ventricular dysfunction
(Khavandi et al, 2009). Leptin and insulin-coupled glucose metabolism only ameliorates the more immediate crisis of heart failure. This suggests that, for these gorillas, the progression of cardiac disease may already be well underway.
The unpredicted significant positive association between LogQUICKI and
LogFerritin (Figure 15) suggests that better insulin sensitivity is somehow associated with increased iron loading for this sample. However, it may also indicate that increased iron loading is disrupting proper insulin secretion (Fernandez-Real et al,
2002). Thus, in this sample, higher QUICKI scores may not indicate normal insulin sensitivity as much as they might suggest improper insulin secretion as the pancreatic
β-cells become dysfunctional with increased iron loading. On the other hand, ferritin may be interfering with glucose uptake. This is consistent with ferritin levels and insulin resistance in human populations (Wrede et al, 2006); however, the precise mechanisms for these interactions remain unknown.
Increasing LVIDd is indicative of left ventricular dilation and increases with age (Lohr, 2005). Hyperprolactinemia has been described as ameliorating left ventricle dilation in patients with heart failure (Limas et al, 2002). In this sample, prolactin demonstrated the strongest trend towards a significant negative association with LVIDd than any other echocardiograph measure. However, prolactin does not increase with age along this cardiac dimension (Table 12).
63
Decreasing EF may well be age-related; however, the preservation of EF is most probably mediated by glucose uptake as hyperinsulinemia and hyperleptinemia ensue. Only with further analysis along age graded classes will this become clearer.
What remains unclear, however, is why age alone should correlate so well with the echocardiograph and biomarker data while most of theses biomarkers do not demonstrate significant associations when age is controlled. Aging hearts do show characteristic morphological and biochemical changes (Lammey et al, 2008), yet, clearly not all gorillas over a certain age should acquire heart disease without an underlying pathophysiology. It should be noted that in the present sample, 50% of the study subjects were over the age of 20. Only one individual over the age of 20 did not have heart disease.
In humans, the relationship between total cholesterol and insulin may underscore the complexity of cholesterol metabolism during the progression of cardiomyopathy. Rauchhaus and colleagues (2003) found lower total serum cholesterol was associated with lower ejection fractions in patients suffering from chronic heart failure. It is also been reported that low QUICKI scores are associated with increased cholesterol levels (Hoenig and Sellke, in press). While the association between QUICKI and cholesterol in this study was non-significant, it did demonstrate a negative trend. This trend continued with prolactin as well. This may indicate that high cholesterol offers a protective function against prolactin induced immune activation (Hu et al, 1998; Rauchhaus et al, 2003).
64
Great Ape Heart Disease
Incidents of captive gorilla cardiovascular disease have been described in the literature. A case of hypertension leading to congestive heart failure and fibrosing cardiomyopathy was highlighted by Miller et al. in 1999. Kenny et al. (1994) found that six out of eight gorillas had atherosclerosis, yet all eight died of aortic dissection.
In Schulman et al’s (1995) review, 11 out of 16 captive gorillas died of fibrosing cardiomyopathy; although, eight out of 19 gorillas had arteriosclerosis and or moderate artherosclerosis, it proved only attributable in one death (Schulman et al,
1995).
Recent reports of great ape deaths due to cardiac events were diagnosed postmortem as fibrosing cardiomyopathy (Lammey et al, 2008; Varki et al, 2009).
This differs from human cardiac disease, which is almost exclusively characterized by arterial occlusions and coronary thrombosis via an inflammatory atherosclerotic pathway (Judd, 2005). By examining the necropsy reports over 14-16 year period at four primate research centers, Varki et al (2009) found that heart disease in chimpanzees accounted for 14-36% of mortality. However, while histological staining confirmed the presence of interstitial fibrosing cardiomyopathy, no major blockages and only mild atherosclerosis was observed in the major blood vessels.
Varki et al (2009) concluded that underlying pathology for heart disease in the chimpanzee is not the same as in humans. However, the symptoms of heart disease in
65
great apes may go unnoticed as diagnosis of heart disease in great apes comes well after the fact.
There is a common route to the different destinations of atherosclerosis, aortic dissection, and fibrosing cardiomyopathy. It is induced via inflammation of the cardiovascular system by reactive oxygen species (Videan et al, 2009). The reason that gorillas and chimpanzees have one outcome, while humans have another, may depend on which genes are upregulated and downregulated during inflammatory crises. Research in the regulation of matrix metalloproteinase clearance of collagen from cardiac tissue (Rockman et al, 2004) and endocytosis of circulating cholesterol
(Ding et al, 2007) suggests that the two apes have diverged genetically in this regard.
Humans, compared to both gorillas and chimps, are phenotypically prone to atherosclerosis despite comparably lower total serum cholesterol loads (Varki et al,
2009). Gorillas and chimpanzees are phenotypically hypercholestolemic compared to humans, yet do not suffer from higher rates of atherosclerotic heart disease and may be less able to degrade interstitial collagens from cardiac tissue (Rockman et al,
2004).
The resulting EF and insulin sensitivity association in this study is in keeping with findings by Videan and colleagues (2009). They found that captive chimpanzees have a greater risk of heart disease than humans and that risk is associated with higher levels of insulin and insulin growth like factor 1. The implications for heart disease in non-human great apes suggest that hyperinsulinemia is associated with cardiomyopathies that induce fibrosis (Videan et al, 2009).
66
Given that much is unknown about the etiology of fibrosing cardiomyopathy
and the implication of inflammation as a possible route to aortic dissection, this study
warns against dismissing great ape heart disease as being so unlike that in humans
that similar biomarkers of inflammation, and the inflammation pathway involved in
metabolic syndrome should go unutilized as a way to assess disease progression and characterization.
Confounding Factors
Confounding variables include the effects of the different anesthetics used during immobilization of the gorillas on each of the serum biomarkers. In addition, several individuals in this sample were taking medication for heart disease. The duration and type of pharmaceutical therapy likely affected the present data. It has been recently documented that long term use of angiotensin I converting enzyme
(ACE) inhibitors may decrease serum leptin concentrations in rats (Santos et al,
2009). While captive gorillas have been prescribed ACE inhibitors, it is not clear what (if any) medications each of the gorillas in this study was taking at the time of serum sampling.
Assessment
This study provided a unique opportunity to explore the multiple variables
associated with metabolic syndrome as they relate to cardiac parameters used in the
diagnosis of heart disease in captive gorillas. Of the serum biomarkers, insulin
67
sensitivity appears to be the measure most associated when controlling for age. The fact that age mediates this association and that this relationship runs counter to predictions, suggests either a signaling or secretion dysfunction of insulin might be explained by interactions with an age-related increase in iron loading. These data suggest that no direct association exists between cardiac thickening or dilation and these biomarkers that are not related to age progression (i.e. ferritin). However, prolactin may prove useful in future prediction on left ventricular dilation.
Study Limitations
Caution should be taken when inferring these upon the larger captive gorilla population. This study was limited in sample size, but not in scope. The use of PCA analysis is an exploratory procedure that is best met when sample size is substantially greater than the number of variables. This study had a diminishing number of cases per PCA given incomplete data and a large number of variables to consider. Indeed, the PCAs proved unable to trim the data set down more than one biomarker. Finally, the correlation analyses demonstrate associations but not causality.
Diagnoses of metabolic syndrome are incomplete without a body mass index.
To date, a measure of body condition has not been standardized for gorillas. This study was not making assumptions about whether the sample subjects have metabolic syndrome, only if the biomarkers associated with metabolic syndrome also associate with these cardiac parameters.
68
Future Directions
Because it is unlikely that more precise techniques in the determination of gorilla insulin sensitivity will gain wide acceptance, given the inherent invasiveness of current procedures, it remains imperative to examine glucose-insulin levels along age-graded, sex graded, and body mass index graded classes. This will provide future research with a basic expectation of insulin sensitivity during captive gorilla growth and development. Thus, QUICKI may prove useful in this regard. However, QUICKI would be more beneficial with multiple samples and a record of the exact length of time each gorilla has been fasting. This may provide us with the basic parameters for gorilla pancreatic β-cell production without the need for extended sedation and insulin-glucose infusions.
Further work is required to determine the nature of the relationship between leptin and prolactin on one hand and insulin on the other. It is not clear whether the results of this study indicate a direct interaction between these hormones and EF preservation or whether these results are spurious given sample size.
69
Chapter 5
Conclusions
Metabolic syndrome is a compound condition that has been investigated by examining fluctuations in inflammatory biomarkers. The choice of biomarkers utilized in this study provided an indication that insulin sensitivity may increase with decreasing cardiac efficiency. However, it is possible that the relationship between increasing insulin with increasing ejection fraction is either spurious or indicative of a complex interaction
between glucose uptake and cardiac cells. It could also suggest improper insulin secretion
at the pancreatic level; however, this assessment is speculative until direct analysis can be
made.
The results of this study also suggest that the relationships among biomarkers are
strongly mediated by age. Therefore, further investigations with larger sample sizes
might find it relevant to establish meaningful age groups, based on EF, IVS, LVPW, and
LVIDd.
69 70
LITERATURE CITED
Ahima RS, Osei SY. 2004. Leptin Signaling. Physiology & Behavior 81:223-241.
Ando K, Fujita T. 2009. Metabolic Syndrome and Oxidative Stress. Free Radical Biology & Medicine 47:213-218.
Baitchman E, Calle P, Clippinger T, Deem S, James S, Raphael B, Cook R. 2006. Preliminary Evaluation of Blood Lipid Profiles in Captive Western Lowland Gorillas (Gorilla gorilla gorilla). Journal of Zoo and Wildlife Medicine 37(2):126- 129.
Bajraktari G, Koltai MS, Ademaj F, Rexhepaj N, Qirko S, Ndrepepa G, Elezi S. 2006. Relationship between insulin resistance and left ventricular diastolic dysfunction in patients with impaired glucose tolerance and type 2 diabetes. International Journal of Cardiology 110:206-211
Bellisari A. 2008. Evolutionary origins of Obesity. Obesity reviews 9:165-180.
Beltowski J. 2006. Leptin and atherosclerosis. Atherosclerosis 189:47-60.
Benefit BR, McCrossin ML. 1995. Miocene hominoids and hominid origins. Annu. Rev. Anthropol. 24:237-256
Ben-Jonathan N, Hnasko R. 2001. Dopamine as a Prolactin (PRL) Inhibitor. Endocrine Reviews 22(6):724-763
Bindon JR, Baker PT. 1997. Bergmann's Rule and the Thrifty Genotype. American Journal of Physical Anthropology 104:201-210.
Björntorp P. 2000. Adipose Tissue. eJIFCC 12(3):1-5.
Blundell J, Cooling J. 2000. Routes to obesity: phenotypes, food choices and activity. British Journal of Nutrition 83(suppl.1):S33-S38.
Blunt W. 1976. The Ark in the Park: The Zoo in the Nineteenth Century. London.
70 71
Boullier A, Bird DA, Chang M-K, Dennis EA, Freidman P, Gillotte-Taylor K, Horkko S, Palinski W, Quehenberger O, Shaw P et al. 2001. Scavenger Receptors, Oxidized LDL, and Atherosclerosis. Annals of the New York Academy of Sciences 941(1):214-223.
Bourne GH, Cohen M. 1975. The Gentle Giants: The Gorilla Story. New York, NY: G.P. Putnam’s Sons
Brennan AM, Li TY, Kelesidis I, Gavrila A, Hu FB, Mantzoros CS. 2007. Circulating leptin levels are not associated with cardiovascular morbidity and mortality in women with diabetes: a prospective cohort study. Diabetologia 50:1178-1185
Brouwers A, Langlois M, Delanghe J, Billiet J, Buyzere MD, Vercaemst R, Rietzschel E, Bernard D, Blaton V. 2004. Oxidized Low-Density Lipoprotein, Iron Stores, and Haptoglobin Polymorphism. Atherosclerosis 176(189-195).
Brydon L, Magid K, Steptoe A. 2006. Platelets, coronary heart disease, and stress. Brain, Behavior, and Immunity 20:113-119
Cefalu WT. 2008. Diabetic dyslipidemia and the metabolic syndrome. Clinical Research & Reviews 2:208-222
Cejkova P, Fojtikova M, Cerna M. 2009. Immunomodulatory Role of Prolactin in Diabetes Development. Autoimmunity Reviews In Press.
Cinti S. 2007. The Adipose Organ Chapter 1. In: Fantuzzi G, Mazzone T, editors. Nutrition and Health: Adipose Tissue and Adipokines in Health and Disease. Totowa, New Jersey: Humana Press.
Clark C, Newgard CB. 2007. Hepatic Regulation of Fuel Metabolism. In: Saltiel AR, Pessin JE, editors. Mechanisms of Insulin Action. Austin, Texas: Landes Bioscience.
Clifford SL, Abernathy KA, White LJT, Tutin CEG, Bruford MW, Wickings EJ. 2003. Genetic studies of western gorillas. In: Taylor AB, Goldsmith ML, editors. Gorilla Biology: A Multidisciplinary Perspective. Canbridge, UK: Cambridge University Press.
Coe JC, Lee GH. 1996. One-Hundred Years of Evolution in Great Ape Facilities in American Zoos. Proceedings of The AZA 1995 Western Regional Conference. Bethesda, MD: American Zoo and Aquarium Association.
Couillard C, Lamarche B, Mauriege P, Cantin B, Dagenais GR, Moorjani S. 1998. Leptinemia is not a risk factor for ischemic heart disease in men. Prospective results from the Quebec Cardiovascular Study. Diabetes Care 21:782-786.
72
Courtney CH, Olefsky JM. 2007. Insulin Resistance. In: Saltiel AR, Pessin JE, editors. Mechanisms of Insulin Action. Austin, Texas: Landes Bioscience.
Cousins D. 1983. Mortality factors in captive gorillas. International Zoo News 30 (3), No. 179, 5-17
Diaz BB, Perez MdCR, Leon ACd. 2006. The Vicious Circle of Leptin and Obesity. Current Nutrition & Food Science 2:361-373.
Di Bello V, Santini F, Di Cori A, Pucci A, Palagi C, Donne MGD, Fierabracci P, Marsili A, Talini E, Giannetti M, Biadi O, Balbarini A, Mariani M, Pinchera A. 2006. Obesity Cardiomyopathy: Is it a reality? An ultrasonic tissue characterization study. Journal of the American Society of Echocardiography 19(8):1063-1071
Dierenfeld E. 1997. Captive Wild Animal NutritionL A Historical Perspective. Proceedings of the Nutrition Society 56:989-999.
Ding K, McDonough SJ, and Kullo IJ. 2007. Evidence for positive selection in the C- terminal domain of the cholesterol metabolism gene PCSK9 based on phylogenetic analysis in 14 primate species.PLoS ONE 2(10): e1098
Ding K, Kullo IJ. 2009. Evolutionary genetics of coronary heart disease. Circulation 119:459-467
Djousse L, Gaziano JM. 2009. Dietary Cholesterol and Coronary Artery Disease: A Systematic Review. Current Atherosclerosis Reports 11:418-422.
Doehner W, Rauchhaus M, Ponilowski P, Godsland IF, von Haehling S, Okonko DO, Leyva F, Proudler AJ, Coats AJS, Anker SD. 2005. Impaired Insulin Sensitivity as an Independent Risk Factor for Mortality in Patients with Stable Chronic Heart Failure. Journal of the American College of Cardiology 46(6):1019-1026
Doran D, McNeilage A, Greer D, Bocian C, Mehlman P, Shah N. 2002. Western Lowland Gorilla Diet and Resource Availability: New Evidence, Cross Site Comparisons, and Reflections on Indirect Sampling Methods. American Journal of Primatology 58:91-116.
Doran-Sheehy DM, Greer D, Mongo P, Schwindt D. 2004 Impact of ecological and social factors on ranging in western gorillas. American Journal of Primatology. 64:207-222
Feistner ATC, Price EC. 2002. Zoos and Universities: Collaborating for Primate Conservation. Evolutionary Anthropology Suppl. 1 (2002):7-11.
73
Ferrante AW. 2007. Macrophages, Adipocytes, and Obesity. In: Fantuzzi G, Mazzone T, editors. Nutrition and Health: Adipose Tissue and Adipokines in Health and Disease. Totowa, New Jersey: Humana Press.
Fernandez-Real JM, Lopez-Bermejo A, Ricart W. 2002. Cross-Talk Between Iron Metabolism and Diabetes. Diabetes 51(2347-2354).
Fleagle JC. 1999. Primate Adaptation and Evolution. San Diego, CA: Academic Press.
Gerstein HC, Pais P, Pogue J, Yusuf S. 1999. Relationship of Glucose and Insulin Levels to a Risk of Myocardial Infarction: A Case-Control Study. Journal of the American College of Cardiology 33(3):612-619.
Grandi AM, Maresca AM, Giudici E, Laurita E, Marchesi C, Solbiati F, Nicolini E, Guasti L, Venco A. 2006. Metabolic Syndrome and Morphofunctional Characteristics of the Left Ventricle in Clinically Hypertensive Nondiabetic Patients. American Journal of Hypertension 19:199-205.
Green A, Basile R, Rumberger JM. 2006. Transferrin and Iron Can Induce Insulin Resistance of Glucose Tansport in Adipocytes. Metabolism Clinical and Experimental 55:1042-1045.
Group TDS. 2004. Prediction of the risk of cardiovascular mortality using a score that includes glucose as a risk factor. The DECODE Study Diabetologia 47:2118- 2128.
Groves C. 2003. A History of Gorilla Taxonomy. In: Taylor AB, Goldsmith ML, editors. Gorilla Biology: A Multidisciplinary Perspective. Canbridge, UK: Cambridge University Press.
Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, Gordon DJ, Krauss RM, Savage PJ, Smith SC, Spertus JA, Costa F. 2005 Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation 112(7):2735- 2752
Guenno GL, Chanseaume E, Ruivard M, Morio B, Mazur A. 2007. Study of Iron Metabolism Disturbances in a Animal Model of Insulin Resistance. Diabetes Research and Clinical Practice 77:363-370.
Halle M, Konig D, Berg A, Keul J, Baumstark MW. 1997. Relationship of serum ferritin concentrations with metabolic cardiovascular risk factors in men without evidence for coronary artery disease. Atherosclerosis 128:235-240.
74
Hatt JM, Liesegang A. 2002. Nutrition of western Lowland Gorillas (Gorilla g. gorilla) and Sumatran Orang-utans (Pongo pygmaeus abelii) in Captivity. Heidleberg, Germany: European Association of Zoo and Wildlife Veterinarians (EAZWV).
Hill JO, Peters JC. 2002. Biomarkers and functional foods for obesity and diabetes. British Journal of Nutrition 88:suppl. 2, S213-S218.
Hoenig MR, and Sellke FW. In press. Insulin resistance is Associated with Increased Cholesterol Synthesis, decreased Cholesterol Absorption and Enhanced Lipid Response to Statin Therapy. Atherosclerosis.
Hoff M, Hoff K, Maple T. 1998. Behavioural Response of a Western Lowland Gorilla (Gorilla gorilla gorilla) Group to the Loss of the Silverback Male at Zoo Atlanta. International Zoo Yearbook 36:90-96.
Holstad M, and Sandler S. 1999. Prolactin protects against diabetes induced by multiple low doses of streptozotocin in mice. Journal of Endocrinology 163:229-234
Holt RIG. 2008. Medical causes and consequences of hyperprolactinaemia. Journal of Psychopharmacology 22(2): S28-S37
Holvoet P, Mertens A, Verhamme P, Bogarts K, Beyens G, Verhaeghe R, Collen D, Muls E, Van der Werf F. 2001. Circulating oxidized LDL is a useful marker for identifying patients with coronary artery disease. Arterioscler Thromb Vasc Biol. 21:844-848
Horkko S, Binder CJ, Shaw PX, Chang M-K, Silverman G, Palinski W, Witztum JL. 2000. Immunological Responses to Oxidized LDL`. Free Radical Biology & Medicine 28(12):1771-1779.
Houseknecht KL, Spurlock ME. 2003. Leptin regulation of lipid homeostasis: dietary and metabolic implications. Nutrition Research Reviews 16:83-96.
Hu Z, Zhuang L, Dufau M. 1998. Prolactin Gene Diversity: Structure and Regulation. TEM 9(3):94-102.
Jackson MB, Ahima RS. 2007. Leptin. In: Fantuzzi G, Mazzone T, editors. Adipose Tissue and Adipokines in Health and Disease. Totowa, New Jersey: Humana Press.
Jehn M, Clark JM, Guallar E. 2004. Serum Ferritin and Risk of the Metabolic Syndrome in U.S. Adults. Diabetes Care 27(10):2422-2428.
Judd SJ, editor. 2005. Cardiovascular Diseases and Disorders Sourcebook. third ed. Detroit: Omnigraphics. 713 p.
75
Junge RE, Mezei LE, Muhlbauer MC, and Weber M. 1998. Cardiovascular evaluation of lowland gorillas. JAVMA 212(3):413-415
Kadar A, Glasz T. 2001. Development of Atherosclerosis and Plaque Biology. Cardiovascular Surgery 9(2):109-121.
Kahn SE. 2003. The Relative Contributions of Insulin Resistance and Beta-Cell Dysfunction to the Pathophysiology of Type 2 Diabetes. Diabetologia 46:3-19.
Katz A, Nambi SS, Mather K, Baron AD, Follmann DA, Sullivan G, Quon MJ. 2000. Quantitative Insulin Sensitivity Check Index: A Simple, Accurate Method for Assessing Insulin Sensitivity In Humans. The Journal of Clinical Endocrinology & Metabolism 85(7):2402-2410.
Kawata K. 2008. Zoo Animal Feeding: A Natural History Viewpoint. Der Zoologische Garten 78:17-42.
Kellerer M, Lammers R, Fritsche A, Strack V, Machicao F, Borboni P, Ullrich A, Haring HU. 2001. Insulin inhibits receptor signalling in HEK293 cells at the level of janus kinase-2: a potential mechanism for hyperinsulinaemia-associated leptin resistance. Diabetologia 44:1125-1132.
Kennedy GC. 1953. The role of depot fat in the hypothalamic control of food intake in the rat. Proc. Roy. Soc. B 140:578-592
Kenny DE, Cambre RC, Alvarado TP, Prowten AW, Allchurch AF, Marks SK, and Zuba JR. 1994. Aortic Dissection: An Important Cardiovascular Disease in Captive Gorillas (Gorilla gorilla gorilla). Journal of Zoo and Wildlife Medicine 25(4): 561-568
Khavandi K, Khavandi A, Asghar O, Greenstein A, Withers S, Heagerty AM, Malik RA. 2009. Diabetic Cardiomyopathy – a distinct disease? Best Practice & Research Clinical Endocrinology & Metabolism 23:347-360
Koeppen BM, Stanton BA, editors. 2008. Berne and Levy Physiology Sixth Edition. Sixth ed. Philadelphia, PA: Mosby Elsevier.
Lammey ML, Baskin GB, Gigliotti AP, Lee DR, Ely JJ, Sleeper MM. 2008. Interstitial Myocardial Fibrosis in a Captive Chimpanzee (Pan troglodytes) Population. Comparative Medicine 58(4):389-394.
Limas CJ, Kroupis C, Haidaroglou A, Cokkinos DV. 2002. Hyperprolactinaemia in patients with heart failure: clinical and immunogenetic correlations. European Journal of Clinical Investigation 32:74-78
76
Lindop GBM. 2007. Pathology of the Heart. Surgery 25(5):189-197.
Lohr JL. 2005. Introduction to Echocardiography. In: Paul A. Iaizzo, editor. Handbook of Cardiac Anatomy, Physiology, and Devices. Totowa, New Jersey: Humana Press
Maple T, Hoff MP. 1982. Gorilla Behavior. New York, New York: Van Nostrand Reinhold Co.
Margetic S, Gazzola C, Pegg GG, and Hill RA. 2002. Leptin: a review of its peripheral actions and interactions. International Journal of Obesity. 26:1407-1433
Martin R, Perez E, He Y, jr RD, Millard W. 2000. Leptin Resistance is Associated with Hypothalamic Lepin mRNA and Protein Downregulation Metabolism 49(11):1479-1484.
Martin RL. 2000. Leptin Resistance Reduces Growth Hormone Secretion and Contributes to the Pathogenesis of Obesity: University of Florida. 180 p.
Masson S, Latini R, Salio M, Fiordaliso F. 2005. Cardiac Fibrosing and Aging. In: John Varga, David A Brenner, Sem H Phan, editors. Fibrosis Research: Methods and Protocols. Totowa, New Jersey: Humana Press
McGaffen KR, Sun CK, Rager JJ, Romano LC, Zou B, Mathier MA, O’Doherty RM, McTiernan CF, O’Donnell CP. 2008. Leptin signaling reduces the severity of cardiac dysfunction and remodeling after chronic ischaemic injury Cardiovascular Research 77:54-63
Medicine Io. 2005. Dietary Reference Intakes for Energy, Carbohydrates, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. Washington, D.C.: Institute of Medicine of the National Academies.
Meehan T, Lowenstine L. 1994. Causes of Mortality in Captive Lowland Gorillas: A Survey of the SSP Population Proceedings AAZV 216.
Mehlman P. 2008. Current Stratus of Wild Gorilla Populations and Strategies for Their Conservation. In: Stoinski T, Steklis H, Mehlman P, editors. Conservation in the 21st Century: Gorillas as a Case Study. New York, New York: Springer.
Mertens A, Holvoet P. 2001. Oxidized LDL and HDL: antagonists in atherothrombosis. FASEB J 15.
Miller C, Schwartz A, jr JB, Bell M. 1999. Chronic Hypertension with Subsequent Congestive Heart Failure in a Weastern Lowland Gorilla (Gorilla gorilla gorilla). Journal of Zoo and Wildlife Medicine 30(2):262-267.
77
Morris JN. 2007. Uses of epidemiology. International Journal of Epidemiology 36:1165- 1172
Neel JV. 1962. Diabetes Mellitus: A “Thrifty” Genotype Rendered Detrimental by “Progress”? Am J Hum Genet 14:353–362.
Negre-Salvayre A, Dousset N, Ferretti G, Bacchetti T, Curatola G, Salvayre R. 2006. Antioxidant and Cytoprotective Properties of High Density Lipoproteins in Vascular Cells. Free Radical Biology & Medicine 41:1031-1040.
Niethammer M, Sieber M, von Haehling S, Anker SD, Munzel T, Horstick G, Genth- Zotz S. 2008. Inflammatory pathways in patients with heart failure and preserved ejection fraction. International Journal of Cardiology 129:111-117
Orino K, Watanabe K. 2008. Molecular, Physiological and Clinical Aspects of the Iron Storage Protein Ferritin. The Veterinary Journal 178:191-201.
Paolisso G, Tagliamonte MR, Galderisi M, Zito GA, Petrocelli A, Carella A, de Divitiis O, and Varricchio M. 1999. Plasma Leptin Level is Associated with Myocardial Wall Thickness in Hypertensive Insulin-Resistant Men. Hypertension 34:1047- 1052
Patterson F, Matevia M. 2001. The Status of Gorillas Worldwide. In: Galdikas, Erickson, Briggs, Sheeran, Shapiro, Goodall, editors. All Apes Great and Small Kluwer Academic/Plenum Publishers.
Plumbtree AJ, McNeilage A, Hall JS, Williamson EA. 2003. The current status of gorillas and threats to their existence at the beginning of the new millennium. In: Taylor AB, Goldsmith ML, editors. Gorilla Biology: A Multidisciplinary Perspective. Canbridge, UK: Cambridge University Press.
Popovich D, Dierenfeld E. 1997. Gorilla Nutrition. In: Ogden J, Wharton D, editors. Management of Gorillas in Captivity: Husbandry Manual, Gorilla Species Survival Plans: American Association of Zoos and Aquariums.
Prolo P, Wong M, Licinio J. 1998. Leptin. The Int J of Biochem and Cell Biol 30:1285- 1290.
Quon MJ. 2001. Editorial: Limitations of Fasting Glucose to Insulin Ratio as an Index of Insulin Sensitivity. The Journal of Clinical Endocrinology & Metabolism 86(10):4615-4617.
Raaz D, Wallaschofski H, Stumpf C, Yilmaz A, Cicha I, Klinghammer L, Daniel WG, Lohmann T, Garlichs CD. 2006. Increased Prolactin in Acute Coronary
78
Syndromes as Putative Co-activator of ADP-stimulated P-Selectin Expression. Horm Metab Res 38:767-772
Rajpathak SN, Wylie-Rosett J, Gunter MJ, Negassa A, Kabat GC, Rohan TE, Crandall J, Group DPPDR. 2009. Biomarkers of Body Iron Stores and Risk of Developing Type 2 Diabetes. Diabetes, Obesity and Metabolism 11:472-479.
Ratcliffe H. 1965. Age and Environment as Factors in the Nature and Frequency of Cardiovascular Lesions in Mammals and Birds in the Philadelphia Zoological Garden. Annals of the New York Academy of Sciences 127(1):715-735.
Rauchhaus M, Clark AL, Doehner W, Constantinos D, Bolger A, Sharma R, Coats AJS, and Anker SD. 2003. The Relationship Between Cholesterol and Survival in Patients With Chronic Heart Failure. Journal of the American College of Cardiology 42(11): 1933-1940
Rector SR, Warner SO, Liu Y, Hinton PS, Sun GY, Cox RH, Stump CS, Laughlin MH, Dellsperger KC, Thomas TR. 2007. Exercise and diet induced weight loss improves measures of oxidative stress and insulin sensitivity in adult with characteristics of metabolic syndrome. Am J Physiol Endocrinol Metab. 293:E500-E506
Remis M. 2003. Are Gorillas the Vacuum Cleaners of the Forest Floor? The Roles of Body Size, Habitat, and Food Preferences on Dietary Flexibility and Nutrition. In: Taylor A, Goldsmith ML, editors. Gorilla Biology: A Mutlidisciplinary Perspective. Cambridge, UK: Cambridge University Press.
Remis MJ, Dierenfeld ES. 2004. Digesta Passage, Digestibility and Behavior in Captive Gorillas Under Two Dietary Regimens. International Journal of Primatology 25(4).
Rietzschel ER, Langlois M, Buyzere MLD, Segers P, Bacquer DD, Bekaert S, Cooman L, Oostveldt PV, Verdonck P, Backer GGD et al. 2008. Oxidized Low-Density Lipoprotein Cholesterol is Associated with Decrease in Cardiac Function Independent of Vascular Alterations. Hypertension 52:535-541.
Robbins, M. & Williamson, L. 2008. Gorilla beringei. In: IUCN 2010. IUCN Red List of Threatened Species. Version 2010.1.
Robert L. 2005. Matrix remodeling and atherosclerosis effects of age. In: John Varga, David A Brenner, Sem H Phan, editors. Fibrosis Research: Methods and Protocols. Totowa, New Jersey: Humana Press
79
Rockman MV, Hahn MW. Soranzo N, Loisel DA, Goldstein DB, and Wray GA. 2004. Positive selection on MMP3 regulation has shaped heart disease risk. Current Biology 14:1531-1539
Rogers M, Abernathy K, Bermejo M, Cipolletta C, Doran D, McFarland K, Nishihara T, Remis M, Tutin C. 2004. Western Gorilla Diet: A Synthesis From Six Sites. American Journal of Primatology 64:173-192.
Ruige JB, Mertens I, Considine RV, Paelinck BP, and Van Gaal LF. 2006. Opposite effects of insulin-like molecules and leptin in coronary heart disease of type 2 diabetes. International Journal of Cardiology 111:19-25
Ryder OA. 2005Conservation genomics: applying whole genome studies to species conservation efforts. Cytogenet Genome Res 108:6-15
Santos EL, de Picoli Souza K, da Silva ED, Batista EC, Martins PJ, D’Almeida V, Pesquero JB. 2009. Long term treatment with ACE inhibitor enalapril decreases body weight gain and increases life span in rats. BioChem. Pharmacol 78(8):951- 958
Sasso FC, Carbonara O, Cozzolino D, Rambaldi P, Mansi L, Torella D, Gentile S, Turco S, Torello R, and Salvatore T. 2000. Effects of Insulin-Glucose Infusion on Left Ventricular Function at Rest and During Dynamic Exercise in Healthy Subjects and Noninsulin Dependent Diabetic Patients. Journal of the American College of Cardiology 36(1):219-226
Schmidt DA, Ellersieck MR, Cranfield MR, Karesh WB. 2006. Cholesterol values in free-ranging gorillas (Gorilla gorilla gorilla and Gorilla beringei) and Bornean Orangutans (Pongo pygmaeus). Journal of Zoo and Wildlife Medicine 37(3):292- 300
Schulman FY, Farb A, Virmani R, Montali RJ. 1995. Fibrosing cardiomyopathy in captive western lowland gorillas (Gorilla gorilla gorilla) in the United States: A retrospective study. Journal of Zoo and Wildlife Medicine 26(1):43-51
Schwenk R, Luiken J, Bonen A, Glatz J. 2008. Regulation of Sarcolemmal Glucose and Fatty Acid Transporters in Cardiac Disease. Cardiovascular Research 79:249-258.
Scott NA, McManamon R, Strobert E, Cipolla GD, Tarazona N, Swenson RB. 1995. In vivo diagnosis of coronary artery disease in a western lowland gorilla (Gorilla gorilla gorilla). Journal of Zoo and Wildlife Medicine 26(1):139-143
Serri O, Li L, Mamputu JC, Beauchamp MC, Maingrette F, Renier G. 2006. The Influences of Hyperprolactinemia and Obesity on Cardiovascular Risk Markers: Effects of Caberoline Therapy. Clinical Endocrinology 64(4):366-373
80
Seufert J, Kieffer TJ, Leech CA, Holz GG, Moritz W, Ricordi C, Habener JF. 1999. Leptin Supression of Insulin Secretion and Gene expression in Human Pancreatic Islets: Implication for the Development of Adipogenic Diabetes Mellitus. Journal of Clinical Endocrinology and Metablosim 84(2):670-676.
Shen GX. 2006. Oxidatively Modified Low-Density Lipoproteins and Thrombosis. In: Cheema SK, editor. Biochemistry of Atherosclerosis. New York, NY: Springer.
Sikaris KA. 2004. The Clinical Biochemistry of Obesity. Clin Biochem Rev 25:165-181
Sierra-Johnson J, Romero-Corral A, Lopez-Jimenez F, Gami AS, Kuniyoshi FHS, Wolf R, Somers VK. 2007. Relation of increased leptin concentrations to history of myocardial infarction and stroke in the United States population. Am J. Cardiol. 100:234-239
Simmen B, Charlot S. 2003. A comparison of taste thresholds for sweet and astringent- tasting compounds in great apes.Biologies 326:449-455
Singh B, Mallika V, Goswami B. 2007. Metabolic syndrome: Diagnosis, potential markers and management -an update. Clinica Chimica Acta 380:4-12.
Söderberg S, Ahren B, Jansson JH, Johnson O, Hallmans G, Asplund K. 1999. Leptin is associated with increased risk of myocardial infartion. Journal of Internal Medicine 246:409-418
Stehno-Bittel L. 2008. Intricacies of Fat. Physical Therapy 88(11):1265-1278
Stewart KJ, Sicotte P, Robbins MM. 2001. Mountain gorillas of the Virungas: a short history. In: Robbins MM, Sicotte P, Stewart KJ, editors. Mountain Gorillas: Three Decades of Research at Karisoke. Cambridge, UK: Cambridge University Press.
Stoinski TS, Lukas KE, Hutchins M. 2008. Zoos and Conservation: Moving beyond a piecemeal approach. In: Stoinski T, Steklis H, Mehlman P, editors. Conservation in the 21st Century: Gorillas as a Case Study. New York, New York: Springer.
Sullivan JL. 1999. Iron versus Cholesterol -Perspectives on the Iron and Heart Disease Debate. Journal of Clinical Epidemiology 49(12):1345-1352.
Swan JW, Ankar SD, Walton C, Godsland IF, Clark AL, Leyva F, Stevenson JC, Coats AJS. 1997. Insulin Resistance in Chronic Heart Failure: Relation to Severity and Etiology of Heart Failure. Journal of the American College of Cardiology 30(2):527-532
Taylor SH. 1971. Insulin and Heart Failure. British Heart Journal 33:329-333
81
Tentolouris N, Argryakopoulou G, Katsilambros N. 2008. Perturbed Autonomic Nervous System Function in Metabolic Syndrome. Neuromolecular Medicine 10:169-178.
Trupkiewicz JG, McNamara TS, Weidenheim KM, Cook RA, Grenell SL, Factor SM. 1995. Cerebral infarction associated with coarctation of the aorta in a lowland gorilla (Gorilla gorilla gorilla). Journal of Zoo and Wildlife Medicine 26(1):123- 131
Tsutamoto T, Wada A, Matsumoto T, Maeda K, Mabuchi N, Hayashi M, Tsutsui T, Ohnishi M, Sawaki M, Fujii M et al. 2001. Relationship Between Tumor Necrosis Factor-Alpha Production and Oxidative Stress in the Failing Hearts of Patients With Dilated Cardiomyopathy. Journal of the American College of Cardiology 37(8):2086-2092.
Tsutsui T, Tsutamoto T, Wada A, Maeda K, Mabuchi N, Hayashi M, Ohnishi M, Kinoshita M. 2002. Plasma Oxidized Low-Density Lipoprotein as a Prognostic Predictor in Patients With Chronic Congestive Heart Failure. Journal of the American College of Cardiology 39(6):957-962.
Van der Horst IC, Zijlstra F, van’t Hof AW, Doggen CJ, der Boer MJ, Suryapranata H, Hoorntje JC, Dambrink JH, Grans RO, and Bilo HJ. 2003. Glucose-insulin- potassium infusion inpatients treated with primary angioplasty for acute myocardial infarction: The glucose-insulin-potassium study: A randomized trial. Journal of the American College of Cardiology 42:784-791
Varki N, Anderson D, Herndon JG, Pam T, Gregg CJ, Cheriyan M, Murphy J, Strobert E, Fritz J, Else JG et al. 2009. Heart Disease is common in humans and chimpanzees, but is caused by different pathological processes. Evolutionary Applications 2(1):101-112.
Videan EN, Heward CB, Chowdhury K, Plummer J, Su Y, and Cutler RG. 2009. Comparison of biomarkers of oxidative stress and cardiovascular disease in humans and chimpanzees (Pan troglodytes). Comparative Medicine 59(3): 287- 296.
Wallander M, Soderberg S, Norhammar A. 2008. Leptin -a predictor of abnormal glucose tolerance and prognosis in patients with myocardial infarction and without previously known Type 2 diabetes. Diabetic Medicine 25:949-955.
Wallaschofski H, Kobsar A, Sokolova O, Siegemund A, Stepan H, Faber R, Eigenthaler M, Lohmann T. 2004. Differences in Platelet Activation by Prolactin and Leptin. Horm Metab Res 36:453-457
Walsh, P.D., Tutin, C.E.G., Oates, J.F., Baillie, J.E.M., Maisels, F., Stokes, E.J., Gatti, S., Bergl, R.A., Sunderland-Groves, J. & Dunn. A. 2008. Gorilla gorilla. In: IUCN
82
2010. IUCN Red List of Threatened Species. Version 2010.1.
Watson RT, Saltiel AR, Pessin JE, Kanzaki M. 2007. Subcellular Compartmentalization of Insulin Signaling Processes and GLUT4 Trafficking Events. In: Saltiel AR, Pessin JE, editors. Mechanisms of Insulin Action. Austin, Texas: Landes Bioscience.
Weinberg E. 1999. Iron Loading and Disease Surveillance. Emerging Infectious Diseases 5(3):346-352.
Welch KD, Van Eden ME, Aust SD. 2001. Modification of Ferritin During Iron Loading. Free Radical Biology & Medicine. 31(8):999-1006
Wells J. 2006. The evolution of human fatness and susceptibility to obesity: an ethological approach. Biol Rev 81:183-205.
Witteles RM, and Fowler MB. 2008. Insulin-Resistant Cardiomyopathy. Journal of the American College of Cardiology 51(2):93-102
Wong VW, Mardini M, Cheung NW, Mihailidou A. in press. High-dose insulin in experimental myocardial infarction in rabbits: protection against effects of hyperglycaemia. Journal of Diabetes and Its Complications (2010)
Wrede CE, Buettner R, Bollheimer LC, Scholmerich J, Palitzsch KD, and Hellerbrand C. 2006. Association between serum ferritin and the insulin resistance syndrome in a representative population. European Journal of Endocrinology 154:333-340
Zar JH. 1984. Biostatistical Analysis.Englewood Cliffs, New Jersey: Prentice-Hall, Inc.
Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. 1994. Positional Cloning of the Mouse Obese Gene and its Human Homologue. Nature 372:425- 432.
Zihlman AL, McFarland RK. 2000. Body Mass in Lowland Gorillas: A Quantitative Analysis. American Journal of Physical Anthropology 113:61-78
Appendix A
Institution Name Studbook Diagnosis A Insulin Glucose QUICKI Leptin Cholesterol oxLDL Prolactin Ferritin ge (mU/l) (mg/dl) (ng/dl) (ng/dl) (U/l) (ng/ml) (ng/dl)
Brookfield Bakari 1848 0 3 .61 99 .56 1.35 341 4.19 30.73 465.85 Louisville Cecil 1490 0 5 4.11 104 .38 1.38 357 4.91 54.17 333.05 Louisville Kicho 1454 0 6 6.14 78 .37 1.36 285 2.98 62.52 411.87 Louisville Bengati 1489 0 7 3.04 66 .43 1.38 235 3.64 104.53 443.75 Louisville Jelani 1451 0 9 2.50 73 .44 1.29 252 4.80 42.02 537.94 Omaha Tatu 1500 1 9 5.58 97 .37 1.24 321 3.46 30.27 403.94 Omaha Kijito 1373 1 12 .54 75 .62 .69 268 4.06 25.52 485.26 Omaha Ktembe 1461 0 12 2.91 103 .40 .86 367 3.46 45.92 530.21 Omaha Samson 1372 1 12 .42 73 .67 .67 338 4.55 25.16 490.00 Omaha Tambo 1337 1 12 2.13 105 .43 8.94 263 3.34 22.43 450.43 Omaha Ngoma 1336 0 13 .54 75 .62 3.64 231 4.58 23.19 539.81 FPZ Little Joe 1295 0 14 .47 44 .76 1.41 186 2.87 47.35 912.32 Omaha Gerry 1308 1 15 .45 87 .63 1.19 187 3.61 59.67 737.70 Columbus Nkosi 1195 0 16 1.42 63 .51 1.14 295 3.00 28.58 1018.08 Louisville Mshindi 971 0 18 6.61 74 .37 10.58 229 2.79 51.33 813.59 Buffalo Koja 967 1 21 .87 110 .50 .98 248 3.06 24.03 597.20 FPZ Kitombe 934 1 22 10.91 76 .34 9.44 208 2.96 28.96 865.10 Cleveland Mokolo 948 1 22 12.88 90 .33 7.56 205 3.41 43.66 539.98 Omaha Motuba 883 1 22 3.25 110 .39 12.85 260 4.57 26.47 652.28 Columbus Macomba 836 1 23 5.54 58 .40 5.63 288 2.85 30.98 711.88 Omaha Mosuba 835 0 23 5.09 135 .35 7.08 355 3.78 24.34 666.19 Cleveland Bebac 872 1 24 4.54 90 .38 6.11 275 3.39 31.31 629.51 Knoxville Ernie 808 1 24 15.03 69 .33 13.75 246 5.81 26.47 838.06 Knoxville Kwashi 796 1 26 5.59 56 .40 12.86 238 5.76 37.85 945.04 Miami Jimmy Jr 716 1 29 6.90 73 .37 5.66 225 3.26 76.63 654.10 Brookfield Ramar 537 1 40 2.74 126 .39 5.74 240 3.82 33.97 1753.93 Bronx Fubo 314 1 42 10.92 85 .34 .94 288 4.79 51.42 599.53 Louisville Frank 265 1 43 1.60 78 .48 5.97 257 2.25 N/A N/A GorillaHaven Joe 268 1 44 9.28 73 .35 11.02 196 3.79 36.57 775.54 Louisville Timmy 282 1 48 3.36 89 .40 9.13 190 7.58 47.47 1030.71
83
Appendix B
Institution Name Studbook # Diagnosis Age EF (%) IVS (cm) LVPW (cm) LVIDd (cm)
Brookfield Bakari 1848 0 3 72 .60 0.60 3.50 Louisville Cecil 1490 0 5 60 1.05 1.03 3.00 Louisville Kicho 1454 0 6 89 1.29 1.07 3.20 Louisville Bengati 1489 0 7 71 .91 0.82 4.43 Louisville Jelani 1451 0 9 66 .93 N/A 3.86 Omaha Tatu 1500 1 9 65 1.70 1.80 6.30 Omaha Kijito 1373 1 12 40 2.00 1.60 6.30 Omaha Ktembe 1461 0 12 55 1.20 1.20 6.50 Omaha Samson 1372 1 12 60 2.00 2.00 7.50 Omaha Tambo 1337 1 12 50 1.40 1.40 5.60 Omaha Ngoma 1336 0 13 58 1.50 1.50 6.00 FPZ Little Joe 1295 0 14 65 1.50 1.00 5.50 Omaha Gerry 1308 1 15 30 1.90 1.90 6.80 Columbus Nkosi 1195 0 16 66 1.50 1.50 3.58 Louisville Mshindi 971 0 18 80 1.34 1.21 5.42 Buffalo Koja 967 1 21 40 1.60 N/A 6.80 FPZ Kitombe 934 1 22 65 1.20 1.20 5.50 Cleveland Mokolo 948 1 22 80 1.60 1.70 6.10 Omaha Motuba 883 1 22 58 1.50 1.40 6.90 Columbus Macomba 836 1 23 46 1.00 0.90 5.71 Omaha Mosuba 835 0 23 70 1.20 1.20 7.50 Cleveland Bebac 872 1 24 42 2.90 2.40 4.80 Knoxville Ernie 808 1 24 52 1.85 N/A 6.09 Knoxville Kwashi 796 1 26 45 2.05 N/A 4.71 Miami Jimmy Jr 716 1 29 61 1.42 1.58 6.63 Brookfield Ramar 537 1 40 43 2.70 2.60 4.50 Bronx Fubo 314 1 42 48 1.25 1.25 6.20 Louisville Frank 265 1 43 33 2.10 2.00 4.40 GorillaHaven Joe 268 1 44 40 2.39 2.54 6.02 Louisville Timmy 282 1 48 60 1.60 1.88 5.90
84