Western Kentucky University TopSCHOLAR®

Masters Theses & Specialist Projects Graduate School

12-2013 Single Nucleotide Polymorphisms Linked to Essential in Kasigau, Kenya Julia Carol Freeman Western Kentucky University, [email protected]

Follow this and additional works at: http://digitalcommons.wku.edu/theses Part of the Biology Commons, Cardiology Commons, Genetics Commons, Medical Education Commons, and the Physiology Commons

Recommended Citation Freeman, Julia Carol, "Single Nucleotide Polymorphisms Linked to Essential Hypertension in Kasigau, Kenya" (2013). Masters Theses & Specialist Projects. Paper 1316. http://digitalcommons.wku.edu/theses/1316

This Thesis is brought to you for free and open access by TopSCHOLAR®. It has been accepted for inclusion in Masters Theses & Specialist Projects by an authorized administrator of TopSCHOLAR®. For more information, please contact [email protected].

SINGLE NUCLEOTIDE POLYMORPHISMS LINKED TO ESSENTIAL HYPERTENSION IN KASIGAU, KENYA

A Thesis Presented to The Faculty of the Department of Biology Western Kentucky University Bowling Green, Kentucky

In Partial Fulfillment Of the Requirements for the Degree Master of Science

By Julia C. Freeman

December 2013

Dedicated to my family, friends, and the people of Kasigau, Kenya

ACKNOWLEDGMENTS

I would like to acknowledge my primary investigator, Dr. Nancy Rice for her

supervision and guidance during this project. Without her persistence and dedication to

improving the lives of the Kasigau people, this project would not be possible. She has

been generous with her time. I am also thankful for my committee members, Dr. Ajay

Srivastava and Dr. Cheryl Davis, for their support and assistance during this project.

Also deserving acknowledgment are Lindsay Williams, Seth Tooley, my WKU

friends, and the numerous PiC:MiK members as they have helped tremendously with data

collection, analysis, and making this work worthwhile. Ms. Naomi Rowland is most

deserving of my gratitude for her persistence to teach me and enlighten me with her

expertise. Without her, this project would have been far less successful and more stressful

for all involved.

This work was supported by the NIGMS of the NIH under award number

8P20GM103436-12, a WKU Office of Research RCAP award, and a WKU Graduate

Studies grant. These funds made this project beyond just an idea on paper.

I am who my parents and family have raised me to become and their love and support have been incredible. Thank you for teaching me more than any classroom ever has and to not let obstacles stand in my way. There truly is no substitute for hard work or the experience of working in complete darkness with light sensitive assays.

Lastly, I would like to mention my gratitude to the participants in Kasigau. The people of Kasigau have allowed me to gain a new perspective on life, both as a global citizen and on an individual level. They have helped to shape me into a better scientist.

iv

CONTENTS

LIST OF FIGURES……………………………………………………. vi

LIST OF TABLES……………………………………………………... vii

ABSTRACT……………………………………………………….. ….. viii

INTRODUCTION ………………………………………………… ….. 1

MATERIALS AND METHODS ……………………………………… 14

RESULTS …………………………………………………………. ….. 22

DISCUSSION …………………………………………………………. 33

APPENDIX A ………………………………………………………….. 38

APPENDIX B ………………………………………………………….. 39

APPENDIX C ………………………………………………………….. 40

REFERENCES …………………………………………………………. 41

v

LIST OF FIGURES

Figure 1: Worldwide prevalence and subtypes of EH by age between males and females.2

Figure 2: Awareness, treatment, and BP control in Kenyan patients with hypertension….6

Figure 3: Renin-angiotensin system…………..………………………………………...…8

Figure 4: TaqMan® genotyping assay...... ………………………………...... ………17

Figure 5: Hardy-Weinberg Equilibrium equation…………………………….…….……20

Figure 6: Pearson’s Chi-Square equation………………………………….....…………..21

Figure 7: AGT rs699 allele frequencies for Kasigau………....……………………....…..24

Figure 8: AGT rs699 genotype distribution for Kasigau…...………………...... ……24

Figure 9: AGTR1 rs5186 allele frequencies for Kasigau……….………...... …..26

Figure 10: AGTR1 rs5186 genotype distribution for Kasigau……….…………………..27

Figure 11: HSD11β2 rs5479 allele frequencies for Kasigau………………...………...…30

Figure 12: HSD11β2 rs5479 genotype distribution for Kasigau…….….….…..………...31

vi

LIST OF TABLES

Table 1: Demographics of Kasigau...... ……………………….….………..15

Table 2: RAS SNP sequence information……………………………….…..…………...19

Table 3: AGT rs699 genotype distribution and allele frequencies per EH classification in

Kasigau...... …...... …....25

Table 4: AGTR1 rs5186 genotype distribution and allele frequencies per EH classification

in Kasigau...... …....…...………………………………..…...... 28

Table 5: HSD11β2 rs5479 genotype distribution and allele frequencies per EH

classification in Kasigau...... …..…………….….….………………32

vii

SINGLE NUCLEOTIDE POLYMORPHISMS LINKED TO ESSENTIAL HYPERTENSION IN KASIGAU, KENYA

Julia C. Freeman December 2013 46 Pages

Directed by: Dr. Nancy Rice, Dr. Cheryl Davis, and Dr. Ajay Srivastava

Department of Biology Western Kentucky University

Hypertension, or high blood pressure (BP), is an ever-growing epidemic in the developing world. Understanding the genetics behind essential hypertension (EH), or hypertension with no known cause, is especially important. In this study, three single nucleotide polymorphisms (SNPs) known to be linked to an increase in susceptibility to

EH were quantified from a cohort of Kenyans living in the Kasigau region. The SNPs are located in three that are part of the renin angiotensin system, the primary regulatory pathway in humans controlling BP. They include: AGT (rs699), AGTR1 (rs5186), and

HSD11β2 (rs5479). Overall, by using a fluorescent-based RT-PCR technique, the genotype distribution of AGT (rs699) was 0.63 C/C, 0.34 C/T, and 0.03 T/T. When evaluated as normotensive, prehypertensive, Stage I, or Stage II categories the allele frequencies for f(C)= 0.77,0.85,0.81, 0.77, respectively, and demonstrated Hardy

Weinberg Equilibrium (HWE) as assessed by Χ2, p < 0.05. The genotype distribution of

AGTR1 (rs5186) was 0.96 A/A, 0.03 A/C, and 0.00 C/C and the genotype distribution of

HSD11β2 (rs5479) was 0.46 A/A, 0.46 A/C, and 0.08 C/C. The majority of genotype

frequencies for each SNP were in HWE, with the exception of the AGT (rs699) SNP

found in the sublocation of Bughuta suggesting other evolutionary selective pressures

may be at work in this subpopulation. The high prevalence of the susceptible C allele for

AGT (rs699) likely implies it is a critical factor in the high prevalence of EH observed in

this population.

viii

INTRODUCTION

Cardiovascular disease (CVD), a group of disorders of the heart and blood

vessels, accounts for an estimated 17,300,000 deaths worldwide a year (World Health

Organization, 2011). These diseases are further categorized into coronary heart disease,

cerebrovascular disease, peripheral arterial disease, deep vein thrombosis and pulmonary

embolism, congenital heart disease, and rheumatic heart disease. Acute events such as

heart attack and stroke are also included as cardiovascular conditions. It is estimated that

over 80% of the deaths from CVD take place in low- and middle-income countries and occur almost equally in men and women (Mathers & Loncar, 2006). Additionally, ethnic differences also can differentially contribute to CVD. Essential hypertension (EH) is considered to be a major modifiable risk factor for the development of CVD (Edwards,

Unwin, & Mugusi, 2000), and EH is the leading risk factor for CVD. EH can also even lead to premature death (Doris, 2002). EH is a heterogeneous disorder with many different causal factors leading to its diagnosis; no single genetic or risk factor can determine its severity. Therefore, there is a great need for research into the causes of EH.

EH has various degrees of severity and is based primarily on blood pressure (BP)

(Tobin, Sheehan, Scurrah, & Burton, 2005). Since BP is a ratio of systolic to diastolic rates, the clinical aspects of classification is determined by two different numbers based on these rates (Tobin et al., 2005). There are three categories of EH classification in addition to normal individuals. The first, and least severe, is normal or normotensive and is classified as having consistent BP readings of 120/80 mmHG or below.

Prehypertensive individuals have repeated BP readings of 121/81-139/89 mmHg, Stage 1

EH is characterized by BP readings of 140/90-159/99 mmHg, and Stage 2 EH features

1

consistent BP readings of 160/100 mmHg or greater (Kaplan & Victor, 2009) and is of

immediate concern for intervention (Tobin et al., 2005).

The prevalence of EH increases with age (Figure 1) for both males and females

from all ethnicities (Gupta, Gupta, & Pednekar, 2004). Individuals with high BP are also

more likely to have other CVD risk factors like high cholesterol levels and larger waist

circumferences (Doris, 2002). By knowing what causes EH, appropriate precautions,

treatments, therapies, and studies may be developed to combat controllable aspects

associated with CVD.

Figure 1: Worldwide prevalence of EH by age and subtype between males and females.

Solid=prehypertensive, checkered=Stage I, hashed=Stage II. (Modified from Gupta et al.,

2004).

Unmanaged EH is correlated with CVD severity. Individuals who have been diagnosed with EH require treatment for the remainder of their lives to prevent CVD 2

from occurring or advancing (Mathers & Loncar, 2006). Treatment and proper diagnosis of EH is critical to prevent or delay health complications. Untreated EH is dangerous as left ventricular hypertrophy (LVH) can occur. LVH serves as a marker for enlargement of the left ventricle of the heart and can lead to death. This contractile overcompensation by the heart leads to damage in the heart, brain, and kidneys of patients with EH (Bauml &

Underwood, 2010) and elevates the risk of premature death (Katholi & Couri, 2011).

EH has a well-known familial aggregation and has been calculated to be approximately 40% genetically determined (Deng, 2007). Among people younger than 50 years, EH occurs 3.8 times more often in those with two or more immediate relatives that developed high BP before age 55 (Giner et al., 2000). While many EH cases are attributed to genetic factors, there is no single genetic determinate known (Rhodes &

Bell, 2009). As a result, EH is classified as a polygenic disease.

Quantitative trait loci (QTL), in addition to genome-wide association techniques, have helped in the understanding of EH (Rhodes & Bell, 2009). Three QTLs have been shown to contribute to lower BP. Epistatic relationships occur between two of the three

QTLs (Iigaya et al., 2009). QTL mapping on chromosome 1 of rats revealed hundreds of genes linked to regulation of BP (Deng, 2007). For example, the cofactor that regulates angiotensin and angiotensinogen receptors, Arrb1, also regulates internalization of the β- adrenergic receptors (Iigaya et al., 2009). This is linked to BP variations due to the products of Arrb1 altering signaling pathways throughout the body. The pathways responsible for regulation of BP are included in those influenced by Arrb1 (DeFea, 2008).

By identifying the genetic targets responsible for BP conditions such as EH, specific therapeutics can be developed to treat and diagnose patients. 3

Kasigau, Kenya demographics

This study is based in Kenya, a developing country of East Africa, with a population of 39 million people (Hendriks et al., 2012). There are on average 15 doctors per 100,000 people within the country. This is drastically different than that of the developed world because there are an estimated 270 doctors per 100,000 people in the

United States (World Health Organization, 2011). Through Western Kentucky

University’s Partners In Caring: Medicine in Kenya (PiC:MiK) program, a partnership between the people of Kasigau, Kenya and Western Kentucky University, the healthcare concerns of Kasigau are addressed. Medical care and education are among the primary goals of this initiative and attempt to bridge the disparity in healthcare. This thesis is a component of the research focused around the cardiovascular health of the Kasigau people.

Kasigau is located within the Eastern Arc Mountains in sub-Saharan Africa between the Taita Hills and Indian Ocean. Within this location, there are scarce resources, inadequate healthcare, and diagnosis, treatment, and prevention of EH and other CVD specifically are inadequate. These all contribute to negative healthcare consequences for the residents of Kasigau. Mount Kasigau, which is part of the Taita Hills, is surrounded by the villages of Jora, Makwasinyi, Ngambenyi, Rukanga, Bungule, and Kiteghe. There is an additional village included in this study as well. Bughuta is located further away from the Taita Hills and does not surround Mount Kasigau directly; however, the village is only a few kilometers drive from the other villages of this study. Collectively, all seven villages are divided into three sublocations (Bughuta, Rukanga, and Makwasinyi) based upon proximity. 4

Preliminary studies have found that in a pilot group (n=159) of individuals, ~67%

of the Kasigau population has EH. Unlike in developed populations, risk factors such as

body mass index and cholesterol did not correlate to the prevalence of EH in this

population (Williams, 2012). Therefore, this preliminary epidemiological work supports

the hypothesis that genetic or behavioral risk factors may greatly contribute to the

observed prevalence of EH in Kasigau.

Burden of CVD

There are 970 million people worldwide living with EH (Edwards et al., 2000).

The majority of these, 640 million, live in the developing world. According to the World

Health Organization (WHO), it is estimated that by 2025 people living with EH will grow

to 1.56 billion worldwide (Addo, Smeeth, & Leon, 2007). CVDs, primarily due to stroke

and coronary heart disease, are the leading cause of death globally (Pickering, 1965) and

account for 30% of all deaths (Miranda, Kinra, Casas, Smith, & Ebrahim, 2008). Deaths

from CVD are expected to grow 17% overall, with the largest increase, 27%, predicted in

Africa (Addo et al., 2007).

The WHO reports that low- to middle- income countries are disproportionately affected by CVD with over 80% of all CVD deaths occurring in such areas (Dongfeng et al., 2010). People of African descent are particularly vulnerable to EH compared to other ethnicities due to both socioeconomic factors and genetics. Many people in Africa are not even aware they are hypertensive (Figure 2) because of the lack of preventative care and awareness surrounding CVD (Hendriks et al., 2012).

5

Figure 2: Awareness, treatment, and BP control in Kenyan patients with hypertension

(Modified from Hendriks et al., 2012).

Recent research has attributed 20% of all deaths in sub-Saharan Africa to

noncommunicable disease. The term noncommunicable disease is used interchangeably

with chronic disease and refers to mental illnesses, diabetes, chronic respiratory diseases,

cancers, CVD, and injuries. Globally, CVD, cancer, and injuries consistently rank as the

top three conditions in low to low-middle income countries (Miranda et al., 2008).

The heightened susceptibility of people in low- to middle- income countries, like

Kenya, to developing EH has three reasons. People in these countries are more exposed to risk factors that are known to lead to CVD and other noncommunicable diseases; people do not have access to prevention programs compared to people in high- income countries; and people do not receive effective and equitable healthcare services because the area lacks availability. Early detection services, which are essential to lessening the rate of CVD in these countries, are the most effective healthcare service (Hendricks et al.,

2012). Among risk factors, high salt diets and limited access to resources like clean drinking water contribute to elevated rates of CVD (Mathers & Loncar, 2006). The

6

poorest people are affected most in any developing country, including Kenya (Jamison et

al., 2006).

Approximately half of Kenya’s population (15 million people) are classified as

very poor and survive on less than 1 USD a day with very limited access to healthcare.

Awareness of EH within Kenya is essential to effective management of CVD (Figure 2).

But unfortunately, the necessary treatment for EH is often difficult to obtain for Kasigau residents due to the general lack of CVD awareness and low income levels.

In Kenya, medication for treating EH costs the nation $24 million annually

(Hendriks et al., 2012). Given that not all EH is effectively controlled pharmaceutically, the Kenyan population and economy are left especially vulnerable. For Kenyans that are aware they have EH, it is often difficult to afford drugs and/or adjust to the necessary lifestyle changes to control and treat their EH. Treatments of cardiovascular conditions often include more than a single type of pharmaceutical and/or lifestyle changes

(Hendricks et al., 2012). Studies have attributed increased CVD rates to the increased price of pharmaceutical treatments (Edwards et al., 2000).

Renin-angiotensin system

One of the key hormonal systems that regulates and maintains BP and ultimately

EH development in the body through controlling salt balance and arterial tone is the renin-angiotensin system (RAS) (Giner et al., 2000). The RAS pathway is composed of renin, angiotensin I-converting enzyme, angiotensinogen, and angiotensin II receptor type

I, predominately, as well as several other components.

7

Each protein in the pathway is encoded by a separate . The genes are angiotensin converting enzyme (ACE), angiotensinogen (AGT), renin (REN), and angiotensinogen II receptor isoform 1a (AGTR1), respectively (Figure 3). AGT encodes the protein angiotensinogen which is cleaved by renin to form angiotensin I. Angiotensin

I is cleaved by ACE to form angiotensin II, and angiotensin II is responsible for stimulating aldosterone secretion (Mills, 2011) that ultimately elevates BP within the arteries of multiple organ and tissue systems (Pickering, 1965). Aldosterone is also regulated by angiotensinogen II receptor isoform 1a (Mills, 2011) which increases intracellular calcium concentration and phosphorylation of proteins (Pickering, 1965).

Figure 3: Renin-angiotensin system. The arrows indicate the hormonal pathway that leads to changes in arterial tone. Angiotensinogen is secreted from the liver, broken down to angiotensin I by renin, and further converted into angiotensin II by angiotensin converting enzyme. Angiotensin II is responsible for numerous physiological effects ultimately resulting in water and salt retention (Public Domain).

8

Single nucleotide polymorphisms

Genetic polymorphisms are defined as a difference in the DNA sequence among

individuals, groups, or populations that affect a variety of traits within the organism and

those in the coding region represent about 90% of the common variation within the

genome (Doris, 2002). Genetic polymorphisms produce an individual’s distinctive

behavioral, physical, and social responses to events, environments, and conditions

throughout their lifetime (Giner et al., 2000). Individuals differ both in appearance and in

their genetic makeup because of polymorphic variations; however, polymorphisms also

have implications in disease susceptibility (Doris, 2002). There are numerous

polymorphisms in RAS genes that have been identified, and several of these correlate

with EH in Asian and European populations (Tobin et al., 2005).

The most stable variation in the human genome is in the form of a single

nucleotide (SNP) (Doris, 2002). It was recently discovered that SNPs are

not distributed equally over the entire portion of the genome but are instead found

concentrated within noncoding regions of DNA (Murray, Granner, Mayes, & Rodwell,

2000). Several SNPs in RAS genes have previously been identified; three were selected

for this study based upon prior correlation to increased EH susceptibility. The SNPS

included in this study are: AGT M235T, hydroxysteroid dehydrogenase 11β2 G534A

(HSD11β2), and the AGTR1 A1166C.

When polymorphisms are identified, they are provided a reference SNP ID (rs)

number. An rs number is defined as a variation at a location on an ideal reference

chromosome and archived in the Single Nucleotide Polymorphism Database (dbSNP), a

9

free public archive of within and among species. The database was

developed and is hosted by the National Center for Biotechnology Information (NCBI) of

the National Institutes of Health (NIH). The rs number and the associated allelic variants

examined in this study are rs699 (AGT M235T), rs5186 (AGTR1 A1166C), and rs5479

(HSD11β2 G534A).

Angiotensinogen

Prior studies have shown that the M235T missense polymorphism (rs699) of AGT

directly mediates a predisposition for EH (Ortlepp et al., 2003). The thymine allele,

which codes for methionine, is the normal variant of AGT; higher angiotensinogen levels

are caused by the C allele via inability of angiotensinogen to be cleaved correctly by

renin, which encodes for threonine. For clarification in this thesis, thymine will be

abbreviated T. The susceptible cytosine allele is known to lead to higher BP and increased risk for cardiovascular conditions (Ortlepp et al., 2003). Genotypes are classified as C/C, T/T, or C/T based on the presence of alleles.

Ethnicity also seems to be a key factor with regard to how the amino acid substitution influences EH (Dongfeng et al., 2010). Asian and white European individuals are more likely to have EH based upon this amino acid of T to C substitution. The C allele compared to those without the variant caused improper BP regulation and end- organ damage (Ortlepp et al., 2003). This was especially true from individuals with family histories of EH, although familial polymorphism rates were not discussed (Poch et al., 2001). Notably absent from current literature on EH is the relation between the rs699 polymorphism and EH in East African populations.

10

Angiotensin II receptor, isoform 1

There are many variants in AGTR1, each of which impact BP (Bonnardeaux et al.,

1994). The role of the AGTR1 gene product is to balance the narrowing of blood vessels and aldosterone secretion by binding angiotensin II (Murphy, Alexander, Griendling,

Runge, & Bernstein, 1991). Within the body, it has also been shown to trigger inflammation by inducing oxidative stress through increasing sensitivity to inflammatory mediators.

Further analysis is required on the specific variants of AGTR1 in order to properly gauge their influence on cardiovascular conditions like EH. Due to it’s predisposition to

EH susceptibility, the A-to-C variant (rs5186) in the 3’ end of the gene that codes the receptor at nucleotide 1166 was linked to a high frequency of elevated BP in white subjects (Amir et al., 2009). This linkage has been essential to proper understanding of the global understanding of AGTR1’s contributions to EH.

The AGTR1 A allele is identified as the normal variant, while the C allele has been classified as the susceptible variant for EH. Genotypes are classified as A/A, A/C, or C/C.

In a study of Japanese populations with severe EH, the frequency of C/C and A/C genotypes is increased in patients with EH compared to normotensive controls (Poch et al., 2001). The genotype C/C for rs5186 is rare (5%) in African American populations because it is attributed to an increased global mortality due to premature heart failure

(Bonnardeaux et al., 1994). There is little research being performed on the polymorphic variations within AGTR1, especially in populations from East Africa.

11

Hydroxysteroid (11-beta) dehydrogenase

The level and activity of the hormone cortisol is determined by HSD11β2 (Persu,

2005). Cortisol production is elevated in stress conditions and mediates the increased recruitment of fat distribution to stressed areas of the human body (Rhodes & Bell, 2009).

HSD11β2 is a microsomal short-chain dehydrogenase that catalyzes the reduction of organic compounds. Within the body, it is responsible for converting the active hormone cortisol to the inactive form cortisone. The amino acid change in the HSD11β2 C534A

(rs5479) polymorphism is essential in the hormone conversion of the active version cortisol to inactive cortisone (Persu, 2005). HSD11β2 also plays a key role in sodium homeostasis through mediating the absorption and release of sodium by inhibiting the mineralocorticoid receptor. Compromised function results in overstimulation of receptors by cortisol leading to sodium retention, hypokalemia, and ultimately EH (Wilson et al.,

1998 & Carvajal et al., 2005). Normally, HSD11β2 regulates excess aldosterone secretions, however, in an individual with EH, this mechanism is altered or completely shut down so that secretions are able to accumulate. These secretions make it difficult for

BP to be maintained. Studies have found that the HSD11β2 rs5479 SNP causes increased

BP (Mune, Rogerson, Nikkilä, Agarwal, & White, 1995). This also causes overcompensation of potentially hazardous elements, such as sodium, being retained within the body. The colon, which are part of the human digestive system, and kidneys, an integral component of the body’s urinary system, are primarily affected by these changes in allele frequency.

12

Goals of research

The goal of this research project was to evaluate the prevalence of three SNPs that

had in previous studies and ethnicities been correlated to an increased risk for EH within

Kasigau, Kenya. This was achieved by performing allelic discriminations and calculating

allelic and genotype frequencies of rs699, rs5479, and rs5186 in individuals from each of

the rural Kasigau villages. Subsequently, allelic genotype frequency allowed for

associations of the RAS SNPs to normotensive, prehypertensive, Stage I hypertensive,

and Stage II hypertensive BP responses and determination of Hardy-Weinberg equilibrium (HWE). This work serves as a model study examining the variation of polymorphic changes in a rural East African population with a high amount of EH

(Williams, 2012). Overall, this work evaluated how polymorphic variation in RAS genes correlate to phenotypic differences of EH in Kasigau, Kenya.

13

MATERIALS AND METHODS

Compliance/informed consent

This study was approved by Western Kentucky University's Human Subjects

Review Board (HR-174, Appendix A) and the Taita District Health Officer in Kenya

(Appendix B). All participants were read an informed consent document and their

voluntary willingness to participate in the study implied their consent (Appendix C).

Participants were also modestly compensated financially for participation.

Study participants

Participants for this study were selected based upon a cross-sectional population

assessment of the Kasigau, Kenya area villages (Jora, Rukanga, Makwasinyi, Bungule,

Kiteghe, and Bughuta) using census data of the region. The area was subdivided into

sublocations: Rukanga, Makwasinyi, and Bughuta which compromise 29%, 21%, and

50% of the total population, respectively. Sublocations consisted of one to four distinct

villages.

Participant criteria also included being 45-75 years of age since the onset of EH occurs primarily in older people. Stratification of participants by sex from the census data was also essential for the overall population structure to match any differences among villages. The recruitment goal was 50% males and 50% nonpregnant females of Kasigau, although actually 66% of participants were females and 34% were males (Table 1).

Overall, 400 individuals were recruited to take part in the study, although only 161 individuals eventually participated in the study.

14

Table 1: Demographics of Kasigau. Total population of Kasigau and number of individuals per sublocation.

Total Pilot Population Study Demographics n % n % Female % Males % Kasigau 13,696 161 107 66% 54 34% Rukanga 3,940 29% 63 39% 47 75% 16 25% Makwasinyi 2,847 21% 31 19% 12 39% 19 61% Bughuta 6,909 50% 67 42% 48 72% 19 28%

Buccal cell acquisition

Buccal cells were acquired from participants using Isohelix DNA Buccal Cell

Swabs (Cell Projects Ltd., Kent, UK; Cat. No: SK-2S). These individually wrapped swabs were advantageous because they could yield up to 5 μg of human genomic DNA and were sealed to avoid contamination. Swabs were labeled by village and participant number in order to correlate genotype data to an individual’s demographic data.

Swabbing occurred during the participant’s clinic visit time and involved the thorough swabbing of the inside of the cheek for one minute. After acquisition of buccal cells, samples were sealed and allowed to dry. Upon arrival in the United States, the samples were stored at 4°C prior to genomic DNA isolation.

Genomic DNA isolation

Genomic DNA was isolated from swabs by standard alkaline lysis using the

QIAamp DNA Mini Kit (Qiagen; Valencia, CA, USA; Cat. No: 51306). The results from the kit and swabs provided 500-2000 ng of DNA per sample while rapidly purifying and creating high quality, ready-to-use DNA. Groups of 24 samples were processed per extraction cycle.

15

To isolate DNA, the swabs were submerged in 1X phosphate buffer saline (PBS).

Upon being exposed to PBS, protease K and buffer AL were added to each sample

according to kit protocol. Protease K was used as the optimal enzyme with the lysis

buffer (Qiagen; Valencia, CA, USA). Samples were immediately capped and vortexed

briefly after addition of buffer AL. A dry water bath (56°C, 10 min) was used for

activation of the protease within each sample. Ethanol (96-100%) was then added to the

contents of the sample. Brief vortexing and centrifugation of capped sample tubes

followed.

The samples were applied to a spin column from the QIAamp DNA Mini Kit and

centrifuged at 8,000 rpm for one minute to isolate genomic DNA. Filtrates were

discarded. The columns were then washed with a series of wash buffers provided in the

kit followed by centrifugation. DNA was eluted from the spin column by the addition of

buffer AE followed by centrifugation (8,000 rpm, 1 min) to complete isolation.

Sample DNA was stored at -20°C. This storage technique was recommended for long-term storage (Qiagen; Valencia, CA, USA). It also maintained the quality of the

DNA with minimal degradation. All genomic DNA was isolated from samples prior to quantification.

Nanodrop analysis

DNA concentrations and purity were determined by the ratio of absorbance of UV light at 260/280 nm using a Nanodrop spectrophotometer (Thermo Scientific;

Wilmington, DE, USA). Two µl of each sample was analyzed to determine final DNA concentration. Typically, a buccal swab would yield 3-23 ng/µl of genomic DNA.

Although variable, 96% of Kasigau genomic DNA samples were measured within this concentration range. 16

Allelic discrimination analysis

Real-time Polymerase Chain Reaction (RT-PCR) was used for genotype and allelic discrimination purposes. The TaqMan SNP Genotyping Assays Protocol (Applied

Biosystems; Foster City, CA, USA) was used for allelic discrimination in this study.

These assays use fluorescent dyes as reporters of DNA binding. Each dye is covalently attached to each primer used in the PCR reaction and includes a blue dye, called FAM, and a green dye, called VIC. A reference dye, ROX, was also used. Either blue or green coloration were identified through the use of Sequence Detection System (SDS) software

(Applied Biosystems; Foster City, CA, USA). Both a reporter dye, which is detected by

SDS, and a quencher, used for binding to DNA, are contained on the 5’ and 3’ end of the assay, respectively (Figure 4).

Figure 4: TaqMan® genotyping assay. (“TaqMan® Genotyping Assay Product Bulletin,”

Applied Biosystems; Foster City, CA, USA). 17

All samples were run in triplicate on a MicroAmp® Optical 96-well reaction plate

(Applied Biosystems; Foster City, CA, USA) in an Applied Biosystems Prism 7300 Real-

Time polymerase chain reaction instrument. Both positive controls and negative controls,

without DNA template, were also included. Triplicate sampling reduces the likelihood of

bias or error. The PCR exploits the 5’ nuclease activity of AmpliTaq Gold® DNA

Polymerase (Applied Biosystems; Foster City, CA, USA) to cleave the assay during the

reaction. This cleavage is responsible for separating the assay’s quencher and reporter

dyes from one another for PCR. By cleaving the assay into such parts, the color intensity

of the reporter dye increases in fluorescence. The fluorescence is only successful if the

target sequence within the genomic DNA is complementary to that of the cleaved assay

sequence. If not cleaved, the reporter and quencher dye remain too close to one another,

and Forster-type energy transfer takes effect between the two dyes. The 3’ end of the

assay is blocked so that polymerization of the assay can be completed during PCR. No

extension of the assay occurs during PCR (Figure 4).

For each sample, following quantification of DNA, water and DNA (in µL) were added to each well so the final volume per well totaled 11.25 µL with a final concentration of 20 ng of DNA per well. Initial DNA concentrations ranged from 1.42-

11.87 ng/µL. Next, a premade master mix was added to each well along with the SNP assay. The master mix contained the following: 50 nM of forward primer labeled with

VIC dye, 50 nM of reverse primer labeled with FAM dye, deoxyribonucleotides with deoxyuridine triphosphates, AmpliTaq Gold DNA polymerase, AmpErase® uracil-N- glycosylase, optimized buffer, and passive reference components. The primer pair sequence for each SNP is provided in Table 2. 18

Table 2: RAS SNP primer sequences. The context sequence is exclusive to the SNP, while the forward and reverse primers are in reference to the surrounding genetic code. Each context sequence has portions identical to the primers.

SNP Context sequence Reverse primer (3’ → 5’) Forward primer (5’ → 3’)

CAGGGTGCTGTCCACACT ACGTTGGATGAAAC ACGTTGGATGTGATAG

GGCTCCC[A/G]TCAGGGA GGCTGCTTCAGGTGC CCAGGCCCAGCT rs699 GCAGCCAGTCTTCCATCC

TGCAGCACTTCACTACCA AGAAAAGTCGG CCCCAAAAGCCAAATC

rs5186 AATGAGC[A/C]TTAGCTAC TTCAGTCCAC CCAC

TTTTCAGAATTGAAGGA

TGCTAGAGTTCACCAAGG ACCAGCACCGGTCA TGCTAGAGTTCACCAA

rs5479 CCCACAC[A/C]ACCAGCAC GTGGCAAGTGT GGCCCACAC

CGGTCAGTGGCAAGTGT

The following were used as cycling conditions: activation (95°C-10 minutes);

denaturing (92°C-15 seconds); extending and annealing (60°C- 70 seconds). The cycling procedure was repeated 50 times. This was performed to sufficiently elevate the fluorescence for SDS to detect within each well.

A final postrun was performed on the plate after PCR cycling. The postrun step

(60°C-l minute) occurred for all 96 wells. This was done in comparison to the prerun of the plate to detect changes in fluorescence.

19

Statistical analysis

The Hardy-Weinberg Equilibrium (HWE) equation (Figure 5) was used to compare the actual genetic composition for this study’s participants to a hypothetical gene structure, probabilities of frequencies and expected values of alleles were derived from observed data and analyzed using the equation for HWE. The HWE equation includes the genotype frequency that is composed of either or both the dominant ‘A’ and/or recessive ‘a’ allele. The ‘A’ allele is known as the common allele, while the ‘a’ allele is deemed the rare allele. The frequencies of the genotypes associated with these alleles allow for further statistical analysis to be performed.

2 2 p +2pq+q =1

Figure 5: HWE equation. p2=frequency of genotype AA, p=frequency of allele A, q2=frequency of genotype aa, q=frequency of allele a (Public Domain).

Genotype frequencies were analyzed in this study using a Pearson’s Chi Square test. The equation is used to determine significance of deviations in frequencies and

HWE models (Figure 6). This calculation was used to compare genotype distributions of

SNPs within different EH classifications for all individual sublocations. The genotype distributions for rs699, rs5186, and rs5479 were also calculated for the total Kasigau population.

20

Figure 6: Pearson’s Chi-Square equation. e=expected, o=observed (Public Domain).

Statistical analysis was performed using a Pearson’s Chi Square test in R (R

Development Core Team, 2008). Probability values are reflections of the area under the curve of Gaussian distribution (bell curve) of genotypes and are reported as p-values in the resulting tables. In order to investigate whether the distribution of each SNP’s genotype differed from one another, a Chi square value was calculated.

21

RESULTS

Prior to allelic discrimination, the concentration and purity of DNA isolated from all buccal cells was determined. All Nanodrop 260/280 ratio concentrations and purities were 1.7-1.9 and considered to be pure DNA. This step was done to assure that the contents of the isolation were DNA and not protein. Ratios of 1.8 are considered pure

DNA. There is a 2.0 approximation ratio that defines pure RNA. A ratio that is approximately 0.6 is referred to as pure protein (Nanodrop, 2007). This is critical for genotype and allelic studies to be performed. If there were high protein or RNA contaminants, the allelic discrimination, polymerase chain reaction efficiency, and genotype determinations would be difficult and inaccurate.

Allele and genotype frequency of the AGT rs699 SNP correlated with EH classification

There is predisposition to EH for individuals who carry the variant C allele of the rs699 SNP. As a whole for AGT rs699, approximately 0.80 of the Kasigau population possesses the C allele, while only approximately 0.20 possess the T allele (Figure 7). This allele distribution was also observed in all three sublocations. When the derived allelic frequency was subsequently assessed as genotype, Kasigau displayed 0.63 C/C, 0.34 C/T, and 0.03 T/T (Figure 8). The sublocations differed, however, in genotype distribution;

Makwasyini and Rukanga had a C/T of ~0.23; however, Bughuta had an equal distribution of C/T and C/C genotypes. When the observed genotypes were compared with the expected frequencies predicted by HWE, Kasigau, Rukanga, and Makwasinyi exhibited a nonsignificant difference, p= 0.34, 0.14, 0.64, respectively whereas the

Bughuta AGT rs699 genotype frequencies deviated from HWE (p ≤ 0.001) (Figure 8). 22

To assess whether allelic frequency varied based upon the severity of EH, the derived allelic and genotypic distributions of rs699 were determined for normotensive, prehypertensive, Stage I EH, and Stage II EH, both for the total Kasigau population, and for the three sublocations. EH classifications were based upon the systolic BP of participants (systolic values only will be used throughout the results). Regardless of sublocation, for AGT rs699 there were more C alleles than T alleles for all EH classifications (Table 3). The T allele frequency for Kasigau ranged from 0.15-0.23, while the C allele frequency ranged from 0.77-0.85 across all four EH classifications. With regard to genotype, the majority of participant genotypes for rs699 were C/C. When the obtained genotype frequencies of AGT rs699 in each EH category were compared with the expected frequencies predicted by HWE, the polymorphism exhibited a nonsignificant difference for all EH categories (Χ2 < 3.84 with one degree of freedom).

The exception to this was prehypertensive individuals from Rukanga (Χ2= 5.4, p ≤ 0.01)

(Table 3).

Based upon these results, the majority of people in Kasigau possess the susceptible allele of the AGT rs699 SNP. While this may be a partial contributor to the high prevalence of EH observed, there is no statistical difference between the frequency of the allele in normotensive versus hypertensive Kasigau participants. More statistical analysis will need to be conducted to confirm these results. In Bughuta, the fact that the genotype frequencies for this SNP did not demonstrate HWE suggests that non-random mating may contribute to this selective pressure as evidenced by traditional village customs.

23

Angiotensinogen Allele Frequencies

n=287 n=128 n=51 n=108 90 80 70 Allele 60 T 50 80.2 83.1 85.0

Percent 75.0 40 C n=71 n=26 n=9 n=36 30 20 10 19.8 16.9 15.0 25.0 0 Kasigau Rukanga Makwasyini Bughuta

Location

Figure 7: AGT rs699 allele frequencies for Kasigau. n= the number of alleles present in each sublocation.

Angiotensinogen Genotypes 80 n=113 n=55 n=22 n=36 70

60 n=61 n=18 n=7 n=36 50 Genotype 40 T/T

Percent 71.4 73.3 50.0 C/T 30 63.1 34.1 50.0 C/C 20 n=5 n=4 n=1 n=0 23.4 10 23.4 2.8 5.2 3.3 0 Kasigau Rukanga Makwasyini Bughuta Χ2= 0.89 2 Χ2= 0.22 Χ2= 8.00 Χ = 2.25 Location p= 0.34 p= 0.14 p= 0.64 p= 0.005*

Figure 8: AGT rs699 genotype distribution for Kasigau. n= the total number of participants in each sublocation. Hardy-Weinberg equilibrium was tested using the Χ2 test. *= p ≤ 0.05.

24

Table 3: AGT rs699 genotype distribution and allele frequencies per EH classification in

Kasigau. HWE, Hardy-Weinberg equilibrium. *= p ≤ 0.05.

Rukanga Genotype Normotensive Prehypertensive Type I EH Type II EH T / T 1 1 0 2 T /C 2 1 5 8 C/C 11 14 13 18 HWE X2= 2.44 X2= 5.4 X2= 0.47 X2= 0.68 p= 0.12 p= 0.01* p= 0.49 p= 0.42 Allele Normotensive Prehypertensive Type I EH Type II EH

T 4 3 5 12 C 24 29 31 44 f (T) 0.14 0.09 0.14 0.21 f (C) 0.86 0.91 0.86 0.79 Makwasinyi Genotype Normotensive Prehypertensive Type I EH Type II EH

T / T 0 0 0 1 T /C 0 2 3 2 C/C 1 9 4 8 HWE X2= N/A X2= 0.14 X2= 0.58 X2= 1.93 p= N/A p= 0.74 p= 0.47 p= 0.20 Allele Normotensive Prehypertensive Type I EH Type II EH

T 0 2 3 4 C 2 20 11 18 f (T) 0.00 0.09 0.21 0.18 f (C) 1.00 0.91 0.79 0.82 Bughuta Genotype Normotensive Prehypertensive Type I EH Type II EH

T / T 0 0 0 0 T /C 9 9 10 9 C/C 9 11 14 9 HWE X2= 2.26 X2= 1.65 X2= 1.59 X2= 2.26 p= 0.16 p= 0.19 p= 0.20 p= 0.16 Allele Normotensive Prehypertensive Type I EH Type II EH

T 9 10 9 8 C 27 38 31 14 f (T) 0.25 0.21 0.23 0.36 f (C) 0.75 0.79 0.78 0.64 Kasigau Genotype Normotensive Prehypertensive Type I EH Type II EH

T/T 1 1 0 3 T/C 12 14 17 18 C/C 17 38 28 30 HWE X2= 0.60 X2= 0.05 X2= 2.5 X2= 0.02 p= 0.52 p= 0.82 p= 0.12 p= 0.89 Allele Normotensive Prehypertensive Type I EH Type II EH

T 14 16 17 24 C 46 90 73 78 f (T) 0.23 0.15 0.19 0.23 f (C) 0.77 0.85 0.81 0.77

25

Allele and genotype frequency of the AGTR1 rs5186 SNP correlated with EH classification

The C allele of AGTR1 rs5186 has been shown to be correlated with elevated BP

(Gillespie et al., 2005). When the Kasigau population was evaluated for this variant, only

0.02 possessed the C allele while 0.98 had the A allele (Figure 9). This allelic distribution was essentially the same in all three sublocations. When allelic distribution was evaluated as genotype frequency for rs5186, Kasigau displayed 0.96 A/A, 0.04 A/C, and 0.00 C/C

(Figure 10). This was not unexpected since the C/C genotype is known to cause premature heart failure (Amir et al., 2009). The only C alleles for AGTR1 rs5186 were identified in heterozygous participants. There was no significant difference in genotypic distribution for rs5186 among sublocations. Moreover, when the observed genotypes were compared with the expected frequencies predicted by HWE, all locations exhibited a nonsignificant difference, p= 0.82, 0.86, 0.93, and 0.91, respectively for Kasigau,

Rukanga, Makwasinyi, and Bughuta (Figure 10).

Angiotensin II receptor Allele frequencies

120 n=356 n=151 n=61 n=144 100 80 Allele A 60 C Percent 98.3 n=6 98.1 n=3 98.4 n=1 98.6 n=2 40 20 1.7 1.9 1.6 1.4 0 Kasigau Rukanga Makwasinyi Bughuta Location

Figure 9: AGTR1 rs5186 allele frequencies for Kasigau. n= the number of alleles present in each sublocation. 26

To assess whether allele frequency and genotype varied based upon severity of

EH, the derived allelic and genotypic distributions of AGTR1 rs5186 were determined for

normotensive, prehypertensive, Stage I EH, and Stage II EH individuals for the total

Kasigau population and for the individual sublocations. For AGTR1 rs5186, there was a

higher frequency of A alleles than C alleles across all EH classifications and locations

(Figure 9). The A allele frequency in Kasigau as a whole ranged from 0.97-0.99 for all

EH classifications, while the C allele frequency ranged from 0.01-0.03 (Table 4).

When the derived genotype frequencies of the AGTR1 rs5186 SNP in each EH

category were compared with the expected frequencies predicted by HWE, the

polymorphism demonstrated nonsignificant differences for all locations across all EH

phenotypes (Χ2 < 3.84 with one degree of freedom) (Table 4). It should be noted that

HWE could not be determined for many of the prehypertensive, Stage I, and Stage II groups. This is to be expected since the C allele is not highly represented in this small

sample size and moreover since a known selection exists against the C/C genotype.

Angiotensin II receptor Genotypes

120 n=175 n=74 n=30 n=71 100 Genotype 80 A/A 60 A/C

Percent 96.7 96.1 96.8 97.3 C/C 40 20 n=6 n=3 n=1 n=2 3.3 3.9 n=0 n=0 3.2 n=0 2.7 n=0 0 Kasigau Rukanga Makwasinyi Bughuta 2 2 2 2 Χ = 0.25 Χ = 0.03 Location Χ = 0.05 Χ = 0.22 p= 0.82 p= 0.86 p= 0.93 p= 0.91

Figure 10: AGTR1 rs5186 genotype distribution for Kasigau. n= the total number of

participants. Hardy-Weinberg equilibrium was tested using the Χ2 test. 27

Table 4: AGTR1 rs5186 genotype distribution and allele frequencies per EH classification in Kasigau. HWE, Hardy-Weinberg equilibrium.

Rukanga Genotype Normotensive Prehypertensive Type I EH Type II EH A/A 10 18 18 29 A/C 1 0 0 2 C/C 0 0 0 0 HWE X2= 0.03 X2= N/A X2= N/A X2= 0.05 p= 0.87 p= N/A p= N/A p= 0.85 Allele Normotensive Prehypertensive Type I EH Type II EH

A 21 36 36 60 C 1 0 0 2 f (A) 0.95 1.00 1.00 0.97 f (C) 0.05 0.00 0.00 0.03 Makwasinyi Genotype Normotensive Prehypertensive Type I EH Type II EH

A/A 2 10 7 11 A/C 0 1 0 0 C/C 0 0 0 0 HWE X2= N/A X2= 0.11 X2= N/A X2= N/A p= N/A p= 0.87 p= N/A p= N/A Allele Normotensive Prehypertensive Type I EH Type II EH

A 4 21 14 22 C 0 1 0 0 f (A) 1.00 0.95 1.00 1.00 f (C) 0.00 0.05 0.00 0.00 Bughuta Genotype Normotensive Prehypertensive Type I EH Type II EH

A/A 27 24 20 10 A/C 1 0 0 1 C/C 0 0 0 0 HWE X2= 0.04 X2= N/A X2= N/A X2= 0.04 p= 0.92 p= N/A p= N/A p= 0.87 Allele Normotensive Prehypertensive Type I EH Type II EH

A 35 48 40 21 C 1 0 0 1 f (A) 0.97 1.00 1.00 0.95 f (C) 0.03 0.00 0.00 0.05 Kasigau Genotype Normotensive Prehypertensive Type I EH Type II EH

A/A 33 50 44 47 A/C 2 1 0 3 C/C 0 0 0 0 HWE X2= 0.39 X2= 0.05 X2= N/A X2= 0.05 p= 0.86 p= 0.94 p= N/A p= 0.83 Allele Normotensive Prehypertensive Type I EH Type II EH

A 68 101 88 97 C 2 1 0 3 f (A) 0.97 0.99 1.00 0.97 f (C) 0.03 0.01 0.00 0.03

28

Allele and genotype frequency of the HSD11β2 rs5479 SNP correlated with EH

classification

The HSD11β2 gene product is responsible for inactivation of cortisol by

converting it to cortisone. When the kidney variant, HSD11β2, is abundant, cortisol levels

increase resulting in salt and water reabsorption leading to increases in BP. Previously,

the A allele in HSD11β2 rs5479 SNP has been identified as a salt sensitive genetic variant

(Dongfeng, 2010). In Kasigau, 0.69 of the population possess the C variant of HSD11β2,

while 0.31 carry the A allele (Figure 11). Allelic frequency was further assessed as

genotype for rs5479; Kasigau displayed 0.46 C/C, 0.46 A/C, and 0.08 A/A (Figure 12).

Moreover, when the observed genotypes were compared with the expected frequencies predicted by HWE, all locations exhibited a nonsignificant difference, p= 0.33, 0.23,

0.78, and 0.87, respectively for Kasigau, Rukanga, Makwasinyi, and Bughuta (Figure

12).

To assess whether allelic frequency varied based upon the severity of EH, the

derived allelic and genotypic distributions of rs5479 were determined for normotensive,

prehypertensive, Stage I EH, and Stage II EH, both for the total Kasigau population, and

for the three sublocations. Regardless of sublocation, for HSD11β2 rs5479 there were

more C alleles than A alleles across all EH classifications (Table 5). The A allele

frequency for Kasigau across all EH classifications ranged from 0.27-0.37, while the C

allele frequency ranged from 0.63-0.73. When the obtained genotype frequencies of

HSD11β2 rs5479 in each EH category were compared with the expected frequencies

predicted by the HWE, the polymorphism exhibited a nonsignificant (X2 < 3.84 with one

degree of freedom) difference for all locations across all EH categories. While not above

the 3.84 value for HWE, both prehypertensive individuals from Rukanga (X2= 3.79, p ≤ 29

0.05) and normotensive participants in Kasigau as a whole (X2= 3.71, p= 0.06) had X2 values that might suggest, with an increase in sample size, that they are not in equilibrium. Interestingly, while odds ratio analyses need to be performed, there is a general trend in all locations but Bughuta, that the susceptible A allele increases as severity of EH increases. An example of this is that there are more A alleles in participants with Stage II EH versus normotensive (Table 5).

Hydroxysteroid (11-beta) dehydrogenase 2 Allele frequencies

80 n=245 n=100 n=40 n=105 70 60

50 Allele n=109 n=50 n=20 n=39 40 A 69.2 66.7 66.7 72.9

Percent C 30 20 30.8 33.3 33.3 27.1 10 0 Kasigau Rukanga Makwasinyi Bughuta Location

Figure 11: HSD11β2 rs5479 allele frequencies for Kasigau. n= the number of alleles present in each sublocation.

30

Hydroxysteroid (11-beta) dehydrogenase 2 Genotypes

60 n=82 n=31 n=13 n=38 n=81 n=38 n=14 n=29 50 Genotype

40 A/A A/C 46.3 41.3 43.3 52.8 30 C/C

Percent 45.8 50.7 46.7 40.3 20 n=14 n=6 n=3 n=5 10 7.9 8.0 10.0 6.9 0 Kasigau Rukanga Makwasinyi Bughuta 2 2 2 2 X = 0.96 X = 1.49 Location X = 0.08 X = 0.03 p= 0.33 p= 0.23 p= 0.78 p= 0.87

Figure 12: HSD11β2 rs5479 genotype distribution for Kasigau. n= the total number of participants in each sublocation. Hardy-Weinberg equilibrium was tested using the X2 test.

31

Table 5: HSD11β2 rs5479 genotype distribution and allele frequencies per EH classification in Kasigau. HWE, Hardy-Weinberg equilibrium. *= p ≤ 0.05.

Rukanga Genotype Normotensive Prehypertensive Type I EH Type II EH A/A 0 0 2 4 A/C 6 10 14 10 C/C 3 5 8 14 HWE X2= 2.27 X2= 3.79 X2= 1.4 X2= 1.5 p= 0.13 p= 0.05* p= 0.23 p= 0.34 Allele Normotensive Prehypertensive Type I EH Type II EH

A 6 10 16 18 C 12 20 30 38 f (A) 0.33 0.33 0.35 0.32 f (C) 0.67 0.67 0.65 0.68 Mwakwasinyi Genotype Normotensive Prehypertensive Type I EH Type II EH

A/A 0 0 1 2 A/C 1 6 4 3 C/C 1 5 2 5 HWE X2= 0.21 X2= 1.57 X2= 0.15 X2= 1.24 p= 0.64 p= 0.21 p= 0.66 p= 0.28 Allele Normotensive Prehypertensive Type I EH Type II EH

A 1 6 6 7 C 3 16 8 13 f (A) 0.25 0.27 0.43 0.35 f (C) 0.75 0.73 0.57 0.65 Bughuta Genotype Normotensive Prehypertensive Type I EH Type II EH

A/A 0 3 1 1 A/C 8 8 9 4 C/C 10 12 10 6 HWE X2= 1.49 X2= 0.58 X2= 0.37 X2= 0.04 p= 0.23 p= 0.39 p= 0.57 p= 0.78 Allele Normotensive Prehypertensive Type I EH Type II EH

A 8 14 11 6 C 28 32 29 16 f (A) 0.22 0.30 0.28 0.27 f (C) 0.78 0.70 0.72 0.73 Kasigau Genotype Normotensive Prehypertensive Type I EH Type II EH

A/A 0 3 4 7 A/C 15 24 25 17 C/C 14 22 21 25 HWE X2= 3.71 X2= 0.96 X2= 0.92 X2= 2.13 p= 0.06 p= 0.28 p= 0.36 p= 0.17 Allele Normotensive Prehypertensive Type I EH Type II EH

A 15 30 33 31 C 43 68 67 67 f (A) 0.26 0.31 0.33 0.32 f (C) 0.74 0.69 0.67 0.68

32

DISCUSSION

The allelic and genotype frequencies of three different SNPs of the RAS, known to influence predisposition to EH (rs699, rs5479, and rs5186), were evaluated from a population of individuals living in the rural villages of Kasigau, Kenya. The allele frequencies were compared across normotensive, prehypertensive, Stage I hypertensive, and Stage II hypertensive BP responses from the same individuals. Additionally, the genotype distribution in each classification was evaluated for HWE. Overall, this work begins to investigate how polymorphic variation in RAS genes influence phenotypic differences of EH in Kasigau, Kenya.

Most of the genotype frequencies for all three SNPs were in HWE when evaluated based upon EH classification; this is consistent with other studies as well

(Rhodes & Bell, 2009). One exception to the observed HWE was the genotype frequency of the rs699 SNP in participants from Bughuta. Additionally, prehypertensive participants from Rukanga are not in HWE of rs699 according to this study.

The most significant finding of this study was that most of the participants possessed the susceptible rs699 variant. To our knowledge, this is the first report to identify a high prevalence of this susceptible allele in a cohort of East African participants. The current study also discovered a geographical difference of the frequency of the alleles in Kasigau. Tribal differences may possibly be contributing to this observation since the sublocation of Bughuta has the most non-Taita persons residing there (Williams, 2012).

The AGT rs699 SNP may be contributing to the severity of EH observed in this study. The ancestral T allele, which codes for methionine, encodes AGT at a normal rate.

Meanwhile, the susceptible variant is responsible for inappropriate amounts of the 33

angiotensinogen protein being produced. The amount of protein within the body can vary depending upon other, nongenetic factors. Excess or deficiencies of the angiotensinogen protein has been shown to have detrimental clinical effects that expand beyond, yet include EH and other CVD (Deng, 2007). Therefore, the amount of angiotensinogen within the majority of Kasigau people is likely altered, which can influence the rate, acquisition, and stage of EH. This study found that in all three sublocations, the T allele frequency was greatest in either Stage I or Stage II EH participants.

While previous studies indicate a high prevalence of EH in Kasigau, it may conceivably be even higher (Williams, 2012). It is possible that many Kasigau individuals with the most severe forms of EH (Stage II EH) were unable to participate in this study because of their poor health due to cardiovascular complications and disease.

This is especially true because premature death is also a suspected effect caused by the rs699 susceptible C allele variant (Giner et al., 2000), something that is particularly concerning based upon the results of this study. The majority of participants possessed at least one C allele.

Globally, CVD and EH increase in populations as susceptible variant alleles increase in frequency (Mills, 2011). In this study, there is a slight increase throughout

Kasigau in EH prevalence for heterozygote carriers of the susceptible allele of rs699.

However, in Kasigau, those with the homozygous ancestral allele condition of rs699 were primarily normotensive.

The A allele is the predominant allele for AGTR1 in this population. This finding though can be attributed to the rare occurrence of the C allele of AGTR1 found globally

(Amir et al., 2009). Previous research has shown that the presence of the susceptible C allele in AGTR1 correlates with increased severity of cardiovascular conditions (Poch et 34

al., 2001). There is a complete loss of function or mutant form of the receptor in individuals possessing the homozygous C/C condition (Poch et al., 2001). The buildup of angiotensin II results from AGTR1 malfunctioning. The lack of C/C genotype in this study was not surprising based upon the results of prior studies (Gillespie et al., 2005).

Only heterozygote individuals carried the C allele for rs5186. In this study, the A/A genotype was the majority genotype, which indicates a functional angiotensin receptor is present in most Kasigau participants. The stage of EH does not appear to be associated with any particular allele or genotype of AGTR1 in this study and Kasigau overall. Allele and genotype frequencies were consistent with those observed in this study for all three sublocations of Kasigau.

This study evaluated the HSD11β2 rs5479 SNP and the susceptibility of developing EH within an East African population. Previous studies have found that the A allele for rs5479 is rarer than the C allele (Friso, Pizzolo, Choi, Guarini, & Castagna,

2008). The current study also found that the C allele was more abundant in Kasigau individuals with EH. of the C allele of rs5479 to the A variant has been shown to lead to extracellular volume expansion of the kidney and adrenal cortex (Friso et al.,

2008). Kasigau participants with Stage I and Stage II EH were more frequently found to possess the A/A genotype than that of normotensive participants in this study.

Previous studies have identified RAS gene polymorphisms and EH within many ethnic groups (Deng, 2007), but none have thoroughly investigated EH in East African populations. There are many reasons for this exclusion, although many of the Kasigau results are comparable to other areas of the world. EH manifests itself universally as the most critical condition related to CVD (Edwards et al., 2000), regardless of the similarities, disparities, and differences that exist among ethnic groups and geography. 35

The EH severity, rates of CVD, and associated genotype data that were determined for other ethnic communities such as Japanese and European descendants (Wang, Poole,

Treiber, Harshfield, & Hanevold, 2006) are similar to the distribution of alleles in

Kasigau. From the International HapMap project (http://hapmap.ncbi.nlm.nih.gov), populations of people from African, American, East Asian, and European decent were used as sources of comparison. This database is not exclusive to RAS genes, but rather to each individual SNP selected for this study.

Included in the HapMap project are subpopulations of participants with African ancestry, specifically Luhya in Webuye, Kenya and Yoruba in Ibadan, Nigeria. This ancestry is most relevant to the Kasigau data because of similar demographic and geographic origins. Nigerians are West African and had different ancestry during African migration. Therefore, the Luhya population is likely the most genetically similar to the participants from Kasigau.

The allelic distribution data of rs699 for the overall Kasigau population was similar to the frequencies reported for Luhya. There were little differences discovered in the comparison of SNPs genotype distributions between Kasigau and Luhya, but the largest difference was discovered as a slightly higher C allele frequency was discovered for AGTR1 in Kasigau than in Luhya. Overall, individuals in the Kasigau population are least like individuals from the Southwest United States and Nigeria, the other two sublocations listed within the SNP comparison database as populations of African ancestry.

This study also found that Bughuta, the most populated of the sublocations, had the least diverse genotype and allelic distribution. There was less genetic diversity in the most populated rural villages of Tanzania as well (Edwards et al., 2000). In parts of the 36

world with moderate to high rates of CVD, investigators have found no significant correlation between genetic diversity and susceptibility to CVD or changes in BP (Gupta et al., 2004).

The results of the present study contribute critical information to a developing database of knowledge about EH, the RAS, and genetics of Kenyan people. This study relied on previously established parameters such as ancestral and susceptible alleles, frequencies of genotypes, frequencies of alleles, and other relevant details related to all three RAS gene SNPs. This study is among the first to study SNPs related to EH in East

African populations. In the future, the results derived from this study will be used to help promote awareness and treatment of how genetic elements impact disease in developing countries. In addition to discovering more about the genetic components of RAS genes and their contributions to the development of EH within the people of Kenya, this study has implications for many additional areas of research.

Finally, environment-gene interactions should also be evaluated. Epigenetic factors, gene-gender interactions, aging effects, and salt sensitivity have been shown to influence the development of EH (Mathers & Loncar, 2006) and need to be addressed in future studies. This is particularly relevant to Kasigau for the genetic SNPs and EH classification differences that resulted from this study.

37

APPENDIX A: Western Kentucky University Institutional Review Board approval (HR-174)

38

APPENDIX B: Taita District approval

39

APPENDIX C: Informed consent document

40

REFERENCES

Addo, J., Smeeth, L., & Leon, D.A. (2007). Hypertension in Sub-Saharan Africa: a

systematic review. Hypertension. 50: 1012-1018.

Amir, O., Amir, R.E., Paz, H., Attias, E., Sagiv, M., & Lewis, B.S. (2009). Relation

between AT1R gene polymorphism and long-term outcome in patients with heart

failure. Cardiology. 112: 151-7.

Applied Biosystems (2010). TaqMan® Universal PCR Master Mix Protocol. Foster City,

CA, USA.

Bauml, M.A. & Underwood, D.A. (2010). Left ventricular hypertrophy: an overlooked

cardiovascular risk factor. Cleveland Clinic Journal of Medicine. 77: 381-387.

Bonnardeaux, A., Davies, E., Jeunemaitre, X., Fery, I., Charrru, A., Clauser, E., Tiret, L.,

Cambien, F., Corvol, P., & Saubrier, F. (1994). Angiotensin II type 1 receptor gene

polymorphisms in human essential hypertension. Hypertension. 24: 63-69.

Carvajal, C.A., Romero, D.G., Mosso, L.M., González, A.A., Campino, C., Montero, J.,

& Fardella, C.E. (2005). Biochemical and genetic characterization of 11 β-

hydroxysteroid dehydrogenase type 2 in low-renin essential hypertensives.

Journal of Hypertension. 23: 71-77.

DeFea, K. (2008). Beta-arrestins and heterotrimeric G-proteins: collaborators and

competitors in signal transduction. British Journal of Pharmacology. 153: 298-

309.

Deng, A.Y. (2007). Genetic basis of polygenetic hypertension. Human Molecular

Genetics. 16: R195-R202.

41

Dongfeng, D., Kelly, T.N., Hixson, J.E., Chen, J., Liu, D., Chen, J.C., Rao, D.C., Mu, J.,

Ma, J., Jaquish, C.E., Rice, T.K., Gu, C., Hamm, L.L., Whelton, P.K., & He, J.

(2010). Genetic variants in the rennin-angiotensin-aldosterone system and salt-

sensitivity of blood pressure. Journal of Hypertension. 28: 1210-1220.

Doris, P.A. (2002). Hypertension Genetics, Single Nucleotide Polymorphisms, and the

Common Disease: Common Variant Hypothesis. Hypertension. 39: 323-331.

Edwards, R., Unwin, N., & Mugusi, F. (2000). Hypertension prevalence and care in urban

and rural areas of Tanzania. Journal of Hypertension. 18: 145-152.

Friso, S., Pizzolo, F., Choi, S.W., Guarini, P., & Castagna, A. (2008). Epigenetic control

of 11 beta-hydroxysteroid dehydrogenase 2 gene promoter is related to human

hypertension. Atherosclerosis. 199: 323–327.

Gillespie, E.L., White, C.M., Kardas, M., Lindberg, M., & Coleman, C.I. (2005). The

impact of ACE inhibitors or angiotensin II type 1 receptor blockers on the

development of new-onset type 2 diabetes. Diabetes Care. 28: 2261-2266.

Giner, V., Poch, E., Bragulat, E., Oriola, J., Gonzalez, D., Coca, A., & de la Sierra, A.

(2000). Renin-Angiotensin System Genetic Polymorphisms and Salt Sensitivity in

Essential Hypertension. Hypertension. 35: 512-517.

Gupta, P.C., Gupta, R., & Pednekar, M.S. (2004). Hypertension prevalence and blood

pressure trends in 88,653 subjects in Mumbai, India. Journal of Human

Hypertension. 18: 907-910.

Hendriks, M.E., Wit, F.W.N.M., Roos, M.T.L., Brewster, L.M., Akande, T.M., de Beer,

I.H., Mfinanga, S.G., Kahwa, A.M., Gatongi, P., Rooy, J., Janssens, G. W.,

42

Lammers, J., Kramer, B., Bonfrer, I., Gaeb, E., van der Gaag, J., Rinke de Wit,

T.F., Lange, J.M.A., & Schultsz, C. (2012). Hypertension in Sub-Saharan Africa:

Cross-Sectional Surveys in Four Rural and Urban Communities. PLoS ONE. 7:

e32638. Doi:10.1371/journal.pone.0032638.

Iigaya, K., Kumagai, H., Nabika, T., Harada, Y., Onimaru, H., Oshima, N., Takimoto, C.,

Kamayachi, T, Saruta, T., & Itoh, H. (2009). Relation of Blood Pressure

Quantitative Trait Locus on Rat Chromosome 1 to Hyperactivity of

Rostralventrolateral Medulla. Hypertension. 53: 42-48.

Jamison, D.T., Feachem, F.G., Makgoba, M.W., Bos, E.R., Baingana, F.K., Hofman, K.J.,

& Rogo, K.O. (2006). Disease and Mortality in Sub-Saharan Africa. Washington,

D.C. World Bank.

Kaplan, N.M., & Victor, R.G. (2009). Kaplan's Essential Hypertension (10th ed.).

Wolters Kluwer Health.

Katholi, R.E., & Couri, D.M. (2011). Left Ventricular Hypertrophy: Major Risk Factor in

Patients with Hypertension: Update and Practical Clinical Applications.

International Journal of Hypertension. 495349: 1.

Mathers, C. D. & Loncar, D. (2006). Projections of global mortality and burden of

disease from 2002 to 2030. PLoS Med. 3 (11): e442.

Mills, R.M. (2011). Epigenetics and hypertension. Current Hypertension Reports. 13: 21-

28.

Miranda, J.J., Kinra, S., Casas, J.P., Smith, G.D., & Ebrahim, S. (2008). Non-

communicable diseases in low- and middle-income countries: context,

43

determinants and health policy. Tropical Medicine and International Health. 13:

1225-1234.

Mune, T., Rogerson, F.M., Nikkilä, H., Agarwal, A.K., & White, P.C. (1995). Human

hypertension caused by mutations in the kidney isozyme of 11 beta-

hydroxysteroid dehydrogenase. Nature Genetics. 10: 394-399.

Murphy, T.J., Alexander, R.W., Griendling, K.K., Runge, M.S., & Bernstein K.E. (1991).

Isolation of cDNA encoding the vascular type-1 angiotensin II receptor. Nature.

351: 233-236.

Murray, R.K., Granner, D.K., Mayes, P.A., & Rodwell, V.W. (2000). Harper's

Biochemistry. 25th Ed. Appleton & Lange, Stanford, CA.

Nanodrop Technologies, Inc. (2007). 260/280 and 260/230 Ratios. Technical Support

Bulletin. DE, USA.

Ortlepp, J.R., Metrikat, J., Mevissen, V., Schmitz, F., Albrecht, M., Maya-Pelzer, P.,

Hanrath, P., Zerres, K., & Hoffmann, R. (2003). Relation between the

angiotensinogen (AGT) M235T gene polymorphism and blood pressure in a large,

homogeneous study population. Journal of Human Hypertension. 17: 555-559.

Persu, A. (2005). 11β-Hydroxysteroid deshydrogenase: a multi-faceted enzyme. Journal

of Hypertension. 23: 29-31.

Pickering, G. (1965). High blood pressure without evident cause: essential hypertension.

British Medical Journal. 2: 1021-1026.

Poch, E., Gonzalez, D., Giner, V., Bragulat, E., Coca, A., & de la Sierra, A. (2001).

Molecular basis of salt-sensitivity in human hypertension: evaluation of rennin-

44

angiotensin-aldosterone system gene polymorphisms. Hypertension. 38: 1204-

1209.

Qiagen® (2010). QiaAmp® DNA Mini and Blood Mini Handbook. 3rd Ed: 37-39.

Valencia, CA, USA.

R Development Core Team. (2008). R: A language and environment for statistical

computing. R Foundation for Statistical Computing, Vienna, Austria.

Rhodes, R.A. & Bell, D.A. (2009). Medical Physiology: Principles for Clinical Medicine.

3rd Ed. Lippincott, Williams, and Wilkins, Baltimore, MD.

Tobin, M.D., Sheehan, N.A., Scurrah, K.J., & Burton, P.R. (2005). Adjusting for

treatment effects in studies of quantitative traits: antihypertensive therapy and

systolic blood pressure. Statistics in Medicine. 24: 2911-2935.

Wang, X., Poole, J.C., Treiber, F.A., Harshfield, G.A., & Hanevold, C.D. (2006). Ethnic

and gender differences in ambulatory blood pressure trajectories: results from a

15-year longitudinal study in youth and young adults. Circulation. 114: 2780–

2787.

Williams, L. (2012). The Prevalence of Essential Hypertension in Kasigau, Kenya.

Honors College Capstone Experience/Thesis Projects. Paper 363.

http://digitalcommons.wku.edu/stu_hon_theses/363.

Wilson, R.C., Dave-Sharma, S., Wei, J., Obeyesekere, V.R., Li, K., Ferrari, P.,

Krozowski, Z.S., Shackleton, C.H.L., Bradlow, L., Wiens, T., & New, M.I. (1998).

A genetic defect resulting in mild low-renin hypertension. PNAS. 95: 10200-

10205.

45

World Heath Organization (2011). Global status report on noncommunicable diseases.

46