A NOVEL 11-BLOCKER BASED ON ANTISENSE OLIGONUCLEOTIDES:
A NEW GENE THERAPY APPROACH IN HYPERTENSION,
CARDIAC HYPERTROPHY AND MYOCARDIAL ISCHEMIA
YUAN CLARE ZHANG
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
Dedicated to my husband Ryan, and my parents.
I would like to thank Dr. M. lan Phillips, Chairman of my supervisory
committee, for everything he has done for me in the past four years. I like his
hands-off style which makes me an independent thinker; I appreciate his faith in
me and enjoy the feeling of being backed up wherever I go; I cherish the
moments we talked about antisense, sense and nonsense. I am grateful for
having the best mentor who is a scientist, teacher, poet, friend and like a father.
I would like to thank Dr. Stephen Baker for his insightful comments and
providing the cell culture for in vitro experiments, and Dr. Jeffrey Hughes for
providing liposomes throughout my study. I also enjoyed the discussions with Dr.
Craig Gelband and his frank remarks.
I would like to thank Dr. William Hauswirth, Dr. Collin Sumners and Dr.
Luiz Belardinelli for their tremendous support and valuable suggestions.
I would like to thank my fellow student, Jon Bui, for his unselfish help in
my research and personal life. I sincerely appreciate Dr. Sara M. Galli, Leping
Shen, Dr. Keping Qian, Birgitta Kimura and Harold Snellen for making this
laboratory a wonderful place to study and stay.
Finally, my deepest thanks go to my parents, my husband and my big
family who have made everything possible.
TABLE OF CONTENTS
0 0 0 0 0 0 0 *
1.1 pi-Adrenoceptors *
1.2 Hypertension: Definition, Prevalence and Risks
1.3 pl-Adrenoceptors and Cardiovascular Diseases
1.4 Why Antisense Gene therapy for pi-Blockade?
1.5 Gene Delivery Vehicles (Viral vs. Nonviral Vectors)
1.6 Specific Aims *
2 GENERAL METHODOLOGY
2.1 Biochemical Methods 0 0 0 .
2.2 Molecular Biology Methods *
2.3 Pharmacological Methods .0 *. 0 0
2.4 Physiological Methods .
2.5 Clinical Pathology Methods *
2.6 Statistical Analysis .
3 OPTIMIZATION OF ANTISENSE DESIGN AND DELIVERY
* 0 0 5 0 5
* S S S S 5
* S S 5 0 0
4 EFFECTS ON BLOOD PRESSURE AND HEART RATE:
pi-AS-ODN VS. ATENOLOL
* 0 .
* 0 0
0 0 0 & 0 0 0 0 0 0 0
& 0 0 0 0 a 0 0 0 0 0 0 0 0 0 0 0
5 MECHANISMS FOR THE ANTIHYPERTENSIVE
EFFECTS OF pl-AS-ODN .
5.1 Introduction .
5.2 Results .
5.3 Discussion 0 .
6 REPEATED ADMINISTRATION OF p3-AS-ODN
AND SAFETY PROFILE 9
6.1 Introduction 0 .
6.2 Results .
6.3 Discussion .
7 PROTECTION BY pl-AS-ODN AGAINST MYOCARDIAL
DYSFUNCTION INDUCED BY ISCHEMIA-REPERFUSION
7.1 Introduction 0 0 0
7.2 Results 0 0 .
7.3 Discussion .
8 CONCLUSIONS AND FUTURE DIRECTIONS .
8.1 Conclusions .* .
8.2 Limitations of the Current Approach .
8.3 Future Directions .. *
PATENT AND PUBLICATIONS *
REFERENCE LIST .
BIOGRAPHICAL SKETCH *
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
A NOVEL pl-BLOCKER BASED ON ANTISENSE OLIGONUCLEOTIDES:
A NEW GENE THERAPY APPROACH IN HYPERTENSION,
CARDIAC HYPERTROPHY AND MYOCARDIAL ISCHEMIA
Yuan Clare Zhang
Chairman: M. lan Phillips
Major Department: Physiology
Objectives: p-Adrenoceptor antagonists (p-blockers) are effective in treating
hypertension, heart failure and myocardial infarction, but they have poor patient
compliance because of the daily regimen and side effects such as central
nervous system (CNS) reactions and nonspecific inhibition of p2-adrenoceptor
(P2-AR). To overcome these problems, I have developed a novel and specific pl-
blocker by targeting 31-adrenoceptor (pl-AR) mRNA with antisense
oligonucleotides (p1-AS-ODN) and optimized systemic delivery with cationic
liposomes. The effects of pi-AS-ODN on blood pressure, left ventricular
hypertrophy (LVH) and myocardial ischemia were tested in spontaneously
hypertensive rats (SHR) and isolated rat hearts subjected to
ischemia/reperfusion. The antihypertensive mechanisms were investigated and
the safety profile of repeated administration of p3-AS-ODN was studied.
Results: A single intravenous injection of 0.5-1 mg/kg l1-AS-ODN with cationic
liposomes significantly decreased blood pressure by 25-35 mmHg for 20-33 days
in SHR (P<0.05), with no effect on heart rate. 11-AS-ODN significantly reduced
the pl-AR levels in the heart and kidney, without affecting p2-AR. The
mechanisms for blood pressure reduction appeared to involve a rapid decrease
in cardiac contractility and a delayed but marked reduction in plasma renin
activity. Quantitative autoradiography revealed no changes in p-AR in the brain,
indicating the absence of CNS effects. pl-AS-ODN is effective with repeated
injections. It not only produced a sustained reduction in blood pressure, but also
significantly reduced LVH in SHR. No toxicity (assessed by liver transaminases,
hematology, and tissue histopathology) or immune response was observed
during long-term treatment. In addition, pretreatment with p1-AS-ODN
significantly attenuated cardiac dysfunction and lipid peroxidation of isolated rat
hearts induced by ischemia-reperfusion.
Conclusion: Systemic injection of pl-AS-ODN delivered by cationic liposomes
can reduce high blood pressure and cardiac hypertrophy, and can protect the
heart from ischemic injury. It is clearly longer-lasting than currently available p-
blockers and likely to have fewer side effects because of its p1-specificity and
lack of CNS effects and bradycardia. Repeated administration is effective without
causing toxicity or immune stimulation. Therefore 1p-AS-ODN is viable as a
potentially new prolonged antihypertensive agent with benefits for cardiac
remodeling and myocardial protection from ischemia.
The sympathetic nervous system (SNS) plays an important role in the
regulation of a wide variety of physiological functions that are essential for
survival and quality of life. Most pbstganglionic sympathetic nerve terminals
release norepinephrine as a neurotransmitter, which exerts its effects through
interaction with adrenoceptors. Adrenoceptors (AR) were first subdivided into a
and p subtypes by Ahlquist in 1948 (1). p-AR was further classified into P, and 13
subtypes in 1967 (2) based on their relative potencies for the endogenous
agonists adrenaline (epinephrine) and noradrenaline (norepinephrine). A third
subtype P3 was discovered in 1989 (3). 3,-AR is the predominant AR subtype
expressed in the myocardium and renal cortex. When activated, P,-AR induces
positive inotropic, chronotropic and lusitropic responses to catecholamines in the
heart, i.e. an increase in myocardial contractility, heart rate and relaxation speed.
It also stimulates renin release from the juxtaglomerular apparatus in the kidney
and upregulates the peripheral renin-angiotensin system (4,5). Therefore the
combined effect of p,-adrenergic activation is an elevation in cardiac output and
blood pressure. P,-AR is essential for normal cardiac performance. Its
importance is manifested by the prenatal death of P,-AR knockout mice (6) and
by heart failure associated with impaired P,-AR functions (7,8).
p,-AR belongs to the superfamily of G-protein-coupled receptors that have
seven hydrophobic transmembrane domains. The ligand-binding site is located
in the transmembrane regions (9), and the third intracellular hydrophilic loop
appears to be most important for coupling 3,-AR to G protein (10). The coding
sequence of human P,-AR is 1434 bp long and contains no introns (11).
Although P,-AR has been characterized for many years, the understanding of its
intracellular signal transduction has largely derived from the study of P,-AR. The
putative signaling pathway for cardiac 3,-AR involves stimulatory G protein (Gs),
activation of adenylyl cyclase (AC), accumulation of cAMP, stimulation of cAMP-
dependent protein kinase A (PKA), and phosphorylation of key target proteins
(including L-type calcium channel [IcL], phospholamban and troponin).
P,-AR undergoes desensitization and down-regulation when exposed to
the agonists. Rapid desensitization occurs after acute activation (within minutes)
which appears to be through receptor phosphorylation and sequestration (12).
Chronic activation leads to receptor down-regulation which involves reduced
mRNA levels and enhanced degradation of P,-AR (13). Diminished density and
functional responsiveness of cardiac 3,-AR are common features of chronic
heart failure and myocardial infarction that result from an excessive adrenergic
drive in these diseases (7,14).
1.2 Hypertension: Definition, Prevalence and Risks
Hypertension, manifested by chronic high blood pressure, affects a large
population and is a silent killer leading to severe cardiovascular problems. The
conceptual definition of hypertension, according to Kaplan in his Clinical
Hypertension (15), is "the level of blood pressure at which the benefits (minus
the risks and costs) of treatment exceed the risks and costs (minus benefits) of
inaction." Numerically, hypertension is defined by the World Health Organization
(WHO) as the blood pressure (BP) higher than 160/95 mmHg (systolic/diastolic
BP). However, hypertension is treated more aggressively in the United States
with the clinical end point <140/90 mmHg. It is estimated that hypertension
afflicts one fourth of adults in America, i.e. 50 million people (16,17), and is the
No. 1 reason for office visits to specialists in internal medicine (18). However,
due to the asymptomatic nature of this disease, many patients are largely
unaware of high BP and therefore have little motivation to seek or follow
treatment. As a result, only about 50% of hypertensives in the U.S. get their BP
The danger of hypertension does not lie in high BP itself, but in the
complications of high BP. Hypertensives generally suffer premature disability or
death from cardiovascular diseases, because chronic untreated high BP
increases the incidence of deadly events such as stroke, heart failure,
myocardial infarction and renal dysfunction (16,19). The mechanisms are
complex and may involve the structural changes in blood vessels and cardiac
tissues termed hypertrophy and remodeling, which eventually lead to organ
damage and functional insufficiency (20).
1.2.1 Hypertension: a Major Risk Factor for Heart Failure
Hypertension and heart failure are closely associated. Heart failure, a
pathophysiological state whereby the heart fails to pump sufficient blood to meet
the oxygen requirements of the body, is life-threatening and affects 4 million
Americans (21). The prevalence is rising with 400,000 new cases every year,
mostly in patients over 65 years old. The best available data on hypertension as
a prelude to heart failure were derived from the Framingham study (22). This
study involved follow-up of more than 5,000 patients for a median of 14 years,
and found that hypertension is the largest single risk factor for the development
of heart failure, accounting for 39% of attributable risk in men and 59% in
women. How hypertension degenerates into heart failure is not completely
understood. However, it is well accepted that the heart responds to a prolonged
pressure overload in three major ways (23): 1) left ventricular hypertrophy (LVH),
2) adrenergic stimulation of heart rate and contractility, and 3) volume expansion.
Mild hypertension increases the risk of LVH by 2-3 times, and severe
hypertension leads to tenfold elevated risk (24). LVH is initially a compensatory
response to hypertension which helps maintain normal cardiac performance, but
gradually progresses to cardiac remodeling and dysfunction. An overactive
adrenergic system, indicated by an increase in norepinephrine (NE) release (25)
and sympathetic nerve traffic (26,27), has been observed in the early stage of
essential hypertension and results in the diminished responsiveness of cardiac
P-adrenergic receptors in failing hearts (7).
1.2.2 Hypertension: a Major Risk Factor for Myocardial Infarction
Hypertension is also a major risk factor for myocardial infarction.
Myocardial infarction occurs when there is an abrupt decrease in the coronary
blood flow after the thrombotic occlusion of a coronary artery previously
narrowed by atherosclerosis. Approximately 1.5 million myocardial infarctions
occur in America every year, with about 30% mortality rate. In the Framingham
Heart Study, 30-year follow-up data revealed that the incidence of myocardial
infarction was proportional to the level of systolic BP in a wide range from
normotension to severe hypertension (28-30). Active treatment of isolated
systolic hypertension in the elderly resulted in a 27% reduction in myocardial
infarction and coronary heart disease (CHD)-related mortality (31). The
mechanisms by which high BP increases the risk of CHD are yet unclear.
Extensive studies on hypertensive rats revealed that Ca2+ is elevated excessively
in myocytes and coronary artery wall of these animals. The Ca2+ overload
increases the incidence of coronary spasm, which occurs in the early phase of
acute myocardial infarction (32). Systemic hypertension also induces LVH, which
reduces coronary flow reserve and renders the heart vulnerable to ischemia-
induced damage (33). In addition, chronic sympathetic activation leads to
endothelial dysfunction (34) and vascular smooth muscle hypertrophy (20) that
further promote the development of atherosclerosis.
1.3 pl-Adrenoceptors and Cardiovascular Diseases
1.3.1 Overactive SNS in Cardiovascular Diseases
Overactive sympathetic activity plays a crucial role in the pathogenesis of
cardiovascular diseases, including essential hypertension, congestive heart
failure and myocardial infarction (35-37). Nearly 30 years ago, Julius et al. (38)
observed that a noticeable fraction of subjects with borderline hypertension had
an increased sympathetic and a decreased parasympathetic drive. This finding
has since been supported by a large body of evidence. A meta-analysis of all
published data (25) showed that plasma NE levels as an indirect marker of
sympathetic tone were significantly elevated in younger patients with essential
hypertension than in normotensive individuals, which was accompanied by an
increase in heart rate and cardiac output (39). Concomitantly, NE spillover rate
(40-42) and muscle sympathetic nerve traffic (27,43) rise progressively with the
development of essential hypertension. Likewise, elevated plasma NE levels
were observed universally in symptomatic and asymptomatic heart failure
(44,45). Myocardial infarction is also associated with reduced baroreflex and
increased sympathetic activation (46). These data indicate the sympathetic
overactivation as a key participant in the initiation and maintenance of these
cardiovascular problems. PI-AR, a central mediator of SNS activity, is therefore
unequivocally involved in these deleterious events.
1.3.2 P3-Adrenoceptors and Hypertension
Pi-adrenergic activation induces positive inotropic and chronotropic
responses in the heart and stimulates renin release in the kidney, thereby
increasing cardiac output and upregulating the renin-angiotensin system (47). As
a result, P,-agonists such as dobutamine can rescue acute cardiac failure by
increasing myocardial contractility and heart rate (48). Pharmacological
interventions by p-blockers are effective in reducing mortality in high risk
arrhythmic patients and in lowering BP (49,50). All p-blockers can inhibit 31-AR
and share structural similarity to p-agonist isoproterenol. Since the introduction of
propranolol in 1964 (51), p-blockers have become the first line therapy for
treating hypertension as recommended by the Joint National Committee and
WHO. They are now proposed as the initial monotherapy for younger and
middle-aged hypertensives, especially white males, and for patients with high
levels of stress (52). The precise mechanisms by which these agents reduce BP
remain unclear (53). However it is generally agreed that the primary effect of p-
blockers is a reduction in cardiac output by 15-20%, resulting from the blockade
of cardiac p1-AR, thereby reducing both heart rate and myocardial contractility
(54). Besides, the inhibition of pi-AR in renal cortex suppresses renin production
and release, which may lead to a diminished peripheral renin-angiotensin system
(55). These concepts are supported by the observation that p-blockers are more
effective for patients with high cardiac output and heart rate (52) and high renin
1.3.3 pl-Adrenoceptors and Heart Failure
The importance of fully functional pi-AR in the maintenance of normal
heart performance is apparent, as the diminished functional responsiveness of
P3-AR in chronic heart failure is closely correlated with the severity of impaired
cardiac functions (7,58). Therefore pf-blockade was considered to contraindicate
heart failure for a long period of time. However, early medical attempts to treat
heart failure with p-adrenergic agonists showed only short-term benefits, and
then precipitated symptoms with chronic use (59). This observation challenged
the prevailing wisdom that p-adrenergic activation improved cardiac functions
and therefore led to the pioneering trial by Dr. Hjalmarson's group in Sweden
treating heart failure with p-blockers in 1975 (60). Since then, their success has
been confirmed by overwhelming clinical and experimental evidence. These data
suggest that in contrast to acute supportive effects, chronic p-adrenergic
activation is detrimental to the heart (37,61).
Several recently completed multi-national, multi-center, large-scale clinical
trials achieved clinical end points in heart failure patients on P-blockers, including
improvement of systolic functions, exercise capacity and quality of life; and
reduction in morbidity, hospitalization and mortality (62-67). In the Metoprolol in
dilated Cardiomyopathy (MDC) trial (n=383), metoprolol treatment in patients
with IDC resulted in a 34% reduction of the primary combined endpoint, total
number of deaths and need for cardiac transplantation (62). In the US Carvedilol
studies (n=1094), carvedilol reduced mortality by 65% in patients with ischemic
and idiopathic cardiomyopathy, which was highly significant (67). This led the
Food and Drug Administration (FDA) to approve carvedilol for chronic heart
failure. p-blockers are also effective in the prevention of heart failure in
hypertensive patients. In the STOP trial (n=1627), old hypertensive patients
treated with p-blockers (atenolol, metoprolol or pindolol) for an average of 25
months showed a significant reduction in heart failure rate and total death
Many mechanisms have been proposed for the protective role of p-
blockers in heart failure [reviewed in (70,71)]. They include the following:
Improvement in cardiac function by restoring the density and functional
sensitivity of p-AR that are desensitized and down-regulated after chronic
sympathetic activation (62).
Reduction in oxygen consumption and metabolic demand by decreasing
heart rate (72,73), cardiac contractility and thereby ventricular workload.
Regression of left ventricular hypertrophy (74,75).
Inhibition of the renin-angiotensin system (23).
Attenuation of cytotoxic and apoptotic effects of catecholamine (76,77).
Furthermore, recent advances in physiological genomics corroborate
these clinical observations. Cardiac overexpression of human p1-AR (78), Gas
(79) or Gaq proteins (80) in transgenic mice improved contractile force only at a
young age, then developed ventricular hypertrophy and systolic dysfunction,
which ultimately resulted in a cardiomyopathic phenotype and premature death.
The mechanisms are not clear, but growing evidence suggests that p1-AR
mediates the proliferative, cytotoxic (76) and apoptotic (77) effects of excessive
NE on cardiomyocytes.
1.3.4 pl-Adrenoceptors and Myocardial Infarction
During acute myocardial infarction, excessive reflex activation of SNS
kicks in and increases platelet aggregation, myocardial ischemia and arrhythmia
(81). Elevated SNS activity has been considered an independent predictor for
sudden cardiac death after myocardial infarction (46). p-blockers, by inhibiting
cardiac pI-AR, can attenuate the detrimental effects of sympathetic activation
and thus protect the heart from severe damage and ventricular fibrillation. Three
clinical trials published in 1981 (82-84) demonstrated for the first time that 3-
blockers started after the onset of myocardial infarction reduced total mortality.
Since then, over 55 randomized trials have been carried out in more than 55,000
patients. A summary of 24 long-term trials including about 25,000 patients
showed that p-blockers reduced total mortality by 20% and sudden cardiac death
by 34% over 2 years of follow-up after myocardial infarction (85). They also
significantly decreased the incidence of recurrent infarction (83). It is now
suggested that patients (especially older patients) with acute myocardial
infarction, who have no contraindications to p-blockers, should be treated with
early intravenous p-blockade followed by oral p-blockers indefinitely (81).
Many types of p-blockers including timolol, metoprolol, propranolol,
acebutolol, pindolol, carvedilol and bisoprolol have been found to significantly
reduce total mortality after myocardial infarction to similar extent. The property
they have in common is pi-blockade. Therefore pl-blockade appears to be the
essential life-saving ingredient.
In animal models of ischemia/reperfusion, the mRNA level and density of
pi-AR increased during acute infarction and decreased after chronic infarction
(14,86-88). Pretreatment with p-blockers significantly reduced infarct size and
prevented ventricular dysfunction compared with control animals (89).
The theoretical benefits of p-blockade in myocardial infarction include a
decrease in oxygen demand, an antiarrhythmic effect, and a positive influence
on myocardial metabolism (90). The decrease in oxygen demand results
primarily from a reduction in heart rate and contractility. Excessive release of
catecholamine during myocardial infarction can be arrhythmogenic. Sudden
cardiac death accounts for almost half of mortality after myocardial infarction,
which is due to myocardial tachyarrhythmia and fibrillation (91). p-blockers can
prevent arrhythmia and are the only drugs that have consistently reduced
sudden death in large trials (85). In addition, p-blockers prevent the
derangement of cardiac energy metabolism. The cellular mechanisms underlying
the beneficial effects of p-blockade in myocardial infarction are being studied
extensively, which seem to involve the protection of intracellular Ca2+
homeostasis and protein phosphorylation (92), antioxidant changes (93),
reduction in inflammatory cytokines (94), and favorable energy metabolism
1.4 Why Antisense Gene therapy for pi-blockade?
1.4.1 Drawbacks of Current p-blockers
It is evident that p-blockers are reliable, successful and relatively safe in
the control of hypertension, heart failure, myocardial infarction and arrhythmia. In
spite of the success, this traditional drug therapy has many shortcomings. For
example, the compliance for p-blockers in hypertensive patients has been less
than 50% by some estimates (97). The reasons are multiple. First, the obvious
asymptomatic nature of the initial phases of hypertension renders this disorder
underdiagnosed and undertreated. Second, p-blockers are associated with a
number of side effects due to nonspecific p2-antagonism (98) and central
nervous system (CNS) reaction (99). Another problem with p-blockers, which is
also true for all the available antihypertensive drugs, is short action, requiring a
mandatory daily regimen. This further discourages patient compliance and
produces a potentially dangerous peak-and-trough fluctuation of 24-hr BP (97).
Most of the unwanted effects of p-blockers are caused by the blockade of
p2-mediated functions, including bronchodilation, vasodilation, and mobilization
of glucose and free fatty acids (98). Limited by pharmacological drug design,
truly pi-specific antagonists have not been discovered, making these adverse
effects inevitable even with highly Pi-selective blockers at therapeutic doses
Bronchodilation and vasodilation: Pulmonary bed and blood vessels
express high levels of 12-AR, activation of which leads to bronchodilation and
vasodilation. The suppression of this receptor leaves the al-mediated
vasoconstriction unopposed and results in elevated airway and vascular
resistance (101,102). Therefore nonselective 3-blockers are generally
considered to contraindicate asthma and peripheral vascular diseases.
Blood sugar. p2-adrenergic activation promotes glycogenolysis and
gluconeogenesis in the liver and thereby helps maintain normal glucose levels in
the plasma. Nonselective p-blockers tend to decrease glucose tolerance. pi-
selective agents are preferred over nonselective agents for patients with insulin-
dependent diabetes mellitus, because nonselective p-blockers can increase the
duration and severity of insulin-induced hypoglycemia, and pi-selective agents
have a less pronounced effect on blood glucose levels (103,104).
Quality of life: Nonselective p-blockers such as propranolol markedly
decrease quality of life when compared with the ACE inhibitor captopril (105). In
contrast, pi-selective drugs cannot be differentiated from ACE inhibitors in terms
of overall quality of life scores (105) and are superior to the calcium channel
blocker nifedipine (106). In patients with chronic heart failure, pi-selective
blockers showed significant advantages over nonselective agents in improving
quality of life and NYHA class (62,63,65,66,107). The reason is that p2-mediated
metabolic functions are largely unimpaired with pi-selective blockade (108,109).
These functions, including mobilization of glucose and free fatty acids, are
important in providing energy supply to and removing unwanted by-products from
exercising muscles. Therefore, post-exercise hypoglycemia is unlikely to occur,
which contributes to lesser fatigue with p3-selective agents (110).
Blood lipid profile: Serum levels of total cholesterol and low-density
lipoprotein (LDL) are generally not affected by p-blockers. But nonselective
agents such as propranolol significantly increase triglyceride (TG) concentration
and decrease high-density lipoprotein (HDL) levels (104,111), which have been
associated with a higher risk of atherosclerosis (112-114). These effects are due
to 32-blockade, as the changes are less apparent with p1-selective agents such
as metoprolol and atenolol (110). With highly p1-selective agents such as
bisoprolol, the changes appear to be minimal or absent (115).
In addition to the side effects induced by p2-inhibition, common complaints
about p-blockers include depression, impotence, vivid dreams and insomnia
(99). Because these effects are more pronounced with lipophilic agents that are
able to cross the blood-brain barrier and enter the brain (116,117), they are
generally attributed to the suppression of central p-AR functions (116).
Most p-blockers have short half-lives in plasma between 3 and 10 hr (90).
Therefore a daily regimen is necessary. This problem is not unique to p-blockers,
and all the antihypertensive drugs available today have to be taken at least once
a day (97). Compulsory daily medication, along with adverse effects, discourages
patients from staying on antihypertensive therapy.
These issues led us to conclude that the traditional pharmacological
approach for p-blockade has reached a plateau. A new and innovative way must
be considered to overcome these problems and to achieve better control of BP
and heart dysfunction. Because hypertension and its cardiovascular
complications are chronic disorders that affect a large population, any
improvement in patient compliance may lead to significant reductions in
morbidity and mortality in the long run and carry with it potential economical
benefits to society.
1.4.2 Antisense Gene Therapy
Ideally, such a new therapy should provide 31-specific inhibition, and
stable and persistent lowering of BP without CNS side effects or toxicity.
Antisense gene therapy for pi-blockade holds promise for achieving all these
The idea of using antisense oligonucleotides (AS ODN) to inhibit gene
expression was first proposed and performed by Zamecnik and Stephenson in
1978 (118). Antisense oligonucleotides are short (typically 15-25 bases in length)
synthesized and chemically modified DNA molecules that are complementary to
the mRNA sequences encoding proteins of interest. They bind to mRNA
transcripts in a specific manner by Watson-Crick base pairing and inhibit the
expression of target proteins.
Several mechanisms of action have been proposed for AS ODN and
demonstrated to work in different systems [reviewed by Crooke (119)]. One
theory is translational arrest by assuming a steric hindrance to prevent ribosomal
assembly and read-through (120-122). A second mechanism works through
activation of RNase H that degrades the RNA strand of an RNA-DNA duplex
(123-125). Thus the mRNA is cleaved irreversibly and cannot be translated into
proteins. In addition, AS ODNs have been designed to inhibit preRNA slicing
(126) and to form triple helix with DNA (127), but these appear to be possible
only in special cases.
In principle, an oligo of 12-15 bases long can provide sufficient affinity and
specificity to its complementary mRNA (128). The specificity of AS ODN is so
great that even two genes that differ by a mutation at a single base can be
inhibited differentially (129). Therefore gene targeting specificity is the most
obvious virtue of antisense strategy, as it is based on individual gene sequences.
This also endows antisense with another advantage, which is versatility.
Antisense can be designed to specifically intervene with the expression of any
gene whose sequence is known. The design of AS ODN is rational and the
production is inexpensive. Besides, AS ODNs are hydrophilic molecules that
cannot cross the blood-brain barrier to cause CNS effects (130).
Specificity, simplicity and easy availability of synthesized oligos, along with
developments in the human genome project, have made antisense technology a
powerful tool for basic and clinical applications. It has been used increasingly to
study gene functions in recent years (131-135). For example, transgenic mice
expressing an antisense RNA against angiotensinogen (AGT) mRNA in the brain
provided convincing data on the role of central AGT in the regulation of BP and
fluid homeostasis (136). AS ODNs have also been explored as therapeutic
agents to suppress harmful components involved in disease development (137),
including cancer (138,139), AIDS (140), autoimmune disorders (141,142) and
restenosis (143-145). Its success is manifested by the recent approval of the first
antisense drug, Vitravene, by FDA to treat cytomegalovirus-induced retinitis.
Furthermore, antisense can be delivered to produce long-term gene inhibition
and therapeutic benefits for treating cardiovascular diseases. Studies in our
laboratories have established that systemic delivery of antisense cDNA directed
at AT1-receptor or angiotensin-converting enzyme by adeno-associated virus
(AAV) results in a long-term attenuation (6 months) of hypertension in
spontaneously hypertensive rats (146) as well as in double transgenic mice
expressing human angiotensinogen and renin (147). The hypertension-
associated ventricular hypertrophy and endothelial dysfunction are concomitantly
reduced. Raizada's group (148,149) used retrovirus to deliver ATI-receptor
antisense which prevented the development of hypertension in SHR, and the
normotensive phenotype was passed to the second generation. In addition,
clinical trials are underway to evaluate the efficacy of AS ODN for treating
cardiovascular problems such as restenosis (narrowing of blood vessels) after
coronary angioplasty. These results indicate that antisense gene therapy is
conceptually sound and technically feasible for the treatment of cardiovascular
diseases including hypertension. Therefore a specific p,-adrenergic inhibition
without CNS effects should be achieved by antisense molecules targeted to P,-
1.5 Gene Delivery Vehicles (Viral vs. Nonviral Vectors)
It is evident that once a therapeutic target is defined, the success of gene
therapy largely depends on the choice of the delivery system. An ideal vector
should possess the following properties: (i) high efficiency of gene transfer; (ii)
low toxicity, non-immunogenicity and non-mutagenesis; (iii) simplicity to produce
on a large scale; (iv) maintained efficacy on repeated administration. Currently,
numerous systems have been developed for gene therapy, which fall into two
categories: viral and nonviral vectors.
1.5.1 Viral Vectors
The four viruses that have been broadly used for gene therapy are
retrovirus, adenovirus, adeno-associated virus (AAV) and herpes simplex virus
(HSV). The characteristics of these viral vectors are summarized in Table 1-1.
Although viral vectors are highly efficient in gene delivery, they suffer
several significant drawbacks that make them currently inappropriate for clinical
use, such as high acute toxicity, immunogenicity, risk for mutagenesis and
insufficient pharmaceutical quantities (viral titers) (150-152). The safety issue
has remained a major concern for viral vectors (153), and the critique became
more intense after the tragic death of a young patient after receiving adenovirus-
delivered gene therapy in 1999 (154,155). This incident alerts scientists of the
immature status of current viral gene therapy and warrants more scrupulous and
extensive study and improvement for viral vectors.
Table 1-1. Characteristics of presently available viral vectors for gene therapy.
In vivo risks
1.5.2 Nonviral Vectors
The latest development in nonviral vectors includes gene gun (156,157),
cationic liposomes (158,159), polymer (160) and liposome/polymer complex
(161). Some drawbacks are associated with these vectors, i.e. lower efficiency
than viral vectors in gene transfer and transient gene expression. In spite of
these limitations, nonviral vectors are now being pursued vigorously as a
promising alternative for viral vector gene therapy, because of their low toxicity,
non-immunogenicity and feasibility to be produced in large quantities [reviewed
in (162,163)]. Liposomes, a traditional drug carrier used in pharmaceutics, have
established safety and pharmacological properties (164,165). They can also
reduce the toxicity of many compounds such as doxorubicin and anthracycline
(166,167). In addition, cationic liposomes are effective vehicles for gene delivery
(168) by facilitating cellular uptake of DNA, protecting DNA from degradation and
extending the circulation time (169). For example, cationic liposomes are used
routinely for in vitro transfection. Their therapeutic potentials as gene therapy
vectors have been demonstrated in clinical trials of a variety of diseases such as
cystic fibrosis (170) and cancer (171-173). Currently 75 clinical trials are using
liposomal vectors for gene delivery which involve approximately 21% of all
patients treated by gene therapy (174).
1.6 Specific Aims
3-blockers are used successfully to treat hypertension, heart failure and
myocardial infarction. However they have poor patient compliance due to the
mandatory daily regimen and side effects resulting from nonspecific 12-
antagonism and CNS reactions. Antisense gene therapy has the potential to
overcome these problems. The objective of this dissertation is to develop an
antisense oligonucleotide specifically against 1,-AR (P,-AS-ODN) and deliver it
systemically by nonviral vector cationic liposomes. Its antihypertensive effects
and underlying mechanisms will be studied in an animal model of human
essential hypertension. The potential benefits on inhibiting hypertension-
associated left ventricular hypertrophy and protecting the heart from myocardial
infarction will be explored. The safety profile of chronic treatment eith p,-AS-ODN
will be established.
AIM 1: To develop a P,-AS-ODN that can inhibit P,-AR expression specifically
and effectively without affecting other receptor subtypes.
AIM 2: To achieve efficient systemic delivery of P,-AS-ODN by cationic
liposomes. A commercially available liposome formula with established efficacy
and safety, DOTAP/DOPE, will be used to improve antisense delivery in vivo.
AIM 3: To test the antihypertensive effects of p,-AS-ODN in spontaneously
hypertensive rats (SHR). BP and HR will be measured by indirect and direct
methods, i.e. tailcuff and radiotelemetry.
AIM 4: To elucidate the mechanisms by which P,-AS-ODN reduces
hypertension. The effects of antisense on cardiac output, peripheral renin-
angiotensin system and CNS will be studied.
AIM 5: To investigate the effects of repeated administrations of P,-AS-ODN
on high BP and left ventricular hypertrophy of SHR. Three repeated
injections will be given to adult SHRs with established hypertension.
AIM 6: To study the safety profile of p,-AS-ODN after long-term treatment.
AIM 7: To explore the protective effects of p,-AS-ODN in an animal model of
myocardial infarction. Four days after antisense treatment, rat hearts will be
isolated and subjected to 30 min global ischemia followed by 30 min reperfusion.
2.1 Biochemical Methods
C6-2B cells and 1-15-5B cells were cultured in DMEM supplemented
with 5% fetal bovine serum (FBS), 100 U/ml penicillin, 0.1 mg/ml streptomycin
and 2.5 tg/ml amphotericin-B at 370C with 5% CO2 for 3-4 days until cells grew
to 80% confluency. The medium was changed to OPTI-MEM I (Life
Technologies, reduced serum medium). pl-AS-ODN or saline (as control) was
added to the medium.
Preparation of Liposomes and ODN/liposome Complex
The cationic lipid 1,2-bis(oleoyloxy)-3-(trimethylammonio)propane
(DOTAP) was mixed with the helper lipid L-a dioleoyl phosphatidylethanolamine
(DOPE, Avanti Polar Lipids) at 1:1 mole ratio. The mixture was evaporated to
dryness with a rotary evaporator at 30C for 10 min. The lipid film was further
dried by nitrogen for 10 min to evaporate residual chloroform and resuspended
in sterile water to a concentration of 5 mg DOTAP/ml. The resultant mixture was
shaken in a water bath at 30-40C for 30 min and stored at 4C overnight. The
suspension was then briefly sonicated, stored at 40C and used within 3 months
(175). The average diameter of DOTAP/DOPE was 170nm.
ODN/liposome complex was prepared on the day of use by mixing desired
amount of ODNs with DOTAP/DOPE to the final DNA concentration of 300.ig/ml
in 5% (w/v) dextrose in water. The mixture was incubated at room temperature
for 40-60 min to facilitate the maturation of the complex. Rats were anesthetized
by inhalation of 6% isofluorene in a drop jar and injected intravenously through
Plasma Renin Activity and Plasma Ang II Levels
Plasma renin activity (PRA) was determined using angiotensin I
radioimmunoassay kit (DuPont). Plasma Ang II levels were measured by radio-
immunoassay as previously described (176).
Determination of MDA Levels in the Myocardium
Malondialdehyde (MDA) levels in the myocardium were measured in
duplicate by a modified method of Ohkawa (177). Briefly, the ventricular tissues
were homogenized. The reaction mixture consisted of 0.1 ml tissue homogenate,
0.4 ml 0.9% NaCI, 0.5 mi 3% sodium dodecylsulfate and 3 ml thiobarbituric acid
reagent (containing equal amount of 0.8% aqueous thiobarbituric acid and acetic
acid) and was heated for 75 min at 95C. Thereafter the mixture was cooled by
addition of 1 ml cold 0.9% NaCI and extracted with 5 ml n-butanol. After
centrifugation at 3,000 rpm for 15 min, the butanol phase was assayed
spectrophotometrically at 532 nm. Tetramethoxypropane (in amounts of 0, 0.1,
0.2, 0.4, 0.8, 1.0 nmole) served as external standard. MDA levels were
expressed as .tmole/g tissue.
2.2 Molecular Biology Methods
AS-ODN were designed according to the basic principles (178),
synthesized and modified by backbone phosphorothioation (Gemini).
At different time points after the single injection of p1-AS-ODN or inverted-ODN
(as control), rats were sacrificed, renal cortex was dissected from left kidneys,
dipped into RNAlater tissue storage buffer (Ambion) immediately and stored at -
200C. Total RNA was extracted using RNAwiz reagent (Ambion) and quantified
by spectrophotometer. RNA samples from 4-5 rats of each time point were
pooled. 5ptg RNA was digested by DNase I, reverse transcribed by Superscript
reverse transcriptase (Life Technologies) at 420C for 50 min, and 1/20 of RT
product was used to run PCR for 20 cycles. PCR primers for renin were 5'-
AGGCAGTGACCCTCAACATTACCAG-3' (forward) and 5'-CCAGTATGCA-
CAGGTCATCGTTCCT-3' (reverse) and yielded products of 362 bp. Primers for
GAPDH were 5'-ATCAAATGGGGTGATGCTGGTGCTG-3' (forward) and 5'-
CAGGTTTCTCCAGGCGGCATGTCAG-3' (reverse) and yielded products of 505
RT-PCR products were subjected to Southern blotting, hybridized with
psoralen-biotin labeled cDNA probes and detected using nonisotopic kits
(Ambion). After exposing membranes to X-ray films, the intensity of renin
mRNAs was quantified by densitometry and normalized with GAPDH mRNA
levels. The experiments were repeated twice.
2.3 Pharmacological Methods
Membrane Preparation and p-AR Binding Assay
Membrane preparation and binding assay were carried out as previously
described (180). Briefly, cells were detached and scraped from the cell culture
plates. Rats were decapitated, and hearts and kidneys were removed. Atria,
valves and fat were dissected from the ventricles. Cells and tissues were
homogenized in ice-cold 50mM Tris-HCI buffer at pH 7.5. Then the suspensions
were centrifuged at 48,000xg for 10 min at 40C. The pellet was resuspended in
ice-cold 50mM Tris-HCI at pH 7.5 with 5mM MgCI2, passed through a mosquito
net to remove large pieces of connective tissues and centrifuged at 48,000xg for
10 min. The resulting pellet was washed once more and diluted in 50mM Tris-
HCI, 5mM MgC12, pH 7.5. Protein content was determined by Lowry's method.
For saturation binding experiments, 100.tg membrane protein was incubated with
six concentrations of 1251-(-)iodocyano-pindolol (I-CYP, NEN Life Science, 6.25-
100pM) in a total volume of 250pl containing 50mM Tris-HCI (pH7.4), 5mM
MgCI2 at 360C for 60 min. For the single point binding, assay was performed with
100pM 1251-CYP. The nonspecific and 32-AR binding levels were determined in
the presence of 1l.M ()-alprenolol and 150nM CGP20712A (RBI), respectively.
Then the reaction mixture was passed through Whatman GF/B glass fiber filter
using Brandel harvester and the bound radioactivity was counted for 1 min. All
binding assays were performed in triplicate, with the results varying by less than
Tissue Preparation and Quantitative Autoradiography
Rats were anesthetized and perfused with 0.9% saline before euthanasia.
Tissues were removed and frozen in dry ice. Coronal sections of brain, horizontal
sections of heart and sagittal sections of kidney (20tm) were cut on a cryostat
(Microm) at -20oC and mounted on microscope slides. Every seventh slide was
stained with hematoxylin and eosin for histology. Autoradiography was
performed according to the previously published protocol (181). Tissue sections
were preincubated in Krebs buffer (NaCI 118.4mM, KCI 4.7 mM, MgS04 1.2mM,
CaCl2 1.27mM, NaH2PO4 10.0mM, pH 7.1) containing 0.1mM guanosine
triphosphate (GTP), 0.1mM ascorbic acid and 10iM phenylmethylsulfonylfluoride
(PMSF) for 30 min at 250C. Sections were then incubated in Krebs buffer
containing 0.1mM ascorbic acid and 10iM PMSF with 100pM I-CYP at 25C for
150 min, in the presence of 1pLM (-)propranolol, 100nM ICI118,551 (p2-selective
antagonist) or 100nM CGP20712A (pi-selective antagonist) to distinguish non-
specific, pi- and P2-bindings. Labeled sections were rinsed in the same buffer,
followed by two 15-min washes at 370C in the buffer and rinsed in distilled water
at 250C. Dried sections were then exposed to X-ray films. The images were
quantitated with a computerized image analysis system (MCID, Imaging
Research) and normalized using 1251-standards. Non-specific binding was less
than 10% of total binding.
2.4 Physiological Methods
Blood samples were collected from the tail. Serum was obtained by
collecting blood with no anticoagulant, incubating on ice for 1 hr followed by
centrifugation at 4000rpm for 15 min. Plasma was obtained by collecting blood in
tubes containing EDTA as the anticoagulant and centrifuged at 4000rpm for 15
Male SHRs (250-350g, Harlan) and Sprague-Dawley rats (200-250g,
Harlan) were kept in cages in a room with a 12-hr light-dark cycle. Animals were
fed standard laboratory rat chow and tap water ad libitum. All care and surgical
conditions were approved by the UF Animal Care Committee.
Telemetric Sensor Implantation. Before implantation, the zero of each
radiotransmitter (TA11PA-C40, Data Sciences) was verified to be < 4 mmHg.
SHR were anesthetized with 100mg/kg ketamine and 15mg/kg xylazine and a
midline abdominal incision was made. A fluid-filled sensor catheter was then
inserted into the right femoral artery and the tip of the catheter was in the
abdominal aorta caudal to the renal arteries. The sensor battery was sewed to
the skin in the peritoneal cavity. The implanted rats were allowed to recover for 1
Jugular Vein Cannulation. One week after telemetric implantation, rats
were anesthetized and a curved catheter made of PE 50 and vinyl tubing was
inserted into the jugular vein. The tubing was led under the skin of the neck and
exposed on the back to allow for drug infusion. Rats were allowed to recover for
24 hrs before experimentation. The catheters were flushed with 100U heparin
everyday to prevent clogging.
Determination of Effects of PI-AS and Atenolol on Cardiovascular
Parameters in Response to p-stimulation
Langendorff Heart Perfusion. 48hrs after injection of 1mg/kg pi-AS-ODN
(n=9) or inverted-ODN (n=6), SHR were anesthetized and sacrificed. Heart
perfusion was performed according to the protocol described previously (182).
Hearts were quickly removed and perfused via the aorta with oxygenated Krebs
buffer (118mM NaCI, 18.75mM NaHCO3, 1.2mM KH2PO4, 4.7mM KCI, 1.2mM
MgSO4, 1.25mM CaCI2, 11.1mM Glucose, 0.01mM EDTA) at a constant flow of
7.0ml/min at 360C. Coronary perfusion pressure (CPP) was measured via a
catheter placed proximal to the aorta and connected to a pressure transducer
(Gould Statham P231D). A latex balloon filled with water and connected to the
pressure transducer was inserted into the left ventricle through the left atrium to
measure left ventricular end-diastolic pressure (LVEDP), left ventricular systolic
pressure (LVSP), and developed left ventricular pressure (dLVP) (dLVP=LVSP-
LVEDP). LVEDP during equilibration was set at 5 to 7mmHg. CPP, LVEDP and
LVSP were continuously recorded on a 4-channel recorder (Astro-Med). After
baselines for dLVP and heart rate (HR) were stable for 5 min, isoproterenol (ISO,
non-specific p-agonist) was given at 0.01, 0.025, 0.05, 0.12pM at 10 min
intervals so as to avoid the effect of tachyphylaxis.
Telemetric Monitoring of Live Animals. The effects of p3-AS-ODN and
atenolol on cardiac dP/dtmax, HR and systolic BP (SBP) were compared in the
same group of SHR (n=4). Two days after catheterization of jugular vein, control
values were taken and 1mg/kg p3-AS-ODN was injected. 48 hrs later, rats were
tested for the effect of p1-AS-ODN. The rats were allowed to recover until all the
cardiovascular parameters returned to control values. Then 1mg/kg atenolol (1i-
selective antagonist) was injected and rats were tested 30 min later. For 1i-
stimulation, SHR were infused with dobutamine (pi-selective agonist) through
jugular vein catheter at 5, 10, 20, 40pg/kg/min. Each dose was given for 5 min
continuously and at 1 hr intervals until all the cardiovascular parameters returned
to baseline so as to avoid the effect of tachyphylaxis. BP and HR were sampled
every 1 min. dP/dtmax was calculated from the slope of the rising pulse pressure
curve and determined every 1 min. The difference between values at each dose
and baseline was denoted as A.
BP and HR Monitoring
Telemetry. Each rat cage was placed on a receiver (RLA1020, Data
Sciences) for measurement of cardiovascular parameters. Data were collected
with a computer-based data acquisition program (Dataquest LabPRO3.0; Data
Sciences). BP and HR were measured every 10 min and averaged every 1-24
hrs. Before treatment, SHR were monitored for a week to get a stable baseline.
Tailcuff. Rats were warmed for 20-30 min in cages on heating pads. The
temperature was controlled at 35-370C. Then rats were placed in a plastic
restrainer kept at 370C. A pneumatic pulse sensor was attached to the tail. After
cuff inflation, systolic BP was determined as the first pulsatile oscillation on the
descending side of pressure curve. Heart rate (HR) was determined by manual
counting of pulse numbers per unit time. BP and HR were recorded by a Narco
physiograph. Data values of each rat were taken as an average of at least four
stable readings. Baseline was determined by averaging three days of
measurements before antisense administration.
SD rats were anesthetized with sodium pentobarbital (40 mg/kg)
intraperitoneally. The hearts were rapidly excised and placed in ice-cold Krebs-
Henseleit buffer (118 mM NaCI, 25 mM NaHCO3, 1.2 mM KH2PO4, 4.7 mM KCI,
1.2 mM MgSO4, 1.25 mM CaCI2, 11 mM Glucose, pH 7.4). Within 1 min, the
hearts were transferred to an Langendorff apparatus and perfused via the aorta
with oxygenated (95% 02 + 5% CO,) Krebs-Henseleit buffer at 370C using a
MasterFlex pump (model 7015-21, Cole-Palmer Instrument). The heart was
placed in a semiclosed circulating water-warmed (370C) air chamber, paced
atrially at a rate of 300 bpm with a Medtronic 530 pacemaker, and perfused at a
constant flow (5.5-6.0 ml/min). LVEDP during equilibration was set at 5-7 mmHg.
Sham controls were continuously perfused with Kreb-Henseleit buffer for 80 min.
Heart from other groups, were equilibrated for 20 min, subjected to 30 min global
ischemia followed by 30 min reperfusion. After completion of the experiment,
hearts were frozen in liquid nitrogen for p-AR radioligand binding assay and
measurement of MDA.
2.5 Clinical Pathology Methods
Liver Transaminase Levels
The plasma levels of alanine aminotransferase (ALT) and aspartate
aminotransferase (AST) were determined by a colormetric method using a kit
(Sigma, Cat. #505-OP). Briefly, plasma samples were incubated with proper
substrates for ALT or AST at 370C for 1 hr and subjected to color reactions. The
absorbency was read at the wavelength of 500 nm and the concentrations of
ALT and AST were calculated from the calibration curve obtained under the
Hematology measurements were carried out in the Clinical Laboratory of
Shands Hospital at University of Florida. The hematology parameters tested
included erythrocyte count (RBC), hematocrit, platelet count and total leukocyte
C-Reactive Protein (CRP)
CRP values were obtained on blood serum in the Clinical Laboratory of
Shands Hospital at University of Florida.
Rats were euthanized by a peritoneal injection of overdose pentobarbital.
Selected tissues, including heart, liver, spleen and kidney, were removed,
trimmed of fat and contiguous tissues and weighed. These tissues were fixed in
10% neutral buffered formalin, embedded in paraffin (Diagnostic Referral
Laboratory, University of Florida), sectioned at the thickness of 5.im, mounted on
gelatin-coated slides, stained with hematoxylin and eosin, and examined under
Gel Immunodiffusion Assay
Specific antibodies to ODN/liposome complex were measured with double
immunodiffusion (Ouchterlony) method on premade discs purchase from ICN
Biomedicals. Briefly, the ODN/liposome complex was loaded in the center well,
and serum samples in the surrounding wells. They were allowed to diffuse at
room temperature in a moisturized and sealed container for up to 7 days. The
occurrence of precipitation bands indicated the presence of antibody-antigen
2.6 Statistical Analysis
All experiments were repeated at least 2 times unless the situation and results
called for more experiments. Various statistical test were performed depending
on the experimental design using Jandel Sigmastat Software. Values were
expressed as mean SEM. Student t test was used to evaluate significant
differences when two groups were used. One way analysis of variance (ANOVA)
and appropriate multiple range test were used when more than two groups were
compared. Pearson product moment correlation was used to assess the
relationship between two parameters. In some studies, a two-way ANOVA were
utilized to evaluate the effects of individual treatments and any interactions
between the treatments. For analysis of parameters (e.g. dP/dt, BP, HR by
telemetry) obtained from the same animal over time, repeated measure of
ANOVA was incorporated. Significance for all tests was set at the 95%
confidence limit. All molecular biology experiments dealing with the comparison
of mRNA by RT-PCR were carried out in the linear range of the assay. The data
were normalized by GAPDH and subjected to appropriate statistics.
OPTIMIZATION OF ANTISENSE DESIGN AND DELIVERY
Antisense oligodeoxynucleotides (AS-ODN) are generally 15-20 bases in
length. Theoretically the length of 12-15 bases is able to provide sufficient
specificity and binding affinity for a target mRNA (128) while short enough to gain
entry into cells (183). The specificity of antisense molecules is so great that even
two genes that differ by a single base mutation can be differentially inhibited
(129). There is a wide range of potential sites within an mRNA sequence for
antisense targeting, the best of which is considered to be in the vicinity of AUG
translation initiation codon. The availability of these sites largely depends on the
folding of the mRNA (184). However currently there is no computer program
sophisticated enough to predict the secondary structure of the target mRNA.
Lacking a sure-fire method of oligo prediction, antisense development has to go
through a trial-and-error test.
The basic principles of antisense design are described below, which are
now published in Methods in Enzymology (178).
1. Identify the sequence: Check GenBank for the mRNA sequence of the target
protein in the species to be studied. If there is more than one lab cloning the
same protein, compare the homology of different reports. These sequences
will point out the controversial bases, which can be due to natural mutations
in variant strains of the same species or sequencing errors. Try to avoid
these debatable regions. Target the ODN to the identical regions, which
ensures reliability of the sequence being used.
2. Target the sites: Although there are several possible sites within the DNA
sequence to target for antisense design, the regions that are considered to be
the best targets for constructing effective AS-ODNs are the 5' cap region and
the area around the AUG translation initiation codon. In our experience, AUG
start codon and nearby bases downstream of AUG in the coding region are
most promising sites for antisense inhibition (185). Usually the length of 15-
20 mer is sufficient to guarantee sequence specificity, which must be
confirmed by checking with GenBank for existing sequences to avoid any
significant homology with other mRNAs.
3. Modify the backbone: AS-ODNs are chemically modified to increase nuclease
resistance and stability. The most commonly adopted modification is
backbone phosphorothioation, in which one non-bridging oxygen of the
phosphodiester linkage is replaced by sulfur (137). Compared to natural
DNA, phosphorothioated ODNs have a greatly extended half-life of 42-56 hrs
in the plasma of rat, monkey and human (186-188). They are also better
substrates to induce RNase H cleavage of mRNA. But phosphorothioation
significantly decreases the melting temperature (Tm) of ODNs and reduces
binding affinity and stability of DNA/DNA and DNA/RNA duplexes. For 15-25
mer, Tm of phophorothioate ODN is usually 7-120C lower than normal ODN,
with AT bases showing more Tm depression than GC (189,190). This fact
must be taken into account in the design of short AS-ODN, because 15-mer
phophorothioate ODN /RNA hybrids may have Tm close to body temperature
4. Avoid pitfalls: Some other basic rules applicable in the design of polymerase
chain reaction (PCR) primers also hold true for AS-ODN (191), e.g.
avoidance of palindromic structure and primer-dimer formation, and choosing
sequence with a balance AT/GC ratio to minimize toxicity and non-specific
binding caused by high GC content. Although a high GC/AT ratio should be
avoided generally, the danger of toxicity and non-specific binding can be
reduced by phosphorothioation.
3.2.1 Design of pi-AS-ODN
My plan was to develop a 15 mer Pi-AS-ODN specific to rat (192) and
human (11) pi-AR mRNA. Thirteen sequences were picked within the conserved
regions between rat and human pi-AR mRNA, which covered a variety of sites
including upstream of AUG, spanning AUG, CDS and 3'-UTR. Seven were
discarded after specificity checking by Blast search and secondary structure
analysis of the ODNs. The remaining six candidates are listed below. They are
modified by backbone phosphorothioation. Their positions refer to rat p1-AR
mRNA (gene accession number: J05561) (192).
#1: 5'- ccg cgc cca tgc cga -3' (64-78, spanning AUG start codon)
#4: 5'- ggt cgt tgt agc agc -3' (691-705, in the CDS)
#6: 5'- cgg tct tga gta aac -3' (1602-1616, in the 3'-UTR)
#7: 5'- ggc cga cga cag gtt -3' (111-125, in the CDS)
#8: 5'- atg agc agc acg atg -3' (269-283, in the CDS)
#10: 5'- ttg gtg agc gtc tgc -3' (335-349, in the CDS)
3.2.2 In vitro and in vivo Screening
The effects of pi-AS-ODN on the expression of P1-AR were examined in
C6-2B and i1-15-5B cell lines. C6-2B is a rat glial tumor cell line. P1-15-5B is a
mouse connective tissue cell line NCTC 2071 permanently transfected with
human p3-AR cDNA. Both cell lines express abundant p1-AR. Cells were treated
with 0.1-10 pM p -AS-ODN or saline (as control) for 3-48 hrs. P1-AR levels on the
cell membrane were measured with radioligand binding assay. Many
experiments were performed testing different doses and time course, but all
yielded negative results. None of the antisense candidates shown above
reduced PI-AR levels in either cell line. A representative experiment was
summarized in Table 3-1 and 3-2.
Reasons for the absence of antisense action were searched for. Although
efficient uptake of 15-18 mer S-ODN in cell cultures was shown in early studies
(183), the possibility still existed that AS-ODN did not enter cells. To answer this
question, cells were treated with fluorescin-labeled AS-ODN and the ODN
uptake was examined under confocal microscopy. 30 min after addition into the
medium, fluorescence was readily observed and widely distributed in the
cytoplasm of the majority of the cells. Another reason could be that cells were
not treated with an optimal dose of AS-ODN for enough time. Therefore,
experiments on dose response (0.1LiM, 1tiM, 2jLM, 5jiM, 10iM) and time course
(3hr, 6hr, 12hr, 24hr, 48hr) were performed and repeated, but still showing no
significant effects of AS-ODN on pj-AR expression. It was noticed that high
doses (5pM, 10p.M) of phophorothioate ODN seemed to be toxic because cell
viability and total protein content per plate were significantly reduced compared
to saline control in some experiments. More experiments were planned, including
liposomal delivery of AS-ODN and effects of AS-ODN on cellular responses to p
1-agonists. However, despite continuous efforts, pi-AS-ODN failed to reduce pi-
AR levels in vitro.
Table 3-1. Effects of p1-AS-ODN on C6-2B cells.
C6-2B Cell Cell count total protein /plate P-AR binding P-AR binding
/plate (mg) (fmol/mg) (% of Control)
Saline 5.4 x 106 0.71 57 100
p1 AS 1 4.4 x 106 0.56 68 119
P1 AS 7 4.8 x 106 0.80 48 84
p1 AS 8 5.2 x 106 0.54 53 93
P1 AS 10 4 x106 0.60 54 95
Cells were grown in DMEM / 5%FBS until 80% confluency and the
medium was changed to OPTI-MEM I (GIBCO BRL, reduced serum medium).
5pM p~-AS-ODN or saline were added to the medium 24hr before a single point
binding assay on membrane proteins at 100 pM 1251-CYP. Non-specific binding
was obtained in the presence of 1pM alprenolol.
Table 3-2. Effects of pf-AS-ODN on p1-15-5B Cells.
p1-15-5B Cell count total protein/plate P-AR binding P-AR binding
Cell /plate (mg) (fmol/mg) (% of Control)
Saline 5.2 x 106 0.53 156 100
p1 AS 1 5.2 x 106 0.50 175 113
p1 AS 7 5.0 x 106 0.49 151 97
31 AS 8 4.8 x 106 0.49 165 106
P1 AS 10 4.8 x 106 0.50 163 104
It was hypothesized that an effective p1-AS-ODN, by suppressing the pi-
AR expression and function in heart and kidney, could efficiently inhibit
hypertension. Hence, the effectiveness of pi-AS-ODN was evaluated based on
the extent of reduction of Pi-AR expression and high BP in rats. To facilitate the
antisense delivery, ODNs were completed by cationic liposomes DOTAP/DOPE
at a (+/-) charge ratio of 0.5.
Antisense efficacy was first tested in normotensive Sprague-Dawley (SD)
rats, which received a single intravenous injection of pi-AS-ODN through tongue
vein. The inhibition of p1-AR levels in the heart was assessed by saturation
receptor binding assay two days after treatment. All antisense candidates
significantly reduced the cardiac p-AR levels to different extents as compared
with saline or inverted-ODN (INV-ODN) control (Table 3-3). Based on the
receptor level, ODN-1 and ODN-7 were better than other candidates.
Table 3-3. Effects of pi-AS-ODNs on cardiac P-AR in rats.
Treatment R KD (pM) Bmax Decrease in
Saline (n=3) 0.9716 553 42.11.6
INV-ODN-1 (n=3) 0.9217 562 38.71.2
AS-ODN-1 (n=3) 0.8650 341 27.50.8* 31%
AS-ODN-7 (n=3) 0.9500 322 26.01.5* 35%
AS-ODN-8 (n=3) 0.9712 363 28.92.1* 27%
Male SD rats (250-300g) were treated with a single i.v. injection of 200 pg
P1-AS-ODNs or INV-ODN for 48 hrs. Membrane proteins were prepared from
heart ventricles. Affinity and density of p-AR were measured by radioligand
binding assay. P<0.01 versus saline or INV-ODN.
The antihypertensive effects of 31-AS-ODNs were then tested in
spontaneously hypertensive rats (SHR) monitored by radiotelemetry. A single
injection of ODN-1 significantly diminished mean arterial pressure by up to 14
mmHg for 8 days (P<0.05), whereas the drop in BP produced by ODN-7 was
less marked (8 mmHg) and for a shorter duration (4 days). The hypotensive
impact of pi-AS-ODN was dose-dependent, with the order 800pg > 400pg > 200
Based on these results, ODN-1 was determined to be the optimal
antisense sequence for pi-AR which reduced Pi-AR expression and high BP
3.2.3 Optimization of Antisense Delivery by Cationic Liposomes
Although chemical modifications like phosphorothioation endow AS-ODN
with nuclease insensitivity and longer half-life, the stability and delivery of AS-
ODN must be further improved to be effective in vivo. Cationic liposomes are
known for their ability to facilitate cellular uptake of nucleic acids and protect
them from degradation in circulation and tissues (165,169). Numerous lipid
formulations have been developed to enhance the transfection of plasmid DNA
and oligonucleotides (168,193). Dioleoyl trimethyl ammoium propane (DOTAP)
was introduced in 1980s and remains one of today's favorite lipids for its
effectiveness and biodegradability (163,194,195). Not only has its usefulness
been proven in bench studies but also explored in clinical trials, e.g. to deliver
CFTR gene for cystic fibrosis. Therefore we chose to use DOTAP coupled with
the neutral lipid dioleoyl phosphatidylethanolamine (DOPE) to deliver p3-AS-ODN
in our study.
However, high transfection efficiency that was seen in vitro could not be
readily achieved in vivo with the same liposome/DNA preparation. This is in part
attributed to the serum resistance of liposome/DNA complex because negatively
charged plasma proteins can bind to cationic lipids and destabilize
liposome/DNA complex (196). It was recently shown that increasing the +/-
charge ratio of liposomes to DNA and inducing the maturation of liposome-DNA
complex by prolonging incubation time could overcome this problem
(169,197,198). The optimal charge ratio of DOTAP:DNA was found to be ;2.0
(198,199). Thus I chose to test five charge ratios ranging from 0 to 3.5 to
optimize the delivery of p3-AS-ODN. In addition, DOTAP/DOPE and ODN were
carefully mixed and incubated at room temperature for 40-60 min before use to
facilitate stabilization of the complex.
Figure 3-1 shows the effect of different (+/-) liposome:ODN charge ratios
on systolic BP of SHR measured by tail cuff (n=6 for each ratio) in a
representative experiment. 0.5 mg/kg pi-AS-ODN alone, i.e. at ratio 0, did not
change SBP, whereas ratio 0.5 significantly reduced SBP by up to 38+5 mmHg
for 7-8 days. When the ratio was increased, the duration of the hypotensive
impact was drastically prolonged to 18-20 days at ratio 1.5, 20-33 days at ratio
2.5 and 3.5, varying with liposome preparations. However the maximum drop in
SBP was greater at ratio 1.5 and 2.5 (-35 mmHg) than at ratio 3.5 (-25 mmHg)
(Table 3-4). Accordingly, the optimal charge ratio of DOTAP:ODN was
determined as 2.5 and employed in the subsequent experiments. It was worth
noting that the antihypertensive effect of pi-AS-ODN varied slightly with different
batches of liposome mixture, probably due to the variation in particle size and
structure. Furthermore, aggregation and precipitation of liposome/ODN must be
avoided to ensure the successful delivery.
Table 3-4. Amplitude and duration of reduction in blood pressure of SHR after a
single i.v. injection of 0.5 mg/kg pi-AS-ODN delivered at different charge ratios of
Charge ratio Maximum reduction in Range of reduction Duration
of liposome/ODN blood pressure (mmHg) (mmHg) (day)
0 2 --
0.5 35-38 24-38 7-8
1.5 28-35 18-35 18-20
2.5 30-34 20-34 20-33
3.5 20-24 15-24 20-33
The purpose of this study is to develop an effective p1-AS-ODN targeted
to rat and human pi-AR mRNA that can reduce pi-AR expression and BP in
hypertensive models. After in vitro and in vivo screening, the best 31-AS-ODN
candidate was selected which was 15 mer in length and spanned the AUG site. It
significantly reduced the p3-AR levels in the heart and lowered high BP in SHR.
The delivery of AS-ODN was improved by cationic liposomes DOTAP/DOPE and
the condition was optimized.
These results support the notion that the region around the translation
initiation codon may be the best target for antisense design. ODN binding to this
area may interfere with the ribosomal assembly and/or form a structural hurdle to
ribosome sliding along the mRNA strand. Assuming that all the six p1-AS-ODN
candidates can bind to pi-AR mRNA, it is conceivable that interrupting the initial
phase of translation may be more effective than later blockade. However, this
assertion is more or less arbitrary, because I do not have evidence to show that
all candidates have assess to 11-AR mRNA at the same affinity. In spite of this,
since it is unlikely to screen every possible region in an mRNA for economical
reasons, I believe the first site to consider for AS targeting is the vicinity of AUG
codon if other criteria can be met.
Efficient gene delivery is vital to the therapeutic application of AS-ODN in
vivo. Among nonviral vectors, cationic liposomes are most widely used. They are
safe, nonimmunogenic and easy to produce on a large scale. However, relatively
low transfection efficiency has been obtained after intravenous administration,
mainly due to the inactivation of cationic liposome by serum. It was shown
recently that increasing the charge ratio (+/-) of liposome to DNA and inducing
the maturation of liposome-DNA complex can overcome this problem (197,198).
The optimal charge ratio of DOTAP:DNA was demonstrated to be =2 (198,199).
Based on these studies, the delivery of AS-ODN was optimized by testing five
charge ratios from 0-3.5. As shown in the results, increasing the charge ratio not
only improved the delivery efficiency but prolonged the duration of p3-AS-ODN
action. The best antihypertensive result (35 mmHg for 20-33 days) was achieved
at ratio 2.5 consistently.
The prolonged effect of ,3-AS-ODN on BP may be attributed to several
reasons. The half-life of phosphorothioate ODN is ~50hrs in rat plasma after
intravenous injection. Li et al. (183) showed that S-ODN could be found in its
intact form for at least 5 days after entry into cells. It has been shown that
liposome encapsulation not only improves the cellular uptake of DNA, but also
protects DNA from degradation and extends its circulation time in vivo (165,169).
In addition, Baker's group (180,200) and other researchers (201) have shown
that p-ARs had a slow turnover rate which took 200-250 hrs to recover in the
heart after irreversible blockade.
Then a question arose, i.e. why pi-AS-ODN reduced P3-AR density on the
cell membrane so rapidly (within 2 days), provided that p-ARs had a slow
turnover rate. It is possible that p3-AS-ODN not only inhibited the translation of p
1-AR, but also decreased the intracellular pool of pi-AR and accelerated the
internalization of the membrane receptors. The recycling of receptors between
cell membrane and cytoplasm is a much faster process than receptor
metabolism. This may explain the discrepancy between the rapid reduction in pi-
AR density and the slow turnover. However to test this hypothesis requires
It is still not understood why p1-AS-ODN did not decrease p1-AR levels in
vitro. Similar protocols had been successfully used in our lab to screen for AS-
ODN against AT1-receptor in cell culture. The possibility that AS-ODN failed to
enter cells was ruled out by the visualization of fluorescin-labeled AS-ODN in the
cytoplasm 30 min after addition into the medium. Further experiments on dose
response and time course did not improve the effects. We stopped in vitro study
after 6 months of pursuing. However, it is possible that AS-ODN delivery in cell
culture could have been improved by liposomes, as the case in vivo. It is also
likely that AS-ODN, after the endocytosis, did not escape endosomes to interact
with target mRNA.
7" / p 7.
i^. ^. /**- d^
\ A IA' A J
-- ratio 0
--W ratio 0.5
--0- ratio 1.5
-A ratio 2.5
--0 ratio 3.5
I I I I I I
0 10 20 30 40 50
Days after injection
Figure 3-1. Improving the antihypertensive effect of pi-AS-ODN by
optimization of liposome:ODN charge ratios.
SHR received a single iv injection of pi-AS-ODN or INV-ODN. 1mg/kg 31-AS-
ODN was delivered by DOTAP/DOPE at charge ratios from 0 to 3.5. 1mg/kg
INV-ODN delivered by DOTAP/DOPE at charge ratio 2.0 served as control (0).
Data represent mean values of each group (n=6). Standard errors were omitted
EFFECTS ON BLOOD PRESSURE AND HEART RATE:
l1-AS-ODN vs. ATENOLOL
Pi-AR plays a crucial role in the regulation and maintenance of
hypertension. p1-adrenergic activation increases cardiac contractility and heart
rate, and stimulates renin production from the kidney. Thus the inhibition of this
receptor can lead to decreased cardiac output and renin-angiotensin activity and
thereby reduce hypertension. This concept has been proven correct by the
antihypertensive action of p-blockers in clinical practice. In developing an
antisense agent for pi-inhibition to reduce hypertension, the rationale borrows
from the success of p-blockers because both approaches can suppress Pi-AR
and therefore p1-AS should be able to lower high BP if p-blockers do. However
these two strategies adopt different pathways for pi-inhibition. Antisense reduces
the total numbers of Pi-AR by interfering with translation, while p-blockers
dampen the function of p-receptors by competing with endogenous p-agonists
like norepinephrine. Due to the intrinsic drawback of pharmacological
intervention, pi-specific antagonists have not been developed which inevitably
results in undesirable antagonism of p2-receptor subtype. This accounts for a
large portion of side effects caused by p-blockers including contraindication of
asthma, peripheral vascular disease and insulin-dependent diabetes mellitus,
and reduced exercise capacity. In contrast, antisense can be designed specific
to its target based on gene sequence. This will preclude adverse effects of
nonspecific antagonism by p-blockers. Furthermore, CNS reaction can be
avoided by antisense owing to the inability of antisense molecules to enter the
brain. Here the effects of pi-AS-ODN on BP and HR of spontaneously
hypertensive rats (SHR) were tested and compared with those of a hydrophilic p
i-selective blocker, atenolol.
4.2.1 Effects of 31-AS-ODN on BP and HR Measured by Tail cuff
Figure 4-1 shows a typical experiment in which SHR was treated with a
single i.v. injection of 1mg/kg pi-AS-ODN delivered by the optimal ratio of
cationic liposomes determined in Chapter 3. BP and HR were measured by the
traditional tail cuff method. Antisense significantly decreased systolic BP for 4
weeks. The maximum drop was 30 mmHg. However, HR was not changed by 1i-
AS-ODN. Inverted-ODN (INV-ODN) as control had no effect on BP or HR.
4.2.2 Effects of 31-AS-ODN on BP and HR Measured by Radiotelemetry
In order to compare the effects of pi-AS-ODN and atenolol, radiotelemetry
was used to monitor BP and HR on a regular basis. Radiotelemetry allows for
accurate and continuous hemodynamic monitoring without inducing stress or
artificial factors. 1mg/kg pi-AS-ODN delivered with liposomes at the (+/-) ratio
0.5 produced a maximum drop of 15mmHg in mean blood pressure (Fig. 4-2).
The antihypertensive effect lasted for 8 days, which was consistent with the
results measured with tailcuff. HR was not significantly altered. In contrast to
AS-ODN, although the onset of hypotensive effect caused by 1mg/kg atenolol
occurred as early as 20 min after injection, it lasted for only 10 hrs (Fig. 4-3). In
addition, atenolol caused considerable bradycardia up to an average of -75 bpm.
All these hemodynamic changes with i.v. atenolol were not detectable within a
The results indicate that the antihypertensive effect of p3-AS-ODN is
significantly longer lasting than that of atenolol. A single iv injection can reduce
BP for 20-30 days with the maximum drop of 30 mmHg. In contrast, the
hemodynamic response to bolus iv injection of atenolol lasts only for hours. As a
result, atenolol must be taken every day to achieve chronic control of BP.
Moreover, atenolol causes the peak-and-trough fluctuation in BP, in contrast to
the steady hypotensive action of antisense. Greater than normal variation in 24-
hr BP is potentially dangerous and associated with early morning cardiovascular
catastrophe. Finally, 31-AS-ODN does not decrease HR as shown by both tail
cuff and telemetry, whereas atenolol induces appreciable bradycardia.
Bradycardia is a common complaint by patients taking p-blockers.
The most surprising finding of this study is the lack of effects on HR of 1i-
AS-ODN. The mechanisms are still under investigation. Several possibilities can
be considered. First, 31-AS-ODN did not affect P2-AR, which played an important
role in the regulation of HR (202,203), while not involved in the cardiac
contraction of rats. Second, antisense inhibition of 31-AR expression is gradual
and to a lesser extent than p-blockers. It is also possible that 11-AR may have a
larger reserve for controlling HR than contractility. However, to answer this
question requires more extensive studies and electrophysiology techniques
which are beyond the scope of this dissertation.
Both tailcuff and telemetry measures of BP showed a significant drop after
antisense treatment. SHR responded to p3-AS-ODN to a greater degree of
hypotension when subjected to tailcuff versus telemetry. In addition, the baseline
measured with tailcuff was consistently higher than that with telemetry by 20-
30mmHg in the same rats. Bazil et al. (204) reported a similar phenomenon.
They compared the cardiovascular parameters recorded by telemetry, tailcuff
and arterial catheter and observed a more sensitive hypotensive effect of
captopril with tailcuff. Tailcuff measurement of BP involves warming and restraint
of rats. It is conceivable that p1-AS, through the suppression of sympathetic
activity, can decrease BP of animals under stress more effectively.
ST --- AS-ODN
170 -o- Saline
0 5 10 15 20 25 30
0 5 10 15 20 25 30 35
Days after injection
Figure 4-1. Effects of P,-AS-ODN on BP and HR of SHR.
16-week old SHR were treated with a single iv injection of 1mg/kg P,-AS-ODN
delivered by DOTAP/DOPE at a (+/-) charge ratio of 2.5. BP (A) and HR (B) were
measured by tail cuff method. Data represent mean SEM (n=6-9 for each
group). *** P<0.001, ** P<0.01, P=0.059 vs. saline or INV-ODN.
Time after treatment (day)
Figure 4-2. Effects of pI-AS-ODN on mean BP
monitored by telemetry.
and HR of SHR
1mg/kg INV-ODN (0) or P,-AS-ODN (0) was injected with liposomes at a (+/-)
charge ratio 0.5. BP (A) and HR (B) were recorded every 10 min and averaged
every 24 hrs. Data represent mean SEM of each group (n=4). P<0.05 vs.
I 1 I j / /
I I / I 1 :
I I *
0 5 10 15 20
0 5 10 15 20
0 5 10 15 20
Time after treatment (hr)
Figure 4-3. Effects of atenolol on mean BP and HR of SHR monitored by
Saline (0) or 1mg/kg atenolol (0) was injected intravenously. BP (A) and HR (B)
were taken every 10 min and averaged every 1 hr. Data represent mean SEM
of each group (n=4). P<0.05 vs. saline.
MECHANISMS FOR THE ANTIHYPERTENSIVE EFFECTS OF Pi-AS-ODN
In the preceding studies, a pl-AS-ODN was developed which lowered BP
of SHR effectively and persistently with a single i.v. injection. The current study is
carried out to understand the mechanisms underlying its antihypertensive
effects. Pf-AR activation induces positive inotropic and chronotropic responses in
the heart, and stimulates renin production and release in the kidney. p-blockers
decrease cardiac output and plasma renin activity, which contribute to the BP
reduction. Therefore I hypothesize that:
i) p1-AS-ODN, through specific inhibition of p1-AR expression, will decrease the
functional sensitivity of pi-AR-mediated responses in the face of sympathetic
activation, and thereby achieve an antihypertensive effect.
ii) Since the pi-AS-ODN is designed specifically to p1-AR, it should not affect
the expression of 12-AR subtype.
iii) p1-AS-ODN cannot cross blood-brain barrier to affect p-AR in the CNS.
5.2.1 Inhibition of Cardiac Contractility
Effects on cardiac p-AR levels
Fig. 5-1 shows that a single i.v. injection of 1mg/kg pi-AS-ODN delivered
with cationic liposomes at a (+/-) charge ratio of 2.5 significantly reduced the p1-
AR density in the SHR hearts for 18 days (P<0.01). The drop was at its
maximum of 47% on day 4, 33% on day 10 and maintained at 29% on day 18. In
contrast, there was no significant change in the Bmax of p2-AR. KD of both
subtypes remained unaltered (Table 5-1). Consequently, the 11/P2 subtype ratio
in the ventricles was diminished from -70/30 to -50/50 by pi-AS-ODN. INV-ODN
had no effect on either subtype.
Table 5-1. Effects of pi-AS-ODN on Bmax and KD of p,- and p2-AR in cardiac
ventricles of SHR.
KD (pM) Bmax (fmol/mg)
KD (pM) Bmax (fmol/mg)
KD (pM) Bmax (fmol/mg)
p,- and p2-AR levels were measured 4 days after
saline, INV-ODN or p,-AS-ODN. Data represent mean
(n=6-10). P<0.01 versus saline control.
a single i.v. injection of
+ SEM of each group
Effects on cardiac contractility and heart rate in response to p-stimulation
48 hrs after injection of 1mg/kg p3-AS-ODN, the cardiac inotropic and
chronotropic responses to p-stimulation were determined in SHR in vitro and in
First, isolated hearts were perfused with incrementing doses of
isoproterenol (ISO, nonselective p-agonist), which enhanced HR and contractility
via activating p-AR. The dLVP-ISO dose-response curve, which reflected the
positive inotropic effect of ISO, was significantly shifted down by p1-AS-ODN
(P<0.05). HR was not significantly decreased, except at one point, i.e. 0.01 PM
ISO (P<0.05) (Fig. 5-2).
In vivo test was performed in conscious SHR monitored by radiotelemetry.
11-AS-ODN significantly (P<0.02) dampened the increase in dP/dtmax (index for
cardiac contractility) in the face of dobutamine (p1-selective agonist) (Fig. 5-3A).
The change in HR was not significantly reduced (Fig. 5-3B), which echoed the
results in isolated perfused hearts. On the other hand, the response of BP to
dobutamine was biphasic (Fig. 5-3C). Dobutmaine elevated systolic BP by 5-
8mmHg at low infusion speed and reduced systolic BP at higher speed. This is
probably due to the partial 32-agonistic activity of dobutamine at high doses and
the consequent vasodilatory effect on BP. Fig. 5-3 also compares the results with
11-AS-ODN and atenolol (1p-selective antagonist). Relative to p3-AS-ODN,
1mg/kg atenolol produced a more profound decline in AdP/dtmax and AHR during
5-hr period of time after injection. Moreover, it reduced basal dP/dtmax (2450295
mmHg/sec versus control 2937277 mmHg/sec) and caused bradycardia (295
12 bpm versus control 365+8 bpm) (P<0.05), while p3-AS did not change basal
contractility (2922 249 mmHg/sec) or HR (36512 bpm). However the effects
of atenolol on AdP/dtmax and AHR were transient. Within 24hrs after atenolol
administration, the inotropic and chronotropic effects of dobutamine had returned
to control level.
5.2.2 Suppression of Peripheral Renin-Angiotensin System
Effects on renal p-AR levels
Scatchard analysis of p-AR binding in renal cortex (Fig. 5-4) indicated that
P1-AR was the major subtype in the control rats, composing 70% of total p-AR.
After p1-AS-ODN injection, the Bmax of pi-AR was significantly diminished by 35%
on day 4 (P<0.05), 29% on day 10 (P<0.05), 23% on day 18 and completely
resumed on day 40. p1-AR reduction in kidney coincided with that in heart, which
were accompanied by the significant drop in BP (P<0.01). In contrast, the p2-AR
level was not affected, nor the affinity of either subtype (data not shown). INV-
ODN had no effect on either subtype.
Kidney slices from the same rats were subject to quantitative
autoradiography in order to display the structural distribution of p-AR (Fig. 5-5). p
i-subtype composed -60% of p-AR levels, which was predominantly localized in
renal cortex and outer band of medulla. p2-subtype was more diffusely
distributed in the kidney at a lower level. This result was consistent with previous
reports(205). Four days after pi-AS-ODN treatment, the overall density of pi-
subtype in kidney was significantly reduced from 23.5+2.1 to 15.4+3.3 fmol/mg
(P<0.05). The diminution in renal cortex was particularly conspicuous due to the
higher basal level. As expected, the distribution and concentration of p2-subtype
remained unchanged in accord with binding results. This further confirms the
specificity of the inhibitory effect of 31-AS-ODN on p3-subtype.
Effects on the peripheral renin-angiotensin system
RT-PCR (Fig. 5-6) revealed that the renin mRNA level in renal cortex was
transiently decreased to 62% of control 4 days after p3-AS-ODN injection. It was
completely reversed by day 18 (Fig. 5-7). On the other hand, PRA and plasma
Ang II levels showed different patterns of reduction, which were significantly
decreased on day 10 and day 18 (P<0.01), but not on day 4. Thus PRA and Ang
II seemed to have a delayed action relative to the reduction in renin mRNA (Fig.
5.2.3 Absence of CNS Effects
Quantitative autoradiography of 6-18 tissue slices in brain, heart and
kidney was analyzed four days after intravenous pi-AS-ODN administration (Fig.
5-8). No changes in the distribution of p-AR in forebrain and brainstem regions
were detected. This indicated the absence of antisense effect on the p-AR
expression in CNS. However, PI-AS-ODN significantly (P<0.05) reduced P1-AR
density in cardiac ventricles (from 30.22.1 to 20.62.5 fmol/mg) and renal
cortex (from 26.43.1 to 17.43.3 fmol/mg). This was consistent with the binding
results. 32-AR was not affected in any tissues.
The data presented above demonstrate that pI-AS-ODN effectively
decreases p3-AR density in heart and kidney, which led to a significant reduction
in cardiac contractility and plasma renin activity. These findings are consistent
with the hypothesis that P3-AS, through the inhibition of p3-AR expression, is able
to render heart, kidney and other tissues less sensitive to sympathetic activation,
which is a major contributing factor in high BP.
A single injection of pi-AS-ODN effectively decreases the cardiac 31-AR
density, which produces a rapid reduction in ventricular contractility and thereby
in cardiac output in response to p-stimulation. Unlike p-blockers, 31-AS-ODN
does not decrease HR. Hence bradycardia is unlikely to be involved in the
antihypertensive mechanisms of pi-AS-ODN.
The inhibition of renin release is regarded as a primary mechanism
second to a decrease in cardiac output for the antihypertensive effects of p-
blockers. Many p-blockers can reduce PRA in patients and experimental animals
(55,206). They are found to be more effective in patients with higher renin
profiles (56,57). In my experiments, 3p-AS-ODN effectively decreased PRA and
Ang II chronically. But the decrease in PRA and Ang II did not occur until -10
days after p1-AS-ODN injection (Fig. 5-7), in contrast to the rapid drop in cardiac
output two days after injection (Fig. 5-2 & 5-3). Thus it appears that the effects of
Pi-AS-ODN on the kidney renin and the circulating renin-angiotensin system are
more delayed than cardiac action. Therefore suppression of cardiac output may
account for the early phase of the antihypertensive effect of 3p-AS-ODN while
the inhibition of renin-angiotensin activity acts as the secondary mechanism
underlying the sustained reduction of BP in SHR.
Plasma renin activity (PRA) is regulated in several steps: synthesis of
renin in kidney, conversion of prorenin into active renin, and secretion of renin
into the circulation. By decreasing p3-AR levels in kidney, p3-AS-ODN
significantly reduced PRA and the subsequent plasma Ang II levels. This is
unlikely through the inhibition of renin expression, because there is no chronic
diminution of renin mRNA levels. Instead, pi-AS-ODN may exert its inhibitory
impact on renin secretion or the conversion of inactive renin to active renin. This
hypothesis is consistent with the observation that p-adrenergic stimulation of
renin expression had a different time course than that of renin secretion
(207,208). Further, p-blockers have been shown to reduce prorenin processing
to active renin without changing total renin (prorenin + PRA) levels in plasma
(55). A flaw in this study is the omission to measure the local renin concentration
in kidney cortex, where the circulating renin comes from. This assay would have
provided an insight in the differential effects of p1-AS-ODN on renin production
In addition to pi-adrenergic activation, renin expression is also regulated
by the negative feedback of ANG II. Therefore the reversal of renin mRNA levels
on day 10 and day 14 after p1-AS-ODN treatment may be explained by the
decreased plasma ANG II concentrations that exerted less inhibition on renin
Antisense is known for the gene targeting specificity. Radioligand binding
assay showed that p1-AS-ODN is specific in inhibiting 31-AR without any effect
on 32-AR. Therefore pi-AS-ODN may be free of the side effects that are caused
by non-specific p2-antagonism of p-blockers.
The blood-brain barrier (BBB) formed by capillary endothelia is only
permeable to small lipophilic molecules with molecular weight <600 dalton
[reviewed by Pardridge (209)]. Owing to their high hydrophilicity, ODNs undergo
negligible transport through BBB and have very limited access to CNS (130).
Coupling with liposomes increases the lipophilicity of DNA, but the large size of
DNA/liposome complex (100-300nm) makes it unable to cross BBB. The cellular
and organ distributions of DNA/liposome complexes with fluorescent labeling
were previously studied in mice after intravenous injection and the results
indicated that the complexes were primarily taken up by capillary endothelial
cells in most of the peripheral organs including lung, heart, kidney and spleen,
but absent in the brain (210). Although the brain retention of liposomes after
peripheral administration was observed in some cases, it is due to entrapment
within the brain microvasculature (211). In our study, autoradiography in brain
revealed no detectable changes in the expression and distribution of p-AR after
intravenous 31-AS-ODN injection. This provided further evidence that our
antisense approach did not have CNS effects. It also suggested that the
inhibition of central p-adrenergic system is unlikely to contribute to the
antihypertensive action of p1-AS.
Time after treatment (day)
Figure 5-1. Pi-AS-ODN decreased the density of PI-AR but not p2-AR in
cardiac ventricles of SHR.
A). Time course of P,-AS-ODN effects on Bmax of pi-AR (0) and P2-AR (0).
B). Bmx of p-ARs 4 days after iv injection of saline (solid bar), 1mg/kg INV-ODN
(open bar) or 1mg/kg p3-AS-ODN (shaded bar). Data represent mean SEM of
each point (n=6). *P<0.01, **P<0.001 vs. saline or INV-ODN.
0.00 0.02 0.04 0.06 0.08 0.10 0.12
E 260 o INV-ODN
0.00 0.02 0.04 0.06 0.08 0.10 0.12
Figure 5-2. P,-AS-ODN reduced ISO-stimulated positive inotropic more than
chronotropic response in isolated perfused SHR hearts.
48 hrs after iv injection of 1mg/kg INV-ODN (0) or Pi-AS-ODN (0), hearts were
perfused with Krebs buffer containing increasing concentrations of ISO. Shown is
the effect of P,-AS-ODN on ISO-induced elevation in developed left ventricular
pressure (A) and heart rate (B). The data represent mean SEM of each group
(n=6-9). P<0.05, ** P<0.01 vs. INV-ODN.
010 20 30 40
0 10 20 30 40
Figure 5-3. In vivo effects of p1-AS-ODN and atenolol on cardiovascular
hemodynamics of SHR in response to pi-stimulation.
The same group of SHR (n=4) was tested for dobutamine-induced hemodynamic
alteration at control levels (0), 48 hrs after p1-AS-ODN injection (0), and 1 hr
after atenolol injection (V). The rats were allowed to fully recover between
treatments. During dobutamine infusion, cardiac dP/dtmax (A), heart rate (B) and
systolic BP (C) of SHR were monitored by telemetry system. Data represent
mean SEM. P<0.05, ** P<0.01 versus control.
Days after injection
Figure 5-4. Effects of 11-AS-ODN on p-AR levels in renal cortex.
SHR were injected with 0.5 mg/kg pi-AS-ODN or INV-ODN with liposomes at
charge ratio 2.0. A) Time course of the changes in Bmax of P1-AR (*) and 32-AR (
0). B) Bmax of p-AR 4 days after iv injection of saline (solid bar), INV-ODN (open
bar) or 11-AS-ODN (shaded bar). Data represent mean SEM of each point
(n=6). P<0.05, ** P<0.01 vs. saline or INV-ODN.
.., I ~t' .
Figure 5-5. Representative autoradiography of p-AR levels in the kidney of
control (n=6) or pi-AS-ODN (n=6) treated SHR 4 days after
1 2 3 4 5 6
wwq' ms w -- a
Figure 5-6. RT-PCR and DNA blot of renin and GAPDH mRNA in renal
cortex of SHR.
Lane 1, saline control; 2, INV-ODN; 3, pf-AS-ODN day 4; 4, p3-AS-ODN day 10;
5, pi-AS-ODN day 18; 6, p1-AS-ODN day 40. Every point represents a pool of
total RNA extracted from 4-5 rats.
r 6 1
= 50 \
E 30 A'
0 10 20 30 40
Days after injection
Figure 5-7. pf-AS-ODN at a single injection exerts a delayed suppression on
A, Effect on renin mRNA levels in renal cortex. B, Effect on plasma renin activity.
C, Effect on plasma angiotensin II levels. Data represent mean + SEM of each
time point (n=6). P<0.01, ** P<0.001 vs. baseline.
Figure 5-8. Autoradiographic illustration of pf-AR levels in different tissues
A, forebrain; B, brainstem; C, heart; D, kidney. Autoradiography was performed 4
days after iv injection of INV-ODN (n=6) or Pf-AS-ODN (n=6).
REPEATED ADMINISTRATION OF p3-AS-ODN AND SAFETY PROFILE
To be therapeutically useful, the pl-AS-ODN has to be effective with
repeated administration and non-toxic. Currently there are at least ten clinical
trials testing AS-ODNs in a variety of diseases [reviewed by Agrawal (137)].
These clinical data have convincingly demonstrated the sustained efficacy with
repeated iv injection of AS-ODNs. Nucleic acids are poor epitopes. Human
bodies usually do not develop immune response to nucleic acids except in rare
autoimmune diseases. This may explain why AS-ODNs maintain effectiveness
during chronic treatment.
Hypertension is associated with pathophysiological changes such as left
ventricular hypertrophy (LVH) (212,213). Long-term antihypertensive therapies
including angiotensin-converting enzyme inhibitors and P-blockers can prevent
and reverse LVH (214). Therefore, if repeated treatment with p1-AS-ODN is
effective in sustained reduction in high BP, it may as well inhibit the development
Safety and pharmacokinetic profiles of several AS-ODNs have been
reported. In these studies, phosphorothioate-ODN without any drug carriers were
given through continuous iv infusion over 3-11 weeks and a wide range of doses
were tested for toxicity. Phosphorothioate-ODNs are well tolerated in humans at
the doses < 2.5 mg/kg/day without any pathological or functional changes
(215,216). The commonly seen side effects at higher doses are transient
thrombocytopenia and fatigue (216,217). Likewise, studies in mice and monkeys
revealed no adverse effects at doses < or = 10 mg/kg/day. Toxicity with large
doses (>10 mg/kg/day) include splenomegaly, lymphoid hyperplasia and
mononuclear cell infiltrates in mice, and complement activation and
thrombocytopenia in monkeys (218,219). However, these adverse events are
transient, reversible and only associated with high dosage (219-221). Lack of
genetic toxicity has been confirmed by different researchers (164,219).
The polyanionic property of phosphorothioate-ODN due to the addition of
sulfur atoms is responsible for some side effects such as thrombocytopenia
(137).Due to the limited cellular uptake without appropriate vectors, AS-ODNs
have to be given at high doses and frequency in these clinical studies. This in
part accounts for the toxicity discussed above.
Liposomes are used as carriers for a wide variety of drugs. They are safe
and non-immunogenic. They can also reduce the acute toxicity of many
compounds such as amphotericin B and doxorubicin. Generally speaking,
liposome formulations tend to protect tissues from the toxic effects of
encapsulated compound and are usually better tolerated than the free drug
(222). With improved uptake by liposomes, AS-ODN can be given at a lower
dose to elicit the same therapeutic benefits. Cationic liposomes will also
decrease the net negative charge of phosphorothioate-ODN and thus attenuate
its nonspecific interaction with serum proteins. Therefore cationic liposomes
should be able to improve the delivery as well as reduce the toxicity of
This study is designed to test the effects of repeated injections of 31-AS-
ODN on BP and cardiac hypertrophy in hypertensive rats. Toxicity and immune
response will be assessed.
6.2.1 Effects of Repeated pI-AS-ODN Treatment on BP
The efficacy of three repeated iv administrations of 1 mg/kg pi-AS-ODN
(n=9) was tested in adult SHR with high BP. Animals treated with three repeated
injections of 1mg/kg INV-ODN (n=7) or saline (n=7) served as controls. Systolic
BP was measured every three days. A second injection of P1-AS-ODN was given
before the BP returned to baseline after the previous treatment.
Three repeated injections of pi-AS-ODN at 16, 20 and 26 days
maintained SBP of SHR at 20-30 mmHg lower than controls for more than 2
months (Fig. 6-1). There was no loss in the antihypertensive efficacy with
repeated treatments. Instead, the 2nd and 3rd injections were more effective than
the 1st one in the extended duration of BP reduction. Control animals treated with
saline or INV-ODN had no change in BP with repeated injections. HR was not
altered in any groups.
6.2.2 Effects of Repeated 31-AS-ODN Treatment on Cardiac Hypertrophy
At the end of 2-month treatment, SHR were sacrificed and the organs
were weighed in a blind manner. The left ventricle of the heart was dissected
free of fat and connective tissues, and weighed by a colleague who was not
aware of the group assignment. Left ventricular hypertrophy (LVH) was indicated
by the ratio of left ventricle weight to body weight (LV/BW). p3-AS-ODN treatment
produced a small but significant (P<0.05) reduction in hypertension-associated
LVH in SHR compared to saline control (Fig. 6-2).
6.2.3 Safety Profile
Liver is a major organ for uptake and clearance of both liposomes and
phosphorothioate-ODN when they are administered separately (222-224).
Therefore large doses of liposome/ODN complex may overload the liver and
cause liver toxicity. Alanine aminotransferase (ALT) and aspartate
aminotransferase (AST) are routinely used as a clinical index for liver function.
Their levels increase substantially upon liver damage. Plasma concentrations of
ALT and AST were measured after each injection, which revealed no significant
difference between AS-ODN and control groups (Fig. 6-3).
Phosphorothioate-ODN has increased negative charges due to sulfur
atoms and may interact with positively charged proteins in the blood and cause
hematological alterations. Reduced platelet counts were reported with high
doses of systemic injection of phosphorothioate-ODN (>5mg/kg/day for weeks)
(225). At the end of 2-month treatment, hematological parameters were
measured. There was no difference in total leukocyte count (WBC), platelet
count, hematocrit and mean platelet volume (MPV) among three groups (Table
Table 6-1. Clinical biochemical and hematological parameters after repeated
administration of P3-AS-ODN.
p31-AS-ODN INV-ODN Saline
Antibody to ODN
74.0 + 3.7
n = 7-9 for each group, ND: non-detectable.
In order to investigate whether animals develop immune responses
against repeated administration of pi-AS-ODN, gel immunodiffusion assay was
employed to detect possible antigen-antibody reaction between liposome/ODN
complex and serum. If the body produces specific antibodies against a certain
antigen, the increase of IgG levels will have different time courses after the initial
and following inoculations. Second encounter with the antigen will mount a more
rapid and greater elevation in IgG production than the first response. Considering
this immune memory, serum samples for antibody detection were collected at
13, 7 and 7 days after the 1st, 2nd and 3rd injection, respectively.
No antibodies to ODN/liposome were detected in immunodiffusion
analysis. Presence of C-reactive protein (CRP) in the blood is a clinical indication
for immune reaction after infection. This parameter was found to be negative in
all animals before and after repeated injections of PI-AS-ODN, INV-ODN or
saline. It provided further evidence for the absence of immune responses.
High doses of phosphorothioate-ODN were associated with hyperplasia of
mononuclear/phagocytic cells in the reticuloendothelial (RE) system (225). Heart,
liver, spleen and kidney were subjected to pathological examination at the end of
the study, which revealed no significant sign of immunopathology or organ
damage (Fig. 6-4 to 6-7). Mononuclear cell infiltration was occasionally observed
in small areas of the heart, but this occurred in all three groups. It is probably due
to the higher background immune response in SHR, compared to WKY rats.
Enlargement of spleen and kidney has been reported in rats receiving
high doses of phosphorothioate-ODN (219,225). In this study, no significant
change was found in the weight of spleen and kidney of all groups (Table 6-2).
Table 6-2. Organ weights after repeated administration of p3-AS-ODN.
pl-AS-ODN INV-ODN Saline
LV/BW (g/kg) 3.10 + 0.07* 3.26 0.06 3.33 + 0.04
spleen (g) 0.64 0.03 0.65 + 0.01 0.62 + 0.1
kidney (g) 1.32 + 0.03 1.34 + 0.08 1.34 0.06
n = 7-9 for each group
P < 0.05 vs. saline control.
Maintained Efficacy and Lack of Immune Response after Repeated
Administration of pJ-AS-ODN by Cationic Liposomes
Any therapy that induces immune response cannot be truly useful in clinic,
because immune response not only increases the danger of acute treatment, but
also renders repeated administration ineffective. Gene therapy is no exception.
Tremendous efforts have been made to avoid immune reaction in gene therapy.
These efforts have mainly focused on the choice and development of
appropriate vectors. Among viral vectors, adenovirus and HSV are known for
acute toxicity and immune response, albeit new generations of vectors have
minimized viral genomes and are better tolerated. Retrovirus has high
transduction efficiency, but is usually limited in dividing cells. AAV is considered
a safe vector, because wild type AAV has not been associated with any
pathology in humans. It has a broad infectivity and can produce long-term
expression of therapeutic genes due to the ability to integrate into the host
genome. However it has its own limitations, i.e. difficulty in production and
relatively low expression [reviewed in (152-154)].
Liposomes, as a nonviral vector, are safe and non-immunogenic. In this
study, liposomal delivery of p3-AS-ODN maintained its antihypertensive
effectiveness upon repeated administration. No immune response or significant
toxicity was observed. These results suggest that the liposome formulation is
technically feasible to achieve sufficient gene delivery and produce prolonged
therapeutic benefits, which are considerably longer lasting than traditional
Inhibition of LVH by Repeated Treatment with 31-AS-ODN
LVH, the disproportionate overgrowth of muscle in the left ventricle, has
historically been considered the natural evolution of hypertension. It is estimated
that the structural remodeling of the heart, including LVH, develops in about 70%
of persons with sustained hypertension (212,213). Hypertrophied cardiac muscle
not only increases myocardial mass and workload, but also contains excess
collagen and increased fibrosis. LVH is associated with higher cardiac risks and
viewed as a strong predictor for heart failure and myocardial infarction (226-229).
In light of the importance of LVH in cardiovascular physiology and patient
prognosis, much attention has been paid to the use of therapies to reverse the
process. Among the antihypertensive agents, angiotensin-converting enzyme
inhibitors are found to be the most potent for reversal of LVH, followed in the
order by calcium channel antagonists and P-blockers (214,230,231). pi-selective
blockade is highly effective in reversing LVH during chronic treatment (232).
My results indicate that repeated injection of 11-AS-ODN is not only able
to reduce high BP, but also attenuate the cardiac remodeling associated with
hypertension. 2-month treatment results in a small but significant reduction in
LVH of SHR. Increasing the treatment period may exert more marked effects, as
greater LVH regression has been observed after longer treatment with p-
blockers (232). Since LVH is a risk factor for heart failure and myocardial
infarction, 31-AS-ODN may as well protect the heart from these cardiac events
by inhibiting LVH. However, whether the beneficial effects of i1-AS-ODN on LVH
is due to direct inhibition of myocyte growth or secondary to the BP reduction is
Limitations of the Current Study
Generally, the clinical assessment of toxicity requires more extensive
measurements of pathological and biochemical parameters than what were
determined in this study. Therefore further experiments should be performed to
provide a complete safety profile of p1-AS-ODN including blood sugar, lipid
profile, and liver and kidney functions. In addition, the toxicity data presented
here were obtained at least 7 days after each injection. Although there was no
adverse effect beyond this time point, it is still possible that ODN/liposome
treatment may cause some acute toxicity.
The results of this study demonstrate that p -AS-ODN can be given
repeatedly to achieve a sustained reduction of hypertension and reduces cardiac
hypertrophy without causing toxic effects or immune stimulation. These data,
together with early results, show that pi-AS-ODN is viable as a potentially new
prolonged antihypertensive agent with benefits on cardiac remodeling.
0 10 20 30 40 50 60
blood blood blood blood Sacrifice
Figure 6-1. Sustained antihypertensive effects of repeated administrations
of 3,-AS-ODN in SHR.
Three repeated iv injections of 1 mg/kg P,-AS-ODN (n=9) delivered by liposomes
were given to adult SHR as indicated by upward arrows. Repeated injection of
saline (n=7) and 1 mg/kg INV-ODN (n=7) served as controls. Systolic BP was
measured by tail cuff method. Blood samples were collected at different time
points as indicated by downward arrows for safety profile analysis. Data
represent mean SEM. P<0.05 vs. saline or INV-ODN.
Figure 6-2. Repeated administration of P,-AS-ODN reduced
hypertrophy in SHR.
SHR were treated with repeated injections of p,-AS-ODN (n=9), saline (n=7) or
INV-ODN (n=7) for 2 months. P<0.05 vs. saline.
I __ __ L __ i __ __
AS-ODN INV-ODN saline
E2222 1st injection
K2. 2nd injection
M 3rd injection
AS-ODN INV-ODN saline
Plasma levels of liver transaminases ALT and AST after
repeated injections of P,-AS-ODN.
Saline and INV-ODN serve as controls. Data represent mean SEM of each
PROTECTION BY 31-AS-ODN AGAINST MYOCARDIAL DYSFUNCTION
INDUCED BY GLOBAL ISCHEMIA-REPERFUSION
Acute myocardial ischemia causes a significant increase in plasma
catecholamine levels, which leads to exacerbation of the myocardial injury and
dysfunction. Acute myocardial ischemia is also characterized by increased
sensitivity of p-ARs to catecholamines (233). P,-AR, being the predominant
subtype in the myocardium, forms an interface between sympathetic nervous
system and heart function. Several experimental studies show that the density
and mRNA of P,-AR are augmented in the myocardium after acute ischemia
(86,234). p-blockers, especially p,-selective blockers, can protect myocardium
against ischemic injury and dysfunction (235), decrease infarct size (236,237)
and reduce the incidence of sudden cardiac death after myocardial infarction
The previous results have shown that P,-AS-ODN can inhibit P,-AR
expression specifically and reduce BP in SHR for a prolonged period with a
single iv injection. This study is designed to test the hypothesis that the use of p,-
AS-ODN may protect myocardium against dysfunction in ischemia-reperfusion in
the isolated rat heart, an in vitro model of myocardial infarction.
7.2.1 Cardiac Dysfunction during Ischemia-Reperfusion
Four parameters were measured as the major indices for cardiac
functions, including coronary perfusion pressure (CPP), left ventricular end-
diastolic pressure (LVEDP), left ventricular systolic pressure (LVSP) and
developed left ventricular pressure (dLVP = LVSP LVEDP). Myocardial
ischemia results in cardiac dysfunction, which is indicated by increased CPP and
LVEDP and decreased LVSP and dLVP. An increase in CPP may result from an
impaired coronary endothelium-dependent relaxation. The increased LVEDP and
decreased LVSP represent insufficient contraction and relaxation of the
myocardium. The net result of these changes is a diminished dLVP whereby the
heart fails to pump enough blood to meet the elevated oxygen demand.
Five groups of SD rats were studied. Sham controls (n=6) were injected
with saline and perfused continuously for 80 min without ischemia/reperfusion
(I/R). In other groups, hearts were perfused for 20 min and subjected to global
ischemia (30 min) followed by reperfusion (30 min). Saline + I/R (n=6) or atenolol
+ I/R (n=7) groups were injected iv with saline or 2mg/kg atenolol 6 hr before the
hearts were excised. AS-ODN + I/R (n=7) or INV-ODN + I/R (n=7) were injected
iv with 1 mg/kg p,-AS-ODN or INV-ODN 4 days before the I/R experiment.
The basal levels of CPP, LVEDP and dLVP were similar in all groups of
rat hearts. In the sham controls, there were minimal (-5%) changes in the
indices of cardiac function. In saline group, 30 min ischemia followed by 30 min
reperfusion resulted in marked cardiac dysfunction, indicated by a significant
increase in CPP and LVEDP, and a decrease in dLVP (P<0.01 vs. pre-ischemia
Treatment with P,-AS-ODN markedly attenuated the ischemia-reperfusion
induced ventricular abnormality, manifested by the preservation of dLVP and
minimization of increase in CPP and LVEDP (P<0.05 vs. saline group). Overall,
AS-ODN treatment appeared to be equivalent to or even better than atenolol in
these effects. AS-ODN restored dLVP to a greater extent than atenolol. INV-
ODN showed no effect on ischemia-induced dysfunction and cannot be
differentiated from saline group. Data on cardiac function parameters from these
experiments are summarized in Fig. 7-1.
7.2.2 Malondialdehyde Levels in Myocardium
Malondialdehyde (MDA), a product of lipid peroxidation, has been used as
an index of tissue injury after ischemia-reperfusion by several investigators
(238;239). MDA levels in myocardium increased significantly after ischemia-
reperfusion (P<0.05 vs. sham control). Pretreatment of rats with AS-ODN and
atenolol significantly attenuated this increase (P<0.05 vs. saline treatment). As
expected, INV-ODN did not affect MDA levels (Fig. 7-2).
7.2.3 P,-AR Levels in Myocardium
p,-AR levels in the left ventricle were measured after ischemia-reperfusion
by radioligand saturation binding assay (Table 7-1). The Bmax of P,-AR was
slightly elevated during ischemia-reperfusion, compared to sham controls. P,-AS-
ODN treatment significantly reduced the 3,-AR density by nearly 50% (P<0.05),
while atenolol or INV-ODN had no effect. p1-AR levels were not significantly
changed in any group. The affinity of both i,- and p1-ARs remained unaltered.
Table 7-1. Bmax of P,- and 3,-ARs in myocardium during ischemia/reperfusion.
Sham Saline INV-ODN AS-ODN Atenolol
Pi-AR 21.52.0 25.44.4 26.53.2 13.80.9* 27.14.0
12-AR 11.20.5 9.00.5 10.20.7 10.70.7 9.70.6
Data represent mean SEM of 7 rats per group. P<0.05 vs. sham, saline, INV-
ODN or atenolol groups.
There is a generalized stimulation of sympathetic nervous system during
myocardial ischemia, manifested by an increase in catecholamine levels in both
plasma and myocardium (240-242), coupled with hypersensitized cardiac p-ARs
(233). These changes are viewed as a compensatory response in an attempt to
preserve cardiac function. However the consequence is detrimental, because the
sympathetic activation increases the chance of tachyarrhythmia and workload of
the heart which in fact exacerbates cardiac injury induced by ischemia.
Consistent with this concept, pharmacological intervention of p-ARs can protect
the heart and reduce mortality and morbidity during and after myocardial
infarction (85), and prevent recurrent infarction (243). It is noteworthy that all the
P-blockers that have shown benefits in myocardial infarction patients can inhibit
p,-AR, and no difference in the therapeutic efficacy has been found between p,-
selective and non-selective blockers. This suggests that 3,-AR is the major
subtype mediating the deleterious effects of overactive sympathetic system on
the heart. However, as discussed earlier, the use of p-blockers is limited by their
intrinsic drawbacks including undesirable p2-antagonism and short action. A new
strategy for specific P,-AR inhibition by antisense may overcome these problems
and provide an effective and longer-lasting protection against myocardial
Therefore, the present study was designed to examine the protective role
of 3,-AS-ODN against cardiac dysfunction after a brief period of ischemia-
reperfusion. The effects of 3,-AS-ODN was compared with those of a selective
pi-blocker atenolol in this process. The results indicate that 30 min ischemia
followed by 30 min reperfusion resulted in significant cardiac dysfunction and
lipid peroxidation in the saline-treated rat hearts. Pretreatment with both 3,-AS-
ODN and atenolol preserved cardiac performance and reduced lipid peroxidation
following ischemia-reperfusion. The protective effects of P,-AS-ODN on cardiac
function were at least equivalent to those of atenolol, and AS-ODN seemed to
provide a better restoration of dLVP than atenolol.
p,-AR, the predominant p-AR subtype in the myocardium, is the major
mediator of sympathetic activation of cardiac functions. Several experimental
studies have shown an upregulation of P,-AR, but not P,-AR levels, during acute
ischemia (86,234). Ihl-Vahl et al. (86) convincingly demonstrated a subtype-
specific and time-dependent increase in mRNA levels of P,-AR during acute
myocardial ischemia. This is consistent with the clinical data that point to the
major role of 3,-AR in ischemia-induced cardiac dysfunction. In the current
experiments, a slight elevation in P,-AR levels was observed after brief ischemia-
reperfusion, which did not reach the statistical significance. Reperfusion may
explain the discrepancy here, because acute ischemia unequivocally increased
3,-AR expression but the following reperfusion usually restored P,-AR to normal
levels in several studies (87,234).
Both 3,-AS-ODN and atenolol suppress p1-AR functions and protect the
heart from severe injury after ischemia/reperfusion. However, these two
approaches are different in several aspects. First, p,-AS-ODN reduced the total
number of functional P,-AR on the cell membrane. Atenolol competed with
endogenous catecholamines and did not affect 3,-AR expression. Second, AS-
ODN was given 4 days prior to ischemia-reperfusion to allow for sufficient
reduction in P,-AR expression. Atenolol was given 6 hrs before excising the heart
because of its rapid onset of action and short half-life. Therefore, 3,-AS-ODN
may provide a prophylactic protection from myocardial infarction in addition to its
antihypertensive benefit. However it cannot replace the intravenous injection of
P-blockers as the acute postinfarction rescue because of its slow effects.
Isolated rat, rabbit or guinea pig heart model provides an inexpensive and
reproducible method to evaluate cardiac function and myocardial metabolic
alterations during ischemia-reperfusion. It has been used extensively as an in
vitro model for acute myocardial ischemia (182,239). In an isolated and