CONTROL OF THE TONE OF CORONARY RESISTANCE ARTERIES

PhD thesis

Mária Szekeres MD

Tutor: László Dézsi PhD Institute of Human Physiology and Clinical Experimental Research Semmelweis University, Faculty of Medicine

Budapest, 2000

-1-

Contents 1. Introduction………………………………………………………………………………………….. 1.1. Significance 2. Specific aims of the study…………………………………………………….……………………... 3. Overview of the literature…………………………………………………….……………………... 3.1. Coronary circulation………………………………………………………………………….. 3.1.1. Vascular endothelium 3.1.2. Vascular smooth muscle 3.1.3. Vascular resistance 3.1.3.1. Coronary vascular resistance 3.1.4. Coronary microcirculation 3.2. Control of coronary microcirculation………………………………………………………… 3.2.1. Hemodynamic effects on coronary circulation………………………………………… 3.2.1.1. Role of perfusion pressure 3.2.1.2. Biomechanical wall characteristics 3.2.1.3. Shear stress 3.2.2. Mechanical effects of cardiac cycle…………………………….……………………… 3.2.3. Metabolic control mechanisms………………………………………………………… 3.2.4. Neural and humoral control mechanisms……………………………………………… 3.2.5. Endothelium-mediated control mechanisms…………………………………………… 3.2.5.1. EDRF and NO………………………………………………………………….... 3.2.5.1.1. Background 3.2.5.1.2. Biosynthesis and intracellular signalling of NO 3.2.5.1.3. Physiological and pathophysiological role of NO 3.2.5.1.4. Role of NO in the control of coronary vascular tone 3.2.5.2. Prostaglandins (PG-s)…………………………………………………………… 3.2.5.2.1. PG biosynthesis 3.2.5.2.2. Role of PG-s in the control of coronary vascular tone 3.2.5.3. Other endothelium-mediated control mechanisms……………………………… 3.2.6. Direct smooth-muscle mediated control mechanisms…………………………………. 3.2.7. Myogenic response (MR)……………………………………………………………… 3.2.7.1. Background, definition of MR 3.2.7.2. Characterization of MR 3.2.7.3. Physiological role of MR: autoregulation 3.2.7.4. MR in the microcirculation 3.2.7.5. Origin of MR, possible role of endothelium 3.2.7.6. Modulation by exercise 3.2.8. Vasoactive metabolites studied………………………………………………………… 3.2.8.1. Adenosine 3.2.8.2. Acetylcholine 3.2.8.3. Bradykinin 3.2.8.4. Substance P 3.2.8.5. NO donors 3.2.8.6. Norepinephrine 3.2.8.7. Serotonin 3.2.9. Coronary autoregulation……………………………………………………………….. 3.2.10. Integration of microvascular control mechanisms…………………………………… 3.3. The coronary microcirculation in myocardial ischemia and reperfusion…………………….. 3.3.1. Effects of short term ischemia, reactive hyperemia 3.3.2. Effects of long term ischemia, myocardial stunning 3.3.3. Endothelial function in myocardial stunning, role of NO 3.3.4. Cardioprotection 3.4. Effect of exercise on the coronary vascular functions………………………………………... 3.5. Methods to study coronary microcirculation…………………………………………………. 4. Materials and Methods………………………………………………………………………………. 4.1. Isolated vessel studies……………………………………………….………………………... 4.1.1. Animals

-2-

9 10 12 12

14 15

18 19 20 21 22

28 30 30 31

34

40 41 42

47 48 50 50

4.1.1.1. Exercise protocol 4.1.2. Equipment 4.1.3. Solutions and drugs 4.1.4. Preparation 4.1.5. Cannulation and mounting 4.1.6. Experimental protocols 4.1.7. Data processing 4.1.8. Statistical analysis 4.2. Isolated heart studies………………………………………………………………………….. 59 4.2.1. Animals 4.2.2. Equipment 4.2.3. Solutions and drugs 4.2.4. Preparation and mounting 4.2.5. Experimental protocols 4.2.6. Data processing 4.2.7. Statistical analysis 5. Results…………………………………………………………………….…………………………. 64 5.1. Isolated vessel studies…………………………………………………………….…………... 64 5.1.1. Biomechanical wall characteristics of rat intramural coronary resistance arteries and arterioles 5.1.2. Myogenic characteristics of rat intramural coronary arterioles 5.1.2.1. Modulatory role of endothelium 5.1.2.2. Effects of daily exercise 5.1.3. Endothelium-dependent and endothelium-independent vasoactive responses of rat intramural coronary resistance arteries and arterioles 5.1.3.1. Endothelium-dependent and endothelium-independent vasoactive responses of rat intramural coronary arterioles 5.1.3.2. Endothelium-dependent and endothelium-independent vasoactive responses of rat intramural coronary resistance small arteries 5.1.4. Intraluminal drug administration in rat intramural coronary resistance arteries 5.1.4.1. Bradykinin-induced responses with intraluminal flow in rat intramural coronary resistance arteries 5.2. Isolated heart studies………………………………………………………………………….. 70 5.2.1. Endothelium-dependent and endothelium-independent vasoactive responses of the rat heart 5.2.2. Effects of an NO donor, GEA 3162 on the biomechanical performance of the isolated ischemic rat heart 6. Discussion of experimental results………………………………………………………………….. 72 6.1. Isolated vessel studies…………………………………………………………….…………... 72 6.1.1. Coronary wall biomechanics 6.1.2. Coronary myogenic response 6.1.3. Function of endothelium in coronary resistance arteries and arterioles 6.1.3.1. In vitro verse in vivo studies 6.1.4. Effects of daily exercise on the coronary arteriolar function 6.1.5. Intraluminal drug administration 6.2. Isolated heart studies………………………………………………………………………….. 84 6.2.1. Function of endothelium in the isolated rat heart 6.2.2. Cardioprotective effects of GEA 3162 in ischemia-reperfusion 7. Conclusions………………………………………………………………………………………….. 87 8. Acknowledgements………………………………………………………………………………….. 90 9. References…………………………………………………………………………………………… 91 10. Publications…………………………………………………………………………………………106

-3-

Abbreviations

ACE ACh ADO ADP AMP ANOVA ANP ATP BK o C cAMP Ca2+ CaCl2 CF cGMP Clcm3 cNOS CO2 CPP CT CVR D ∆ DAG Dinc DMSO dP/dt EDHF EDRF EDTA EGTA Einc ET-1 EX Fe Ftang g GAPDH GMP h H+ HCO3H 2O 2 HPO42H2PO4 HR 5-HT

INDO iNOS Ip. IP3 K+ K+ATP KCl kg KH2PO4 l LAD L-NAME L-NNA L-NNA LVP M M m2 MAP-kinases

Angiotensin converting enzyme Acetylcholine Adenosine Adenosine-di-phosphate Adenosine-mono-phosphate Analysis of variances Atrial natriuretic peptide Adenosine-tri-phosphate Bradykinin Celsius degree Cyclic AMP Calcium ion Calcium-cloride Coronary flow Cyclic-GMP Cloride Centimeter cube (ml) Constitutive NOS Carbon dioxide Coronary perfusion pressure Computer tomography Coronary vascular resistance Diameter Change Diacyl glicerol Incremental distensibility Dimethylsulfoxide Left ventricular pressure change vs. time (contractility) Endothelium-derived hyperpolarizing factor Endothelium-derived relaxing factor Ethylene-diamine-tetraaceticacid Ethylene-glycol-bis-βaminoethylether-tetraacetic-acid Incremental elastic modulus Endothelin Exercised Iron Circumferential tension Grams Glyceraldehide-3-phosphatedehydrogenase Guanosine-mono-phosphate Wall thickness Hydrogen ion Bicarbonate ion Hydrogen peroxide Hydrogen-posphate ion Hydrogen-posphate ion Heart rate Serotonin (5-hydroxytryptamine)

mg Mg2+ MgSO4 MI min ml µl mm mM µm mmHg mmol mph MR MS Na+ NaCl NADPH NaHCO3 NaNO2 NE nKR NMR nNOS NO NOS . O2O2 . OH P PD PET PG

-4-

Indomethacin Inducible NOS Intraperitoneal Inositol triphosphate Potassium ATP sensitive K+ channels Potassium-cloride Kilograms Potassium-hydrogen-phosphate Liter Left anterior descending NG-nitro-L-arginine-methylester NG-nitro-L-arginine Nω-nitro-L-arginine Left ventricular pressure Muscarinic Molar/liter Meter square Mitogen-activated protein kinases Miligrams Magnesium ion Magnesium-sulphate Myogenic index Minute Mililiters Microliters Milimeters Milimolar Micrometers Mercury milimeters Milimolar Miles per hour Myogenic response Myocardial stunning Sodium Sodium-cloride Nicotinamide-dinucleotide phosphate Sodium-hidrogen-carbonate Sodium-nitrite Norepinephrine Normal Krebs-Ringer Nuclear magnetic resonance Neuronal NOS Nitric oxide Nitric oxide synthase Superoxide radical Oxygen Hydroxyl radical Pressure Passive diameter Positron emission tomography Prostaglandin

PGI2 pH Pi pO2 PP PSS RAS RI Ro sec SED SEM SNAP SNP SO42SOD SP δtang TX U VIP

Prostacyclin -log[H+] Phosphate ion Partial oxygen pressure Perfusion pressure Physiological salt solution Renin-angiotensin system Inner radius Outer radius Second Sedentary Standard error of the mean S-nitroso-N-acetyl-penicillamine Sodium-nitroprusside Sulphate ion Superoxide-dismutase Substance P Circumferential wall stress Thromboxane Units Vasoactive intestinal peptide

-5-

Összefoglaló Koronária rezisztencia artériák tónusának szabályozása. Dr. Szekeres Mária Témavezető: Dr. Dézsi László PhD program címe: A vérkeringési rendszer normális és kóros működési mechanizmusai. Semmelweis Egyetem, Általános Orvostudományi Kar, Doktori Iskola, Budapest, 2000 Jelen

tanulmány

szabályozásával

a

koronária

foglalkozik.

A

rezisztencia koronária

artériák

tónusának

mikrocirkuláció

adaptív

szabályozó

mechanizmusainak részletes irodalmi összefoglalása során kiemeltük a jelen kísérleti munkával kapcsolatos témákat, pl. biomechanikai, miogén és endotélium-függő adaptációs mechanizmusok. Különös hangsúlyt fektettünk a kísérleteinkben alkalmazott vazoaktív anyagok (pl. bradikinin, adenozin, acetilkolin, prosztaglandinok, stb.) hatásainak elemzésére. Ezen kívül a kísérletesen előállított iszkémia és reperfúzió esetében tapasztalható miokardiális adaptációs mechanizmusok megbeszélésével különös figyelmet fordítunk a kardioprotekció kérdésére. Jelen tanulmány céljaként intramurális

koronária

rezisztencia

artériák

biomechanikai,

fiziológiai

és

farmakofiziológiai szabályozó mechanizmusainak részletes leírását tűztük ki, valamint ezen paraméterek változását terveztük megfigyelni miokardiális iszkémia és rendszeres fizikai aktivitás esetén. Kísérleteinket izolált, perfundált patkány szíveken, valamint izolált, kanülált patkány koronária mikroereken végeztük. Az intramurális koronária mikroerek preparációs technikájának kifejlesztése lehetőséget nyújtott a koronária vaszkuláris kontroll mechanizmusok szegmentális tanulmányozására. Eredményeink azt jelzik, hogy izolált patkány intramurális rezisztencia artériák geometriai, elasztikus és kontraktilis tulajdonságai szegmentális megoszlást mutatnak. Kisebb erek spontán tónusa és relatív falvastagsága nagyobb, amely kisebb falfeszülést és elasztikus modulust, és nagyobb disztenzibilitást eredményez a nagyobb erekhez képest. Intramurális patkány koronária arteriolák jelentős miogén választ mutatnak, amelyben a nitrogén monoxid és konstriktor prosztaglandinok szerepe meghatározó.

-6-

Rendszeres fizikai aktivitás növelte a koronária arteriolák miogén válaszát és a nitrogén monoxid értónust befolyásoló szerepét. Az érfal simaizon hipertrófia és a csökkent falfeszülés hozzájárulhat a koronária vazodilatációs rezerv megnövekedéséhez a rendszeres fizikai aktivitást követően. Endotélium-függő és direkt simaizom-függő vaszkuláris funkciók egyensúlya a koronária artériás rendszerben szegmentális variabilitást mutat. Bradikinin nitrogén monoxid-függő vazodilatációt okozott kis rezisztencia artériákon és arteriolákon egyaránt. Nagy dózisú acetilkolin a nagyobb erekben dilatációt, a kisebb erekben azonban konstrikciót eredményezett. Izolált szíven bradikinin koronária áramlás fokozódást idézett elő, míg acetilkolin áramláscsökkentő hatása alapján növelte a koronáriaerek tónusát az izolált arteriolákon megfigyelt hatásának megfelelően. Kimutattuk továbbá, hogy nitrogén monoxid és konstriktor prosztaglandinok meghatározó szerepet játszanak a nyugalmi koronária tónus beállításában. Izolált, iszkémiás patkány szíven egy új nitrogén monoxid donor, GEA 3162 kardioprotektív hatását figyeltük meg. Reméljük, hogy jelen tanulmány eredményeivel hozzájárulunk a koronária mikrocirkuláció szabályozó mechanizmusai szegmentális megoszlásának, valamint a miokardiális iszkémia és rendszeres fizikai aktívitás által előidézett adaptációs mechanizmusok pontosabb megértéséhez.

-7-

Summary Control of the tone of coronary resistance arteries. Mária Szekeres MD Tutor: László Dézsi PhD PhD program: Physiological and pathological mechanisms of the cardiovascular system. Semmelweis University, Faculty of Medicine, School of PhD Studies, Budapest 2000 In the present study adaptive control mechanisms of coronary resistance vessels have been studied. An exhaustive summary of the literature is given of the controlling mechanisms of coronary microcirculation with a special emphasis on topics related to our experimental work, i.e. biomechanical, myogenic, and endothelium-mediated adaptation mechanisms. Effects of vasoactive agonists on coronary resistance vessels used in our studies (bradykinin, adenosine, acetylcholine, prostaglandins etc.) are handled with special attention. In addition, we also discuss myocardial adaptation mechanisms in respect to experimental ischemia and reperfusion focusing on the mechanisms of cardioprotection. The aims of the study are outlined to give an extensive description

the

biomechanical,

physiological,

pharmacophysiological

control

mechanisms of intramural coronary resistance vessels as well as how these parameters are changing in such conditions as myocardial ischemia and chronic exercise training. The experiments included in this study were carried out on isolated, perfused rat hearts as well as on isolated, pressurized rat coronary microvessels applying in vitro microangiography. A methodology has been developed for preparation, isolation and cannulation of intramural coronary microvessels. This allowed us to examine segmental differences in coronary vascular controlling mechanisms. It has been found, that in isolated rat intramural coronary resistance arteries and arterioles geometric, elastic and contractile properties are changing along the vascular tree, showing a characteristic segmental specificity. Toward smaller coronary arteries spontaneous tone and relative wall thickness increased, which decreased average wall stress, that in turn seems to explain the observed higher distensibility and lower elastic modulus in these vessels. Examining myogenic behavior of intramural rat coronary arterioles we found, that they exhibit substantial myogenic responses greatly influenced

-8-

by nitric oxide and constrictor prostaglandins. Daily exercise augmented myogenic response and enhanced the role of nitric oxide in the control of coronary vascular diameter. The observed arteriolar smooth muscle hypertrophy and decreased arteriolar wall stress may be beneficial in enhancing coronary vasodilator reserve during exercise. Balance of endothelium-dependent and direct smooth muscle-dependent vascular functions showed segmental variability in different segments of the coronary arterial tree. Bradykinin induced vasodilations mediated by nitric oxide both in small coronary arteries and in arterioles. Higher doses of acetylcholine, however, induced dilation in larger vessels but constriction in smaller vessels. On isolated hearts, bradykinin induced increase in coronary flow, but acetylcholine a flow reduction due to an increased coronary vascular resistance, which is in accord with its effect on arterioles. Furthermore, it has been demonstrated, that both nitric oxide and constrictor prostaglandins play a substantial role in setting the basal level of coronary vascular tone. On isolated ischemic rat hearts a beneficial effect of a new nitric oxide donor, GEA 3162 was found. We hope, the data presented in this study will improve our understanding of the segmentally specific control processes of the coronary microcirculation, their adaptation to exercise as well as to myocardial ischemia.

-9-

1. Introduction Diseases of the cardiovascular system represent the major source of morbidity and mortality in the national statistical records in developed countries. It is a great challenge for the basic research work to discover biophysical, physiological, pathophysiological and pharmacological mechanisms in this field in order to open up new strategies in the diagnostic and therapeutic approach for these diseases. Coronary resistance vessels represent the majority of the hemodynamic resistance of the heart, they are the major site of metabolic control processes, by altering vascular resistance they greatly determine myocardial functions (38, 40, 137, 141, 173, 187). Thus, coronary resistance arteries and arterioles play an essential role in the physiological and pathological cardiac functions, they are the targets for many drug actions (43, 96, 108, 173). Abnormalities of coronary microvessels lead to such cardiac diseases, as microvascular angina and other ischemic heart diseases (43, 108, 170). There is an increasing importance of the research of coronary microcirculatory mechanisms, e. g. on the control of microvascular tone, which in turn determines microvascular resistance and limits cardiac functions. Long term control mechanisms of the coronary microcirculation also have utmost importance, e.g. how the hemodynamic resistance is altered in some physiologic and pathophysiologic conditions, such as in exercise training, or in hypertension and diabetes. Such observations can lead to new therapeutic approaches. 1.1. Significance The experimental studies included in this study were performed on Langendorff rat hearts and on isolated rat coronary resistance arteries and arterioles. All small artery and arteriole segments used in this study were in intramural position. Different segments from the course of the left anterior descending coronary artery along from the orifice to the apex were isolated. Small coronary arteries or arterioles in superficial location have been isolated from several animal species (126, 137, 138, 139, 141, 205, 206, 212, 243). The adjustment of their vascular tone to the changes of arterial pressure in vivo is considered to be a continuous, dynamically controlled process (51, 70, 159).

- 10 -

Thus, assuming that intramural coronary microvessels make up the greatest number and volume of coronary vessels (96), it has a great importance of studying coronary resistance vessels of intramural position. Because of preparation difficulties, there are only few studies available on coronary resistance arteries of the rat with intramural position or the position of the vessels is not properly addressed (79, 187, 188). We studied myogenic mechanisms on isolated, cannulated intramural coronary arterioles, such as myogenic response, that is characterized by a decrease in vascular diameter to step increases in intraluminal pressure. This is one of the most important determining factor of coronary vascular resistance, and in turn of the autoregulation of blood flow (70, 96, 106, 108, 137, 139, 173, 176, 212, 227, 228). We studied vascular wall biomechanics on isolated, cannulated intramural small coronary arteries and arterioles. This is another important factor in determining vascular resistance (1, 20, 33, 48, 49, 60, 81, 149, 166, 167, 176, 179, 180, 207, 233, 237). And we studied pharmacological responses to different vasoactive agents on these vessels, and their dependence on the function of endothelium, which has a great importance according to previous studies (35, 79, 130, 131, 137, 139, 141, 173, 206, 212, 225). On the other hand, isolated heart studies give us an overall information about the coronary vascular resistance, which can be computed from the coronary flow and from the coronary perfusion pressure. Studying cardiac performance on this preparation has a great importance in ischemia-reperfusion studies (4, 10, 27, 75, 91, 107, 151, 153, 158, 232, 256). Studying the coronary microcirculation by two different technics (Langendorff heart and isolated segments) we had a possibility to compare the results obtained. Segmental differences, their contribution to the overall vascular resistance could be evaluated. 2. Specific aims of the study Our aim was to shed light on a number of potentially biological problems with important clinical relevance, such as coronary microvascular functions in respect of myogenic mechanisms, endothelial functions and biomechanical characteristics.

- 11 -

The primary focuses of the study were: 1. Comparing endothelial and smooth-muscle directed functions in determining coronary vascular resistance both in an overall manner in isolated rat hearts, and at different levels of the coronary artery branching system on isolated rat coronary vessels. 2. Comparing

segmentally

specific

coronary

vascular

functions

including

biomechanical wall characteristics, endothelial and smooth-muscle directed functions in different levels of the coronary resistance microvasculature. 3. Studying specific myogenic and pharmacological responses of the smallest resistance coronary vessels, the arterioles, and further examining their behavior in response to daily exercise. The specific aims were: Methodological aims: To introduce a technique for isolation and cannulation of intramural coronary small arteries and arterioles. And to introduce a technique for intraluminal drug application into isolated pressurized microvessels. Physiological aims: 1. To determine, how endothelium-dependent and -independent vasoactive agents, such as bradykinin, acetylcholine, adenosine are involved in coronary flow adjustments, what is their relative contribution and physiological significance in the coronary circulation. And to specify the extent of endothelial involvement in the control of coronary circulation of the rat heart by determining the balance of endothelium-mediated and direct smooth-muscle-mediated actions in intact hearts compared with data on other species in the literature.

- 12 -

2. To specify cardiac biomechanical performance altered by myocardial ischemia and reperfusion and its modulation by a NO-donor, GEA 3162, which is proposed to have a beneficial effect in ischemia-reperfusion. 3. To examine, whether endothelial function is altered by low-perfusion ischemia and reperfusion in isolated rat heart. 4. To investigate, how geometric, elastic and contractile properties of rat intramural coronary arteries are changing along the vascular tree of resistance arteries and arterioles, and how these parameters depend on the value of intraluminal pressure as well as on the extent of smooth muscle tone in vessels with different morphological caliber. 5. To characterize the myogenic mechanisms of rat intramural coronary arterioles, and to elucidate its modulation by nitric oxide and prostaglandins. 6. To characterize the effects of daily exercise training on the structure (e.g. wall thickness) and function of rat intramural coronary arterioles focusing on the characteristics of the myogenic response of its modulation by nitric oxide and prostaglandins. 7. To specify segmental differences in the role of the endothelium and smooth muscle in the regulation of the tone and responses of rat intramural coronary resistance arteries and arterioles by elucidating the actions of endothelium-dependent and independent vasoactive agents, such as bradykinin, acetylcholine, substance P, adenosine, nitroprusside and norepinephrine. And to specify the extent of involvement of nitric oxide in the control of the tone of coronary resistance arteries and arterioles. 8. To characterize the role of prostaglandins in the modulation of coronary arteriolar tone.

- 13 -

9. To characterize concentration-response curves for serotonin and bradykinin recorded at extraluminal application of the agents on intramural coronary arteries and to further characterize a potential intraluminal/extraluminal asymmetry of the bradykinin-response with additional intraluminal application of bradykinin during constant intraluminal flow.

- 14 -

3. Overview of the literature As a consequence of its clinical importance, the number of publications in the literature on coronary vascular resistance, on its long term and short term control mechanisms, the morphology, the pathology, pharmacology of coronary resistance vessels is enormously high. Still we wanted to give an exhaustive analysis of existing literature in connection with the biomechanical, physiological and pharmacological properties of coronary resistance vessels forming the scope of our experimental study. As a compromise, in this chapter, a logically structured analysis of coronary resistance artery function will be given, but specific emphasis will be placed on publications referring to the topic of our experimental studies. Thus, this study is limited not to be extended to some important mechanisms in the control of coronary vascular resistance, such as neural reflex mechanisms with central nervous representations. To save space, the Discussion was shortened, facts of literature needed to explain our experimental data of the Results are shortly mentioned, and for a more detailed overview the Reader will be referred back to this chapter. 3.1. Coronary circulation 3.1.1. Vascular endothelium Coronary vascular endothelium, in the inner layer of the vessel (intima) consists of a continuous layer of squamous epithelial cells. In arterioles, the cells are reported to be 2 µm thick, 10-20 µm wide and 30-50 µm long (176) with tapering ends and oriented with the long axis parallel to the direction of flow. Its great importance is suggested by the fact that the total surface area of the endothelial cells in the human body reaches 720 m2. Endothelium has an ability to affect vascular contractility by their frequent projection through fenestrations to the internal elastic lamina and may make contact with the vascular smooth muscle cells within the tunica media (176). Among several important functions of the vascular endothelium it synthetizes a number of substances that play a substantial role in the control of vascular tone by modulating the contractile state of vascular smooth muscle (3.1.2.). Vascular endothelium synthetizes relaxing

- 15 -

factors, from which the endothelium-derived relaxing factor (EDRF), also called nitric oxide (NO, 3.2.5.1.) is the most prominent. Other relaxing factors are prostaglandin I2 (PGI2) and endothelium-derived hyperpolarizing factor (EDHF, 11, 76, 86, 222). Most of the relaxing factors (NO, PGI2) inhibit platelet aggregation. Endothelial cells also synthesize contracting factors (EDCF-s), e.g. thromboxane A2 -which also activates platelet aggregation-, and endothelin (76, 121, 157). Endothelial cells synthesize enzymes, e. g. angiotensin converting enzime, several clotting factors, hormon sensitive lipase, etc. Endothelial cells possess several receptors, which determine the actions of different vasoactive substances modulating endothelial function (3.2.8.). Endothelial cells sense hemodynamic effects e.g. shear stress, which is described to modulate endothelial function, and in turn vascular resistance (3.2.1.3.). Furthermore, endothelium has a great importance as a physical, mechanical and metabolic barrier. 3.1.2. Vascular smooth muscle Vascular smooth muscle is located in the tunica media. The number of smooth muscle layers within the media of small arteries decreases with decreasing vessel diameter from approximately six layers in 300-µm vessels to a monolayer in 30-50 µm arterioles. The volume fraction of smooth muscle cells within the media of small arteries is about 70-85 %. Smooth muscle cells are circumferentially arranged in small arteries within the media. Smooth muscle cells appear to be mechanical connected mainly through membranous contacts (dense bodies). Even in small arteries they appear to form electrical syncytium lacking gap junctions. Models of elastic and contractile elements explain the structure of small arteries, which are supposed to determine passive and active wall characteristics (176). Vasoactive materials on specific receptors may modulate the funtions of vascular smooth muscle cells directly, or indirectly by released mediators from endothelial cells (3.2.5., 3.2.6.). 3.1.3. Vascular resistance Small arteries consist of endothelial (intima, 3.1.1.), smooth muscle (media, 3.1.2.) and adventitial layers. However, in smallest resistance vessels e. g. in the

- 16 -

coronary circulation, adventitial layer is extremely small, not definable during isolation (4.1.4.). All layers determine vessel wall characteristics, thus also vascular diameter, which is the major determinant for the vascular resistance. The resistance of a vessel segment (R) is calculated by the division of pressure difference (∆P) and intraluminal flow (Q): R=∆P/Q. Hemodynamic resistance (R) of blood vessels is determined by inner vessel radius (r), length (l) and blood viscosity (η) according to Hagen-Poiseuille’s law: R=8ηl/πr4 (87, 176). In large vessels, due to the pulsatile flow characteristics, impedance can be calculated instead of the resistance. However the majority of the circulatory resistance is determined by small arteries less than 500 µm, which participate actively in the regulation of peripheral resistance in different tissues. These resistance vessels give the greatest impediment against intraluminal blood flow (173, 176). Vascular diameter is determined by transmural pressure and mechanical (elastic and contractile) characteristics of the vessel wall (3.2.1.). Vascular tone, the actual sustained contractile state of the vascular smooth muscle (when active diameter is compared to the passive diameter, 4.1.7.) is controlled by several factors. Blood vessels possess an intrinsic tone, which exists after isolation, thus obviously originated from the vascular smooth muscle itself. In vivo resistance vessels of most vascular beds possess an extrinsic, active sympathetic tone. Furthermore, vascular tone is modulated by the vascular endothelium, metabolic, humoral, neural and hemodynamic factors. These, complicated and compound mechanisms adjusting vascular tone will control overall peripheral resistance throughout the microcirculation (12, 69, 87, 176).

- 17 -

3.1.3.1. Coronary vascular resistance In case of coronary circulation, coronary vascular resistance (CVR) is calculated from the changes of coronary perfusion pressure and coronary blood flow, which in turn dynamically influences coronary vascular diameter. The coronary blood flow is adjusted to meet the myocardial oxygen and metabolic demand in normal conditions. Thus, the actual diameter of coronary vessels is controlled by several factors, i.e. by perfusion pressure, mechanical effects of cardiac cycle, metabolic, humoral, neural factors and intrinsic properties of vessel wall (11, 12, 112, 173) -discussed in details in 3.2. Small coronary arteries and arterioles have been assumed to represent the majority of the hemodynamic resistance of the heart (38, 173), which varies with subepicardial, subendocardial and intramural location (38, 40). Compared with larger vessels, they are the major site of metabolic control, have large spontaneous tone (137, 141, 173, 187), exhibit myogenic response (136, 137, 138, 139, 173, 212) and have large vasodilatory reserve (137, 141, 173). In experimental conditions CVR can be calculated from coronary perfusion pressure and coronary flow. In constant pressure-perfused Langendorff rat hearts CVR determines coronary blood flow, whereas in constant flow-perfused hearts, CVR determines the measured perfusion pressure (4.2.2.). 3.1.4. Coronary microcirculation In the present study we focus on the coronary resistance vasculature. In the experimental work described in this thesis, coronary vasculature of the rat heart have been studied. Concerning the small sizes of the rat heart arterial vessels, all vessels are defined as resistance-sized vessels. In the Langendorff rat heart preparation we used (4.2.), microcirculatory processes are studied in the complete vascular bed included perfused vessels starting from larger, but resistance-sized coronary arteries, arterioles, capillaries continued to the venous system, the right heart chambers and the pulmonary artery. Changes in the coronary vascular resistance in perfused hearts give us an overall information about the coronary microcirculation. Meanwhile, in the studies performed on isolated coronary segments of rat heart (4.1.), segmentally specific microcirculatory

- 18 -

processes are being assessed. Thus, with the isolation of different sized resistance arteries and arterioles we can get information about the regulatory processes specificly on a certain part of the coronary microcirculation. 3.2. Control of coronary microcirculation Many different regulatory factors influence the caliber of coronary arterioles and small arteries. Each of these factors appear to have the potential to affect total coronary vascular resistance (173). Due to the limited extent of this study, there is no possibility to give all the details of the mechanisms controlling coronary circulation. Importantly, myocardial blood flow must be suited to the metabolic demand of the heart. Thus, changes in metabolic demand of the heart (e.g. in exercise) substantially influence coronary blood flow. In consequence of an inbalance between myocardial oxygen supply and demand, myocardial ischemia develops (11, 28, 96, 173, 193). In the followings we analyze several factors that influence myocardial blood flow (112, 193), e.g. perfusion pressure, shear stress, myogenic, metabolic and neural factors with a special emphasis on the role of the vascular endothelium. The most important controlling factors are summarized in Fig. 3.-1. There is a nonuniformity of the vasomotor responsiveness throughout the coronary arterial tree, that suggests a heterogeneity of microvascular control mechanisms (141, 147, 173). Generally it can be assumed, that local regulatory mechanisms, such as metabolic, myogenic regulation, endothelium-mediated regulation is dominant on the microcirculatory site targeting mostly resistance-sized arteries and arterioles (12, 38, 40, 96, 137, 139, 147, 173). 3.2.1. Hemodynamic effects on coronary circulation Pressure and flow dynamics in the circulation is primarily determined by the cardiac function and vascular resistance. Hemodynamic effects also involve the viscous characteristics of blood which markedly influence the dynamics of the blood flow. Blood flow, continuously changing throughout the cardiac cycle elicits a dynamic shear force on the endothelial surface of blood vessels.

- 19 -

3.2.1.1. Role of perfusion pressure As in any perfusion circuit, blood flow is related to the available driving pressure. However, the relationship between perfusion pressure and flow is more complex in the coronary circulation than in other vascular beds (11, 69). Coronary perfusion pressure (CPP), a basic determinant of coronary blood flow, which is primarily controlled by the aortic pressure continuously changing during the cardiac cycle. CPP will also determine transmural pressure through the coronary vessel wall, which is extensively influenced by the mechanical effects of the cardiac contraction, thus the developed intramyocardial pressure (3.2.2., 67, 96). However, the relationship of CPP to coronary blood flow involves many factors of coronary and myocardial physiology. In case of normal CPP, mean coronary blood flow depends mainly on the myocardial metabolism (96). Furthermore, with the changes in CPP, coronary blood flow is greatly determined by the intrinsic myogenic and biomechanical vessel wall characteristics (3.2.1.2., 3.2.7). When perfusion pressure changes in the coronary circulation, vessels actively react and change their diameter in order to keep coronary blood flow in a small range (called autoregulation). This, intrinsic, myogenic mechanism of vessel wall (discussed in details in 3.2.7.), exists mainly in the microcirculatory level, primarily determines autoregulation of coronary blood flow (3.2.9., 12, 96, 137, 139, 147, 173, 194). CPP influences biomechanical wall characteristics, that involves pressure-diameter or pressure-volume relations, and is also determined by the elasticity of the vessel wall (3.2.1.2.). Furthermore, CPP, along with blood flow can independently determine myocardial function, and consequently myocardial oxygen consumption, which is called as Gregg phenomen. Thus, an increase in myocardial oxygen consumption secondary to increased myocardial perfusion implies an inverse causal relationship between function and perfusion as compared to the relation during metabolic regulation, where increases in function are supposed to cause increases in perfusion (11, 67, 84, 93). In experimental conditions perfusion pressure is adequately controlled. If we suddenly increase perfusion pressure in constant pressure perfused Langendorff rat heart experiments by elevating the height of reservoirs (4.2.2.), coronary blood flow slowly decreases as a result of a reactive constriction of coronary vasculature (autoregulation).

- 20 -

On the other hand, if CPP is decreased, coronary blood flow increases (28, 204). In isolated vessel studies perfusion pressure, i. e. intraluminal pressure can also be controlled adequately (4.1.2.). Increases in intraluminal pressure in isolated small arteries and arterioles in several vascular beds induce active constriction (myogenic response) conversly, decreases in pressure induce vasodilation (3.2.7., 99, 137, 139, 162, 194, 212, 228). 3.2.1.2. Biomechanical wall characteristics Passive and active biomechanical characteristics of the vessel wall determine its hemodynamic functions (96, 176). For a thorough description of the elastic behavior of the vessel wall, circumferential wall stress, vascular compliance, distensibility, elastic modulus, and active strain have to be examined. Elastic properties of a vessel segment depend on location, morphological caliber, intraluminal pressure, as well as on the contractile state of the smooth muscle. Such measurements can be made in vitro on isolated and cannulated segments (1, 13, 14, 20, 33, 48, 49, 60, 63, 81, 149, 166, 167, 168, 179, 207, 233, 237), on vascular ring studies (187) or even in in vivo conditions (215, 223). Biomechanical parameters as a function of location (e. g. cardiac and extracardiac), transmural pressure, and contractile state of smooth muscle are available both for larger (14, 20, 33, 48, 49, 60, 166, 167, 179), smaller arteries (1, 13, 14, 237) and on resistance-sized arteries and arterioles (63, 81, 149, 207, 233). Specificly, biomechanical studies were performed e. g. on rat, dog, rabbit carotid arteries (48, 60), dog carotid, iliac or splenic arteries (166, 167), on rat carotid artery and aorta (14, 33), dog carotid and vertebral artery (20), human umbilical artery (179), on rat tail artery (13, 14), rat saphenous artery (1, 237) and on hamster cheek pouch (60), rat coronary (207, 233), porcine coronary (81), rat mesenteric (149) small arteries and arterioles. Biomechanical wall characteristics basicly characterize elastic behavior of the vessel wall as function of intraluminal pressure. However, in order to measure elastic and contractile behavior of vessel wall, preconditioning pressure cycles must be applied in order to eliminate inhomogenous smooth muscle constrictions (4.1.6., 233). Different parameters that characterize vascular wall behavior (listed above) can be measured and further calculated from pressure-diameter relations and dimensions (e.g. wall thickness)

- 21 -

of blood vessels. In the Methods section, calculations of outer, inner radius, wall thickness, vascular incremental distensibility, elastic modulus, circumferential wall stress are demonstrated (4.1.7, 20, 33, 48, 49, 166, 168). In addition, vascular compliance, that characterizes vessel wall elasticity, is defined by the slope of pressure (P)-volume (V) relation at a given pressure (∆V/∆P). Volume relation can be calculated also as cross sectional area (81, 96, 168). Dividing this value by the initial volume or cross sectional area values gives distensibility (96, 166, 168). Average circumferential (tangential) wall stress increases with increasing in pressure and vessel size (20, 48, 49, 149, 167, 168, 233). Wall stress or tension can be also measured on isolated ring studies (187). Several studies demonstrate wall stressstrain (relative increase in radius) relationship (60, 149, 168, 207). Incremental distensibility decreases with increasing intraluminal pressure (20, 149, 233, 237). However, incremental elastic moduli showed an increase with increasing intraluminal pressure (13, 14, 20, 33, 48, 49, 60, 149, 166, 167, 179, 207, 233, 237). Contractile characteristics of vascular wall, such as spontaneous myogenic tone and agonist-induced active tone (active strain) also characterizes vascular wall. Spontaneous and active tones are mostly calculated as the percent contraction of the diameter from the passive diameter (168, 4.1.7.). Spontaneous tone is developed in isolated vessels from most of the vascular beds as an indicator of viability of the preparations (51, 63, 137). Spontaneous tone generally increases with decreasing vessel size (51, 105, 227, 228, 233). Agonist-induced active tone as a function of intraluminal pressure also characterizes vessel wall behavior (20, 187, 233). In case of coronary small arteries, mathematical models have been applied to describe the supposed in vivo mechanical characteristics of the vessel wall (30, 249). Some direct measurements are also available for porcine subepicardial and subendocardial and rat septal small arteries (78, 81, 136, 137, 207, 212). Intramural coronary small arteries are exposed to a very complicated pattern of transmural pressure changes during the cardiac cycle (30, 96). Therefore, it can be supposed that the mechanical properties of their wall will play an important role in shaping the actual values of the lumen diameter and hydrodynamic resistance. Pourageaud and Freslon (207) measured elastic properties of right interventricular rat coronary arteries. As a result, they did not found changes in elastic modulus as a function of wall stress in

- 22 -

different contractile states. A decreased elastic modulus in response to NO-synthase inhibition was described when it was plotted as a function of strain. On rat intramural small coronary arteries, Szekeres et al. (233) found, that elastic and contractile properties are altering with morphological caliber with the position along the coronary microcirculatory branching system. 3.2.1.3. Shear stress Dynamic mechanical forces and heterogenous distribution of blood flow at different circulatory levels elicit variable local shear forces onto the blood vessel wall, which further regulates vascular tone (11). Thus, besides myogenic and metablic regulatory mechanisms, flow-dependent regulation has a significant importance in determining vascular diameter as a local regulatory mechanism. It has been demonstrated in several vascular beds, that increasing blood flow through inducing increasing mechanical forces (shear stress) onto the blood vessels induces vasodilation. In the last two decades since the recognition of the role of endothelium in the control of vascular tone, new evidence has developed on the importance of the flow-dependent vascular reactivity with the involvment of the endothelium (133). Flow-induced vasodilation has been observed in several vascular beds, such as in the skeletal muscle (128, 129, 133, 230) and coronary circulation (138, 139, 141, 173, 174, 206), which exists both on large vessels (11, 97) and on small, resistance-sized vessels (128, 129, 133, 138, 139, 141, 173, 174, 206, 230). Local flow-induced mechanical forces primarily act on the endothelial layer of blood vessels causing vasodilation. Endothelial cells must act as “flow sensors” and must transduce the hydrodinamically evoked signals (11). Changes in flow may elicit mechanotransduction probably by conformational change in macromolecules of the extracellular matrix such as glycosaminoglycans (23), and an integrin signalling is proposed (174). In response to flow several intracellular signalling mechanisms can be activated (133). Initiating Ca2+ influx into the endothelial cells, endothelium-derived nitric oxide and prostaglandins can be released which in turn relax vascular smooth muscle (11, 133). It is also supposed, that the response of an artery to intraluminal flow is a result of an interaction of constrictor and dilator influences. (23).

- 23 -

A greater significance of flow-induced responses was suggested in the microcirculation, and it has been demonstrated in the recent years (133, 138, 141). In vivo studies showed that an increase in blood flow velocity elicit vasodilation in the skeletal muscle microcirculation (128), which is dominantly mediated by endotheliumderived prostaglandins (129). In isolated gracilis arterioles, both endothelium-derived nitric oxide and prostaglandins mediate flow-induced dilation (130). In case of increased hemodynamic load, e.g. in exercise training (3.4.) increased blood flow is supposed to be associated with increased flow shear stress that induces endothelial nitric oxide synthesis (146, 185, 219, 229, 231, 250). Endothelium-dependent, flow-induced vasodilation was observed in the coronary microcirculation e.g. on porcine (138, 139, 141, 173, 174) and rat coronary arteries (206), which has been demonstrated to be mediated by endothelium-derived nitric oxide (12, 139). However, vasodilatory responses to flow are heterogenous within the coronary microcirculation (141, 173). Kuo et al. (141) found, that large coronary arterioles around 100 µm are the most sensitive to increases in flow compared to even smaller or larger vessels. This suggests, that in the coronary microcirculation, flowinduced responses are the dominant in the intermediate-sized resistance vessels, whereas smaller vessels are more susceptible to metabolic and pressure-induced mechanisms (141, 147, 173). However, in the microcirculation, flow-induced vasodilation is demonstrated to be opposed by neurogenic and pressure-induced myogenic constriction (133, 139, 173, 230), which suggests, that the microvascular tone in vivo is adjusted by a complicated interaction of several vasoactive control mechanisms (3.2.10.). 3.2.2. Mechanical effects of cardiac cycle Mechanical effects of the myocardium during the cardiac cycle substantially affects coronary circulation by extravascular compression of blood vessels during systole. Thus, the transmural pressure across the vascular wall is determined by the origin of extravascular and intravascular pressures (67, 96). Systolic myocardial contraction squeezes out the blood contained in coronary vessels, enhances coronary venous outflow, and thereby promotes subsequent arterial inflow. Furthermore, the extravascular mechanical effects may depend upon location, systolic flow inhibition or

- 24 -

reversal of flow is more pronounced in the inner myocardial layer, i.e in intramural vessels (11, 28, 96). In order to explain the effects of cardiac contraction on the coronary blood flow, the vascular waterfall model was proposed. According to this model, blood vessels are compressed by the intramyocardial pressure. Thus, intramyocardial circulation has a characteristics of a collapsible tube compressed by intramyocardial tissue pressure, which acts as instantaneous effective outflow pressure of the coronary circulation when it exceeds coronary venous pressure (11, 28). On the basis of this observation, increases in heart rate or contractility enhances coronary vascular resistance (11). Extravascular compression is not uniformly distributed across the left ventricular wall, it decreases from subendocardium to subepicardium eliciting an inhomogeneity in the transmural distribution of myocardial blood flow (11, 67, 93, 96). Under physiological conditions with intact coronary autoregulatory reserve, a uniform transmural myocardial blood flow distribution is maintained. However, in the presence of coronary stenosis with compromised autoregulatory functions, extravascular compression is of major importance for the transmurally nonuniform, preferentially subendocardial manifestation of myocardial ischemia (11). 3.2.3. Metabolic control mechanisms Local metabolic control is thought to be the most important mechanism by which increases in myocardial metabolic demand and oxygen consumption are matched by increases in coronary blood flow (67, 173). Although an increase in metabolic activity undoubtedly produces an increase in vasoactive metabolites at all levels of coronary vasculature, metabolic control mechanisms appear to be prominent in the coronary microcirculation, furthermore, in the smallest coronary arterioles (141, 147, 173, 193). A number of factors, including metabolites are released from the working myocardium which act to provide an appropriate arteriolar tone (11). During increased metabolic demand due to an increased cardiac performance (e.g. exercise) or in hypoxia and ischemia caused by altered coronary perfusion, local, vasodilator metabolites are released from the myocardial cells and from the coronary vasculature. Consequently, increased metabolic demand induces a reactive vasodilation in coronary circulation, and

- 25 -

thus myocardial blood supply can be adjusted to the increased demand in the intact circulatory bed (11, 67, 112). Adenosine has a significant contribution to coronary blood flow adaptations during rapid changes in cardiac performance and to hypoxic and ischemic vasodilation (11, 67, 193). Adenosine is formed in the myocardial cells with increasing myocardial metabolism and oxygen demand, when energy consumption is increased with ATP utilization. Diffusing to blood vessels it causes a substantial vasodilation (11, 67, 193, 3.2.8.1.). Previous studies suggest, that small coronary arterioles are the primary target for the effects of adenosine (11, 141, 173). When myocardial metabolic demand increases, due to a relative underperfusion, the developed hypoxia, hypercapnia and acidosis greatly contributes to the reactive vasodilation. Due to the anaerobic metabolism, lactic acid is released and acidosis develops (12, 67, 87, 193). Other metabolic factors, like prostaglandins and bradykinin may be also released from cardiac and vascular cells due to a relative hypoxia contributing to the reactive vasodilation. Osmolarity and ions also contribute to the local metabolic regulation, most importantly potassium ion (67). Activation of ATP-sensitive K+ channels may be important in the metabolic regulation of microvascular resistance (173). Thus, changes in metabolic activity of the heart will result in changes in levels of vasoactive metabolites released from the myocardial and vascular tissue, and thus, through altered coronary vasomotor tone, myocardial blood flow is consequently altered. When metabolic demand of the heart increases, vasodilator metabolites are released, preferentially adenosine, which chiefly act on small resistance vessels, causing in turn a reactive vasodilation (141, 173, 193). 3.2.4. Neural and humoral control mechanisms Nervous stimulation to the heart can affect coronary blood flow both directly and indirectly. The direct effects result from the direct action of the neurotransmitter substances on the coronary vessels, acetylcholine from vagus nerves and norepinephrine from the sympathetic nerves, these fibers are divisions of the autonomic nervous system. Nerves originating from the cardiovascular centers in the central nervous

- 26 -

system determine important cardiovascular reflexes (e.g. baroreceptor reflex). Furthermore, other brain areas (e.g. hypothalamus) have great impact in the control of coronary vascular tone (67, 87, 104, 112, 236). The transmitters are released from the perivascular nerves in the vessel wall. The indirect effects result from secondary changes in coronary blood flow caused by increased or decreased activity of the heart (87). The indirect effect play a far more important role in the control of coronary blood flow. For example, sympathetic stimulation increases heart rate and contractility, and thus the rate of metabolism, which, in turn, by increased metabolic demand of the heart (3.2.3.) increases coronary blood flow (67, 112). Both sympathetic and vagal vasomotion may occur at the level of epicardial coronary arteries, in the resistive microcirculation and in the collateral microcirculation (11, 16, 67, 112). Parasympathetic (vagal) stimulation has only a slight direct, and rather an endothelium-dependent effect to dilate coronary vessels (87). The transmitter acetylcholine is released and may elicit vasodilation or vasoconstriction depending on the function of the endothelium or the relative distribution of muscarinic receptors on the endothelial and smooth muscle cells (3.2.8.2.). However, in the coronary circulation of mammals and humans, cholinergic constriction may operate (117). The possibility of its overhelming constricting effect would be in accord with the proposed role for acetylcholine in some types of coronary spasm (193, 3.2.8.2.). There is much more extensive sympathetic innervation of the coronary vessels (87). Sympathetic neurotransmitters norepinephrine and epinephrine can elicit either vascular constrictor or dilator effect, depending on the presence of the different types of receptors. Both α and β receptors exist on coronary vessels with the predominance of β receptors, from which α causes vasoconstriction and β vasodilation. Although αvasoconstrictor effects play a major role in the control of coronary vascular tone in normal conditions as balanced with dilator mechanisms (i.e. metabolic effects, 67, 110, 112, 164, 236), overhelming α-mechanisms seem to be a major predisposing factor eliciting vasospastic angina (16, 87, 193), and its role is increased in myocardial ischemia (111, 112). Subtypes of α and β receptors may have slightly diverging effects. α1 and α2 receptors located on the coronary vascular smooth muscle cells cause vasocontriction. However, α2 receptors located on the presynaptic nerve terminals diminish the release of the neurotransmitter, norepinephrine. Furthermore, α2 receptors

- 27 -

located on endothelial cells are responsible for the release of nitric oxide thus eliciting vasodilation. Generally it is proposed, that α1 and α2 receptor subtypes are nonuniformly distributed among coronary vessels with different sizes. In large epicardial arteries, α1 receptor subtype is dominant, whereas in coronary resistance vessels α2 receptor subtype. Similarly, β1 receptors are seem to be dominant in large epicardial arteries, whereas β2 receptors are dominant in the resistance vessels (11, 193, 3.2.8.6.). Other neurotransmitters exist in the coronary circulation, e.g. neuropeptide-Y, which is an adrenergic cotransmitter exerting vasoconstriction, thus it may also play a role in the development of coronary spasm (193). To summarize these extremely complex control mechanisms, the chief regulatory factor appears to be metabolic vasodilation opposed by vasoconstrictor αadrenergic activity along with vasodilator β-stimulation and cholinergic activity (11, 67, 112, 164, 193). A number of circulating hormones can modify coronary vasomotor tone. Angiotensin II existing in the circulation can be also produced by local reninangiotensin system (RAS). Locally released kinins like bradykinin evidently play a role in the control of coronary circulation (3.2.8.3.). Potentially also involved are circulating norepinephrine and epinephrine, vasopressin, atrial natriuretic peptide (ANP), vasoactive intestinal peptide (VIP), neuropeptide-Y and endothelin (11). Although vasopressin, angiotensin II, Neuropeptide-Y and catecholamines are mostly strong vasoconstrictors, vasopressin, angiotensin II, ANP, VIP, bradykinin, catecholamines may elicit a dual effect by their action both on endothelium (vasodilation) and vascular smooth muscle (vasoconstriction), as summarized in Fig. 3.-2. (11). 3.2.5. Endothelium-mediated control mechanisms The decisive role of the endothelium in vasomotor control became clear when Furchgott and Zawadzki (77) demonstrated endothelium dependency of the relaxation to acetylcholine in isolated arteries, and the new vasorelaxing mediator was called endothelium-derived relaxing factor (EDRF). Endothelium-dependent relaxation of blood vessels is produced by a large number of agents (e. g. acetylcholine, ATP, ADP, substance P, bradykinin, histamin, serotonin), which is mostly limited to certain species

- 28 -

and vascular beds. Relaxation results from release of a labile EDRF, called nitric oxide (NO). NO stimulates guanylate cyclase of the vascular smooth muscle, with the resulting increase in cyclic GMP activating relaxation. Also there is an evidence for other

relaxing

factors

including

prostaglandin

I2

and

endothelium-derived

hyperpolarizing factor (EDHF) (9, 11, 12, 34, 76, 86, 114, 144, 222, 247). Flow-induced shear stress also stimulates release of endothelium-derived vasodilators (11, 76, 128, 130, 138, 139). Endothelium-dependent relaxation occurs in large arteries and in resistance vessels as well, it is generally more pronounced in arteries than in veins (169). Relaxing factors also inhibit platelet adhesion and aggregation. However, endothelium-derived contracting factors appear to be responsible for endotheliumdependent contractions produced by some products of arachidonic acid metabolic pathway (constrictor prostaglandins) or by endothelin (76, 157). Several agonists (e.g. acetylcholine, bradykinin, histamin, serotonin, substance P, ATP) acting on endothelial receptors may release endothelium-derived relaxing factors causing vascular relaxation (11, 76, 247, Fig. 3.-3.). However, by their direct action on smooth muscle they may cause vasoconstriction. The resulting vascular effect depending on the balance of the dilator and constrictor functions. In case of endothelial damage vasoconstricton may occur by the stimulation of otherwise dilator agonists (e.g. acetylcholine, 11, 247, Fig. 3.-2.). Thus, responses to endothelium-dependent vasodilators may be used as an index of functional integrity for appropriate autacoid release from endothelial lining of the coronary bed (11). However, in some species and vascular beds vasodilators may cause vasoconstriction with intact endothelial function. For example, acetylcholine causes vasodilation in rabbit, dog, guinea pig coronary arteries (5, 54, 253), whereas it causes vasoconstriction porcine coronary arteries (137, 139, 212). Furthermore, acetylcholine may have a diverse action on rat and human coronary arteries, it may cause vasoconstriction even with intact endothelium (8, 11, 55, 64, 72, 79, 183, 205, 206, 244, 253, 254). Endothelium-dependent vasodilation may be impaired in several cardiovascular pathologies, such as in coronary artery diseases, atherosclerosis, diabetes, hypertension and microvascular angina, due to the decreased NO production as a result of endothelial cell damage. Thus the overhelming constrictor effects may induce vasospasm. Furthermore, the diminished NO production may accelerate coronary artery disease by promoting interactions between platelets and the

- 29 -

vessel wall, and thus thrombus formation and atherosclerosis (11, 59, 64, 65, 150, 160, 170, 247, 255). 3.2.5.1. EDRF and NO 3.2.5.1.1. Background In early experiments it was demonstrated on isolated rabbit aorta strips, that endothelium-dependent relaxation by acetylcholine results from the release of a diffusable relaxing substance, later termed as EDRF (76, 77). Within a few years the release of EDRF has also been demonstrated by using superfused endothelial cells and the EDRF released from these cells is bioassayed by the endothelium-free preparation of artery downstream. The preparation used dowstream to bioassay-released EDRF is either a superfused ring or strip of artery attached to a force transducer or a perfused segment of artery whose diameter or resistance to flow is continuously monitored (76, 216). In early work on rabbit and dog arteries, endothelial cells were found to be required for most of the relaxation induced by ATP, ADP, and all of the relaxation induced by substance P and calcium ionophore A23187 (76). A large number of additional agents that produce endothelium-dependent relaxation of arteries have been found, although, depending on the agent, the endothelium-dependent relaxation may be limited to certain species and certain vascular beds. Later, endothelium-derived relaxing factor has been identified as nitric oxide (76, 102, 195). The initial evidence was obtained in a study of the characteristics of the transient relaxation of rabbit aorta produced by a factor that is generated in NaNO2 solutions on acidification. This factor has been shown to be NO, which is generated as an intermediate during the decomposition of nitrous acid. It was found, that NO is a simple radical gas, a shortlived labile molecule, which forms covalent bonds fairly easy. NO is not stored, but diffuses freely from its site of formation being soluble both in water and lipid. The reactivity of NO is related to its redox properties forming redox couples. Vasorelaxing effect of NO is blocked by hemoglobin, by generators of oxygen, and it is markedly potentiated by superoxide-dismutase (SOD, 6, 76). Ignarro et al. (102) have presented both biological and chemical evidence that EDRF released from endothelial cells in

- 30 -

both bovine pulmonary artery and veins is NO. The strongest evidence that EDRF is NO has been obtained by the Moncada group (195), who not only showed that infused NO and EDRF released by bradykinin from cultured porcine aortic endothelial cells possess similar characteristics in various tests that used endothelium-free strips of rabbit aorta for bioassay, but also that the amount of NO determined by chemoluminescence released by bradykinin could quantitatively account for the relaxation of the bioassay strip produced by EDRF liberated by the peptide. They also reported, that the source of NO released from cultured porcine aortic endothelial cells is L-arginine. They also demonstrated, that inhibition of endothelium-dependent relaxation of rabbit aorta by Nmonomethyl-L-arginine

decreased

NO-release.

NO

can

be

detected

by

chemiluminescence method (5), by measurement of its metabolic end-products (nitrite and nitrate, 11) or directly with an electrochemical NO sensor (246). 3.2.5.1.2. Biosynthesis and intracellular signalling of NO Biosynthesis of NO: Moncada et al. (165) have found, that the precursor of NO synthesis is Larginine. Endothelial cells kept in the absence of L-arginine had lost their ability to synthetize NO, however, addition of L-arginine induced NO synthesis. NO is synthetized in several cells and tissues in the organism, for example, in vascular endothelial cells, myocytes, leukocytes and neurons (3.2.5.1.3.). In vascular endothelial cells NO synthesis can be induced by vasoactive agonists and mechanical effects (Fig. 3.-3., 11, 76): 1. neurotransmitters: acetylcholine, substance P, ATP, norepinephrine, VIP 2. local mediators: bradykinin, arachidonic acid, histamine, substance P 3. hormones: catecholamines, vasopressin, oxytocin, angiotensin II, ANP 4. hemostasis factors and platelet mediators: thrombin, adenine nucleotides, serotonin, platelet-derived growth factor 5. mechanical or hydrodynamic effects: flow-induced shear stress NO biosynthesis is well described. When agonist binds to its membrane receptor, through activating G-proteins, by inositol triphosphate (IP3)-mediated release of Ca2+ from intracellular stores and by opening membrane Ca2+ channels intracellular

- 31 -

Ca2+ level increases. Activation of thyrosine and mitogen-activated protein (MAP)kinases may be involved in the agonist-induced Ca2+-influx and Ca2+-signalling (127, 68). Ca2+ bound to calmodulin in turn activates nitric oxide synthase (NOS) enzyme, which catalyzes the hydroxylation of the nitrogen in the guanidino group of L-arginine (Fig. 3.-4.). The process incorporates molecular oxygen into NO and citrulline. Then, bioproduct of the reaction, citrulline is recycled back to L-arginine incorporating one nitrogen (6, 11, 59, 175). Isoforms of NOS: Several isoforms of NO synthase have been identified. There are two main classes: constitutive NOS (cNOS) is present in the endothelium and neural tissue and the inducible enzyme (iNOS) formed in activated immune cells and vascular smooth muscle cells (6, 73, 90). Three main NOS isoforms have been characterized. All isoforms utilize L-arginine as a substrate and require the cofactors reduced nicotinamide-dinucleotide phosphate (NADPH), tetrahydrobiopterin, flavin adenine dinucleotide and flavin mononucleotide (71, 175). 1. Isoform I (NOS1), which was originally classified as constitutive NOS, was purified from the brain. It is present in certain neurons, mostly in non-cholinergic nonadrenergic nerves (3.2.5.1.3.). Thus is called neuronal NOS (nNOS), although it is present in certain epithelial cells, human skeletal muscle, kidney macula densa cells, in pancreatic islet cells and in endothelial cells of brain vessels (71, 124, 155). Isoform I is a Ca2+ and calmodulin-dependent enzyme (59). 2. Isoform II of NO synthase (NOS2), which is called as iNOS can be induced in macrophages (and other cells) with bacterial endotoxin and cytokines. When induced in macrophages, iNOS produces large amounts of NO that represent a major cytotoxic activity (71). The activity of iNOS is not regulated by Ca2+, however, the amino acid sequence of the enzyme demonstrated a binding site for calmodulin (71). iNOS has been demonstrated to present also in vascular smooth muscle cells, in cardiac myocytes, in endocardial endothelium, fibroblasts and microvascular endothelium (90, 124, 218). 3. Isoform III NOS (NOS3) also called as constitutive endothelial NOS (eNOS) is present in endothelial cells. Like NOS1, NOS3 is also regulated by Ca2+ and

- 32 -

calmodulin. Although the eNOS isoform was originally characterized in large conduit vessel endothelium, it is now known to be expressed within the heart in the endothelium both of the endocardium and of the coronary vasculature, in cardiac myocytes as well as in some blood elements (e.g. monocytes and platelets, 71, 90, 124, 218). Signalling effects of NO: NO relaxes blood vessels by binding to iron in the heme at the active site of guanylate cyclase, thereby activating the enzyme to generate cGMP (Fig. 3.-4., 59, 177). The increase in cGMP is associated with protein kinase activation and altered phosphorilation of numerous endogenous smooth muscle proteins. cGMP can also decrease the activity of phospholipase C thus decreasing formation of inositol phosphates, which also appears to be mediated by a cGMP-dependent protein kinase. These effects probably lower cytosolic free calcium concentrations, resulting in a decreased phosphorilation of myosin light chain and relaxation. Other target sites for cGMP-regulated effects in smooth muscle could be membrane transport of calcium and other cations and protein phosphatase regulation (59, 177). Futhermore, NO exerts its effects by its reaction with SH (thiol) groups of proteins and with proteins that contain metal ions. Essential components of NO response therefore include ion channel proteins, enzymes, surface receptors, and transcription factors - all of which contain either transition metals or thiol strategically located at the allosteric and reactive sites (73). NO may also act independent of cGMP pathway through Ca2+-activated K+channels (26) and also by enhancing ADP ribosylation of platelet proteins by inhibiting glyceraldehide-3-phosphate-dehydrogenase (GAPDH) activity (59). Blocking NO synthesis: After NO was discovered, it was a substantial step to discover inhibitors of NO biosynthesis. Previously several inhibitors were used, which act on different levels of NO biosynthesis and action, e.g. hemoglobin, methylene blue, gossypol (11, 17, 54, 90, 196), from which most of them appeared to be nonspecific.

- 33 -

Up today, several L-arginine analogues have been used, which act by competing with L-arginine at the active site of NOS, thus they specifically block NO biosynthesis, which can be reversed by L-arginine (6, 11, 90). These NOS inhibitors have been essential for evaluating the role of NO in physiological and pathophysiological processes. Thus, with blocking NO synthesis we may assume, that whether the effects of vasoactive substances are mediated by NO, or what is the extent of NO mediation in their effect. For example if the effect of a vasoactive substance (e.g. bradykinin) is mediated by NO, the vasodilation induced by the substance will be decreased or eliminated after blocking NO synthesis. Furthermore, infusion of NOS inhibitors increases blood pressure and peripheral resistance indicating that vascular tone is influenced by a continuous release of NO (6). Up today, several L-arginine analogues are applyed to block NO synthesis: 1. NG-nitro-L-arginine (L-NNA or L-NOARG), which we applyed in our studies (see 4., 29, 79, 90, 126, 175), 2. NG-nitro-L-arginine-methylester (L-NAME, 6, 90, 226, 244), 3. NG-monomethyl-L-arginine (L-NMMA, 6, 11, 47, 85, 90, 226, 244), 4. N-iminoethyl-l-ornithin (L-NIO, 226), 5. L-NG-aminoarginine (85). These analogues are described to inhibit NO synthesis on both the constitutive and inducible isoforms of NOS (90). Up to now specific NOS inhibitors have been discovered. For example, inducible NOS can be inhibited by mercaptoethylguanidine (156), or by cyclohexamide (90). 3.2.5.1.3. Physiological and pathophysiological role of NO NO is formed by a variety of cells with different biological actions (165). This signalling molecule appears to function primarily in three broad categories (73): 1. at the endothelial cell to cause relaxation 2. during neurotransmission to facilitate central nervous system function 3. in cell-mediated immune responses to facilitate immunologic function Furthermore, NO plays a role in the vascular and secretory functions of different organs, e.g. in the lung, kidney, pancreas and gastrointestinal system.

- 34 -

NO plays a substantial role in the cardiovascular system by its action on blood vessels and heart. NO, synthetized by endothelial NOS, is released from endothelial cells and induces vasodilation, inhibits platelet aggregation and adhesion (6, 11, 73, 76, 124). NO produces vasodilation to a variety of agonists, a variety of locally released mediators, metabolites, circulating hormones, e.g. bradykinin, acetylcholine, substance P, serotonin, histamin, thrombin, ATP, ANP, VIP, norepinephrine, angiotensin II, etc., however, the responses to agonists show a great heterogeneity among tissues and species (6, 11, 76, 247). Thus, NO plays a substantial role in adjusting vascular tone throughout the circulatory system and in the redistribution of blood flow to changes in hemodynamic load (e.g. in exercise, or in circulatory shock). With intact endothelial function, NO keeps blood vessels in a relative vasodilatory state, thus it also controls adequate tissue perfusion and blood pressure. This controling effect may be exerted by local metabolic, neural, humoral control mechanisms, or by flow-induced shear stress. In diseases which affect the circulatory system, e.g. in atherosclerosis, hypertension, diabetes, due to the damaged endothelial layer, and thus a diminished NO synthesis, dilatory agonists on their direct effects on the smooth muscle may cause vasoconstriction, and vasospasm. Furthermore, abnormal NO genes may play a role in the etiology of atherosclerosis and hypertension (11, 59, 76, 140, 193, 247, 252). Thus, the sclerotic narrowing of the lumen, a higher vascular tone and the prevalence of vasospasm predispose to increased blood pressure with stroke events and myocardial ischemic diseases. Diminished NO production may accelerate coronary artery disease by promoting interactions between platelets and vessel wall. In diseased vessels, products of activated platelets such as thromboxane, serotonin, adenine nucleotides and platelet-derived growth factor may lead to vasoconstriction and proliferation of smooth muscle, which accelerates the coronary artery disease and coronary spasm (59, 252). In experimental

conditions,

applying

NO

synthase

inhibitors

may

also

cause

vasoconstriction and spasm (6, 11, 59, 76, 125, 193, 204, 205, 244, 247, 252). In the clinical therapy, NO has great importance due to its similarity to nitrovasodilators. It was proved, that these drugs are metabolized in the target cells to a nitrosothiol and/or NO. Both NO and nitrovasodilators act through the stimulation of guanylate cyclase and subsequent formation of cyclic GMP (6, 160, 177). Furthermore, nitrates and other NO donors can substitute for endogenous NO in the case of diseased

- 35 -

endothelium or when due to other pathologies when NO production is decreased (2, 224, 248). Till now nitrates are widely used as a primarily important therapeutic agent in several cardiovascular pathologies, most importantly in coronary heart diseases (252). Nitric oxide has a significant impact on the cardiac function. The presence of cNOS and iNOS have been verified within the myocardium. cNOS is expressed in myocytes, endocardium, endothelial cells and neurons in the myocardium, and there is an evidence for iNOS in the myocytes, endocardium, small vessel endothelium, vascular smooth muscle cells and immune cells in the myocardium. NO influences myocardial inotropic and chronotropic responses by either constitutive and inducible isoforms of NO synthase (90, 124, 218). NO donors in some species and preparations may exert a negative inotropic effect as well as an enhancement of diastolic relaxation. However, excessive NO production induced by e.g. cytokines may contribute to the myocardial depression associated with conditions such as sepsis, myocarditis, cardiac transplant rejection, dilated cardiomyopathy (6, 59, 90, 124, 218). Thus, it is suggested, that NO influences normal cardiac physiology, and may play an important role in the pathophysiology of certain disease states associated with cardiac dysfuntion (90, 124). Meanwhile, nitric oxide and nitric oxide donors have been demonstrated to exert cardioprotective actions in myocardial ischemia and reperfusion mostly due to a reversal of impaired endothelium-dependent vasodilation and direct cytoprotective effect on myocardial cells (150, 3.3.4.). NO has been proved to play a significant role as a neurotransmitter or a cotransmitter in the central and peripheral nervous system. In neurons of the brain, NO is synthetized by nNOS, it may be a modulator of neurotransmission, and it probably plays a role in the autonomic neurotransmission as well (6, 11, 59). In the autonomic nervous system, NO may be the non-cholinergic, non-adrenergic (NCNA), actually called “nitroxidergic” neurotransmitter or coreleased factor in various vascular beds, in the contactile and secretory tissue (6, 59, 71, 73). NO plays a role as a mediator in the immunoreactions. Local, inflammatory mediators, e.g. bradykinin, histamine induce NO synthesis in endothelial cells during inflammation, and thus cause vasodilation (76). NO is also produced by macrophages and leukocytes by iNOS induced by cytokines and bacterial toxins exerting bactericidal,

- 36 -

microbicidal and tumoricidal actions, but it may lead to cytotoxicity in excessive amount (11, 59, 71, 73). The actions of NO, however may be controversial. Whereas it has important and beneficial effect in the maintenance of tissue perfusion by a vasodilatory state, in the nervous transmission and in the activity of immune system (etc.), excessive NO levels may mediate other cardiovascular pathologies. Thus, although diminished NO synthesis may be relevant to abnormalities in hypertension, coronary artery disease and hypoxia, excessive, induced NO synthesis may induce cytotoxicity, further aggravate hypotension e.g. in septic shock (6, 59, 71, 73, 252). 3.2.5.1.4. Role of NO in the control of coronary vascular tone Several studies have been addressed to examine the significance of NO along with other endothelium-derived mediators in the coronary circulation. Modulation of coronary tone by the release of endothelial autacoids was first analyzed and understood in large epicardial coronary arteries, and later investigations pointed to the coronary microvasculature as well. Previous studies showed an enhanced coronary vascular tone by blocking NO synthesis on isolated vessel (205, 243) and isolated heart studies (29, 125, 204), or in vivo studies (19, 101), which indicates a continuous, basal release of NO in the coronary vessels. Thus, it is obvious that NO plays an important role in regulating coronary vascular tone (12, 125, 173, 204, 205, 243). Agonist-induced vasoactive responses may be (139) or may not be mediated by NO (79). Most of the previous studies show that endothelium-dependent vasodilators: bradykinin, acetylcholine, substance P elicit vasodilation on coronary resistance arteries in which endothelium-dependent nitric oxide plays a substantial role (141, 173). Although in large coronary arteries substance P-induced vasodilation depend on endothelium or NO (211, 239), in rat coronary resistance arteries it may not be mediated by NO (79, 136, 3.2.8.4.). Bradykinin elicits a dose-dependent vasodilation on small coronary arteries of several animals and humans that is at least partially mediated (47, 126, 205, 243) or not mediated by NO (79, 3.2.8.3). Acetylcholine may induce vasodilation or vasoconstriction on coronary arteries of different species with intact endothelium (8, 46, 54, 55, 79, 108, 139, 144, 163, 183, 205, 209, 243, 244, 3.2.8.2).

- 37 -

Acetylcholine-induced vasodilation may be partially, fully, or may not be mediated by NO (54, 79, 211, 244). Thus, it can be assumed according to the recent observations, that endotheliumderived nitric oxide plays a substantial role in the control of coronary vascular tone. 3.2.5.2. Prostaglandins (PG-s) The eicosanoids are derivatives of arachidonic acid which are formed in platelets and various parts of the vascular wall or in other tissues (11, 73). Eicosanoids include prostaglandins (PG-s), thromboxanes (TX-s), hydroxy or hydroperoxy fatty acids and leukotrienes. Prostaglandins and thromboxanes are the products formed following cyclooxygenase rearrangement of arachidonic acid. The other group called lipooxygenase products, since leukotriene formation is catalyzed by lipooxygenases. These two families are compounds of hormone-like molecules, they have short-lived influence and they are effective at very low concentrations (73). This study focuses on the prostaglandin-thromboxane system emphasizing their possible vascular actions. Among their vascular actions, prostaglandins are involved in several physiological and pathophysiological processes (73): 1. The inflammatory response and the development of inflammatory abnormalities such as rheumatoid arthritis and psoriasis. 2. Production of pain and fever. 3. Induction of blood coagulation (by augmented platelet aggregation). 4. Regulation of blood pressure by affecting vasodilation and relaxation of smooth muscle cells and small arteries and veins. 5. Regulation of uterine contraction during labor. 6. Regulation of sleep and wake cycles. 3.2.5.2.1. PG biosynthesis The essential prerequisite for PG formation is the availibility of its precursor fatty acid, arachidonic acid in the plasma membrane. The first step of PG biosynthesis is catalyzed by cyclooxygenase, which can be blocked by substances like indomethacin,

- 38 -

aspirin, ibuprofen or diclofenac (11, 129, 130, 154). As a result of further enzymatic processes, different types of prostaglandins such as PGE2, PGI2 (called prostacyclin), PGF2α, PGD2, PGH2 are synthetized. From PGH2 TXA2 is further synthetized (73). PGI2 is mainly synthetized by endothelial cells, it inhibits platelet aggregation and induces vasodilation, whereas TXA2 is mainly synthetized by platelets and promotes platelet aggregation and adhesion. Since PGI2 and TXA2 are very labile, their more stable metabolites are 6-keto-PGF1α and TXB2, respectively, which can be mostly detected in experimental preparations (11, 73, 80, 119). Among the physiologic stimuli of arachidonic acid release and PG biosynthesis -mostly of PGI2- are hormones and drugs such as angiotensin II, bradykinin, ATP, histamine, ACE inhibitors, nitrovasidilators, dipyridamole, nifedipine, NO-donors and catecholamines (11, 73). 3.2.5.2.2. Role of PG-s in the control of coronary vascular tone Prostaglandins and thromboxanes synthetized from arachidonic acid may exert divergent vascular actions. Different PG types listed in 3.2.5.2.1. exert different intracellular pathway mechanisms, however, their classification and mechanism of action is not completely understood. PGI2 and PGD2 act by increasing intracellular cAMP level and thus cause vasodilation. PGI2 is known to be a potent vasodilator synthetized by endothelial cells in the largest amount in response to several agonists (e.g. bradykinin, ACE inhibitors, ATP). Furthermore, PGI2 inhibits platelet adhesion and aggregation, thus inhibits thrombus formation and prevents atherosclerosis (11, 73). On the other hand, TXA2/PGH2 and PGF2α through IP3-DAG system cause vasoconstriction by activating protein kinase C and increasing intracellular Ca2+ level in the vascular smooth muscle cells. TXA2 activates and promotes aggregation of platelets further promoting vascular diseases in atherosclerosis (11, 45). In experimental conditions, PGF2α is used to preconstrict coronary arteries in order to test agonistinduced responses (143, 188, 209). However, recently, a TXA2/PGH2 agonist, U46619 is mostly used for this purpose (37, 47, 208, 209, 220). Furthermore, applying a specific TXA2/PGH2 antagonist, SQ 29,548 helps to differentiate between the proportion of constrictor and dilator prostaglandins in controlling vascular tone in different vascular beds (61, 100). PGE2 is known to be mostly a vasodilator, although it may exert diverse

- 39 -

actions by inducing both cAMP and IP3-DAG system (45). Thus, among eicosanoids, TXA2 and PGI2 are the most important factors in adjusting vascular tone and platelet functions. The effective balance between the counteracting effects of TXA2 and PGI2 is shifted more towards PGI2 in the intact vascular beds (11). The prostaglandin-thromboxane system plays a major role in the control of vascular tone in the peripheral circulation, e.g. in the skeletal muscle circulation (113, 128, 129, 130), however, its role in the coronary circulation is controversial. Most of the studies reveal that prostaglandins have no substantial or only a minor role in the control of coronary vascular tone under physiological conditions (11, 12, 95, 184, 189), but they may become important in certain pathologic conditions of hypoxia and increased platelet activation as may occur at the site of a severe coronary stenosis (11, 12, 119, 184, 189). It has also been shown that coronary vasculature syntethize prostaglandins, mostly PGI2, PGE2 and PGF2α , which may have divergent actions on coronary vascular resistance of different species (80, 119, 184). For example, PGE2, which is known as a dilator prostaglandin, enhanced coronary vascular resistance in rat heart (119). Inhibition of prostaglandin synthesis may not change (3, 113, 119, 189) or increase (95, 11) or in certain conditions even decrease (119) coronary vascular resistance. It is still proposed that prostaglandins might be involved in the autoregulation of coronary blood flow (3, 119). In agonist-induced coronary vascular responses PG-s may play a major role. Endothelium-derived PGI2 has been demonstrated to play a major role in the vasodilatory effect of several agonists, such as bradykinin, ATP, histamine and catecholamines (11, 86, 144, 154, 190). Although, blockade of PGI2 synthesis by indomethacin did not modulate bradykinin-induced vasodilation in porcine and rat coronary arterial preparations (79, 208, 3.2.8.3.). 3.2.5.3. Other endothelium-mediated control mechanisms An important endothelium-derived vasodilator besides from NO and PGI2 is endothelium-derived hyperpolarizing factor (EDHF). EDHF has been shown to hyperpolarize vascular smooth muscle cells via activation of K+ channels, and it contributes to the dilator effects of bradykinin in different vascular beds (15, 86, 126).

- 40 -

EDHF seems to be activated by an arachidonic acid metabolic pathway, as it has been demonstrated, that EDHF-type vasodilation has been inhibited by phospholipase A2 and cytochrome P450 inhibitors in the rat coronary circulation. Thus, EDHF also seems to play an important role in the control of coronary circulation (15). Besides endothelium-derived vasodilators, endothelium-derived constrictor factors may play a role in the regulation of vascular tone. The most important and potent constrictor is endothelin (ET-1). A continuous basal release of ET-1 may contribute to a constrictor resting tone, thus supplementing myogenic and α-adrenergic constrictor coronary tone (11, 203, 242). ET-1 exerts a strong vasoconstriction in all vessel types by increasing intracellular Ca2+ level through IP3-DAG activation. ET-1 causes a dosedependent

vasoconstriction

in

the

coronary

circulation

especially

in

the

microcirculation, that may induce severe myocardial ischemia and acute heart failure (11, 72, 203). 3.2.6. Direct smooth-muscle mediated control mechanisms Vascular smooth muscle cells possess several receptors that further activate second messenger pathway mechanisms and eliciting relaxation or constriction (Fig. 3.5.). Several agonists eliciting endothelium-mediated actions (e.g. bradykinin, acetylcholine, thrombin, serotonin, histamine) may directly act on the vascular smooth muscle causing mostly vasoconstriction (3.2.5.). However some agonists, e.g. sodiumnitroprusside (SNP) or other nitrates and NO-donors, and adenosine (ADO) act exclusively on the vascular smooth muscle cells independent of endothelium causing vasodilation. SNP and ADO have been demonstrated to cause prominent vasodilation in several vascular beds, such as in coronary arteries (136, 138, 139, 141, 174, 199, 212, 227, 228, 3.2.8.1., 3.2.8.5.). Their vascular relaxations can be exerted by activating cAMP (by catecholamines on α-adrenoceptors, prostaglandins, especially PGI2) and cGMP (by nitrates, NO donors, ANP) systems and K+ channels (by adenosine, bradykinin). On the other hand, vasoconstriction may be elicited by inhibiting cAMP system (by acetylcholine, catecholamines on α2 adrenoceptors and TXA2) and activating IP3-DAG system (by catecholamines on α1 adrenoceptors, serotonin, ATP, ET-1, angiotensin II, acetylcholine, some prostaglandins, 11, 45).

- 41 -

Among several types of K+ channels, ATP sensitive K+ channels (K+ATP) have been demonstrated to play a substantial role in the regulation of coronary vascular tone (134). K+ATP channels mediate vasodilation induced by agonists, e.g. adenosine, bradykinin or shear stress, and may participate in the effect of EDHF (15, 134, 178, 213). Furthermore, K+ATP channels play an important role in coronary microvascular vasomotion during autoregulation, ischemia and reactive hyperemia (134). Another K+ channel, large conductance Ca2+-activated K+ channel has been demonstrated to mediate adenosine-induced vasodilation (35, 3.2.8.1.). 3.2.7. Myogenic response (MR) The myogenic mechanism is thought to be importantly involved in the local regulation of blood flow (18, 52, 106). 3.2.7.1. Background, definition of MR Myogenic response (MR) is defined as an active contraction of the vessel to increase of intravascular pressure (18, 38, 40, 51, 52, 99, 106, 137, 161, 162, 194, 212, 227, 228). Mulvany and Aalkjer (176) described, that myogenic response studied in small arteries (< 500 µm), arterioles (< 100 µm) is a stretch-response or pressureinduced response, in both calcium ions are involved. Myogenic response can be studied rather in cannulated and pressurized vessel preparations in vitro where in vivo variables are eliminated or controlled (neurohumoral, metabolic factors, etc., 51, 52, 105, 136, 161). Hystorical perspective Discovery of MR is credited to Bayliss in 1902 (18), when he recorded large increases in the volume of the dog hindlimb following release of brief aortic occlusions. He considered this response too rapid to be mediated by accummulation of metabolites and thought it reflected the same mechanism by which isolated arteries constricted following sudden distension. Several investigators tended to confirm Bayliss’ findings mostly attributing local vascular regulation primarily to chemical and neural

- 42 -

mechanisms until Folkow (69) demonstrated that denervated preparations develop pressure-dependent vascular tone. In an early review Folkow (70) discussed myogenic hypothesis. He proposed, that myogenic mechanisms originate from vascular smooth muscle, which has an automatic pacemaker function to provide vasomotion and active smooth muscle tone. In late 60-s Johnson pioneered the application of techniques for quantitating the myogenic response in the microcirculation, which lead to intense investigation over the subsequent decade. In a review Johnson (106) discussed myogenic concept as an active response to stretch and he proposed its homeostatic significance as the autoregulation of tissue blood flow. Development of isolated vessel techniques in the 80-s enabled more careful quantification of MR and its underlying mechanisms in small arteries and arterioles, where the effect of pressure could be clearly distinquished from flow, metabolic, neural and endothelial influences (52). 3.2.7.2. Characterization of MR Fig. 3.-6. demonstrates possible diameter changes of a vessel in response to increase in transmural pressure. Response A is considered to be a nonmyogenic, as for vessel does not constrict to increases in perfusion pressure, however it still keeps a constant diameter exhibiting a minimal autoregulation. Responses B and C are considered myogenic because of active decrease in diameter to increases in pressure after an initial increase in low pressures (194). Myogenic response can be further characterized by the myogenic index (MI) (see calculations in 4.1.7.). The MI is an indicator of the relative slope of the active pressure-diameter relation for an arteriole; the more negative the value, the more powerful the myogenic responsiveness of that vessel (51). The range of maximum myogenic responsiveness is demonstrated to be around the in vivo pressures (38, 40, 51, 105). 3.2.7.3. Physiological role of MR: autoregulation It has been demonstrated, that vascular resistance is regulated by several factors, e. g. metabolic, myogenic, flow-shear stress and neurohumoral factors (3.1.3., 12, 173).

- 43 -

Myogenic component largely contributes to the autoregulation of tissue blood flow, that means to keep a relatively constant flow in a wide pressure range (3.2.9.). Autoregulation exists in several vascular beds, e.g. in the kidney, brain, heart, skeletal muscles (32). Autoregulation is best quantified by the the effectiveness of myogenicity, which can be quantified by a gain factor (G) assuming that flow is related to the fourth power of the diameter (Fig. 3.-7.). Thus, the resulting distribution in vessel contractile behavior permitted calculation of the upper and lower limit of the myogenic range for vessels studied and for positive as well as negative pressure steps. The gain is then G=1{[(P2D24/P1D14)-1] / [(P2-P1)/P1]}, where P is transmural pressure, D is the vessel diameter, subscripts 1 and 2 refers to initial and final conditions of a pressure step. Thus, a value of G=1 implies perfect flow autoregulation and values less than unity indicate insufficient myogenicity to maintain a constant blood flow (194). 3.2.7.4. MR in the microcirculation The characteristics of the myogenic response vary among species and tissues and also depend on the size and hierarchy of vessels in the microcirculation with an inverse relationship of vessel size and myogenic responsiveness (51, 52, 105, 136, 162, 212, 227, 228). Microvessels of several vascular beds have been demonstrated to exhibit myogenic response, e. g. rat skeletal muscle (94, 99, 228) and mesenteric arterioles (227), kidney arterioles (92), rat cerebral arteries (194), hamster cheek pouch arterioles (50, 51, 63, 105). In coronary microvessels, myogenic reactivity is thought to be an important local control mechanism, which plays a role in the regulation of myocardial blood flow (12, 38, 40, 137, 173, 212). Small coronary arteries and arterioles exhibit myogenic responses both in animals (136, 137, 139, 197, 212, 234) and humans (115, 162). However, in some studies although coronary arterioles exhibit substantial vascular tone, with increasing intravascular pressure vessels do not constrict or do not keep constricted on high pressures (35, 78), or with lowering intravascular pressure vessels do not dilate (212). Porcine subepicardial or subendocardial (136, 137, 139, 212, 251), dog subepicardial (35) and rat interventricular or septal (78, 197) coronary arteries and

- 44 -

arterioles have been studied. The characteristics of the myogenic mechanism of intramural (i.e. intramyocardial) coronary arteries are likely to be different from those of epicardial vessels, if for no other reason but that intravascular and intramural pressure changes continuously during the cardiac cycle (40, 96, 234). 3.2.7.5. Origin of MR, possible role of endothelium The mechanism eliciting and modulating myogenic responses is still not completely elucidated. The forces generated in response to pressure originate from vascular smooth muscle and thought to be initiated by the stretch of the vascular wall (52, 106, 161). Vascular smooth muscle may have a "pacemaker activity" that establishes basal vascular tone (70, 106). Myogenic response may have a steady state and a dynamic or rate-sensitive component, thus there is a continuous adjustment of vascular tone to the changes in pressure (159). According to Meininger and Davis (161), the possible mechanisms involved in the vascular myogenic response are: 1) altered membrane properties leading to activation of ion channels; 2) modulation of biochemical cell-signaling pathways within vascular smooth muscle; 3) lengthdependent changes in contractile protein function; and 4) endothelial-dependent modulation of vascular smooth muscle tone. Cellular mechanisms may involve stretchactivated channels and voltage-dependent calcium channels on vascular smooth muscle cells (50, 161), however a Ca2+-independent mechanism is also indicated (50). In response to stretch, not only vascular smooth muscle cells, but also endothelial cells may be activated by their dystorsion that activates ion channels on their surface, too (161). Davis and Hill (52) proposes several transduction mechanisms mediating MR. Pressure/stretch may activate different membrane ion channels on vascular smooth muscle cell (e.g. voltage-gated Ca2+ channels, mechanosensitive Na+, Ca2+, K+, Clchannels), different active transporters (Na+-K+ ATP-ase, Ca2+ ATP-ase, K+-H+ ATPase, Na+/Ca2+ exchanger etc.), second messenger systems (Ca2+, IP3-DAG), G-proteins, protein kinase-C, enzymes of PG metabolism, structure of cytoskeleton and extracellular matrix (e.g. integrin), and, most of all, intercellular communication mechanisms.

- 45 -

Previous studies reported that although myogenic responses are generated exclusively by vascular smooth muscle (137, 197, 227), the strength of myogenic tone can be modulated by factors released from the endothelium (11, 12, 92, 94, 161, 204, 234). Such mechanisms however, may be different in vascular beds of various organs. In subepicardial porcine coronary arterioles or in rat coronary arteries and arterioles, myogenic tone (78, 137) is not affected by the endothelium; yet endothelial factors, such as nitric oxide have been demonstrated to play an important role in the control of coronary vascular tone (12, 125, 204, 3.2.5.1.4.). 3.2.7.6. Modulation by exercise Exercise training is described to induce functional and structural adaptations in the cardiovascular system (3.4., 132, 147, 148, 198, 229). It elicits advantageous redistribution of blood flow to improve perfusion of the working skeletal muscles and the heart (132, 146). Furthermore it may alter vasoreactivity altering endothelial and smooth muscle-dependent vascular control mechanisms (146, 147, 172, 185, 198, 199, 229, 250). Exercise training significantly influences the functions of large coronary arteries (146, 192, 198), and coronary microvessels (146, 148, 171, 172, 198, 199). Exercise training has been demonstrated to augment myogenic constriction of rat skeletal muscle arterioles, porcine epicardial and rat intramural coronary arterioles, however the possible mechanism is not yet explained (116, 171, 229). 3.2.8. Vasoactive metabolites studied 3.2.8.1. Adenosine Adenosine (ADO) plays a critical, but probably not a solitary role in the local metabolic regulation of the coronary circulation (193, 3.2.3.). ADO is formed within the myocardial cell when, as a result of hypoxia, ischemia, or vigorous heart work, high energy phosphate compounds are broken down (67, 193). As ATP falls, ADP, AMP and Pi rises. These changes in the energy status of the myocardial cell are thought to activate pathways breaking AMP into ADO, acting at the level of 5’-nucleotidase. This enzyme

- 46 -

converts AMP into ADO at the inner border of the sarcolemma. Most of the adenosine leaves the cell to reach the extracellular space, where it acts on the arteriolar vessel wall as a vasodilator. ADO penetrates into the vascular smooth muscle cell and reaches the circulation, where it is broken down by adenosine deaminase. Agents such as dipyridamole inhibit adenosine deaminase, allow ADO to accumulate, and increase coronary vasodilation, thus vasodilatory reserve can be measured (67, 193, 3.2.9.). Under a variety of myocardial activity states with different metabolic rates, there is a good correlation between myocardial oxygen consumption, the release of adenosine and coronary flow (11). ADO acts on purinergic receptors. A1 subtypes are myocardial receptors and A2 are vascular receptors (193). The vascular A2 receptors situated on the vascular smooth muscle cells stimulate the formation of cAMP, thereby giving second messenger mechanism for coronary artery dilation in addition to the hyperpolarization resulting from opening of K+ channels. K+ATP channels have been demonstrated to mediate vasodilation induced by ADO and thus EDHF may be involved (134, 178, 213). Another K+ channel, large conductance Ca2+-activated K+ channel has been demonstrated to mediate adenosine-induced vasodilation (35, 3.2.6.). ADO, an endothelium-independent vasodilator has been demonstrated to cause prominent vasodilation in several vascular beds, such as in coronary arteries from different species and humans (37, 57, 91, 141, 199, 235). There is a heterogenous vasodilatory action of ADO on coronary vessels. ADO exerts its vasodilatory action chiefly on small resistance vessels in the coronary circulation (11, 141, 173, 178). 3.2.8.2. Acetylcholine Acetylcholine (ACh) is a parasympathetic neurotransmitter (3.2.4.). Vagal fibers on the basis of their acetylcholinesterase content have been identified in the coronary circulation in many species and humans (11, 117, 145). However, the effects of exogenous acetylcholine on the coronary circulation are still the matter of a great controversy (11). The pioneering observation by Furchgott and Zawadski (77) that damage to the endothelial layer reverses acetylcholine-induced relaxation to constriction in isolated aortic preparations has initiated broad research on EDRF. ACh, such as vagal

- 47 -

stimulation may elicit vasodilation or vasoconstriction depending on the function of the endothelium or the relative distribution of muscarinic (M) receptors, i.e. M receptors located on the vascular smooth muscle cells -causing vasoconstriction- and the endothelial cells –causing vasodilation (31, 89, 191, 217). Thus, the resulting vascular effect of ACh, such as of several other agonists (e.g. bradykinin, serotonin, 3.2.5.) depends on the balance of the dilator and constrictor functions. When acting on endothelial muscarinic receptors ACh releases endothelium-derived relaxing factors, mostly NO causing vascular relaxation and hypotension (11, 76, 117, 214, 247, Fig. 3.3.). Furthermore, it has been demonstrated, that ACh on endothelial M3 receptors releases EDHF causing vascular relaxation in guinea pig coronary artery (89). Thus in case of endothelial damage vasoconstricton may occur by activating M receptors on vascular smooth muscle cells inducing IP3-DAG second messenger system or decreasing cAMP level (11, 36, 214, 247). However, the effect of ACh in the coronary circulation may be greatly controversial depending on the species (83, 117). ACh causes a prominent endotheliumdependent vasodilation in monkey, rabbit, dog, cat and guinea pig coronary arteries (5, 8, 54, 83, 144, 145, 191, 209, 242, 253). ACh induces a strong vasodilation in all size classes of coronary arteries and arterioles in anaesthetized dogs and cats (145). In dogs, ACh-induced vasodilation in small coronary arteries and arterioles was appeared to be mediated by NO (109). In another study, although ACh-induced vasodilation of large epicardial coronary arterioles appeared to be mediated by NO, in small coronary arterioles not NO, but a cytochrome P450 metabolite seemed to play a substantial role in the ACh-induced vasodilation (253). Monkey coronary arteries responded to ACh with concentration-dependent relaxation, which seemed to be mediated by NO (191). Although, removal of endothelium reversed ACh-relaxation to contraction, the latter response was abolished by blockade of PG synthesis (indomethacin). This suggests, that the contraction is associated with the release of vasoconstrictor PG-s from subendothelial tissues (191). ACh-induced vasodilation in rabbit hearts was appeared to be mediated by both NO and PGI2 (144). However, in porcine coronary arteries and in rat hearts, ACh may cause strong vasoconstriction even with intact endothelium (55, 83, 137, 139, 182, 183, 186, 212, 217, 243, 254), such as in lower ranked animals, like fish species, amphibians, reptiles and birds (117). ACh-induced coronary vasoconstriction in

- 48 -

rat hearts may be mediated by M2 receptors and constrictor prostaglandins (183). Yang et al. (254) found, that whereas muscarinic receptor activation is responsible for the ACh-induced vasoconstriction in rat hearts, both PG-s and NO are released during ACh infusion and modulate its cardiac effects. On the other hand, ACh induced strong vasodilation in right septal and left ventricular coronary arteries of rats (57, 72, 205, 206). Dilation to ACh in rings of rat coronary artery was mediated by NO (244). Whereas in rat coronary artery preparations of septal and LAD branches ACh-induced vasodilation did not appear to be mediated by NO, nor by prostaglandins, but by a cytochrome P450 product (79). Furthermore, interestingly, ACh relaxed bovine coronary artery strips in low concentrations, whereas contracted in high concentrations (31). Acetylcholine may have a diverse action on human coronary arteries as well, it may cause vasoconstriction even with normal endothelium (8, 11, 64, 83, 117, 170). In isolated human small coronary arteries, ACh paradoxically caused vasoconstriction with a functionally intact endothelium shown by the relaxation response to substance P, another endothelial NO stimulant (8). In angiographically normal coronary arteries of patients with microvascular angina or hypertension, the predominant response to intracoronary ACh appears to be dilation, although constriction is also observed (11, 16, 64, 65, 170, 211). Vasodilation to ACh in human coronary arteries appears to be mediated by NO (211). Interestingly, in epicardial coronary arteries obtained from autopsy ACh induced vasoconstriction in spite of the vasodilatory effect of bradykinin and substance P (239). Furthermore, Miller at al. (163) found a topical heterogeneity in response to ACh in the human coronary circulation: atrial vessels constricted, whereas ventricular vessels dilated. In atherosclerotic segments, however, the predominant response appears to be constriction due to an impaired endothelial NO synthesis in the coronary microcirculation (11, 16, 37, 140). Apart from the integrity of endothelium, the marked negative chronotropic and inotropic effects of ACh alters both metabolic and extravascular components determining coronary blood flow (11, 16, 217). Thus, in vivo experimental studies and studies on isolated hearts and isolated vessels may give diverse results. Although acetylcholine may be a strong vasodilator on isolated rat coronary arteries (72, 79, 205, 244), in isolated rat hearts it may cause either vasodilation (increase in coronary blood flow or decrease in perfusion pressure, 196) or paradoxically vasoconstriction with

- 49 -

intact endothelium (decrease coronary blood flow or increase in perfusion pressure, 55, 183, 186, 217, 253, 254). Thus, acetylcholine may be an important factor in the regulation of coronary vascular tone mostly as a neurotransmitter in the perivascular cholinergic nerves. However, its action on coronary vessels is controversial depending greatly on species and on its negative chronotropic and inotropic cardiac actions as well as on the functional integrity of the endothelium. 3.2.8.3. Bradykinin Increasing evidence shows that bradykinin (BK), a pentapeptide is involved in the normal physiological control processes of the coronary microcirculation. This substance is formed in the myocardium and in the vascular tissue, and it may play an important role mediating vasomotor responses in vivo (86, 88). Furthermore, BK is formed in damaged and inflammed tissues inducing pain sensation (87). Bradykinin, has importance in several coronary pathological states as well (86, 88, 202). BK acts on different receptors both on the endothelium and on vascular smooth muscle cells, thus its effect is usually a combination of an endothelium-dependent vasodilator and a direct smooth

muscle-dependent

vasoconstrictor

action.

The

endothelium-dependent

vasodilation of BK is mostly due to the activation of nitric oxide synthase (62, 86, 139, 144, 196, 202), however, a non-NO-dependent relaxation to BK was also observed, which may be mediated by prostacyclin and EDHF or an other K+-sensitive mechanism (15, 86, 98, 126, 144, 190). BK exerts endothelium-dependent vasodilation on constitutive B2 receptors located at the endothelial cells (17, 88, 190, 202) and its constrictor effect on smooth muscle cells is mediated by B1 receptors (62). Thus, BK is often used as an index of functional integrity of the endothelium (11). There is a substantial heterogeneity of its vasodilatory actions among different vascular beds (190). Specific studies show, that bradykinin is a potent vasodilator on coronary arteries of different species and humans on both isolated vessel and heart preparations, such as in situ (15, 17, 55, 56, 57, 62, 74, 79, 86, 88, 98, 137, 138, 144, 163, 196, 202, 206, 208, 212, 235, 243). BK-induced vasodilation appeared to be the most prominent in the coronary microcirculation (79, 137, 139, 163, 212). Although

- 50 -

BK-induced vasodilation was mostly found to be mediated by NO (57, 62, 86, 139, 144, 202, 235), on pig coronary arterioles BK-induced relaxation was only minimally mediated by NO (47). On small coronary arteries of Dahl rats BK-induced vasodilation was found not to be mediated by either NO or prostaglandins, rather by a cytochrome P450 product (79). On the other hand, both NO and prostaglandins appeared to mediate BK-induced vasodilation on juvenile pig coronary arteries (208). On isolated rat heart vasodilation to BK was found to be mediated by NO (196), and besides NO and prostaglandins, a cytochrome P450 metabolite, epoxyeicosatrienoic acid may have a significant role in the BK-induced effect (74). Interestingly, in isolated rat hearts, the duration, but not the magnitude of BK-induced vasodilation was found to be reduced with NO-synthesis blockade (17). In another study on isolated rat hearts, vasodilation to BK was even found to be augmented with the blockade of NO and prostaglandin synthesis (15). Recently, it has been suggested, that bradykinin may contribute to the physiological control of the coronary circulation (86). In addition, bradykinin is released from the damaged myocardium and helps reducing vascular resistance in the affected area acting on constitutive endothelial B2 receptors. Somewhat later, inducible B1 receptors may appear on the vascular smooth muscle cells and on endothelial cells with diverse contractile actions (88). The effects of bradykinin are influenced by the rate of its local degradation, which is mostly done by angiotensin converting enzyme (ACE). ACE inhibition induces vasodilation probably mediated by BK. Thus, ACE inhibition is a crucial component of present therapy in several cardiovascular pathologies (201). 3.2.8.4. Substance P Substance P (SP) is a peptide neurotransmitter in the nervous system being responsible for pain transmission or as a local mediator accompanying axon reflexes (11, 76, 87). SP has an endothelium-dependent relaxing effect similarly to BK and ACh and other agonists (8, 11, 136, 239). In the coronary circulation SP is a potent endothelium-dependent vasodilator (8, 64, 79, 136, 139, 141, 163, 174, 211, 239). For example, porcine (136, 139, 141, 174), rat (79) and human coronary arteries and arterioles (8, 22, 64, 163, 211, 239) dilate in

- 51 -

response to SP. Although in rat coronary artery rings SP had no effect with a potent ACh-induced vasodilation (244). SP-induced vasodilation in the human and guinea pig coronary circulation was found to be mediated by NO (8, 163, 211, 239). On the other hand, SP-induced vasodilation -similarly to ACh and BK- was found to be mediated by a cytochrome P450 product, rather than by NO and prostaglandins in coronary arteries of Dahl rats (79). Furthermore, SP has been demonstrated to activate an outward K+ current, it induces membrane hyperpolarization and increases cytosolic Ca2+ concentration in porcine coronary endothelial cells (221). Similarly to other endothelium-dependent vasodilators, SP-induced vasodilation is suppressed in atherosclerosis and hypertension (64, 211). 3.2.8.5. NO donors Nitrates have been used in the pharmacological treatment of cardiac diseases for over one hundred years, at the beginning in the treatment of angina pectoris (see: 248). Later, the scale of clinical indications has widened toward other diseases where the vasodilating effect of these drugs could be of benefit. Nitrates and other NO donors release NO and thus can substitute for endothelium-derived NO in the case of a diseased endothelium when the production of NO is reduced and abnormally changed to toxic products (248). They may dilate coronary arteries even in case of coronary artery diseases or atherosclerosis. Thus, NO donors improve myocardial perfusion in coronary heart diseases and lower arterial blood pressure in hypertension (160). Furthermore, NO donors have been shown to reduce cardiac work through peripheral vasodilation and thus alleviate myocardial injury by decreasing myocardial oxygen requirements (2, 248). Till now nitrates are widely used as a primarily important therapeutic agent in several cardiovascular pathologies, most importantly in coronary heart diseases and in hypertension (248, 252). In the treatment of certain ischemic pathological states of the heart organic nitrates have a beneficial effect obviously through NO release (2, 10, 160, 3.3.4.). Among organic nitrates, nitroglycerin, isosorbide dinitrate and isosorbide-5mononitrate are potent vasodilators that have been used for over 100 years in the management of myocardial ischemia (160, 248). These agents exert their therapeutic

- 52 -

action by intracellular conversion to NO or possibly to S-nitrosothiols in vascular smooth muscle thus activating guanylate cyclase similarly to NO (3.2.5.1.). Till recent years, a variety of NO donors have been used in experimental studies, such as sodium nitrite, sodium-nitroprusside (SNP), SIN-1 or molsidomine, pirsidomine, SNAP, GEA compounds, etc. (118, 120, 124, 151, 224, 232, 243, 248). Besides their vasodilator and hypotensive actions, they have been shown to protect against ischemia and reperfusion injury in several species and in different heart models by decreasing infarct size or improving contractile functions (10, 75, 107, 151, 158, 224, 232, 3.3.4.). SNP has been used in experimental studies to demonstrate prominent endothelium-independent vasodilation in several vascular beds, such as in coronary arteries (27, 105, 136, 137, 138, 139, 141, 174, 199, 205, 212, 227, 228, 235). 3.2.8.6. Norepinephrine Norepinephrine (NE) released from sympathetic nerves has a significant influence of coronary vascular tone and myocardial perfusion (210, 3.2.4.). Sympathetic neurotransmitters norepinephrine and epinephrine can elicit either vascular constrictor or dilator effect, depending on the presence of the different types of receptors. Both α and β receptors exist on coronary vessels, from which α causes vasoconstriction and β vasodilation (3.2.4.). α -vasoconstrictor effect seems to be a major predisposing factor eliciting vasospastic angina (16, 87, 193). Subtypes of α and β receptors may have slightly diverging effects. α1 and α2 receptors located on the coronary vascular smooth muscle cells cause vasocontriction. However, α2 receptors located on endothelial cells elicit vasodilation by the release of NO (7, 11, 76). α1 and α2 receptor subtypes are nonuniformly distributed among coronary vessels with different sizes. In large epicardial arteries, α1 receptor subtype is dominant, whereas in coronary resistance vessels α2 receptor subtype (11, 193). In dog coronary microcirculation it was found, that α1-adrenergic activation predominates in small arteries and α2-adrenergic activation predominates in arterioles (109). Chilian (42) found, that α2 receptors are preferentially distributed in arterioles, whereas α1-adrenergic receptors are located throughout the coronary microcirculation. Coronary vascular β receptors appear to respond chiefly to circulating catecholamines (193). Both β1 and β2 receptor subtypes exist on coronary

- 53 -

vessels, β1 receptors are dominant in large epicardial arteries, whereas β2 receptors are dominant in the resistance vessels. NE on β receptors exerts vasodilation by increasing cAMP levels in vascular smooth muscle cells (11, 87, 193). Under experimental conditions, administration of NE usually elicits vasodilation in the coronary circulation by activating β and endothelial α2 receptors (11). However results are contradictory. Previous studies on large coronary arteries have demonstrated direct α-adrenoceptor-mediated vasoconstriction (210), whereas studies on small coronary vessels have generally found minimal α-adrenoceptor-mediated constriction (182, 210). NE predominantly dilated porcine coronary microvessels, both by βadrenoceptor activation and by stimulating release of EDRF (182, 210). NE induces vasoconstriction both on α1 and α2 receptors in large coronary arterioles in the epicardial coronary circulation of intact beating dog hearts, whereas it induces vasodilation in small coronary arterioles mediated predominantly by α2 receptors (39). In the presence of β-adrenergic blockade (propranolol), NE induced vasoconstriction in greyhound and dog coronary arteries, which was increased by endothelium removal and mediated by α1-adrenoceptor activation (7). On the other hand, NE relaxed pig coronary arteries with intact endothelium in the presence of propranolol by EDRF release mediated by endothelial α2 receptors (7). In human coronary arteries, NE-induced vasodilation mediated by β-adrenoceptors is converted to vasoconstriction in atherosclerosis (22). Interestingly, NE induced vasoconstriction in isolated rat hearts, which was augmented by inhibition of NO synthesis (196). Furthermore, NE induced vasoconstriction in rat coronary LAD rings (199). 3.2.8.7. Serotonin Serotonin (5-hydroxytryptamine, 5-HT) is a neurotransmitter and mediator that participates in several physiological and pathophysiological mechanisms such as in central nervous system analgesia, sleep, gastrointestinal functions, inflammation, blood flow control and in hemostasis (11, 87). 5-HT can have either vasoconstrictor or vasodilator effect depending on the condition or the area of circulation. Similarly to other agonists, 5-HT acts via receptors both on the endothelium and on vascular smooth muscle cells, thus their effect is usually a combination of an endothelium-dependent

- 54 -

vasodilator and a direct smooth muscle-dependent vasoconstrictor action. Thus, with intact endothelium, 5-HT may cause vasodilation, although, with damaged endothelium it may exert vasoconstriction (11, 59, 87, 143). Serotonin is released from aggregating platelets and is thought to be responsible for the additional smooth muscle component of ischemia in coronary thrombosis (59). In the coronary circulation, serotonin exerts divergent actions. Localization of these receptors on endothelial and smooth muscle cells may show species specificity. 5HT induces contractions acting directly on vascular smooth muscle cell 5-HT1 and 5HT2 receptors (which may activate IP3-DAG or cAMP systems). Some direct vascular smooth muscle relaxation effects through 5-HT1 receptors are also supposed. It also induces indirect endothelium derived relaxation through 5-HT1 receptors (122, 143). However, in most of the studies, 5-HT is a potent vasoconstrictor in the coronary circulation (122, 143, 187, 188, 205, 206, 244), such as in rat (143, 187, 188, 206, 244) and human (122) coronary arteries. Earlier reports obtained on subepicardial arteries and intramyocardial artery rings show, that the net result of the effect of serotonin on coronary vessels with diameters over 90 µm is constriction (143). On the other hand, in pig (243) coronary arteries and in isolated rat and rabbit hearts (27, 91) 5-HT induced endothelium-dependent or endothelium-independent relaxations. However, serotonin may be used as a preconstrictor in isolated coronary vessel studies in rats (205, 206, 4.1.). 3.2.9. Coronary autoregulation Autoregulation refers to the intrinsic mechanisms, which maintain blood flow constant when the perfusion pressure is varied. Thus, it allows many organs and tissues (e.g. kidney, brain, heart) to adjust their vascular resistance and maintain a relatively constant blood flow in the presence of changes in arterial pressure (11, 32, 52, 67, 106). The coronary circulation exhibits substantial autoregulatory capacity. Coronary blood flow is maintained relatively constant when perfusion pressure varies between 60-160 mmHg. Under normal conditions, blood pressure is kept within relatively narrow limits by the baroreceptor reflex mechanisms so that changes in coronary blood flow are primarily caused by caliber changes of the coronary resistance vessels in response to

- 55 -

metabolic demands of the heart (21). Coronary autoregulatory adjustments involve primarily coronary arteriolar vessels less than 150 µm. Coronary flow reserve, i.e. the potential for increasing coronary blood flow above resting flow, is a measure of autoregulatory capacity. In the human coronary circulation at unchanged coronary perfusion pressure the coronary blood flow can be increased by four- to fivefold (11). In isolated, perfused heart experiments, sudden changes in perfusion pressure induce coronary autoregulation measured by changes in coronary blood flow (28, 204). In isolated, perfused vessels, changes in perfusion pressure induce adjustments of vascular diameter, that can be best explained by myogenic mechanisms (3.2.7). Two main theories have been proposed to be responsible for the autoregulation phenomenon: 1., myogenic and 2., metabolic theories (32). 1. The myogenic theory suggests that vascular smooth muscle constricts when it is stretched secondary to an increase in perfusion pressure, and consequently, vessels dilate due to a decreased perfusion pressure. This mechanism exists in various tissue preparations, however, it is best elicited in the resistance vasculature. In coronary small arteries and arterioles, numerous studies have demonstrated that myogenic mechanism has a significant physiological importance in establishing arteriolar tone, and thus contributing to the autoregulation of tissue perfusion (12, 18, 67, 70, 96, 106, 137, 139, 161, 173, 197, 204, 212, 227, 228, 251, 3.2.7.). 2. The metabolic theory suggests, that an increase in arterial blood pressure initially increases blood flow to a tissue or organ. The increased blood flow washes out vasodilator substances in the area, thus vascular resistance increases and blood flow returns to normal. Several vasodilator substances have been suggested, including CO2, H+, adenosine, lactate, prostaglandins, K+, phosphate ions, low O2 levels. In the coronary circulation, adenosine has been suggested to play a substantial role in the autoregulatory control (135). Furthermore, K+ATP channels play an important role in coronary microvascular vasomotion during autoregulation in dogs (134). 3.2.10. Integration of microvascular control mechanisms Several regulatory factors influence the caliber of coronary resistance arteries and arterioles Fig. 3.-1.), they may also influence each other in a complicated fashion.

- 56 -

An important mechanism of interactions among the various regulatory factors relates to the manner by which they act in a concerted integrative manner to maintain proper oxygen delivery to the myocardium (173). Local regulatory mechanisms may show a segmental distribution in most microvascular systems. This is supported by several studies, which described that metabolic, myogenic, neural, flow shear stress-dependent mechanisms and active and passive biomechanical wall properties depend upon vessel size, furthermore, a nonuniformity of vasomotor responses is described along the coronary arterial tree (141, 147, 173, 233). Coronary vasomotor responses may exert great variability among species and greatly depend not only on the vessel caliber, but also on their location of subepicardial, subendocardial or intramyocardial positions due to the mechanical effects of the myocardium (11, 38, 40, 96, 136, 173, 233). Most of the controlling factors contribute to the determination of the microvascular resistance of the heart (38, 40, 141, 173). Fig. 3.-8. illustrates a simplified sheme in which metabolic, myogenic and flowdependent mechanisms are integrated to optimize the network responses to alterations in metabolism. This sheme is likely to be too simple, it neglects other important controlling factors, e.g. neurohumoral control mechanisms and biomechanical wall characteristics. It is based on data measured by Kuo et al. (141) that demontrates longitudinal gradients for metabolic, myogenic and flow-dependent responses. The figure depicts the feed-forward sequences (+) and the negative feedback loops (-). If myocardial oxygen consumption would suddenly increase during hemodynamic stimulatory effects associated with exercise or excitement, there would be an increased production of metabolites of the heart. Meanwhile, an increased metabolism induces vasodilation, which occurs preferentially in the resistance vasculature, mainly in small coronary arterioles. Small coronary arterioles are also more sensitive to adenosine than upstream vessels (137, 141, 173). Metabolic dilation at this resistance-sized vasculature would decrease upstream pressure. This would produce myogenic vasodilation of intermediate-sized arterioles. Myogenic and metabolic vasodilation would also cause a decrease in network resistance, allowing greater flow through the upstream larger arterioles and small arteries (141), thereby recruiting flow-dependent vasodilation of these vessels. This upstream dilation allows pressure to be transmitted to downstream segments, which would stop further vasodilation and induce negative feedback

- 57 -

mechanisms. The overall decrease in resistance and thus the increase in flow would induce washout of vasodilatory metabolites and thus with negative feedback mechanisms the metabolic demands and also the coronary blood flow would return to baseline level. It is also demonstrated, that myogenic constriction effectively counteracts flow-induced dilation in the coronary microvasculature (139), such as is other vascular beds (230), which may be a key mechanism in controlling blood flow in vivo. However, in case of coronary artery diseases with severe stenosis and impaired endothelial function these adjusments would not occur due to impaired flow-mediated and metabolic vasodilation (173, 247). Vasoregulatory mechanisms have a great dependence on endothelial functions. Most of the vasoactive agents, like acetylcholine, bradykinin, histamin, serotonin, substance P, ATP have dual actions on the vessel wall by causing endotheliumdependent

vasodilation

and

concomitantly

direct

smooth-muscle-mediated

vasoconstriction. Thus, their vasoactive effects depend on the balance of endotheliumderived and smooth-muscle-directed functions, which may vary between species and different vascular beds and greatly depend on the integrity of the endothelium. In case of intact endothelium some otherwise vasodilator factors, e.g. acetylcholine, histamin, serotonin still may cause vasoconstriction in certain vascular beds (11, 83, 137, 139, 143, 182, 186, 187, 188, 205, 212, 217, 243, 244, 254). Fig. 3.-2. summarizes several factors influencing vasomotor tone. With intact endothelium, vasoactive factors, e.g. hemostasis

factors

released

from

platelets

(serotonin,

thrombin,

ADP),

neurotransmitters and peptides (acetylcholine, bradykinin, substance P), hormones (angiotensin, catecholamines, histamin) or hydrodynamic stimuli (shear stress) may contribute to an enhanced vasodilation by releasing relaxing factors from endothelium (NO, PGI2). On the other hand, when endothelium is damaged due to atherosclerosis or coronary artery disease, many of these factors will enhance vasoconstriction by their actions directly on the vascular smooth muscle including products of activated platelets that may accelerate the progression of the coronary artery disease (11, 59, 64, 65, 150, 160, 170, 247, 255). 3.3. The coronary microcirculation in myocardial ischemia and reperfusion

- 58 -

Myocardial ischemia develops when oxygen supply to heart is insufficient to meet the oxygen demand. This may appear either with an insufficient blood flow in coronary artery diseases or coronary occlusion, and with an increased metabolic demand of the heart e.g. during exercise (87, 93, 123). Modelling myocardial ischemia and reperfusion has a great impact concerning the high significance of ischemic heart diseases in patients. There are two fundamental reasons for explosive growth of interest in this phenomenon. First, it is now recognized, that spontaneous reperfusion after coronary spasm or thrombosis is common in patients with coronary artery disease. Second, with the advent of interventional recanalization, an increasingly large number of patients are subjected to coronary reperfusion in an effort to preserve left ventricular function (24, 93). A wide range of in vitro and in vivo model systems has been utilized to study cardiac reperfusion injury, in most of which isolated perfused and in situ hearts and papillary muscle preparations are applied (10, 24, 75, 107, 123, 151, 158, 224, 232). Description of myocardial energetics and metabolism will be omitted. 3.3.1. Effects of short term ischemia, reactive hyperemia When blood supply to a tissue is blocked for a few seconds and then is unblocked, the flow through the tissue ususally increases to four to seven times normal, the increased flow will continue for a few seconds. This phenomenon is called reactive hyperemia (21, 67, 87). After short periods of vascular occlusion, the extra blood flow during the reactive hyperemia phase lasts long enough to repay almost exactly the tissue oxygen deficit that has occured during the period of occlusion. This mechanism emphasizes the close connection between local blood flow regulation and delivery of oxygen and other nutrients to tissues. In a possible mechanism metabolite-mediated dilation in arterioles accelerates blood flow in the feeder arteries. This increases adenosine levels (67), the shear stress on the arterial endothelium and can induce vasodilation by release of EDRF (NO). Although intact endothelium was reported to be a necessary requirement for reactive hyperemia, NO is probably not the only determinant of reactive hyperemia in the coronary circulation (21, 29, 87). 3.3.2. Effects of long term ischemia, myocardial stunning

- 59 -

The events in parallel with the return of oxygen into hypoxic or ischemic organ are generally referred to as reperfusion or reoxygenation injury. Postischemic dysfunction termed as myocardial stunning (MS) defines the mechanical dysfunction that persists after reperfusion despite the absence of irreversible damage (24, 25, 93). In accordance with this definition, MS is a relatively mild, sublethal injury that must be kept quite distinct from myocardial infarction. Thus, in reperfusion period following ischemia, the depressed myocardial functions gradually improve (58, 93). In experimental models of isolated or in situ hearts, MS can be examined in regional or global myocardial ischemia with complete or intermittent coronary occlusions or with low flow method (25, 123). Mechanisms of MS MS is probably a multifactorial process that involves complex sequences of cellular perturbations and the interaction of multiple pathogenetic mechanisms. Several different hypotheses are proposed in the pathogenensis of MS (24, 25, 142): A. Most likely mechanisms 1. Generation of oxygen-derived free radicals 2. Excitation-contraction uncoupling due to sarcoplasmic reticulum dysfunction 3. Calcium overload B. Other proposed mechanisms 1. Insufficient energy production by mitochondria 2. Impaired energy use by myofibrils 3. Impairment of sympathetic neural responsiveness 4. Impairment of myocardial perfusion 5. Damage of extracellular collagen matrix 6. Decreased sensitivity of myofilaments to calcium A unifying hypothesis for the pathogenesis of MS is proposed in Fig. 3.-9. This proposal integrates and reconciles different mechanisms into a unifying pathogenetic hypothesis (24, 25). Transient reversible ischemia followed by reperfusion could result

- 60 -

in increased production of superoxide radicals (O2-) through several mechanisms, including 1) increased activity of xanthine oxidase, 2) activation of neutrophils, 3) activation of arachidonate cascade, 4) accumulation of reducing equivalents during oxygen deprivation 5) derangements of intramitochondrial electron transport system resulting in increased univalent reduction of oxygen, and 6) autoxidation of catecholamines and other substances. Superoxide dismutase (SOD) dismutates .O2- to hydrogen peroxide (H2O2); in the presence of catalytic iron, .O2- and H2O2 interreact to generate the hydroxyl radical (.OH). H2O2 can also generate .OH in the absence of .O2provided that other substances (such as ascorbate) reduce Fe (III) to Fe (II). .O2- and .OH attack proteins and polyunsaturated fatty acids, causing enzyme inactivation and lipid peroxidation, respectively. In reversible ischemia, the intensity of this damage is not sufficient to cause cell death but is sufficient to produce dysfunction of key cellular organelles. Postulated targets of free radical damage include 1) the sarcolemma, with consequent loss of selective permeability, impairment of calcium-stimulated ATPase activity and calcium transport out of the cell, and impairment of Na+, K+-ATPase activity (resulting increased calcium influx and cellular calcium overload); 2) the sarcoplasmic reticulum, with consequent impairment of calcium-stimulated ATPase activity and calcium transport (resulting in impaired calcium homeostasis causing excitation-contraction uncoupling); and 3) possibly other structures, such as the extracellular collagen matrix (with consequent loss of mechanical coupling) or the contractile proteins (with consequent decreased sensitivity to calcium). At the same time, reversible ischemia-reperfusion could cause cellular Na+ overload due to 1) inhibition of sarcolemmal Na+, K+-ATPase and 2) acidosis and Na+-H+ exchange. This could further exaggerate calcium overload by increased Na+-Ca+ exchange. An increase in free cytosolic calcium would activate phospholipases and other degradative enzymes and further exacerbate injury. The consequence of this complex series of perturbations is a reversible depression of contractility (24, 25). Despite the considerable progress, the pathogenesis of MS has not been definitively characterized (25). Among the numerous mechanisms proposed, three have emerged as likely contributing factors: 1, generation of oxygen radicals, 2, calcium overload, and 3, decreased responsiveness of contractile filaments to calcium. The oxyradical hypothesis and the calcium hypothesis thus may represent different steps of

- 61 -

the same pathophysiological cascade. Thus, generation of oxyradicals may cause calcium overload, and both of these processes could lead to damage and dysfunction of myofilaments (25). Markers of MS In experimental studies a wide range of markers have been used to study the damage evident at reperfusion. These markers include the ones which are identified and measured in patients to diagnose acute myocardial infarction (93, 123, 240): 1. Biochemical: mitochondrial function, oxygen utilization, energy metabolism, calcium transport and content 2. Conductance: electrocardiogram, arrhythmias 3. Lytic: release of intracellular enzymes, release of myoglobin 4. Mechanical: contractility, peak and resting tension, coronary flow, cardiac output, systemic blood pressure, ventricular pressure, contractility (dP/dt), ventricular wall thickening 5. Pathological: infarct size, myocardial edema, capillary permeability, ultrastructural and hystochemical changes Most of ischemia-reperfusion studies monitor mechanical functions of the left ventricle: left ventricular pressure (LVP), dP/dt (a measure of cardiac contractility), heart rate in isolated heart studies with systemic hemodynamic parameters in in situ studies. LVP, dP/dt may be markedly depressed following ischemia. In the reperfusion period, cardiac functions gradually improve, although LVP and dP/dt parameters usually reach only to about 70 % of the preischemic level (91, 158, 232). In other studies, infarct size with area of risk may be determined, enzyme markers and development of arrhythmias may be followed, or even altered metabolic fuctions and myocardial energetics can be monitored as the measure of ischemia (58, 75, 123, 151, 153, 158, 224, 240). 3.3.3. Endothelial function in myocardial stunning, role of NO

- 62 -

Endothelial integrity is a substantial condition to maintain adequate myocardial perfusion. Endothelial cell dysfunction in coronary artery diseases, in atherosclerosis or hypertension etc. may aggravate myocardial ischemia (3.2.5.). On the other hand, in the experimental myocardial stunning, endothelial cells may be damaged further aggravating impaired myocardial perfusion. After induced ischemic cardiac arrest and reperfusion, myocardial blood flow and oxygen consumption are decreased, and the magnitude of this decrease is proportional to the degree of injury to myocytes, as judged by cardiac function and metabolism. There are several possible explanations for decreased coronary flow after arrest. First, interstitial edema, intracellular edema or both could cause extravascular compression of the coronary arteries and prevent normal blood flow. A second mechanism could be appropriate autoregulation resulting from the decreased metabolic demands of the myocardium after arrest. Third, coronary arterial smooth muscle damage could augment vascular tone and impair the ability of the vascular smooth muscle to relax. Finally, endothelial cell damage could impair the ability of the coronary endothelium to release EDRF (NO), prostacyclin or both, and thus increasing arterial tone indicated by impaired endothelium-dependent relaxation to agonists. Decreased coronary blood flow may in turn lead to ischemia and additional tissue damage during reperfusion (91, 150, 224, 226). In isolated rabbit heart after global ischemia and reperfusion, the smooth muscle vasodilatory response induced by adenosine was still intact, however, there was a marked impairment of endothelium-dependent relaxation of the coronary arteries induced by serotonin. Thus, endothelial cell dysfunction is most likely the cause of the decreased coronary flow observed after global cardiac ischemia and reperfusion in the isolated rabbit heart (91). Endothelial dysfunction reduced endothelium-dependent relaxation in isolated coronary rings and isolated perfused cat hearts. This endothelial dysfunction may be due to production of superoxide radicals which incativate NO. Endothelial generation of superoxide radicals acts as a trigger mechanism for endothelial dysfunction which is then amplified by platelet aggregation and neutrophyl adherence and diapedesis into the ischemic region enhancing post-reperfusion ischemic injury. Thus, the reduction of NO release may increase the ischemia-reperfusion-related damage of myocytes by predisposing coronary vessels to vasospasm, thrombosis and

- 63 -

neutrophyl adherence. Agents that preserve endothelial function or inhibit neutrophyl activation (e.g. superoxide dismutase, prostacyclin analogues, thromboxane synthase inhibitors) can protect against endothelial dysfunction and myocardial injury (2, 91, 150, 224, 248). This endothelial cell damage may induce not only a global reduction of myocardial perfusion, but also an inhomogenous reperfusion deficit in certain regions eliciting depressed cardiac functions, and thus, exogenous NO may elicit advantageous redistribution of blood flow to improve cardiac functions. Thus, NO donors may exert their cardioprotective actions in ischemia-reperfusion by improving regional cardiac perfusion (2, 93, 181, 232, 3.3.4.). On the other hand, ischemia-reperfusion impaired endothelium-dependent relaxation in coronary microvessels, but not in large coronary arteries (209). Although, endothelium-dependent relaxation to bradykinin was still preserved in reperfusion followed by low flow ischemia in isolated rat hearts indicating, that the increased coronary resistance and altered myocardial performance is not obligatory by a result of an irreversible endothelial damage (232). However increased NO synthesis is shown to be protective in ischemiareperfusion, excessive NO levels due to an increased iNOS activity may aggravate myocardial ischemia by inducing cytotoxicity and hypotension (6, 59, 71, 73, 252). 3.3.4. Cardioprotection Cardioprotection in ischemia and reperfusion may be exerted by several mechanisms. Among these mechanisms vasodilation and thus advantageous redistribution of tissue perfusion, direct cytoprotective effects, advantageous metabolic effects and antihemostasis have to be mentioned. Several agents have been studied and proposed to exert cardioprotection (e.g. bradykinin, adenosine, NO donors, L-arginine), most of them exert their actions through NO (4, 10, 27, 59, 75, 150, 151). In several studies nitrates have been shown to be beneficial for the myocardium against ischemia and reperfusion injury. Nitrovasodilators (sodium nitrite, sodium nitroprusside; molsidomine and its metabolite SIN-1; nitroglycerin, pirsidomine, SPM5185, GEA compounds, etc.), NO-synthesis stimulators (bradykinin, ACE inhibitors, Larginine) and by releasing protective metabolites ischemic preconditioning have been shown to protect against ischemia and reperfusion injury on several animal heart models

- 64 -

(4, 10, 27, 75, 107, 151, 153, 158, 232, 256). This protective action was shown in in vitro experiments, as it was indicated by decreased release of ischemic marker enzymes and improved metabolic and contractile functions (4, 153, 232, 256). However, in in vivo studies NO and NO donors were suggested to elicit their myocardial protective actions by decreasing myocardial contractile functions probably due to a peripheral vasodilation reducing ventricular filling pressure. In addition, the decrease in infarct size indicates myocardial protective action of NO in several in vivo studies (2, 10, 75, 107, 151, 158). On the other hand, some studies did not show beneficial actions of NO, conversely, NO may have a direct myocardial depressing effect and even may increase the infarct size, especially if it is present in excess amount (6, 200). The mechanisms of protective actions of NO and NO donors are not entirely understood. NO induces vasorelaxation and thus improves myocardial perfusion. In addition, NO scavenges superoxide radicals, inhibits platelet aggregation, reduces neutrophyl adherence, and by these mechanisms it can alleviate the injury induced by ischemia and reperfusion (2, 10, 75, 150, 151). In the treatment of certain ischemic pathological states of the heart organic nitrates have a beneficial effect obviously through NO release (2, 9, 10, 160). Exogenously administered NO may protect endothelial cell function by replacing endogenous NO in ischemia (224). Furthermore, an inhomogenous reperfusion deficit, i.e. an imbalance between myocardial perfusion and function in ischemic myocardium at regional level induced by locally increased vascular resistance in postischemia may be attenuated in response to NO donors, and thus advantageous redistribution of perfusion may be obtained by improving ischemic zone nutrient blood flow. Thus, NO-donors, like nitroglycerin and pirsidomine were observed to cause a redistribution of blood flow from the non-ischemic areas to the ischemic areas of the myocardium. This may explain the improved myocardial functions postischemia in response to NO donors (2, 93, 158, 181, 232). Another possibility is a direct cytoprotective action of the NO donor (150). 3.4. Effect of exercise on the coronary vascular functions Exercise training has been demonstrated to induce cardiovascular adaptations involving the resistance vasculature (132, 146, 229, 231). Adaptations to chronic

- 65 -

physical exercise can be different after a short term (e. g. 1-4 weeks) or a long term endurance exercise training (e. g. 16-20 week, 132, 147, 171, 172, 192, 199, 219, 229, 231, 250). Some investigators have studied the effects of long term, chronic exercise training on the functions of the cardiovascular system (146, 147, 172, 198, 199), which is supposed to be beneficial in preventing or delaying coronary arterial diseases (146, 147, 185). These adaptations in the coronary vasculature involve increased vasodilator capacity and enhanced transport capacity associated with enhanced capillary diffusion capacity (146, 147, 199). Exercise training elicits adaptations mostly in the resistance vasculature, which are the major sites of metabolic and myogenic control mechanisms. Early vascular adaptations to exercise training often involve altered vasoactive responses with increased vasodilator and altered vasoconstrictor responses (146, 147, 148, 172, 199). Endothelium-dependent vasodilator responses are often increased in exercise training associated with an increased NOS gene expression in endothelial cells indicating the role of nitric oxide in the enhanced vasodilator capacity (146, 147, 219, 231). Altered hemodynamic parameters and increased blood flow in exercise training is supposed to be associated with increased flow shear stress that induces endothelial nitric oxide synthesis (146, 185, 219, 229, 231, 250). Thus, the increased vasodilatory reserve in exercise training may be explained by altered vascular functions, i.e. increased endothelium-dependent and endothelium-independent vasodilator responses and/or altered vasoconstrictor responses (132, 146, 147, 172, 199, 229), although vasoactive responses may be unaltered (147, 148, 192, 199). Furthermore, the long term adaptation mechanisms in chronic exercise training may involve structural remodelling of the vasculature and the cardiac muscle by angiogenesis or vascular and cardiac hypertrophy (146, 185). Thus exercise training may have beneficial effects not only in healthy individuals by decreasing risk factors and preventing the occurence of certain pathological conditions, but also it may improve patients’ conditions in coronary heart diseases or in atherosclerosis (146, 147, 185). 3.5. Methods to study coronary microcirculation The investigation of coronary microcirculatory function entails several difficulties, notably gross movement attributed to cardiac contraction, that are not found

- 66 -

in other microcirculatory beds (44). Thus, among in vivo methodologies in vitro studies may add additional information about the specific localization of vascular and metabolic control mechanisms in the myocardial tissue. Methodologies of in vitro studies are essential in order to study vascular wall myogenic or pharmacological responses not influenced by in vivo existing factors, e.g. metabolic, humoral and neural influences (51, 105, 136, 162, 212). In case of heart, these strategies have been applied as a way to avoid the limitations imposed by the motion and thickness of the intact heart (44). Microvessels of interest can be dissected from any region of the heart, including human heart, and can be cannulated and perfused at various pressures and flows in order to study myogenic or humoral control mechanisms (44, 78, 79, 81, 96, 108, 137, 139, 141, 162, 173, 176, 206, 207, 212, 225, 233). Also isolated vascular ring preparations applied onto wire myographs (47, 126, 143, 176, 187, 188, 244) are widely used for studying pharmacological responses, as for its relatively easy application, although it has several limitations. In vascular ring studies artificial vessel geometry is controlled by the applied tension measured between the wires, not by intraluminal pressure. Furthermore, the investigation of microvessels is limited. Apart from isolated vessel methodologies, isolated heart preparations are used to study physiological and metabolic parameters of heart independently of its environment (53). Several preparations of isolated perfused heart is currently used, mainly the Langendorff heart with a retrograde perfusion system and the working heart model. Isolated heart methods allow to study coronary microcirculation by measuring coronary flow or perfusion pressure changes in constant pressure or constant flowperfused conditions, respectively. Thus, effects of various drugs on coronary resistance and cardiac performance can be examined. This system may also reproduce different pathological conditions, such as ischemia, reperfusion and hypoxia (53, 54, 58, 74, 91, 183, 196, 217, 232, 254). Coronary microcirculation can be also investigated in situ in anaesthetized animals (38, 40, 44, 108, 112, 202). Furthermore, human clinical studies may provide valuable information of the cardiovascular control mechanisms (64, 65, 86, 160, 170, 255). All these methodologies together are essential in order to obtain a proper concept of coronary vascular control mechanisms. Different measurement techniques may be applied both in vivo and in vitro in order to obtain data on regional coronary blood flow or vascular diameter. In humans,

- 67 -

Doppler flow angiography and contrast echocardiography techniques are mostly applied (64, 65, 86, 93, 170, 193, 240, 255). In in situ or isolated beating hearts, epi-illumination with intravital microscopy may give comprehensive information about microvascular diameter mostly in the subepicardial layer (40, 41, 44, 108, 241), and recently a method for directly viewing subendocardial and intramural vessels in beating hearts has been developed with the aid of a needle probe (44). Furthermore, coronary blood flow can be measured with hydrogen clearance technique (238) and with radiolabelled microspheres (67, 93, 193). Techniques of CT, NMR and PET with injected contrast materials or radioisotopic tracers are often used in animal studies providing three dimensional views of entire coronary microcirculation and thus useful means of studying vascular structure and function (44, 53, 93, 193, 240). In addition, technological advances in cell isolation and identification involving fluorescence cell sorting, magnetic beads and antibodies have greatly facilitated the feasibility of isolating endothelial cells from specific microvascular beds (44).

- 68 -

4. Materials and Methods 4.1. Isolated vessel studies 4.1.1. Animals Male Wistar and Sprague-Dawley rats were used in our studies, weighing around 350-450 grams. All rats were derived from Charles River Laboratories. Wistar rats used in the protocols performed in New York Medical College were ordered directly from Charles River Laboratories and kept further in the college by specially trained assistants. Protocols, applied in the studies were approved by Animal Care Committee of New York Medical College. Sprague-Dawley rats used in protocols performed at Semmelweis University were obtained from Charles River Laboratories and further bred in the animal house of the Institute of Human Physiology and Clinical Experimental Research. All rats were anaesthetized with Pentobarbital (50 mg/kg intraperitoneally) before preparations. 4.1.1.1. Exercise protocol Some rats in the exercise study (5.1.2.2.) were exercised on a treadmill. The training was carried out 5 days a week for 4 weeks. The length of the time on treadmill was initially 5 min/day with the speed of 0.5 mph at grade 0 and progressively increased to 20 min/day with 0.8 mph, grade 1 by the end of the 1st week, to 40 min/day with 1 mph, grade 3 by the end of the 2nd week and to a maximum of 60 min/day with 1 mph, grade 5 by the beginning of the 4th week. Rats ran up to exhaustion on the 5th day of the 1st and last week. Sedentary rats were also handled, but were not made to run the treadmill. 4.1.2. Equipment In the description of the equipment we refer to the numbers of different experimental protocols we used (4.1.6.).

- 69 -

Measurements of perfusion (intraluminal) pressure Pressures at both sides of the segments were continuously measured with Gould pressure transducers calibrated with a mercury manometer. In the experimental protocol No. 1 intraluminal pressure of the segments was controlled by an air-pressurized reservoir vessel and a continuous flow pump (Harvard Apparatus) on one side of the vessel, meanwhile the other side was closed. In experimental protocol No. 6 intraluminal pressure was controlled by pressure servo-controlled peristaltic pump (Living Systems, Burlington, VT) smoothed with a rubber balloon built into the line and connected to one side of the vessel, meanwhile the other side was controlled by a continuous flow pump. In protocols No. 2, 3 and 4 a pressure servo-controlled syringereservoir (Living Systems, Burlington, VT) was connected to one side of the vessel, meanwhile the other side was closed. In protocol No. 5 we used a hydrostatic-reservoir system for adjusting intraluminal pressure. In flow experiments (protocol No. 6) intraluminal flow through the segment was adjusted by pressure servo control system. Measurements of vessel diameter In protocols No. 1 and 6 outer diameter of the segments was continuously and automatically measured by in vitro microangiometry (Fig. 4.1.-1a.,b.). The microangiometer setup was constructed in the workshop of our Department. In this setup the glass-bottomed tissue bath was positioned in the light path of a microscope. A magnified picture of the vessel was formed with the aid of a videocamera (Philips LDH 0702/20) and a monitor (Philips Computer Monitor 80). A specifically developed microcomputer evaluated the signal from the camera and automatically positioned two light markers adjusted to the contours of the vessel. The distance between the two light spots, i. e. the outer diameter of the segment was continuously measured (Fig. 4.1.-1a., Fig. 4.1.-6.). In protocols No. 2, 3, 4 and 5 we measured inner diameter of the segments. In protocol No. 5 we used the same microangiomerer setup, but in a manual mode with a continuous manual adjustment of the 2 light spots. In protocols No. 2, 3 and 4 the inner diameter of the arterioles was measured with a calibrated image-sharing monitor (Instrumentation for Physiology and Medicine, San Diego, CA) with a continuous manual adjustment of the shared image of the vessel.

- 70 -

Intraluminal drug administration In protocol No. 6 a continuous flow pump (Cole Palmer) was applied for intraluminal drug administration. Measurement of the weight of the vessel The weight of the vessel was measured with an ultramicrobalance (Cahn C31, accuracy about 0.1 microgram). During the equilibration of the balance, very small pieces of wet tissue were drying up quickly and the measured weight as a function of time approximated a value which was in a 1 % range of the dry weight of the segments (determined by standard methods, 99 oC for 1 hour in dry hot oven). In case of larger branches the wet weight could also be determined from the early linearly decreasing part of the weight curve. Smaller vessels dried out quickly, thus inducing unacceptable error in the directly measured wet weight values. Dry weight was measured in these segments, and wet weight was computed using the water content found in larger coronary resistance arteries (81.9 %). All vessels are rich in cellular elements in this series, thus a similar water content can be supposed. Volume of wall material was computed using 1.06 g/cm3 specific gravity value. The vascular wall was supposed to be incompressible during the measurements. Recordings Pressure and diameter signals were digitized by an A/D converter (PCL 7/8, Adventech Corporation) and in protocols No. 1 and 6 were transmitted into an Pentium PC computer for data storage and further processing. In protocols No. 2, 3, 4 data were recorded on an Omega Engineering chart recorder (Stamford, CT) and in No. 5 with an OH-850 Radelkis chart recorder (Budapest). Calibrations were made with a mercury manometer and a micrometer etalon. 4.1.3. Solutions and drugs In protocols No. 1 and 6 tissue bath contained normal Krebs-Ringer (nKR) solution, composition of which was (mmol/l): 144.9 Na+, 4.5 K+, 2.5 Ca2+, 1.2 Mg2+, 1.2

- 71 -

-

H2PO4 /HPO42-, 3.6 SO42-, 125.9 Cl-, 22.5 HCO3-, 5.56 glucose (Sigma Ltd., St. Louis, MO). It was thermostated at 37 oC and bubbled with 95 % O2 and 5 % CO2 which kept the pH at 7.4. In protocols No. 2, 3, 4 and 5 we used physiological salt solution (PSS), which contained in mM 110 NaCl, 5 KCl, 2.5 CaCl2, 1 MgSO4, 1 KH2PO4, 24 NaHCO3 and 10 dextrose (Sigma Ltd., St. Louis, MO - Budapest), which was thermostated at 37 oC and bubbled with 20 % O2 and 5 % CO2 in case of arterioles (No. 2, 3 and 4) and with 95 % O2 and 5 % CO2 in case of small arteries (No. 5). The pH was kept at 7.4. Passive diameter (PD) values were obtained with the addition of papaverine, a smooth muscle relaxant in protocol No. 1. In all other protocols we applied Ca2+-free Krebs-Ringer or PSS solution containing 1.0 mM EGTA (Sigma Ltd., St. Louis, MO). Adenosine, acetylcholine, bradykinin, norepinephrine, sodium-nitroprusside, nitro-L-arginine, nitro-L-arginine-methylester, 5-hydroxytryptamine (serotonin), and U46619 were obtained from Sigma Ltd. (St. Louis, MO). Indomethacin, prostaglandins E2 and I2 , SQ 29,548 were obtained from Cayman Chemical (Ann Arbor, MI). Prostaglandin F2α and papaverine were obtained from Chinoin-Sanofi (Budapest). Chemicals were prepared on the day of the experiment by dissolving in nKR or PSS. Indomethacin was dissolved in sodium-bicarbonate solution, which did not change the pH of the perfusate. Prostaglandin E2 and SQ29,548 were dissolved in ethanol (in 10-2 M concentration) and further diluted in nKR or PSS. Prostaglandin I2 was dissolved in a buffer containing (10-2 M) sodium-bicarbonate and (10-2 M) sodium-carbonate on a pH at 10.0, and further diluted in nKR or PSS at the moment of administration maintaining a pH of 7.4. 4.1.4. Preparation After anaesthesia of the rat, the chest was quickly opened and the heart was removed and placed into a preparation dish containing cold, oxygenized PSS (protocols No. 2, 3, 4 and 5) or nKR (1 and 6) solution. In all of our studies segments of the branching system of the left anterior descending (LAD) coronary artery were prepared by careful microdissection. Fig. 4.1.-2. shows the branching of LAD of the rat heart visualized in a video microscope and Fig. 4.1.-3. represents an illustration of a possible

- 72 -

branching course, which may vary among individuals and tribes, although we may find general similarities in the branching forms. In our studies main branches, close to the coronary orifice (7. on Fig. 4.1.-3.), end branches somewhat below halfway between the orifice and the apex (called proximal end branch, 15. around 13. on Fig. 4.1.-3.), end branches close to the apex (called distal end branch, 15. between 9. and 16. on Fig. 4.1.3.), first order side branches (11. and 13. on Fig. 4.1.-3.), second order side branches (12. and 14. on Fig. 4.1.-3.) and apical branches (16. on Fig. 4.1.-3.) were prepared (56, 57, 116, 233, 234, 235). Fig. 4.1.-4. shows a prepared and in situ perfused LAD branching system of the rat heart. All segments studied were of intramural position. Preparation was made by Zeiss and Leica microscopes (Wild, Switzerland and Wien, Austria) with zoom lenses. The dissection was started by removing the visceral epicardial layer above the main course of LAD at the orifice near the aorta. When going downward on LAD, the epicardial layer and the superficial heart muscle layers were removed by blunt dissection with microforceps (Dumont Biologie, Fine Science Tools, USA), untill the vessel was clearly seen. The heart muscle fibers and and connective tissue elements directly covering and surrounding the vessels were carefully removed by sharp dissection using finely polished microforceps and 27 Gauge needles that were further polished. Side branches could be prepared at least down to about 10 µm. Then all side branches were tied with individual filaments teased from a multifilament suture thread (6-0) with the aid of two pairs of microforceps. Tied branches had to be carefully cut above the tie (distal from the vessel) with a microscissor (MiniVannas, Fine Science Tools, USA) without any distraction onto the prepared segment. Once segment is distracted or pulled either radially or axially, endothelial layer may be detached and damaged. When the sides of the vessels are prepared free, careful preparation is continued to the base of the vessel with fine microforceps and a needle. Once the desired segment is prepared completely, the in situ length was measured. Then the segment was cut out, removed and placed into a tissue bath. Fig. 4.1.-5. shows microphotographs of a prepared segment. 4.1.5. Cannulation and mounting

- 73 -

The segment was cannulated first at one side and any remnants of clotted blood had to be removed from the lumen at low perfusion pressure. Both ends of the segments were cannulated with plastic (protocols No. 1 and 6) or glass (protocols No. 2, 3, 4 and 5) microcannulas (Fig. 4.1.-1b.), the size of which was chosen taking into consideration the in situ measured caliber of the segments. After cannulation of both sides, vessels were checked for leaks by the stability of perfusion pressure in the manual mode of the pressure servo-control pump, or by the stability of the fluid level in the pressurereservoir. Those vessels with observable leaks were discarded. The in situ axial length of the segments was set with a micrometer screw and kept constant throughout the experiment. The picture of cannulated and pressurized vessel segments as appear on the monitor are shown in Fig. 4.1.-6. 4.1.6. Experimental protocols In this study we used the following protocols: 1. Biomechanical wall characteristics of rat intramural coronary resistance arteries and arterioles. 2. Myogenic characteristics of rat intramural coronary arterioles. 3. Myogenic characteristics of rat intramural coronary arterioles, effects of daily exercise. 4. Endothelium-dependent and endothelium-independent vasoactive responses of rat intramural coronary arterioles. 5. Endothelium-dependent and endothelium-independent vasoactive responses of rat intramural coronary resistance arteries. 6. Extraluminal and intraluminal administration of bradykinin with continuous intraluminal flow in rat intramural coronary resistance arteries. Protocols 1, 5 and 6 were performed in the Institute of Human Physiology and Clinical Experimental Research, Semmelweis University, Faculty of Medicine, Budapest, Hungary and protocols 2, 3, and 4 were performed in the Department of Physiology, New York Medical College, Valhalla, NY, USA.

- 74 -

Protocol No.1: Biomechanical wall characteristics of rat intramural coronary resistance arteries and arterioles. In this protocol LAD segments (n=31) were prepared from different branching levels with different morphological sizes (see 4.1.4.). Briefly, main branches (n=5), proximal end branches (n=9), distal end branches (n=4), first order side branches (n=8) and second order side branches (n=5) were prepared (233). During the series of measurements one of the cannulas was kept closed (no flow measurements), but fresh solution was flushed through the lumen between the measurements. All segments were incubated at 50 mmHg intraluminal pressure for 30 minutes. During the incubation period, all segments developed active spontaneous tone. Then conditioning pressure cycles were applied by increasing and then decreasing intraluminal pressure between 0150-0 mmHg values at a slow rate of about 0.5 mmHg/sec by infusing nKR solution into the lumen or removing it through the other cannula with the aid of an airpressurized reservoir vessel and a continuous flow pump. Myogenic response was often observed in the first cycle, while its extent diminished in the forthcoming cycles as an effect of the mechanical conditioning of the segments. With this methodology biomechanical wall characteristics could be studied (233), such as in other previous studies (20, 33, 48, 49, 60, 81, 166, 167, 179, 180). Three cycles were applied in oxygenized nKR tissue bath and the upward loop of the second cycle was analyzed (spontaneously contracted segment). Then the pressure was set again at 50 mmHg and 7.5 x 10-6 M PGF2α was added into the tissue bath. This dose induced maximal contraction in pilot experiments. The ensuing contraction was recorded for 10 minutes and then the pressure cycles were repeated. The upward loop of the second cycle was analyzed for quantifying PGF2α - induced contraction. Then the drug was washed out and papaverine was added into the tissue bath in a concentration of 2.8 x 10-4 M. Intraluminal pressure cycles were repeated and the upward loop of the third cycle was analyzed in order to characterize the properties of the fully relaxed segment. Original plots from a characteristic segment are shown on Fig. 5.1.1.-1. At the end of the experiments vessel length and weight was measured (4.1.2.) for further calculations (4.1.7.).

- 75 -

Protocol No. 2: Myogenic characteristics of rat intramural coronary arterioles. Intramural LAD coronary arteriole segments (n=28, distal, apical segments, see 4.1.4.) were prepared and pressurized in no-flow condition and superfused with PSS at a rate of 15 ml/min (234). Arterioles were allowed to equilibrate for 1 hour in PSS at 60 mmHg intraluminal pressure, while all segments included in the study developed active spontaneous tone. Then intraluminal pressure was decreased and increased from 2 to 150 mmHg in 10 mmHg pressure steps and the steady state diameter at each step was measured (after ~5 min). Nω-nitro-L-arginine (L-NNA, 10-5 M), an inhibitor of nitric oxide synthase, indomethacin (INDO, 2.8 x 10-5 M), an inhibitor of prostaglandin synthesis and adenosine (ADO, 5 x 10-5 – 5 x 10-4 M) were administered at 60 mmHg intraluminal pressure into the superfusion fluid then incubated with each for 20-30 min. In the first group of experiments (n=10) after the control pressure-diameter relationship was obtained in PSS, the pressure steps were repeated in the presence of L-NNA and then again with the simultaneous presence of L-NNA and INDO. In the second group of experiments (n=10) after control responses, the pressure-diameter relationships were obtained first in the presence of INDO, and then in the presence of both INDO and LNNA. In the third group of experiments (n=8) after control responses, the pressurediameter relationships were obtained first in the presence of L-NNA, and then in the additional presence of ADO. The concentration of ADO was adjusted to allow the segment to reach its control diameter (in PSS) at 60 mmHg and then also at 2 mmHg. PD was obtained in Ca2+-free PSS. The vessels were incubated for 20 minutes, step increases in intraluminal pressure were repeated and the PD of arterioles at each pressure step was obtained. Protocol No. 3: Myogenic characteristics of rat intramural coronary arterioles, effects of daily exercise. In this study LAD coronary arteriole segments (distal, apical segments, see 4.1.4.) of exercised (EX, n=12) and sedentary (SED, n=20) rats were compared (116). The same protocol with L-NNA and INDO was used as in protocol No. 2. Briefly, after equilibration period steps of 10 mmHg of intraluminal pressure were obtained from 2 to 150 mmHg and the steady state diameter at each step was measured. L-NNA and INDO

- 76 -

were administered at 60 mmHg pressure into the superfusion fluid. In the first protocol after the control pressure-diameter relationship was obtained in PSS, the pressure steps were repeated in the presence of L-NNA and then again with the simultaneous presence of L-NNA and INDO. In another group of experiments after control responses, the pressure-diameter relationships were obtained first in the presence of INDO, then in the presence of both INDO and L-NNA. PD was obtained in Ca2+-free PSS. An original tracing of a myogenic curve appears on Fig. 5.1.2.1.-1. Protocol No. 4: Endothelium-dependent and endothelium-independent vasoactive responses of rat intramural coronary arterioles. Intramural LAD coronary arteriole segments (n=22, distal, apical segments, see 4.1.4.) were prepared and pressurized in no-flow condition and superfused with PSS at a rate of 15 ml/min (235). Arterioles were allowed to equilibrate for 1 hour in PSS at 60 mmHg intraluminal pressure and this pressure was maintained throughout the experiments. All segments included in the study developed active spontaneous tone. In the first group of experiments (n=9) dose-response curves of bradykinin (BK, 10-9-10-5 M), acetylcholine (ACh, 10-9-10-4 M) and sodium-nitroprusside (SNP, 10-9-10-4 M) were obtained. Agents were administered extraluminally, while superfusion was stopped, and each administration was followed by a 5-minute washout period. Then, Nω-nitro-Larginine (L-NNA, 10-5 M) was administered into the superfusion fluid. After 20-30 min. incubation, dose-response curves of BK, ACh and SNP were repeated. In a second group of experiments (n=7) dose-response curves of adenosine (ADO, 10-9-10-4 M), norepinephrine (NE, 10-9-10-4 M), substance P (SP, 10-9-10-5 M), prostaglandin E2 (PGE2, 10-9-10-5 M), prostaglandin I2 (PGI2, 10-9-10-5 M) and indomethacin (INDO, 107

-10-4 M) were obtained. In another set of experiments (n=6), effects of SQ29,548 (10-6

M) was obtained. This dose of SQ29,548 completely blocked the 10-7 M U46619induced constriction. PD was obtained in Ca2+-free PSS. Protocol No. 5: Endothelium-dependent and endothelium-independent vasoactive responses of rat intramural coronary resistance arteries. Small intramural LAD coronary artery segments (n=15, proximal end branches, see 4.1.4.) were prepared and pressurized in no-flow condition and superfused with PSS

- 77 -

at a rate of 15 ml/min (57). Arteries were allowed to equilibrate for 1 hour in PSS at 60 mmHg intraluminal pressure and this pressure was maintained throughout the experiments. All segments included in the study developed active spontaneous tone. In some experiments, a complete dose-response curve of U46619 (5x10-10 M - 10-6 M) was obtained in order to calculate the optimal dose of preconstriction. After preconstriction with U46619 (5x10-8 M - 10-7 M), dose-response curves of bradykinin (BK, 10-9-10-5 M), acetylcholine (ACh, 10-9-10-4 M), adenosine (ADO, 10-9-10-4 M), norepinephrine (NE, 10-9-10-4 M) and sodium-nitroprusside (SNP, 10-9-10-4 M) were obtained (this latter in separate experiments). Agents were administered extraluminally, while superfusion was stopped, U46619 was administered before each dose-series, and administration of each dose-series was followed by a 5-minute washout period. Then, Nω-nitro-L-arginine (L-NNA, 10-5 M) was administered into the superfusion fluid. After 20-30 min. incubation, dose-response curves of BK, ACh, ADO, NE and SNP (this latter in separate experiments) were repeated. PD was obtained in Ca2+-free PSS. Protocol No. 6: Extraluminal and intraluminal administration of bradykinin with continuous intraluminal flow in rat intramural coronary resistance arteries. In this protocol we introduced a simple method of intraluminal drug administration (56). Small intramural coronary artery segments (n=10, proximal end branches, see 4.1.4.) were allowed to equilibrate for 30 minutes and a continuous flow was applied throughout the experiment. Inflow pressure was provided with a pressure servo-controlled pump (Living Systems Inc.) and was continuously set to 60 mmHg, whereas outflow from the segment entered a reservoir the pressure of which was kept constant at 40 mmHg by hand-made adjustments when needed using a continuous withdrawal flow pump (Harvard Apparatus). This arrangement resulted in an approx. 50 mmHg intraluminal pressure and a continuous flow of about 1 µl/sec, which was maintained throughout the experiment. Serotonin, and then with serotonin in the bath, bradykinin were given extraluminally in cumulative doses (10-10-10-5 M), then additional intraluminal 10-5 M bradykinin infusion was applied. Ten minutes were allowed for the effect to develop fully for each dose. At intraluminal application the agent was infused via a side-cannula glued to the main line, using a continuous flow pump (Cole-Parmer) at a rate of 2 µl/sec which exceeded the original flow from the

- 78 -

inflow pressure-controlled pump. The drug solution immediately reached the conic end of the inflow cannula and then the lumen of the vessel. It also filled up slowly the inflow line in backward direction. The servo control pump kept the preset, 60 mmHg pressure by backward movement. This arrangement enabled us to apply the drug intraluminally, with minimum dead space and with no disturbance to either pressure or flow. At the final stage of the protocol, Ca2+-free KR solution was added and the relaxed diameter of the segments was measured. 4.1.7. Data processing In protocols No. 1 and 6 after A/D conversion diameter and pressure values were recorded on Pentium PC using Labtech Notebook software. In protocols No. 2, 3, 4 and 5 diameter and pressure values were recorded on a chart recorder. Further computations were made with Excel 5.0 or 6.0 and Cricket Graph programs. Protocol No. 1. Pressure-diameter recordings were smoothed with a curve fitting method (Fig. 5.1.1.-1b.). From pressure-diameter curves and wall thickness data (obtained from wet and dry weight and length measurements of the segments, see also 4.1.6.), further calculations were made. Inner radius, wall thickness, elastic modulus, distensibility, tangential wall stress, spontaneous myogenic tone, active tone (against nKR and against full relaxation) were calculated and plotted as a function of the intraluminal pressure for spontaneously contracted, PGF2α-contracted, and papaverine-relaxed segments. These biomechanical parameters were calculated as follows (167, 168). RO = D 2 , where D is the measured outer diameter and Ro is the outer radius expressed in micrometers. RI = (RO2 − m ρ lπ ) ,

where RI is the inner radius, Ro is the outer radius, m is the wet weight of the vessel segment measured as described above,

ρ

is the specific weight of the vascular tissue

which was taken to be 1.06 g/cm3, and l is the length of the segment between the two ties.

- 79 -

h = RO − RI , where h is the wall thickness and Ro and RI are the outer and inner radii.

Ftang = PRI , where Ftang is the circumferential (tangential) tension, P is the intraluminal pressure and RI is the inner radius.

σ tang = PRI h , where σ tang is the average circumferential (tangential) wall stress, P is the intraluminal pressure, RI is the inner radius and h is the wall thickness.

Dinc = ∆V V∆P , where Dinc is the incremental distensibility, ∆V is the change in vessel lumen volume in relation to the initial volume of V in response to pressure change of ∆P. 2

2

2

Einc = 2RI RO∆P (RO − RI )∆RO

where Einc is the incremental elastic modulus, Ro and RI are the outer and inner radii,

∆Ro is the change in outer radius in response to a pressure change of ∆P. Spontaneous tone and PGF2α-induced tone of the segments were computed as percent contraction of the outer diameter compared with that in papaverine relaxation at the same pressure level (this latter is expressed as active strain). Comparisons between PGF2α -induced tone and spontaneous tone (nKR) were also made. Outer radius and circumferential tension data were also normalized to the values measured at 150 mmHg (6.1.1.). Protocols No. 2 and 3. Diameter values were also normalized to the corresponding passive diameter and expressed as percent values. Myogenic tone was defined as percent contraction of passive diameter. In order to further characterize the myogenic response, myogenic index (MI) was also calculated as follows (51, 105, 162, 227): MI=100x[(r2-r1)/r1]/(P2P1), where r1, r2 are the radii measured between two consecutive pressure steps (P1 and P2). Protocols No. 4 and 5.

- 80 -

Diameter values were also expressed as changes from control values. Spontaneous myogenic tone of the segments was computed as percent contraction of the passive diameter. Changes in percent of passive diameter were also calculated for some doseresponses (6.1.3.). Protocol No. 6. Active diameter values for each dose were calculated as percent values of passive diameter. Spontaneous myogenic tone of the segments was computed as percent contraction of the passive diameter. 4.1.8. Statistical analysis Protocol No. 1. Statistical analysis of the data was made with linear regression as a function of vascular caliber, computing the significance levels of the correlation coefficients, comparing the variances between different groups of vessels with ANOVA and with paired and unpaired t tests. Differences were declared to be significant at p values < 0.05. Values are expressed as means ± SEM. Protocols No. 2 and 3. Data are presented as means ± SEM. Statistical significance was calculated with repeated measures ANOVA followed by Tukey`s post hoc test. Paired and unpaired t tests were applied for paired and unpaired data. P < 0.05 was considered statistically significant. Protocols No. 4 and 5. Data are presented as means ± SEM. For statistical analysis paired t test and ANOVA were applied. P < 0.05 was considered to declare statistically significant difference. Protocol No. 6.

- 81 -

Data are presented as means ± SEM. For statistical analysis paired t test and one way ANOVA with Fisher’s post hoc test were applied. P < 0.05 was considered to declare statistically significant difference. 4.2. Isolated heart studies 4.2.1. Animals Male Wistar and Sprague-Dawley rats weighing around 250-350 grams were used. In the study performed at the University of Tampere (Finland) Wistar rats were used bred in the animal house of the University. In the study performed in the Institute of Human Physiology and Clinical Experimental Research, Semmelweis University, Faculty of Medicine, Sprague-Dawley rats were used, approved from Charles River Laboratories and further bred in the animal house of the Institute of Human Physiology. In ischemia-reperfusion studies (protocol No. 2, 4.2.5.) after ip. 50 U/kg heparin (Sigma Ltd.) injection rats were killed by dislocation of the neck. All other rats (in protocol No. 1, 4.2.5.) were anaesthetized with Pentobarbital (50 mg/kg ip.) after ip. heparin administration. 4.2.2. Equipment In the description of the equipment we refer to the numbers of different experimental protocols we used (4.2.5.). Measurement of pressure and flow In protocol No. 1 rat hearts were perfused with constant pressure according to Langendorff method (Fig. 4.2.-1.). Perfusion system built by Experimetria Ltd. (Budapest), and was further modified in our department. Perfusion pressure (PP) was measured by Gould or Electromedics pressure transducer and maintained at 60 mmHg with perfusion reservoirs. Two reservoirs were set in the system in order to switch between different solutions. Left ventricular pressure (LVP) was measured by Gould or Electronics pressure transducer connected to a latex or PVC balloon inserted into the

- 82 -

left ventricle and also connected to a syringe reservoir in order to set end diastolic pressure (preload of the heart). Coronary flow (CF) was measured by a 2 mm flow probe put into the route of the perfusion fluid close to the heart and connected to an ultrasound flowmeter (Transonic T 106, Itaha, NJ). In protocol No. 2 rat hearts were perfused with constant flow. Perfusion system was built in the department (at the University of Tampere). LVP was measured by an ISOTEC pressure transducer (Hugo Sachs Electronic, Hugstetten, Germany) and PP was measured by a Harvard Apparatus 377 type pressure transducer. Constant flow was set and maintained by an Ismatec peristaltic pump. Calibrations were made with mercury manometer and a graduated cylinder. Infusion of drugs Drugs were infused into the perfusion cannula through a fixed side cannula close to the heart with a Cole Parmer infusion pump with the rate of 1 % of the actual coronary flow. Recordings In protocol No. 1 data were recorded on Grass polygraph recorder. In protocol No. 2 data were digitalized onto an IBM personal computer. PP recordings were made on Grass polygraph and on IBM PC in the two sets of experiments, respectively. Calibrations were made with mercury manometer. 4.2.3. Solutions and drugs Perfusion of isolated hearts in protocol No. 2 was made by Krebs-Henseleit solution containing in mmol/l: NaCl 118, KCl 4.7, CaCl2 2.52, Mg2SO4 1.66, NaHCO3 24.88, KH2PO4 1.18, and glucose 11 (Sigma Ltd.). In protocol No. 1 a modified KrebsHenseleit solution was applied containing in mmol/l: NaCl 118, KCl 4.7, CaCl2 3.0, Mg2SO4 1.2, NaHCO3 25, KH2PO4 1.2, Na-EDTA 0.5, D-glucose 10 and Na-pyruvate 1.5 (Sigma Ltd.). It was thermostated at 37 oC and bubbled with 95 % O2 and 5 % CO2 gas mixture. The pH was adjusted to 7.4, pO2 was maintained between 400-500 mmHg.

- 83 -

Adenosine, acetylcholine, bradykinin, and nitro-L-arginine were obtained from Sigma Ltd. (St. Louis, MO). GEA 3162 was obtained from GEA Ltd. (Copenhagen, Denmark). Chemicals were prepared on the day of the experiment by dissolving in Krebs. GEA 3162 (267.5 microgram) was dissolved in 1 ml of dimethylsulfoxide (DMSO) which was further diluted in Krebs-Henseleit solution. This dose of DMSO alone had no effect in preliminary studies. 4.2.4. Preparation and mounting Rat hearts were rapidly excised, placed into cold, oxygenized Krebs solution and fixed to the perfusion cannula on a low flow perfusion. Perfusion was started immediately. Hearts were cleared from extra tissue and pulmonary artery was cleared free. A latex or PVC balloon was inserted into the left ventricle through a small slit on the left atrium for measuring the left ventricular pressure. In this setup perfusion fluid entered the coronary circulation by closing aortic valves retrogradely, thus drained mostly through the pulmonary artery. 4.2.5. Experimental protocols In this study we used the following protocols: 1. Endothelium-dependent and endothelium-independent vasoactive responses of the rat heart 2. Effects of an NO donor, GEA 3162 on the biomechanical performance of the isolated ischemic rat heart Protocol No. 1 was performed in the Institute of Human Physiology and Clinical Expreimental Research, Semmelweis University, Faculty of Medicine, Budapest, Hungary, and protocol No. 2 was performed in the Department of Pharmacology, Clinical Pharmacology and Toxicology, University of Tampere, Finland.

- 84 -

Protocol No. 1: Endothelium-dependent and endothelium-independent vasoactive responses of the rat heart Hearts were perfused according to Langendorff with constant pressure (n=10). Perfusion pressure was kept at 60 mmHg. Hearts were allowed to stabilize for 30 minutes. Adenosine (ADO, 2.5 x 10-8-2.5 x 10-6, bradykinin (BK, 10-8-10-6 M) and acetylcholine (ACh, 5.5 x 10-8-10-6 M) were infused into the coronary system in a dosedependent manner. Agents were infused until the maximum response was reached (~ 5 minutes) followed by a 5-minute washout period. Then, Nω-nitro-L-arginine (L-NNA, 10-4 M), an inhibitor of nitric oxide synthesis was administered into the perfusion fluid (by switching to the other reservoir). After 10-minute stabilization, dose-responses were repeated with a continuous perfusion with L-NNA. Protocol No. 2: Effects of an NO donor, GEA 3162 on the biomechanical performance of the isolated ischemic rat heart Hearts were perfused according to Langendorff with constant flow (n=33, 232). The perfusion flow was kept at 16 ml/min. Hearts were allowed to stabilize for 30 minutes. Ischemia was induced by decreasing the flow rate to 0.8 ml/min for 30 minutes which was followed by a 40 and 45 minute reperfusion period in two sets of experiments, respectively. Hearts were divided into two groups: control and GEA 3162 - treated specimens. Intracoronary infusion of GEA 3162 was started 5 minutes before ischemia and was given until the end of the experiments at a rate of 1% of the actual coronary flow entering the heart at 10-5 M concentration. This concentration has been found to be optimal in inducing coronary vasodilation in Langendorff preparations. In the first set of experiments the experimental protocol was as follows: Maximal LVP values were recorded in control state, in 5th minute of ischemia and in 1st, 2nd, 3rd, 5th, 10th, 15th, 20th, 25th, 30th and 40th minute of reperfusion. Ten contraction cycles were recorded in each case. Ten control and ten GEA 3162-treated hearts were studied. In the second set of experiments PP recordings were made for 5 minutes in every 30 seconds, except in ischemia where samples were taken in every 5 minutes. Intracoronary BK infusion was given for 5 minutes at 10-6 M concentration before

- 85 -

ischemia. The infusion of GEA 3162 was started in the GEA 3162-treated group after BK infusion and it was continued until the 30th minute into reperfusion. Then after a 5 minute washout-period BK infusion was introduced again for 5 minutes. Seven control and six GEA 3162-treated hearts were studied in this set of experiments. 4.2.6. Data processing and calculations Protocol No. 1. Data were further processed in Excel 6.0 and Cricket Graph programs. Data were also expressed as changes and percent changes from control values. Protocol No. 2. After acquisition of data on an IBM PC it was processed by using a specific program (Cardiotonic Tester Measure and Analysis Software made by Arvi Salo, Tristar Enterprise Corporation, Tampere, Finland). Data were further processed in Excel 5.0 and Fig-P programs. Data were also expressed as percentage or normalized values of initial values. Protocols No. 1 and 2. The heart rate (HR), the first derivative of LVP (dP/dt), the maximal and minimal values of dP/dt were calculated and monitored. 4.2.7. Statistical analysis Protocol No. 1. Data are presented as means ± SEM. For statistical analysis paired t test and ANOVA were applied. P < 0.05 was considered to declare statistically significant difference. Protocol No. 2. The results were analyzed by the SOLO statistical program applying repeated measures ANOVA and Student's t-tests. Significanct differences were considered at p

- 86 -

values less than 0.05. Values are expressed as means ± SEM. At the beginning of the reperfusion period fibrillation did occur in some cases. These hearts were excluded from further evaluation. Minor arrhythmia could be seen in most of the experiments in connection with ischemia and reperfusion. The abnormal beats were excluded from the analysis.

- 87 -

5. Results 5.1. Isolated vessel studies 5.1.1. Biomechanical wall characteristics of rat intramural coronary resistance arteries and arterioles (protocol No. 1) Original pressure-diameter plots are presented in Fig. 5.1.1.-1a. Panel b. represents mathematical smoothing of the original plot. Outer radii of intramural small coronary arteries from different sites along the branching system of the LAD coronary artery were significantly different at all pressure levels (fig. 5.1.1.-2.). Mean ± SEM values at 100 mmHg intraluminal pressure were: main branches, 288 ± 38 µm; proximal end branches, 203 ± 5 µm; distal end branches, 140 ± 15 µm; first order side branches, 135 ± 30 µm; second order side branches, 99 ± 2 µm (p < 0.001 when the five groups were compared). However, there was no significant difference between the course of the curves corresponding to the first order side branches and the distal end branches. In order to get a better resolution of the parameters, grouping was reaccomplished according to the vessel caliber measured in nKR at 100 mmHg. Vessels have been divided into four groups with inner diameters of 50-150, 150-250, 250-350, and >350 µm. Fig. 5.1.1.-3. shows inner radius values as a function of intraluminal pressure for spontaneously contracted (a.) and for fully relaxed segments (b.). Wall thicknesses at the same time did not change with caliber in the range of small arteries studied (not shown), marking a characteristic change in their geometric proportions. Wall thicknesses at 100 mmHg intraluminal pressure were 33 ± 4, 27 ± 5, 27 ± 4, 30 ± 4 µm, for the four caliber groups (100-200-300-400 µm), respectively. Large differences in inner radii, no differences in wall thickness resulted in significant differences in the average circumferential wall stress values as a function of morphological caliber (significance level of the correlation coefficient, p < 0.001 for each pressure level, both in spontaneously contracted and in relaxed segments (Fig. 5.1.1.-4a., b.) Incremental distensibility decreased with increasing intraluminal pressure (Fig. 5.1.1.-5.). The vessels with smallest caliber (50-150 µm group) were significantly more

- 88 -

distensible than larger vessels when compared in the intraluminal pressure range of 70130 mmHg (p < 0.05). Vessels with smaller calibers presented significantly larger incremental distensibility. The significance level of the linear correlation coefficient in intraluminal pressure range of 70-130 mmHg was p < 0.05. Incremental distensibilities measured at 100 mmHg intraluminal pressure were 35.3 ± 12.1, 18.6 ± 6.6, 15.6 ± 3.1 and 8.1 ± 2.5 x 10-6 m2/N for vessels of groups with calibers around 100, 200, 300, 400 µm, respectively. Distensibility (measured at the same intraluminal pressures) did not seem to depend on morphological caliber in case of relaxed vessels. Incremental elastic modulus increased with increasing intraluminal pressure and with increasing vessel caliber (Fig. 5.1.1.-6.). Elastic moduli measured at 100 mmHg intraluminal pressure were 0.22 ± 0.05, 1.01 ± 0.33, 1.12 ± 0.14 and 3.03 ± 0.77 x 106 N/m2 for the four groups, respectively (p < 0.001). Elastic moduli of relaxed segments also showed significant correlation with vascular caliber at intraluminal pressures of 40-140 mmHg (p < 0.01). All segments expressed spontaneous tone. (Pressure-radius curves obtained in nKR solution were substantially different from the papaverine-relaxed curves). Percent differences in radii are shown in Fig. 5.1.1.-7a. Vessels of the small caliber group expressed significantly higher spontaneous tone than larger vessels (p < 0.05). At 100 mmHg intraluminal pressure spontaneous contraction of outer radii of these mechanically preconditioned vessels was 18.8 ± 4.5, 8.4 ± 4.4, 9.7 ± 3.7 and 8.3 ± 3.8 % for the four groups, respectively. When PGF2α (7.5 x 10-6 M) was given to spontaneously contracted segments at 50 mmHg intraluminal pressure, a further significant contraction could be observed (p < 0.05), which was maximal at calibers of 250-350 µm compared with the other groups (p < 0.05, Fig. 5.1.1.-7c.). PGF2α - induced contraction measured during intraluminal pressure cycles was significantly higher in the 250-350 µm group when compared with the others at 80-110 mmHg pressures (p < 0.05, not shown). Vessels with smaller and wider morphological calibers seemed to express less PGF2α-induced tone compared with state in spontaneous tone. PGF2αcontracted and papaverine-relaxed conditioned pressure-diameter curves are compared in Fig. 5.1.1.-7b. Maximal contractions showed a significant inverse correlation with morphological caliber in the 30-50 mmHg intraluminal pressure (p < 0.05).

- 89 -

Fig. 5.1.1.-8. and 9. summarize the geometric and elastic properties of small intramural coronary arteries as a function of caliber (anatomical size, for the four caliber groups of 100-200-300-400 µm, respectively), intraluminal pressure and contractile state during repeated cycles of intraluminal pressure as they were found in these studies. Contraction decreased, relaxation increased the isobar incremental tangential elastic moduli (p < 0.05 at each pressure level, Fig. 5.1.1.-9). 5.1.2. Myogenic characteristics of rat intramural coronary arterioles (protocols No. 2 and 3) 5.1.2.1. Modulatory role of endothelium Intramural coronary arterioles of rats developed active spontaneous tone of 29.0 ± 1.7 % of their passive diameter (171.8 ± 3.7 µm) at 60 mmHg. Arterioles exhibited significant changes in diameter to step changes in intraluminal pressure from 2 to 150 mmHg. A typical recording of the effect of increases in pressure on the diameter of arterioles is depicted in Fig. 5.1.2.1.-1. Fig. 5.1.2.1.-2a. shows that the diameter of coronary arterioles increased in response to low pressures. Higher intraluminal pressures (above 30-40 mmHg) elicited constrictions which became maximal at ~ 60 mmHg and were maintained up to 150 mmHg intraluminal pressure. At 60 mmHg, the active diameter of arterioles was 67.3 ± 2.7 % of their passive diameter (Fig. 5.1.2.1.-2b.). To characterize the strength of myogenic constrictions, myogenic index (Fig. 5.1.2.1.-2c.) was calculated. The maximum negative peak of the myogenic index was around 50-70 mmHg intraluminal pressure. Role of nitric oxide (and prostaglandins) Administration of the nitric oxide synthase inhibitor L-NNA significantly reduced the initial diameter of arterioles at 2 mmHg by ~30 % compared to control and inhibited the dilation to lower intraluminal pressures (Fig. 5.1.2.1.-3., p < 0.05). At 60 mmHg L-NNA significantly constricted coronary arterioles by 42.5 ± 4.3 % (p < 0.05). Further increases in intraluminal pressure with L-NNA did not decrease the diameter as they did in the control condition. In the presence of L-NNA, additional administration

- 90 -

of indomethacin (INDO), an inhibitor of prostaglandin synthesis, shifted the pressurediameter curve upward (from the one obtained with L-NNA alone) starting from 60 mmHg to 150 mmHg, the difference between the curves becoming significant at higher pressure values (above 110 mmHg, p < 0.05). In a separate group of experiments, in the presence of L-NNA, an appropriate concentration of adenosine (ADO, 5 x 10-5 to 5 x 10-4 M) was administered in order to restore the initial diameter (at 2 mmHg) to the level prior to administration of L-NNA (Fig. 5.1.2.1.-4.). Then the pressure-diameter relationship was obtained in the simultaneous presence of L-NNA and ADO. We found that although the initial diameter of arterioles was restored by ADO in the presence L-NNA, the dilations in response to lower pressures and constrictions to higher pressures, observed in the control condition, were still absent. Role of prostaglandins (and nitric oxide) Administration of the prostaglandin synthesis inhibitor indomethacin (INDO) alone had no significant effect on the initial diameter and the increase in diameter of coronary arterioles at lower pressures (below 50 mmHg), but it elicited a significant upward shift of the pressure-diameter curve at the higher pressure range (above 50 mmHg) compared to that of control (at 70 and above 110 mmHg, p < 0.05, Fig. 5.1.2.1.5.). In the presence of INDO, additional administration of L-NNA significantly reduced the initial diameter by ~30% (at 2 mmHg, p < 0.05) and eliminated the constrictions to higher pressures. 5.1.2.2. Effects of daily exercise The body weight, wet heart weight, heart weight to body weight ratio of SED and EX rats were 405.3 ± 8.5g, 1.32 ± 0.04g, 3.26 ± 0.1g/kg and 382.8 ± 6.7g, 1.3 ± 0.03g, 3.4 ± 0.07g/kg respectively (p