Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Role of cGMP-dependent protein kinase in the negative inotropic effects of cGMP-elevating agents in the… MacDonell, Karen Loraine 1996

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1996-091244.pdf [ 8.24MB ]
Metadata
JSON: 831-1.0087291.json
JSON-LD: 831-1.0087291-ld.json
RDF/XML (Pretty): 831-1.0087291-rdf.xml
RDF/JSON: 831-1.0087291-rdf.json
Turtle: 831-1.0087291-turtle.txt
N-Triples: 831-1.0087291-rdf-ntriples.txt
Original Record: 831-1.0087291-source.json
Full Text
831-1.0087291-fulltext.txt
Citation
831-1.0087291.ris

Full Text

ROLE OF cGMP-DEPENDENT PROTEIN KINASE IN THE NEGATIVE INOTROPIC EFFECTS OF cGMP-ELEVATING AGENTS IN THE MAMMALIAN VENTRICLE by K A R E N L O R A I N E M A C D O N E L L B. Sc. (Pharm.), Dalhousie University, 1984 M. Sc. (Pharm. Sci.), University of British Columbia, 1991 A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in The Faculty of Graduate Studies Division of Pharmacology and Toxicology Faculty of Pharmaceutical Sciences We accept this thesis as conforming to the required standard University of British Columbia February, 1996 © Karen L. MacDonell, 1996 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholady purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. DE-6 (2/88) ABSTRACT The role of guanosine 3':5'-cyclic monophosphate (cGMP) and c G M P -dependent protein kinase (PKG) in the regulation of cardiac contractility is controversial. There is evidence in the literature which suggests that c G M P is involved in the muscarinic receptor agonist-mediated antagonism of the positive inotropic effects of cAMP-elevating agents in the mammalian ventricle. Support for this relationship includes observations that a.) acetylcholine elevates c G M P and mediates a negative inotropic effect in (3-adrenoceptor-stimulated ventricular tissue in a time- and concentration-dependent manner and b.) some of the contractile effects of muscarinic receptor agonists can be mimicked by c G M P analogues and inhibited cGMP-lowering agents. The proposal that c G M P can mediate negative inotropic effects is difficult to reconcile with reports the nitrovasodilator sodium nitroprusside (SNP) markedly increased tissue c G M P levels in several ventricular preparations but had no effect on contractile activity (Rodger and Shahid, 1984; Lincoln and Keely, 1980, 1981). A possible explanation for these conflicting results was proposed by Lincoln and Keely (1980, 1981). They hypothesized that some cGMP-elevating agents, such as muscarinic receptor agonists, increase c G M P in a functional pool which can specifically activate P K G and, thereby, mediate a negative inotropic effect. Cyclic GMP-elevating agents, such as SNP, which fail to elevate c G M P in the requisite pool would be unable to activate P K G and, as a result, would not reduce ventricular contractility. The aim of the present study was to test this hypothesis. The experimental approach was to compare the effects of carbachol, S N P and atrial natriuretic peptide (ANP) on contractility, c G M P content and P K G activity in rat intact ventricular tissue and freshly isolated rat ventricular cardiomyocytes. Carbachol induced a marked negative inotropic effect in intact, perfused hearts, ventricular strips and isolated cardiomyocytes in the presence of isoproterenol. The negative inotropic effect of carbachol in the intact ventricle was associated with insignificant changes in tissue c G M P content but, in isolated cardiac myocytes, with significant, although small, increases in cellular c G M P levels. S N P and A N P had no effect on isoproterenol-stimulated contractility in ventricular strips and intact hearts. Furthermore, S N P did not change the positive inotropic effect of isoproterenol in isolated cardiomyocytes. The absence of negative inotropic effects by A N P and S N P was observed in the presence of marked increases in intact ventricular c G M P levels. S N P also increased c G M P levels by up to 8-fold in isolated cardiomyocytes. The presence of P K G in both the intact ventricle and in isolated ventricular cardiomyocytes confirmed by MonoQ anion exchange chromatography and Western blotting. The elution profile indicated that the conditions of the P K G assay were very selective for measuring P K G activity. Attempts were made to confirm the reliability of the P K G assay, after which agonist-mediated P K G activation was assessed. Carbachol had no significant effect on P K G activity at a concentration which exerted a marked negative inotropic effect in isoproterenol-stimulated ventricular tissue and cardiomyocytes. Conversely, P K G was activated in the presence of S N P or A N P in ventricular tissue and was activated by S N P in myocytes and this activation occurred without changes in the positive inotropic iii effects of isoproterenol. P K G activation may have occurred in the presence of carbachol but may have been underestimated by the conditions of the assay. The results of this study demonstrate that the negative inotropic effects of muscarinic receptor agonists occur in the presence of a small increase in c G M P levels and in the absence of significant activation of P K G . Larger increases in ventricular c G M P content and P K G activity were not sufficient to mediate a negative inotropic effect in the presence of S N P or ANP. Under the conditions of this study, P K G activation cannot be ruled out as playing a role in the negative inotropic effects of muscarinic receptor agonists but elevation of c G M P content and activation of P K G are not sufficient to inhibit contractility in the rat ventricle. iv TABLE OF CONTENTS Page ABSTRACT ii TABLE OF CONTENTS v LIST OF FIGURES xi LIST OF TABLES xiii LIST OF ABBREVIATIONS xiv ACKNOWLEDGEMENTS xviii DEDICATION xix 1.0 INTRODUCTION 1 1.1 cGMP-elevating systems. 2 1.2 Effects of cholinergic agonists in the heart. 5 1.2.1 Effects of muscarinic receptor agonists on atrial contractility. 6 1.2.2 Possible mechanisms for cholinergic inhibition of atrial contractility. 7 1.2.2.1 Modulation of ion channels. 7 1.2.2.2 Elevation of c G M P levels. 8 1.2.3 Effects of muscarinic receptor agonists on ventricular contractility. 9 1.2.4 Possible mechanisms for cholinergic inhibition of ventricular 10 contractility. 1.2.4.1 Decreased cAMP levels. 10 1.2.4.2 Phosphoprotein phosphatase activation. 12 1.2.4.3 Cyclic cGMP. 13 1.2.4.3.1 Role of c G M P in regulation of amphibian 13 ventricular contractility. v Page 1.2.4.3.2 Role of c G M P in regulation of mammalian 14 ventricular contractility. 1.3 Effects of c G M P analogues on ventricular contractility. 16 1.4 Nitric oxide and the ventricle. 18 1.4.1 Effect of nitric oxide donors on ventricular contractility. 18 1.4.2 Relationship between SNP-mediated c G M P elevation and 21 ventricular contractility. 1.5. Atrial natriuretic peptide and the ventricle. 23 1.5.1 Effects of A N P on ventricular contractility. 23 1.5.2 Relationship between ANP-mediated c G M P elevation and 25 ventricular contractility. 1.6 Cyclic GMP-dependent protein kinase and ventricular contractility. 27 1.6.1 Evidence for the presence of cGMP-dependent protein kinase in 28 the mammalian ventricle. 1.6.2 Evidence for a role of cGMP-dependent protein kinase in 29 negative inotropic effects of cGMP-elevating agents. 1.6.2.1 Regulation of l C a . 29 1.6.2.2 Phosphorylation of regulatory proteins. 29 1.6.2.3 cGMP-dependent protein kinase inhibitors. 31 1.6.2.4 Measurement of cGMP-dependent protein kinase 32 activity in intact ventricular tissue. 1.7 Improvement in cGMP-dependent protein kinase assay. 33 1.8 Summary and rationale for proposed experiments. 35 vi Page 2.0 MATERIALS AND METHODS 37 2.1 Chemicals and materials. 37 2.2 Animals. 40 2.3 Measurement of contractile properties of rat cardiac preparations. 40 2.3.1 Preparation of Langendorff-perfused rat hearts. 40 2.3.2 Measurement of contractility of Langendorff-perfused rat hearts. 41 2.3.3 Preparation of rat right ventricular strips. 41 2.3.4 Measurement of contractility of rat right ventricular strips. 42 2.3.5 Preparation of single rat ventricular cardiomyocytes. 42 2.3.6 Measurement of contractility of single isolated rat ventricular 44 cardiomyocytes. 2.4 c G M P and cAMP estimation. 45 2.4.1 c G M P and cAMP estimation in rat ventricular tissue. 45 2.4.2 c G M P and cAMP estimation in suspensions of rat ventricular 46 isolated cardiomyocytes. 2.5 Preparation of soluble and particulate fractions of rat cardiac tissue. 47 2.5.1 Preparation of soluble and particulate fractions of rat ventricular 47 tissue. 2.5.2 Preparation of soluble and particulate fractions of rat isolated 48 ventricular cardiomyocytes. 2.6 c G M P - and cAMP-dependent protein kinase assay. 49 2.7 Chromatographic separation of protein kinases in rat ventricular extracts. 50 2.8 SDS-Polyacrylamide gel electrophoresis and immunoblotting of rat 51 ventricular and isolated ventricular cardiomyocyte chromatographic eluate. vii Page 2.9 Preparation of drug solutions. 53 2.10 Protein determination. 54 2.11 Statistical analysis. 55 3.0 RESULTS 57 3.1 Contractility of rat ventricular preparations. 57 3.1.1 Effects of carbachol, S N P and A N P on contractility of rat 57 Langendorff-perfused heart in the presence of isoproterenol. 3.1.2 Effects of carbachol, SNP and A N P on contractility of rat right 61 ventricular strips in the presence of isoproterenol. 3.1.3 Direct effects of carbachol, S N P and A N P on contractility of 65 rat right ventricular strips. 3.1.4 Baseline contractile function and effect of isoproterenol on 70 contractility of single, isolated cardiomyocytes. 3.1.5 Effect of muscarinic receptor agonists and S N P in presence of 82 isoproterenol on the contractility of single, isolated cardiomyocytes. 3.2 Cyclic nucleotide levels in rat cardiac preparations. 83 3.2.1 Effects of carbachol, A N P or S N P in the presence of isoproterenol 83 on c G M P levels in ventricle of Langendorff-perfused heart. 3.2.2 Effects of isoproterenol, carbachol, A N P and S N P on c A M P levels 90 in ventricle of Langendorff-perfused heart. 3.2.3 Effect of muscarinic receptor agonists and S N P in presence of 90 isoproterenol on c G M P content of isolated ventricular cardiomyocytes. 3.2.4 Effect of muscarinic receptor agonists and S N P in presence of 95 isoproterenol on cAMP content of isolated ventricular cardiomyocytes. 3.3 cGMP-dependent protein kinase in the rat ventricle. 98 viii Page 3.3.1 Characterization of the assay of cGMP-dependent protein kinase. 98 3.3.1.1 Activation of cGMP-dependent protein kinase by 98 c G M P in vitro. 3.3.1.2 Phosphorylation of BPDEtide by cGMP-dependent 101 protein kinase. 3.3.1.3 Temporal linearity of cGMP-dependent protein 101 kinase activity. 3.3.1.4 Chromatography of cGMP-dependent protein 106 kinase. 3.3.1.5 Western blotting of cGMP-dependent protein kinase. 112 3.3.1.6 Effect of putative kinase inhibitors on c G M P - 116 dependent protein kinase activity in vitro. 3.3.1.7 Temporal characteristics of P K G activity ratio 119 during assay. 3.3.2 Activation of cGMP-dependent protein kinase in rat intact 122 ventricular preparations. 3.3.2.1 Effect of carbachol, S N P and A N P on c G M P - 122 dependent protein kinase activity in ventricle of Langendorff-perfused heart. 3.3.2.2 Effect of carbachol, S N P and A N P on c G M P - 127 dependent protein kinase activity in isolated ventricular cardiomyocytes. 4.0 DISCUSSION 132 4.1 Effects of muscarinic receptor agonists on contractility and c G M P 133 content in the ventricle. 4.2 Effects of S N P on contractility and c G M P content in the ventricle. 139 4.3 Effects of A N P on contractility and c G M P content in the ventricle. 144 4.4 Effects of carbachol, SNP , and A N P on ventricular cAMP content. 148 ix Page 4.5 Effects of carbachol, S N P and A N P on ventricular cGMP-dependent 151 protein kinase activity. 4.6 Relationship between contractility and P K G activation by c G M P 161 analogues. 4.7 Possible role of enzymatically-released NO in muscarinic 165 receptor-mediated negative inotropy. 5.0 SUMMARY AND CONCLUSIONS 169 6.0 BIBLIOGRAPHY 171 x LIST OF FIGURES Page 1. cGMP-elevating mechanisms. 3 2. Effect of isoproterenol on contractility of Langendorff-perfused rat 58 heart. 3. Effect of carbachol, A N P or SNP on Langendorff-perfused rat 63 ventricular contractility in the presence of isoproterenol. 4. Twitch tension amplitude of rat ventricular strips in presence of 66 isoproterenol. 5. Effect of carbachol, A N P and S N P on rat right ventricular strip 68 contractility in the presence of isoproterenol. 6. Temporal effects of carbachol, A N P and S N P on amplitude of rat right 71 ventricular strip twitch tension. 7. Concentration-dependent effects of carbachol, A N P and S N P on 73 amplitude of rat right ventricular strip twitch tension. 8. Protocol for drug treatment of single rat ventricular cardiomyocytes. 75 9. Effect of isoproterenol and carbachol on the contraction of a single, rat 80 ventricular cardiomyocyte. 10. Effect of isoproterenol in the presence and absence of acetylcholine on 84 cell shortening of isolated rat ventricular cardiomyocytes. 11. Effect of isoproterenol in the presence and absence of S N P on the 86 contraction of a single, rat ventricular cardiomyocyte. 12. Effects of carbachol, A N P or S N P in the presence of isoproterenol on 88 c G M P levels in ventricle of Langendorff-perfused hearts. 13. Effects of carbachol, A N P or S N P in the presence of isoproterenol on 91 cAMP levels in ventricle of Langendorff-perfused hearts. xi P a g e 14. Effect of muscar in ic receptor agonists and S N P in the p resence of 93 isoproterenol on c G M P levels in rat ventricular card iomyocytes. 15. Effect of muscar in ic receptor agonists and S N P in the p resence of 96 isoproterenol on c A M P levels in rat ventricular card iomyocytes. 16. In vitro activation of c G M P - d e p e n d e n t protein k inase by c G M P . 99 17. Phosphory lat ion of peptide substrate, B P D E t i d e , by soluble c G M P - 102 dependent protein k inase. 18. Tempora l linearity of c G M P - d e p e n d e n t protein k inase assay . 104 19. M o n o Q anion exchange column chromatography of cycl ic nucleot ide- 107 dependent protein k inase activity in rat ventricular t issue. 20. M o n o Q anion exchange column chromatography of cycl ic nucleot ide- 109 dependent protein k inase activity in rat card iomyocytes. 21 . Immunoblotting of MonoQ-fract ionated protein k inases from rat 113 ventricular t issue and cardiomyocytes. 22. Effect of putative k inase inhibitors on c G M P - d e p e n d e n t protein k inase 117 activity in vitro. 23. Effect of assay incubation time on activity ratio of soluble c G M P - 120 dependent protein k inase from agonist-treated rat ventricle. 24. Effect of carbachol , A N P and S N P , in the p resence of isoproterenol, on 123 soluble c G M P - d e p e n d e n t protein k inase activity ratio in rat ventricular t issue. 25. Effect of carbachol , A N P and S N P on soluble c G M P - d e p e n d e n t 128 protein k inase activity in rat isolated ventricular card iomyocytes. xii LIST OF TABLES Page 1. Effects of carbachol, S N P and A N P on contractility of rat Langendorff- 62 perfused hearts in presence of isoproterenol. 2. Effect of isoproterenol and carbachol on rat ventricular cardiomyocyte 78 contractility. 3. Effect of isoproterenol and S N P on rat ventricular cardiomyocyte 79 contractility. 4. Total soluble and particulate cGMP-dependent protein kinase activity 126 in ventricle of agonist-treated, Langendorff-perfused rat heart. 5. Total soluble and particulate cGMP-dependent protein kinase activity 131 in isolated rat ventricular cardiomyocytes. xiii LIST OF ABBREVIATIONS °C degrees Celsius + dP/dt maximal rate of rise in developed pressure - dP/dt maximal rate of decline in developed pressure u micro (1 x 10"6) 8-bromo-cGMP 8-bromo guanosine 3':5'-cyclic monophosphate A ampere A N O V A analysis of variance A N P atrial natriuretic peptide ATP adenosine triphosphate BCIF 5-bromo-4-chloro-indolyl phosphate BPDEtide a peptide substrate of P K G (RKISASEFDRPLR) BSA bovine serum albumin CaCI 2 calcium chloride cAMP adenosine 3':5'-cyclic monophosphate Ci curie c G M P guanosine 3':5'-cyclic monophosphate C 0 2 carbon dioxide C u S 0 4 cupric sulfate d dextro DNase deoxyribonuclease DTT dithiothreitol E C 5 0 , E C 6 0 concentration which causes 50 or 6 0 % of maximal effect xiv EGTA f F P L C 9 9 HCI H E P E S i.p. IBMX IU K a KCI kDa K H 2 P 0 4 Ki KT5823 K m m m M max. LVP ethylene glycol bis((5-aminoethyl ether) N,N'-tetraacetic acid fempto (1 x 10"15) fast protein liquid chromatography relative centrifugal force gram hydrogen chloride A/-2-hydroxyethylpiperazine-A/'-2-ethanesulfonic acid intraperitoneal 3-isobutyl-1 -methylxanthine international units concentration for half-maximal activation potassium chloride kilo Dalton potassium dihydrogen phosphate inhibition constant, concentration for half-maximal inhibition (8R,9S, 11 S)-(-)-9-methoxy-9-methoxycarbonyl-8-methyl-2,3,9,10-tetrahydro-8,11 -epoxy-1 H,8H, 11 H-2,7b, 11 a-triazadibenzo(a,g)cycloocta(cde)-trinden-1-one Michaelis-Menten constant, substrate concentration giving one-half of maximal velocity of reaction. litre metre milli (1 x 10"3), if a prefix molar maximal left ventricular pressure xv MgCI 2 magnesium chloride M g S 0 4 magnesium sulfate M r relative molecular mass n nano (1 x 10"9) N 2 nitrogen N a 2 C 0 3 sodium carbonate NaCl sodium chloride N a H C 0 3 sodium hydrogen carbonate, sodium bicarbonate NaOH sodium hydroxide NBT nitroblue tetrazolium NMMA L-/V-mono-methyl-arginine NO nitric oxide NOS nitric oxide synthase 0 2 oxygen p pico(1 x 10"12) P A G E polyacrylamide gel electrophoresis P D E phosphodiesterase P K A cAMP-dependent protein kinase P K G cGMP-dependent protein kinase PKI PKA synthetic inhibitor P M S F phenylmethylsulfonyl fluoride Rp-8 -CPT- Rp-8-(4-chlorophenylthio)-guanosine 3',5' cyclic c G M P S monophosphorothioate s second xvi S D S sodium dodecyl sulfate S E M standard error of the mean SIN-1 3-morpholinosydnonimine SNK Student-Neuman-Keuls S N P sodium nitroprusside S P A scintillation proximity assay SR sarcoplasmic reticulum TCA trichloroacetic acid TEMED N,N,N',N'-tetramethylethylenediamine urn micron (1 x 10"6 metre) w/v weight per volume xvii ACKNOWLEDGEMENTS First and foremost, I would like to extend my sincere thanks to my supervisor, Dr. Jack Diamond, for his patient, kind and thoughtful guidance throughout the course of my studies at the University of British Columbia. I would also like to express my appreciation to the members of my supervisory committee, Drs. Kathleen MacLeod, Roger Brownsey, David Godin, Helen Burt and Alan Mitchell for their constructive criticism and suggestions regarding my research. Dr. Glen F. Tibbits generously provided access to his cardiomyocyte contractility apparatus in the Cardiac Membrane Research Laboratory at Simon Fraser University. This was crucial for an important component of my research. I would like to acknowledge his generosity and thank Dr. Tibbits and the members of CMRL for kindly welcoming me into their laboratory during the long winter of '93 - '94. I am also appreciative of the gift of PKG antibodies from Dr. Steve Pelech and the use of his research facilities at Kinetec Biotechnology Corp. for the Western blot experiments. The technical expertise of Mr. Harry Paddon in helping me with the immunoblotting experiments is gratefully acknowledged. The laboratory of Dr. Jack Diamond was a warm and friendly place to spend my Ph. D. studies and I thank, particularly, Dr. Yong-jiang Hei, Dr. Ashwin Patel, Dr. John Langlands, Mr. James Hennan and Ms. Danielle Zammit for their friendship. I would also like to thank the Medical Research Council of Canada for financial support in the form of a studentship and the Heart and Stroke Foundation of Canada for their financial support in the form of a research traineeship. I am also appreciative of the very generous financial assistance which my family extended to me. xviii to my family, my sisters, Ann Marie and Paula and my brother, Stan and my mom, Erma Dwyer, with love and respect and to kele & the future xix 1.0 INTRODUCTION Force generation by the mammalian myocardium can be modulated by exogenous signalling molecules such as neurotransmitters, hormones and paracrine substances. These agents generally bind to receptors at the surface of myocardial cells and stimulate the intracellular synthesis of second messengers which may trigger a cascade of biochemical events and ultimately regulate the ability of the myocardium to generate force, that is, mediate an inotropic effect. The most important positive inotropic influence on the heart arises from signalling events (such as those associated with activation of the sympathetic nervous system) which elevate adenosine 3':5'-cyclic monophosphate (cAMP), leading to activation of cAMP-dependent protein kinase (PKA) and phosphorylation of specific proteins involved in calcium handling and responsiveness during contraction (Endoh, 1995). It was observed in the early 1970's that the synthesis of another cyclic nucleotide, guanosine 3':5'-cyclic monophosphate (cGMP), may have an opposite effect on the heart. George et al. (1970, 1972, 1975) observed an association between a negative inotropic effect and elevated c G M P content in intact heart and atrial preparations exposed to the parasympathetic post-ganglionic neurotransmitter, acetylcholine. Goldberg et al. (1975) proposed the "yin-yang" hypothesis to explain the apparently antagonistic functional relationship between cAMP and c G M P on the contractility of the heart, as well as other tissues such as the uterus and vascular smooth muscle. Even after two decades, the veracity of this hypothesis remains an issue of contention. While the role of c A M P in mediating a positive inotropic effect is firmly established, the role of c G M P in regulating cardiac contractile force is controversial (see Lohmann et al., 1991 for a review). The following 1 is a review of the evidence for and against such a role for c G M P and the presentation of a proposal to test the hypothesis that c G M P mediates a negative inotropic effect in the mammalian ventricle through the activation of cGMP-dependent protein kinase (PKG). 1.1 GMP-elevating systems. In spite of the observation over 25 years ago that specific agonists can elevate c G M P in the heart, the exact mechanism by which this occurs in the myocardium is not known. The vasculature is a tissue in which cGMP-elevating systems are well characterized (see Waldman and Murad, 1987 and Schmidt et al., 1993 for reviews). As shown in figure 1, in endothelial cells of blood vessels, receptor-mediated activation of G-proteins by agonists such as acetylcholine (ACh) ultimately leads to increased intracellular calcium which, in turn, activates calcium-dependent constitutive nitric oxide synthase (cNOS). The conversion of /-arginine to /-citrulline is catalyzed by this enzyme and the short-lived signal transduction molecule, nitric oxide (NO), is formed. NO diffuses within and between cells where it has several biological effects, including activation of soluble guanylyl cyclase (sGCy) in smooth muscle cells and the resulting catalysis of the cyclization of the oc-phosphate on GTP, forming the small, water-soluble second messenger, cGMP. Agents such as sodium nitroprusside (SNP) [Na 2Fe(CN) 5NO] and 3-morpholinosydnonimine (SIN-1) bypass plasma membrane receptor-associated events, acting as NO donors and thereby directly activating the heme-containing, heterodimeric soluble guanylyl cyclase (sGCy). S N P is thought to release NO through a reduction-oxidation reaction (Smith et al., 1990) and can also release NO in a photochemical reaction (Wolfe and Swinehart, 1975): 2 ACh p G C y myocyte Figure 1. c G M P elevating mechanisms. See text for details. [Fe(CN)5NO]2" + H 2 0 -> [Fe(CN) 5H 20] 2" + NO c G M P production can be pre-empted by NO scavenging agents which include methylene blue, hemoglobin, and LY83583. c G M P levels can also be elevated by activation of plasma membrane-spanning particulate guanylyl cyclase (pGCy) (Schulz et al., 1991b). The extracellular portion of the cyclase protein contains the specific receptor for atrial natriuretic peptide (ANP) which is secreted by atrial cardiomyocytes (Cantin and Genest, 1985). Binding of A N P stimulates the intracellular portion of the receptor protein to catalyze the conversion of GTP to cGMP, independently of effects by NO. Termination of the c G M P signal is the role of cyclic nucleotide-specific phosphodiesterases. c G M P is hydrolyzed to inactive G M P , a process which is catalyzed primarily by calcium/calmodulin-dependent cyclic nucleotide phosphodiesterases, cGMP-specific phosphodiesterases and cGMP-stimulated phosphodiesterases (Sonnenburg and Beavo, 1994). Several receptors for c G M P are known (see Lincoln and Cornwell, 1993 for a review) and the first to be identified was cGMP-dependent protein kinase (EC 2.7.1.37) (Kuo and Greengard, 1970). The activity of two phosphodiesterases, cGMP-stimulated phosphodiesterase and cGMP-inhibited phosphodiesterase, are regulated upon binding of c G M P while cGMP-binding phosphodiesterase binds c G M P without a change in its catalytic function. The gating properties of several ion channels are directly modulated by c G M P in specific tissues such as the retina, olfactory and renal epithelium, and spermatozoa (see Cornwell and Lincoln, 1993 for a review). Upon binding of c G M P or cAMP, cyclic nucleotide-gated channels are permeant to mono-and divalent cations and are responsible for increases in intracellular calcium levels in 4 specific sensory and non-sensory tissues (Yau, 1994; Kaupp, 1995). Recently, evidence has been reported for the presence of messenger RNA for a cyclic nucleotide-gated channel in the rabbit and bovine heart but no transcript for the channel was present in the rat heart (Biel et al., 1994; Distler et al., 1994). Whether or not these transcripts express functional channel proteins which impact on the contractility of the heart is unknown. 1.2 Effects of cholinergic agonists in the heart. The sensitivity of the heart to modulation by the vagus nerve led to the discovery of neurotransmitters, the first identified being acetylcholine (see Mayer, 1980 and Loffelholz and Pappano, 1985, for reviews). The heart has a variable density of innervation by post-ganglionic parasympathetic neurons. The densest innervation is found in the sino-atrial and atrio-ventricular nodes and the His-Purkinje conducting system. In amphibian and avian hearts, the atria and ventricles have a similar level of innervation while, in mammalian hearts, the ventricle has about one-fifth of the atrial density. Vagal activity, via release of acetylcholine from post-ganglionic nerve terminals and binding to M 2 muscarinic receptors on effector cells, has several depressant effects on the heart, including negative chronotropy (decreased beating rate), negative dromotropy (slowed conduction of electrical impulses), and negative inotropy (decreased force of contraction). The former two effects are a result of muscarinic receptor stimulation in the electrical conduction system of the heart (SA and AV nodes, His-Purkinje system). Negative inotropic effects of the parasympathetic nervous system were initially detected only in the atria. Inhibition of ventricular contractility by acetylcholine was reported after it was appreciated that an elevated 5 level of sympathet ic nervous system activity was required for the express ion of the negat ive inotropic effects of acetylchol ine (DeGees t et al., 1965; Levy et al., 1971). Pre-synapt ic (pre-junctional) inhibition of noradrenal ine re lease by acetylchol ine can account for some of the inhibition of sympathet ic activity (Muschol l , 1980) but negative inotropy can a lso be detected in the ventricle and atrium in the p resence of p-adrenoceptor agonists, indicating that a post-junctional mechan ism must a lso be present. 1.2.1 Effects of muscarinic receptor agonists on atrial contractility. Muscar in ic receptor agonists induce a rapid (within seconds) and persistent dec rease in basal atrial tension. The p D 2 va lues (negative log of the concentrat ion which has 5 0 % of maximal effect) for acetylchol ine and carbacho l range between 7.0 and 6.6 , with carbachol being somewhat more potent. Depending on the spec ies , maximal inhibition of basa l force ranges from 5 0 % to nearly a 1 0 0 % (Endoh and Yamash i t a , 1981, M a c L e o d , 1986; Du et al., 1994, 1995). Muscar in ic receptor agonists not only reduce basa l atrial tension but effectively antagonize the effects of var ious posit ive inotropic agents in the atrium, including a- and (3-adrenoceptor agonists, phosphodies terase inhibitors and elevated extracel lular ca lc ium (Endoh and Yamash i ta , 1981; M a c L e o d , 1986; Jakob et al., 1989). The p D 2 va lues for muscar in ic receptor agonists are essential ly the s a m e as in the p resence and a b s e n c e of posit ive inotropic agents. The maximal reversal of the positive inotropic effect is dependent on the nature of the posit ive inotropic agent; carbachol inhibits the contracti le effects of a-and (3-adrenoceptor agonists and phosphodiesterase inhibitors by 80 - 100%, but dec reases the posit ive inotropic effects of e levated extracel lular ca lc ium by 6 approximately 5 0 % (Endoh and Yamash i ta , 1981; M a c L e o d , 1986; Jakob et al., 1989, Du e ra / . , 1994, 1995). 1.2.2 Possible mechanisms for cholinergic inhibition of atrial contractility. Since muscar in ic receptor agonists can reduce atrial contractility in the a b s e n c e of posit ive inotropic agents, and in the presence of posit ive inotropic agents which do not e levate c A M P levels, mechan isms of action by muscar in ic receptor agonis ts in atria are likely to be independent of direct effects on the cAMP-s igna l l i ng sys tem. Ev idence for ion channe l regulation and c G M P signall ing in the mechan ica l effects of muscar in ic receptor agonists in atria are d iscussed below. 1.2.2.1 Modulation of ion channels. Muscar in ic receptor agonists enhance an atrial outward potass ium current, W c h ) . which leads to hyperpolarization of atrial myocytes. This , in turn, inhibits ca lc ium influx through the L-type calc ium channel ( I C a ) and mediates a negative inotropic effect (Ten Eick et al., 1976). Per tuss is toxin-sensit ive guanine nucleotide binding proteins link the muscar in ic receptor to potass ium channe ls without the involvement of a second messenger (Szabo and Otero, 1990). Direct negat ive inotropy and reversal of a - and (3-adrenoceptor positive inotropy could ar ise from I«(ACh) activation. In addit ion to regulation via activation of I K ( A C h ) , atrial I C a may be inhibited by muscar in ic receptor agonists as a result of inhibition of adenylyl cyc lase such that less c A M P would be avai lable to activate the channel (Pappano and Inoue, 1984). 7 1.2.2.2 Elevation of cGMP levels. Reports from the mid-1970's noted a positive correlation between muscarinic receptor agonist-induced c G M P elevation and negative inotropy in various atrial preparations (George et al., 1975; Goldberg et al., 1975) and a causal relationship between c G M P elevation and negative inotropy was suggested. Since then, several lines of evidence have developed which argue against a cause-effect relationship between the two phenomena (see Linden and Brooker, 1979 for review). Firstly, marked decreases in atrial force by muscarinic receptor agonists occur at lower concentrations than are necessary for significant elevation of c G M P (Diamond et al., 1977; Brooker, 1977). Secondly, very high levels of c G M P could be attained following treatment of atrial preparations with SNP, yet no inhibition of contractility was observed (Diamond et al., 1977; Katsuki et al., 1977). The absence of a negative inotropic effect by NO donors was also apparent in a- and p-adrenoceptor-stimulated atrial preparations (Nawrath er a/., 1995). The effects of another cGMP-elevating agent, A N P , on atrial contractility is sparsely documented. In the only report which considered this question in the mammalian atrium, no changes in guinea pig atrial force of contraction were observed nor was there any effect on beating rate in the presence of A N P II (23 amino acid form of ANP) (Baum et al., 1986). Finally, muscarinic receptor agonists mediate a negative inotropic effect in the atria, even when c G M P elevation is blocked. The cGMP-lowering agent, LY83583, prevented c G M P elevation by acetylcholine but had no effect on the direct negative inotropic effect in rabbit atrial strips (Diamond and Chu, 1985). Similarly, carbachol reversed the positive inotropic 8 effects of forskolin when c G M P levels were significantly e levated by carbacho l or when c G M P elevation was blocked by L Y 8 3 5 8 3 (MacLeod and D iamond, 1986). In contrast to this ev idence for a dissociat ion between negative inotropy and c G M P in atria, c G M P ana logues have been reported to dec rease atrial contracti le force (Baumner and Nawrath, 1995; Nawrath et al., 1995) and mimic some of the electrophysiological propert ies of acetylchol ine (Nawrath, 1977; Kohlhardt and Haap , 1978). The reason for these conflicting data is not c lear but it should be kept in mind that there is ev idence in non-card iac t issue which indicates that c G M P ana logues have effects which do not reflect the effects of endogenous c G M P . For example , 100 u M 8 - b r o m o - c G M P induced a rapid reduction in tension in rat myometrial strips yet N O donors, which stimulate the endogenous production of cycl ic G M P , had no effect on myometrial contractility (Diamond, 1983). Thus , notwithstanding the contrary data from studies with ana logues , there is some ev idence for a dissociat ion between agonist-mediated c G M P elevat ion and negative inotropy in atria. 1.2.3 Effects of muscarinic receptor agonists on ventricular contractility. The response of the ventricle to muscar in ic receptor agonists is genus - and spec ies-dependent and, in some spec ies , is inf luenced by a pre-exist ing posit ive inotropic effect mediated specif ical ly by cAMP-e leva t ing agents. B a s a l force development (i.e., in the absence of a positive inotropic agent) is dec reased by muscar in ic receptor agonists in amphibian and avian ventricular t issue but is usual ly unchanged or even increased in the mammal ian ventricle (Endoh and H o n m a , 1979; Jakob et al., 1989). If a direct positive inotropic effect is exp ressed , the p D 2 va lue for acetylchol ine and carbachol is approximately 4.5 (Korth and Ku lhkamp, 1987). The 9 ferret is unique among mammals, in that acetylcholine directly inhibits basal ventricular force (Hongo et al., 1993). As mentioned above, in most mammalian species, the negative inotropic effect of muscarinic receptor agonists in the ventricle is only measurable in the presence of specific positive inotropic agents, that is, those which increase cAMP levels (e.g., p-adrenoceptor agonists and phosphodiesterase inhibitors). This anti-adrenergic effect has been referred to as "accentuated antagonism" (Levy, 1971). As in the atria, the negative inotropic effect of muscarinic receptor agonists is rapid and persistent and approaches a complete reversal of the contractile effects of cAMP-elevating agents, with pD 2 values of approximately 6.8 for carbachol and 6.0 for acetylcholine (Endoh and Yamashita, 1981; Korth and Kulhkamp, 1987). The net effect of muscarinic receptor agonists on the c A M P -stimulated ventricle is a reduction in potency and efficacy of cAMP-elevating agents in enhancing the force of ventricular contraction, with no reduction of basal contractility. Thus, muscarinic receptor agonists are effective negative inotropic agents in the ventricle but usually require conditions to reveal this effect which differ from the atrium, namely, a background of cAMP-specific increased contractility. 1.2.4 Possible mechanisms for cholinergic inhibition of ventricular contractility. 1.2.4.1 Decreased cAMP levels. The requirement for the presence of cAMP-elevating agents for expression of an inotropic effect by muscarinic receptor agonists in the ventricle suggests that muscarinic receptor binding impacts the cAMP signalling system. Muscarinic receptor agonists inhibited adenylyl cyclase in vitro (Murad et al., 1962; Watson et al., 1988) via 10 a mechanism which probably involves Gj, the inhibitory guanine nucleotide binding protein (Hazeki and Ui, 1981; Gilman, 1987). Adenylyl cyclase inhibition may account for muscarinic receptor-mediated decreases in ventricular agonist-stimulated cAMP levels which were detected in some studies (Keely et al., 1978; Inui et al., 1982; Katano and Endoh, 1993). On the other hand, reductions in c A M P by muscarinic receptor agonists have not always been observed at concentrations which reduce cAMP-mediated positive contractile effects. A variety of muscarinic receptor agonists failed to induce changes in cAMP levels in guinea pig or rabbit ventricular tissues or guinea pig isolated cardiomyocytes treated with either isoproterenol or phosphodiesterase inhibitors (Watanabe and Besch, 1975; MacLeod and Diamond, 1986; Schmeid and Korth, 1990; Gupta et al., 1994a). Although the relationship between changes in cAMP levels and negative inotropic effects of muscarinic receptor agonists remains controversial, it appears that the lowering of c A M P levels by muscarinic receptor agonists through inhibition of cAMP formation or enhancement of c A M P degradation is not an absolute requirement for their negative inotropic effects. The only known receptor for cAMP in the ventricle is PKA, which phosphorylates, among other substrates, the L-type calcium channel or a closely associated regulatory protein (MacDonald et al., 1994). Calcium influx into the ventricular cell via L-type calcium channels is regarded as a key determinant of contractile force (Sperelakis et al., 1994). Muscarinic receptor agonists antagonize the PKA-mediated enhancement of I C a in a manner consistent with the mechanical effects of muscarinic receptor agonists on the cAMP-stimulated ventricle. For example, muscarinic receptor agonists do not alter the amplitude of basal I C a , which is consistent 11 with the lack of an inotropic effect in the unstimulated ventricle. However, they reverse the stimulatory effect on l C a of cAMP-elevating agents such as isoproterenol, 3-isobutyl-1-methylxanthine (IBMX), papaverine and forskolin (Hescheler et al., 1986; Mubagwa et al., 1993; MacDonald et al., 1994) and this is consistent with their ability to specifically antagonize the positive inotropic effects of cAMP-elevating agents. The mechanism behind muscarinic receptor agonist-mediated depression of I C a is unclear, in view of the controversy regarding control of cAMP levels by muscarinic receptor agonists described above, and the failure of cholinergic agents to decrease P K A activity in rat or guinea pig ventricular tissue or cardiomyocytes in the presence of isoproterenol, forskolin or IBMX (Bartel et al., 1993; Gupta et al., 1994a). Evidence exists for a role for c G M P in the cholinergic effect on l C a since methylene blue and LY83583, both cGMP-lowering agents, decreased the efficacy of carbachol in reducing IBMX-stimulated I C a in the guinea pig ventricular cardiomyocyte (Mubagwa et al., 1993; Levi et al., 1994). The possible role for c G M P as a mediator of the negative inotropic effects of muscarinic receptor agonists will be discussed in detail below. 1.2.4.2 Phosphoprotein phosphatase activation. In view of the evidence for a negative inotropic mechanism that does not involve a decrease in cAMP levels, a reasonable hypothesis is that muscarinic receptor agonists may activate phosphoprotein phosphatases and reverse PKA-mediated phosphorylation. Muscarinic receptor agonists decreased (3-adrenoceptor-mediated phosphorylation of various substrates, including troponin 1 and phospholamban, in a phosphatase-dependent manner (Gupta et al., 1994a). The mechanism involved in dephosphorylation has been suggested to be the antagonism of cAMP-mediated 12 activation of phosphatase inhibitor-1 by cholinergic agents (Ahmad et al., 1989; Gupta et al., 1993). This would lead to reversal of cAMP-mediated inhibition of type 1 phosphatase activity and, ultimately, enhanced dephosphorylation. c A M P levels were not substantially altered nor was PKA activity changed by acetylcholine (Gupta et al., 1993), suggesting that muscarinic receptor agonists can decrease the effects of c A M P -elevating agents by a cAMP-independent mechanism. Acetylcholine also reversed the effects of isoproterenol on the inwardly rectifying K + current in an okadaic acid-sensitive manner in guinea pig ventricular cardiomyocytes (Koumi et al., 1995). Okadaic acid is an inhibitor of type 1 and 2A phosphatase. Although it is not known if cholinergic regulation of phosphatase inhibitor-1 activity involves dephosphorylation of a PKA-specific residue or phosphorylation on a different residue, it is interesting to note that a cerebellar phosphatase inhibitor protein is a substrate for P K G (Aitken et al., 1981). 1.2.4.3 Cyclic cGMP. 1.2.4.3.1 Role of cGMP in regulation of amphibian ventricular contractility. Support for the "yin-yang" hypothesis has been provided in several studies carried out on amphibian ventricles. Flitney and Singh (1981) noted that acetylcholine depressed frog ventricular contractility, over a range of concentrations, in a manner which positively correlated with c G M P levels and inversely correlated with c A M P levels. Inhibition by a phosphodiesterase inhibitor of both the negative inotropic effect and the cAMP-lowering effect raised the possibility that muscarinic receptor agonists modulate amphibian ventricular contractility through cGMP-mediated activation of a cAMP 13 phosphodisterase. The contractile effects of muscarinic receptor agonists in the frog ventricle were complemented by the effects of c G M P on I C a (Fischmeister and Hartzell, 1987). Intracellular dialyzed c G M P had no effect on basal frog I C a but shifted the concentration-response curve for cAMP-mediated stimulation of I C a to the right and reduced the maximal effect of cAMP. 8-Bromo-cGMP, a potent P K G activator but a weak regulator of phosphodiesterases, was without effect. c G M P did not reduce I C a stimulated with hydrolysis-resistant 8-bromo-cAMP and phosphodiesterase inhibitors reduced the inhibitory effects of c G M P on cAMP-stimulated I C a . Comparable results have been observed with exogenous cGMP-elevating agents such as acetylcholine and SIN-1 in the amphibian ventricle (Fischmeister and Schrier, 1989; Mery et al, 1994). These results suggested that c G M P enhanced the activity of the cAMP-specific phosphodiesterase (hence the blunted inhibition in presence of IBMX) in a manner which was independent of P K G activation (hence the ineffectiveness of 8-bromo-cGMP) , such that cAMP-mediated stimulation of I C a was reversed and provided a mechanism of action for muscarinic receptor agonists. Therefore, in the amphibian ventricle, c G M P and cAMP mediate opposing actions, as proposed by the "yin-yang" hypothesis, and also interact functionally with c G M P stimulating the catabolism of cAM P. 1.2.4.3.2 Role of cGMP in regulation of mammalian ventricular contractility. Muscarinic receptor agonists elevate ventricular c G M P levels but, as mentioned above, the mechanism by which muscarinic receptor agonists stimulate c G M P formation in the heart is not completely understood. In the mammalian ventricle, levels 14 of c G M P were increased in rabbit and guinea pig isolated cardiomyocytes treated with 10 - 100 pM carbachol (Cramb et al., 1987; Stein et al., 1993) and muscarinic receptor binding was an essential step (Stein et al., 1993). Preliminary data suggest that the M 2 cholinergic receptor is functionally linked to ventricular c G M P elevation (Gupta et al., 1994a). Calcium-dependent constitutive nitric oxide synthase (NOS III) has been detected in rat ventricular cardiomyocytes (Schulz et al., 1991a; Balligand et al., 1995) and formation of NO upon muscarinic receptor binding has been reported in some studies (Balligand et al., 1993) but not others (Stein et al., 1993). Thus, the ventricular cardiomyocyte has the required complement of intracellular machinery to generate c G M P upon M 2 muscarinic receptor binding. The process may involve NO-mediated activation of soluble guanylyl cyclase but further investigation is required to substantiate this proposal. Elevation of c G M P by muscarinic receptor agonists in the mammalian ventricle has been shown to correlate with negative inotropy on a time- and concentration-dependent basis (Watanabe and Besch, 1975; Keely and Lincoln, 1978; Lincoln and Keely, 1980, 1981). Maximal increases in c G M P levels by muscarinic receptor agonists are in the range of 2- to 4-fold above control levels (Watanabe and Besch, 1975; Keely et al., 1978; Cramb et al., 1987). The relationship appears to be more than incidental, since blockade of c G M P elevation impairs the ability of carbachol to reverse the positive inotropic effects of cAMP-elevating agents. Blockade of the cGMP-elevating effects of 3 pM carbachol by 10 pM LY83583 was associated with a significant reduction of the negative inotropic effect of carbachol in forskolin-treated rabbit papillary muscles (MacLeod and Diamond, 1986). Thus, while c G M P is 15 apparently not involved in the negative inotropic effect of muscarinic receptor agonists in mammalian atria, c G M P may contribute to the mechanical effects of cholinergic agonists in the ventricle. Very few studies have measured the effect of muscarinic receptor agonists on c G M P levels and contractility in isolated ventricular cardiomyocytes. Isolated cells may be valuable tools in view of the difficulty inherent in ascribing the cellular source of c G M P when measured in intact tissues which contain a variety of cell types. A single brief report correlated c G M P levels with negative inotropy induced by cholinomimetics in isolated cardiomyocytes; Stein and Schweiger (1990, abstract) reported significant increases in cellular c G M P and inhibition of isoproterenol-stimulated cell shortening by 10 uM carbachol. Whether this was a causal relationship or simply coincidental was not established. 1.3. Effects of cGMP analogues on ventricular contractility. If the negative inotropic effect of muscarinic receptor agonists in the ventricle are mediated by cGMP, exogenous administration of membrane permeable analogues of c G M P should mimic this action. Antagonism of the inotropic effects of p-adrenoceptor agonists was observed with A/ 2-2'-0-dibutyryl-cGMP ( 1 - 1 0 uM) in vitro (Watanabe and Besch, 1975) and 8-bromo-cGMP ( 1 - 1 0 mM) in vivo (Hare et al., 1995). Interpretation of these findings is complicated, however, by reports of the absence of an inhibitory effect of these c G M P analogues under similar conditions. For example, the contractility of p-adrenoceptor-stimulated canine (Endoh and Shimizu, 1979) or ferret ventricular tissues (Shah et al., 1991) was not inhibited by 100 uM 8-bromo-cGMP, and 50 uM 8-bromo-cGMP did not decrease cell shortening in rat p-16 adrenoceptor-stimulated cardiomyocytes (Shah et al., 1994). Another difference between the mechanical effects of muscarinic receptor agonists and c G M P analogues in the ventricle is that, in some studies, 8-bromo-cGMP directly depressed force generation (i.e., in the absence of elevated cAMP levels). The magnitude of the mechanical effect varies considerably among studies. For example, 100 uM 8-bromo-c G M P decreased ferret papillary muscle twitch tension by only 4 % to 7% (Shah et al., 1991; Smith et al., 1991) while 300 uM 84oromo-cGMP decreased cat papillary muscle twitch tension by 3 0 % (Nawrath, 1976). Conversely, neither dibutyryl c G M P (100 - 300 uM) nor 8-bromo-cGMP (300 uM) reduced twitch tension in canine ventricular muscle (Endoh and Shimizu, 1979; Endoh and Yamashita, 1981). c G M P analogues share some of the effects of muscarinic receptor agonists on I C a . In rat and guinea pig ventricular cardiomyocytes, c G M P and 8-bromo-cGMP produced dose-dependent inhibition of I C a which had been stimulated with cAMP, 8-bromo-cAMP, or the phosphodiesterase inhibitor, IBMX (Levi et al., 1989; Mery et al., 1991; Mubagwa et al., 1993). Differences between c G M P analogues and muscarinic receptor agonists in terms of I C a inhibition have also been noted. Low concentrations of intracellularly dialyzed c G M P enhanced, rather than inhibited, cAMP-stimulated I C a in guinea pig myocytes (Ono and Trautwein, 1991) and 8-bromo-cGMP suppressed basal I C a in human and chick fetal ventricular cells and adult rabbit ventricular cells (Bkaily et al., 1993; Tohse et al., 1995). Thus, the data from experiments with c G M P analogues on contractility and I C a are, in some cases, supportive of a role for c G M P in mediating the effects of muscarinic receptor agonists in the ventricle and, in other cases, incompatible with such a role. 17 1.4 Nitric oxide and the ventricle. NO was acknowledged as an important regulatory signal in the vasculature in the 1980's but interest in this molecule in the myocardium arose only recently with reports of inotropic effects of endothelium-derived and endocardium-derived NO (Smith et al., 1992). As shown in figure 1 and as described previously, the requisite components of a NO-generating system in cells of endothelial origin (coronary endothelium and endocardium) and mesothelial origin (cardiomyocytes) are likely to be present in the myocardium. An experimental approach to elevating NO levels in myocardial preparations involves the use of NO donors, such as S N P and 3- SIN-1. 1.4.1 Effect of nitric oxide donors on ventricular contractility. If c G M P is responsible for the negative inotropic effects of muscarinic receptor agonists, then cGMP-elevating agents such as S N P should also have negative inotropic effects and changes in c G M P levels should quantitatively relate to the extent of inhibition of contractility. However, this does not appear to be the case. The ventricle is believed, generally speaking, to be unresponsive to NO-releasing agents so that, as in the atrium, numerous reports describe a lack of direct contractile effects by agents such as SNP, glyceryl trinitrate and SIN-1 in the intact mammalian ventricle (Endoh and Yamishita, 1981; Lincoln and Keely, 1981; Inui et al., 1982; Roger and Shahid, 1984, Thelan et al., 1992, Ishibashi, 1993) and in isolated cardiomyocytes (Stein et al., 1993). For example, the failure of S N P to alter the amplitude of ventricular contractile force development was observed over a concentration range of 0.1 - 30 pM in perfused rat hearts (Lincoln and Keely, 1980, 1981) and 0.1 - 10 pM in guinea pig working hearts (Grocott-Mason et al., 1994). Similarly, exposure of rabbit papillary 18 muscle to 1 mM S N P for periods of up to 13 min failed to alter the amplitude of twitch tension (Roger and Shahid, 1984). In light of the fact that muscarinic receptor agonists generally induce negative inotropic effects in the mammalian ventricle only in the presence of cAMP-elevating agents and assuming that muscarinic receptor agonists produce their contractile effects by elevating c G M P levels, the lack of a direct inhibitory effect of NO donors on ventrcular contractility is not an unexpected finding. If c G M P is involved in the negative inotropic effects of muscarinic receptor agonists, NO donors which elevate c G M P should mimic these negative inotropic effects in the cAMP-stimulated ventricle. On the contrary, p-adrenoceptor stimulation had no influence on the lack of inotropic effects by S N P in several intact ventricular preparations. Hongo et al. (1993) observed that S N P (0.1 - 1 mM) did not change the maximal amplitude or time-course of contraction in 0.1 pM isoproterenol-stimulated ferret papillary muscle. Likewise, 1 mM S N P had no effect on the pD 2 value of phenylephrine in mediating a p-adrenoceptor-stimulated increase in contractile force in canine ventricular trabeculae (Endoh and Yamashita, 1981) and 5 pM S N P did not reduce 0.1 pM adrenaline-stimulated ventricular contractile force in perfused rat hearts (Keely and Lincoln, 1978). Authentic NO gas (52 - 500 nM) was without effect on the amplitude of twitch tension in rat papillary muscles in the presence of noradrenaline (0.01 - 1 pM) and, at a concentration of 500 nM, only slightly decreased tension in 5 pM noradrenaline-stimulated muscles (Weyrich et al., 1994). Thus, the p-adrenoceptor-stimulated mammalian ventricular myocardium is generally resistant to inhibition of force generation by exogenous sources of NO. 19 Under some circumstances, cardio-depressant effects of NO-releasing agents have been observed although the effects are usually very small. For example, 1 uM S N P decreased the amplitude of twitch tension by 3 % in ferret ventricular muscles (Smith et al., 1991) and 10 uM S N P decreased maximal left ventricular pressure by 6 % in perfused ferret hearts (Fort et al., 1991). Small negative inotropic effects of NO donors have also been detected in cardiomyocytes but, as in intact ventricular preparations, the effects were modest, as illustrated in the study by Brady et al. (1993) in which guinea pig ventricular cardiomyocyte shortening declined from 5.8% of resting cell length to 4 .7% in the presence of 10 uM SNP. It has been noted by Weyrich et al. (1994) that the slight negative inotropic effects induced by NO donors occur at uM concentrations, in excess of concentrations which mediate therapeutic effects ( « 500 nM) such as vasodilation and adhesion of neutrophils and platelets to endothelial cells. When considered as a whole, NO donors have little, if any, effect on the inotropic properties of the ventricle and, in fact, have been used as "pure vasodilator" tools, that is, as controls which dilate blood vessels but which do not alter cardiac contractility (Semigran etal., 1994). It has been suggested that the primary direct mechanical effect of S N P is mediation of an earlier onset of ventricular relaxation with relatively little effect on maximal force or maximal rates of contraction and relaxation (Grocott-Mason et al., 1994; Paulus et al., 1994). The relationship of this effect to elevation of c G M P has not been systematically evaluated although it is interesting to note that 8-bromo-cGMP enhanced the onset of relaxation in ventricular cardiomyocytes in some circumstances (Shah et al., 1994) but not in others (Tao and McKenna, 1994). When present, this 20 positive lusiotropic effect was more apparent when cells were examined at relatively unphysiological conditions since 8-bromo-cGMP (50 pM) reduced the time to peak shortening by 17% in rat cardiomyocytes at 25°C and a stimulation frequency of 0.5 Hz but decreased this parameter by only 10% at 35°C and 2 Hz (Shah et al., 1994). Therefore, a role for endogenous c G M P in a positive lusiotropic effect of S N P is uncertain. 1.4.2 Relationship between SNP-mediated cGMP elevation and ventricular contractility. A possible explanation for the lack of effect of S N P and other NO donors on ventricular contractility is the possibility of defective NO-associated cGMP-elevating capabilities in the ventricle. Organic nitrates, such as glyceryl trinitrate, require intracellular biotransformation to release NO (Bennett et al., 1994). Metabolism of glyceryl trinitrate occurs in ventricular tissue but does not result in a substantial rate of c G M P synthesis (Ishibashi et al., 1993). On the other hand, S N P markedly elevates ventricular c G M P levels. Very few studies have thoroughly investigated regulation of ventricular c G M P levels by SNP, particularly in relation to ventricular contractility. However, based on the information which is available, S N P appears to be an efficacious cGMP-elevating agent, producing time- and dose-dependent increases in c G M P (EC 5 0 « 10 pM) with very little effect on ventricular contractile force (Keely et al., 1978; Lincoln and Keely, 1980, 1981). S N P is considerably more effective in elevating c G M P than are muscarinic receptor agonists. Very high concentrations of S N P (3 mM, 10 min) increased ventricular c G M P levels by 40-fold, without causing a negative inotropic effect in the canine ventricle (Endoh and Yamishita, 1981). On the contrary, 21 Inui et al. (1982) reported that S N P decreased isoproterenol-stimulated force development in rabbit ventricular tissue and elevated tissue c G M P content. Interpretation of this apparent association is complicated by the fact that different tissue types were used to measure c G M P (ventricular strips) and contractility (papillary muscles). Furthermore, c G M P measurements were made after exposure to 100 uM S N P for 30 - 60 s but contractility was determined in cumulative dose response curves allowing for exposure to S N P (10~6 - 10"3 M) over a total period of approximately 40 min. Time- and concentration-dependent studies of the relationship between S N P and c G M P in isolated ventricular cardiomyocytes have not been done nor has the effect of S N P on contractility and c G M P levels in p-adrenoceptor-stimulated cardiomyocytes been evaluated. The use of single concentrations of S N P (10 uM, 10 min) have shown significant increases in c G M P without direct effects on cell shortening (Stein et al., 1993). In contrast, another NO-releasing agent, SIN-1, dose-dependently decreased basal shortening in rat ventricular cardiomyocytes (EC 5 0 250 uM) but the role of c G M P in this effect was unclear since SIN-1 significantly decreased shortening at a concentration (1 uM) three orders of magnitude less than required to significantly elevate c G M P levels (Schluter et al., 1994). Effects of S N P on I C a suggest that the channel activity can be inhibited by S N P in a manner similar to carbachol and 8-bromo-cGMP. S N P (10 uM) reduced IBMX-stimulated I C a in guinea pig ventricular cardiomyocytes in a NO-dependent manner, based on its sensitivity to inhibition by methylene blue (Levi et al, 1994). If the only 22 action of NO in the ventricle is to activate guanylyl cyclase, the effects of S N P on IC a could reasonably be considered cGMP-dependent. 1.5 Atrial natriuretic peptide and the ventricle. A N P is a biosynthetic product of mammalian atrial cardiomyocytes which is released from granules into the cardiac and peripheral circulation upon atrial distention (see Wildey et al., 1992 for a review). A N P consists of 28 amino acid residues with an essential disulfide bond forming a 17-amino acid loop. A N P - A and A N P - B receptors have intrinsic guanylyl cyclase activity and A N P - C receptors bind A N P but do not catalyze the formation of c G M P (Anand-Srivastava and Trachte, 1993). A N P binding sites have been detected in coronary endothelia (Bianchi et al., 1985) and ventricular cardiomyocytes (McCartney et ai, 1990). 1.5.1 Effects of ANP on ventricular contractility. Parallels can be drawn between the response of the ventricle to treatment with NO donors and with A N P since A N P has also been observed to be without any effect on ventricular force in some cases (Hiwatari et al., 1986, Criscone et al., 1987; Bohm et al., 1988; Shimizu et al., 1988; Yanigisawa and Lefer, 1988; Dubois Rande et al., 1991; Hutter, 1991; Mikoluc and Wisniewska, 1994; Semigran et al., 1994) and, in other cases, to induce a moderate negative inotropic effect (Meulemans et al., 1988; Neyes and Vetter, 1989; Vaxelaire et al., 1989; McCall and Fried, 1990; Rankin and Swift, 1990; Stone et al., 1990; Smith et al., 1991). Therapeutically, A N P may improve left ventricular function under some conditions. For example, when infused into the coronary circulation at concentrations up to 2.3 nM in normal humans, A N P had no effect on cardiac function. However, patients with congestive heart failure responded 23 to 2.3 nM A N P with hemodynamic changes secondary to peripheral vasodilation, without changes in ventricular contractility (Dubois-Rande et al., 1991). The absence or presence of a negative inotropic effect of A N P cannot be explained on the basis of species specificity since both results have been variously reported in isolated hearts and ventricular preparations from guinea pig, rat, human, cat and dog. The question of genus specificity is not possible to determine due to the paucity of studies in non-mammalian cardiac preparations. Considering that the K d values of A N P binding to A N P receptors are 0.1 - 1 nM (Wildey et al., 1992) and that physiological A N P concentrations in humans are in the 0.001 - 1 nM range in plasma and are in the 0.1 - 1 nM range in the coronary sinus (Raine et al., 1986; Hirata et al., 1988), negative inotropic effects of A N P , when detected, were primarily observed at pharmacological concentrations and were fairly moderate in magnitude. For example, no changes in ventricular function were detected after peripheral infusion with 0.07 nM or 1.3 nM A N P in rabbits and a concentration of 2.5 nM A N P was required to decrease the rate of ventricular contraction by 14%, without any changes in maximal left ventricular pressure (Rankin and Swift, 1990). Only a 3 % decrease in twitch tension was seen in isolated ferret papillary muscle in the presence of 400 nM A N P (Smith et al., 1991) and 100 nM atriopeptin III (rat A N P , amino acids 10 to 33) decreased feline right papillary muscle twitch tension by only 5 % (Meulemans et al., 1988). An exception to these findings is the observation of a direct negative inotropic effect of 10 pM A N P in rat isolated cardiomyocytes (Neyes and Vetter, 1989). 24 Larger maximal depressor effects have been observed in some isolated ventricular cardiomyocyte preparations than were detected in intact preparations, e.g., in the range of 30 - 3 5 % decrease in basal cell shortening at concentrations from 1 nM to 250 nM (Neyes and Vetter, 1989; Vaxelaire et al., 1989). These effects of A N P persisted in the presence of isoproterenol in adult rat cardiomyocytes (Neyes and Vetter, 1989) but were absent in isoproterenol-stimulated avian cells (Vaxelaire et al., 1989). The negative inotropic effects of A N P seen in isolated cells required longer time periods to develop in comparison to rapidly-acting muscarinic receptor agonists. The onset of negative inotropy was within 4 - 5 minutes and maximal effects were achieved after 6 - 30 min (Vaxelaire et al., 1989; MacCall and Fried, 1990). Thus, as was the case for SNP, the evidence for a negative inotropic effect of A N P is inconclusive. The absence of a detectable negative inotropic effect in the absence and presence of a cAMP-elevating agent is a frequent finding and in those cases where decreases in force were detected, the effects were fairly modest and, with some exceptions, required high concentrations of the peptide. 1.5.2 Relationship between ANP-mediated cGMP elevation and ventricular contractility. Ventricular cardiomyocytes can generate elevated levels of c G M P in the presence of A N P , demonstrating the presence of functional A N P receptors, at least in the rabbit (Cramb et al., 1987), neonatal and adult rat (Neyes and Vetter, 1989; MacCall and Fried, 1990) and the embryonic chick (Vaxelaire et al., 1989). The potency of A N P in increasing c G M P in the ventricle (EC 5 0 « 10 nM) (Cramb et al., 25 1987) is similar to that in smooth muscle and endothelial cells (Hamet et al., 1989) and the maximal effect approaches 4 0 % of the maximal effect of S N P (Cramb et al., 1987). The correlation between c G M P and ANP-mediated negative inotropy is not very strong in the mammalian ventricle. Significant c G M P elevation by A N P was detected in rabbit cardiomyocytes at > 1 nM (Cramb et al., 1987) but inhibition of basal contractility occurred at considerably lower concentrations, ie, 10 pM A N P (EC 5 0 70 pM), in rat cardiomyocytes (Neyes and Vetter, 1987). Furthermore, ANP-mediated negative inotropy was not detected in cAMP-stimulated chick cardiomyocytes even though c G M P levels were significantly increased (Vaxelaire et al., 1989). Although the contractile effects of A N P do not parallel c G M P levels very well in the ventricle, an apparent cGMP-mediated inhibition of ventricular f C a by A N P has been observed in some mammalian preparations. Bkaily et al. (1993) noted that A N P reduced both basal and cAMP-stimulated (isoproterenol and intracellular^ dialyzed cAMP) I C a in human fetal ventricular cardiomyocytes by up to 7 0 % . 8-Bromo-cGMP mimicked the effects of A N P on basal I C a and inhibited IBMX-stimulated I C a in rabbit ventricular cardiomyocytes (Tohse et al., 1995). Therefore, A N P may inhibit mammalian ventricular l C a but the physiological significance of this is not clear (ie., the magnitude of the inhibitory effect on the calcium current is not consistently borne out functionally as a marked decrease in ventricular contractile force). It has been proposed that the contractile effects of A N P in the mammalian ventricle, when they occur, may correlate better with a decrease in cAMP levels (MacCall and Fried, 1990) through the inhibition of adenylyl cyclase (Anand-Srivastava and Cantin, 1986). However, definitive conclusions are difficult to draw, given the paucity of literature 26 concerning the relationship between A N P , cyclic nucleotides, and ventricular contractility. 1.6 Cyclic GMP-dependent protein kinase and ventricular contractility. A possible approach to resolving the controversy regarding the role of c G M P in mediating a negative inotropic effect in mammalian ventricle is through an evaluation of the activity of c G M P receptors upon agonist-induced c G M P elevation. The postulated downstream receptors for c G M P in the heart are primarily cGMP-stimulated cAMP phosphodiesterase (PDE II) (Fischmeister and Hartzell, 1987) and cGMP-dependent protein kinase (PKG). As noted above, PDE II activation by cGMP-elevating agents may play a significant role in decreasing the contractility of the amphibian ventricle while a phosphodiesterase-independent mechanism may exist in the mammalian ventricle, possibly involving activation of P K G (Lincoln and Keely, 1980, 1981; Mery et al., 1991; Levy et al., 1994). Binding of c G M P to P K G results in removal of intrasubunit pseudosubstrate inhibition associated with the regulatory domain of each subunit (R), and an increase in phosphotransferase activity associated with the catalytic domain (C). This leads to partial activation (2 mol cGMP/ mol PKG) and finally to full activation (4 mol cGMP/ mol PKG) (Wolfe et al., 1987). The reaction may be depicted as follows: (RC) 2 + 4cGMP (RC) 2 - cGMP 2 + 2cGMP <-» (RC) 2 - cGMP 4 inactive partially active fully active 27 1.6.1 Evidence for the presence of cGMP-dependent protein kinase in the mammalian ventricle. Biochemical evidence for the presence of P K G in the mammalian heart was obtained (Kuo, 1974) several years after the discovery of the enzyme (Kuo and Greengard, 1970). P K G , as a homodimer, is most concentrated in smooth muscle, platelets, lung, and cerebellar Purkinje cells and exists as two isoforms, la and ip, each subunit of which has a molecular weight of approximately 78 - 80 kDa (Hofmann et al., 1992; Francis and Corbin, 1994). Another isozyme, type II, may be monomeric and is found primarily in the plasma membrane of intestinal epithethial cells (DeJonge, 1981). Whole heart content of P K G , as measured by biochemical activity, was similar to that in vascular smooth muscle (Kuo, 1974) but immunocytochemical methods indicated that the primary cellular source for cardiac P K G was smooth muscle cells of the coronary vasculature and not myocardial cells (Ecker et al., 1989). Nevertheless, immunoblotting (Western blotting) methods have revealed the presence of P K G , possibly the la isozyme (Keilbach et al., 1992), in cardiomyocytes of the rat (Mery et al., 1991), although biochemical verification is not available in ventricular myocytes. P K G immunoreactivity has been detected in both soluble and particulate (detergent-extracted) fractions of the rat heart (Keilbach et al., 1992), contrary to initial reports that cardiac P K G was exclusively soluble (Lincoln and Keely, 1981), although the distribution of P K G within the cardiomyocyte itself is not yet established. 28 1.6.2 Evidence for a role of cGMP-dependent protein kinase in negative inotropic effects of cGMP-elevating agents. 1.6.2.1 Regulation of ICa. Indirect evidence for a role of P K G in the negative inotropic effects of c G M P -elevating agents may be found in studies of the effects of c G M P analogues on ICa. The amplitude of mammalian ventricular ICa, which had been pre-stimulated by c A M P -elevating agents including phosphodiesterase inhibitors, could be inhibited by 8-bromo-c G M P (Levi et al., 1989), a poor inhibitor of cAMP phosphodiesterase (IC 5 0 = 24 uM) (Francis et al., 1988) but a relatively potent activator of P K G (K a for la P K G « 2.5 nM, K a for ip P K G * 210 nM) (Sekhar et ah, 1992). This suggested that activation of P K G was a more likely mechanism for cGMP-associated IC a inhibition than activation of cAMP phosphodiesterase. The role of P K G was further substantiated by the observation that infusion of cardiomyocytes with catalytically active P K G inhibited p-adrenoceptor-stimulated IC a in a manner which mimicked 8-bromo-cGMP (Mery et al., 1991; Ono and Trautwein, 1991). Inhibition of IC a in the presence of phosphodiesterase inhibitors by carbachol, S N P and A N P to a similar extent as seen with 8-bromo-cGMP (Bkaily et al., 1993; Levi et al., 1994) suggests that P K G could be involved in some of the electrophysiologic effects of these agonists. 1.6.2.2 Phosphorylation of regulatory proteins. The level of phosphorylation of the cardiac L-type calcium channel or related proteins has not been determined but other proteins which regulate contractility are in greater abundance and are more readily evaluated for modulation by phosphorylation. 29 Phospholamban is one such protein. P K A activation by cAMP-elevating agents results in phospholamban phosphorylation in vivo (England and Shahid, 1987; Garvey et al., 1988), and this in turn leads to an accelerated rate of calcium sequestration and a greater release of calcium during systole. Functional studies of a mutant strain of mice which is phospholamban-deficient suggest that phospholamban phosphorylation is critical for the expression of a positive inotropic effect by cAMP-elevating agents such as (3-adrenoceptor agonists (Luo et al., 1994). Phosphorylation of another regulatory protein, troponin I, decreases the sensitivity of the myofibrils to calcium and accelerates relaxation, as seen during stimulation with cAMP-elevating agents (Robertson et al., 1982). Purified P K G phosphorylates both phospholamban and troponin I in vitro on residues which can also be phosphorylated by P K A (Lincoln and Corbin, 1978; Raeymakers et al., 1988). Purified P K G also catalyzes the phosphorylation of the calcium-release channel in the canine sarcoplasmic reticulum which should increase the opening probability of the channel (Takasago et al., 1991). Since PKA-mediated phosphorylation of these proteins is associated with the positive inotropic effects of such agents as (3-adrenoceptor agonists, it is difficult to foresee a PKG-mediated negative inotropic effect arising from phosphorylation of phospholamban, troponin 1 or the calcium-release channel. Furthermore, in vitro evidence for phospholamban and troponin 1 phosphorylation by P K G was not corroborated by results from in vivo studies. For example, significant phosphorylation of phospholamban required 30 minutes of exposure to S N P (0.01 - 1 mM) in rat cultured neonatal cardiomyocytes and no phosphorylation of phospholamban was seen in perfused guinea pig hearts in the presence of carbachol (1 pM, 2 min) or 8-30 bromo-cGMP (200 uM, 15 min) (Huggins et al., 1989). Conversely, A N P (1 - 1000 nM) increased phospholamban phosphorylation within 5 min in rat cultured neonatal cardiomyocytes but the relationship of this effect to P K G activation is unclear since significant accumulation of c G M P may require up to 30 min of exposure to A N P in the cultured neonatal myocyte preparation (McCall and Fried, 1990). As described above, dephosphorylation, rather than phosphorylation, has been associated with the negative inotropic effects of muscarinic receptor agonists. Cholinomimetic agents can inhibit isoproterenol- and forskolin-stimulated phospholamban and troponin I phosphorylation in an apparently cAMP-independent manner, i.e., without affecting agonist-stimulated PKA activation (Gupta et al., 1993; Bartel et al., 1993). A possible role for c G M P and, thereby, P K G in protein dephosphorylation is suggested by the observation that S N P (1 mM, 1 min) decreased phosphorylation of phospholamban and troponin I in isoproterenol-stimulated rat hearts (Bartel etal., 1993). 1.6.2.3 Inhibitors of cGMP-dependent protein kinase. Evaluation of the role for P K G in the negative inotropic effects of c G M P -elevating agents in intact tissue and cell preparations has been hampered by the lack of useful P K G inhibitors. The putative P K G inhibitor, KT5823 (Kase et al., 1987), was used in a single study to test the relationship between c G M P analogue-mediated contractile effects and P K G activity in the ventricle. KT5823 blocked the modest negative inotropic effect of 8-bromo-cGMP on rat cardiomyocytes (Shah et al., 1994). These results must be interpreted cautiously, since Wyatt et al. (1991) reported that KT5823 has no inhibitory effect on P K G in vitro and does not alter PKG-mediated 31 events in intact cells (neutrophils). Wyatt et al. (1991) and others (Wahler and Dollinger, 1995) also reported effects of KT5823 which were unique to the compound and apparently unrelated to P K G inhibition. 1.6.2.4 Measurement of cGMP-dependent protein kinase activity in intact ventricular tissue. If a role for P K G in the mediation of the negative inotropic effects of agents such as carbachol, S N P and A N P is to be verified, direct measurement of kinase activity in agonist-treated tissue and cellular preparations should reveal a positive correlation between activation of P K G and inhibition of contractility. This has been the case for other protein kinases, such as PKA and PKC, where their activation state correlated well with the functional effects of specific agonists in intact tissues (Murray and England, 1987; Langlands and Diamond, 1993). Measurement of P K G activity in agonist-treated intact vascular tissue revealed positive correlations between kinase activation and smooth muscle relaxation (Jiang et al., 1992), which supported the well established role for c G M P and P K G in modulation of vascular tone (Lincoln, 1989). The only reports of direct measurements of hormonally-activated P K G in the heart to date have implicated P K G activation in the negative inotropic effects of muscarinic receptor agonists and also provided an interesting explanation for some of the discrepancies noted above between agonist-induced c G M P elevation and negative inotropy (Lincoln and Keely, 1980, 1981). It was observed that acetylcholine decreased the force of contraction, increased c G M P levels, and activated P K G in a concentration-dependent manner in the isolated, perfused rat heart. Negative inotropy correlated well with P K G activation at concentrations between 0.03 and 3 pM 32 acetylcholine. On the other hand, S N P (1 - 30 uM) increased c G M P to a much larger extent than did acetylcholine but contractility was unchanged and, interestingly, P K G activity remained near control levels. Lincoln and Keely suggested that activation of P K G might be an essential step in the chain of events leading to inhibition of contractile force by muscarinic receptor agonists and that S N P failed to affect cardiac contractility because it elevated c G M P in a pool which does not have access to P K G . Intracellular compartmentalization of cyclic nucleotides has been postulated to explain conflicting information with regard to the effects of cAMP-elevating agents in the heart (Buxton and Brunton, 1983) and the reports of Lincoln and Keely (1980, 1981) were the first and only persuasive evidence for such a process pertaining to cGMP. An important potential complication of this work stems from the use of intact ventricular tissue. As mentioned above, intact tissues are multicellular and morphometric studies estimate the contribution of cardiomyocytes to the volume of the adult rat ventricle to be in the order of 7 5 % (Olivetti et al., 1980). It is possible that the contribution of c G M P and P K G from non-cardiomyocyte cells, particularly PKG-rich smooth muscle cells, may obscure agonist-induced changes in the cardiomyocyte itself. 15 years have passed since the publication of the provocative work by Lincoln and Keely (1980, 1981), and there has been no confirmation by others that the reported agonist-induced changes in tissue c G M P levels and P K G activity are reproducible and reflect events within the cardiomyocyte. 1.7 Improvement in cGMP-dependent protein kinase assay. Traditionally, P K G has been measured in purified preparations and crude extracts from agonist-treated intact tissue by the determination of 3 2 P incorporation into 33 histones, typically histone H2B , from [y- P]- labeled A T P (Lincoln and Keely , 1980, 1981; F iscus et al., 1984, 1985; Lincoln et al., 1988). Histone is inexpensive and readily avai lable but is not an ideal substrate for quantitative analys is of P K G activity because it increases the binding of c G M P to P K G (Tse et al., 1981) and directly increases P K G activity (Walton and Gi l l , 1981). Histone H 2 B is a lso a substrate for other cycl ic nucleot ide-dependent and - independent k inases (Lincoln and Keely , 1980, 1981; F i scus et al., 1984). Furthermore, P K G - m e d i a t e d incorporation of phosphate into histone var ies depending on the commerc ia l source of the substrate (Flockerzi et al., 1978). Recent ly , a P K G substrate named B P D E t i d e has been des igned , based on the phosphorylat ion site for P K G on cGMP-spec i f i c , c G M P - b i n d i n g phosphod ies terase (Colbran et al., 1992). B P D E t i d e is 16-fold more speci f ic for P K G than P K A , making it the most speci f ic substrate for P K G to be descr ibed. By compar ison , the site on histone H 2 B which is phosphorylated by P K G is just as effectively phosphory lated by P K A (Colbran et al., 1992). Cyc l ic nucleot ide- independent phosphorylat ion of B P D E t i d e a lso appears much lower than corresponding controls with histone H 2 B (e.g., compare J iang et al., 1992 and F iscus et al., 1984). Th is sugges ts that B P D E t i d e is less likely to be non-specif ical ly phosphorylated than histone H 2 B . B P D E t i d e , has been used successfu l ly to detect agonist-mediated P K G activation in intact vascu lar t issue (Jiang et al., 1992). Until the study reported here, B P D E t i d e had not been used to investigate agonist-mediated P K G activation in card iac t issue. 34 1.8 Summary and rationale for proposed experiments. The background introduced above illustrates the conflicting evidence for and against a role for c G M P in mediating the negative inotropic effects of muscarinic receptor agonists, S N P and A N P in the mammalian ventricle. Evidence consistent with such a mechanism is reasonably convincing in the case of muscarinic receptor agonists where concentration- and time-dependent correlations between c G M P and negative inotropy are consistently observed. The evidence is far less convincing in the case of S N P and A N P , since c G M P elevation by these agents is often not accompanied by corresponding negative inotropic effects. A paucity of information from isolated cardiomyocytes in terms of agonist-mediated changes in cyclic nucleotide levels and contractility hampers the analysis of the problem. A unifying hypothesis was proposed by Lincoln and Keely (1980, 1981) which might account for to these discrepancies. Lincoln and Keely hypothesized that P K G activation was an essential step in cGMP-mediated negative inotropy and, in order for P K G activation to occur, elevation of c G M P was required within a functionally specific pool. Muscarinic receptor agonists, which mediate a negative inotropic effect, could elevate c G M P in a pool that had access to P K G , but SNP, which generally does not mediate a negative inotropic effect, did not elevate c G M P in the requisite compartment and did not activate P K G . Evidence which supports this hypothesis has not been independently confirmed. The present study was undertaken to test the hypothesis that c G M P mediates the negative inotropic effects of specific agonists by the selective activation of P K G in 35 mammalian ventricular tissue and cardiomyocytes. To this end, the following experiments were performed with rat heart muscle preparations: 1. The contractile effects of carbachol, S N P and A N P were examined in the intact heart perfused in the presence of isoproterenol and in right ventricular strips in the presence and absence of isoproterenol. Attempts were made to correlate contractile changes in the intact ventricle with changes in c G M P and c A M P levels and P K G activity. 2. The contractile effects of carbachol, acetylcholine and S N P were examined in isoproterenol-stimulated isolated ventricular cardiomyocytes and were compared to changes in c G M P and cAMP levels and P K G activity. Characterization of ventricular P K G and validation of the P K G assay were important components of the study. 36 2.0 MATERIALS AND METHODS 2.1 Chemicals and materials Chemica l s and materials were purchased from or provided by the following sources : A m e r s h a m International. (Little Chalfont. Buck inghamshi re . England) BIOTRAK® c G M P scintillation proximity assay kit, BIOTRAK® c A M P scintil lation proximity a s s a y kit, [y - 3 2 P] adenos ine 5'-tr iphosphate. B a c h e m Cal i fornia (Torrance. Cal i fornia. U.S.A.) c G M P - d e p e n d e n t protein k inase substrate (BPDEt ide) , atrial natriuretic peptide (rat A N P , 1 - 28 amino acids). B D H Inc. (Vancouver. British Co lumbia , Canada ) sod ium hydrogen carbonate, d-g lucose, tr ichloroacetic ac id. B I O L O G Life Sc ience Institute (Bremen. Germany) R p - 8 - C P T - c G M P S . Biomol R e s e a r c h Laborator ies. Inc. (Plymouth Meet ing. P A . U S A ) a catalytic subunit of porcine heart c A M P - d e p e n d e n t protein k inase, bovine lung 1a c G M P - d e p e n d e n t protein k inase ho loenzyme. B io -Rad Laborator ies (Hercules. Cal i fornia. U S A ) acry lamide (99.9%), ammonium persulfate, 5-bromo-4-chloro-indolyl phosphate, chymotrypsin, glycine, goat anti-rabbit IgG alkal ine phosphatase conjugate, nitroblue tetrazol ium, r ibonuclease, N,N,N',N'-tetramethylethyldiamine, sod ium dodecy l sulfate, N,N'-methylene-bis-acry lamide, protein assay kit, and vitamin B 1 2 Boehr inger Mannhe im G m b H (Mannheim. Germany) 37 bovine serum albumin (fraction V) Calbiochem-Novobiochem Corp. (San Diego. CA. U.S.A.) KT5823, microcystin-LR Carnation Inc. (Toronto. Ontario. Canada) Skim milk powder. DuPont NEN Research Products (Boston. MA. U.S.A.) [y-3 2P] adenosine 5'-triphosphate. Fisher Scientific Co. (Fair Lawn. New Jersey. USA) calcium chloride, copper sulfate, o-phosphoric acid, potassium chloride, potassium dihydrogen phosphate, magnesium acetate, magnesium chloride, magnesium sulfate, sodium carbonate, sodium chloride, sodium hydroxide, sodium tartrate, Scintiverse® scintillation fluid. GIBCO Life Technologies (Grand Island. NY. U.S.A.) basal minimum Eagle amino acid solution, Joklik-modified minimal essential medium (powder), M199 medium (powder). Hyclone Laboratories. Inc. (Logan. Utah. USA) fetal bovine serum. Kinetec Biotechnology Corp. (Vancouver. British Columbia. Canada) Pre-stained molecular mass standards, antibody to C terminus of cGMP-dependent protein kinase, antibodies to N termini of types la and l(3 cGMP-dependent protein kinase. Medigas (Vancouver. British Columbia. Canada) Oxygen, 5% carbon dioxide/ 95% oxygen. 38 M T C Pharmaceut ica ls (Cambridge, Ontario. Canada) Somnotol® (sodium pentobarbital). Organon Techn ika . Inc. (Toronto. Ontario. Canada ) Hepar in sulfate. S i g m a Chemica l C o . (St. Louis. U S A ) acetylchol ine, adenos ine 3':5'-cycl ic monophosphate, adenos ine 5'-tr iphosphate, benzamid ine, apoprotinin, /-arginine, ascorb ic ac id , porcine c A M P - d e p e n d e n t protein k inase ho loenzyme, carbachol , diethyl ether, dimethyl formamide, dithiothreitol, ethylenediaminetetraacet ic ac id , Folin and Ciocal teu 's phenol reagent, guanos ine 3':5'-cycl ic monophosphate , (3-glycerophosphate, 3-isobutyl-1-methylxanthine, /-isoproterenol, leupeptin, pepstatin A , phenylmethylsulfonyl f luoride, physost igmine, protein k inase inhibitor (PKI), sod ium pyruvate, sod ium nitroprusside, soybean trypsin inhibitor, Triton® X -100 , tris base , tris-HCI, trypan blue solut ion. Scheoc je r and Schue l l (Keene. N H . U S A ) Protran® nitrocellulose membrane. Wha tman Ltd. (Maidstone. Kent. England) Phosphoce l lu lose paper (P81) Worthington B iochemica l Corp . (Freehold, N J r U S A ) co l lagenase (type II, C L S II). 3 9 2.2 Animals. Male Wistar rats (200 - 300 g) were obtained from the Animal Care Facility, University of British Columbia and housed 2 to 3 to a cage. Animals had free access to water and food. 2.3 Measurement of contractile properties of rat cardiac preparations 2.3.1 Preparation of Langendorff-perfused rat hearts. Hearts were removed from male Wistar rats (200 - 300 g) at least 15 min after administration of heparin (2200 lU/kg, i.p.) and sodium pentobarbital (80 mg/kg, i.p.) and were placed in ice-cold, aerated (5% C 0 2 in 0 2 ) Chenoweth-Koelle buffer of the following composition (mM): NaCl (120), KCI (5.6), CaCI 2 (2.18), MgCI 2 (2.1), N a H C 0 3 (19), /-arginine (0.1), and d-glucose (10). After removal of non-cardiac tissue, hearts were attached by the aorta to the Langendorff-perfusion apparatus (Langendorff 1895, 1897; Fawzi, 1985) and perfused via a peristaltic pump (#7553-20, Cole-Parmer, Chicago, IL, U.S.A.) in a retrograde direction with Chenoweth-Koelle buffer (35°C) at 10 mL/min (19 mL/min with heart unattached to perfusion apparatus). A water-filled («100 uL) latex balloon, attached to PE-50 tubing, was inserted into the left ventricle via the left atrium and left ventricular pressure was measured using a Statham P23AA pressure transducer (Stratham-Gould Instruments, Cleveland, Ohio). Diastolic pressure was 0 - 4 mm Hg. Hearts were paced at 255 beats/min by means of two platinum electrodes placed on either side of the ventricles (5 ms pulse, « 2V, Grass SD9 square wave stimulator). This electrode configuration was used in order to bypass the effects of agents on atrioventricular electrical conduction. Maximal left ventricular pressure (LVP) and rates of rise and fall of left ventricular pressure (positive and negative dP/dt) 40 were recorded on a G r a s s model 79D polygraph. Max imal L V P and posit ive and negat ive dP/dt were measured and ana lyzed using a microcomputer (software des igned by Ro land Burton, University of British Co lumb ia , Vancouver , C a n a d a ) . 2.3.2 Measurement of contractility of Langendorff-perfused rat hearts. After perfusion for at least 20 min, hearts were perfused with 1) increasing concentrat ions of isoproterenol (isoproterenol serial concentrat ion-response curves, 0.1 - 100 nM) or 2) 1 nM isoproterenol for 2 - 3 min, fol lowed by a washout of the drug and then a repeat exposure to isoproterenol in the p resence and a b s e n c e of 10 u M carbacho l or 10 uM S N P , or 3) 1 nM isoproterenol for 2 - 3 min, fol lowed by a washout , a 5 min pre-incubation with 100 nM A N P and then the addit ion of 1 n M isoproterenol. Drugs were added to one of two perfusion buffer reservoirs (35°C) and wash-out of a drug effect was ach ieved by switching the source of perfusion buffer to the second (control) reservoir v ia a 2-way stopcock. The reservoir containing drug w a s foil-wrapped and room lighting was d immed during S N P treatment. At speci f ied t imes during drug treatment, the heart was f reeze-c lamped with metal tongs pre-cooled in liquid N 2 . The atrial t issue from the frozen heart was tr immed away and the ventricular t issue w a s stored at -70°C until b iochemical assays were performed. 2.3.3 Preparation of rat right ventricular strips. Rats were anesthet ized as descr ibed above and the hearts were p laced in warmed (35°C), aerated (5% C 0 2 in 0 2 ) Chenoweth-Koe l le buffer. The right ventricle w a s d issected from the heart and cut into 4-6 strips. O n e end of e a c h strip w a s impaled by 2 platinum electrodes and the other end was c lamped with a micro-c lamp and attached to a G r a s s F T 0 3 force-displacement t ransducer. T i ssues were mounted 41 in 20 ml_ water-jacketed baths containing Chenoweth-Koelle buffer (35°C, aerated with 5 % C 0 2 in 0 2 ) , placed under 1 g resting tension, and stimulated to contract with electrical pulses (5 ms, 1 Hz, < 10 V). Contractile activity was recorded on a polygraph (Grass 7 D). The tissues equilibrated for 60 min during which time resting tension was frequently re-adjusted to 1 g and the buffer was replaced every 15 min. 2.3.4 Measurement of contractility of rat right ventricular strips. Maximal twitch tension was evaluated in the presence or absence of drugs. After the initial equilibration period, drug treatments were performed by adding drugs from stock solutions directly into the tissue bath. Cumulative concentration-response curves to isoproterenol (10"1° - 1 0 " 5 M) were obtained by the addition of drug every 2 - 3 min (at steady state of twitch tension). In separate experiments, the direct effects of carbachol (100 pM), S N P (100 pM) or A N P (100 nM) on maximal twitch tension were assessed for up to 9 min. At least one strip from each ventricle was not exposed to drugs and was used as a control. In another series of experiments, the effects of carbachol (10 pM), S N P (100 pM) or A N P (100 nM) on maximal twitch tension in the presence of isoproterenol were determined. Tissues were exposed to isoproterenol (30 nM) until twitch tension reached steady- state ( 2 - 3 min), at which time carbachol, S N P , or A N P was added to the bath and tension was monitored for up to 6 min. At least one strip from each ventricle was exposed to only isoproterenol and was used as a control. Lighting was dimmed and baths were foil-wrapped during S N P treatment. 2.3.5 Preparation of isolated rat ventricular myocytes. The isolation procedure was modified from the protocol of Kryski et al., 1985. Rats were anesthetized as described above and hearts were removed and placed in 42 ice-cold, oxygenated (100%O 2) M199 medium containing 20 mM A/-2-HEPES, 5 mM N a H C 0 3 a n d 5 mM sodium pyruvate. Distilled water purified by filtration through a 0.22 um filter was used for all buffers. Each heart was perfused in a retrograde fashion using a modified Langendorff apparatus at a flow rate of 9 mL/min (peristaltic pump #2500-001, Harvard Apparatus, South Natick, MA, U.S.A.) with nominally calcium-free Joklik-modified minimal essential medium (MEM) containing 20 mM H E P E S , 5 mM N a H C 0 3 and 5 mM sodium pyruvate at pH 7.4 and 37°C with constant aeration with 100% 0 2 . After 5 min, the perfusate was changed to M E M with collagenase (1.03 mg/ml, 242 units/mg, type II, class 2), C a 2 + (50 uM) and bovine serum albumin (0.1 %, BSA, fraction V) and a recirculating perfusion was performed for 20-30 min with increases in C a 2 + to 100 uM and 200 uM at 10 and 15 min. The ventricles were removed, minced, and agitated in the same buffer used in the recirculating perfusion (including 200 uM C a 2 + ) for 8 min in a water-jacked beaker at 37°C. The suspension was filtered through a 200 um nylon mesh and the filtrate centrifuged at approximately 60g in a Dynac bench-top centrifuge (Becton, Dickenson Co., Parsippany, NJ, U.S.A) for 45 s. The pellet was re-suspended in M E M (200 uM C a 2 + , 1 % BSA) and the cells were allowed to sediment at 1g for 10 min at room temperature. This washing procedure was repeated twice with M E M containing firstly 500 uM and then 1 mM C a 2 + . Finally, the cells were counted and examined for viability using trypan blue exclusion (70% mean viability) and suspended in M199 (BSA 1%, fetal bovine serum 4%) for 30 min at room temperature until used in contractility studies or cyclic nucleotide studies. 43 2.3.6 Measurement of contractility of single isolated rat ventricular myocytes. Isolated cells were plated for 45 min at room temperature onto circular glass coverslips (22 mm diameter) which had been previously etched with concentrated nitric acid, thoroughly washed with ultrapure H 2 0 and incubated at room temperature for >1 hour with laminin (9 pg/coverslip, diluted in M199) (Haddad et al., 1988). Coverslips were then placed in a temperature-controlled chamber and viewed through a Nikon Diaphot TMD inverted microscope. Cells were constantly perfused at 5 mL/min with modified Krebs -HEPES buffer (with or without added drugs) with the following composition (mM): 118 NaCl, 4.96 KCI, 1.19 K H 2 P 0 4 , 1.19 M g S 0 4 , 1.8 CaCI 2 , 5 sodium pyruvate, 20 H E P E S , 5 N a H C 0 3 , 10 d-glucose, and basal minimum Eagle amino acid solution (including 0.1 mM l-arginine) at 30°C and pH 7.4 with constant aeration with 100% 0 2 . Perfusion was run in darkness when S N P was applied; thus, the drug was exposed to visible light only during the brief time in which it flowed through the cell chamber. A single cell was electrically stimulated with a 5 ms pulse at 0.5 Hz and « 4.5 V (1.5 x threshold) using a Grass S11 dual output digital stimulator. The cell image (phase-contrast optics) was detected by a COHU C C D video camera and contractility was assessed by video edge detection instrumentation (Steadman et al., 1988) with a temporal resolution of 16.7 ms and a spatial resolution of <0.1 pm. The digitized output was evaluated with the computer program Oscar 2.06 (Photon Technology International, Inc.). This software utilized the Savitzky-Golay method of differentiation to generate rates of contraction and relaxation. Only rod-shaped cells with no membrane blebs which showed no spontaneous contractions between 44 electrical stimulations and which demonstrated a minimum cell shortening of 5 % were used. Cells were tested within 8 hours of isolation. The average cell length at diastole was 110.0 ±4.0 urn (n=32). The treatment protocol for isolated ventricular cardiomyocytes was as follows: following an initial control period, an electrically stimulated cell was perfused with buffer containing 1 nM isoproterenol for > 2 min while parameters of contractility were monitored. The buffer was then changed back to control solution until contractility returned to baseline ( « 10 min). The cell was then perfused with isoproterenol and carbachol ( 1 - 1 0 uM), acetylcholine ( 1 - 1 0 uM) or S N P (10-100 uM) for at least the same period of time. After washout, treatment with isoproterenol alone was repeated to establish reproducibility of the isoproterenol effect. 2.4 cGMP and cAMP estimation. 2.4.1 cGMP and cAMP estimation in rat ventricular tissue. Frozen ventricular tissue from agonist-treated Langendorff-perfused rat heart was crushed using 4 strokes of a liquid nitrogen-cooled French press. Approximately 75 - 100 mg of the frozen, crushed ventricular tissue was placed in a liquid N 2-chilled capsule (1 mL capacity) with a chilled metal pestle and pulverized in a Vari-Mix III dental amalgam mixer (10 s at medium speed, 5 s at high speed). Ice-cold TCA (6% w/v, 750 uL) was added to the contents of the capsule and the mixture was agitated (10 s at high speed) in the Vari-Mix. The homogenate was removed and the capsule was rinsed with 750 uL TCA , which was added to the homogenate. The homogenate was centrifuged in a Heraeus Contifuge 28RS centrifuge (2,000g for 15 min at 4°C). The supernatant was extracted 4 times in 5 mL water-saturated diethyl ether (upper 45 ether phase was discarded after each wash). The remaining aqueous phase was warmed in a hot water bath to evaporate any remaining ether and was weighed to determine its volume. A sample of the aqueous phase was diluted by 10-fold in 0.05 M acetate buffer and assayed in duplicate for c G M P using a commercially available cGMP-scintillation proximity assay (SPA) radioimmunoassay (acetylation protocol) and a sample was diluted 4-fold in 0.05 M acetate buffer and assayed in duplicate for cAMP (cAMP-SPA radioimmunoassay, non-acetylation protocol). The TCA-insoluble pellet was assayed in duplicate for protein using a modified Lowry procedure (Markwell et al., 1981). Ventricular cyclic nucleotide content was calculated per mg protein. 2.4.2 cGMP and cAMP estimation in suspensions of rat ventricular isolated cardiomyocytes. Cardiomyocytes (2 ml_, approximately 125,000 cells) were suspended in modified Krebs -HEPES buffer (pH 7.4, 30°C) and treated with either 1 nM isoproterenol, 1 or 10 pM acetylcholine and 50 pM physostigmine plus isoproterenol, 1 or 10 pM carbachol plus isoproterenol or 10 or 100 pM S N P plus isoproterenol in 50 ml_ conical-bottomed plastic tubes (foil-wrapped when S N P was used). After 2 or 6 min, 4 .33 ml_ ice-cold ethanol was added and the mixture vortexed and left on ice for 30 min after which it was centrifuged at 2000gf for 15 min. Each treatment was paired with a control (no addition) sample from the same suspension of cardiomyocytes. The treated and control pair were selected from the total suspension of cardiomyocytes at the same time and exposed to ethanol at the same time. The supernatant was dried (Speed-Vac) and stored at -70°C until c G M P or cAMP determination using commercially available radioimmunoassay kits ( cGMP-SPA or c A M P - S P A 46 radioimmunoassay) with an acetylation protocol. The dried sample was resuspended in 400 uL of 0.05 M acetate buffer for c G M P determination and diluted further by 125-fold in acetate buffer for cAMP determination. The ethanol-insoluble pellet was assayed for protein using a modified Lowry procedure (Markwell et al., 1981) and cyclic nucleotide content was calculated per mg protein. 2.5 Preparation of soluble and particulate fractions of rat cardiac tissue. 2.5.1 Preparation of soluble and particulate fractions of rat ventricular tissue. Frozen ventricular tissue from agonist-treated Langendorff-perfused rat heart was crushed using 4 strokes of a liquid nitrogen-cooled French press. Approximately 150 mg of the frozen, crushed ventricular tissue was placed in a liquid N 2-chilled capsule (1 mL capacity) with a chilled metal pestle and pulverized in a Vari-Mix III dental amalgam mixer (10 s at medium speed, 5 s at high speed). Ice-cold homogenization buffer (5 volumes) was added to the capsule and the mixture was agitated in the Vari-Mix (10 s at high speed). The composition of the buffer was as follows: 10 mM H E P E S (pH 7.4), 1mM EDTA, 10 mM DTT, 1 mM IBMX, 125 mM KCI, 1 mM benzamidine, 10 pg/mL leupeptin, 10 pg/mL pepstatin A, 1 ug/mL apoprotinin. P M S F (1 mM final concentration) was added directly to the homogenization buffer in the capsule immediately prior to pulverization. The homogenate was transferred to cold 1.5 mL microcentrifuge tubes and centrifuged in a Heraeus Contifuge 28RS centrifuge (30,000g for 5 min at 4°C). The supernatant (soluble fraction) was placed on ice and immediately assayed for kinase activity as described below. Since each sample was individually homogenized, centrifuged and immediately assayed for kinase activity, no 47 more than 7 min passed from the initiation of homogenization to initiation of soluble kinase activity measurement. The particulate fraction was prepared from the pellet remaining after removal of the soluble fraction. The pellet was rinsed with 500 pL of ice-cold homogenization buffer and then resuspended in 5 volumes of homogenization buffer which included 0.1 % (w/v) Triton X-100 detergent as a solubilizing agent (Uno et al. , 1976). The mixture was placed on ice and gently vortexed 3 times for 10 s each every 10 min. After 30 min, the mixture was centrifuged as before and the supernatant (particulate fraction) was assayed in duplicate as described below. Soluble and particulate fractions were used in protein kinases assays, chromatographic separation and immunoblotting. 2.5.2 Preparation of soluble and particulate fractions of rat isolated ventricular cardiomyocytes. Cardiomyocytes were isolated as described above. Cells prepared from a single rat heart were evenly divided into two 50 ml_ conical tubes containing 10 mL of modified Krebs -HEPES buffer (approximately 5 x 10 6 cells). One suspension was treated with 1 nM isoproterenol for 2 min and the other suspension was treated with 1 nM isoproterenol and 10 uM carbachol, 10 uM S N P or 100 nM A N F for 2 min. The cells were then pelleted in a Dynac centrifuge (speed at #70 for 10s), the buffer removed, and the cell pellet rapidly frozen by immersion of the tube in liquid N 2 and stored at -70°C until required for preparation of subcellular fractions. Preparation of the soluble and particulate fractions from cell pellets was essentially as described above for ventricular tissue. Soluble and particulate fractions were used in protein kinase assays, chromatographic separation and immunoblotting. 48 2.6 cGMP- and cAMP-dependent protein kinase assay. A phosphocellulose paper assay was used to measure the phosphotransferase activity of P K G and PKA in rat ventricular tissue and cardiomyocytes. P K G activity was determined by measuring the transfer of 3 2 P from [y - 3 2P] ATP to the serine-4 residue of the synthetic peptide, BPDEtide (RKISASEFDRPLR) . P K G activity was measured in a total volume of 70 pL containing 150 pM BPDEtide, 10 mM H E P E S , 4 mM magnesium acetate, 150 pM ATP (2.5pCi/tube for assay of soluble and particulate fractions of rat ventricular tissue and cardiomyocytes, 1 pCi/tube for assay of MonoQ fractions), 5 pM PKA synthetic inhibitor (PKI), 35 mM p-glycerophosphate, 100 ng/mL microcystin-LR, 0.5 mM EGTA, and 5 pM heparin in the presence and absence of 5 pM cGMP. KT5823 (2 pM) or R p - 8 - C P T - c G M P S (2 or 10 pM) were present in the appropriate experiments, which were conducted in dimmed ambient lighting. No-substrate blanks were determined for each sample of rat ventricular tissue and cardiomyocyte crude extracts (soluble and particulate fractions) in the presence and absence of added c G M P to correct for phosphorylation of endogenous substrates. The reaction was initiated by adding 20 pL of sample ( « 100 pg soluble or 10 pg particulate fraction protein of rat ventricular tissue or cardiomyocytes, or MonoQ column eluate) to 50 ut reaction cocktail. The reaction proceeded for 4 min at 4°C (or as otherwise specified) and was stopped by spotting 50 pL of the reaction mixture onto 2 x 2 cm squares of P81 phosphocellulose paper, pre-coated with 1 mM ATP and 10 mM K H 2 P 0 4 . Measurement of activity at zero time (temporal linearity study) was accomplished by adding 50 pL of the reaction mixture to 20 pL sample, rapidly mixing (the mixture was drawn into and out of pipette), and immediately spotting 50pL mixture 49 onto P81 paper. The paper was air-dried for 20 s and dropped into approximately 500 mL of gently-stirred 0 .5% o-phosphoric acid, washed 4 times for 10 min each, and then air-dried. Each paper was placed in a 7.5 mL scintillation vial containing 2 mL Scintiverse scintillation fluid. Fluorescence of the scintillant due to p-radiation emitted by the paper was detected by a Beckman LS 6000TA liquid scintillation counter. The assay conditions allowed for the estimation of P K G activity using less than 0 .05% of A T P and BPDEtide. P K G activity was expressed as pmols or fmols of phosphate incorporated into substrate per min per mg protein or uL of eluate. The extent of P K G activation was assessed by calculating the activity ratio, which is the ratio of activity in the absence of added c G M P to kinase activity in the presence of sufficient c G M P to maximally activate the enzyme. Calculation of K m for phosphorylation of BPDEtide by P K G was accomplished using Prism® software (Prism® version 1.03 software, Graphpad Software, Inc., San Diego, CA, U.S.A.). PKA activity in eluates from the MonoQ chromatography of soluble fractions of rat ventricular tissue and soluble and particulate fractions of rat cardiomyocytes was measured using the same assay conditions as the P K G assay except that PKI was excluded and samples were assayed in the presence and absence of 5 uM cAMP, rather than cGMP. 2.7 Chromatographic separation of protein kinases in rat ventricular extracts. Cyclic nucleotide protein kinase activity was analyzed in soluble fractions of rat ventricular tissue and soluble and particulate fractions of rat ventricular cardiomyocytes chromatographically separated using a Pharmacia MonoQ anion exchange column 50 (HR5/5) in a Pharmacia F P L C system (Pharmacia LKB Biotech, Uppsala, Sweden). All procedures were done at 4°C. The column, pre-equilibrated with buffer A (10 mM tris-HCI, pH 7.4, 2 mM EDTA, 1 mM DTT, 0.2 u mesh filtered), was loaded with up to 22 mg of sample protein (using a 10 mL sample loader) at a flow rate of 0.5 mL/min. After the column was washed with 5 mL of buffer A to remove unbound protein, buffer B (buffer A plus 400 mM NaCl) was applied to the column in combination with buffer A so as to develop a linear gradient of NaCl (0 - 400 mM). Eluate was collected at a rate of 0.5 mL/min in 40 x 0.5 mL fractions which were assayed for P K G and P K A activities. Recovery of kinase activity was reported as the amount of kinase activity eluted from the column (in pmole/min) as % of the total kinase activity applied to the column (in pmole/min). Selected fractions were immunoblotted to determine the presence of P K G . 2.8 SDS-Polyacrylamide gel electrophoresis and immunoblotting of rat ventricular and isolated ventricular cardiomyocyte chromatographic eluate. A Bio-Rad Protein II electrophoresis unit was used to perform the separation of proteins by sodium dodecyl sulfate polyacryamide gel electrophoresis (SDS-PAGE) . Samples (200uL) were digested by boiling for 3 min with digestion buffer (final concentrations: 2 % w/v SDS, 120 mM tris-HCI, pH 6.8, 10% glycerol, 5 % B-mercaptoethanol, 0.004% bromophenol blue). Molecular mass pre-stained standards were treated similarly. Molecular masses of standards were 97.4 kDa (rabbit muscle phosphorylase b), 66.2 kDa (bovine serum albumin), 55.3 kDa (hen egg white albumin), 33.2 kDa (glutathione S-transferase), 23 kDa (soybean trypsin inhibitor) and 51 15.6 (cytochrome C). Commercially available purified kinases (bovine lung la P K G , 0.1 mg/mL; porcine heart PKA (0.1 mg/mL) and P K A catalytic subunit, 0.05 mg/mL) were diluted as follows: 50 pL kinase solution, 25 pL distilled water and 25 pL digestion buffer. Treatment of purified kinases was then the same as for the samples. Slab gels were cast according to the method of Laemmli (1970). The separating gel contained 1 1 % (w/v) total acrylamide (acrylamide:N,N'-methylene bisacrylamide = 37.5:1), 375 mM tris-HCI (pH 8.8), 0 . 1 % SDS (w/v), 0.042% (w/v) ammonium persulfate and 0 .03% (w/v) N,N,N',N'-tetramethylethylenediamine (TEMED). The stacking gel contained 4 % (w/v) total acrylamide, 125 tris-HCI (pH 6.8), 0 . 1 % SDS (w/v), 0.08% (w/v) ammonium persulfate and 0.05% (w/v) TEMED. Samples (110 pL), purified kinases (20 pL) and pre-stained molecular mass standards (110 pL) were introduced into sample wells in the stacking gel and the upper and lower tank filled with running buffer (25 mM tris, pH 8.3, 192 mM glycine and 0 . 1 % w/v SDS). Proteins were electrophoretically separated on the gel by applying a constant current (10 mA per gel, overnight). At the end of the electrophoresis, the stacking gel was discarded and the separating gel was mounted in a Hoefer TE 50 Transphor (Hoefer Scientific Instruments) unit for transfer of the resolved proteins onto a nitrocellulose membrane (Tobwin et al., 1979). Transfer was achieved by applying a constant current of 250 mA for 3 hours at 4°C across a bath solution (20% (w/v) methanol, 20 mM tris, 120 mM glycine and 0.008% SDS). The nitrocellulose membrane was treated for 2 hours in blocking buffer, which consisted of 3 % (w/v) skim milk powder in TTBS (20 mM tris-HCI, pH 7.4, 0.5 M NaCl, and 0.05% (w/v) Tween 20) to eliminate non-specific binding and then washed with TTBS ( 4 x 5 min). The membrane was then probed with a polyclonal, affinity-purified 52 antibody raised against a peptide sequence (CDEPPPDDNSGWDIDF) derived from the carboxyl terminus of the la isoform of P K G . The membrane was exposed to the primary antibody (1/200 diluted in antibody buffer which was TTBS and 0 .05% sodium azide) overnight at room temperature. The membranes were then washed ( 4 x 5 min) in TTBS and incubated for 2 h with the secondary antibody (1/2000 diluted goat anti-rabbit IgG alkaline phosphatase conjugate). The membranes were washed again with TTBS ( 4 x 5 min) and rinsed twice for 2 minutes each with TBS (TTBS without Tween 20). Immunologically recognized proteins were detected by the color reaction due to the interaction of alkaline phosphatase and its substrates, nitroblue tetrazolium (NBT) and 5-bromo-4-chloro-indolyl phosphate (BCIF). The reaction was performed in 50 mL A P buffer (100 mM tris-HCI, pH 9.5, 100 mM NaCl and 5 mM Mg Cl 2) in which NBT (170 u.L of a 50 mg/mL stock in dimethyl formamide) and BCIF (340 u L of a 50 mg/mL stock in 7 0 % dimethyl formamide) were added. 2.9 Preparation of drug solutions. Stock solutions of acetylcholine, carbachol, isoproterenol, and S N P were prepared in distilled water on the day of use. Ascorbic acid (1 mg/mL) was present in stock solutions of isoproterenol. Solutions of S N P were protected from light. Solutions of A N P were prepared in either 0.05 M acetic acid or distilled water and were stored in aliquots at -20°C until required for use. Either solvent provided chemically stable conditions for the peptide (personal communication, Bachem California, Technical Services). KT5823 was prepared in dimethylsulfonylformamide and was stored at -20°C until required for use. 53 2.10 Protein determination. Protein in ethanol-precipitated cardiomyocyte preparations and in TCA-precipitated ventricular preparations (from cyclic nucleotide studies) was measured using the method of Lowry et al. (1951) as modified by Markwell et al. (1981). Briefly, ethanol-precipitated cardiomyocyte pellets were incubated in 250 uL DNase I (0.5 mg/mL) overnight at room temperature to digest ethanol-precipitated DNA. Reagent A (1.75 mL, 2 % N a 2 C 0 3 , 0.4% NaOH, 0.16% sodium tartrate, 1 % SDS) was added, the mixture vigorously vortexed and incubated for 2 h at 30°C. After the pellet was solubilized, a 40 - 60 uL aliquot of sample was diluted to 1 mL with distilled water, to which 3 mL reagent B (100:1 ratio of reagent A : 4 % C u S 0 4 ) was added. After at least 10 min, 0.3 mL of Folin and Ciocalteu's phenol reagent (1 N) was added with vigorous vortexing. After 45 min, and within 40 min thereafter, absorbance of light at 750 nm was determined and protein concentration estimated using a BSA standard curve. Protein in the soluble and particulate fractions of sample used in P K G assays and in sample applied to MonoQ columns was assayed using a commercially available assay (Bio-Rad), based on the method of Bradford (1976). The Lowry method was not used for these preparations because of interference by components of the homogenization buffer, such as H E P E S (Peterson, 1979). Soluble fractions (5 uL) or particulate fractions (30 uL) were diluted to 1 mL with distilled water and mixed with 3 mL diluted Bio-Rad dye (Coomassie brilliant blue solution). Absorbance was read at 595 nm for comparison with a BSA standard curve. 54 2.11 Statistical analysis. Results are expressed as means ± S E M . Effects of carbachol, S N P , or A N P on the contractility of rat right ventricular strips or intact heart, in the presence of isoproterenol, were compared to isoproterenol alone at the same time point using 1-way A N O V A followed by Bonferroni's test or the non-parametric Dunn's test, as appropriate. Effects of carbachol, S N P or ANP, in the absence of isoproterenol, on the contractility of rat right ventricular strips were compared to controls (untreated) at the same time point using 1-way ANOVA followed by Bonferroni's test. Contractility of isolated ventricular cardiomyocytes was evaluated by the one-way repeated measures A N O V A followed by Student-Newman-Keuls (SNK) multiple comparison procedure if the data demonstrated normal distributions and equal variances. Otherwise, the nonparametric test, the Friedman repeated measures analysis of variance on ranks, was used. In one instance (table 2, isoproterenol 1 nM plus carbachol 10 pM, rate of relaxation), Duncan's multiple range test was used because a significant difference between groups was detected in the ANOVA but the SNK procedure failed to identify specific differences between groups. Cyclic nucleotide levels, P K G activity ratios or absolute P K G activity levels in rat ventricle treated with isoproterenol plus carbachol, SNP , or A N P were compared to ratios or levels in the presence of isoproterenol alone using 1-way A N O V A followed by Bonferroni's or Dunn's test. Differences in cyclic nucleotide levels or P K G activity in isolated cardiomyocytes were statistically tested using the paired Student's f test. Results were considered significant when P < 0.05. SigmaStat for Windows Version 1.0 (Jandel Scientific, San Mateo, CA, U.S.A.) and Statisica (Statsoft, Inc., 55 Tulsa, OK, U.S.A.) software were used for statistical analyses. N values in contractility studies indicate number of hearts, ventricular strips, or single cells, and in cyclic nucleotide and protein kinase studies indicate number of hearts, ventricular strips, or cell suspensions. 56 3.0 RESULTS 3.1 Contractility of rat ventricular preparations. 3.1.1 Effects of carbachol, SNP and ANP on contractility of Langendorff-perfused rat heart in the presence of isoproterenol. A major objective of this thesis was to repeat the studies by Lincoln and Keely (1980, 1981) which formed the basis for their proposal that P K G can be selectively activated in the ventricle by negative inotropic agents. In those studies, the paced, Langendorff-perfused rat heart was the model used to evaluate the contractile effects of acetylcholine and SNP. Accordingly, the paced, Langendorff-perfused intact rat heart was used in the present study to examine the effects of carbachol, S N P or A N P on ventricular contractility. In the present study, rat hearts were stimulated with isoproterenol during exposure to the aforementioned agents because muscarinic receptor agonists produce a negative inotropic effect in mammalian ventricles primarily in the presence of cAMP-elevating agents (see Loffelholz and Pappano, 1985 for review). The intact rat heart was also a convenient preparation to use in the present study because it yields a much larger mass of tissue per sample, in comparison to ventricular strips or isolated cardiomyocytes, thus providing a larger source of enzyme with which to refine the conditions of the P K G assay. As shown in figure 2, isoproterenol caused a concentration-dependent increase in maximal LVP and positive or negative dP/dt when serially applied in concentration-response experiments. The concentration approximating the E C 6 0 (1 nM) was used for further experiments. Increases in contractility by isoproterenol 57 Figure 2. Effect of isoproterenol on contractility of Langendorff-perfused rat heart. Effect of serial applications of a range of concentrations of isoproterenol on maximal rates of left ventricular relaxation (A) and contraction (B) and left ventricular pressure (C) were expressed as percent change in contractility relative to that prior to each drug application. Prior to drug exposure, values for maximal LVP, negative dP/dt, and positive dP/dt were 61.6 ± 6.4 mm Hg, -721.9 ± 78.2 mm Hg/s, and 1465.7 ± 255.3 mm Hg/s, respectively, n = 7. 58 isoproterenol (nM) 59 tended to reach a maximum within 2 min and decline thereafter. Given the transient nature of the isoproterenol effect, the experimental protocol for evaluation of contractile effects of carbachol, SNP , and A N P on isoproterenol-stimulated hearts was as follows: hearts were perfused with 1 nM isoproterenol for at least 2 min and and then subjected to wash out (perfusion 1). After contractility returned to baseline, the hearts were perfused with A.) isoproterenol alone for 2 min or B.) isoproterenol in combination with carbachol or S N P for 2 min or C.) A N P for 7 min with the last 2 min including isoproterenol. The latter drug treatments are referred to as perfusion 2. In each heart, the effect of isoproterenol on contractility after perfusion 2 was calculated as a percent of the effect of isoproterenol after perfusion 1. Similarly, the effect of isoproterenol plus carbachol, SNP, or A N P on contractility during perfusion 2 was calculated as a percentage of the effect of isoproterenol alone during perfusion 1 in the same heart. Changes in the isoproterenol-mediated positive inotropic effect due to carbachol, SNP , or A N P were tested by comparing the relative effect of isoproterenol in perfusion 2 to the relative effect of isoproterenol plus carbachol, S N P , or A N P in perfusion 2. Given the limited literature on the temporal profile of ANP-mediated c G M P elevation in the heart (Cramb et al., 1987; Neyes and Vetter, 1989), the protocol for pre-perfusion of hearts with A N P was used in an effort to allow a reasonable time for significant accumulation of c G M P prior to isoproterenol exposure. The average contractile parameters of Langendorff-perfused hearts prior to exposure to drugs were as follows: maximal LVP = 101.3 ± 4.1 mm Hg, +dP/dt = 60 2243 ± 124 mm Hg/s, -dP/dt = -1206 ± 56 mm Hg/s (n=41). After the first exposure to 1 nM isoproterenol for 2 min (perfusion 1), contractility was as follows: maximal LVP = 137.5 ± 5.2 mm Hg, +dP/dt = 3378 ± 172 mm Hg/s, -dP/dt = -2150 ± 81 mm Hg/s (n=41). As seen in table 1 and as illustrated in figure 3, contractility after the second exposure to isoproterenol alone was generally less than after the first exposure. This decrease was greatly magnified if the second exposure to isoproterenol was co-incident with perfusion with 10 uM carbachol. Within 2 min, carbachol inhibited isoproterenol-stimulated maximal LVP and positive or negative dP/dt to values less than those seen prior to exposure to drugs. In contrast, changes in maximum LVP and positive or negative dP/dt in the presence of isoproterenol plus 100 nM A N P or 10 uM S N P after 2 min were not significantly different from isoproterenol alone. These results demonstrate that 10 uM carbachol elicts a marked negative inotropic effect in the p-adrenoceptor-stimulated ventricle but 10 uM S N P or 100 nm A N P do not affect contractility under similar conditions. 3.1.2 Effects of carbachol, SNP and ANP on contractility of rat right ventricular strips in the presence of isoproterenol. Small negative inotropic effects of S N P or A N P have been reported in some ventricular preparations, such as ferret ventricular strips (Smith et al., 1991). To test whether or not small negative inotropic effects by S N P or A N P were occuring in the intact rat heart study described above but were not detected due to sensitivity limitations, the effects of carbachol, S N P or A N P on the contractility of electrically-stimulated rat right ventricular stripswere examined. This preparation exhibits very stable basal and isoproterenol-stimulated contractility, in terms of amplitude of twitch 61 Table 1. Effects of carbachol, SNP and ANP on contractility of rat Langendorff-perfused hearts in presence of isoproterenol. Agent Control 1 -dP/dt (-mm Hg/s) ISOInM Control 2 ISO 1nM ± Agent 1165 ±108 2072 ±164 1263 ±114 1935 ±172 CARB10 uM SNP10 uM ANP 100 nM 1148 ±107 1954 ±143 1223 ±72 2434 ±161 1485 ±198 2011 ±215 1232 ±113 1365 ±93 665 ± 83 2287 ±176 1423 ±205 1892 ±292 Agent Control 1 +dP/dt (mm Hg/s) ISOInM Control 2 ISO 1nM ± Agent 2140 ±222 3430 ± 291 2226 ± 238 3182 ±296 CARB10 uM SNP 10 |JM ANP 100 nM 2137 ±253 3183 ±314 2341 ±211 3931 ±324 2603 ±366 2998 ±408 2250 ±294 1595 ±204 2544 ± 281 3621 ± 292 2446 ±394 2814 ±469 Agent Maximal LVP (mm Hg) Control 1 ISOInM Control 2 ISOInM ± Agent CARB 10 uM SNP10 uM ANP 100 nM 89 ±9 101 ±8 106 ±6 108 ±49 124 ± 13 133 ±10 153 ±8 119 ± 53 94 ±8 100 ±9 104 ±7 95 ±43 123 ±10 77 ±7 131 ±8 105 ±47 Perfused hearts (Control 1) were treated with isoproterenol for 2 min (ISO 1nM), washed out (Control 2) and re-perfused for 2 min with isoproterenol (n=15) or isoproterenol plus carbachol 10 uM (CARB 10 uM) (n=10), isoproterenol plus SNP 10 uM (SNP 10 uM) (n=11), or isoproterenol plus ANP 100 nM (ANP 100 nM) (n=5). Values are means ± SEM. Asterisk (*) indicates significantly different from effect of second exposure to isoproterenol alone, 1-way ANOVA, P<0.05. 62 Figure 3. Effect of carbachol, A N P or S N P on Langendorff-perfused rat ventricular contractility in the presence of isoproterenol. Rat hearts were perfused with 1 nM isoproterenol for 3 min, washed, and re-challenged with isoproterenol for 2 min (ISO) (n=15), isoproterenol plus 10 uM carbachol for 2 min (ISO + CARB) (n=10), 100 nM A N P for 7 min of which the final 2 min included isoproterenol (ISO + ANP) (n=5), or isoproterenol plus 10 uM S N P for 2 min (ISO + SNP) (n=11). Effect of treatment was expressed as % decrease in negative dP/dt as compared to negative dP/dt after first exposure to isoproterenol (at 2 min). Average negative dP/dt values prior to drug exposure are described in the text. Asterisks (*) indicate significant difference from isoproterenol alone, 1-way ANOVA, Dunn's test, P < 0.05. 63 % change from first ISO exposure ^ a> en oo ro o o o o o o o o I I I I I I I I tension, so it was reasoned that small mechanical effects would be detected, if present. Based on a cumulative concentration-response curve generated for the effects of isoproterenol on ventricular strip twitch tension amplitude, an E C 6 0 of 30 nM was estimated for isoproterenol (figure 4). This concentration of isoproterenol was used to increase the baseline contractility of ventricular strips prior to the addition of carbachol, SNP , or A N P and, in those experiments, isoproterenol increased baseline twitch tension from 1.98 ± 0.18 g to 2.81 ± 0.25 g. (n=23). A significant inhibitory effect of 10 uM carbachol was apparent within 1 min and was maximal within 3 min. At 3 min, tension had fallen to 8.0 ± 0 .03% (n=6) above baseline, from 47 ± 0.14% above baseline in the presence of isoproterenol alone (n=7). Neither A N P (100 nM) nor S N P (100 pM) changed the isoproterenol-stimulated amplitude of twitch tension of rat ventricular strips during 6 min of treatment (figure 5). These results complement the findings of a B-adrenoceptor-dependent negative inotropic effect of carbachol and the absence of such an effect by S N P and A N P in the perfused rat heart (figure 3). 3.1.3 Direct effects of carbachol, SNP and ANP on contractility of rat right ventricular strips. While muscarinic receptor agonists produce a negative inotropic effect in the ventricle primarily in the presence of cAMP-elevating agents, it was of interest to test whether or not S N P or A N P had negative inotropic effects in the absence of c A M P -elevating agents. To examine this question, the maximal twitch tension development of electrically-stimulated rat right ventricular strips (in the absence of 65 Figure 4. Twitch tension amplitude of rat ventricular strips in presence of isoproterenol. Rat right ventricular strips were incubated with cumulative concentrations of isoproterenol (•). Amplitude of twitch tension was measured at each concentration after steady state was achieved and was expressed as percent of maximal twitch tension, n=3. 66 [isoproterenol] (log M) 67 Figure 5. Effect of carbachol, ANP and SNP on rat right ventricular strip contractility in the presence of isoproterenol. After approximately 2.5 min incubation with 30 nM isoproterenol, rat right ventricular strips were exposed to 10 uM carbachol (•) (n=6), 100 nM A N P (T) (n=4), or 100 uM S N P (A) (n=6) for 6 min. Twitch amplitude was compared to strips treated with 30 nM isoproterenol alone (O) (n=7) at the same time point. Asterisks (*) indicate significant differences from isoproterenol alone at same time point, 1-way ANOVA, Bonferroni adjustment, P < 0.05. 6 8 % c h a n g e i n m a x i m a l t e n s i o n isoproterenol) was measured in the presence of 100 pM carbachol, 100 pM S N P or 100 nm A N P . As shown in figure 6, no changes in baseline twitch tension were observed in the presence of 100 pM S N P or 100 nM A N P over the course of 9 min. Carbachol (100 pM) caused a small positive inotropic effect between 1.5 and 7.5 min of incubation. High concentrations of agonists were used in this experiment as a means to challenge the tissue with concentrations which were likely to markedly increase c G M P levels (Cramb et al., 1987). It was considered that negative inotropic changes may only be apparent at lower concentrations of agonists, particularly in the case of S N P and ANP. The concentration-response relationships were, therefore, evaluated at a lower range of concentrations. The twitch tension of strips in the presence of cumulative concentrations of carbachol (0.1 - 10 pM), S N P (0.1 - 10 pM), and A N P (1 - 100 nM) was not significantly different from the contractility of control strips assessed at the same point in time (figure 7). The slopes of the lines reflect the gradual decline in baseline contractility over time from an average initial twitch tension of 1.40 ± 0.08 g, independent of exposure to drugs. These results demonstrate a lack of direct negative inotropic effects of carbachol, S N P and A N P on a temporal and concentration-dependent basis in the rat right ventricle. 3.1.4 Baseline contractile function and effect of isoproterenol on contractility of single, isolated cardiomyocytes. The drug treatment protocol is diagrammatically presented in figure 8 and is described in Methods, section 2.3.6. 70 Figure 6. Temporal effects of carbachol, A N P and S N P on amplitude of rat right ventricular strip twitch tension. Rat right ventricular strips were exposed to 100 uM carbachol (•) (n=3), 100 nM A N P (T) (n=3), or 100 uM S N P (A) (n=7) for 9 min. Twitch amplitude was expressed as percentage of tension prior to exposure to drugs. Average twitch tension was 1.8 ± 0.2 g (n=17). Asterisks (*) indicate significant difference from control (O) (n=4) at same time point, 1-way ANOVA, Bonferroni adjustment, P < 0.05. 71 72 Figure 7. Lack of effects of carbachol, ANP and SNP on amplitude of rat right ventricular strip twitch tension. Rat right ventricular strips were exposed to 0.1 - 10 uM carbachol (•) (n=3), 1 - 100 nM A N P (T) (n=3), or 0.1 - 10 uM S N P (A) (n=4). Drugs were added to the baths every 4 - 5 min in increasing concentrations. Accordingly, tension was measured at 5, 10, and 15 min after the initiation of the drug treatments. Twitch amplitude was expressed as percentage of tension prior to exposure to drugs. Control tension was 96.3 ± 0.8% at 5 min, 94.3 ± 2 . 1 % at 10 min and 96.2 ± 3.4% at 15 min (n=4). None of the points was significantly different from the contractility of control strips at the same point in time. Average initial baseline twitch tension was 1.40 ± 0.08 g. 73 c 80 -o •</> c 70 -a> 60 -o • •• | 50 -40 -"E • • 30 -o 20 -10 -0 -0.001 0.01 0.1 1 10 concentration of agonist (|jM) 74 Figure 8. Protocol for drug treatment of single rat ventricular cardiomyocytes. See text in Methods, section 2.3.6 for details. Upward deflections indicate cell shortening. 75 90 95 -100 -CONTROL ISO 1 nM ISO + ISO 1 nM CARB 10uM O) 105 -C O — 110 -I <D O 115 120 4 sec 76 Rat ventricular cardiomyocytes which fulfilled minimum requirements for appearance and contractility were examined for baseline parameters (tables 2 and 3). These cells exhibited contractile activity, in terms of time to reach peak contraction and time to relax, which was similar to that of intact ventricular tissue (Mclvor et al., 1988), and a larger extent of shortening and shorter times to peak tension and relaxation than reported in other isolated cardiomyocyte preparations (Xiao et al., 1993; Shah et al., 1994), indicating that the isolation procedure used in this study allowed for preparation of cardiomyocytes which retained relatively normal myocardial contractile characteristics. The effect of a B-adrenoceptor agonist on contractility of the rat cardiomyocytes was examined using a low concentration of isoproterenol (1 nM). All parameters examined were significantly different from control values at 2 min when considered collectively from tables 2 and 3. This concentration was used in all further experiments. An important effect of isoproterenol is to increase the magnitude of shortening during the contraction. This can be seen in figure 9 (trace B) which is representative of the average values shown in tables 2 and 3. A concomitant increase in the rates of change in cell length also occurs as reflected in the greater slopes of traces during contraction and relaxation. The increases in time to peak contraction and time to 7 5 % relaxation are relatively small in comparison to the markedly enhanced increase in cell contraction and relaxation rates induced by 1 nM isoproterenol (tables 2 and 3). The net effect is one of increased magnitude of cell shortening at greater rates of change in length. 77 Table 2. Effect of isoproterenol and carbachol on rat ventricular cardiomyoctye contractility. Initial ISO (1nM) ISO(1nM) +CARB(1uM) Repeat ISO(1nM) Control Drug (% increase) Control Drugs (% increase) Control Drug (% increase) shortening (% diastolic length) 9.9 ±0.5 57.0 ± 11.7 * 8.1 ±0.7 39.3 ± 9.5 * ± 8.1 ±0.8 58.4 ± 9.4 * time to peak shortening (ms) 154 ±8 9±4 * 143 ±9 16 ± 8 * 144 ±9 11 ±3 * time to 75% relaxation (ms) 277 ± 12 2 + 3 261 ± 13 6 ±4 267 ± 14 0±4 rate of contraction (um/s) 141 ± 13 66 ±18 * 115 ±13 35 ±10 * ± 115 ± 14 57 ±11 * rate of relaxation (|jm/s) 121 ± 11 66 ±14 * 99 ±14 52 ±15 * ± 98 ± 16 94 ±24 * Initial ISO(1nM) ISO (1nM) + CARB (10uM) Repeat ISO (1nM) Control Drug (% increase) Control Drugs (% increase) Control Drug (% increase) shortening (% diastolic length) 10.6 ± 1.6 79.3 + 30.6 * 10.2 ± 1.6 19.5 ±6.3 ** 10.4 ± 1.4 65.6 ± 20.9 * time to peak shortening (ms) 166 ±7 13+4 * 161 ±6 12 ± 1 173 ±6 9±3 * time to 75% relaxation (ms) 309 ± 14 2±3 289 ±13 8±3 * 304 ± 12 4±5 rate of contraction (um/s) 153 ±30 80 ± 30 * 145 ± 30 14±6 ± 148 ±29 66 ±23 * rate of relaxation (|jm/s) 135 ±33 100 ±48 * 129 ±31 24 ± 10 t 135 ±32 61 ±20 * Drugs treatments were for 2 min. Values are means ± SEM, no. of cells = 7-9. ISO, isoproterenol, CARB, carbachol. * indicates significant difference from corresponding control value, p < 0.05, Wilcoxon signed rank test or paired Student's t-test. t indicates significant difference from first Iso 1nM treatment value, p < 0.05, ANOVA repeated measures test. +. indicates significant difference from first and second Iso 1 nM values, p < 0.05, Friedman repeated measures on ranks or ANOVA repeated measures tests. Table 3. Effect of isoproterenol and SNP on rat ventricular cardiomyocyte contractility. Inital ISO (1nM) ISO (1nM)+ SNP(10uM) Repeat ISO (1nM) Control Drug (% increase) Control Drugs (% increase) Control Drug (% increase) 2 min 6 min 2 min 6 min 2 min 6 min shortening (% diastolic length) 13.7 ±0.9 34.5 ±7 .3 * 53.3± 18.5 * 11.7 ± 0.5 40.7 ±4.5 * 64.6 ±8.5 * 11.8 ± 0.9 38.9 ±4 .6 * 69.3 ±7.6 * time to peak shortening (ms) 156 ± 6 9 ± 4 * 11 ± 4 * 144 ± 8 1 0 ± 4 * 1 7 ± 6 * 136 ± 6 11 ± 2 * 20 ± 6 * time to 75% relaxation (ms) 257 ± 8 13 ± 3 * 1 0 ± 2 * 234 ± 8 11 ± 2 * 1 8 ± 4 * $ 234 ± 8 5 ± 2 * 7 ± 4 rate of contraction (um/s) 187 ± 15 42 ± 7 * 55 ± 9 * 178 ± 1 8 44 ± 9 * 60 ± 8 * 173 ±21 42 ± 7 * 79 ± 10 * rate of relaxation (um/s) 172 ± 18 42 ± 9 * 46 ± 12 * 162 ± 18 46 ± 12 * 61 ± 1 2 * 163 ± 2 3 48 ± 12 * 82 ±21 * Inital ISO (1nM) ISO (1nM) + SNP100uM Repeat ISO (1nM) Control Drug (% increase) Control Drugs (% increase) Control Drug (% increase) 2 min 6 min 2 min 6 min 2 min 6 min shortening (% diastolic length) 10.2 ±1.0 67.1 ±11.1 * 86.7 ±20.1 * 9.3 ±0.9 56.2 ±10.1 84.0 ±19.9 * 8.9 ±0.8 73.4 ±15.6 100 ±26.4 * time to peak shortening (ms) 169 ± 7 8 ± 3 8 ± 4 144 ± 7 1 9 ± 4 * 24 ± 6 * 139 ± 7 23 ± 5 * 25 ± 7 * time to 75% relaxation (ms) 275 ± 11 8 ± 4 11 ± 4 * 259 ± 7 7 ± 2 * 11 ± 5 * 253 ± 1 2 9 ± 7 1 7 ± 6 * rate of contraction (um/s) 137 ± 1 5 83 ± 18 * 111 ± 2 7 * 137 ± 1 3 66 ±19 * 81 ±19 * 133 ± 1 5 75 ± 16 * 100 ± 3 5 * rate of relaxation (um/s) 122 ± 19 98 ± 2 8 * 122 ± 2 4 * 113 ± 16 93 ±21 * 100 ±28 * 119 ± 12 76 ± 1 2 * 96 ± 2 4 * Values are means ± SEM, no. of cells = 6-9. ISO, isoproterenol. * indicates significant difference from corresponding control value, p < 0.05, Wilcoxon signed rank test or paired Student's t-test.; $ indicates significant difference from corresponding ISO 1 nM values, p < 0.05, Friedman repeated measures on ranks or ANOVA repeated measures tests. Figure 9. Effect of isoproterenol and carbachol on the contraction of a single, rat ventricular cardiomyocyte. Shortening of a single cell is shown under control conditions (A), after 2 min exposure to 1 nM isoproterenol (B) and after simultaneous exposure to 1 uM carbachol and 1 nM isoproterenol for 2 min (C). Control values: shortening, 9 .9% of diastolic length; time to peak contraction, 190 ms; time to 7 5 % relaxation, 340 ms; rate of contraction, 108 um/s; rate of relaxation, 84 um/s. ISO values: shortening, 14.3% of diastolic length; time to peak contraction, 190 ms; time to 7 5 % relaxation, 310 ms; rate of contraction, 159 um/s; rate of relaxation, 149 um/s. Carbachol and ISO values: shortening, 11.3%; time to peak contraction, 190 ms; time to 7 5 % relaxation, 310 ms; rate of contraction, 118 um/s; rate of relaxation, 119 um/s. 80 81 3.1.5 Effect of muscarinic receptor agonists and SNP in presence of isoproterenol on the contractility of single, isolated cardiomyocytes. The effects of carbachol and acetylcholine on contractile function were assessed in the presence of isoproterenol. Carbachol (1 and 10 uM) antagonized the positive inotropic action of 1 nM isoproterenol at 2 min (table 2) and significant negative inotropy by 1 uM carbachol could be detected at 1 min. At 1 min, 1 uM carbachol prevented the elevation in cell shortening induced by isoproterenol (cell shortening changes from paired controls: initial 1 nM isoproterenol treatment, 22.0 ± 7.3%; 1 uM carbachol and isoproterenol, 6.4 ± 5.5%; repeat isoproterenol treatment, 16.1 ± 3 .0%; n=11). Carbachol (10 uM) did not significantly reduce isoproterenol-stimulated cell shortening at 1 min using repeated measures analysis of variance (P= 0.12). Cell shortening and rates of contraction and relaxation were depressed at 2 min by both concentrations of carbachol. This is diagrammatically shown in figure 9 where the magnitude of the peak contraction and slopes of contraction and relaxation are lower in the presence of both carbachol and isoproterenol (trace C) in comparison to isoproterenol alone (trace B). Carbachol (10 uM) reduced isoproterenol-stimulated contractility at 2 min to an extent that the rates of contraction and relaxation were not significantly different from control. Neither the time to peak shortening nor the time to 7 5 % relaxation was altered by the combination of carbachol and isoproterenol in comparison to isoproterenol alone. 82 Acetylcholine (1 and 10 uM) had a similar effect on isoproterenol-stimulated cell shortening (figure 10). S N P did not alter isoproterenol-induced changes in ventricular contractility (table 3, figure 11). Using the same drug treatment protocol as illustrated in figure 8, except using S N P instead of a muscarinic receptor agonist, S N P (10 and 100 uM for 2 min) had no significant effect on 1 nM isoproterenol-stimulated cell shortening, rates of contraction and relaxation, or on times to contract and relax. An absence of an effect by S N P was also apparent at 6 min, except for a slightly prolonged time to 7 5 % relaxation with 10 uM SNP and isoproterenol in comparison to isoproterenol alone. This effect was not seen at a 10-fold higher concentration of S N P . 3.2 Cyclic nucleotide levels in rat cardiac preparations. 3.2.1 Effects of carbachol, ANP or SNP in the presence of isoproterenol on cGMP levels in ventricle of rat Langendorff-perfused hearts. Cyclic G M P levels were measured in ventricular tissue from agonist-treated, Langendorff-perfused hearts which had been frozen after assessment of contractility (Results, section 3.1.1). Carbachol (10 uM), in the presence of isoproterenol, did not significantly increase total tissue c G M P levels after 2 min when compared to levels in the presence of isoproterenol alone (figure 12). Conversely, both 100 nM A N P and 10 uM SNP, in the presence of isoproterenol, significantly increased c G M P levels after a 2 min perfusion by approximately 3 fold above levels in the presence of isoproterenol alone (figure 12). Isoproterenol alone (1 nM) did not significantly increase c G M P levels above those found in untreated tissue (untreated = 290 ± 30 fmol cGMP/mg protein, n = 5; isoproterenol = 403 ± 47 fmol cGMP/mg protein, n = 83 Figure 10. Effect of isoproterenol on cell shortening of isolated rat ventricular cardiomyocytes in the presence and absence of acetylcholine. Cells were exposed to isoproterenol for 2 min (solid bars) and the change from control in maximal amplitude of shortening is reported. After washout (as described in Materials and Methods) cells were exposed to isoproterenol plus acetylcholine (1 or 10 uM) for 2 min (open bars). Finally, cells were washed and again exposed to isoproterenol alone for 2 min (gray bars). Asterisks (*) indicate significant difference from isoproterenol alone, repeated measures ANOVA, P<0.05. ISO, isoproterenol, ACh , acetylcholine. n= 5-6. 84 % i n c r e a s e i n c e l l s h o r t e n i n g Figure 11. Effect of isoproterenol in the presence and absence of SNP on the contraction of a single, rat ventricular cardiomyocyte. Shortening of a single cell is shown under control conditions (A), after 2 min exposure to 1 nM isoproterenol (B) and after 2 min simultaneous exposure to 100 pM S N P and 1 nM isoproterenol (C). Control values: shortening, 11.6% of diastolic length; time to peak contraction, 180 ms; time to 7 5 % relaxation, 310 ms; rate of contraction, 165 pm/s; rate of relaxation, 157 pm/s. ISO values: shortening, 22 .9%; time to peak contraction, 220 ms; time to 7 5 % relaxation, 365 ms; rate of contraction, 402 pm/s; rate of relaxation, 281 pm/s. SNP and ISO values: shortening, 2 3 . 5 % of diastolic length; time to peak contraction, 220 ms; time to 7 5 % relaxation, 345 ms; rate of contraction, 434 pm/s; rate of relaxation, 310 pm/s. 86 87 Figure 12. Effects of carbachol, ANP or SNP in the presence of isoproterenol on cGMP levels in ventricle of Langendorff-perfused hearts. c G M P was measured in rat hearts which were perfused with 1 nM isoproterenol for 3 min, washed, and re-challenged with isoproterenol for 2 min (open bar), isoproterenol + 10 uM carbachol for 2 min (diagonal bar), 100 nM A N P for 7 min of which the final 2 min included isoproterenol (gray bar), or isoproterenol + 10 uM S N P for 2 min (solid bar). Homogenates of ventricular tissue were treated with TCA and the soluble fractions assayed for cGMP. Numbers in bars indicate sample numbers. Asterisks (*) indicate significant differences from isoproterenol alone, 1-way ANOVA, Dunn's method, P < 0.05. 88 89 10). Based on these findings in total tissue preparations, c G M P elevation does not correlate with the mediation of a negative inotropic effect by carbachol, S N P or A N P in the ventricle. This dissociation was also noted in rat ventricular strips in which 100 pM S N P did not alter twitch tension (figure 6) but increased tissue c G M P levels to 2.5-fold above control after 9 min (control ventricular strip c G M P content = 106.3 ± 26.3, 100 pM SNP-treated strip c G M P content = 243.4 ± 9.7, n = 4 - 5). 3.2.2 Effects of isoproterenol, carbachol, ANP and SNP on cAMP levels in ventricle of Langendorff-perfused heart. Ventricular tissue in which c G M P levels had been measured (see section 3.2.1) was also assessed for total tissue cAMP content. After perfusion of Langendorff-mode rat hearts with 1 nM isoproterenol for 2 min, total ventricular tissue c A M P levels were significantly increased above control (control = 12.0 ± 2.6 pmol cAMP/mg protein, n = 5; isoproterenol = 18.8 ± 0.7 pmol cAMP/mg protein, n = 10). As shown in figure 13, neither 10 pM carbachol, 100 nM A N P nor 10 pM S N P altered the ability of isoproterenol to increase cAMP levels in the rat ventricle. 3.2.3 Effect of muscarinic receptor agonists and SNP in presence of isoproterenol on cGMP content of isolated ventricular cardiomyocytes. Unlike the effect of carbachol in intact cardiac preparations, muscarinic receptor agonists were able to significantly increase c G M P levels in suspensions of cardiomyocytes in the presence of isoproterenol. After 2 min incubation, acetylcholine (10 pM, pre-incubated with physostigmine 50 pM for 1 min) significantly increased c G M P levels over paired control levels (figure 14). Similarly, 90 Figure 13. Effects of carbachol, ANP or SNP in the presence of isoproterenol on cAMP levels in ventricle of Langendorff-perfused hearts. c A M P was measured in the same samples as shown in figure 12. Homogenates of ventricular tissues were treated with TCA and the soluble fractions assayed for cAMP. Numbers in bars indicate sample numbers. No differences from isoproterenol alone were found, 1-way ANOVA, Dunn's method, P < 0.05. 91 25 - , 92 Figure 14. Effect of muscarinic receptor agonists and SNP in the presence of isoproterenol on cGMP levels in rat ventricular cardiomyocytes. Suspensions of cardiomyocytes were exposed to isoproterenol (1 nM) plus carbachol (1 or 10 pM), acetylcholine (10 pM), or S N P (10 or 100 pM) for 2 min or plus S N P (100 pM) for 6 min. c G M P content of the treated cells (open bar) was then measured as described in Methods and compared to paired controls (closed bar). Physostigmine (50 pM) was present in cell suspensions for 1 min prior to introduction of acetylcholine and isoproterenol. Asterisks (*) indicate significant differences from control, paired Student's f-test, P < 0.05. n = 6 - 12. C A R B , carbachol; ACh , acetylcholine. 93 600 • 5 500 -O ° - 400 -E CARB CARB ACh SNP SNP SNP 1 uM 10 MM 10 MM 10 U M 100 LIM 100 U M 2 min 2 min 2 min 2 min 2 min 6 min 9 4 I uM and 10 uM carbachol significantly increased c G M P levels. Isoproterenol (1 nM) was added to the cell suspensions at the same time as the muscarinic receptor agonist. These results demonstrate that muscarinic receptor agonists have c G M P -elevating activity in cardiomyocytes stimulated with p-adrenergic agonists. S N P (10 pM), in the presence of 1 nM isoproterenol, significantly increased c G M P after 2 min and 100 pM S N P increased c G M P after 2 min and 6 min in comparison to paired controls (figure 14). Thus, S N P can elevate c G M P in cardiomyocytes to at least 8-fold over control levels at concentrations which had no effect on isoproterenol-induced increases in contractility. Isoproterenol alone had no effect on cellular c G M P content after 2 or 6 min (control c G M P for 2 min = 90 ± II fmol/mg protein, isoproterenol 1 nM for 2 min = 85 ± 14 fmol/mg protein; control for 6 min = 60.8 ± 4.3 fmol/mg protein, isoproterenol 1 nM for 6 min = 55.6 ± 4.2 fmol/mg protein). 3.2.4 Effect of muscarinic receptor agonists and SNP in presence of isoproterenol on cAMP content of isolated ventricular cardiomyocytes. Carbachol (1 and 10 pM) and S N P (100 pM) were examined as to their effect on c A M P content in cardiomyocytes in the presence of isoproterenol (1 nM). As shown in figure 15, 1 nM isoproterenol significantly increased cAMP levels in suspensions of cardiomyocytes after 2 and 6 min. In agreement with results seen in the intact ventricle, neither carbachol nor S N P altered the ability of isoproterenol to increase cAMP under these conditions. 95 Figure 15. Effect of muscarinic receptor agonists and SNP in the presence of isoproterenol on cAMP levels in rat ventricular cardiomyocytes. Suspens ions of card iomyocytes were exposed to 1 nM isoproterenol in the p resence and a b s e n c e of carbachol (1 or 10 pM) or S N P (100 pM) for 2 min or the p resence and absence S N P (100 pM) for 6 min. c A M P content of the cel ls w a s then measured as descr ibed in Materials and Methods. E a c h treatment (open bar) is significantly different (*) from paired control (closed bar), paired Student 's f-test, P < 0.05. ISO, isoproterenol; C A R B , carbachol . n = 4 - 8. 96 97 3.3 cGMP-dependent protein kinase in the rat ventricle. 3.3.1 Characterization of the assay of cGMP-dependent protein kinase. Prior to the measurement of P K G activity in agonist-treated rat ventricle, attempts were made to refine the condit ions of the assay . The a s s a y method w a s based on that of J iang et al. (1992) which descr ibed the measurement of P K G activation by cGMP-e leva t i ng agents in the intact porcine coronary artery. The treated t issue is homogen ized rapidly (within 30 s) with a minimum of dilution with homogenizat ion buffer (5 vo lumes) . Centrifugation of the homogenate is brief; 15 min at 30,000a; w a s used by J iang et al. (1992) which we abbreviated even further to 5 min at 30,000g. The incubation of the soluble fraction with assay reagents is performed at 4°C. T h e s e condit ions are used to address an important technical considerat ion, namely the requirement for the inhibition of dissociat ion of c G M P from its binding sites on P K G and thus maintenance of the activation state of the enzyme during the assay . Other condit ions of the assay were examined as fol lows. 3.3.1.1 Activation of cGMP-dependent protein kinase by cGMP in vitro. The speci f ic activity ratio is the parameter used to quantify the activation state of P K G . It is the ratio of P K G - d e p e n d e n t in vitro phosphorylat ion of substrate in the a b s e n c e of added c G M P to PKG-dependen t in vitro phosphorylat ion of substrate in the p resence of a maximally-activating concentrat ion of c G M P . Thus , the in vitro activation of P K G by c G M P was a s s e s s e d so as to ensure that the concentrat ion of c G M P used in the assay (5 uM) was sufficient to maximally activate the enzyme . A s shown in figure 16, soluble P K G activity was maximal at approximately 100 nM 98 Figure 16. In vitro activation of cGMP-dependent protein kinase by cGMP. The soluble fraction of control rat ventricular t issue w a s assayed for 4 min at 4°C for P K G activity in the presence of c G M P (0.001 - 100 pM). P K G activity was exp ressed as pmol P 0 4 incorporated into B P D E t i d e per min per mg soluble fraction protein. 99 PKG activity (pmol P04/min/mg protein) o ro w 01 o> I I I I I I I c G M P , one-fiftieth of the concentration used in the assay. Identical results were found when particulate P K G was assayed (data not shown). 3.3.1.2 Phosphorylation of BPDEtide by cGMP-dependent protein kinase. A major improvement in the assay for P K G was the design of a PKG-specif ic peptide substrate, known as BPDEtide, which is described as the most specific substrate for P K G now available (Colbran et al., 1992). Since published reports of the kinetics of phosphorylation of this peptide by P K G utilized purified kinase and this research project primarily involves assessment of P K G activity in crude soluble and particulate fractions of cardiac tissue, the K m for BPDEtide in crude fractions was determined. The concentration-dependent phosphorylation of BPDEtide by soluble P K G is shown in figure 17. The K m for phosphorylation of BPDEtide is in the range of 100 - 200 pM for soluble and particulate P K G . BPDEtide is used in the assay at a concentration of 150 pM, and not higher concentrations, for the following reasons: very little substrate is phpsphorylated during the assay at 150 pM BPDEtide (no substrate depletion), concentrations of substrate above K m allow for non-PKG-mediated phosphorylation and therefore decrease the specificity of the assay (Eisenthal and Danson, 1993), and the peptide is expensive. 3.3.1.3 Temporal linearity of cGMP-dependent protein kinase activity. The P K G assay demonstrated linear phosphorylation of substrate with respect to time from 1 to 8 min at 4°C in the presence of 5 pM cGMP. Figure 18A shows cumulative data from 5 samples of ventricular soluble fractions. A single 101 Figure 17. Phosphorylation of peptide substrate, BPDEtide, by soluble cGMP-dependent protein kinase. Soluble fraction of rat ventricular tissue was assayed for 4 min at 4°C in the presence of various concentrations of BPDEtide as described in Materials and Methods. PKG activity was expressed as pmol P 0 4 incorporated into BPDEtide per min per mg soluble fraction protein. Data shown are from two separate experiments. 102 K M = 106 LIM 0 500 1000 1500 2000 2500 BPDEtide (\iM) 103 Figure 18. Temporal linearity of cGMP-dependent protein kinase assay. A. Soluble fraction of control rat ventricular tissue was assayed at 4°C for P K G activity with assay incubation times of 1, 2, 4, and 8 min. n = 5. B. Soluble fraction of one control rat ventricular tissue sample was assayed at 4°C for P K G activity with assay incubation times of 0, 1, 2, 4, and 8 min. P K G activity was expressed as pmol P 0 4 incorporated into BPDEtide per mg soluble fraction protein. 104 30 A incubation time (min) 105 ventricular sample was assayed at zero time as well as at 1, 2, 4 and 8 min and, as shown in figure 18B, a small amount of phosphorylation can be detected at the zero time point. This value is essentially a blank and it can be subtracted from each data point in the line. So too, the extrapolated y-intercept of the line in figure 18A can be subtracted from each data point. Both lines would, therefore, pass through the origin and the assay remains linear. 3.3.1.4 Chromatography of PKG in rat ventricle and ventricular cardiomyocytes. The identity of P K G was verified using chromatographic and immunoblotting techniques. The isoelectric point of P K G is 5.7 and the enzyme, therefore, is negatively charged at neutral pH (Takai et al., 1975). Thus, anion exchange chromatography is a useful tool to separate P K G from other cellular proteins and was the primary technique utilized in the first successful efforts to separate P K G from P K A (Kuo and Greengard, 1970). In the present study, cyclic nucleotide-dependent protein kinase in the soluble fraction of ventricular tissue (figure 19) and the soluble (figure 20A) and particulate (figure 20B) fractions from isolated cardiomyocytes were resolved using MonoQ anion exchange chromatography. In order to demonstrate P K G activity, fractions were assayed in the presence PKI, a specific peptide inhibitor of PKA, either in the presence or absence of 5 uM c G M P . cGMP-dependent activity eluted as a single distinct peak, with a small early shoulder of activity, at the same point in the elution profiles (fractions 27 - 32) of the soluble fraction of ventricular tissue and cardiomyocytes and in the particulate fraction of cardiomyocytes. Distinct peaks which correspond to different P K G isozymes (la and 106 Figure 19. MonoQ anion exchange column chromatography of cyclic nucleotide-dependent protein kinase activity in rat ventricular tissue. Fractions from MonoQ chromatography (0 - 400 mM NaCl gradient elution) of rat ventricular soluble extracts were assayed for c G M P - and cAMP-dependent kinase activity. Total amount of protein loaded onto the column was 14.5 mg soluble protein. Assay incubations were conducted at 30°C for 10 min. PKA activity was measured in the presence (-A-) and absence (-v-) of 5 pM cAMP, in the absence of PKI. P K G activity was measured in the presence (-•-) and absence (-o-) of 5 pM c G M P , in the presence of 5 pM PKI. Profiles are representative of cGMP-dependent kinase activity in 4 preparations and cAMP-dependent kinase activity in 3 preparations. 107 0 3 6 9 12 15 18 21 24 27 30 33 36 39 fraction 108 Figure 20. MonoQ anion exchange column chromatography of cyclic nucleotide-dependent protein kinase activity in rat cardiomyocytes. Fractions from MonoQ chromatography (0 - 400 mM NaCl gradient elution) of isolated rat ventricular cardiomyocytes were assayed for c G M P - and cAMP-dependent kinase activity in soluble (A) and particulate (B) extracts. Total amount of protein loaded onto the columns was 22 mg soluble protein (A) and 4.58 mg particulate protein (B). Assay incubations were conducted at 30°C for 10 min. PKA activity was measured in the presence (-A-) and absence (-v-) of 5 pM cAMP, in the absence of PKI. P K G activity was measured in the presence (-•-) and absence ( -o) of 5 pM c G M P , in the presence of 5 uM PKI. Profiles are representative of cGMP-dependent kinase activity in 1 particulate and 3 soluble preparations and cAMP-dependent kinase activity in 1 particulate and 2 soluble preparations. 109 4000 3500 -J g 3000 -I J> 2500 -] co o 2000 -I 8 1500 -I B 500 400 A c E 0 ~ 300 -I i_ o a o o c O a. o E 200 -I 100 A + c G M P + PKI O - c G M P + PKI + cAMP v - cAMP type I P K A P K G 6 9 12 15 18 21 24 27 30 33 36 39 fraction 110 ip) have been resolved by ion exchange chromatography of crude and purified tissue extracts (Lincoln et al., 1988; Wolfe et al., 1989). The presence of a single peak of P K G activity may, therefore, reflect the predominance of a single P K G isozyme in the rat ventricle. Cyclic nUcleotide-independent kinase activity was negligible under conditions used to assay P K G . P K A activity was detected in MonoQ fractions in the presence and absence of 5 pM cAMP, in the absence of PKI. Two distinct peaks of cAMP-dependent activity were seen in the MonoQ fractions (figures 19 and 20). The absence of these peaks in the presence of PKI (plus cGMP) suggests that they correspond to type I and II PKA. Furthermore, the identification of the first peak as type I P K A was corroborated by its elution at relatively low ionic strength (Lincoln, 1983). Another cAMP-dependent peak, which corresponds to the P K G peak, was seen in figures 19 and 20A and presumably represents activation of P K G by 5 pM cAMP. This peak may also have contributed to the broad shoulder on the type II P K A peak seen in the particulate fraction from the myocyte preparations (figure 20B). This peak was both c A M P - and cGMP-dependent, although c G M P was a considerably more effective activator. Cross-activation by cAMP and cGMP, greater sensitivity to c G M P as an activator, and insensitivity to PKI all serve to identify this peak as P K G . Recovery of P K A activity from MonoQ columns was 103 ± 23 .5% (n=4). The chromatographic profiles demonstrate several points: the assay conditions used to detect PKG-dependent phosphorylation were very selective for measuring P K G activity (no peaks of activity in fractions 7 - 10 or 31 - 35 in the presence of PKI, ± cGMP). Also, very little cyclic nucleotide-independent protein 111 kinase activity was detected using standard assay conditions (± c G M P , + PKI). Furthermore, the profiles show that P K G can be partially activated by c A M P but more than 5 uM cAMP is required for full activation (see small peak in fractions 26 -31 in the presence of 5 uM cAMP). 3.3.1.5 Western blotting of cGMP-dependent protein kinase. Western blots of peak P K G and PKA activities were performed using a polyclonal antibody raised against the C terminus of type la P K G (a gift of Dr. Steve Pelech, Kinetec Biotechnology Corp., Vancouver, B.C.) to confirm the identity of P K G in the MonoQ elution profiles. Selected fractions from profiles shown in figures 19 and 20, and commercially available purified P K G and P K A holoenzymes and P K A catalytic subunit were subjected to S D S - P A G E electrophoresis. The protein was transferred to a nitrocellulose membrane and probed with P K G antibody. Secondary anti-rabbit goat antibody, conjugated with alkaline phosphatase, was used to visualize immunoreactive bands. Preliminary experiments using P K G antibody against proteins in cardiac and non-cardiac tissues demonstrated that P K G was in relatively low abundance in the heart, as compared to more PKG-rich tissues such as vascular smooth muscle. To ensure visualization of P K G immunoreactivity in ventricular samples, concentrated soluble and particulate fractions were applied to MonoQ columns (the maximal amount of protein applied was 22 mg) and the primary P K G antibody was used at a low dilution (1:200). As shown in figure 21, these conditions led to very rapid development of color during the alkaline phosphatase reaction and many non-specific bands of immunoreactivity were 112 Figure 21. Immunoblotting of MonoQ-fractionated protein kinases from rat ventricular tissue and cardiomyocytes. Selected fractions from MonoQ chromatography of rat ventricular tissue and cardiomyocytes (figures 19 and 20) were resolved on 1 1 % S D S - P A G E for Western blotting with an antibody raised to the C-terminus of type la P K G as described in Materials and Methods section. Commercially obtained P K G and PKA holoenzymes and P K A catalytic subunit were included as controls. Positions of proteins of known molecular mass are indicated by their molecular mass values (kDa) on the left of the figure. 113 114 visible. The strongest immunoreactivity was seen in the presence of commercially obtained P K G at the molecular mass appropriate for authentic P K G ; M r values for type la and Ip P K G monomer are 78 and 80, respectively (Wolfe et al., 1989). Specificity of the antibody for P K G was confirmed by the absence of binding to purified P K A holoenzyme or catalytic subunits. Marked immunoreactivity was detected at a similar molecular mass as purified P K G in lanes which corresponded to P K G peaks in the MonoQ profiles (figure 19, fraction 28; figure 20A, fraction 29; figure 20B, fraction 30). Bands at a similar molecular mass as purified P K G stained less intensely in lanes corresponding to type II PKA, although considering the absence of PKG-l ike immunoreactivity in the presence of type I P K A (fraction 8 [soluble ventricle] and fraction 10 [soluble cardiomyocyte]) and commercially purified PKA, this probably reflects an overlap of P K G into the nearby type II P K A fractions in the MonoQ chromatography. Therefore, Western blotting for P K G immunoreactivity resulted in the identification of P K G protein in MonoQ fractions which showed P K G activity. This confirms the presence of P K G in the intact ventricle as well as the isolated cardiomocyte, in both the soluble and particulate fractions, and also serves as a verification of the conditions of the assay in terms of specificity for P K G . Staining in fraction 8 (particulate cardiomyocyte) at the molecular mass corresponding to P K G is most likely to be an artifact, perhaps due to spilling of commercially purified P K G into this lane during sample loading. This appears to be the case since there is no biochemical evidence for P K G activity in fraction 8 115 particulate cardiomyocyte) and the staining of the band is most intense on the portion closest to the purified P K G band. As mentioned above, the singularity of the P K G peak in MonoQ chromatography is suggestive of the presence of a single isozyme in the ventricle. Attempts were made to elucidate the isozyme profile in ventricular tissue and cardiomyocytes by immunoblotting with isozyme-specific antibodies. The antibodies, raised to the N-termini of P K G isozymes types la and IB, did not specifically recognize P K G and, therefore, were not helpful in identifying specific P K G isozymes (blots not shown). 3.3.1.6 Effect of putative kinase inhibitors on cGMP-dependent protein kinase activity in vitro. Effective, specific inhibitors are important tools in the identification of protein kinase activity. Of particular interest in this study were P K G inhibitors which could penetrate biological membranes and thus be useful in elucidating the role of P K G in the mechanical effects of carbachol, S N P and ANP. Two putative, membrane permeant P K G inhibitors were investigated as to their inhibitory effects on P K G activity in vitro, for consideration in contractility studies. KT5823 (or KT5822) has a reported K| for P K G of 2.4 nM in in vitro studies (Kase et al., 1987). However, when used at a thousand-fold greater concentration, KT5823 did not significantly inhibit soluble or particulate rat ventricular P K G activity in vitro (figures 22A and 22B). Rp-8 - C P T - c G M P S has a reported Kj for P K G of 0.5 uM (Butt et al., 1994). As seen in figures 22A and 22B, 4 to 20 fold higher concentrations had limited efficacy; particulate P K G activity was marginally decreased by 2 uM R p - 8 - C P T - c G M P S and 116 Figure 22. Effect of putative kinase inhibitors on cGMP-dependent protein kinase activity in vitro. KT5823 (2 uM) (diagonal bar), and R p - 8 - C P T - c G M P S (Rp-CPT) (2 pM, gray bar; 10 pM, solid bar) were included in the assay for total P K G activity (30°C for 4 min) in soluble (A) or particulate (B) rat ventricular extract. P K G activity in the presence of KT5823 or Rp -8 -CPT -cGMPS was compared to activity in the same extract in the absence of these agents (open bars). Asterisks (*) indicate significant difference, paired student's t-test, P<0.05, n=5-8. 117 particulate P K G activity (pmol/min/mg) s 0 , u b | e p K G a c t j v j t y ( p m o l / m i n / m g ) soluble and particulate activities decreased by 15% to 2 0 % at 10 pM. Concentrations of both inhibitors used in these experiments exceeded the reported Kj for inhibition of other kinases, such as PKA (K| for inhibition of P K A by KT5823 is 37.4 nM, by R p - 8 - C P T - c G M P S is 8.3 pM) (Kase et al., 1987; Butt ef al., 1994). It was concluded that KT5823 and Rp -8 -CPT -cGMPS are not effective P K G inhibitors and, if used at concentrations sufficient to substantially inhibit P K G , are unlikely to be specific P K G inhibitors. Based on these findings, neither agent was used in intact tissue contractility studies. 3.3.1.7 Temporal characteristics of PKG activity ratio during assay. Bound c G M P dissociates from P K G if conditions become more dilute (e.g., homogenization, assay incubation) and this process is accelerated at temperatures commonly used for enzyme assays (e.g., 30°C). If the cyclic nucleotide dissociates from P K G during the course of tissue extract preparation and handling, the assay will underestimate the activation state of the kinase in agonist-treated tissue. An aim of this project was to determine a time span in which to incubate cardiac soluble fractions with substrate, such that there was sufficient time for measurable substrate phosphorylation with a minimum dissociation of c G M P from P K G . Agonist-treated ventricular tissue was assayed at 4°C, due to the 100-fold increased affinity of P K G for bound c G M P at this temperature relative to 30°C (Francis and Corbin, 1994) and the reaction was stopped at 1, 2, 4 and 8 min. As shown in figure 23, the P K G activity ratio (defined in Materials and Methods) in ventricular tissue treated with carbachol and S N P was changing at early time points. After 4 min of incubation, the activity ratio was measurable and stable. It was reasoned that the 4 min time point 119 Figure 23. Effect of assay incubation time on activity ratio of soluble cGMP-dependent protein kinase from agonist-treated rat ventricle. Langendorff-perfused rat hearts were treated with 1 nM isoproterenol for 2 min (O) (n=6), 10 uM carbachol and 1 nM isoproterenol for 2 min (•) (n=6) or 10 uM S N P and 1 nM isoproterenol for 2 min (A) (n=5) and then rapidly frozen. Soluble fractions of ventricular tissue were prepared and assayed for P K G activity as described in Materials and Methods. Aliquots of the reaction mixture were removed from the reaction vessel at specified times, the reaction stopped, and P K G activity ratios were determined. 120 0.7 -0.0 i 1 1 1 , 1 1 1 1-1 2 3 4 5 6 7 8 incubation time (min) 121 would provide a more reliable estimation of P K G activation than earlier time points because the activity ratio was not changing with time and activation of P K G by known cGMP-elevating agents could be measured. It should be noted, however, that underestimation of the degree of P K G activation is likely to occur under these conditions. 3.3.2 Activation of cGMP-dependent protein kinase in rat intact ventricular preparations. 3.3.2.1 Effect of carbachol, SNP and ANP on cGMP-dependent protein kinase activity in ventricle of Langendorff-perfused rat heart. Following validation of the assay as, described above, agonist-induced activation of P K G in intact ventricular preparations was measured. Firstly, the activity ratio of P K G was assessed in the soluble fraction of ventricular tissue from agonist-treated, Langendorff-perfused rat heart. When compared to isoproterenol alone, 10 pM carbachol in the presence of 1 nM isoproterenol did not have a significant effect on P K G activity after 2 min (figure 24). In contrast, the P K G activity ratio significantly increased by approximately 5 fold after a 2 min exposure to either 100 nM A N P plus isoproterenol or 10 pM S N P plus isoproterenol, when compared to isoproterenol alone. Isoproterenol (1 nM) had no significant effect on the ventricular soluble P K G activity ratio in comparison to control (control = 0.05 ± 0.02, n = 5; isoproterenol = 0.13 ± 0.04, n = 10). The absence of P K G activation by carbachol and the marked activation by A N P and S N P correlated well with changes in total tissue c G M P levels in ventricular tissue (figure 12). A comparison of these 122 Figure 24. Effect of carbachol, ANP and SNP, in the presence of isoproterenol, on soluble cGMP-dependent protein kinase activity ratio in rat ventricular tissue. Rat hearts were perfused with 1 nM isoproterenol for 3 min, washed, and re-challenged with isoproterenol for 2 min (ISO), isoproterenol plus 10 uM carbachol for 2 min (ISO + CARB), 100 nM A N P for 7 min of which the final 2 min included isoproterenol (ISO + ANP) or isoproterenol plus 10 uM S N P for 2 min (ISO + SNP). The soluble fraction of ventricular tissue was assayed for P K G activity which was expressed as the ratio of phosphorylation of substrate/min/mg protein at 4°C in the absence and presence of added cGMP. Numbers in bars indicate sample number. Asterisks (*) indicate significant difference from isoproterenol alone, 1-way ANOVA, Bonferroni's method, P < 0.05. 123 — 1.0 - , a. 1 o I S O I S O I S O I S O + + + C A R B A N P S N P 124 findings with the results of contractility experiments, represented in figures 3 and 5, illustrates a dissociation between P K G activation and agonist-mediated negative inotropy in the rat isoproterenol-stimulated intact ventricle (carbachol mediated a negative inotropic effect with no significant activation of P K G , whereas marked activation of P K G by A N P and S N P was not associated with a negative inotropic effect). In addition to assessment of the activity ratio of P K G in the soluble fraction of tissue extracts, total P K G activity in both soluble and particulate (Triton X -100-extracted) fractions of agonist-treated rat ventricle was measured in an attempt to detect possible translocation of the enzyme between subcellular compartments. Translocation of P K G from the soluble fraction to the particulate fraction, for example, to the sarcoplasmic reticulum (SR) (Cornwell et al., 1991) and the cytoskeleton (Pryzwansky et al., 1995), has been suggested to be an important event in the mediation of a functional effect by cGMP-elevating agents. If translocation occurred, decreases in total activity in one fraction would be coincident with increases in the other fraction, at least on a qualitative level. As shown in table 4, none of the agents used (isoproterenol, carbachol, A N P or SNP) significantly altered levels of total soluble or Triton X-100-extracted P K G . Total particulate P K G activity in the presence of 100 nM A N P appears larger than for other treatments (although it is not statistically different) but this average includes an outlier value (40.37 pmol/min/mg). If the outlier is excluded, the average particulate P K G activity in the presence of 100 nM A N P is 12.86 ± 1.18 pmol/min/mg. This table also illustrates that, since total kinase activity in the presence of drugs did not increase 125 Table 4: Total soluble and particulate cGMP-dependent protein kinase activity in ventricle of agonist-treated, Langendorff-perfused rat heart. Control cGMP-dependent protein kinase activity (+ 5 uM cGMP, 4°C; pmol/min/mg protein) soluble particulate n 3.49 ±0.31 9.73 ±0.73 ISO 1 nM for 2 min 4.12 ±0.42 9.75 ±1.70 10 CARB 10 pM + ISO 1 nM for 2 min 3.82 ±0.75 9.40 ±1.62 5 ANP 100 nM for 7 min + ISO 1 nM for 2 min 4.29 ± 0.58 18.36 ±5.58 5 SNP 10 pM + ISO 1 nM for 2 min 4.20 ±0.46 8.37 ±1.05 6 ISO, isoproterenol. C A R B , carbachol. n, number of ventricular tissue preparations. 126 above control levels, activation of protein kinases unrelated to P K G was not detected. 3.3.2.2 Effect of carbachol, SNP and ANP on cGMP-dependent protein kinase activity in isolated ventricular cardiomyocytes. The cellular source of P K G activity measured in intact ventricular tissue must be considered, since the heart contains a variety of cell types, including cardiomyocytes, neural tissue, fat, smooth muscle cells, fibrocytes and endothelium. The focus of this study is the activity of P K G in the mechanically functional unit of the heart, the cardiomyocyte. The relative contribution of different cell types to the total mass of the heart and the content and activation state of P K G in different cell types may lead to erroneous extrapolations of P K G activity from the intact ventricle to the cardiomyocyte. Accordingly, it was important to assess the effects of carbachol, A N P and S N P on P K G activity in agonist-treated isolated ventricular cardiomyocytes. To this end, myocytes from a single heart were isolated and divided into two aliquots. One aliquot from each heart was incubated with 1 nM isoproterenol for 2 or 6 min and the other was incubated with isoproterenol and either 10 pM carbachol for 2 min, 100 nM A N P for 6 min, or 10 - 100 pM S N P for 6 min. As shown in figure 25, carbachol (10 uM for 2 min) did not significantly increase P K G activity over paired control levels, similarly to results from the ventricle. While significant P K G activation by 10 uM S N P for 6 min was not demonstrated in this study, significant activation of P K G did occur in the presence of 100 uM S N P for 6 min. This concentration of S N P did not alter the contractility of rat ventricular strips or isoproterenol-stimulated ventricular cardiomyocytes after 6 min. 127 Figure 25. Effect of carbachol, ANP and SNP on soluble cGMP-dependent protein kinase activity in isolated rat ventricular cardiomyocytes. Isolated cardiomyocytes from a single rat heart were divided into two suspensions. One suspension was treated with 1 nM isoproterenol for 2 or 6 min (solid bars) while the other was treated with 1 nM isoproterenol plus either 10 pM carbachol for 2 min (CARB), 100 nM A N P for 6 min (ANP), or 10 - 100 pM, S N P for 6 min (SNP) (open bars). Cells were then rapidly centrifuged, frozen in liquid N 2 and stored at -70°C. The 30,000g soluble fraction of cardiomyocytes was assayed for P K G activity which was expressed as the ratio of phosphorylation of substrate/min/mg protein at 4°C in the absence and presence of added cGMP. Numbers in bars indicate sample number. Asterisks (*) indicate significant difference from isoproterenol alone, paired Student's f-test, P < 0.05. 128 129 A N P (100 nM for 6 min) increased the soluble P K G activity ratio by approximately 3-fold. As noted earlier, this concentration of A N P had no effect on the contractility of rat ventricular strips (in the presence or absence of isoproterenol), or intact rat hearts (in the presence of isoproterenol). The correlation between c G M P elevation and P K G activation was not as good in the isolated cardiomyocytes (figure 14 and figure 25) as it was in the intact ventricular preparations (figure 12 and figure 24), although there was a similar pattern of elevation and activation. Underestimation of cardiomyocyte P K G activation may account for this variance. Total P K G activity was measured in the soluble and Triton X-100 (0.1%)-extracted fractions of agonist-treated cardiomyocytes to ascertain whether translocation of P K G activity occurred between soluble and particulate fractions during agonist treatment. As shown in table 5, no evidence could be seen for inverse changes in total soluble and particulate P K G activity levels in the presence of 10 uM carbachol, 100 nM A N P or 100 uM S N P (plus 1 nM isoproterenol) when compared to paired cell suspensions (1 nM isoproterenol alone). No evidence for activation of non-PKG kinases was apparent. These effects on P K G activity in the intact ventricle and isolated cardiomyocytes demonstrate that P K G may be activated in the presence of agents which do not exhibit a negative inotropic effect in intact ventricular preparations (ANP and SNP) or isolated cardiomyocytes (SNP). Furthermore, P K G is not significantly activated in the presence of carbachol at a concentration which has a marked negative inotropic effect on isoproterenol-stimulated intact ventricular tissue and isolated cardiomyocytes. 130 Table 5. Total soluble and particulate cGMP-dependent protein kinase activity in isolated rat ventricular cardiomyocytes. C A R B 10 uM + ISO 1 nM ISO 1 nM for 2 min (pair to above) S N P 10 pM + ISO 1 nM for 6 min ISO 1 nM for 6 min (pair to above) cGMP-Dependent Protein Kinase Activity (+ 5 (iM c G M P , 4°C, pmol/min/mg protein) soluble particulate n 6.2 ±0.7 10.0 ±1.1 6 6.4 ±0.5 10.8 ±0.6 6 4.1 ±1.0 12.2 ±0.6 5 4.8 ±1.7 11.6 ±0.9 5 S N P 100 pM + ISO 1 n M f o r 6 min 4.4 ± 0.7 11.7 ±2.5 6 ISO 1 nM for 6 min (pair to above) 6.2 ±2.0 7.1 ±0.7 6 A N P 100 nM + ISO 1 n M f o r 6 m i n 3.7 ± 0.6 11.9 ±0.9 5 ISO 1 nM for 6 min (pair to above) 3.7 ±0.3 11.0 ±1.6 5 ISO, isoproterenol. C A R B , carbachol. n, number of cell suspensions. 131 4.0 DISCUSSION It has been suggested that the negative inotropic effects of muscarinic receptor agonists in the heart are mediated by increases in c G M P levels and activation of P K G . In agreement with this, negative inotropic effects have also been reported, in some studies, with other cGMP-elevating agents such as S N P and A N P . However, at the present time, there is considerable controversy regarding the precise relationship between c G M P elevation, P K G activation and negative inotropy in mammalian ventricular tissues. Part of this controversy involves data showing that S N P can increase c G M P levels in a variety of cardiac preparations without affecting contractility. A possible explanation for this apparent discrepancy was provided in earlier reports by Lincoln and Keely (1980, 1981), who found that acetylcholine increased c G M P levels and activated P K G in intact rat hearts while it exerted a negative inotropic effect on the hearts. S N P on the other hand, markedly elevated c G M P in these preparations but did not activate P K G and did not exert a negative inotropic effect on the hearts. Lincoln and Keely therefore suggested that activation of P K G was a necessary step in the chain of events leading to cGMP-mediated negative inotropy, and that S N P had no negative inotropic effects because it increased c G M P in a pool which did not have access to the kinase. The results of the studies in this thesis do not support the hypothesis of Lincoln and Keely. In the present study, S N P had no effect on contractility of isolated, perfused rat hearts, ventricular strips or freshly isolated ventricular cardiomyocytes, even though it markedly increased c G M P levels and activated P K G in these preparations. These and other results demonstrated a lack of correlation between c G M P levels, P K G activity and contractility in several ventricular preparations stimulated by the B-adrenoceptor agonist, isoproterenol. The results of 132 the studies described in this thesis are discussion of the results follows. Effects of carbachol briefly summarized below and a more detailed Effects of SNP and ANP • Induced a marked negative inotropic effect in isoproterenol-stimulated perfused rat hearts, ventricular strips and isolated cardiomyocytes. • Did not affect c G M P levels in intact ventricular tissue but significantly elevated c G M P levels in isolated cardiomyocytes • Did not affect the activity of soluble P K G in intact ventricular tissue or isolated cardiomyocytes. • Did not affect the contractility of isoproterenol-stimulated perfused rat hearts and ventricular strips (and S N P had no effect on the contractility of isoproterenol-stimulated isolated cardiomyocytes). • Markedly elevated c G M P levels in intact ventricular tissue (and S N P elevated c G M P in isolated cardiomyocytes to a greater extent than carbachol). • Significantly activated soluble P K G in intact ventricular tissue and isolated cardiomyocytes. 4.1 Effects of muscarinic receptor agonists on contractility and cGMP content in the ventricle. If c G M P is the mediator of the negative inotropic effects of muscarinic receptor agonists, then other cGMP-elevating agents such as S N P and A N P should have 133 similar effects on cardiac contractility. However, in terms of modulation of ventricular contractility, muscarinic receptor agonists differ markedly from S N P and A N P . Carbachol inhibited all the measured parameters of cardiac contractility in isoproterenol-treated intact hearts and ventricular strips, consistent with the well-documented role of the parasympathetic nervous system as the major inhibitory system of the heart (Endoh, 1995). "Accentuated antagonism" by muscarinic receptor agonists of p-adrenoceptor agonist effects (Levy, 1971) was exemplified by the absence of direct inhibitory effects of carbachol in untreated right ventricular strips (figures 6 and 7) and the presence of inhibition of strips which were pre-stimulated with isoproterenol (figure 5). Similarly, in intact hearts, carbachol exerted a negative inotropic effect when it was perfused concurrently with isoproterenol (figure 3, table 1). Reversal of ventricular strip twitch tension to almost control levels by 10 pM carbachol in the presence of 30 nM isoproterenol is comparable to the degree of negative inotropy seen with carbachol in the presence of sub-maximal concentrations of c A M P -elevating agents in a variety of ventricular tissue preparations, including canine ventricular trabeculae (Endoh and Shimizu, 1979; Endoh and Yamashita, 1981), human papillary muscle (Jakob et al., 1989), and guinea pig papillary muscle (Korth and Kulhkamp, 1987). In addition to inhibition of maximal force and pressure development, carbachol inhibited the stimulatory effects of isoproterenol on maximal rates of contraction and relaxation in the rat intact heart, in keeping with observations made in vivo (Landzberg et al., 1994; Hare et al., 1995), in intact hearts (Watanabe and Besch, 1975; Lindemann and Watanabe, 1985) and in ventricular muscle preparations (Mclvor et al., 1988). 134 Inhibition of contractility to levels below control by carbachol in intact hearts differs from previous studies which demonstrated a muscarinic receptor agonist-mediated reversal of the positive inotropic effect of cAMP-elevating agents without a reduction in contractility to below control levels (Watanabe and Besch, 1975; Keely and Lincoln, 1978; Lindemann and Watanabe, 1985). The cause of the greater inhibitory effect seen in this study is not clear but may result from a number of factors, including the use of a higher concentration of muscarinic receptor agonist (10 uM carbachol) than was used in previous reports and the relatively low concentration of c A M P -elevating agent (1 nM isoproterenol). Watanabe and Besch (1975) reported a reversal by 0.1 uM acetylcholine of the effects of 1 nM isoproterenol or 0.1 uM histamine on developed tension from 140% of control to slightly less than control (90%) in guinea pig isolated hearts. These authors did not use higher concentrations of acetylcholine, so it is not clear whether a further reduction in contractility below control levels would have been detected. Since tension in isoproterenol-stimulated ventricular strips was not diminished beyond control levels by carbachol (figure 5), it is likely that the difference in pharmacological response is due to a difference in the nature of the cardiac preparations. One such difference is that the perfused heart has an intact electrical conducting system and, although attempts were made to avoid atrio-ventricular blockade of electrical impulse conduction by placement of the stimulating electrodes on either side of the ventricles (across the breadth of the heart), cholinergic inhibition of electrical conduction within the ventricles may have occurred and may have led to the inhibition of contractility to below control levels. This would be consistent with reports of direct inhibition of contractility by cholinomimetics in intact hearts (Lincoln and Keely, 135 1980, 1981). Diminished conduction velocity would be less likely to occur in ventricular strips due to the comparatively short distance between stimulating electrodes in this preparation (< 0.5 cm). The use of isolated cardiomyocytes obviates a concern arising from the use of intact ventricular preparations containing heterogeneous cell types which can act as confounding sources of cyclic nucleotides and kinases. As well, the direct effects of agents on cardiomyocytes can be monitored, without complication by potential interactions between cardiomyocytes and other cells which may influence cardiomyocyte contractility. Interactions of both an inhibitory and excitatory nature, although not clearly defined, have been described between endothelial cells and cardiomyocytes (Smith et al., 1992). Cardiomyocytes were isolated from the myocardium by collagenase digestion, which raises the concern that the functional effects of the cells may be impaired by proteolytic damage. Considering that the cardiomyocyte preparation described herein demonstrated functional characteristics which compare very favorably with other published reports of contractile activity in ventricular cardiomyocytes (Stein et al., 1993; Xiao and Lakatta, 1993; Shah et al., 1994) and with intact cardiac preparations, this would seem to indicate that the preparation was representative of healthy myocardial cells with intact extracellular receptors. The cardiomyocyte preparation demonstrated a vigorous positive inotropic response to a low dose of isoproterenol (1 nM) both in terms of the extent of cell shortening and in the rates of contraction and relaxation. The time-frame of the isoproterenol-stimulated contraction was not abbreviated. A decrease in duration of 136 contraction has been seen at higher concentrations of this B-adrenoceptor agonist in some isolated cardiomyocyte preparations (Xiao and Lakatta, 1993) but not others (Balligand et al., 1995). These results are consistent with the effect of isoproterenol in the intact ventricular preparation where, for example, in the ferret papillary muscle, 1 nM isoproterenol enhanced only the amplitude of tension development, whereas a reduction of the time-to-peak tension and the time to 5 0 % relaxation required concentrations of 5 to 10 nM (Okazaki et al., 1990). The concentration of isoproterenol used in the present study is below the E C 5 0 for cell shortening (Harding et al., 1988) and was chosen so as to generate a significant positive inotropic effect which could potentially be surmounted by a weak negative inotropic signal. The muscarinic receptor agonists, acetylcholine and carbachol, had a negative inotropic effect which was statistically significant at 2 min on isoproterenol-stimulated cardiomyocytes. As mentioned above, negative inotropy by muscarinic receptor agonists is a well recognized effect in intact myocardial preparations in the presence of cAMP-elevating agents, but this effect has not been extensively characterized in isolated ventricular cardiomyocytes. In the present study, the amplitude and rate of shortening of the isoproterenol-stimulated cardiomyocyte contraction was depressed in the presence of carbachol or acetylcholine (figures 8 and 10), as was shown in intact preparations (figures 3 and 5, table 1), in the absence of changes in the time-frame of the contraction. This is comparable to results in the isoproterenol-stimulated ferret papillary muscle where 1 uM acetylcholine depressed the extent of twitch tension development and also reduced the rates of change in twitch tension, without changing the overall time frame of the contraction (Mclvor et al., 1988). Recently, Balligand et al. 137 (1995) reported a very similar inhibition of isoproterenol (1 pM)-stimulated cell shortening amplitude and rates of contraction and relaxation by 1 pM carbachol in adult rat ventricular cardiomyocytes. The present study demonstrates that c G M P levels can be elevated by carbachol and acetylcholine in the isoproterenol-stimulated rat ventricle, although increases reached statistical significance only in isolated cardiomyocytes. The lack of statistical significance in the c G M P elevation by 10 pM carbachol in intact hearts was not altogether unexpected, even though lower concentrations of the agonist have been shown to significantly elevate c G M P levels in various intact ventricular tissues (Lincoln and Keely, 1980, 1981; Inui et al., 1982; MacLeod and Diamond, 1986). Muscarinic receptor agonists are relatively ineffective cGMP-elevating agents in comparison to S N P in the heart, maximal c G M P increases being approximately 2-fold above basal (Lincoln and Keely, 1980, 1981; MacLeod and Diamond, 1986). Phosphodiesterase inhibitors are required to demonstrate more substantial c G M P elevations by cholinomimetics (Endoh and Honma, 1979; Cramb et al., 1987), suggesting that the potential capacity for muscarinic receptor agonist-mediated c G M P elevation is muted by phosphodiesterase activity. Considering that phosphodiesterase activity can vary between individual hearts (Endoh and Honma, 1979) and possibly even between cells of the same heart (Kojda et al., 1996), it may be possible that phosphodiesterase activity was relatively high in the intact ventricular tissue studied herein. However, it should be noted that the qualitative trend in c G M P levels in the presence of carbachol, S N P and A N P in intact ventricular tissue (figure 12) is consistent with the reported 138 relative efficacy of these agents in elevating cardiac c G M P levels (Lincoln and Keely, 1980, 1981; Cramb era/., 1987). The individual cardiomyocyte is clearly a source of muscarinic receptor agonist-stimulated c G M P synthesis. Cramb et al. (1987) noted a 2-fold increase in rabbit ventricular cardiomyocyte c G M P content using a higher concentration of carbachol (100 uM, in the presence of a phosphodiesterase inhibitor) than was used in this thesis. A muscarinic receptor-mediated increase in adult ventricular cardiomyocyte c G M P content in the absence of a B-adrenergic agonist or a phosphodiesterase inhibitor has also been reported by Stein et al. (1993). Thus, it is apparent that muscarinic receptor agonists are linked to a guanylyl cyclase in the cardiomyocyte itself. The nature of the transduction mechanism between receptor binding and cyclase activation remains unclear and has been postulated to be mediated by nitric oxide (Balligand et al., 1993, 1995) or a nitric oxide-independent mechanism (Stein et al., 1993). 4.2 Effects of SNP on contractility and cGMP content in the ventricle. Since negative inotropic effects of muscarinic receptor agonists usually require the concomitant exposure of ventricular tissue to cAMP-elevating agents, it was reasoned that other cGMP-elevating agents such as S N P would produce a negative inotropic effect under the same conditions. Of the few studies which have investigated the functional relationship between NO donors and B-adrenoceptor agonists in the ventricle, most have failed to detect a SNP-mediated negative inotropic effect (Keely and Lincoln, 1978; Endoh and Yamashita, 1981; Hongo et al., 1993). For example, Hongo et al. (1993) observed that S N P (0.1 - 1 mM) did not change the maximal 139 amplitude or time-course of contraction in 0.1 pM isoproterenol-stimulated ferret papillary muscle. In agreement with these findings, the present study demonstrated a lack of negative inotropic effect by S N P on isoproterenol-stimulated intact heart maximal LVP and maximal positive or negative dP/dt, on isoproterenol-stimulated ventricular strip twitch tension and on amplitude, rate and time-course of cell shortening in isoproterenol-stimulated cardiomyocytes. This is the first reported characterization of the effects of S N P on rates of contraction and relaxation in p-adrenoceptor-stimulated intact hearts and on contractile properties of isoproterenol-stimulated cardiomyocytes. A range of concentrations of S N P (0.1 - 100 pM) was tested for a contractile effect in ventricular strips in order to detect mechanical effects which may arise only at specific concentrations. A recent report described positive inotropic effects of NO donors at low concentrations and negative inotropic effects at high concentrations in rat cardiomyocytes (Kojda et al., 1996) but a similar pattern was not observed in the present study. No direct negative inotropic effect in ventricular strips, in terms of reduction in maximal twitch tension in the absence of p-adrenoceptor stimulation, was observed in the presence of SNP. Even the pharmacological concentration of 100 pM S N P (the reported therapeutic concentration of NO donors is 0.5 pM (Weyrich et al., 1995)), failed to change twitch tension over the course of 9 min. These results are in agreement with an absence of direct effects of S N P and other NO-releasing agents on twitch tension in canine and rabbit ventricular tissue (Endoh and Yamashita, 1981; Inui et al., 1982; Rodger and Shahid, 1984), on maximal ventricular pressure or force in rat and guinea pig perfused hearts (Lincoln and Keely, 1980, 1981; Thelan et al., 1992; 140 Grocott-Mason, 1994) and on maximal cell shortening in guinea pig isolated ventricular cells (Stein et al., 1993). Thus, according to these results, S N P is similar to carbachol in that it failed to elicit a negative inotropic effect in the absence of p-adrenoceptor stimulation. However, small direct negative inotropic effects of S N P have been observed in some ventricular intact tissue and cardiomyocyte preparations (Fort et al., 1991; Smith etai, 1991; Brady era/., 1993; Kojdaefa/., 1996). For example, S N P (10 uM, 10 min) decreased guinea pig isolated cardiomyocyte shortening amplitude from 5.8% to 4 . 7 % without changes in the times to maximal contraction or to full relaxation (Brady et al., 1993). The physiological importance of such negative inotropic effects is as yet unknown but, considering the magnitude of the effect, it is probably fairly minor. While the rat ventricular myocardium, with or without p-adrenoceptor stimulation, did not respond to S N P with changes in contractility, a mechanism for stimulation of c G M P synthesis by S N P was apparent in the rat heart. S N P (10 uM, 2 min) increased c G M P levels by 3-fold in ventricular tissue from intact hearts and S N P (100 uM, 9 min) elevated cellular c G M P levels by approximately 2.5-fold over control levels in ventricular strips. Several cell types, including smooth muscle cells, were probable sources of c G M P in these preparations, but it is clear that cardiomyocytes were involved since S N P increased c G M P levels by 2.5- to 8-fold over control levels in suspensions of cardiomyocytes after 2 - 6 min. These results are in accordance with previous studies in rabbit and guinea pig isolated ventricular cardiomyocytes (Cramb et al., 1987; Stein et al., 1993). SNP increased c G M P to a larger extent than did cholinergic agonists in all three ventricular preparations without changing contractility. 141 This suggests that c G M P formation was concurrent with, but not responsible for, the negative inotropy caused by the muscarinic agonists in the rat isoproterenol-stimulated ventricle. Several studies corroborate observations in the present study of a dissociation between elevation of c G M P by S N P and negative inotropy in intact ventricular tissues (Endoh and Yamashita, 1981; Lincoln and Keely, 1980, 1981; Rodger and Shahid, 1984) and isolated cardiomyocytes (Stein et al., 1993). In contrast, direct negative inotropic effects of another NO donor, 3-morpholinosydnonimine (SIN-1), were detected in rat isolated cardiomyocytes but this effect correlated poorly with c G M P levels since cell shortening was diminished by 1 pM SIN-1 but 1 mM SIN-1 was required to significantly elevate c G M P (Schluter et al., 1994). Moreover, the negative inotropic effect of SIN-1 could be partially accounted for by an active, c G M P -independent metabolite, SIN-C. In a recent report, concentrations of NO donors which induced small increases in cellular c G M P content (1.5- to 2.5-fold above basal levels) produced a direct positive inotropic effect and enhanced the positive inotropic effect of isoproterenol in rat isolated cardiomyocytes (Kojda et al., 1996). Increases in both basal and isoproterenol-stimulated cellular cAMP levels were associated with the NO donor-mediated positive inotropic effects and the authors concluded that low concentrations of c G M P inhibited cAMP phosphodiesterases, such that the resulting increase in cAMP levels may have mediated an activation of PKA. Elevation of cellular c G M P to higher levels (3.5- to 6-fold above basal levels) were associated with activation of P K G activity (based on its sensitivity to the putative P K G inhibitor, KT5823) which antagonized the positive inotropic effect of PKA activation, resulting in 142 a net negative inotropic effect. Cyclic G M P levels in rat isolated cardiomyocytes in the present study were elevated to both relatively low (<2.5-fold) and high (8-fold) extents by S N P , yet neither a positive nor a negative inotropic effect was observed. This suggests that neither cyclic GMP-mediated stimulation nor inhibition of c A M P phosphodiesterases by S N P occurred in the present study, a contention which is supported by measurements of cAMP levels in intact ventricular tissue and cardiomyocytes, as described below. It should be noted that, in the Kojda et al. (1996) study, basal c G M P levels were 1 0 - 2 0 times higher and cAMP levels 10 times lower than reported herein. These differences, in addition to cGMP-sensitive cAMP metabolism observed in the Kojda et al. (1996) study but not the present study, may reflect differing patterns of phosphodiesterase isozyme content and catalytic activity between cardiac preparations of the same species, as suggested previously (Endoh and Honma, 1979). Only one other study examined the effect of a NO donor in the presence of a cAMP-elevating agent on c G M P levels in the same tissues as were used for contractile experiments. Keely and Lincoln (1978) described a 3-fold increase in ventricular tissue levels of c G M P from perfused rat hearts after treatment with 5 uM S N P and 0.1 uM adrenaline for 1 min, data which are in close accordance with those presented here. On the contrary, Inui et al. (1982) reported that S N P decreased isoproterenol-stimulated force development in rabbit ventricular tissue and elevated tissue c G M P content. Interpretation of this apparent association is complicated by the fact that different tissue types were used to measure c G M P (right ventricular strips) and contractility (papillary muscles). Furthermore, c G M P measurements were made after 143 exposure to 100 pM S N P for 30 - 60s but contractility was determined in cumulative concentration-response curves allowing for exposure to S N P (10"6 - 10"3 M) over a period of approximately 40 min. cGMP-independent contractile effects of free radical (-NO) generating agents, such as SNP, are an important consideration since nitrosylation of essential biomacromolecules by redox-active NO donors may have pharmacological and toxicological consequences (Freeman, 1994). Evidence exists for cGMP-independent modification of cellular function by SNP-derived NO in the cardiovascular system at physiologically relevant concentrations (10"8 - 10"4 M) (Brune et al., 1993; Gupta et al., 1994b). S-nitrosylation of the cardiac glycolytic enzyme glyceraldehyde-3-phosphate dehydrogenase (GAPDH) by S N P and SIN-1 leads to further covalent modification (ADP-ribosylation) and, ultimately to loss of catalytic activity (Brune and Lapetina, 1989; Boyd et al., 1993). The functional consequences of NO-mediated G A P D H inactivation in the heart have not been described, but it is interesting to note that S N P inhibited glycolysis in rat atria under normal and hypoxic conditions (Laustiola et al., 1983). In addition to its glycolytic activity, G A P D H is involved in transcription and DNA repair and is associated with the cytoskeleton (Brune et al., 1993). It is reasonable to anticipate that NO covalently modifies other cardiac proteins in a cGMP-independent manner, particularly iron-sulfur- and sulfhydryl-containing proteins, which may alter metabolic and contractile function. 4.3 Effects of ANP on contractility and cGMP content in the ventricle. A N P ( 1 - 1 0 0 nM), like SNP and carbachol, did not directly reduce ventricular strip twitch tension. Furthermore, A N P , like SNP, bore little similarity to muscarinic 144 receptor agonists in modulating B-adrenoceptor-stimulated ventricular contractility. No anti-adrenergic effect of A N P was detected in intact hearts or ventricular strips in the presence of isoproterenol. It is unlikely that a potential effect, apparent only at high concentrations, was overlooked since the typical test concentration (100 nM) was several orders of magnitude above physiological peripheral and coronary sinus levels (Raine et al., 1986; Dubois-Rande et al., 1991) and above coronary sinus levels seen in pathological conditions, such as congestive heart failure (Dubois-Rande et al., 1991). These findings are in agreement with observations of a lack of contractile effect of up to 1 uM A N P in human, guinea pig, and rat ventricular strips or papillary muscles, in the presence or absence of isoproterenol stimulation (Hiwatari et al., 1986; Criscione et al., 1987; Bohm et al., 1988). An intact endocardium was required for the expression of a small, slowly developing negative inotropic effect (3 - 5 % reductions in twitch tension and onset of relaxation after 10 to 15 min) by 1 - 100 nM A N P in cat papillary muscles (Meulemans etal., 1988) and by 400 nM in ferret papillary muscles (Smith et al., 1991), comparable to the slight, endocardium-independent effect of S N P which has been observed in some studies (Smith et al., 1991; Brady et al., 1993). A slight endocardial-dependent decrease in ventricular strip contractility may have been overlooked in the present study if the viability of the endocardium in rat ventricular strips was compromised. Although manipulation of the strips may not be as disruptive to the endocardium as the detergent treatment used in the studies cited above to damage ventricular endocardial cells, care was taken to handle the strips gently. There is no reason to believe that the endocardium in the intact heart was compromised but decreases in contractility on the 145 scale of 5 % would be less than the standard error of the measured parameters and, therefore, it cannot be conclusively stated that subtle reductions in intact heart contractility did not occur. However, in agreement with the results reported herein, no negative inotropic effects of A N P have been detected in several other studies of endocardium-intact cardiac preparations. Hutter (1991) reported an absence of inhibition of maximal positive dP/dt by 0.1 - 100 pM A N P in rat isolated working hearts, and intracoronary infusions of A N P (0.4 - 1.3 nM) failed to alter positive dP/dt in humans with normal cardiac function or congestive heart failure (Duboise-Rande et al., 1991). Likewise, intracoronary infusions of very high concentrations of A N P (0.3 - 30 pM) in anesthetized dogs did not alter ventricular myocardial maximal force (F) or maximal positive or negative dF/dt (load-independent indices) or left ventricular maximal positive or negative dP/dt (Shimizu etal., 1988). Similarly to SNP, 100 nM A N P significantly elevated ventricular tissue c G M P content by 3-fold in perfused rat hearts, thus supporting a dissociation between c G M P and negative inotropy in the mammalian ventricle. The cellular source of A N P -stimulated c G M P in intact ventricular tissue probably includes the cardiomyocyte, since high affinity A N P binding sites (K d = 12 pM) have been detected in rat cardiomyocytes (Neyes and Vetter, 1989) and ventricular c G M P content of cardiomyocytes dose-dependently rises in the presence of A N P (EC 5 0 » 10 nM) (Cramb et al., 1987). A N P receptors are also expressed on cardiac endothelial cells and vascular smooth muscle cells (Cantin and Genest, 1985) and these sources probably contributed to A N P -stimulated c G M P elevations in intact ventricles. 146 The ineffectiveness of A N P in reducing ventricular contractility in in vitro studies has been ascribed to physical binding of A N P to the experimental apparatus (e.g. to plastic tubing or glassware) and some investigators added bovine serum albumin to perfusion solutions to prevent loss of the peptide (Hotter, 1991). Inactivation of A N P in this manner was not a substantial concern in the present study, since A N P significantly increased tissue c G M P content several fold higher than carbachol and to an extent comparable to SNP. The results of the present study contribute to a very small body of literature which relates c G M P content with ventricular contractility in the presence of A N P . Interpretation of earlier reports is hampered by the fact that no attempts were made to measure c G M P in the same preparations as were used for contractility studies. Smith et al. (1991) reported a 4-fold increase in c G M P by 400 nM A N P in ferret papillary muscle and, in separate tissue samples, only a 3 % decline in twitch tension and indices of relaxation time. Neyes and Vetter (1989) noted that rat cardiomyocyte shortening amplitude declined from approximately 5 .5% to 3 .7% of resting cell length ( « 3 5 % decrease) in the presence of 100 nM A N P and, in separate experiments, 100 nM A N P increased c G M P content by about 2-fold. However, the two phenomena were likely to be functionally unrelated, since much higher concentrations of A N P were required to elevate c G M P levels (100 nM ANP) than were necessary to diminish cell shortening (EC 5 0 ~ 70 pM). Likewise, peripheral infusion of physiologic concentrations of A N P in normal humans resulted in elevated plasma c G M P levels and vasodilation but no direct cardiac effects were detected (Roy et al., 1989). Based on these reports and the data in the present study, tissue and cellular c G M P content correlate poorly 147 with changes in ventricular contractility in the presence of A N P . This is similar to the dissociation between c G M P content and contractility seen in the presence of SNP, further supporting the contention presented in the previous section that c G M P elevation is not responsible for the negative inotropy caused by muscarinic agonists in the rat isoproterenol-stimulated ventricle. 4.4 Effects of carbachol, SNP, and ANP on ventricular cAMP content. It has been suggested that muscarinic receptor agonists may decrease isoproterenol-stimulated contractility by either preventing the isoproterenol-produced formation of cAMP via inhibition of adenylyl cyclase (Murad et al., 1962; Endoh, 1979; Watson et al., 1988) or by stimulating the catabolism of cAMP via activation of c G M P -stimulated-cAMP-phosphodiesterase (Fischmeister and Hartzell, 1987). Muscarinic receptor agonists have been shown to decrease p-adrenoceptor- and phosphodiesterase inhibitor-stimulated increases in cAMP levels and contractility in ventricular preparations (Endoh and Honma, 1979; Inui et al., 1982) but in other studies, including the present one, muscarinic receptor agonists elicited a negative inotropic effect without altering isoproterenol- or forskolin-stimulated c A M P levels (Watanabe and Besch, 1975; MacLeod and Diamond, 1986). Assessment of total tissue levels of cAMP, like the situation with cGMP, may be confounded by the contribution of cAMP from non-cardiomyocyte cells. It was, therefore, of particular interest to determine the effects of muscarinic receptor agonists on isolated cardiomyocyte cAMP content in the presence of isoproterenol. Of the studies which have evaluated the effect of muscarinic receptor agonists on cAMP levels in isolated cardiomyocytes, Gupta et al. (1994a) noted that a variety of muscarinic receptor 148 agonists did not depress isoproterenol-elevated cAMP levels in guinea pig ventricular cardiomyocytes but Kubalak et al. (1991) showed a concentration-dependent antagonism of isoproterenol-stimulated cAMP elevation by carbachol in canine ventricular cardiomyocytes. The results of our study are in agreement with those of Gupta et al. (1994a). Differences from the Kubalak et al. (1991) report may result from a 24 hour lag between cell isolation and agonist treatment in the latter study, during which time cellular differentiation may have occurred. Katano and Endoh (1993) also noted antagonism of isoproterenol-stimulated cAMP levels in freshly isolated rat ventricular cardiomyocytes by carbachol, although a relatively high concentration of carbachol (100 uM) was used in that study. Since no significant depression in isoproterenol-stimulated cAMP levels by carbachol was observed in either intact ventricular tissue or isolated cardiomyocytes under the conditions of the present study, it can be suggested that neither adenylyl cyclase inhibition nor cAMP phosphodiesterase activation played an important role in the negative inotropic effect of muscarinic receptor agonists in this study. These results cannot rule out the possibility that signalling events downstream from cAMP production, such as phosphatase activation and/or reversal of cAMP-dependent protein phosphorylation, may result from muscarinic receptor activation. The cGMP-elevating effect of A N P is consistently observed in target organs but A N P receptor-mediated decreases in adenylyl cyclase activity via a G, protein-coupled mechanism have also been reported in several tissues, including cultured cardiac myocytes (Anand-Srivastava et al., 1986). McCall and Fried (1990) correlated an ANP-mediated direct negative inotropic effect with pertussis toxin-sensitive decreases 149 in cAMP in cultured neonatal rat ventricular cardiomyocytes. Adenylyl cyclase inhibition by A N P may be more evident in cultured cells since depression of cAMP levels was not apparent in isoproterenol-stimulated ventricular cardiomyocytes in this study or in freshly isolated rabbit cardiomyocytes in the presence or absence of isoproterenol (Cramb et al., 1987). It is interesting to note that a chemical interaction between S N P and isoproterenol did not occur with their simultaneous use. Isoproterenol is susceptible to oxidation and, since S N P can participate in reduction-oxidation reactions as an oxidizing agent (Smith et al., 1990), S N P could potentially oxidize and inactivate isoproterenol. The absence of a decrease in isoproterenol-stimulated contractility or cAMP accumulation indicated that a significant chemical interaction did not occur in the present study. Some studies have noted positive contractile and electrophysiological effects of cGMP-elevating agents (NO donors, NOS inhibitors, intracellularly dialysed cGMP) which were associated with inhibition of cAMP degradation (Ono and Trautwein, 1991; Klabunde et al., 1992; Kojda et al., 1996). SNP-mediated inhibition of cAMP phosphodiesterases, with a resultant increase in cAMP levels, was not observed in this study, in keeping with observations of a lack of a SNP-mediated positive inotropic effect. The results of the present study are in agreement with the lack of elevation of cAMP levels by S N P in canine, rabbit and rat intact ventricular tissues (Keely and Lincoln, 1978; Endoh and Yamashita, 1981; Rodger and Shahid, 1984). 150 4.5 Effects of carbachol, SNP and ANP on ventricular cGMP-dependent protein kinase activity. A role for c G M P could still be implicated in mediating the negative inotropic effects of muscarinic receptor agonists if these agents, but not S N P or A N P , selectively activate a cGMP-regulated protein and, thereby, decrease contractile function. Thus, the objective of the final series of experiments in this study was to test the hypothesis that P K G is activated in the heart specifically by cGMP-elevating agents which mediate a negative inotropic effect and not by cGMP-elevating agents which do not affect contractility. Compartmentalization of c G M P has been proposed as the means by which P K G could be selectively activated (Lincoln and Keely, 1980, 1981). An answer to the question of intracellular compartmentalization of components of the c G M P cascade was sought by measuring P K G activity in agonist-treated intact tissue as well as in isolated cardiomyocytes. Efforts were made to confirm the existence of P K G in the heart. P K G , unlike its homologue PKA, is differentially distributed in the body, with high concentrations detected particularly in tissues containing significant amounts of smooth muscle cells, e.g., blood vessels, trachea, lung parenchyma, and parts of the gastrointestinal tract (Lincoln et al., 1977; Keilbach et al., 1992). Enzymatic activity ascribed to P K G has been found in the intact heart (Lincoln et al., 1976; Flockerzi et al., 1978; Lincoln and Keely, 1980, 1981) but cell-specific distribution of P K G could not be determined since intact cardiac tissues were used. Separation of protein kinase-containing fractions of rat ventricle by MonoQ anion exchange chromatography demonstrated that the assay conditions used in the present study selectively detected cGMP-dependent activity 151 without interference from PKA or cyclic nucleotide-independent protein kinase activity. The latter effect, due either to PKG-unrelated kinase activity or to catalytically active proteolytic fragments of the P K G homodimer, was a substantial confounding source of kinase activity in previous studies, based on the detection of significant phosphorylation of histone in the absence of c G M P in crude and partially purified extracts of rat cardiac tissue (Lincoln and Keely, 1980, 1981). The present study also demonstrates the presence of P K G , as measured by enzymatic activity, both in intact ventricular tissue and, for the first time, in isolated cardiomyocytes. Total soluble P K G recovered from chromatography of intact ventricular tissue was comparable to total soluble P K G in isolated cells, as calculated on the basis of activity per mg of loaded protein. It is apparent that cardiac P K G activity levels are low in general but that cardiomyocytes contribute significantly to total cardiac P K G content. In contrast, immunological techniques have detected P K G immunoreactivity almost exclusively in smooth muscle cells of cardiac tissue with very little immunoreactivity in cardiomyocytes (Joyce et al., 1984; Ecker et al., 1989). The reason for the difference between the immunocytochemical studies and the biochemical results described herein is not clear but may be related to differences in the immunoreactivity of P K G between smooth muscles and cardiomyocytes. It might be argued that contamination of the cardiomyocyte preparation with PKG-rich cells, such as smooth muscle cells, could account for the apparent cardiomyocyte-specific P K G activity in the present study. However, contamination was minimal because repeated gravity sedimentation of isolated cardiomyocytes removed essentially all non-card iomyocyte cells. Cardiomyocytes constitute approximately 2 5 % of all myocardial cells (Bugaisky, 1988) 152 but, due to their large volume, comprise in excess of 7 5 % of the myocardial mass (Olivetti et al., 1980). The approximately 10-fold greater density of cardiomyocytes in comparison to smooth muscle cells allows purification by sedimentation of cardiomyocytes from the smaller, more buoyant smooth muscle cells. Microscopic visualization of cardiomyocyte suspensions showed that the vast majority of cells were cardiomyocytes, primarily viable rod-shaped cells with a component of rounded cells of a comparable volume, i.e. dead cardiomyocytes. Thus, cardiomyocytes from rat ventricle were considerably enriched in number relative to non-cardiomyocyte cells by the isolation procedure, although in the absence of histological evidence the exact purity of the cardiomyocyte preparation cannot be specified. In relation to total cardiomyocyte P K G activity, the results of this study indicate that levels of particulate P K G activity are substantial. This is based on the observation that the total amount of P K G activity recovered from MonoQ chromatography of particulate cardiomyocyte extracts was similar to the recovery of P K G activity from soluble extracts, when adjusted for the amount of protein loaded on to the columns. This differs from previous descriptions of intracellular distribution of cardiac P K G activity in rat intact cardiac tissue. Lincoln and Keely (1981) reported that greater than 9 5 % of total P K G catalytic activity was in the soluble extract and Ecker et al. (1989) estimated that 7 5 % of total P K G was soluble, as detected by immunoblotting. The level of detergent-extractable P K G activity cannot be exactly quantified by the present study, since entrapment of soluble P K G in the particulate fraction during the initial sedimentation of cell extracts (30,000g for 5 min) may have occurred. 153 The existence of P K G in the heart was further substantiated by the detection of PKG-specific immunoreactivity in MonoQ fractions which expressed P K G enzymatic activity. This confirms other reports of P K G immunoreactivity in bovine and rat intact hearts (Ecker et al., 1989; Keilbach et al., 1992) and in rat isolated cardiomyocytes (Mery et al., 1991). Classification according to isozyme identity was not possible due to a lack of availability of adequate isozyme-specific P K G antibodies but, since Keilbach et al. (1992) detected only the la isozyme in the intact rat heart, the P K G isozyme profile in isolated cardiomyocytes is likely restricted to the la isozyme. The close similarity between MonoQ profiles of soluble P K G activity from intact hearts and cardiomyocytes supports this contention, since the presence of both isozymes in a single tissue gives rise to two distinct peaks in anion exchange chromatography (Lincoln et al., 1988; Wolfe et al., 1989). The la and ip isozymes have a very high sequence identity, differing only in the first 89 or 104 amino acids of the N-terminus (Francis and Corbin, 1994). Differential functional specificity of the two isozymes has not been described but it is interesting to note that the ip isozyme has a 2.5-fold to 10-fold lower affinity for c G M P than does the la isozyme (Ruth et al., 1991; Sekhar et al., 1992), and predominates in smooth muscle tissues which contain high levels of soluble guanylyl cyclase activity (Keilbach et al., 1992). It has been hypothesized that the low-affinity form of P K G plays a role in blunting the effects of high c G M P levels in tissues such as blood vessels (Keilbach et al., 1992). The corollary that the effects of low levels of c G M P in tissues such as the heart are accentuated through activation of the high affinity la P K G isozyme has not yet been tested. 154 Measurement of agonist-induced activation of P K G in intact tissues and cells is a more problematic procedure than is the case for PKA (Fiscus and Murad, 1988). The catalytic subunits of the PKA heterotetramer dissociate upon c A M P binding to the regulatory subunits. Thus, the activation state of the enzyme can be preserved during assay procedures by conditions, such as dilution, which prevent re-association of the catalytic and regulatory subunits. On the contrary, the catalytic portion of P K G is contiguous with c G M P binding sites and conditions must be used which limit dissociation of the c G M P - P K G complex. This is complicated by the tendency of c G M P to dissociate from P K G upon dilution during tissue homogenization. At 20°C, c G M P dissociates from P K G with a tVl of 0.6 min (McCune and Gill, 1979). The absence of agonist-induced P K G activation measured at an assay temperature of 30°C in the present study confirms the lability of c G M P - P K G binding. Fortunately, the binding affinity of c G M P for P K G is increased 100-fold at 0°C (Francis et al., 1988) and it is feasible to measure agonist-mediated activation if tissue extracts are rapidly prepared and assayed at 0°C with brief incubation times (Fiscus et al., 1984). This was confirmed by the assessment of P K G activity in agonist-treated intact ventricles at various assay incubation times (0°C) (figure 23). The results indicated a persistence of P K G activation during the course of the assay. Even using low temperatures for the preparation and assay of tissue extracts, underestimation of P K G activation remains a possibility, particularly considering that, of the two distinct c G M P binding sites on each P K G monomer, site 2 has a lower affinity for c G M P and the nucleotide dissociates from this site rapidly, even at 0°C (tA < 10 s) (MacKenzie, 1982; Corbin and Doskeland, 155 1983). Therefore, detection of only a portion of agonist-mediated P K G activation may occur. Another concern in the P K G assay of crude extracts is that activation of P K G may reflect exposure to c G M P during tissue disruption rather than within the intact cell. This could arise if c G M P was elevated in intracellular pools which had no access to P K G in the intact cell but, upon homogenization, spuriously increased P K G activity. This issue was considered by Jiang et al. (1992) who reported agonist-mediated P K G activation in porcine coronary arteries. c G M P levels detected in 10 pM SNP-treated porcine coronary arteries were 10-times higher than agonist-stimulated levels in ventricular tissues described herein. Even so, the authors concluded that S N P -stimulated P K G activity was underestimated by dilution of cellular constituents in 2 volumes of homogenization buffer (2 mL per gram of tissue) rather than exaggerated by in vitro activation. In addition to comparatively low control and agonist-stimulated c G M P levels in the present study, ventricular tissue and cardiomyocyte homogenates were prepared in 5 volumes of buffer, which created conditions even less conducive to spurious activation during processing than in the Jiang et al. (1992) study. Furthermore, it can be reasoned that, since c G M P tends to dissociate from P K G during the early stages of the assay (Fiscus et al., 1984; cf. figure 23), the force of mass action is towards underestimation of P K G activation, not overestimation by in vitro association. Palmer et al. (1980) introduced the use of charcoal in homogenization buffers as a tool to adsorb free cyclic nucleotides which might lead to spurious kinase activation. Fiscus et al. (1984) found that 3 pM methacholine and 300 pM S N P significantly activated canine trachealis regardless of the presence or absence of 156 charcoal (3 mg/mL) in the homogenization buffer, although the degree of activation by S N P was somewhat attenuated in the presence of charcoal. Charcoal was not added to the homogenization buffer in the present study because, in addition to adsorbing free cyclic nucleotides, charcoal lowers total protein (Fiscus and Murad, 1988), may bind and inactivate P K G (Lincoln, 1983) and may promote dissociation of PKG-bound c G M P . Thus, the effects of charcoal on P K G activity ratios are complicated and cannot be unambiguously interpreted. Since the early 1980's, changes in the P K G assay have considerably improved its reliability such that more accurate descriptions of agonist-induced P K G activation in intact tissues can be made. A key difference in the present study is that the assay of soluble P K G activity ratios was conducted in rapidly processed extracts (maximum of 7 min processing time) at 0°C for a brief incubation time (4 min) as opposed to more prolonged processing (> 20 min) and assaying at 30°C (incubation time periods were not described), used in the Lincoln and Keely (1980, 1981) reports. If the latter conditions were used, no agonist-mediated P K G activation could be detected in this study or in others which used smooth muscle preparations (Fiscus et al., 1984). The use of the novel, PKG-specific BPDEtide substrate served to improve the sensitivity of the assay by elimination of cyclic nucleotide-independent protein kinase interference, as demonstrated by the absence of cyclic nucleotide-independent peaks in MonoQ elution profiles and the detection of low control P K G activity ratios in crude extracts. Activity ratios of untreated and isoproterenol-treated ventricular tissues were generally < 0.20. This is 30-50% lower than the activation level of control intact tissues, including the ventricle, when assayed with histone H2B as a substrate (Lincoln and Keely, 1980, 157 1981; Fiscus et al., 1984, 1985) and is comparable to basal activity ratios in porcine coronary arteries assayed with BPDEtide (Jiang etal., 1992). The results of the present study demonstrated that ventricular tissue P K G activity, as measured by the ratio of endogenously active P K G to total P K G , in the presence of carbachol, SNP , or A N P correlated with c G M P levels in the same intact ventricular tissues. Even though S N P and A N P stimulate c G M P synthesis by different pathways, they were equally effective in increasing tissue c G M P content and activating P K G . Similarly, the lack of a significant effect of carbachol on tissue c G M P content was reflected in an absence of P K G activation. In isolated cardiomyocytes, the effects of carbachol and S N P on P K G activity mirrored changes in intact tissue P K G activity on a qualitative basis, although the increases in cardiomyocyte c G M P content by 10 pM carbachol and 10 pM S N P were not reflected in significant cellular P K G activation. Only at the higher concentration of 100 pM S N P was significant cellular P K G activation detected. Thus, c G M P content appeared to be a more sensitive marker of agonist-stimulated signalling in cardiomyocytes. A strength of the intact tissue studies was that c G M P content and P K G activity, as well as contractility, were measured in the same agonist-treated samples. Measurements of these parameters in different samples of cardiomyocytes may have complicated the assessment of the quantitative relationship between c G M P levels and P K G activation. Furthermore, considering that only small increases in c G M P levels (e.g., 2-fold) were needed for P K G activation in intact tissues (Lincoln and Keely, 1980, 1981; Fiscus et al., 1984, 1985), and with the aforementioned caveat in mind regarding the possibility that some PKG-bound c G M P may dissociate from the kinase during the assay, it cannot be ruled out that P K G was 158 activated by carbachol in both ventricular preparations and by low concentrations of S N P in cardiomyocytes but that the level of activation was underestimated by the conditions of the assay. Nevertheless, it is unlikely that the degree of carbachol-mediated P K G activation approached that produced by S N P or A N P , since the latter agonists consistently increased ventricular c G M P content by several fold more than did carbachol. The effects of carbachol, S N P and A N P on tissue and cellular P K G activity bore no positive relationship to the eliciting of a negative inotropic effect. In the presence of isoproterenol, carbachol had no significant effect on the P K G activity ratio at a concentration which exerted a marked negative inotropic effect in intact ventricular tissue and isolated cardiomyocytes. Furthermore, P K G was activated in the presence of S N P and A N P at 2 min (100 nM A N P and 10 uM S N P in ventricular tissue) and 6 min (SNP 100 uM in myocytes) without any effect on the positive inotropic effects of isoproterenol. Thus, the dissociation between mechanical effects and c G M P elevating actions by these agents described in the present study, and in various other reports, can be extended to the level of kinase activation. The absence of P K G activation by carbachol in both intact tissue and cardiomyocytes demonstrates a lack of evidence for subcellular compartmentalization of c G M P and does not support the hypothesis that selective activation of P K G by muscarinic receptor agonists mediates their negative inotropic effects, as was first proposed by Lincoln and Keely (1980, 1981). Activation of P K G may be reflected by not only the activity ratio, but also by changes in levels of total activity in various subcellular compartments. Data presented herein indicate that P K G is associated with both soluble and particulate (Triton X -100-159 extracted) subcellular fractions and it is possible that translocation of P K G between fractions may be necessary for the expression of the functional effects of c G M P -elevating agents. It has been observed in a study of PKG-mediated phosphorylation of vimentin, an intermediate filament subunit protein, that, in addition to c G M P elevation, agonist-stimulated co-localization of P K G with its substrate was required to observe a secretory response in human neutrophils (Wyatt et al., 1993; Pryzwansky et al., 1995). Interestingly, agonist-stimulated intracellular translocation of P K G was essential because activation of P K G by c G M P analogues, which did not stimulate translocation, was not sufficient for phosphorylation of substrates and expression of a functional effect. This suggests the necessity of another signalling event (possibly an increase in intracellular Ca 2 + ) in conjunction with elevation of c G M P for the mediation of P K G translocation and expression of biological activity. By extension, if translocation of P K G within intracellular compartments of ventricular tissue occurred in the presence of carbachol but not in the presence of SNP or A N P , one could ascribe a role for P K G in the functional effects of muscarinic receptor agonists specifically. However, under the conditions of this study, no evidence for translocation between soluble and particulate fractions was observed in intact ventricular tissues or isolated cardiomyocytes in the presence of carbachol, S N P or ANP, because total P K G activity in soluble and particulate fractions remained unchanged regardless of the treatment conditions. It should be noted that the procedure used in this study to prepare soluble and particulate fractions did not provide highly purified subcellular preparations. Thus, these experiments should be considered as preliminary and, as such, do not entirely rule out P K G translocation. More firm conclusions regarding movement of P K G to 160 specific subcellular locations await the evaluation of kinase activity in highly purified cytosolic and membrane preparations. 4.6 Relationship between contractility and PKG activation by cGMP analogues. Since S N P and A N P activate cardiac P K G but do not mediate a negative inotropic effect under the conditions of this study, agents which directly activate P K G , such as specific c G M P analogues, should also fail to alter ventricular contractility. As mentioned in the Introduction, the effects of c G M P analogues on ventricular contractility are not clear and some of the confusion may arise from the use of analogues which have PKG-independent effects. For example, the ability of the membrane permeant analogue dibutyryl-cGMP to mimic the inhibitory effects of acetylcholine on isoproterenol-stimulated ventricular contractility was considered strong evidence for a role for c G M P in the functional effects of muscarinic receptor agonists (Watanabe and Besch, 1975). However, dibutyryl-cGMP also inhibits phosphodiesterases (Komas et al., 1991), activates P K A (K a « 2 uM) more effectively than P K G (K a « 20 uM) (Francis et al., 1988), requires metabolism to A/2-monobutyryl-c G M P in order to modestly activate P K G (K a « 2 uM) (Francis et al., 1988) and releases free butyrate which may produce effects which are unrelated to P K G . Therefore, functional effects of dibutyryl-cGMP cannot be readily associated with P K G activation. The vast majority of studies which have investigated the effects of c G M P analogues on contractility have used 8-bromo-cGMP, which is an effective activator of P K G (K a for la P K G * 2.5 nM, K a for Ip P K G « 210 nM) (Sekhar et al., 1992). 161 However, 8-bromo-cGMP does not bind only to P K G ; it is a substrate (albeit a poor one) for three different phosphodiesterases (Butt et al., 1992), it inhibits c G M P -inhibited phosphodiesterase (K, « 8 pM) (Butt et al., 1992), and it activates c G M P -stimulated phosphodiesterase at concentrations of 1 - 100 pM (Komas et al., 1991). Moreover, 8-bromo-cGMP alters contractility in some non-cardiac muscles in a manner which does not reflect the effects of elevated endogenous c G M P . In the rat myometrium, for example, 100 pM 8-bromo-cGMP rapidly reduced tension but elevation of endogenous c G M P levels by several NO donors had no effect on tension (Diamond, 1983). Even 7.5-fold increases in intracellular c G M P levels by 0.5 mM S N P did not alter rat vas deferens contractility yet 100 pM 8-bromo-cGMP was an effective relaxant (Diamond and Janis, 1978; Diamond, 1983). Therefore, caution must be used in attributing the functional effects of c G M P analogues to activation of P K G . Effects of 8-bromo-cGMP on contractiity of the ventricle have been observed only in a few cases (Nawrath, 1976; Smith et al., 1991; Shah et al., 1991, 1994; Brady et al., 1993) and, when present, ranged from substantial (Nawrath, 1976) to minor (Shah et al., 1991, 1994; Smith et al., 1991). It is interesting to note that 8-bromo-c G M P was consistently without negative inotropic effects in ventricular preparations treated with cAMP-elevating agents (Endoh and Shimizu, 1979; Endoh and Yamashita, 1981; Hongo et al., 1993; Shah et al., 1991, 1994). Inhibition of cGMP-mediated negative inotropic effects by cAMP-elevating agents sharply contrasts with the requirements for muscarinic receptor agonist-mediated contractile actions and is consistent with competition between P K G and PKA for a common phosphorylation site. Cardiac phospholamban and troponin I are phosphorylated in vitro by P K G at the 162 same serine residues as can be phosphorylated by PKA (Lincoln and Corbin, 1978; Raeymakers et al., 1988). A contractile correlate of the phosphorylation of these proteins is accelerated relaxation. It is possible that the earlier onset and faster rate of relaxation which has been observed in the presence of S N P (Grocott-Mason et al., 1994; Paulus et al., 1994) and 8-bromo-cGMP (Shah et al., 1994) results from P K G -mediated phosphorylation of one or the other of these substrates. The magnitude of the relaxant effect may be dependent on the pre-existing activity of P K A in the tissue. Clearly, this is an incomplete explanation of the contractile effects of c G M P -elevating agents and c G M P analogues, in light of the fact that these agents inhibit cAMP-stimulated I C a in the mammalian ventricle (see MacDonald et al., 1994 and Sperelakis et al., 1994 for reviews). The work of Levi et al. (1989) and Mery et al. (1991, 1994) points to regulation of IC a by P K G in a manner opposite to, but dependent upon, PKA activity because SNP, 8-bromo-cGMP and catalytically active P K G (intracellularly perfused) inhibited IC a much more effectively in the presence of activated P K A than in its absence. Unlike the case for the putative PKG-mediated phosphorylation of phospholamban or troponin I, no functional equivalent exists for cGMP-mediated IC a inhibition because S N P and 8-bromo-cGMP have little effect on isoproterenol-stimulated force development. Thus, for reasons that are not clear, the electrophysiological effects of P K G in IC a do not result in corresponding negative inotropic effects in intact ventricular tissues or cardiomyocytes. Cross-activation of cyclic nucleotide-dependent protein kinases by c G M P and cAMP has been proposed as a mechanism for PKG-mediated smooth muscle relaxation by cAMP-elevating agents (Cornwell and Lincoln, 1989; Jiang et al., 1992) 163 and for PKA-stimulated chloride transport in colonic epithelial cells by cGMP-elevating agents (Forte et al., 1992). Relatively high concentrations of agonists were required for cross-activation, so it is not surprising that the low concentration of 1 nM isoproterenol used in the present study did not increase P K G activity in ventricular tissue in the present study. Concentrations of up to 100 pM S N P and 100 nM A N P were used in this study and one might propose that cross-activation of P K A countered a P K G -mediated negative inotropic effect by SNP and A N P . This is unlikely to be of practical importance due to the relatively low capacity of the ventricle to synthesize c G M P upon agonist stimulation. In tissues where cGMP-elevating agents cross-activated PKA, c G M P levels were elevated to the nmol/mg protein range, as in colonic epithelial cells where cellular c G M P content increased by 2,000-fold, from 0.77 pmol/mg protein to 1,598 pmol/mg protein in the presence of 1 pM heat-stable enterotoxin (Forte et al., 1992). In rat cardiomyocytes, S N P (100 pM, 6 min)-stimulated increases in c G M P levels were an order of magnitude less than control levels of c A M P so, considering that P K A exhibits a > 50-fold greater affinity for cAMP than for c G M P (Doskeland et al., 1983), cross-activation of PKA by S N P and A N P in cardiomyocytes is very unlikely. Furthermore, if S N P and A N P cross-activated PKA as well as activated P K G , the unlikely scenario of perfectly balanced functional effects of each kinase must be envisioned because a range of concentrations of these agonists neither directly increased nor decreased ventricular strip contractility. As mentioned earlier, observations of cGMP-associated enhancement of cAMP signalling has been ascribed to cGMP-mediated inhibition of cAMP phosphodiesterases (Ono and Trautwein, 1991; Klabunde et al., 1992; Kojda et al., 1996) but the lack of an effect of S N P or A N P on 164 isoproterenol-st imulated c A M P levels argues against such a mechan ism in the present study. 4.7 Possible role of enzymatically-released NO in muscarinic receptor-mediated negative inotropy. The three cGMP-e leva t i ng agents used in the present study increase c G M P by different mechan isms, and it is in these dif ferences that a hypothesis regarding the mechan ism for muscar in ic receptor agonist-mediated negative inotropy may be formulated. The two agonists which do not dec rease contractility, as measured in this study and others, or which may have minor negative inotropic or posit ive lusiotropic effects (Brady et al., 1993, Grocot t -Mason et al., 1994), either do not st imulate formation of N O (ANP) or do so in a non-enzymat ic, non-receptor-regulated manner ( S N P ) (Smith et al., 1990). O n the other hand, the agonist which has a marked negative inotropic effect increases N O through a M 2 -muscar in ic receptor-mediated mechan ism (Bal l igand et al., 1995). It is tempting to speculate that a crucial factor in muscar in ic receptor agonist-mediated negative inotropy is the re lease and act ion of N O in a strictly regulated manner (possibly through the containment of muscar in ic receptor-assoc ia ted N O to a speci f ic intracellular locale) and that elevat ion of c G M P by itself is insufficient for production of mechanica l effects. In freshly isolated rat cardiomyocytes, muscar in ic receptor agonis ts stimulate (by an unknown mechanism) constitutive, C a 2 + - d e p e n d e n t nitric oxide syn thase type III ( N O S III) which is primarily assoc ia ted with the particulate fraction (Bal l igand et al., 1995). Thus , muscar in ic receptor agonists may stimulate N O production by activation of N O S III local ized in particular subcel lular areas. It is difficult to fo resee, at first 165 glance, a containment of N O at its site of synthesis because a hallmark character ist ic of N O is that it rapidly and readily diffuses across membranes and its s ignal is dampened primarily by its brief half-life and its effects are not limited by solubility. However , it should be noted that ev idence for the extensive d ispers ion of N O has been accumulated from studies in t issues such as blood vesse ls and the brain (see Garthwaite, 1991 and Nathan, 1992 for reviews). The myocard ium differs from these t issues in that it contains high levels of myoglobin (Wittenberg, 1970) which forms a relatively stable adduct with N O (NO-heme) (Ignarro, 1991). Thus , myoglobin may represent a N O buffer within the cardiomyocyte (Ishibashi et al., 1993). Accord ingly , diffusion of N O away from specif ical ly local ized N O S III upon muscar in ic receptor stimulation may be spatially limited within the cardiomyocyte, such that it acts only near its site of formation. Similarly, a c c e s s of muscar in ic receptor-associated N O to cytosol ic guanylyl cyc lase may also be limited, which is consistent with the low eff icacy of muscar in ic receptor agonists in elevating c G M P levels. Target ing of N O S III to the particulate fraction is signif icant in v iew of the apparent NO-dependent inhibition of sarco lemmal ICa by muscar in ic receptor agonists. Bal l igand et al. (1995) recently suggested that inhibition of cAMP-s t imu la ted I C a by carbacho l in rat ventricular card iomyocytes w a s NO-dependent , because the effects of carbachol were reversed by L-A/-mono-methyl-arginine ( N M M A ) , a speci f ic N O S antagonist, and by hemoglobin, an N O scavenger . Methy lene blue and L Y 8 3 5 8 3 a lso inhibit the negative effect of carbachol on cAMP-s t imu la ted l C a in the adult rat and gu inea pig ventricular cardiomyocyte (Mubagwa et al., 1993; Levi et al., 1994). Methy lene blue and L Y 8 3 5 8 3 inhibit NO-media ted p rocesses by poorly def ined 166 mechanisms. LY83583 was initially considered to be a direct inhibitor of soluble guanylyl cyclase (Muisch et al., 1988) but it has also been shown to increase soluble guanylyl cyclase activity in the presence of S N P (Gupta et al., 1994b) and to modify the activity of other enzymes (Luond et al., 1993). LY83583 and methylene blue share the capacity to generate superoxide anion which quenches NO (Barbier and Lefebvre, 1992; Marczin et al., 1992). Superoxide anion (-0"2) chemically reacts with NO to form peroxynitrite and, thence, hydroxyl radical and - N 0 2 (Freeman, 1994) and, therefore, methylene blue and LY83583 can bind and destroy NO. Interpretation of the effects of these inhibitors should, therefore, include a role for NO and, perhaps only secondarily, for c G M P . The functional correlate of I C a inhibition by cholinomimetics, i.e., inhibition of the positive inotropic effects of cAMP-elevating agents, is also sensitive to antagonism by methylene blue and hemoglobin in rat ventricular cardiomyocytes (Balligand et al., 1995) and by LY83583 in rabbit papillary muscles (MacLeod and Diamond, 1986). The effects of the NOS inhibitor NMMA have been variously reported as inhibitory of the negative inotropic effects of cholinomimetics in.the in vivo canine heart (Hare et al., 1995) and without effect on the negative inotropic effects of carbachol in forskolin-stimulated rabbit papillary muscles (Hui et al., 1995). Observations that induction of cardiomyocyte NOS II activity by cytokines depresses cardiomyocyte contractile activity complements a postulated role for enzymatically generated NO in negative inotropic processes (Hung and Lew, 1993; Tao and McKenna, 1994). The exact spatial pattern of NO release from S N P is not known but is likely to be more diffuse than that by muscarinic receptor-mediated stimulation of NOS III and may not elevate NO in the required subcellular location to a sufficient concentration to 167 produce a negative inotropic effect. The greater efficacy of S N P in elevating c G M P in comparison to carbachol may mirror a more widespread access of the drug to the cardiomyocyte cytosol where it can release NO and thereby activate soluble guanylyl cyclase. However, in the absence of localized NO production, the c G M P elevation by itself may not consistently decrease contractile function. Inhibition of cAMP-stimulated IC a by S N P (Levi et al., 1994), A N P (Bkaily et al., 1993; Tohse et al., 1995), and 8-bromo-cGMP (Levi et al., 1991, 1994; Mery et al., 1991; Ono and Trautwein, 1991) suggests that c G M P can inhibit L-type calcium channels of the sarcolemma in a manner analogous to muscarinic receptor agonists under experimental conditions. The absence of a substantial negative inotropic effect by these agents supports the contention that other signalling events, such as enzymatically-generated NO, are required to translate cGMP-mediated inhibition of IC a into a negative inotropic effect. 168 5.0 SUMMARY AND CONCLUSIONS The principal observat ions of this study were as fol lows: 1. Carbacho l markedly depressed contractility in isoproterenol-st imulated perfused rat hearts, rat right ventricular strips and isolated rat card iomyocytes. It did not significantly change c G M P levels in intact ventricular t issues and increased c G M P levels to a relatively smal l extent in isolated ventricular card iomyocytes. T h e negative inotropic effects of carbachol a lso occurred in the absence of signif icant changes in P K G activity and c A M P levels in ventricular t issues and card iomyocytes. 2. Total ventricular c G M P levels and P K G activity were elevated by S N P in the a b s e n c e of changes in contractility in isoproterenol-st imulated intact hearts and isoproterenol-st imulated ventricular cardiomyocytes. 3. Similarly to S N P , total ventricular c G M P levels and P K G activity were elevated by A N P in the absence of changes in contractility in isoproterenol-st imulated intact hearts. A N P a lso activated P K G in isoproterenol-st imulated ventricular card iomyocytes. 4. Neither carbachol , S N P , nor A N P altered the ability of isoproterenol to e levate c A M P levels in intact ventricular t issues. Isoproterenol-st imulated c A M P elevat ion was a lso unchanged by carbachol and S N P in isolated card iomyocytes. The necessi ty for c G M P elevation and P K G activation in muscar in ic receptor agonist -mediated negative inotropy cannot be ruled out because smal l , albeit somet imes insignificant, increases in c G M P were noted in the ventr icular preparat ions used in the present study and activation of P K G may have occurred but possib ly was 169 not detected due to technical limitations of the assay. However, it is clear that a cGMP/PKG signalling mechanism is insufficient to mediate a negative inotropic effect in the ventricle because S N P and A N P increased c G M P levels and P K G activity to greater extents than did carbachol, yet did not decrease contractility. Thus, these results do not confirm the hypothesis proposed by Lincoln and Keely (1980, 1981) that an apparent compartmentalization process mediates the selective activation of P K G by specific cGMP-elevating agents which have negative inotropic capabilities. Further studies into the role of other signalling processes by muscarinic receptor agonists, such as NO produced specifically by NOS, may shed light on the mechanism of muscarinic receptor-mediated negative inotropy in the mammalian myocardium. 170 6.0 BIBLIOGRAPHY Ahmad, Z., F. J . Green, H. S. Subuhi, and A. M. Watanabe. Autonomic regulation of type 1 protein phosphatase in cardiac muscle. J . Biol. Chem. 264: 3859-3863, 1989. Aitken, A., T. Bilham, P. Cohen, D. Aswad, and P. Greengard. A specific substrate from rabbit cerebellum for guanosine-3':5'-monophosphate-dependent protein kinase. Ill Amino acid sequences at the two phosphorylation sites. J . Biol. Chem. 256: 3501-3506, 1981. Anand-Srivastava, M. B., and M. Cantin. Atrial natriuretic factor receptors are negatively coupled to adenylate cyclase in cultured atrial and ventricular myocytes. Biochem. Biophys. Res. Commun. 138: 427-436, 1986. Anand-Srivastava, M. B., and G. J . Trachte. Atrial natriuretic factor receptors and signal transduction mechanisms. Pharmacol. Rev. 45: 445-497, 1993. Balligand, J - . L , R. A. Kelly, P. A. Marsden, T. W. Smith, and T. Michel. Control of cardiac muscle cell function by an endogenous nitric oxide signaling system. Proc. Natl. Acad. Sci. U.S.A. 90: 347-351, 1993. Balligand, J - . L , L. Kobzik, X. Han, D. M. Kaye, L. Belhassen, D. S. O'Hara, R. A. Kelly, T. W. Smith, and T. Michel. Nitric oxide-dependent parasympathetic signaling is due to activation of constitutive endothelial (type III) nitric oxide synthase in cardiac myocytes. J . Biol. Chem. 270: 14582-14586, 1995. Barbier, A. J . M., and R. A. Lefebvre. Effect of LY83583 on relaxation induced by non-adrenergic non-cholinergic nerve stimulation and exogenous nitric oxide in the rat gastric fundus. Eur. J . Pharmacol. 219: 331-334, 1992. Bartel, S. , P. Karczewski, and E. G. Krause. Protein phosphorylation and cardiac function: cholinergic-adrenergic interaction. Cardiovasc. Res. 7: 1948-1953, 1993. Baum, T., E. J . Sybertz, R. W. Watkins, S . Nelson, W. Coleman, K. K. Pula, N. Prioli, M. Rivelli, and A. Grossman. Hemodynamic actions of a synthetic atrial natriuretic factor. J . Cardiovasc. Pharmacol. 8: 898-905, 1986. Baumner, D., and H. Nawrath. Effects of inhibitors of cGMP-dependent protein kinase on atrial heart and aortic smooth muscle from rats. Eur. J . Pharmacol. 273: 295-298, 1995. Bennett, B. M, B. J . McDonald, R. Nigam, and W. C. Simon. Biotransformation of organic nitrates and vascular smooth muscle cell function. Trends Pharmcol. Sci. 15: 245-249, 1994. 171 Bianchi, C , J . Gutkowska, G. Thibault, R. Garcia, J . Genest, and M. Cantin. Radioautographic localization of 1 2 5 l -atrial natriuretic factor (ANF) in rat tissues. Histochem. 82: 441-452, 1985. Biel, M., X. Zong, M. Distler, E. Bosse, N. Klugbauer, M. Murakami, V. Flockerzi and F. Hofmann. Another member of the cyclic nucleotide-gated channel family, expressed in testis, kidney, and heart. Proc. Natl. Acad. Sci. USA 91: 3505-3509, 1994. Bkaily, G., N. Perron, S. Wang, A. Sculptoreanu, D. Jacques and D. Menard. Atrial natriuretic factor blocks the high threshold C a 2 + current and increases K + current in fetal single ventricular cells. J . Mol. Cell. Cardiol. 25: 1305-1316, 1993. Bohm, M. F. Diet, B. Pieske, and E. Erdmann. h-ANF does not play a role in the regulation of myocardial force of contraction. Life Sci. 43: 1261-1267, 1988. Boyd, R. S. , L. E. Donnelly, J . R. Allport, and J . MacDermont. Sodium nitroprusside promotes N A D + labelling of a 116 kDa protein in NG108-15 cell homogenates. Biochem. Biophys. Res. Commun. 197: 1277-1282, 1993. Bradford, M. M. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248-254, 1976. Brady, A. J . B., J . B. Warren, P. A. Poole-Wilson, T. J . Williams, and S. E. Harding. Nitric oxide attenuates cardiac myocyte contraction. Am. J . Physiol. 265 (Heart Circ. Physiol. 34): H176-H182, 1993. Brooker, G. Dissociation of cyclic G M P from the negative inotropic action of carbachol in guinea pig atria. J . Cyclic. Nucleotide Res. 3: 407-413, 1977. Brune, B., S. Dimmeler, L. M. Vedia, and E. G. Lapetina. Nitric oxide: a signal for ADP-ribosylation of proteins. Life Sci. 54: 61-70, 1993. Brune, B. and E. G. Lapetina. Activation of a cytosolic ADP-ribosyltransferase by nitric oxide-generating agents. J . Biol. Chem. 264: 8455-8458, 1989. Bugaisky, L. B. Isolation and culture of human adult cardiac myocytes. In: Biology of Isolated Adult Cardiac Myocytes, edited by W. A. Clark, R. S. Decker, and T. K. Borg. New York: Elsivier, 1988, pp. 272-275. Butt, E., M. Eigenthaler, and H-. G. Genieser. (Rp) -8-pCPT-cGMPS, a novel c G M P -dependent protein kinase inhibitor. Eur. J . Pharm. 269: 265-268, 1994. Butt, E., C. Nolte, S. Schulz, J . Beltman, J . A. Beavo, B. Jastroff, and U. Walter. Analysis of the functional role of cGMP-dependent protein kinase in intact human 172 platelets using a specific activator 8-para-chlorophenylthio-cGMP. Biochem. Pharmacol. 43: 2591-2600, 1992. Buxton I. L. O., and L. L. Brunton. Compartments of cAMP and protein kinase in mammalian cardiomyocytes. J . Biol. Chem. 258: 10233-10239, 1983. Cantin M. and J . Genest. The heart and the atrial natriuretic factor. Endocrine Rev. 6: 107-127, 1985. Colbran, J . L , S. H. Francis, A. B. Leach, M. K. Thomas, H. Jiang, L. M. McAllister, and J . D. Corbin. A phenylalanine in peptide substrates provides for selectivity between c G M P - and cAMP-dependent protein kinases. J . Biol. Chem. 267: 9589-9594, 1992. Corbin, J . , and S. O. Doskeland, Studies of the two different intrachain cGMP-binding sites of cGMP-dependent protein kinase. J . Biol. Chem. 258: 11391-11397, 1983. Cornwell, T. L. and T. M. Lincoln. Regulation of intracellular C a 2 + levels in cultured vascular smooth muscle cells. Am. J . Physiol. 264: 1146-1155, 1989. Cornwell, T. L, K. B. Pryzwansky, T. A. Wyatt, and T. M. Lincoln. Regulation of sarcoplasmic reticulum protein phosphorylation by localized cyclic GMP-dependent protein kinase in vascular smooth muscle cells. Mol. Pharmacol. 40: 923-931, 1991. Cramb, G., R. Banks, E. L. Rugg, and J . F. Aiton. Actions of atrial natriuretic peptide (ANP) on cyclic nucleotide concentrations and phosphatidylinositol turnover in ventricular myocytes. Biochem. Biophys. Res. Commun 148: 962-970, 1987. Criscone, L. , R. Burdet, H. Hanni, B. Kamber, A. Truog, and K. G. Hofbauer. Systemic and regional hemodynamic effects of atriopeptin II in anesthetized rats. J . Cardiovasc. Pharmacol. 9: 135-141,1987. DeGeest, H., Levy, M. N., Zieske, H., and R. I. Lipman. Depression of ventricular contractility by stimulation of the vagus nerve. Circ. Res. 17: 222-235, 1965. DeJonge, H. R. Cyclic GMP-dependent protein kinase in intestinal brush borders. Adv. Cyclic Nucleotide Res. 14: 315-333, 1981. Diamond, J . Lack of correlation between cyclic G M P elevation and relaxation of nonvascular smooth muscle by nitroglycerin, nitroprusside, hydroxylamine and sodium azide. J . Pharmacol. Exp. Ther. 225: 422-426, 1983. Diamond, J . , and E. B. Chu. A novel cyclic GMP-lowering agent, LY83583, blocks carbachol-induced cyclic G M P elevation in rabbit atrial strips without blocking the negative inotropic effects of carbachol. Can. J . Physiol. Pharmacol. 63: 908-911, 1985. 173 Diamond, J . , and R. A. Janis. Increases in cyclic G M P levels may not mediate relaxant effects of sodium nitroprusside, verapamil and hydralazine in rat deferens. Nature (London) 271: 472-473, 1978. Diamond, J . , R. E. Ten Eick, and A. J . Trapani. Are increases in cyclic G M P levels responsible for the negative inotropic effects of acetylcholine in the heart? Biochem. Biophys. Res. Commun. 79: 912-918, 1977. Distler, M., M. Biel, V. Flockerzi and F. Hofmann. Expression of cyclic nucleotide-gated cation channels in non-sensory tissues and cells. Neuropharmacol. 33: 1275-1282, 1994. Doskeland, S . O., D. 0greid, R. Ekanger, P. A. Sturm, J . P. Miller, and R. H. Suva. Mapping of the two intrachain cyclic nucleotide binding sites of adenosine cyclic 3', 5' -phosphate dependent protein kinase I. Biochem. 22:1094-1101, 1983. Du, X. Y., R. G. Schoemaker, E. Bos, and P. Saxena. Different pharmacological responses of atrium and ventricle: studies with human cardiac tissue. Eur. J . Pharmacol. 259: 173-180, 1994. Du, X. Y., R. G. Schoemaker, E. Bos, and P. Saxena. Characterization of the positive and negative effects of acetylcholine in the human myocardium. Eur. J . Pharmacol. 284: 119-127, 1995. Dubois-Rande, J . L., S . Adnot, C. Benvenuti, P. Merlet, L. Hittinger, S . Sediame, E. Chabrier, P. Braquet, and A. Castaigne. Hemodynamic response to intracoronary infusion of atrial natriuretic factor in patients with normal or altered left ventricular function. J . Cardiovasc. Pharmacol. 17: 608-614, 1991. Ecker, T., C. Gobel, R. Hullin, R. Rettig, G. Seitz, and F. Hoffman. Decreased cardiac concentration of c G M P kinase in hypertensive animals: an index for cardiac vascularization? Circ. Res. 65: 1361-1369, 1989. England, P. J . , and M. Shahid. Effects of forskolin on contractile responses and protein phosphorylation in the isolated perfused rat heart. Biochem. J . 246: 687-695, 1987. Eisenthal, R. and M. J . Danson (eds) Enzyme Assays: A Practical Approach. Oxford: IRL Press, 1993, pp. 109-111. Endoh, M. Correlation of cyclic A M P and cyclic G M P levels with changes in contractile force of dog ventricular myocardium during cholinergic antagonism of positive inotropic actions of histamine, glucagon, theophylline and papaverine. Japan. J . Pharmacol. 29: 855-864, 1979. 174 Endoh, M. The effects of various drugs on the myocardial inotropic response. Gen. Pharmacol. 26: 1-31, 1995. Endoh, M. and M. Honma. Effects of papaverine and its interaction with isoproterenol and carbachol on the contractile force and cyclic nucleotide levels of the canine ventricular myocardium. Naunyn-Schmiedeberg's Arch. Pharmacol. 306: 241-248, 1979. Endoh, M, and T. Shimizu. Failure of dibutyryl and 8-bromo-cyclic G M P to mimic the antagonistic action of carbachol on the positive inotropic effects of sympathetic amines in the canine isolated ventricular myocardium. Japan J . Pharmacol. 29: 423-433, 1979. Endoh, M. and S. Yamashita. Differential responses to carbachol, sodium nitroprusside and 8-bromo-guanosine 3',5'-monophosphate of canine atrial and ventricular muscle. Br. J . Pharmacol. 73: 393-399, 1981. Fawzi, A. Role of superficial calcium binding sites in the inotropic responses of isoproterenol and oubain. Ph. D. thesis, University of British Columbia, 1985. Fischmeister, R. and H. C. Hartzell. Cyclic guanosine 3',5'-monophosphate regulates the calcium current in single cells from frog ventricle. J . Physiol. 387: 453-472, 1987. Fischmeister, R., and A. Schrier. Interactive effects of isoproterenol, forskolin and acetylcholine on C a 2 + current in frog ventricular myocytes. J . Physiol. 417: 213-239, 1989. Fiscus, R. R., and F. Murad. cGMP-dependent protein kinase activation in intact tissues. Methods Enzymol. 159: 150-159, 1988. Fiscus, R. R., T. Torphy, and S. E. Mayer. Cyclic GMP-dependent protein kinase activation in canine tracheal smooth muscle by methacholine and sodium nitroprusside. Biochim. Biophys. Acta 805: 382-292, 1984. Fiscus, R. R., R. M. Rapoport, S . A. Waldman, and F. Murad. Atriopeptin II elevates cyclic G M P , activates cyclic GMP-dependent protein kinase and causes relaxation in rat thoracic aorta. Biochim. Biophys. Acta 846: 179-184, 1985. Flitney, F.W. and J . Singh. Evidence that cyclic G M P may regulate cyclic A M P metabolism in the isolated frog ventricle. J . Mol. Cell. Cardiol. 13: 963-79, 1981. Flockerzi, V., N. Speichermann, and F. Hofmann. A guanosine 3':5'-monophosphate-dependent protein kinase from bovine heart muscle. J . Biol. Chem. 253: 3395-3399, 1978. 175 Fort, S. and M. J . Lewis. Regulation of myocardial contractile performance by sodium nitroprusside in the isolated perfused heart of the ferret. Br. J . Pharmacol. 102: 351P, 1991 (Abstract). Forte, L. R., P. K.Thorne, S. L. Eber, W. J . Krause, R. H. Freeman, S. H. Francis, and J . D. Corbin. Stimulation of intestinal Cl" transport by heat-stable enterotoxin: activation of cAMP-dependent protein kinase by cGMP. Am. J . Physiol. (Cell Physiol. 32): C607-C615, 1992. Francis, S . H. and J . D. Corbin. Structure and function of cyclic nucleotide-dependent protein kinases. Annu. Rev. Physiol. 56: 237-272, 1994. Francis, S . H., B. D. Noblett, B. W. Todd, J . N. Wells, and J . D. Corbin. Relaxation of vascular and tracheal smooth muscle by cyclic nucleotide analogs that preferentially activate purified cGMP-dependent protein kinase. Mol. Pharmacol. 34: 506-517, 1988. Freeman, B. Free radical chemistry of nitric oxide: looking at the dark side. Chest 105: 79S-84S, 1994. Garthwaite, J . Glutamate, nitric oxide and cell-cell signalling in the nervous system. Trends Neurolog. Sci. 14: 60-67, 1991. Garvey, J . L., E. G. Kranias, and R. J . Solaro. Phosphorylation of C-protein, troponin I and phospholamban in isolated rabbit hearts. Biochem. J . 249: 709-714, 1988. George, W. J . , Ignarro, L. J . , Paddock, R. J . , White, L. A., and P. J . Kadowitz. Oppositional effects of acetylcholine and isoproterenol in isometric tension and cyclic nucleotide concentration in rabbit atria. J . Cyclic Nucleotide Res. 1: 339-347, 1975. George, W. J . , J . B. Poison, A. G. OToole, and N. D. Goldberg. Elevation of guanosine 3',5'-cyclic phosphate in rat heart after perfusion with acetylcholine. Proc. Natl. Acad, Sci. U.S.A. 67: 305-312, 1970. George, W. J . , R. D. Wilkerson, and P. J . Kadowitz. Influence of acetylcholine on contractile force and cyclic nucleotide levels in the isolated perfused rat heart. J . Pharmacol. Exp. Ther. 184: 22-235, 1972. Gilman, A. G proteins: transducers of receptor-generated signals. Annu. Rev. Biochem. 56: 615-649, 1987. Goldberg, N. D., M. K. Haddox, S. E. Nichol, D. B. Glass, C. H. Sanford, F. A. Kuehl, Jr., and E. Estensen. Biologic regulation through opposing influences of cyclic G M P and cyclic A M P : the yin-yang hypothesis. Adv. Cyclic Nucleotide Res. 5: 307-330, 1975. 176 Grocott-Mason, R., S. Fort, M. J . Lewis, and A. M. Shah. Myocardial relaxant effect of exogenous nitric oxide in isolated ejecting hearts. Am. J . Physiol. 266 (Heart Circ. Physiol. 35): H1699-H1705, 1994. Gupta, R. C., J . Neumann, P. Boknik, and A. M. Watanabe. M 2 -specific muscarinic cholinergic receptor-mediated inhibition of cardiac regulatory protein phosphorylation. Am. J . Physiol. 266 (Heart Circ. Physiol. 35): H1138-H1144, 1994a. Gupta, R. C , J . Neumann, and A. M. Watanabe. Comparison of adenosine and muscarinic receptor-mediated effects on protein phosphatase inhibitor-1 activity in the heart. J . Pharmacol. Exper. Ther. 266: 16-22, 1993. Gupta, S. , C. McArthur, C. Grady, and N. B. Ruderman. Stimulation of vascular N a + -K + -ATPase activity by nitric oxide: a cGMP-independent effect. Am. J . Physiol. 266 (Heart Circ. Physiol. 35): H2146 - H2151, 1994b. Haddad, J . , M. L. Decker, L. C. Hsieh, M. Lesch, A. M. Samarel, and R. S. Decker. Attachment and maintenance of adult rabbit cardiac myocytes in primary cell culture. Am. J . Physiol. 255 (1 Part 1): C19-C27, 1988. Harding, S. E., G. Vescovo, M. Kirby, S. M. Jones, J . Gurden, P. A. Poole-Wilson. Contractile responses of isolated adult rat and rabbit cardiac myocytes to isoproterenol and calcium. J . Mol. Cell. Cardiol. 20: 635-647, 1988. Hare, J . M., J . F. Keaney, J . L. Balligand, Loscalzo, and T. W. Smith. Role of nitric oxide on parasympathetic modulation of p-adrenergic myocardial contractility in normal dogs. J . Clin. Invest. 95: 360-366, 1995. Hazeki, O., and M. Ui. Modification by islet-activating protein receptor-mediated regulation of cyclic A M P accumulation in isolated rat heart cells. J . Biol. Chem. 256:2856-2862,1981. Heschler, J . , M. Kameyama, and W. Trautwein. On the mechanism of muscarinic inhibition of the cardiac Ca current. Pfluegers Arch. 407: 182-189, 1986. Hirata, Y., T. Serizawa, O. I. Kohmoto, T. Sugimoto, H. Matsuoka, M. lizuka, M. Ishii, T. Sugimoto, A. Miyata, K. Kangawa, and H. Matsuo. Estimation of the secretion rate of atrial natriuretic peptide from the coronary sinus in coronary artery disease. Am. J . Cardiol. 62: 56-58, 1988. Hiwatari, M., K. Satoh, J . A. Angus, and C. I. Johnston. No effect of atrial natriuretic factor on cardiac rate, force and transmitter release. Clin. Exper. Pharmacol. Physiol. 13: 163-168, 1986. 177 Hofmann, F., W. Dostmann, A. Keilbach, W. Landgraf, and P. Ruth. Structure and physiological role of cGMP-dependent protein kinase. Biochim. Biophys. Acta 1135: 51-60, 1992. Hongo, K., E. Tanaka, S. Kurihara. Mechanism of the effects of acetylcholine on the contractile properties and C a 2 + transients in ferret ventricular muscles. J . Physiol. 461: 185-199, 1993. Huggins, J . P., E. A. Cook, J . R. Pigott, T. J . Mattinsley, and P. J . England. Phospholamban is a good substrate for cGMP-dependent protein kinase in vitro, but is not in intact cardiac and smooth muscle. Biochem. J . 260: 829-835, 1989. Hui, J . , A. Tabatabaei, and K. M. MacLeod. L-NMMA blocks carbachol-induced increases in c G M P levels but not decreases in tension in the presence of forskolin in rabbit papillary muscles. Cardiovasc. Res. 30: 372-376, 1995. Hung, J . , and W. Y. Lew. Cellular mechanisms of endotoxin-induced myocardial depression in rabbits. Circ. Res. 73: 125-134, 1993. Hotter, J . H. Action of synthetic atrial natriuretic factor on contractility and coronary perfusion in isolated working rat hearts. Eur. J . Pharmacol. 193: 127-129, 1991. Ignarro, L. J . Heme-dependent activation of guanylate cyclase by nitric oxide: a novel signal transduction mechanism. Blood Vessels 28: 67-73, 1991. Inui, J . , O. -E . Brodde, and H. J . SchOmann. Influence of acetylcholine on the positive inotropic effect evoked by a- or p-adrenoceptor stimulation in the rabbit heart. Naunyn-Schmiedeberg's Arch. Pharmacol. 320:152-159,1982. Ishibashi, T., K. Hamaguchi, T. Kawada, H. Ohta, H. Sage, and S. Imai. Relationship between myoglobin contents and increases in cyclic G M P produced by glyceryl trinitrate and nitric oxide in rabbit aorta, right atrium and papillary muscle. Naunyn-Schmied. Arch. Pharmacol. 347: 553-561, 1993. Jakob, H., Oelert, H., Rupp, J . , and H. Nawrath. Functional role of cholinoceptors and purinoceptors in human isolated atrial and ventricular heart muscle. Br. J . Pharmacol. 97: 1199-1208, 1989. Jiang, H., J . L. Colbran, S. H. Francis, and J . D. Corbin. Direct evidence for cross-activation of cGMP-dependent protein kinase by cAMP in pig coronary arteries. J . Biol. Chem. 267: 1015-1019, 1992. Joyce, N. C , P. DeCamilli, and J . Boyles. Pericytes, like vascular smooth muscle cells, are immunocytochemically positive for cyclic GMP-dependent protein kinase. Microvasc. Res. 28: 206-219, 1984. 178 Kase, H., K. Iwahashi, S . Nakanishi, Y. Matsuda, K. Yamada, M. Takahashi, C. Murakata, A. Sato, and M. Kaneko. K-252 compounds, novel and potent inhibitors of protein kinase C and cyclic nucleotide-dependent protein kinases. Biochem. Biophys. Res. Commun. 142: 436-440, 1987. Katano, Y. and M. Endoh. Cyclic A M P metabolism in intact rat ventricular cardiac myocytes: interaction of carbachol with isoproterenol and 3-isobutyl-1-methylxanthine. Mol. Cell. Biochem. 119: 195-201, 1993. Katsuki, S. , Arnold, W. P., and F. Murad. Effects of sodium nitroprusside, nitroglycerin and sodium azide on levels of cyclic nucleotides and mechanical activity of various tissues. J . Cyclic Nucleotide. Res. 3: 239-247, 1977. Kaupp, U. B. Family of cyclic nucleotide gated ion channels. Curr. Opinion Neurobiol. 5: 434-442, 1995. Keely S. and T. M. Lincoln. On the question of cyclic G M P as the mediator of the effects of acetylcholine on the heart. Biochim. Biophys. Acta 543: 252-257, 1978. Keely, S. L., T. M. Lincoln, and J . D. Corbin. Interaction of acetylcholine and epinephrine on heart cyclic AMP-dependent protein kinase. Am. J . Physiol. 234: H432-H438, 1978. Keilbach, A., P. Ruth, and F. Hofmann. Detection of c G M P dependent protein kinase isozymes by specific antibodies. Eur. J . Biochem. 208: 467-473, 1992. Klabunde, R.E., N. D. Kimber, J . E. Kuk, M. C. Helgren, and U. Forstermann. Methyl-L-arginine decreases contractility, cGMP, and cAMP in isoproterenol-stimulated rat hearts in vitro. Eur. J . Pharmacol. 22: 1-7, 1992. Kohlhardt, M., and K. Haap. 8-Bromo-guanosine-3',5'-monophosphate mimics the effect of acetylcholine on slow response action potential and contractile force in mammalian atrial myocardium. J . Mol. Cell. Cardiol. 10: 72-74, 1978. Kojda, G., K. Kottenburg, P. Nix, K. D. Schulter, H. M. Piper, and E. Noak. Low increase in c G M P induced by organic nitrates and nitrovasodilators improves contractile response of rat ventricular myocytes. Circ. Res. 78: 91-101, 1996. Komas, N., A. Le Bee, J . C. Stoclet, and C. Lugnier. Cardiac cGMP-stimulated cyclic nucleotide phosphodiesterases: effects of c G M P analogues and drugs. Eur. J . Pharmacol. (Mol. Pharmacol) 206: 5-13, 1991. Korth, M., and V. Kulhkamp. Muscarinic receptors mediate negative and positive inotropic effects in mammalian ventricular myocardium: differentiation by agonists. Br. J . Pharmacol. 90: 81-90, 1987. 179 Koumi, S. , J . A. Wasserstrom, and R. E. Ten Eick. B-Adrenergic and cholinergic modulation of the inwardly-rectifying K + current in guinea pig ventricular myocytes. J . Physiol. 486: 647-659, 1995. Kryski, A., Jr., K. A. Kenno, and D. L. Severson. Stimulation of lipolysis in rat heart myocytes by isoproterenol. Am J . Physiol. 248: H208-H216, 1985. Kubalak, S. W., W. H. Newman and J . G. Webb. Differential effect of pertussis toxin on adenosine and muscarinic inhibition of cyclic A M P accumulation in canine ventricular myocytes. J . Mol. Cell. Cardiol. 23: 199-205, 1991. Kuo J . F., and P. Greengard. Cyclic nucleotide-dependent protein kinases. IV. Isolation and partial purification of a protein kinase activated by guanosine 3',5'-monophosphate. J . Biol. Chem. 245: 2493-2498, 1970. Kuo, J . F. and P. Greengard. Cyclic nucleotide-dependent protein kinases. J . Biol. Chem. 245: 2493-2498, 1970. Laemmli, R. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685, 1970. Landzberg, J . S. , Parker, J . D., Gauthier, D. F., and W. S. Colucci. Effects of intracoronary acetylcholine and atropine on basal and dobutamine-stimulated left ventricular contractility. Circ. 89: 164-168, 1994. Langendorff, O. Untersuchungen am uberlebenden saugetierherzen. Pfluegers Arch, ges Physiol. 61: 291, 1895. Langendorff, O. Untersuchungen am uberlebenden saugetierherzen. II. Abhandlung. Uber den einfluss von warme und calte auf das herz der warmblutigen tiere. Pfluegers Arch, ges Physiol. 66: 355, 1897. Langlands, J . M., and J . Diamond. Translocation of protein kinase C in bovine tracheal smooth muscle strips: the effect of methacholine and isoprenaline. Eur. J . Pharmacol. 227: 131-138, 1992. Laustiola K., P. Vuorinen, H. Vapautalo, and T. Metsa Ketela. S N P inhibits lactate formation in rat atria: is c G M P involved? Acta Pharmacol. Toxicol. Copenh. 52: 195-200, 1983. Levi, R. C , G. Alloatti, and R. Fischmeister. Cyclic G M P regulates the Ca-channel current in guinea pig ventricular myocytes. Pfluegers Arch. 413: 685-687, 1989. Levi, R. C , G. Alloatti, C. Penna, and M. P. Gallo. Guanylate-cyclase-mediated inhibition of cardiac l C a by carbachol and sodium nitroprusside. Pfluegers Arch. 426: 419-426, 1994. 180 Levy, M. N. Sympathetic-parasympathetic interactions in the heart. Circ. Res. 29:437-445, 1971. Lincoln, T. M. cGMP-dependent protein kinase. Meth. Enzymol. 99: 62-71, 1983. Lincoln, T. M., and J . D. Corbin. Purified cyclic GMP-dependent protein kinase catalyzes the phosphorylation of cardiac troponin inhibitory subunit (TN-I). J . Biol. Chem. 253: 337-339, 1978. Lincoln, T. M., and T. L. Cornwell. Intracellular cyclic G M P receptor proteins. F A S E B J . 7: 328-338, 1993. Lincoln, T. M., C. L. Hall, C. R. Park, and J . D. Corbin. Guanosine 3':5'-cyclic monophosphate binding proteins in rat tissues. Proc. Natl. Acad. Sci. U.S.A. 73: 2559-2563, 1976. Lincoln, T. M. and S. L. Keely. Effects of acetylcholine and nitroprusside on cyclic GMP-dependent protein kinase in the perfused heart. J . Cyclic Nuc. Res. 6: 83-91, 1980. Lincoln, T. M, and S. L. Keely. Regulation of cardiac cyclic GMP-dependent protein kinase. Biochim. Biophys. Acta 676: 230-244, 1981. Lincoln, T. M., M. Thompson, and T. L. Cornwell. Purification and characterization of two forms of cyclic GMP-dependent protein kinase from bovine aorta. J . Biol. Chem. 263: 17632-17637, 1988. Lindemann J . P, and A. M. Watanabe. Muscarinic cholinerigic inhibition of p-adrenergic stimulation of phospholamban phosphorylation and C a 2 + transport in guinea pig ventricles. J . Biol. Chem. 260: 13122-13129, 1985. Linden, J . , and G. Brooker. The questionable role of cyclic guanosine 3':5'-monophosphate in heart. Biochem. Pharmacol. 28: 3351-3360, 1979. Loffelholz, K., and A. J . Pappano. The parasympathetic neuroeffector junction of the heart. Pharmacol. Rev. 37: 1-24, 1985. Lohmann, S. M., R. Fischmeister, and U. Walter. Signal transduction by c G M P in heart. Basic. Res. Cardiol. 86: 503-514, 1991. Lowry, O. H., N. J . Rosebrough, A. L. Farr, and R. J . Randall. Protein measurement with the phenol reagent. J . Biol. Chem. 193: 265-275, 1951. Luo, W., I. L. Grupp, J . Harrer, S. Ponniah, G. Grupp, J . J . Duffy, T. Doetschman, and E. G. Kranias. Targeted ablation of the phospholamban gene is associated with 181 markedly enhanced myocardial contractility and loss of p-agonist stimulation. Circ. Res. 75: 401-409, 1994. Luond R. M., J . H. McKie, and K. T. Douglas. A direct link between LY83583, a selective repressor of cyclic G M P formation and glutathione metabolism. Biochem. Pharmacol. 45: 2547-9, 1993. MacDonald, T. F., S. Pelzer, W. Trautwein, and D. J . Pelzer. Regulation and modulation of calcium channels in cardiac, skeletal and smooth muscle cells. Physiol. Rev. 74: 365-507, 1994. MacKenzie III, C. W. Bovine lung cyclic GMP-dependent protein kinase exhibits two types of specific cyclic GMP-binding sites. J . Biol Chem. 257: 5589-5593, 1982. MacLeod, K. M. Adrenergic-cholinergic interactions in left atria: interaction of carbachol with alpha- and beta-adrenoceptor agonists. Can. J . Physiol. Pharmacol. 64: 597-601, 1986. MacLeod, K. M., and J . Diamond. Effects of the cyclic G M P lowering agent LY83583 on the interactions of carbachol with forskolin in rabbit isolated cardiac preparations. J . Pharmacol Exper. Ther. 238: 313-318, 1986. Marczin, N., U. S. Ryan, and J . D. Catravas. Methylene blue inhibits nitrovasodilator-and endothelium-derived relaxing factor-induced cyclic G M P accumulation in cultured pulmonary arterial smooth muscle cells via generation of superoxide. J . Pharmacol. Exper. Ther. 263: 170-179, 1992. Mayer, S. E. Neurohumoral transmission and the autonomic nervous system. In: The Pharmacological Basis of Therapeutics, 6 t h Ed., edited by A. G. Gilman, L. S . Goodman and A. Gilman, New York: MacMillan, 1980, pp. 62-63. McCall , D., and T. A. Fried. Effect of atriopeptin II on Ca influx, contractile behavior and cyclic nucleotide content of cultured neonatal rat myocardial cells. J . Mol. Cell. Cardiol. 22: 201-212, 1990. McCartney, S. , J . F. Aiton, and G. Cramb. Characterization of atrial natriuretic peptide receptors in bovine ventricular sarcolemma. Biochem. Biophys. Res. Commun. 167: 1361-1368, 1990. McCune P. W., and G. M. Gill. Positive cooperativity in guanosine 3':5'-monophosphate binding to guanosine 3':5'-monophosphate-dependent protein kinase. J . Biol. Chem 254: 5083-5091, 1979. Mclvor, M. E., C. H. Orchard, and E. G. Lakatta. Dissociation of changes in apparent myofibrillar C a 2 + sensitivity and twitch relaxation induced by adrenergic and cholinergic stimulation in isolated ferret cardiac muscle. J . Gen. Physiol. 92: 509-529, 1988. 182 Mery, P. F., Lohmann, S. M., Walter, U, and R. Fischmeister. C a 2 + current is regulated by cyclic GMP-dependent protein kinase in mammalian cardiac myocytes. Proc. Natl. Acad. Sci. USA 88: 1197-1201, 1991. Meulemans, A. L , K. R. Sipido, S. U. Sys, and D. L. Brutsaert. Atriopeptin III induces early relaxation of isolated mammalian papillary muscle. Circ. Res. 62: 1171-1174, 1988. Mikoluc B., and R. J . Wisniewska. The effect of C-terminal fragment of ANF-ANF(24-28)OH on the cardiovascular system in rat. Gen. Pharmacol. 25: 661-665, 1994. Mubagwa, K., Shirayama, T., Moreau, M., and A. J . Pappano. Effects of P D E inhibitors and carbachol on the L-type Ca Current in guinea pig ventricular myocytes. Am. J . Physiol. 264 (Heart Circ. Physiol. 33): H1353-H1363, 1993. Mulsch, A., R. Busse, S. Liebau, and U. Fbrstermann. LY 83583 interferes with the release of endothelium-derived relaxing factor and inhibits soluble guanylate cyclase. J . Pharmacol. Exp. Ther. 247: 283-288, 1988. Muscholl, E. Peripheral muscarinic control of norepinephrine release in the cardiovascular system. Am. J . Physiol. 239 (Heart Circ. Physiol., 8): H713-H720, 1980. Murad, F., Y. - M . Chi, T. W. Rail, and E. W. Sutherland. Adenyl Cyclase III. The effect of catecholamines and choline esters on the formation of adenosine 3',5'-phosphate by preparations from cardiac muscle and liver. J . Biol. Chem. 237: 1233-1238, 1962. Murray K. J . and P. J . England. Protein kinases as mediators of hormone action. Sci. Prog. 71: 221-238, 1987. Nathan, C. Nitric oxide as a secretory product of mammalian cells. F A S E B J . 6: 3051-3064, 1992. Nawath, H. Cyclic A M P and cyclic G M P may play opposing roles in influencing force of contraction in mammalian myocardium. Nature 262: 509-511, 1976. Nawrath, H. Does cyclic G M P mediate the negative inotropic effect of acetylcholine in the heart? Nature 267: 72-74, 1977. Nawrath, H., Baumner, D., Rupp, J . and H. Oelert. The ineffectiveness of the NO-cyclic G M P signaling pathway on the atrial myocardium. Br. J . Pharmacol. 116: 306-3067, 1995. 183 Neyes L. and H. Vetter. Action of atrial natriuretic peptide and angiotensin II on the myocardium: studies in isolated rat ventricular cardiomyocytes. Biochem. Biophys. Res. Commun. 163: 1435-1443, 1989. Okazaki, O., N. Suda, K. Hongo, M. Konishi, and S. Kurihara. Modulation of C a 2 + transient and contractile properties by p-adrenoceptor stimulation in ferret ventricular muscles. J . Physiol. 423: 221-240, 1990. Olivetti, G., P. Anversa, and A. V. Loud. Morphometric study of early postnatal development in the left and right ventricular myocardium of the rat. Circ. Res. 46: 503-512, 1980. Ono, K., and W. Trautwein. Potentiation by cyclic G M P of p-adrenergic effect on C a 2 + current in guinea-pig ventricular cells. J . Physiol. 443: 387-404, 1991. Palmer, W. K., J . M. McPherson, and D. A. Walsh. Critical controls in the evaluation of cAMP-dependent protein kinase activity ratios as indices of hormonal action. J . Biol. Chem. 255: 2663-2666, 1980. Pappano, A. J . , and D. Inoue. Development of different electrophysiological mechanisms for muscarinic inhibition of atria and ventricles. Fed. Proc. 43: 2607-2612, 1984. Paulus, W. J . , P. J Vantrimpont, and A. J . Shah. Acute effects of nitric oxide on left ventricular relaxation and diastolic distensibility in humans. Circ. 89: 2070-2078, 1994. Peterson, G. L. Review of the Folin phenol protein quantitation method of Lowry, Rosebrough, Farrand Randal. Anal. Biochem. 100: 201-220, 1979. Pryzwansky, K. B., T. A. Wyatt, and T. M. Lincoln. Cyclic GMP-dependent protein kinase is targeted to intermediate filaments and phosphorylates vimentin in A23187-stimulated human neutrophils. Blood 85: 222-230, 1995. Raeymakers, L., F. A. Hofmann, and R. Casteels. Cyclic GMP-dependent protein kinase phosphorylates phospholamban in isolated sarcoplasmic reticulum from cardiac and smooth muscle. Biochem. J . 252: 269-273, 1988. Raine. A. E. G., P. Erne, E. Burgisser, F. B. Muller, P. Bolli, F. Burkart, and F. R. Buhler. Atrial natriuretic peptide and atrial pressure in patients with congestive heart failure. N. Engl. J . Med. 315: 533-537, 1986. Rankin A. J . and F. V. Swift. The inotropic effect of atrial natriuretic factor in the anesthetized rabbit. Pfluegers Arch. 417: 353-359, 1990. 184 Robertson, S. P., J . D. Johnson, M. J . Holroyde, E. G. Kranias, J . D. Potter, and R. J . Solaro. The effects of troponin-I phosphorylation on the Ca 2 + -binding properties of the Ca-regulatory site of bovine cardiac troponin. J . Biol. Chem. 257: 260-3, 1982. Rodger I. W., and M. Shahid. Forskolin, cyclic nucleotides and positive inotropism in isolated papillary muscles of the rabbit. Br. J . Pharmacol. 81: 151-159, 1984. Roy, L. F., R. I. Ogilvie, P. Larochelle, P. Hamet, and F. H. H. Leenan. Cardiac and vascular effects of atrial nartiuretic factor and sodium nitroprusside in healthy men. Circ. 79: 383-392, 1989. Ruth, P., W. Landgraf, A. Keilbach, B. May, C. Egleme, and F. Hofmann. The activation of expressed cGMP-dependent protein kinase isozymes I alpha and I beta is determined by the different amino-termini. Eur. J . Biochem. 202: 1339-1344,1991. Schluter, K. D., M. Weber, E. Schraven, and H. M. Piper. NO donor SIN-1 protects against reoxygenation-induced cardiomyocyte injury by a dual action. Am. J . Physiol. 267 (Heart Circ. Physiol. 36): H1461-1466, 1994. Schmeid R. and M. Korth. Muscarinic receptor stimulation and cyclic AMP-dependent effects in guinea-pig ventricular myocardium. Br. J . Pharmacol. 99: 401-417, 1990. Schmidt, H. H. H. W., S . M. Lohmann, and U. Walter. The nitric oxide and c G M P signal transduction system: regulation and mechanism of action. Biochim. Biophys. Acta 1178: 153-175, 1993. Schulz, R. J . A. Smith, M. J . Lewis, and S. Moncada. Induction and potential biological relevance of a Ca 2 + - independent nitric oxide synthase in the myocardium. Br. J . Pharmacol. 104: 21-24, 1991a. Schulz, S. , P. S. T. Yuen, and D. L. Garbers. The expanding family of guanylyl cyclases. Trends Pharmacol. Sci. 12: 116-120, 1991b. Sekhar, K. R., R. J . Hatchett, J . B. Shabb, L. Wolfe, S. H. Francis, J . N. Wells, B. Jastorff, E. Butt, M. M. Chakinala, and J . D. Corbin. Relaxation of pig coronary arteries by new and potent c G M P analogs that selectively activate type la, compared with type IB, cGMP-dependent protein kinase. Mol. Pharmacol. 42: 103-108, 1992. Semigran, M. J . , C. N. Aroney, H. C. Herrmann, G. W. Dec, C. A. Boucher, and M. A. Fifer. Effects of atrial natriuretic peptide on left ventricular function in hypertension. Hypertension 24: 271-279, 1994. 185 Shah, A. M., M. J . Lewis, and A. H. Henderson. Effects of 8-bromo-cyclic G M P on contraction and on inotropic response of ferret cardiac muscle. J . Mol. Cardiol. 23: 55-64, 1991. Shah, A., H. A. Spurgeon, S. J . Sollot, A. Talo, and E. G. Lakatta. 8-Bromo-cGMP reduces the myofilament response to C a 2 + in intact cardiac myocytes. Circ. Res. 74: 970-978, 1994. Shimizu, A., M. Ueda, H. Watanabe, K. Hina, N. Yamada, S. Kusachi, D. Saito, S . Haraoka, and T. Tsuji. Enhancement of coronary conductance by a -human atrial natriuretic polypeptide without effects on myocardial contractility. Arzneim. -Forch. Drug Res. 38(H): 1572-1577, 1988. Smith, J . A., A. M. Shah, S. Fort, and M. J . Lewis. The influence of endocardial endothelium on myocardial contraction. Trends Pharmacol. Sci. 13: 113-116, 1992. Smith, J . A, A. M. Shah, and M. J . Lewis. Factors released from endocardium of the ferret and pig modulate myocardial contraction. J . Physiol. 439: 1-14, 1991. Smith, R. P., D. E. Wilcox, H. Kruszyna, and R. Kruszyna. Nitroprusside: a potpourri of biologically reactive intermediates. In: Biological Reactive Intermediates, edited by C. M. Witmer, R. R. Snyder, D. J . Jollow, G. F. Kalf, J . J . Kocsis, and I. G. Sipes. New York: Plenum Press, 1990, vol. IV, pp. 365-369. Sonnenburg, W. K. And J . A. Beavo. Cylic G M P and regulation of cyclic nucleotide hydrolysis. Adv. Pharmacol. 26: 87-114, 1994. Sperelakis, N., Z. Xiong, G. Haddad, and H. Masuda. Regulation of slow calcium channels of myocardial cells and vascular smooth muscle cells by cyclic nucleotides and phosphorylation. Mol. Cell. Biochem. 140: 103-117, 1994. Steadman, B. W., K. B. Moore, K. W. Spitzer, and J . H. Bridge. A video system for measuring motion in contracting heart cells. IEEE Trans. Biomed. Eng. 35: 264-272, 1988. Stein, B., A. Drogemuller, A. Mulsch, W. Schmitz, and H. Scholz. Ca + + -dependent constitutive nitric oxide synthase is not involved in the cyclic GMP-increasing effects of carbachol in ventricular cardiomyocytes. J . Pharmacol. Exp. Ther. 266: 919-925, 1993. Stein, B., and C. Schweiger. Effects of carbachol in the presence of isoprenaline on contractile response and cAMP and c G M P content in ventricular cardiomyocytes. Naunyn-Schmiedeberg's Arch. Pharmacol. 341 (Suppl): R47, 1990 (Abstract). Stone J . A., P. H. M. Backx, and H. E. D. ter Keurs. The effect of atrial natriuretic factor on force development in rat cardiac trabeculae. Can. J . Physiol. Pharmacol. 68: 1247-1254, 1990. 186 Szabo. G, and A. S. Otero. G-Protein mediated regulation of K + channels in heart. Ann. Rev. Physiol. 52: 293-305, 1990. Takai, Y., K. Nishiyama, H. Yamamura, Y. Nishizuka. Guanosine 3':5'-monophosphate-dependent protein kinase from bovine cerebellum: purification and characterization. J . Biol. Chem. 250: 4690-4695, 1975. Takasago, T., T. Imagawa, K. Furukawa, T. Ogurusu and M. Shigekawa. Regulation of the cardiac ryanodine receptor by protein kinase-dependent phosphorylation. J . Biochem. 109: 163-170, 1991. Tao, S. and T. M. McKenna. In vitro endotoxin exposure induces contractile dysfunction in adult rat cardiac myocytes. Am. J . Physiol. (Heart Circ. Physiol. 36): H1745-H1752, 1994. Ten Eick, R., Nawrath, H., MacDonald, T. F., and W. Trautwein. On the mechanism of the negative inotropic effect of acetylcholine. Pfluegers Arch. 361: 207-213, 1976. Thelan, K. I., A. Dembinska-Kiec, D. Pallapies, T. Simmet, and B. A. Peskar. Effect of 3-morpholinosydnonimine (SIN-1) and NG-nitro-L-arginine (NNA) on isolated perfused anaphylactic guinea pig hearts. Naunyn-Schmiedeberg's Arch. Pharmacol. 345: 93-99, 1992. Tobwin, H., Staehlin, T., and J . Gordon. Electrophoretic transfer of proteins from acrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76: 4350-4354, 1979. Tohse, N., H. Nakaya, Y. Takeda, and M. Kanno. Cyclic GMP-mediated inhibition of L-type C a 2 + channel activity by human natrieuretic peptide in rabbit heart cells. Br. J . Pharmacol. 114:1076-1082,1995. Tse, J . , C. W. MacKenzie III, and T. E. Donnelly. A sensitive and specific binding assay for cyclic GMP-dependent and cAMP-dependent protein kinases in rat liver. Int. J . Biochem. 13: 1071-1079, 1981. Uno, I., T. Ueda, and P. Greengard. Differences in properties of cytosol and membrane-derived protein kinases. J . Biol. Chem. 251: 2192-2195, 1976. Vaxelaire, J . F., S . Laurent, P. Lacolley, V. Briad, H. Schmitt, and J . B. Michel. Atrial natriuretic peptide decreases contractility of cultured chick ventricular cells. Life Sci. 45: 41-48, 1989. Wahler, G. M., and S. J . Dollinger. Nitric oxide donor SIN-1 inhibits mammalian cardiac calcium current through cGMP-dependent protein kinase. Am. J . Physiol. 268 (Cell Physiol. 37): C45-C54, 1995. 187 Waldman, S. A. and F. Murad. Cyclic G M P synthesis and function. Pharmacol. Rev. 39: 163-196, 1987. Walton, G. M., and G. N. Gill. Regulation of cyclic nucleotide-dependent protein kinase activity by histones and poly (L-arginine). J . Biol. Chem. 256: 1681-1688, 1981. Watanabe, A. M., and H. R. Besch, Jr. Interaction between cyclic adenosine monophosphate and cyclic guanosine monophosphate in guinea pig ventricular myocardium. Circ. Res. 37: 309-317, 1975. Watson, J . M., S. M. Vogel, D. J . Cotterell, and M. L. Dubocovich. Cholinergic antagonism of p-adrenergic stimulated action potentials and adenylate cyclase activity in rabbit ventricular cardiomyocytes. Eur. J . Pharmacol. 155: 101-108, 1988. Weyrich, A. S. , X. Ma, M. Buerke, T. Murohara, V. E. Armstead, A. M. Lefer, J . M. Nicolas, A. P. Thomas, D. J . Lefer, and J . Vinten-Johansen. Physiological concentrations of nitric oxide do not elicit an acute negative inotropic effect in unstimulated cardiac muscle. Circ. Res. 75: 692-700, 1994. Wildey, G. M., K. S. Misono, and R. M. Graham. Atrial natriuretic factor: biosynthesis and mechanism of action. In: The Heart and Cardiovascular System: Scientific Foundations, 2 n d ed., edited by H. A. Fozzard, E. Haber, R. B. Jennings, A. M. Katz, and H. E. Morgan, New York: Raven Press, 1992, pp. 1777 - 1796. Wittenberg, J . B. Myoglobin-facilitated oxygen diffusion: role of myoglobin in oxygen entry into muscle. Physiol. Rev. 50:559-636, 1970. Wolfe, L., Corbin, J . D., and S. H. Francis. Characterization of a novel isozyme of cGMP-dependent protein kinase from bovine aorta. J . Biol. Chem. 264: 7734-7741, 1989. Wolfe, L., S . H. Francis, L. R. Landiss, and J . D. Corbin. Interconvertible cGMP-free and cGMP-bound forms of cGMP-dependent protein kinase in mammalian tissue. J . Biol. Chem. 262: 16906-16913, 1987. Wolfe, S. K., and J . H. Swinehart. Photochemistry of pentacyanonitrosylferrate(2"), nitroprusside. Inorg. Chem. 14:1049-1053, 1975. Wyatt, T. A., K. B. Pryzwansky, and T. M. Lincoln. KT5923 activates human neutrophils and fails to inhibit cGMP-dependent protein kinase phosphorylation of vimentin. Res. Commun. Chem. Pathol. Pharmacol. 74: 3-14, 1991. Wyatt, T. A., T. M. Lincoln, and K. B. Pryzwansky. Regulation of human neutrophil degranulation by LY-83583 and L-arginine: role of cGMP-dependent protein kinase. Am. J . Physiol. (Cell Physiol. 34): C201-C211, 1993. 188 Xiao, R-. P., and E. G. Lakatta. p r Adrenoceptor stimulation and p 2-adrenoceptor stimulation differ in their effects on contraction, cytosolic C a 2 + , and C a 2 + current in single rat ventricular cells. Circ. Res. 73: 286-300, 1993. Yanagisawa A. and A. M. Lefer. Pharmacological actions of synthetic atrial natriuretic factor on coronary vascular smooth muscle. Japan. Circ. J . 52: 1436-1439, 1988. Yau, K-. W. Cyclic nucleotide-gated channels: an expanding new family of ion channels. Proc. Natl. Acad. Sci. USA 91: 3481-3483, 1994. 189 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0087291/manifest

Comment

Related Items