Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Role of cGMP-dependent protein kinase in smooth muscle relaxation Patel, Ashwinkumar Ishverbhai 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_1996-091465.pdf [ 5.89MB ]
Metadata
JSON: 831-1.0087242.json
JSON-LD: 831-1.0087242-ld.json
RDF/XML (Pretty): 831-1.0087242-rdf.xml
RDF/JSON: 831-1.0087242-rdf.json
Turtle: 831-1.0087242-turtle.txt
N-Triples: 831-1.0087242-rdf-ntriples.txt
Original Record: 831-1.0087242-source.json
Full Text
831-1.0087242-fulltext.txt
Citation
831-1.0087242.ris

Full Text

ROLE OF cGMP-DEPENDENT PROTEIN KINASE IN SMOOTH MUSCLE RELAXATION. by ASHWINKUMAR ISHVERBHAI PATEL B.Sc. (Hons)., University of London, 1982 M.Sc, Memorial University of Newfoundland, 1986 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT 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 The University of British Columbia March, 1996 © Ashwinkumar I Patel, 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 scholarly 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. Department of PrrMMAcfr jTlCAL ^^C£ The University of British Columbia Vancouver, Canada DE-6 -(2788) Abstract. Elevations in cyclic guanosine 3'-5'-monophosphate (cGMP) and activation of cGMP-dependent protein kinase (PKG) are believed to be responsible for nitrovasodilator-induced relaxation of vascular smooth muscle (VSM). In contrast to VSM, in tissues such as rat vas deferens and rat distal colon (non-VSM), cGMP elevation is not associated with relaxation. The hypotheses tested in this thesis were that (1) in VSM, cGMP elevation and PKG activation are correlated with relaxation, and (2) in non-VSM, the lack of correlation between cGMP elevation and relaxation is due to (a) lack of PKG activation and/or (b) low levels of PKG. Studies were carried out to optimize the assay conditions used to measure PKG activity. When PKG and cAMP-dependent protein kinase (PKA) were resolved by MonoQ anion exchange chromatography it was found that the assay conditions were specific for PKG since this activity was not found where Type I PKA activity was found. Furthermore immunoblotting with PKG antibodies showed that immunoreactive blots were only seen in those fractions which exhibited PKG activity. Concentration-dependent studies with sodium nitroprusside (SNP) and nitroglycerin (GTN), demonstrated a good correlation between elevation of cGMP levels, activation of PKG and relaxation in rabbit aorta. In temporal studies, PKG was activated at the earliest time point studied at which the preparations were known to be relaxed by SNP and GTN. ii In rat vas deferens, 0.1 mM SNP elevated cGMP levels and significantly increased PKG activity ratios but no inhibition of contraction took place. A higher concentration of SNP (5.0 mM), resulted in a more marked increase in PKG activity ratio (3-fold compared to control) but despite this increase the phenylephrine-induced contraction was not inhibited. Similarly, in the rat distal colon, atrial natriuretic factor (ANF) significantly increased cGMP levels and activated PKG but did not inhibit spontaneous contractions. Another possible reason why, despite activation of PKG, there is no inhibition in distal colon might be due to low total levels of PKG in this tissue compared to rabbit aorta. Total PKG levels in both proximal and distal colon were lower than those in the rabbit aorta and yet, when PKG was activated in these tissues by ANF, the contractions in proximal, but not in the distal colon, were inhibited. Taken together, these data suggest that the ability of some, but not all, smooth muscles to relax in response to cGMP-elevating agents cannot be explained solely on the basis of differences in total PKG activity in the muscles. It is possible that the lack of relaxation in response to PKG activation in muscles such as the rat vas deferens and distal colon may be due to (1) a lack of co-localization of PKG with a substrate, or (2) lack of critical protein substrates downstream from PKG activation. ^ J. Diamond. Ph.D. Research Supervisor. iii TABLE OF CONTENTS Page ABSTRACT ii TABLE OF CONTENTS iv LIST OF FIGURES viii LIST OF TABLES x LIST OF ABBREVIATIONS xi LIST OF AMINO ACID CODES xiv ACKNOWLEDGMENTS xv 1.0. INTRODUCTION 1 1.1 Historical perspective: from contraction to relaxation. 2 1.2. Mechanism of cGMP elevation: Role of nitric oxide (NO). 5 1.3. Evidence for a role of cGMP in smooth muscle relaxation. 9 1.4. Cyclic GMP-dependent protein kinase. 14 1.5. Is PKG involved in smooth muscle relaxation? 18 1.6. How is vascular tone reduced by cGMP/PKG? 20 1.7. cGMP in non-vascular smooth muscle. 25 1.8. Summary and rationale for proposed experiments. 27 1.9. Hypothesis. 29 2.0. Specific goals of the present investigation. 30 2.0. MATERIAL AND METHODS 31 2.1. Chemicals and materials. 31 iv 2.2. Animals. 33 2.3. Isolated tissue preparation. 34 2.3.1. Preparation and handling of rabbit aorta. 34 2.3.2. Preparation and handling of rat vas deferens. 35 2.3.3. Preparation and handling of rat proximal and 35 distal colon. 2.4. Isolated tissue treatment protocol. 36 2.4.1. Rabbit aorta. 36 2.4.2. Rat vas deferens. 38 2.4.3. Rat proximal and distal colon. 38 2.5. cGMP estimation in rabbit aorta, rat vas deferens 39 and rat proximal & distal colon. 2.6. Preparation of extracts and assay of cGMP- and 40 cAMP-dependent protein kinase. 2.6.1. Extraction of soluble and particulate fraction. 40 2.6.2. Assay of cGMP-and cAMP-dependent protein kinase. 41 2.7. Column chromatography of PKG and PKA in 42 muscle extracts. 2.8. Immunoblotting. 43 2.9. Preparation of drug solutions. 45 2.10. Protein determination. 45 2.11. Statistical analysis. 45 3.0. RESULTS 47 3.1. Optimization of assay parameters. 47 v 3.1.1. Linearity of phosphotransferase activity. 47 3.1.2. Determination of kinetic parameters of BPDEtide 47 phosphorylation and ATP utilization in crude soluble rabbit aorta homogenate. 3.1.3. Effect of different incubation times on PKG activity ratios. 52 3.1.4. Effect of PKI on phosphorylation of BPDEtide. 55 3.1.5. Chromatographic separation and identification of PKG 58 in rabbit aorta, rat vas deferens, rat proximal and distal colon. 3.1.6. Inhibition of PKG phosphotransferase activity in vitro 69 3.2. Effect of nitrovasodilators on contractility, cGMP levels 72 and PKG activity ratio in smooth muscle. 3.2.1. Effects of SNP and GTN on contractility, PKG activity 72 ratio and cGMP levels in rabbit aorta. 3.2.1.1. Sodium nitroprusside. 72 3.2.1.2. Nitroglycerin. 78 3.2.2. Effect of SNP on contractility, PKG activity ratio and 82 cGMP levels in rat vas deferens. 3.2.3. Effect of nitrovasodilators on PKG activity in soluble and 87 particulate fractions in rabbit aorta and rat vas deferens. 3.2.3.1. Determination of optimal Triton X-100 concentration. 87 3.2.3.2. Determination of soluble and particulate PKG 91 activity in rabbit aorta and rat vas deferens. 3.2.4. Effect of ANF on contractility, PKG activity ratio and 92 cGMP levels in rat colon. 3.3. Comparision of total PKG activity in different smooth muscles. 95 3.4. Attempts to inhibit nitrovasodilator-induced relaxation in 99 rabbit aorta with PKG inhibitors. vi 4.0. DISCUSSION 106 4.1. Optimization of assay conditions. 106 4.1.1. Choice of substrate. 107 4.1.2. Is the PKG activity ratio a good reflection of in vivo 110 activation? 4.1.3. Is the measured phosphotransferase activity due to PKG? 111 4.2. PKG activity in smooth muscle preparations. 117 4.2.1. Studies with SNP and GTN in rabbit aorta. 117 4.2.2. Studies with SNP in selected non vascular smooth muscles. 123 4.2.3. Effect of Rp-8-pCPT-cGMPS on smooth muscle relaxation. 129 4.3. Future Directions. 131 4.4. Summary. 133 5.0. BIBLIOGRAPHY 136 vii LIST OF FIGURES 1. Time course of cGMP-dependent protein kinase assay. 48 2. Determination of kinetic parameters of BPDEtide and ATP 50 in rabbit aorta. 3. Effect of reaction time on PKG activity ratios in rabbit aorta. 53 4. Effect of cAMP-dependent protein kinase inhibitor on rabbit aorta soluble PKG activity. 56 5. MonoQ column chromatography of rabbit aorta and rat vas 60 deferens PKG and PKA. 6. MonoQ column chromatography of rat colon PKG and PKA. 62 7. Immunoblots of PKG- and PKA-containing fractions. 64 8. MonoQ column chromatography and immunoblot of rabbit 67 aorta PKG. 9. Effect of Rp-8-pCPT-cGMPS on PKG activity in rabbit aorta, 70 rat vas deferens, rat proximal and distal colon. 10. Effects of GTN and SNP on PE (1.0 uM) -induced 73 contraction in rabbit aorta. 11. Effect of SNP on the PE-induced contractile response 83 in rat vas deferens. 12. Determination of optimal Triton X-100 concentration for 89 PKG extraction. 13. Distribution of PKG activity in rat vas deferens and rabbit 93 aorta following treatment with SNP or GTN. 14. Effects of ANF on spontaneously-contracting rat distal and 96 proximal colon. 15. Total PKG activity levels in rabbit aorta, rat vas deferens 100 and rat proximal and distal colon. viii 16. Effect of Rp-8-pCPT-cGMPS on SNP and GTN-induced 104 relaxation in rabbit aorta medial strips. ix LIST OF TABLES 1. Effect of sodium nitroprusside (0.1 -10 uM / 2 min) on PKG activity, cGMP levels and contractility in phenylephrine-contracted 75 rabbit aorta. 2. Time course of sodium nitroprusside (10 uM) effect on PKG 77 activity, cGMP levels and contractility in phenylephrine-contracted rabbit aorta. 3. Effect of nitroglycerin (0.1-10 uM/ 2 min) on PKG activity, cGMP 79 levels and contractility in phenylephrine-contracted rabbit aorta. 4. Time course of nitroglycerin (10 uM) effect on PKG activity, cGMP 81 levels and contractility in phenylephrine-contracted rabbit aorta. 5. Effect of sodium nitroprusside on PKG activity, cGMP levels 85 and contractility in PE-induced contractions in rat vas deferens. 6. Effect of atrial natriuretic factor (ANF) (0.1 uM) on PKG activity, 98 cGMP levels and contractility in spontaneously contracting rat proximal and distal colon. x LIST OF ABBREVIATIONS ACh acetylcholine ANF atrial natriuretic factor ATP adenosine 5'-triphosphate BCIP 5-bromo-4-chloro-3-indolyl phosphate [y-32P] ATP ATP in which the phosphorous at the y position is radioactive BSA bovine serum albumin °C degrees Celsius cAMP adenosine 3', 5'-cyclic monophosphate CKII casein kinase II CPM counts per minute DMF N,N-dimethylformamide DTT dithiothreitol EDTA ethylenediamine tetraacetic acid FPLC fast protein liquid chromatography GTN nitroglycerin g gram (s) (mass), gravitational acceleration, 9.8 m/s2 (centrifugal acceleration) cGMP guanosine 3', 5'-cyclic monophosphate h hour i.e. id est (that is) i.v. intravenous(ly) XI IBMX 3-isobutyl-1-methylxanthine kDa kilodalton kg kilogram M molar mg milligram min minute ml millilitre mM millimolar MW molecular weight uCi microCurie ug microgram ul microlitre uM micromolar nM nanomolar NBT nitroblue tetrazolium PAGE polyacrylamide gel electrophoresis PKA cyclic AMP-dependent protein kinase PKC protein kinase C PKG cyclic GMP-dependent protein kinase PKI peptide inhibitor of cAMP-dependent protein kinase PMSF phenylmethylsulfonyl fluoride pmol picomole xii s second SDS sodium dodecyl sulfate S.E.M. standard error of the mean SNP sodium nitroprusside TCA trichloroacetic acid TEMED N,N,N',N'-tetramethyl ethylenediamine xiii LIST OF AMINO ACID CODES Name Three letter code One letter code Alanine Ala A Arginine Arg R Asparagine Asn N Aspartic acid Asp D Cysteine Cys C Glycine Gly G Glutamic acid Glu E Glutamine Gin Q Histidine His H Isoleucine lie 1 Leucine Leu L Lysine Lys K Methionine Met M Phenylalanine Phe F Proline Pro P Serine Ser S Threonine Thr T Tryptophan Trp W Tyrosine Tyr Y Valine Val V xiv ACKNOWLEDGMENTS I would like to extend my sincere gratitude to my supervisor, Dr. J. Diamond, for his guidance, support and patience throughout the course of my study. I would also like to express my sincere thanks to the members of my supervisory committee, Dr. J. Axelson, Dr. D. Godin, Dr. K. MacLeod and Dr. S. Pelech for their constructive criticism and invaluable guidance and their understanding during difficult times. I am especially thankful to Dr. S. Pelech for providing the antibodies for the Western blot studies and for the use of his research facilities at Kinetek Biotechnology Corp. In addition I would like to acknowledge the help and assistance of Mr. Harry Paddon in performing the Western blot studies. I am appreciative of my colleagues for their constant encouragement and support. In particular, I wish to thank Ms. Karen MacDonell, Dr. Y. Hei, Dr. J. Langlands and Mr. James Hennan for their help and suggestions during the course of my studies. I would like to acknowledge the financial support from the Heart and Stroke Foundation of BC and Yukon and Eli Lilly Inc. in the form of predoctoral traineeships. xv DEDICATION. With love To my mother, father and Jyoti, for their care, support and understanding, and who were always there for me xv i 1.0. INTRODUCTION Cyclic guanosine 3'-5'-monophosphate (cGMP) was first discovered in rat urine by Ashman et al. (1963) and it was several years later that investigators began to study possible physiological roles of cGMP. It is interesting to note that several biochemical aspects of cGMP metabolism, such as the enzymes which catalyze its synthesis (Ishikawa et al., 1969) and its breakdown (Beavo et al., 1970), for example, were investigated well before its physiological effects on smooth muscle or any other muscle were studied. This was mainly due to the fact that research on cGMP was heavily influenced by progress made in understanding the role of another cyclic nucleotide, cyclic adenosine 3'-5'-monophosphate (cAMP), which was discovered much earlier (Sutherland and Rail, 1957). The enzymes responsible for the synthesis and degradation of cAMP (adenylyl cyclase and phosphodiesterases, respectively) had been widely studied, and a cAMP-dependent protein kinase (PKA) had been identified, the activation of which was suggested to be responsible for many of the effects of cAMP (see review by Flockhart and Corbin, 1982). Analogous experiments, using biochemical techniques similar to those used to study cAMP, were easily applied to the study of cGMP. As a result, the enzymes responsible for the synthesis and degradation of cGMP were identified, as noted above and, in addition, Kuo and Greengard (1970) first reported the presence of a protein kinase (cGMP-dependent protein kinase ([PKG]) in lobster tail muscle which was selectively activated by cGMP. 1 Since the original discovery of cGMP, this nucleotide has been shown to be an important intracellular second messenger for a wide variety of drugs and hormones (see review by Waldman and Murad, 1987). More recent studies have shown that it may also be involved in more diverse processes, such as insulin secretion (Schmidt et al., 1992), amylase secretion (Rogers et al., 1988), chloride secretion in the gut (Forte et al., 1992 ) and long-term potentiation in the hippocampus (Schuman and Madison, 1991). Most, if not all, of the physiological effects of cGMP are now thought to be mediated by the regulation of three proteins, namely PKG (Lincoln, 1989; Walter, 1989; Francis and Corbin, 1994), cGMP-gated ion channels (Fesenko et al., 1985; Kaupp, 1991) and cGMP-dependent phosphodiesterase (Beavo, 1988). It is not my intention to review the evidence for cGMP-gated ion channel or cGMP-dependent phosphodiesterase regulation by cGMP. Rather, I intend to review studies which provide evidence which implicates cGMP and PKG in smooth muscle relaxation. It is also beyond the scope of this review to describe every report published on this topic and therefore I will deliberately concentrate on those aspects which are most relevant to this thesis. 1.1 Historical perspective: from contraction to relaxation. The first demonstration of hormonally-induced increases in cGMP levels in smooth muscle involved agents which were known to cause smooth muscle 2 contraction. For example, Lee et al. (1972) showed that ACh and other cholinomimetics increased cGMP, but not cAMP, levels in intestinal smooth muscle. In the rat vas deferens, 10 uM ACh caused a contraction and elevated cGMP levels (Schultz et al., 1973), whereas in the rat uterus stimulation by methacholine, serotonin, oxytocin, and PGF 2 a , which induce contractions, led to elevations of cGMP levels ranging from 150 to 500% of control values (Goldberg et al., 1973). In contrast, cAMP and agents which increased cAMP levels in the rat uterus had the opposite effect, namely relaxation (Triner et al., 1971). These studies and others led to the postulation by Nelson Goldberg of the "Yin Yang" hypothesis with respect to the role of cyclic nucleotides as second messengers (Goldberg et al., 1973). This hypothesis had as its central tenet that cAMP and cGMP mediated mutually antagonistic actions inside the cell and was based on evidence from a variety of systems. With respect to smooth muscle, the hypothesis suggested that elevation of cAMP mediated relaxation, whereas elevation of cGMP mediated contraction. Other studies provided support for this contention. For example, in both bovine dorsal digital vein and the canine saphenous vein, PGF 2 o c contracted the tissues and increased cGMP values (45% to 160%), whereas the vasodilators PGE 2 and isoproterenol relaxed the blood vessels and increased cAMP levels (Dunham et al., 1974). The dibutyryl analog of cGMP was shown to potentiate contractions induced by electrical field stimulation and norepinephrine (NE) in guinea pig vas deferens (Wikberg and Andersson, 1978). 3 All these studies supported the hypothesis originally put forward by Goldberg et al. (1973). However, a subsequent series of studies conducted on the rat uterus, guinea pig uterus and taenia coli provided evidence against the "Yin Yang" hypothesis as it applied to smooth muscle (Diamond and Hartle, 1976; Diamond and Holmes, 1975). For example, when contraction was induced by KCI in the rat myometrium, cGMP levels were decreased, rather than increased (Diamond and Holmes, 1975). Furthermore even when increases in cGMP could be demonstrated during contractions in the guinea pig myometrium and taenia coli, the increases in cGMP occurred after the initiation of contraction (Diamond and Hartle, 1976). These studies suggested that cGMP elevation was not the cause of the contraction but might, in fact, be a result of contraction. More evidence against the "Yin Yang" hypothesis was provided by experiments in which the effects of nitroglycerin (GTN) were studied (Diamond and Holmes, 1975; Diamond and Blisard, 1976). KCI-induced contractions of rat uterus were relaxed by GTN, which also caused an increase in cGMP levels (Diamond and Holmes, 1975). Similar results were found with canine femoral arteries by Diamond and Blisard (1976), in which phenylephrine (PE)-induced contractions were relaxed by GTN, which also increased cGMP levels. These two studies studied (Diamond and Holmes, 1975; Diamond and Blisard, 1976) were the first demonstration that a vasodilator (GTN) could induce cGMP elevation in smooth muscle while causing relaxation and argued against the 4 "Yin Yang" hypothesis. Studies in other laboratories also cast doubt on the hypothesis. Schultz et al. (1977) showed that sodium nitroprusside (SNP) increased cGMP in rat vas deferens, and other studies by Katsuki et al. (1977a, 1977b) showed that several agents which activated guanylyl cyclase also increased cGMP levels and induced relaxation in a large number of smooth muscle preparations. As a result of the studies described above, the original suggestion that cGMP is a mediator of smooth muscle contraction is no longer accepted. It was then suggested that, not only was cGMP not responsible for contraction, but it might, in fact, be responsible for the smooth muscle relaxation caused by a variety of agonists (Schultz et al., 1977; Katsuki et al., 1977a). The evidence for a correlation between cGMP elevation and relaxation of vascular smooth muscle (VSM) now appears to be very strong, particularly for directly acting nitrovasodilators such as GTN, SNP and sodium nitrite, for drugs such as ACh and others which act via the endogenously released endothelial derived relaxing factor (EDRF), and for atrial natriuretic factor (ANF) (see reviews by Ignarro and Kadowjtz, 1985; Waldman and Murad 1987; Lincoln, 1989). 1.2. Mechanism of cGMP elevation: Role of nitric oxide (NO). As mentioned above, initial studies had shown that nitro-containing drugs such as GTN could increase cGMP levels and relax smooth muscle preparations (Diamond and Holmes, 1975, Diamond and Blisard, 1976). This, together with the 5 later studies by DeRubertis and Craven (1976) and Katsuki et al. (1977a), which showed that compounds such as azide, hydroxylamine and nitrite activated crude preparations of guanylyl cyclase and increased cGMP levels, led to further studies in this field. Activation of guanylyl cyclase and elevation of cGMP levels by nitrogen-containing compounds, and by nitric oxide (NO) itself, were also reported by Arnold et al. (1977), Craven and DeRubertis (1978) and Craven et al. (1979), leading to the suggestion that NO was responsible for enzyme activation by these compounds (Murad et al., 1978). This was indeed found to be the case, as NO gas was shown to activate cytosolic guanylyl cyclase, to elevate cGMP levels and to relax pre-contracted bovine coronary artery strips (Gruetter et al., 1979). In the same study, both methylene blue and methemoglobin, which were reported to inhibit NO activation of guanylyl cyclase (Murad etal., 1978), also inhibited the NO-induced relaxation and activation of guanylyl cyclase. These findings suggested that cGMP formation was probably causally associated with smooth muscle relaxation. Similar results were observed with chemical agents related to NO (Ignarro et al., 1981; Gruetter et al., 1981a, 1981b). The direct release of NO from compounds containing the nitro group was shown by Ignarro et al. (1980a, 1980b). These studies all supported the hypothesis that cGMP was involved in smooth muscle relaxation. The mechanism of NO production from compounds such as GTN and SNP is not clearly known. It is generally thought that SNP releases NO spontaneously (Feelisch and Noack, 1987). However a more recent study reported that NO 6 release from SNP took place only after a one-electron reduction with accompanying cyanide loss had occurred (Bates et al., 1991). The production of NO from GTN is thought to involve an enzymatic and a non-enzymatic mechanism (Anderson et al., 1994; Ignarro, 1989). The non-enzymatic pathway is thought to involve the reaction of GTN with thiols, such as cystein, to form unstable S -nitrosothiols (Ignarro et al., 1981). The formation of nitrosothiol intermediates from GTN appears to depend on the denitrification of the R-O-nitric oxide2 to a reactive thiol intermediary (Ignarro et al., 1981). Since S -nitrosothiols are unstable they rapidly decompose to yield NO (Ignarro, 1989). Organic nitrates are also enzymatically metabolized to yield NO via the involvement of glutathione-S-transferase (Lau et al., 1992) and the cytochrome P 4 5 0 related enzymes (McDonald and Bennet, 1990). The physiological significance of cGMP elevation by NO was addressed after the discovery of EDRF (Furchgott and Zawdazki, 1980). EDRF is a vascular relaxing factor released from endothelial cells upon stimulation by a number of agents such as ACh, ATP, histamine, serotonin, bradykinin and substance P, to name but a few (Furchgott, 1984). Subsequently, a number of studies reported that EDRF-dependent relaxation of VSM, from a variety of sources, was associated with cGMP accumulation (Holzman, 1982; Diamond and Chu, 1983; Rapoport and Murad, 1983; Ignarro et al., 1984). Methylene blue was also found to inhibit the EDRF-induced relaxation and cGMP elevation (Holzman, 1982; Ignarro et al., 1984) and Ignarro et al. (1984) suggested that EDRF mimicked NO in activating guanylyl cyclase. In later studies, Ignarro et al. (1986) demonstrated that EDRF released 7 from both artery and vein activated purified soluble guanylyl cyclase and that this activation was inhibited by methylene blue. Further studies revealed that the activation of guanylyl cyclase by EDRF required heme, as did the activation by NO (Ignarro et al., 1987b). These and other similarities led Ignarro et al. (1988) and Furchgott (1988) to independently suggest, at a 1986 meeting, that EDRF was NO or an unstable nitroso compound related to it. Evidence that EDRF was NO or a compound very similar to NO was independently reported by Ignarro et al. (1987a) and Palmer et al. (1987). This conclusion was based on the very similar actions of NO and of EDRF released from bovine aorta and vein (Ignarro et al., 1987a) and from cultured endothelial cells (Palmer et al., 1987). It is now generally accepted that EDRF is NO or a closely related intermediate which spontaneously releases NO (Ignarro, 1990). These studies have shown that both the nitrovasodilators and EDRF utilize a common pathway, NO, to activate soluble guanylyl cyclase, resulting in generation of cGMP. The soluble guanylyl cyclase is a heme protein and consists of 2 heterodimers with molecular masses of 70 kDa ( p subunit) and 82 kDa ( a subunit) (reviewed by Goy, 1991). Studies with expression of soluble guanylyl cyclase subunits in cells have shown that co-expression of both subunits was required for enzyme activity (Breuchler et al., 1991). The heme moiety can be removed to yield heme-deficient soluble guanylyl cyclase: however the guanylyl cyclase does not respond to NO in the absence of heme (Ignarro et al., 1984). NO is thought to bind to the heme moiety of soluble guanylyl cyclase and dislocates the heme-iron 8 resulting in a conformational change in the cyclase. This then activates the catalytic site of the enzyme (Ignarro et al., 1984). In addition to NO, carbon monoxide and hydroxyl groups (OH) can also activate soluble guanylyl cyclase leading to cGMP generation (Schmidt etal., 1993). In contrast to soluble guanylyl cyclase, the particulate guanylyl cyclase consists of a single transmembrane protein with molecular masses of 130 kDa to 180 kDa and does not contain heme (Chinkers et al., 1989). There are at least three forms of particulate guanylyl cyclase and these have similar intracellular regions but different extracellular domains to which peptide ligands such as ANF bind (Nakane and Murad, 1994). ANF and brain natriuretic factor (BNF) bind to the 'A' form of particulate guanylyl cyclase whereas natriuretic factor C binds to the 'B' form of particulate guanylyl cyclase. A third type of particulate guanylyl cyclase, the 'C form, is activated upon binding with the heat stable enterotoxin from Escherichia Coli (Nakane and Murad, 1994). 1.3. Evidence for a role of cGMP in smooth muscle relaxation. To confirm that nitrovasodilator-induced increases in cGMP mediate smooth muscle relaxation, certain criteria have to be satisfied. These criteria, which have been modified from those suggested for cAMP by Sutherland and co-workers (Robison et al., 1968), are as follows: 9 1. The drug must be capable of stimulating guanylyl cyclase in a cell-free system. 2. The drug-induced relaxation of smooth muscle must be accompanied by increases in tissue cGMP levels which should correlate with relaxation in a concentration- and time-dependent manner. 3. Analogs of cGMP should be able to mimic the relaxant effects of nitrovasodilators in the appropriate smooth muscle preparations. 4. Inhibitors of phosphodiesterases and guanylyl cyclase should be able to potentiate and inhibit, respectively, the relaxation caused by nitrovasodilators. With regard to the first point in the list above, it has been consistently shown that a wide range of compounds, which are sources of NO, activate, in a concentration-dependent manner, soluble guanylyl cyclase from a broad spectrum of smooth muscles (DeRubertis and Craven, 1976; Katsuki et al., 1977a; Arnold et al., 1977; Craven et al., 1979; Gruetter et al., 1979, Waldman and Murad, 1987). EDRF released from bovine pulmonary artery and vein has also been shown to activate soluble guanylyl cyclase and to increase cGMP levels (Ignarro et al., 1986). ANF, by contrast, does not activate soluble guanylyl cyclase, but does activate particulate guanylyl cyclase (Winquist et al., 1984). At about the same time that studies were reporting the activation of guanylyl cyclases by a variety of agents, a number of studies reported strong correlations 10 between the ability of these agents to increase cGMP levels and induce smooth muscle relaxation in both a time- and concentration-dependent manner. The first of these comprehensive studies was performed with bovine coronary artery (Kukovetz et al., 1979). These authors studied the relaxing effects of GTN, SNP and sodium nitrite (NaN02) on pre-contracted bovine coronary artery and found very tight concentration-dependent correlations between drug-induced cGMP elevation and relaxation. Moreover, the increase in cGMP preceded the relaxation. When the fold increase in cGMP was plotted against relaxation (on a probit scale) a correlation coefficient of +0.98 was obtained. Using the same preparation, Gruetter et al. (1981a) noted that the elevation of cGMP preceded the relaxation induced by NO, GTN, SNP and NaN02. Similar results have been reported in other smooth muscle preparations such as rat aorta (Keith et al., 1982) and bovine mesenteric artery (Axelsson et al., 1979). A concentration- and time-dependent relaxation of pre-contracted rabbit aorta by ACh (via EDRF) was demonstrated by Diamond and Chu (1983). Similar results have also been reported for bovine intrapulmonary arteries (Ignarro et al., 1984). All of these studies support the role of cGMP in smooth muscle relaxation. However, there is some evidence which does not appear to be consistent with this hypothesis, particularly in non-VSM (Diamond and Janis, 1978; Diamond, 1983) and this is discussed later (see Section 1.7). If cGMP is involved in relaxation, then analogs of cGMP should, theoretically, relax pre-contracted smooth muscle. The availability of membrane permeable cGMP analogs such as 8-Br-cGMP, which activates PKG (Kuo et al., 1978), has 11 made it possible to study the role of cGMP in relaxation. For example, Schultz et al. (1979) showed that 1 to 100 uM 8-Br-cGMP induced concentration-dependent relaxation of pre-contracted rat and rabbit aorta. This analog also inhibited PE-induced contraction of the rat vas deferens. In addition, 8-Br-cGMP inhibited the frequency of oxytocin-induced contractions (Schultz et al., 1979) in rat uterus and also the KCI-induced contraction of the rat uterus (Diamond, 1983). Kukovetz ef al. (1979) and Napoli etal. (1980) both reported concentration-dependent relaxation of bovine coronary artery by 8-Br-cGMP. These studies are viewed as fairly strong evidence for the cGMP hypothesis. The last criterion has also, in general, been satisfied. If the vasodilator-induced generation of cGMP can be blocked by inhibiting guanylyl cyclase, theoretically, the relaxant response should also be inhibited or diminished. Methylene blue, which inhibits guanylyl cyclase (Gruetter ef al., 1979), has been used extensively for these kinds of experiments. In the rat aorta, methylene blue (10 uM) significantly shifted the GTN and SNP-induced relaxation curve to the right and also attenuated cGMP increases by about 75% (Keith et al., 1982). In a later study, the same authors found very similar results in a non-VSM preparation (Keith et al., 1983). In bovine coronary artery, 10 uM methylene blue inhibited the GTN (1 uM)-induced elevation of cGMP and relaxation by 80 and 50%, respectively. EDRF-induced relaxation was also shown to be inhibited by methylene blue together with a reduction in cGMP elevation (Ignarro et al., 1984). Martin et al. (1985) reported that both methylene blue and hemoglobin inhibited the ACh-12 induced relaxation in endothelium intact rabbit aorta. Use of another compound which inhibits cGMP elevation, LY83583, has shown that it also blocks EDRF-induced elevation of cGMP as well as relaxation (Diamond, 1987). If inhibition of cGMP elevation leads to attenuation of relaxation, the converse should also be possible. Inhibition of phosphodiesterase (PDE) should enhance or maintain vasodilator-induced cGMP elevation and this should be reflected in an enhancement of the relaxation. A number of studies indeed confirm to a large extent that such modulation of PDE leads to enhanced relaxation. Kramer and Wells (1979) studied the effects of a number of PDE inhibitors on their ability to modulate cGMP levels and to cause relaxation, in pig coronary arteries. They found that (a) there was a strong correlation between the IC5 0 value for hydrolysis of cGMP and the E C 5 0 value for the relaxation, and (b) the ability to elevate cGMP levels and to induce relaxation showed a strong correlation. Kukovetz et al. (1979) reported that the elevation of cGMP by GTN and SNP in pig coronary arteries was significantly potentiated in the presence of M&B 22,948 (Zaprinast), a cGMP-PDE inhibitor. At the same time, M&B 22,948 significantly potentiated the relaxations induced by GTN and SNP. Similar results have also been reported in other vascular smooth muscle preparations such as the bovine intrapulmonary artery and vein (Ignarro et al. 1987b). In contracted canine trachealis muscle (a non-VSM preparation), the SNP-induced elevation of cGMP and relaxation were potentiated by M&B 22,948 (Zhou and Torphy, 1991). While all of these studies were done using M&B 22,948, similar results have been observed 13 using other PDE inhibitors such as milrinone, isomazole and LY 95115 (Kaufman et al., 1987), which led the authors to suggest that the enhanced relaxation was probably due to increases in cGMP. While the criteria listed above have been satisfied in general, there are data that do not conform to the hypothesis. Examples of non-conformity include (a) relaxation by nitrovasodilators in the absence of cGMP elevation (Diamond and Chu, 1983), (b) non-relaxation despite elevation of cGMP (Diamond, 1983), and (c) disproportionate changes in cGMP and relaxation with various cGMP elevating agents. These aspects have been critically reviewed by Nakatsu and Diamond (1989) and suggest that, despite data which strongly support the notion of cGMP being involved in relaxation, this may not always be the case. 1.4. Cyclic GMP-dependent protein kinase. Cyclic GMP-dependent protein kinase was first discovered by Kuo and Greengard (1970) in lobster tail muscle as a kinase which was specifically activated by cGMP. It has been hypothesized that, following generation of cGMP, activation of cGMP-dependent protein kinase and subsequent phosphorylation of various proteins results in reduction of smooth muscle tone (see reviews by Lincoln, 1989; Walter, 1989; Lincoln and Cornwell, 1991). Cyclic GMP-dependent protein kinases (PKG) are serine/threonine kinases which transfer the y-phosphoryl group from ATP onto serine or threonine residues in 14 target proteins (Butt et al., 1983; Francis and Corbin, 1994). The distribution of PKG is not as widespread as that of PKA and is found mainly in different types of smooth muscle, cerebellum, platelets and lung tissue (Lohmann etal., 1981; Joyce et al., 1986; Walter, 1981). The levels of PKG and PKA are similar in vascular smooth muscle (Francis et al., 1988), but in other tissues PKA levels are considerably higher (Lincoln and Corbin, 1983) than that of PKG. Platelets have high levels of PKG (about 3 uM) compared to smooth muscle (Eigenthaler et al., 1992). Two types of PKG have been identified in mammalian tissue. Type I (PKG I) is mainly a cytosolic enzyme (Gill et al., 1976; Lincoln et al., 1988; Wolfe et al., 1989a) , although some reports indicate that about 25% may be in the particulate fraction (Ives et al., 1980). Type II PKG (PKG II) was originally described in intestinal epithelial cells (DeJonge, 1976; DeJonge, 1981) and is tightly bound to the membrane via the N-terminal by a 12-15 kDa segment, although a recent report (Uhler, 1993) suggests that PKG II from mouse brain may be cytosolic. PKG I is a dimer consisting of two identical subunits (molecular mass of about 76-78 kDa per subunit) (Lincoln et al., 1988; Wolfe et al., 1989a), whereas PKG II is a monomer, although a recent report suggests that mouse intestinal PKG II exists as a dimer (Gamm et al., 1995). PKG II has not been reported to be involved in regulation of smooth muscle tone. 15 Purification (Lincoln et al., 1988, Wolfe et al., 1989a, 1989b) and cloning of PKG I (Wernet et al., 1989; Sandberg et al., 1989) led to the identification of two isozymes: l a and 1(3. These isozymes appear not to differ in substrate specificity or catalytic rate and, based on amino acid sequences, they only differ in the first «100 amino acids in the N-terminal region (Wernet et al., 1989; Sandberg et al., 1989). The distribution of PKG l a is somewhat different from that of PKG ip. PKG l a is the predominant form of PKG I in bovine lung, trachea, cerebellum and uterine smooth muscle (Wolfe et al., 1989a), whereas both PKG l|3 and PKG l a are found, in approximately equal quantities, in porcine coronary arteries, human and bovine aorta (Shekhar et al., 1992; Wolfe et al., 1989a). Most of the structural information available is for PKG I and has been extensively reviewed elsewhere (Lincoln and Corbin, 1983; Butt, et al., 1993; Francis and Corbin, 1994). Basically, the PKG subunit consists of five functional domains: (1) a dimerization domain located at the extreme NH2 terminus, (2) an autophosphorylation and autoinhibitory domain which is carboxyl terminal to the dimerization domain, (3) two cGMP-binding sites, (4) the catalytic domain and (5) the carboxyl-terminal domain. cGMP binds to the two cGMP binding sites and the rate of dissociation of cGMP distinguishes the two sites: one is a fast site (site 2) and the other is a slow site (site 1) (Corbin and Doskeland, 1983). The catalytic site of PKG is inhibited by the autoinhibitory region (domain 2) in the absence of cGMP. Under conditions of elevated cGMP levels, binding of 16 cGMP to its binding sites induces a conformational change which relieves the inhibition on the catalytic site and results in subsequent phosphorylation of intracellular targets. Binding of cGMP to the slow site can partially activate PKG (Corbin and Doskeland, 1983), unlike PKA, but occupation of both sites is required for full activation. Upon binding of cGMP, the catalytic domain does not dissociate, as is the case with PKA (Lincoln et al., 1976; Takai et al., 1976). E 2 + 2cGMP <-> E2-cGMP2 + 2 cGMP <-> E2-cGMP4 (partially active) (fully active) In addition, binding of cGMP to these sites displays different kinetics between PKG la and PKG Ip. cGMP activates the PKG ip isozyme with a higher Ka value (1.27 uM) compared to the PKG la isozyme (0.12 uM) (Ruth et al., 1991), suggesting that higher amounts of cGMP may be required to activate PKG ip than PKG la. Cyclic AMP can also bind to these sites but with a 50- to 200-fold lower affinity (Francis and Corbin, 1994) and it would appear that cAMP would not be able to activate PKG at usual tissue concentrations. However, under certain conditions activation may still be possible. Autophosphorylation of the autophosphorylation domain of PKG ip in the presence of cGMP results in a 6-fold decrease in the Ka of cAMP (Langdarf et al., 1986). Similarly, autophosphorylation of PKG ip lowers Ka for cAMP and cGMP (Smith et al., 1992). This represents another method by which cGMP may activate PKG (partially) and increases the range of cyclic nucleotide 17 concentrations which could result in full activation. Additionally, autophosphorylation could also, theoretically, sustain the physiological effects of extracellular signals that elevate cGMP only transiently (Smith et al., 1993). 1.5. Is PKG involved in smooth muscle relaxation? The next question is, does activation of PKG mediate the cGMP-induced relaxation? Unfortunately, it has been difficult to answer this question with any certainty, but more recent studies have reported data which lend support to a role for PKG in mediating the relaxant effects of cGMP-elevating agents. There are very few studies in which PKG activation (as indicated by activity ratio) has been measured in intact tissues. In one of those studies, strong correlations between increases in PKG activity ratio and relaxation have been reported in rat aorta treated with ANF (Fiscus et al., 1983). Sodium nitroprusside has also been reported to activate PKG in rat thoracic aorta (Fiscus et al., 1985), although relaxation was not measured in this study. In the pig coronary artery, Jiang et al. (1992) reported that SNP induced a concentration-dependent activation of PKG. None of the above mentioned studies, however, measured cGMP levels, PKG activity, and relaxation simultaneously. Finally, a very recent study in bovine carotid artery failed to detect activation of PKG, even though the blood vessel was fully relaxed in the presence of SNP (Bergh etal., 1995). 18 Studies with isolated aortic cells also provide further evidence for the involvement of PKG. In primary (non-passaged) rat aorta cells, vasopressin-induced increases in cytosolic Ca2+ (measured using fura-2, a Ca 2 + sensitive dye) were reduced by both ANF and 8-Br-cGMP (Cornwell & Lincoln, 1989). Since reducing cytosolic Ca2+ ultimately leads to a reduction in tone, the effect of ANF and 8-Br-cGMP should be to relax smooth muscle. However, when the study was repeated using cultured cells which had been passaged many times, ANF and 8-Br-cGMP had no inhibitory effect on vasopressin-induced increase in Ca2+ levels. It was noted that levels of PKG were considerably reduced in these passaged cells. When purified PKG was added back to the passaged cells, the C a 2 + lowering effect of ANF and 8-Br-cGMP was restored, which appears to be strong evidence that PKG is required for the relaxant effects of cGMP in these cells. Studies using cGMP and cAMP analogs have provided another line of evidence for the role of PKG in smooth muscle relaxation. For example, Francis et al. (1988) investigated the ability of a number of cAMP and cGMP analogs to relax pre-contracted porcine coronary artery and guinea pig trachealis. They found that the ability of the analogs (EC50) to relax these tissues correlated well with the Ka of the analogs for activating purified PKG rather than PKA 19 1.6. How is vascular tone reduced by cGMP/PKG? The studies mentioned above support the role for PKG in smooth muscle relaxation, but the underlying mechanisms by which activation of PKG could cause relaxation are not known. Phosphorylation of specific proteins in rat aorta by SNP, EDRF and 8-Br-cGMP has been demonstrated by Rapoport et al. (1982, 1983) but the identity of these proteins could not be determined. Since an increase in free ionized intracellular calcium concentration is critical to contraction of smooth muscle (Somlyo and Himpens, 1989; Morgan and Suematsu, 1990), any intervention that affects the state of contraction must, logically, alter either the intracellular concentration of calcium or the calcium requirement of the contractile machinery or both. It is not surprising that most of the research directed at delineating the intracellular mechanisms of cGMP-induced relaxation has focused on how cGMP and PKG may modulate calcium concentrations. The experimental basis for such a direction stems from studies which indicated changes in calcium concentrations in the presence of nitrovasodilators. For example, phosphorylase b kinase activity, which is dependent on calcium concentration, was found to be decreased after elevation of cGMP by GTN and activation of PKG by 8-Br-cGMP (Axelsson et al., 1985). The use of calcium-sensitive dyes further revealed the effect of nitrovasodilators on intracellular calcium. For example, Morgan and Morgan (1984) first showed that 20 nitrovasodilators reduced intracellular calcium levels in aequorin-loaded ferret vein strips. Reports with similar findings, in a number of preparations, soon followed (Kobayashi et al., 1985; Rashatwar et al., 1987; Karaki et al., 1988). A similar calcium lowering effect was also observed with ANF in rat aortic cells (Cornwell and Lincoln, 1989). In view of the complexity of calcium homeostasis in smooth muscle (Somlyo and Himpens, 1989; Morgan and Suematsu, 1990), no clear picture has emerged as far as regulation of calcium by cGMP is concerned, and an attempt will be made to present the various possible ways such regulation may be taking place. One of the sources of calcium for agon/'sMnduced contractions is the inositol 1,4,5-trisphosphate (IP3) - induced calcium release from sarcoplasmic reticulum (SR), following phosphatidylinositol 4,5-bisphosphate (PIP2) hydrolysis (Berridge, 1993). Rapoport (1986) first demonstrated the reduction of PIP2 hydrolysis in rat aorta by SNP and ANF. These agents relaxed norepinephrine-contracted rat aorta and decreased inositol monophosphate accumulation (an indication of PIP2 hydrolysis). 8-Br-cGMP and ACh (via EDRF) also had similar effects on PIP2 hydrolysis. What is not known is how cGMP/PKG inhibits IP3 accumulation. In rat aorta and cultured bovine aortic cells, cGMP has been reported to attenuate vasopressin-stimulated G-protein activation and also uncouple the activated G-protein and phospholipase C, resulting in decreased IP3 production (Hirata et al., 1990). They suggested that this attenuation was probably mediated through a cGMP-dependent phosphorylation since the effect required ATP. However, a direct effect of PKG on G protein phosphorylation has not been observed and it may be 21 that the inhibitory effect of cGMP/PKG may be an indirect one, in which phosphorylation of a regulatory protein for phospholipase C occurs. Another site where cGMP/PKG may act is at the IP3 receptor site on the SR. Komalavilas and Lincoln (1994) have reported that the purified IP3 receptor from the cerebellum can be phosphorylated by PKG. They also found that rat aortic cultures treated with SNP or ANF phosphorylated the IP3 receptor. This would explain the observation by Meisheri et al. (1986) who found that ANF inhibited intracellular calcium release in smooth muscle. Increased sequestration of calcium by SR has also been demonstrated in skinned cultures of rat aortic smooth muscle (Twort and van Breemen, 1988) in the presence of cGMP and PKG. One mechanism by which increased sequestration may occur is by activating the SR Ca2+-ATPase. PKG has been shown to phosphorylate phospholamban, the regulatory protein of SR Ca2+-ATPase in vitro (Saracevic et al., 1989; Raeymakers et al., 1988) and in early passaged cells (Cornwell et al., 1991) from aortic smooth muscle. If phosphorylation of phospholamban results in activation of SR Ga 2 +-ATPase, such phosphorylation would ultimately lead to increased sequestration of calcium into the SR and reduction in smooth muscle tone. In the latter study, phosphorylation of phospholamban correlated well with activation of Ca2+-ATPase and a reduction in intracellular calcium. However, these studies were done in vitro, and attempts to demonstrate in vivo phosphorylation have not met with success. Huggins et al. (1989) found that the inhibition of tension in pre-contracted rabbit 22 aorta by SNP did not correlate with increased phosphorylation of phospholamban. Therefore, the role of phospholamban as an in vivo target of PKG is not clear at present. In contrast, in vivo inhibition of the GTN- and ANF-induced relaxation of rabbit aorta by Ca2+-ATPase inhibitors (thapsigargin and cyclopiazonic acid) has been demonstrated, indicating that the Ca2+-ATPase may be involved in sequestration of calcium in response to cGMP-elevating agents and subsequent activation of PKG (Luo et al., 1993). The activation of the plasma membrane (PM) Ca2+-ATPase by cGMP and subsequent extrusion of calcium is another route by which intracellular calcium levels could be reduced. Most of the studies on PM Ca2+-ATPase have been done using isolated purified Ca2+-ATPase preparations (Rashatwar et al, 1987; Vrolix et al., 1988; Yoshida et al., 1991). Rashatwar et al. (1987) demonstrated that purified PKG activated PM Ca2+-ATPase activity by 4-fold in rat aortic cells. Later, Vrolix et al. (1988) reported that PKG increased Ca 2 + uptake and Ca2+-ATPase activity in porcine aorta. They further suggested that the mechanism of activation of the Ca 2 + -ATPase pump was probably by stimulation of phosphatidylinositol 4-phosphate (PIP) production by PKG rather than a direct action on the pump, since the stimulatory effect of PKG on the pump was only seen in the presence of phosphatidylinositol (PI). Yoshida et al. (1991) reported that the PM Ca2+-ATPase activity from porcine aorta was also increased in the presence of PKG but PI was not necessary for the effect of PKG. They reported that a 240 kDa protein was phosphorylated in a concentration-dependent manner by PKG and that this was not 23 the pump itself. In the absence of the 240 kDa protein, PKG had no effect on pump activity. This led Yoshida et al. (1991) to suggest that PKG phosphorylated the 240 kDa protein, which activated the PM Ca2+-ATPase pump by a mechanism that is still not clearly known. In addition to the above mechanisms by which cGMP/PKG may modulate calcium levels and therefore muscle tone, there is an additional mechanism by which relaxation may be achieved. This is by the hyperpolarization of the + membrane by activation of calcium-activated potassium (K ) channels. Studies with K channel openers have shown that there is a close relationship between relaxation and hyperpolarization (Hamilton et al., 1986; Nelson et al., 1990), and hyperpolarization may inhibit Ca 2 + influx through voltage gated Ca 2 + channels. Some recent reports have provided data which strongly implicate modulation of K channels by PKG-mediated phosphorylation (Robertson et al., 1993; Taniguchi et al., 1993: Alioua etal., 1995). Robertson et al. (1993) demonstrated, that in rabbit cerebral artery smooth muscle cells, NO and CPT-cGMP, a PKG activator, + + activated K channels by 2-fold and that purified PKG activated the K channels by 8-fold. Similar results have also been observed in canine coronary artery smooth muscle cells by Taniguchi et al. (1993). In a very recent report, Aliou et al. (1995) confirmed that PKG could activate K channels in airway smooth muscle. They also found that PKG phosphorylated two proteins, amongst others, which co-migrated with the same relative molecular mass as the 62 and 30 kDa subunits of the purified K channels, suggesting that the K channels are up-regulated by PKG. Such an 24 effect would hyperpolarize the membrane, inhibit voltage-gated channels and result in an inhibition of calcium influx which would, in turn, lead to relaxation. It should be noted that the above m e c h a n i s m s represent only s o m e of the possible m e c h a n i s m s by which nitrovasodilators induce relaxation and a combination of these m e c h a n i s m s may be involved simultaneously. However , it is not clear which, if any, of these m e c h a n i s m s is actually responsible for the smooth m u s c l e relaxation attributed to c G M P . This area has been reviewed recently by Lincoln etal. (1994). 1.7. cGMP in non-vascular smooth muscle. A s descr ibed above, there is generally good ev idence for a role for c G M P as a mediator of smooth muscle relaxation, particularly for vascular smooth muscle . T h e ev idence for such a role for c G M P in s o m e other types of smooth m u s c l e s is not so convincing. For example, in contrast to the results in V S M , the use of nitrovasodilators such as S N P , G T N , sodium azide and hydroxylamine, in t issues s u c h as rat vas deferens and uterus, has provided ev idence which suggests that there is a poor relationship between the ability of these c o m p o u n d s to increase c G M P and to c a u s e relaxation. O n the one hand, in rat vas deferens, very large increases in c G M P levels (16-fold) by S N P had no effect on P E - i n d u c e d contractions (Diamond and Janis , 1978). O n the other hand, G T N , which produced a lower (4-fold) increase in c G M P levels in rat vas deferens, inhibited P E - i n d u c e d 25 contractions (Diamond, 1983). This is in contrast to the situation in the rat aorta in which a smaller (2-fold) increase in cGMP by SNP resulted in relaxation (Lincoln, 1983). In the spontaneously-contracting rat myometrium, 0.5 and 5 mM SNP produced 2- and 6-fold (respectively) increases in cGMP, but had no effect on contraction. On the other hand, 0.5 mM GTN completely inhibited the spontaneous contractions, but did not significantly elevate cGMP levels (Diamond, 1983). Similar dissociations were also seen in potassium-contracted rat myometrium. Studies with the other nitrovasodilators, hydroxylamine and sodium azide, also show similar dissociations. Hydroxylamine (1mM), despite increasing cGMP levels by 3-fold, had no effect on spontaneous contractions whereas sodium azide completely relaxed the uterine strips but the elevation of cGMP was less (1.4-fold) than that seen with hydroxylamine. When the extent of cGMP elevation was plotted against % inhibition of uterine contraction, no correlation was found (r = - 0.05) (Diamond, 1983). Similar inconsistencies have been recently observed in the longitudinal muscle of the rat colon, in which there was an association between NO-induced relaxation and elevation of cGMP in the proximal colon, but not in the distal colon where NO increased cGMP levels but had no relaxant effect (Suthamnatpong et al., 1993). In the same study, a similar pattern of association (proximal end) and dissociation (distal end) was also seen with ANF, which activates the particulate guanylyl cyclase (Winquist et al., 1984). Therefore, dissociations between cGMP 26 elevation and relaxation are evident even in the same organ, depending on what part of the organ is studied. Some support for a role for cGMP as a mediator of relaxation even in the so called "non-responsive" types of muscle has come from studies using cGMP analogs. In the rat uterus, 8-Br-GMP has been shown to inhibit both the frequency of oxytocin-induced contractions (Schultz et al., 1979) and spontaneous contractions (Diamond, 1983). Potassium-induced contractions of the rat uterus were also inhibited by 8-Br-cGMP (Diamond, 1983). This analog also inhibited NE-induced contractions of rat vas deferens (Schultz et al., 1979). Thus, in contrast to VSM, the role of cGMP elevation and relaxation is not very clear in non-VSM preparations. 1.8. Summary and rationale for proposed experiments. It is clear from the above that cGMP elevation is related to relaxation in VSM. Since the actions of cGMP (in terms of VSM relaxation) are thought to be mediated via activation of PKG (see above), it would be important to directly demonstrate such activation. Unfortunately, such data are scarce. To date, there have been only four reports demonstrating PKG activation by nitrovasodilators in intact smooth muscle preparations (Fiscus et al., 1983, 1984, 1985; Jiang etal., 1992). The lack of such data stems from the difficulty in measuring PKG, partly due to a lack of a specific PKG substrate. The three earlier reports used histone H2B as a substrate 27 for the assay of protein kinase but this substrate is not specific for PKG. The PKG assay used by Fiscus et al. (1983, 1984, 1985) also lacked sensitivity and required very high levels of [y-32P] ATP. Partly because of the difficulties inherent in the assay, most of the reports cited above lack time- and concentration-dependence data. Even when time- and concentration-dependence data have been reported, the different parameters were determined in separate strips of muscle. One objective of the present study, therefore, was to simultaneously measure nitrovasodilator-induced cGMP elevation, PKG activation and relaxation in the same muscle strip using an improved assay for PKG recently described by Corbin and co-workers (Jiang et al., 1992). This assay makes use of BPDEtide, a sensitive and specific substrate for PKG (Colbran et al., 1992). It was hoped that these studies would provide a better insight into the relation between PKG activation and relaxation of vascular smooth muscle by nitrovasodilators. In contrast to VSM, in some types of non-VSM the evidence for a correlation between cGMP elevation and relaxation is not as clear. In rat vas deferens and myometrium, SNP has no relaxant effect, whereas it is a very potent relaxant in VSM. A similar dissociation between cGMP elevation and relaxation has also been observed in the rat distal colon, whereas in the proximal colon there appeared to be an association between cGMP elevation and relaxation. The question which needs to be addressed is that, if cGMP elevation can cause vascular smooth muscle relaxation, why does an equal or greater elevation of cGMP by various agents fail to produce relaxation in some non-vascular smooth muscles? One possible 28 explanation for the reported dissociation is that in non-responsive smooth muscle (i.e. muscle which does not relax in the presence of elevated cGMP), the SNP-induced cGMP elevation is compartmentalized, such that it has no access to PKG. If it is assumed that activation of PKG is required for relaxation, it follows that without PKG activation no relaxation would take place. Another possible explanation for the lack of correlation between cGMP and relaxation in some tissues is that the PKG is activated but is unable to relax the tissue because of a lack of appropriate protein substrates(s) in the non-responsive tissue. With the exception of one study in the canine trachea (Fiscus et al., 1984), there are virtually no studies which have investigated the effects of nitrovasodilators on PKG activation in non-vascular smooth muscle. As noted above, this is at least partly due to the difficulty in measuring PKG activation. 1.9. Hypothesis. The hypotheses tested in this thesis were that (1) in VSM, cGMP elevation and PKG activation are correlated with relaxation, and (2) in non-VSM, the lack of correlation between cGMP elevation and relaxation is due to (a) lack of PKG activation and/or (b) low levels of PKG. 29 2.0. Specific goals of the present investigation. Specific goals of the study were: 1. To optimize assay conditions to measure PKG activation in crude soluble homogenates from various smooth muscle preparations. 2. To study concentration- and time-dependent effects of SNP and GTN on cGMP levels, PKG activity ratios and tension in PE-contracted rabbit aorta. 3. To determine whether, in rat vas deferens and distal colon, the lack of relaxation in response to cGMP elevation is due to a lack of PKG activation. 4. If PKG was found to be activated in the vas deferens and distal colon, another objective would be to investigate other possible explanations for the lack of relaxation in these tissues. For example, (a) the total amount of PKG in various tissues (as reflected by PKG phosphotransferase activity in the presence of added cGMP) would be compared in an attempt to determine whether the total amount of PKG in the tissue has any bearing on whether or not it relaxes, and (b) attempts would be made to determine whether translocation of PKG from one compartment to another plays a role. 30 2.0. MATERIAL AND METHODS 2.1. Chemicals and materials. Chemicals and materials were purchased from the following sources: Amersham International pic. (Little Chalfont. Buckinghamshire. England) BIOTRAK ® cGMP scintillation proximity assay kit, [ y-32P]-adenosine 5'-triphosphate Bachem Inc. (Torrance. California. U.SA) cGMP-dependent protein kinase substrate (BPDEtide), atrial natriuretic factor (rat ANF) BDH Inc. (Vancouver. British Columbia, Canada) sodium hydrogen carbonate, D-glucose, trichloroacetic acid Biolog Life Science Institute. (Bremen. Germany ) Rp - 8 - (4 - Chlorophenylthio) - guanosine - 3',5' - cyclic monophosphothioate (Rp -8 -pCPT-cGMPS) Biomol Research Laboratories Inc. (Plymouth Meeting. PA. USA) cAMP-dependent protein kinase holoenzyme, a catalytic subunit of cAMP-dependent protein kinase (porcine heart), cGMP-dependent protein kinase 1a holoenzyme (bovine lung) 31 Bio-Rad Laboratories. (Hercules. California. USA). Acrylamide (99.9%), ammonium persulfate, goat anti-rabbit IgG alkaline phosphatase conjugate, N,N,N',N'-tetramethyl ethylenediamine (TEMED), sodium dodecyl sulfate (SDS), N,N'-methylene-bis-acrylamide (bis-acrylamide), protein assay dye (161-0400) Carnation Inc. (Toronto. Ontario. Canada) Skim milk powder DuPont NEN ® Research Products (Boston. MA. USA) [ y-32P]-adenosine 5'- triphosphate Fisher Scientific Co. (Fair Lawn. New Jersey. USA) Calcium chloride, o-phosphoric acid, magnesium acetate, magnesium sulphate, potassium chloride, potassium phosphate (monobasic), potassium phosphate (dibasic), Scintiverse ® scintillation fluid, Organon Technika Inc. (Toronto, Ontario. Canada) Heparin sulfate MTC Pharmaceuticals. (Cambridge. Ontario. Canada) Somnotol ® (sodium pentobarbital) Parke Davis Inc. (Scarborough, Ontario. Canada) Nitrostat ® tablets (0.6 mg nitroglycerin per tablet) 3 2 Peninsula Laboratories Inc. (Belmont. California. U.S.A.) Atrial natriuretic factor (rat ANF) Sigma Chemical Co. (St. Louis. Missouri. USA) Adenosine 5'-triphosphate (ATP), benzamidine, ethylenediamine tetra acetic acid (EDTA), protein kinase inhibitor (PKI), adenosine 3', 5'-cyclic monophosphate (cAMP), guanosine 3', 5'-cyclic monophosphate (cGMP), L-phenylephrine, leupeptin, phenylmethylsulfonyl fluoride (PMSF), sodium nitroprusside, 3-isobutyl-1-methylxanthine (IBMX), soybean trypsin inhibitor, Triton ® X-100, Trizma base, Trizma HCI Whatman Ltd. (Maidstone. Kent. England) Phosphocellulose paper (P81) 2.2. Animals. White New Zealand rabbits of either sex (2.5 to 3.0 kg) and male Wistar rats (250 to 300 g) were obtained from the Animal Care Unit, University of British Columbia and housed either individually (rabbits) or three to a cage (rats). All animals had free access to water and food. 33 2.3. Isolated tissue preparation. 2.3.1. Preparation and handling of rabbit aorta. The rabbits were first injected (intravenously into the ear vein) with sodium pentobarbital (65 mg/kg) and then exsanguinated. The descending thoracic aortas were excised, placed in Krebs-bicarbbnate buffer of the following composition (mM): KCI (4.75), K H 2 P 0 4 (1.2), MgS0 4 (1.2), CaCI2 (2.5), NaCI (118), NaHC0 3 (25), D-glucose (11.12) and gassed with 95% 02/5% CO2, and carefully trimmed of adhering fat and connective tissue. A stainless steel rod was then inserted into the lumen and the aorta cut into a helical strip. The endothelial layer was removed by gently rubbing a glass rod across the exposed lumen. From this strip, smaller segments ( « 7 mm wide and 15-18 mm long) were cut. These strips were then set up for isometric tension recording by attaching stainless steel metal clips to the two ends. One end was anchored to a hook at the bottom of the isolated organ bath and the other end attached to a force-displacement transducer (Grass Model FT03C) which was connected to a Grass Model 7D polygraph recorder. Strips were equlilibrated for 60 min at 37° C under 2 g tension in Krebs-bicarbonate buffer and gassed with 95% 02/5% CO2. When determining the effect of the PKG inhibitor (Rp-8-pCPT-cGMPS), rabbit aorta medial strips were used. Medial strips were separated from the adventitia of rabbit aorta according to the method of Becker et al. (1990). These 34 strips (10 mm x 5 mm) were then set up for isometric tension recording as above, but under a 1 g preload tension. 2.3.2. Preparation and handling of rat vas deferens. Male Wistar rats (275-300 g) were sacrificed by carbon dioxide exposure. Vasa deferentia were quickly and carefully (such that the tissue was not stretched) removed and placed in warm (37° C) Krebs-bicarbonate buffer of the following composition (mM): KCI (4.75), KH 2 P0 4 (1.2), MgS0 4 (1.2), CaCI2 (2.5), NaCI (118), NaHCC-3 (25) and D-glucose (11.12) and aerated with 95% 02/5% CO2. The vas deferens ( « 30 mm long) was then gently sqeezed to remove any material in the lumen. The ends were tied off with 4-0 thread and one end was then anchored to the bottom of the isolated organ bath and the other end tied to a force-displacement transducer (Grass Model FT03C) which was connected to a Grass Model 7D polygraph recorder. The vas deferens was equilibrated for 90 min under 2 g of tension and aerated with 95% 02/5% C 0 2 The buffer was changed every 15 min. 2.3.3. Preparation and handling of rat proximal and distal colon. Male Wistar (275-300g) rats were sacrificed by carbon dioxide exposure. The proximal colon (defined as the ascending colon up to the transverse colon) and the distal colon (defined as the descending colon) were cut out and placed in warm 35 (37° C) Tyrodes buffer of the following compostion (mM): KCI (2.7), NaH2PC>4 (0.4), MgCI2 (1.05), CaCI2 (1.8), NaCI (136.9), NaHC0 3 (11.9) and D-glucose (5.6). The segments were flushed thoroughly with Tyrodes buffer (to remove fecal material). After cleaning, 2.0 cm segments (2 each from proximal and distal colon, per rat) were set up for tension recording in isolated organ baths under a 1g pre-load tension and aerated with 95% 02/5% C 0 2 . The segments were set up such that the stainless steel clips were attached to both the oral and anal ends of the segment. The anal end was attached to an anchor in the organ bath. The oral end was then tied (via a thread on the clip) to a force-displacement transducer (Grass Model FT03C) which was connected to a Grass Model 7D polygraph recorder and the segments were equilibrated for at least 30 min during which time the buffer was changed every 10 min. 2.4. Isolated tissue treatment protocol. 2.4.1. Rabbit aorta. After the initial equilibration period, rabbit aortic strips were contracted by adding phenylephrine (PE) (final bath concentration 1.0 uM which is equivalent to an EC 8 0) for 5 min, after which the tissues were washed with buffer for a further 60 min (4 washes). Tension was adjusted, as required, to maintain pre-load tension. The strips were again contracted with 1.0 uM PE for 5 min and then washed as 36 before. After a further 60 min (4 washes in between) the strips were again contracted with 1.0 uM PE. When strips had contracted for 5 min (steady-state contracture), nitroglycerin (GTN) was added to the bath in a concentration-dependent manner (0.01 uM to 10 uM for 2 min) or time-dependent manner (10 uM for various time periods of 30 s to 10 min). Each strip was treated with only one concentration/time of GTN. A similar protocol was also used for the SNP study (0.1 to 10.0 uM SNP for 2 min and 10 uM SNP for 30 s to 2 min). The strips were then frozen between tongs pre-cooled in liquid nitrogen at the relevant time periods and stored for biochemical analysis (cGMP levels and PKG activity ratio). The degree of relaxation was then calculated as a percentage of maximal contraction induced by 1.0 uM PE. At least one strip from each rabbit was treated as a control (i.e. did not receive any drug). In experiments in which the effects of PKG inhibitors were studied, medial strips rather than intact rabbit aortic strips were used. Contractions of medial strips were induced by adding PE (1.0 uM final bath concentration) for 5 min and the strips washed 4 times over the next 60 min. The contraction was repeated an hour later. After repeated washing of the strips and after the tension reached baseline, some of the strips were exposed to Rp-8-pCPT-cGMPS (30 uM) for 40 min. At least two strips from each rabbit were designated as control, and did not receive the inhibitor. At the end of the 40 min period, the strips were contracted with 1.0 uM PE and 20 min later either GTN (0.01 nM to 10 uM) or SNP (0.01 nM to 10 uM) was added in a cumulative manner to relax the medial strips. One control strip was 37 relaxed with GTN and the other with SNP. Relaxation was then calculated as a percentage of the maximal contraction induced by 1.0 uM PE. 2.4.2. Rat vas deferens. After an initial equilibration period, vasa deferentia were contracted with 3 uM PE (equivalent to an EC 4 0), washed twice with buffer after peak tension was generated and allowed to relax until tension reached baseline (within a min). This was repeated two more times at 5 min intervals. After the third contraction, the strips were allowed to re-equilibrate for a 30 min period. One strip from each rat was always treated as a control and the other with SNP. Treated strips were treated either with 0.1 mM SNP or 5 mM SNP for 90 s. During the last 30 s of the 90 s period the strips were contracted with 3 uM PE and then frozen between tongs pre-cooled in liquid nitrogen. The tension in control strips was then compared to tension in SNP treated strips on a paired basis. 2.4.3. Rat proximal and distal colon. Thirty min after the segments started to spontaneously contract, one segment, from each of the proximal and distal colon, was exposed to 0.1 uM ANF for 2 min, then frozen between tongs pre-cooled in liquid nitrogen and stored for biochemical analysis. The other (control) segments received an equivalent volume (200 pi) of the vehicle (0.05 M acetic acid) and were frozen, as above, 2 min later. Thus, with this protocol, each treated segment had its paired control segment. 38 2.5. cGMP estimation in rabbit aorta, rat vas deferens, rat proximal and distal colon. Frozen samples (15 to 50 mg) were placed in a 1 ml capsule, pre-cooled by placing on dry ice, containing a chilled metal pestle. Samples were pulverized using a Vari-Mix III dental amalgam mixer (5 s at medium speed). Then 0.5 ml of ice-cold TCA (6 % w/v) was added to the capsule and the pulverized tissue homogenized by subjecting the capsule to a 15 s medium and 5 s high speed agitation. The homogenized tissue was then removed and the capsule washed with 0.5 ml TCA, which was added to the first homogenate (total volume 1.0 ml). The crude homogenate was then centrifuged in either a Sorval RC2-B centrifuge (SM 24 rotor) or a Heraeus Contifuge 28RS centrifuge (rotor 3744) at 2000 x g for 15 min. The supernatant was aspirated and the TCA extracted 4 times with 5 ml water-saturated ether (4°C). The upper ether phase was discarded after each wash. Any remaining ether was evaporated by placing the glass tubes in a hot (« 70° C) water bath. cGMP levels were then determined in the aqueous solution by using a commercially available cGMP radioimmunoassay kit (cGMP-SPA [scintillation proximity assay] Amersham). The samples were acetylated, as suggested by Harper & Brooker (1975), to increase the sensitivity of the assay. Tissue cGMP levels were calculated as pmole cGMP per gram (g) wet weight of tissue. 39 2.6. Preparation of extracts and assay of cGMP- and cAMP-dependent protein kinase. 2.6.1. Extraction of soluble and particulate fractions. At least 80 mg of frozen tissue were placed in a pre-cooled (dry ice) 1 ml capsule containing a chilled metal pestle and pulverized by using a Vari-Mix III dental amalgam mixer (5 s at medium speed). The pulverized sample was then homogenized (15 s at medium and 5 s at high speed agitation) in 5 volumes of ice-cold buffer of the following composition: 100 mM potassium phosphate (pH 6.8), 1.0 mM IBMX , 10.0 mM EDTA and 10.0 mM 2-mercaptoethanol. The homogenate was then centrifuged at 12,000 x g for 15 min in either a Sorval RC2-B centrifuge (SM 24 rotor) or a Heraeus Contifuge 28RS centrifuge (rotor 3744). The homogenization step did not take more than 1 min. The resultant supernatant was then immediately used in the PKG assay. For those studies in which PKG activity was measured in the particulate fraction, the pellet was first rinsed with 200 ul of homogenization buffer and then resuspended in 5 volumes of homogenization buffer, now supplemented with 0.1% (w/v) Triton X-100 and kept on ice. The resuspended pellet was agitated on a low speed vortex for 3 x 10 s every 5 min, for 30 min. The resuspended pellet was then centrifuged at 12,000 x g for 15 min and the supernatant assayed for PKG activity. 40 2.6.2. Assay of cGMP- and cAMP-dependent protein kinase. The PKG activity was determined by measuring the transfer of the [y-32P] phosphoryl group of ATP to an acceptor peptide substrate. The unreacted ATP was resolved from the radioactive peptide substrate according to the method of Witt and Roskoski (1976). The reaction was initiated by adding 20 pi of muscle extract or MonoQ fractions (see section 2.7 below) to 50 pi of a reaction mixture containing 40 mM Tris buffer (pH 7.4), 2 mM magnesium acetate, 200 uM [ y-32P] ATP (specific activity « 300 cpm/pmol for assay of soluble and particulate fractions of smooth muscle preparations and « 150 cpm/pmol for assay of MonoQ fractions), 100 pM IBMX, 150 pM BPDEtide (RKISASEFDRPLR) and 1 pM synthetic protein kinase inhibitor (PKI) (Cheng et al., 1986) to inhibit PKA. This concentration of PKI has been shown to inhibit PKA activity in rat vas deferens by at least 97 % (Hei at al., 1990). The reaction was allowed to proceed for 10 min at 0-4 °C (30°C for MonoQ fractions) either in the absence or presence of 5 pM cGMP. To correct for phosphorylation of endogenous substrates, no-substrate (i.e. no BPDEtide) blanks were determined. The reaction was terminated by spotting 50 pi of the reaction mixture onto 1 x 2 cm P81 phosphocellulose paper (Whatman). The paper was then dropped into 0.5% phosphoric acid and washed 4 times for 5 min each. The papers were allowed to dry, placed in scintillation vials, scintillant added and counted either in a Packard Tricarb or Beckmann LS500 liquid scintillation counter for 2 min/vial. PKG activity was expressed as pmol of phosphate transferred per minute per mg of protein. The degree of PKG activation was determined by 41 calculating the PKG activity ratio. This ratio is the activity in the absence of added cGMP (reflecting endogenous PKG activation) divided by the activity in the presence of 5 uM cGMP (total activity). PKA activity in the fractions eluted from the MonoQ column chromatography of the various smooth muscle preparations was measured using the same assay conditions as for PKG, with two exceptions. Firstly, the assay was conducted in the absence or presence of 5 uM cAMP and, secondly, PKI was not added to the reaction mixture. 2.7. Column chromatography of PKG and PKA in muscle extracts . Both PKG and PKA were separated using a Pharmacia MonoQ® anion exchange column (HR5/5) coupled to a FPLC® system (Pharmacia LKB Biotech, Uppsala, Sweden). Frozen smooth muscle samples were first homogenized in buffer A (5 mM Tris-HCI, pH 7.4, 2 mM EDTA, 1 mM DTT) containing soybean trypsin inhibitor (10 ug/ml), benzamidine (1 mM), leupeptin (2 ug/ml), pepstatin (10 ug/ml) and phenylmethylsulfonyl fluoride (PMSF) (1 mM) using the same protocol as in Section 2.4. The soluble extract was then diluted in homogenization buffer to yield a concentration of 12 mg/10 ml. This was then loaded onto the MonoQ column, pre-equilibrated with buffer A, at a flow rate of 0.5 ml/min. The column was developed at 0.5 ml/min with a linear NaCI gradient of 0 - 400 mM in buffer A and 40 fractions, 0.5 ml each, were collected. All procedures were done at 4°C. The fractions were then assayed, as described above for PKG and PKA activity at 30°C. 42 Fractions which showed maximal PKG and PKA activity were then immunoblotted for the presence of PKG. 2.8. Immunoblotting. Separation of PKG was accomplished in a Bio-Rad Protean II Electrophoresis unit by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS PAGE) according to the method of Laemmli (1970). The separating gel contained 11% (w/v) total acrylamide (acrylamide: N,N'-methylene bisacrylamide = 37.5:1), 375 mM Tris-HCI, pH 8.8, 0.1% (w/v) SDS, 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 mM Tris-HCI pH 6.8, 0.1% (w/v) SDS, 0.08% (w/v) ammonium persulfate and 0.05% (w/v) TEMED. Aliquots (200 pi) effractions from column chromatography of muscle extracts were boiled for 3 min in 2% (w/v) SDS, 120 mM Tris-HCI, pH 6.8, 10% glycerol, 5% p-mercaptoethanol and 0.004% bromophenol blue. Molecular mass pre-stained standards (Kinetek Biotechnology Corp, Vancouver, B.C. Canada) used were 97.4 kDa (rabbit muscle phosphorylase b), 66.2 kDa (bovine serum albumin), 55.3 kDa (hen egg white ovalbumin), 33.2 kDa (glutathione-S-transferase), 23 kDa (soybean trypsin inhibitor) and 15.6 kDa (cytochrome C) and were treated similarly. Samples (120 pi) 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 43 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, SF, California, USA) unit for transfer of the resolved proteins onto a nitrocellulose membrane (Protran®, Schleicher & Schuell, Keene, NH, USA) (Tobwin et al., 1979). Transfer was achieved by applying a constant current of 250 mA for 3 h at 4°C across a bath solution [20% (v/v) methanol, 20 mM Tris, 120mM glycine and 0.008% (w/v) SDS]. The nitrocellulose membrane was treated for 2 h in blocking buffer which consisted of 3% (w/v) skim milk powder in TTBS [20 mM Tris-HCI, pH 7.4, 0.5 M NaCI and 0.05% (w/v) Tween 20] to eliminate non-specific binding and then washed with TTBS (4x5 min). The membrane was then probed with a polyclonal, affinity-purified antibody raised against a peptide sequence (CDEPPPDDNSGWDIDF) derived from the C-terminus of the 1a isoform of PKG. The membrane was exposed to the primary antibody (1/250 diluted (v/v) in antibody buffer which was TTBS and 0.05% (w/v) sodium azide) overnight at room temperature. The membranes were then washed (4x5 min) in TTBS and incubated for 2 h with the secondary antibody (1/2000 diluted (v/v) goat anti-rabbit IgG alkaline phosphatase conjugate). The membranes were washed again with TTBS (4x5 min) and rinsed twice for 2 min 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-3-indolyl phosphate (BCIP). The reaction was performed in 50 ml alkaline 44 phosphatase buffer (100 mM Tris-HCI, pH 9.5, 100 mM NaCI and 5 M MgCI2) to which NBT (170 pi of a 50 mg/ml stock in 100% DMF) and BCIP (340 pi of a 50 mg/ml stock in 70% DMF) were added. 2.9. Preparation of drug solutions. Solutions of phenylephrine were prepared in distilled water. SNP was prepared in distilled water immediately before use. GTN was prepared by dissolving a tablet of Nitrostat® in distilled water immediately prior to use. Rp-8-pCPT-cGMPS was prepared in distilled water and stored at -20° C until required for use. Further dilutions were made in distilled water. 2.10. Protein determination. Protein concentrations were determined using dye reagents (Bio-Rad), based on the method of Bradford (1976). To determine the protein concentration, the sample was mixed with the dye reagent, which was Coomassie brilliant blue G-250, and absorbance measured at 596 nm. The protein standard curve was obtained by using BSA as the standard. 2.11. Statistical analysis. Values in the drug-treated groups were compared to their respective controls using a Students t test statistical program (SigmaStat V 1.0, Jandel Scientific, San 45 Mateo, California, USA). A probability (p) of less than 0.05 was accepted as the level of significance. In all experiments, at least one tissue was treated as control. All values are expressed as the mean ± standard error of the mean (S.E.M.). 46 3.0. RESULTS 3.1. Optimization of PKG Assay Parameters. 3.1.1. Linearity of phosphotransferase activity. Since PKG activity, using BPDEtide as a substrate, has been previously measured in only three smooth muscle preparations (porcine coronary artery, bovine carotid artery, human umbilical artery), various parameters of the assay conditions were determined to ensure that the conditions of the PKG assay were optimal in rabbit aorta. The first point to consider was whether the reaction (i.e. phosphotransferase activity) was linear over time. To determine this, total phosphorylation in presence of 5 uM cGMP was measured in rabbit aorta control strips (PE-stimulated, 1 uM, EC 8 0) at various time periods as shown in Figure 1. The graph clearly shows that phosphorylation, as indicated by the amount of phosphate (Pi) transferred to the substrate, is linear for at least 15 min (the maximum time measured) after initiation of the reaction. 3.1.2. Determination of kinetic parameters of BPDEtide phosphorylation and ATP utilization in crude soluble rabbit aorta homogenate. Since ATP is the source of Pi, and BPDEtide the acceptor of Pi, it was of interest to determine whether the concentrations of ATP (200 uM) and BPDEtide (150 uM) (Jiang et al., 1992) were optimal for the PKG assay in the rabbit aorta. 47 Figure 1. Time-course of cGMP-dependent protein kinase assay. Phosphotransferase activity of crude soluble homogenates from rabbit aorta were assayed using BPDEtide (150 uM) in the presence of 5 uM cGMP. The reaction was allowed to proceed for various time periods and stopped by spotting 50 pi of the mixture onto P81 phosphocellulose paper. Free [y-32P] ATP was removed by a phosphoric acid (75 mM) wash protocol (4x5 min). The P81 papers were then counted by liquid scintillation counting. Each point is the mean (± S.E.M.) of n = 5 experiments. The average protein concentration was 0.4 ug/ul. 48 Time (min) 49 Figure 2. Determination of kinetic parameters of BPDEtide and ATP in rabbit aorta. The Km's for both BPDEtide and ATP were determined by measurement of phosphotransferase activity of PKG in rabbit aorta crude soluble homogenates when (A) BPDEtide concentration was varied (ATP concentration was fixed at 200 pM) and (B) ATP concentration was varied (BPDEtide concentration was fixed at 150 pM). PKG phosphorylation (rate) was plotted against both BPDEtide and ATP concentrations. Insets are double reciprocal Lineweaver-Burke plots. Kinetic parameters were plotted by a computer program (ENZFITTER). 50 Accordingly, Km values of both ATP and BPDEtide were determined in crude soluble fractions from control rabbit aortic strips. Figure 2A shows an example of the effect of varying BPDEtide on phosphorylation in aorta from one rabbit. In this study, the apparent K m for BPDEtide was 159 pM. When ATP concentration was the dependent variable, the apparent Km for ATP was 70 uM (Figure 2B). Similar results were obtained in two different experiments. 3.1.3. Effect of different reaction times on PKG activity ratios. Since cGMP has been reported to dissociate from its binding sites on PKG rather rapidly ( r 1 / 2 of « 0.6 min, McCune & Gill, 1979), it was important to determine the optimal reaction time period (time during which phosphotransferase activity was measured) which would give consistent results. Experiments were conducted in which PKG activity ratios were determined in control, GTN (130 pM/2 min)- and SNP (10 pM/2 min)-treated rabbit aorta at different reaction times. In all experiments, reactions were started after homogenization and centrifugation of crude extracts, which took about 17 minutes. Figure 3 shows the results from this study. During the first 5 min of reaction, PKG activity ratios changed considerably compared to the ratios at the 1 min period. For example, in GTN-treated strips, activity ratios diminished from an initial value of 0.63 ±0 .1 to 0.36 ± 0.03 and in SNP-treated strips from 0.43 ± 0.16 to 0.32 ± 0.03 at the 5 min period. A similar reduction in activity ratio was also seen in control strips (0.56 ±0.1 to 0.20 ± 0.08) 52 Figure 3. Effect of reaction time on PKG activity ratios in rabbit aorta. PKG activity ratios in control (•), GTN- (•, 130 uM/2 min) and SNP- (•, 10 uM/2 min) treated rabbit aorta were determined at different time periods after initiation of the reaction, as outlined in Materials and Methods. Activity ratios are the activity in the absence of added cGMP (reflecting endogenous PKG activation) divided by the activity in the presence of 5 pM cGMP (total activity). Each data point is the mean (±S.E.M.) of n = 4 experiments. * = significant difference (p<0.05) compared to control. NS = no significant difference (p>0.05) compared to control. 53 54 during the same time period. Activity ratios remained relatively constant between the 5 min and 15 min reaction time period. The activity ratios at 5, 10 and 15 min were significantly elevated in GTN-treated strips compared to control values (p<0.05). The activity ratios were more consistent at 5, 10 and 15 min, as reflected by smaller standard error values when compared to ratios measured at 1 and 2 min. 3.1.4. Effect of PKI on phosphorylation of BPDEtide. PKG assays were generally performed with a suitable amount of PKA inhibitor (PKI) present in the reaction mixture. This was done to block any phosphorylation, by cGMP-activated PKA, of the substrate. Accordingly, the effects of 1.0 uM synthetic PKI (Cheng et al., 1986) on basal (-cGMP) and total (+cGMP) PKG activities were determined in rabbit aorta. Basal PKG activity in the absence of PKI was 12.5 pmol/min/mg and this was reduced in the presence of PKI to 10 pmol/min/mg ( « 25% inhibition) (see Figure 4). Total (+cGMP) PKG activity was also reduced in the presence of PKI by about 12% (from 162 to 143 pmol/min/mg). These data indicate that phosphorylation by PKA of BPDEtide does take place but not to a great extent, since total phosphorylation (+cGMP) was reduced by only 12%). PKG activity ratios were the same in the presence or absence of PKI. 55 Figure 4. Effect of cAMP-dependent protein kinase inhibitor on rabbit aorta soluble PKG activity. In control rabbit aortic strips basal (-cGMP) and total (+cGMP) PKG activities were measured in the presence and absence of PKI (1.0 pM). The bars represent mean values from 2 experiments. 56 57 3.1.5. Chromatographic separation and identification of PKG in rabbit aorta, rat vas deferens, rat proximal and distal colon. PKG activity in crude soluble fractions from the above tissues was resolved by using MonoQ anion exchange chromatography on an FPLC system. The collected fractions were then each assayed for PKG activity in the absence and presence of 5 pM cGMP. To block out any effect PKA may have had on phosphorylation, 1 pM PKI was added. The fractions were also assayed for PKA activity using the same protocol as for the PKG assay, except that no PKI was present and the assay was done in the absence and presence of 5 pM cAMP. Figure 5 shows that two main peaks were resolved by MonoQ anion exchange chromatography of soluble fractions from rabbit aorta (A), one of which was PKG (peak at fraction 28) and another which was PKA (fraction 10). Although we do not have direct proof, based on the elution profiles previously reported (Silver et al., 1982; Hei et al., 1990) the PKA peak most probably reflects Type I PKA activity. A smaller PKG peak was also observed around fraction 22. In the rat vas deferens (Figure 5B), three peaks were resolved: two for PKA (peak activities fractions 10 and 34) and one for PKG (peak activity at fraction 27). The two PKA peaks in the vas deferens correspond to PKA Type I (fraction 10) and Type II (fraction 34). In the vas deferens, the isozyme profile for both types of PKA was very similar to that seen previously (Hei et al., 1991) in rat vas deferens. The 58 PKG activity was much higher than that of PKA activity in both tissues, using the relatively more specific PKG substrate BPDEtide. Figure 6 shows that three distinct peaks were resolved in both the rat proximal (Figure 6A) and distal colon (Figure 6B): two for PKA and one for PKG. In contrast to the rabbit aorta and the vas deferens, the PKA Type II activity is much greater and in the case of the proximal colon is greater than the activity for PKG. However, maximum PKG activity is much lower (275 - 400 fmol/min/pl) than that seen in the rabbit aorta (6500 fmol/min/pl). The chromatographic study shows that the assay conditions used to detect PKG activity were selective for PKG, since there was no PKG activity in the Type I PKA peaks. The data also show that, as expected, 5 pM cAMP could also activate PKG: notice the increase in PKG activity in rabbit aorta (fractions 26 to 31) and vas deferens (fractions 24 - 29) (+cAMP, -PKI) (inverted triangles Figure 5A and B) which parallels the activity in the presence of cGMP (+) and PKI. In the colon, there was a similar activation of PKG by 5 pM cAMP (fractions 22 - 28 in proximal colon and fractions 23 - 28 in the distal colon). In addition, there was very little cyclic nucleotide-independent phosphorylation. To confirm the identity of PKG in those fractions which exhibited both PKG and PKA activities, Western blots were performed. The fractions which showed maximum activity were subjected to SDS-PAGE, transferred to a nitrocellulose 59 Figure 5. MonoQ column chromatography of rabbit aorta and rat vas deferens PKG and PKA. Crude smooth muscle extracts (15 mg) from control rabbit aorta (A) and rat vas deferens (B) were fractionated on a MonoQ column using a Pharmacia FPLC system with a linear gradient of NaCI (0-400 mM). The gradient was applied over 40 fractions of 0.5 ml each and the fractions collected and assayed for PKG (in the presence and absence of 5 uM cGMP) and PKA (in the presence and absence of 5 pM cAMP) phosphotransferase activity. Phosphotransferase activity was measured at 30°C for 10 min as described in Section 2.6.2. Similar results were obtained when the experiment was repeated. Results shown represent data from a single experiment. 60 7000 10 15 20 25 30 MonoQ fraction number 61 Figure 6. MonoQ column chromatography of rat colon PKG and PKA. Crude smooth muscle extracts (15 mg) from rat proximal colon (A) and distal colon (B) were fractionated on a MonoQ column using a Pharmacia FPLC system with a linear gradient of NaCI (0-400 mM). The gradient was applied over 40 fractions of 0.5 ml each and the fractions collected and assayed for PKG (in the presence and absence of 5 pM cGMP) and PKA (in the presence and absence of 5 pM cAMP) phosphotransferase activity. Phosphotransferase activity was measured at 30°C for 10 min as described in Section 2.6.2. Similar results were obtained when the experiment was repeated. Results shown represent data from a single experiment. 62 300 0 5 10 15 20 25 30 35 40 Mono-Q fraction number 63 Figure 7. Immunoblots of PKG- and PKA-containing fractions. MonoQ resolved fractions from smooth muscle which contained maximum kinase activity were probed with antibody raised against a peptide sequence from the C-terminus of the la isoform of PKG. The numbers under each smooth muscle type refer to fraction numbers resolved by MonoQ which were probed with the antibody (see Figures 5 and 6). Controls included PKA holoenzyme (PKA), the catalytic subunit of PKA (PKA catalytic) and PKG la holoenzyme (bovine lung). 64 o CO Aorta Distal colon Q Prox colon Vas deferens 6 6 ° 10 28 8 28 30 Q- o_ n. 9 27 30 10 27 34 65 membrane and then probed with antibodies to PKG as indicated in Section 2.8. The results from the immunoblots are shown in Figure 7. The blots show that maximum immunoreactivity was found in those fractions which had the greatest PKG activity and with the PKG la holoenzyme control. There was no immunoreactivity in the PKA standards, indicating that the antibodies were PKG-specific. There was no immunoreactivity with tissue fractions which corresponded to Type I PKA except in fraction 10 of the aorta where moderate immunoreactivity was present at 60 kDa and 45 kDa. The identity of these bands is unclear at this time but are unlikely to be due to PKA (see above). In the above experiments* only the peak fractions were immunoprobed with antibodies. It was of interest to test whether the immunoreactive bands closely paralleled the PKG activity of the fractions resolved by MonoQ column chromatography. Accordingly, PKG activity was resolved in rabbit aorta as above and fractions were probed with antibodies to PKG using the same protocol. Figure 8 shows the MonoQ profile of PKG activity in rabbit aorta. Peak PKG activity was seen at the same location as before (see Figure 5A). The small peak seen at fraction 22 was also consistent with a similar peak observed before and shown in Figure 5A. There was no cyclic nucleotide-independent phosphorylation observed. Strong immunoreactive bands were observed in those fractions which exhibited high PKG activity (fractions 27 to 31). There were two bands in each fraction and the higher molecular weight bands were more intense than the lower bands. Keilbach et al. (1992) have reported similar findings in COS-7 cells transfected with 66 Figure 8. MonoQ column chromatography and immunoblot of rabbit aorta PKG. Crude smooth muscle extracts (10 mg) from rabbit aorta were fractionated on a MonoQ column using a Pharmacia FPLC system with a linear gradient of NaCI (0-400 mM). The gradient was applied over 40 fractions of 0.5 ml each, and the fractions assayed for PKG phosphotransferase activity (in the presence and absence of 5 pM cGMP) (upper panel). Phosphotransferase activity was measured at 30°C for 10 min as described in Section 2.6.2. Selected fractions were resolved on 11%(w/v) SDS-polyacrylamide gels and probed with antibody raised against a peptide sequence from the C-terminus of the la isoform of PKG. The immunoreactive bands in fractions probed with the antibody are shown in the bottom panel. 67 68 the cDNA of the ip isozyme, except that in their studies the major band had the lower molecular mass (77 kDa) compared to the minor band (81 kDa). There was no immunoreactive band, at the same molecular mass position (77 - 81 kDa), for fraction 22 and 23, which showed PKG activity. However weak immunoreactive bands were observed in fractions 22 and 23, but at a lower molecular mass position (66.2 kDa) 3.1.6. Inhibition of PKG phosphotransferase activity in vitro. Further attempts to validate that the phosphotransferase activity measured in our experiments was due to PKG were made using Rp-cGMP analogs which have been reported to be relatively selective PKG inhibitors (Butt et al., 1990). The inhibition by Rp-8-pCPT-cGMPS (0.3 - 1000 pM) of PKG-mediated phosphorylation of BPDEtide was studied in crude soluble extracts of rabbit aorta, rat vas deferens, rat proximal and distal colon. Initial attempts to measure inhibition at 0°C were unsuccessful and subsequently inhibition was measured at 30°C. Rp-8-pCPT-cGMPS inhibited, in a concentration-dependent manner, PKG (+cGMP)-mediated phosphorylation of BPDEtide in all the tissues studied (Figure 9). 69 Figure 9. Effect of Rp-8-pCPT- cGMPS on PKG activity in rabbit aorta, rat vas deferens, rat proximal and distal colon. The effects of various concentrations of Rp-8-pCPT-cGMPS on total PKG (+cGMP) activity were determined in crude soluble fractions from rabbit aorta and rat vas deferens (A) and rat proximal and distal colon (B) under assay conditions as described in Section 2.6.2 but at 30°C. Each point is the mean (± S.E.M.) of n = 5 experiments. 70 Log (M) Rp-8-pCPT-cGMPS Log (M) Rp-8-pCP-cGMPS 71 3.2. Effect of nitrovasodilators on contractility, cGMP levels and PKG activity ratio in smooth muscle. 3.2.1. Effects of SNP and GTN on contractility. PKG activity ratio and cGMP levels in rabbit aorta. The effects of SNP and GTN on the above parameters in rabbit aorta were studied in both a concentration-dependent and time-dependent manner. No previous studies have been reported in which contractility, PKG activity ratio and cGMP levels have been measured simultaneously in the same muscle strips. In preliminary experiments, it was found that the tonic contraction was maintained for at least 30 min. Figure 10 shows representative contractions, induced by 1 pM PE (EC80), of rabbit aorta and the general protocol used in this study. 3.2.1.1. Sodium nitroprusside. PE-contracted rabbit aorta strips were exposed to 0.1, 1.0 and 10.0 pM SNP for 2 min, as outlined in the Materials and Methods Section. These concentrations of SNP were found to inhibit the PE response by approximately 50, 85 and 98%, respectively, based on preliminary cumulative SNP concentration-response curves obtained in PE-contracted aorta (data not shown). Two min exposures to SNP (0.1, 72 Figure 10. Effects of GTN and SNP on PE (1.0 uM) -induced contraction in rabbit aorta. Representative scanned images of responses in two strips (A and B) from one rabbit are shown. After an initial equilibration period, strips were stimulated with 1.0 pM PE for 5 min. Sixty min later, strips were stimulated with PE and treated either as control (A) or SNP-treated (B) and frozen (double lines) after 2 min (concentration study) or different time periods (temporal study). The same protocol was followed in strips where GTN was used instead of SNP. 73 A r J • • = P E 1 LIM B t S N P O 5 min 74 TABLE 1. EFFECT OF SODIUM NITROPRUSSIDE (0.1 -10 uM / 2 min) ON PKG ACTIVITY, cGMP LEVELS AND CONTRACTILITY IN PHENYLEPHRINE-CONTRACTED RABBIT AORTA. cGMP (pmol/g tissue) pmol PO4/ min/ mg (-cGMP/+cGMP) Activity Ratio Relaxation (%) Control SNP (0.1 uM) 0.37 ±0.06 1.0 ± 0 . 2 * 44 ± 6 / 281 ± 56 55 ± 8 / 2 7 3 ±41 0.16 ±0.02 0.22 ± 0 . 0 4 * 18 ± 4 * Control SNP (1.0 uM) 0.27 ± 0.04 3.7 ± 0 . 5 * 44 ± 3 / 2 3 5 ±31 63 ± 12/233 ±47 0.19 ±0.02 0.28 ±0.02 * 58 ± 4 * Control SNP (10.0 uM) 0.28 ±0.03 1 6 ± 4 * 54 ± 8 / 283 ± 34 87 ±13* / 247 ±31 0.19 ±0.02 0.36 ± 0 . 0 3 * 71 ± 3 * Note: Values are mean (± S.E.M.) for n = 11, 6 and 10 for 0.1, 1.0 and 10.0 uM SNP, respectively . Means were compared to their controls by a paired Student's t test. * = Significant difference from respective control (p<0.05). 1.0 and 10.0 uM) induced concentration-dependent relaxations of PE-contracted rabbit aorta while at the same time significantly elevating cGMP levels by approximately 3-, 14- and 57-fold, respectively (see Table 1). PKG activity ratios were significantly increased by all three concentrations of SNP (1.3-, 1.4- and 1.9-fold by 0.1, 1.0 and 10.0 pM SNP, respectively). These increases correlated reasonably well with relaxation (18, 58 and 71%, respectively). The data also showed that total PKG (+cGMP) activity was not significantly changed by SNP. The increases in activity ratio in these experiments were due to increases in endogenous or basal PKG activity (-cGMP). The temporal effect of 10 pM SNP was also determined in PE-contracted rabbit aorta. The protocol was similar to that described above except that the strips were frozen at 10 s, 30 s and 2 min following the addition of SNP. The data in Table 2 demonstrate that as early as 10 s after addition of SNP, cGMP was significantly elevated (by 12-fold) and this was accompanied by a significant (1.3-fold) increase in the PKG activity ratio and inhibition of the contraction by 7 %. The increase in PKG activity ratio reached a maximum (2-fold increase) 30 s after SNP, when cGMP level was further elevated (28-fold) and the strips were relaxed by 29 %. Two min after SNP, the PKG activity ratio had not increased any further (1.9-fold increase), cGMP was even further elevated (57-fold) and relaxation had reached (71 %). Total PKG activity (+cGMP) was somewhat lower in SNP-treated strips compared to control strips at all time periods; however, this decrease was not significant. 76 TABLE 2. TIME COURSE OF SODIUM NITROPRUSSIDE (10 uM) EFFECT ON PKG ACTIVITY, cGMP LEVELS AND CONTRACTILITY IN PHENYLEPHRINE-CONTRACTED RABBIT AORTA. cGMP (pmol/g tissue) pmol PO4/ min/ mg Activity Relaxation (-cGMP/+cGMP) Ratio (%) Control SNP (10 s) 0.33 ± 0.05 3.7 ± 1.1 * 54 ± 15/323 ±99 58 ± 18/277 ±98 0.17 ±0.02 0.23 ± 0 . 0 3 * 7 ± 1 * Control SNP (30 s) 0.26 ± 0.04 7.5 ± 1 . 8 * 57 ± 19/342 ±94 82 ±31 / 234 ±53 0.17 ±0.02 0.34 ± 0.04 * 29 ± 3 * Control SNP (2 min) 0.28 ±0.03 1 6 ± 4 * 54 ± 8 / 283 ± 34 87 ± 12*/247 ±31 0.19 ±0.02 0.36 ± 0 . 0 3 * 71 ± 3 Note: Values are mean (± S.E.M.) for n = 8, 5 and 10 for 10, 30 and 120 s, respectively. Means were compared to their controls by a paired Student's t test. * = Significant difference from respective control (p<0.05). 3.2.1.2. Nitroglycerin. PE-contracted rabbit aortic strips were exposed to 0.01, 0.1, 1.0 and 10.0 pM GTN, as outlined in the Materials and Methods Section. These concentrations of GTN were found to inhibit the PE response by 30, 63, 85 and 94%, based on preliminary cumulative GTN concentration-response curves obtained in aorta (data not shown). A two min exposure to GTN induced concentration-dependent relaxation of PE-contracted rabbit aorta at all concentrations studied, as shown in Table 3. The lowest concentration of GTN studied (0.01 pM) significantly relaxed the aorta by 23%. However the cGMP level and PKG activity ratio both appeared to be increased, although the increases were not statistically significant (p>0.05). At 0.1 pM GTN, the magnitude of the increase in the cGMP level was greater (2.6-fold) and this increase was statistically significant (p<0.05). This was accompanied by a significant rise in the PKG activity ratio (1.7-fold) and a greater relaxation (64%). With higher concentrations, GTN further increased cGMP levels and PKG activity ratios and induced greater relaxation. At all GTN concentrations studied, basal PKG (-cGMP) activity was increased compared to control basal activity levels (Table 3 column 3), although this increase was statistically significant (p<0.05) only at higher concentrations of GTN. Total PKG (+cGMP) activity was not changed significantly at any concentration of GTN. As indicated in Section 2.6.2. phosphorylation of endogenous substrates was controlled for by determination of non-BPDEtide 78 TABLE 3. EFFECT OF NITROGLYCERIN (0.1 -10 uM / 2 min) ON PKG ACTIVITY, cGMP LEVELS AND CONTRACTILITY IN PHENYLEPHRINE-CONTRACTED RABBIT AORTA. cGMP (pmol/g tissue) pmol PO4/ min/ mg (-cGMP/+cGMP) Activity Ratio Relaxation (%) Control GTN (0.01 uM) 1.04 ±0.15 1.2 ±0.2 13 ±4/121 ±27 21 ± 3 / 1 5 7 ±14 0.11 ±0.02 0.14 ±0.02 23 ± 8 * Control GTN (0.1 pM) 1.04 ±0.15 2.7 ± 0 . 3 * 13 ± 4 / 1 2 1 ±27 23 ± 6 / 1 2 3 ±25 0.11 ±0.02 0.19 ±0.03 * 64 ± 6 * Control GTN (1.0 uM) 1.04 ±0.15 5.5 ± 1 . 4 * 13 ±4 /121 ±27 2 9 ± 4 * / 1 1 6 ± 1 7 0.11 ±0.02 0.26 ± 0 . 0 3 * 71 ± 3 * Control GTN (10.0 uM) 1.6 ±0.4 11.2 ± 1.6* 21 ± 4 / 1 4 2 ± 1 5 40 ± 2 * / 1 5 0 ± 18 0.14 ±0.01 0.29 ± 0 . 0 2 * 74 ± 4 * Note: Values are mean (± S.E.M.) for n = 5, 5, 5 and 9 for 0.01, 0.1, 1.0 and 10.0 pM GTN, respectively. Means were compared to their controls by a paired Student's t test. * = Significant difference from respective control (p<0.05). blanks. In rabbit aorta, phosphorylation of endogenous substrates by PKG accounted for 11.2 ± 1.8 % (n=12) of total PKG phosphorylation. The temporal effect of 10 pM GTN was also determined in PE-contracted rabbit aorta. The protocol was similar to that described above except that the strips were frozen at 0.5, 1, 2, 5 and 10 min after addition of GTN. The data in Table 4 show that as early as 30 s after addition of GTN, cGMP was significantly elevated (6.1-fold) and this was accompanied by a significant increase in the PKG activity ratio (1.5-fold) and relaxation of the contraction by 37 %. The effect of GTN on cGMP levels was transient and reached a peak value (7.4-fold increase) at 2 min after GTN. Thereafter, there was a decline in the degree of elevation of cGMP levels (see Table 4) at the later time periods. However, at all time periods the elevation of cGMP was significantly (p<0.05) higher than control values. A somewhat similar time-course profile has also been reported in the rat aorta (Keith et al., 1982). The increase in activity ratio was also maximal at the 2 min time point, where the activity ratio was elevated by 2.1-fold. Thereafter, the fold elevation in activity ratio was somewhat lower (1.8- and 1.5-fold at the 5 and 10 min periods, respectively). In spite of the decline in cGMP levels and PKG activity ratios, GTN-induced relaxation was sustained throughout the 10 min period of the experiment. 80 TABLE 4. TIME COURSE OF NITROGLYCERIN (10 uM) EFFECT ON PKG ACTIVITY, cGMP LEVELS AND CONTRACTILITY IN PHENYLEPHRINE-CONTRACTED RABBIT AORTA. cGMP (pmol/g tissue) pmol PO4/ min/ mg (-cGMP/+cGMP) Activity Ratio Relaxation (%) Control 1.5 ±0.2 29 ± 5 / 1 7 9 ±11 0.16 ±0.02 GTN (30 s) 9.2 ± 0 . 8 * 44 ± 2 * / 1 7 8 ±16 0.25 ±0 .02* 37 ± 6 * Control 1.5 ±0.3 34 ± 5 / 233 ± 43 0.15 ±0.01 -GTN (1 min) 11 ± 1 * 5 8 ± 4 * / 2 2 3 ± 1 9 0.27 ±0.02 * 67 ± 4 * Control 1.6 ±0.4 22 ± 4 / 1 4 3 ±15 0.14 ±0.01 _ GTN (2 min) 11 ± 2 * 40 ± 2 * /151 ± 18 0.29 ± 0.04 * 74 ± 4 * Control 1.9 ±0.4 25 ± 5 / 1 8 0 ±37 0.13 ±0.01 _ GTN (5 min) 7.0 ± 1 . 5 * 37 ± 7 * / 1 6 ±27 0.23 ±0 .02* 82 ± 4 * Control 1.7 ±0.5 20 ± 4 / 1 4 9 ±28 0.13±0.01 -GTN (10 min) 6.4 ± 1 . 4 * 44 ± 1 4 * / 2 2 0 ±67 0.20 ±0.01 * 78 ± 5 * Note: Values are mean (± S.E.M.) for n = 4, 6, 9, 9 and 7 for 30 s, 1, 2, 5 and 10 min, respectively . Means were compared to their controls by a paired Student's t test. * = Significant difference from respective control (p<0.05). 3.2.2. Effect of SNP on contractility. PKG activity ratio and cGMP levels in rat vas deferens. One of the objectives of this thesis was to determine whether or not SNP could activate PKG in the rat vas deferens. Rat vasa deferentia were stimulated with a submaximal (EC4 0) concentration (3.0 uM) of PE as outlined in Section 2.4.2. The PE-induced contractions were reproducible and consistent (see left panel of Figure 11). The contractions were characterized by a phasic contraction, which attained a maximum at about 30 s, at which point they were washed with buffer (W) and allowed to return to baseline (within a minute). SNP (0.1 mM)) had no inhibitory effect on the PE response (panel B, right trace) in any of the tissues studied (control tension = 611 ± 90 mg; in presence of 0.1 mM SNP been previously reported to markedly elevate cGMP without affecting the contractile = 574 ± 40 mg). SNP (0.1 mM) significantly increased cGMP levels by 7-fold over control values (Table 5). The PKG activity ratio also significantly increased from 0.12 (control) to 0.19 (SNP), a 1.6-fold increase. The total phosphorylation appears to be higher in the SNP-treated strips, but this increase was not statistically significant (p>0.05). This increase in total phosphorylation levels was due to an abnormally high PKG activity in vas deferens from two of the rats; activity in the other rats was more consistent. Since 0.1 mM SNP elevated both cGMP levels and PKG activity ratios without inhibiting the contractile response, we decided to study the effect of a higher concentration of SNP (5.0 mM) on vas deferens. This concentration of SNP has 82 Figure 11. Effect of SNP on the PE-induced contractile response in rat vas deferens. Representative scanned images of responses are shown from one rat. After an equilibration period, strips were stimulated with 3.0 pM PE for 30 s, three times at 5 min intervals. Strips were washed with buffer (W) and tension returned to baseline. After a 30 min period, one of the vasa deferentia was exposed (for 90 s) to 0.1 mM SNP (B) and stimulated again with PE during the last 30 s and frozen as indicated by double lines at the peak of the PE contraction. The other vas deferens (A) which was not exposed to SNP, was frozen 30 s after PE administration. The same protocol was followed when 5 mM SNP was used. 83 A. CONTROL B. SNP 84 TABLE 5. EFFECT OF SODIUM NITROPRUSSIDE ON PKG ACTIVITY, cGMP LEVELS AND CONTRACTILITY IN PE-INDUCED CONTRACTIONS IN RAT VAS DEFERENS. cGMP (pmol/g tissue) pmol PO4/ min/ mg (-cGMP/+cGMP) Activity Ratio Tension (mg) Control SNP (0.1 mM) 0.62 ±0.15 4.8 ± 0 . 8 * 1.4 ±0.4 / 1 2 ± 9 3.3 ± 0.6 * /16 ± 4 0.12 ±0.02 0.19 ± 0 . 0 2 * 611 ±90 574 ± 40 NS Control SNP (5.0 mM) 1.9 ±0.3 20 ± 3 * 2.5 ± 0 . 3 / 2 9 ± 4 10 ± 4 * /31 ± 13 0.08 ±0.01 0.27 ± 0.03 * 546 ± 50 560 ± 40 NS Note: Values are mean (± S.E.M.) for n = 9 and 5 for 0.1 mM and 5 .0 mM SNP, respectively. Means were compared to their controls by a paired Student's t test. * _ = Significant (p<0.05) difference from control. N S = Not significantly (p>0.05) different from control. been previously reported to markedly elevate cGMP without affecting the contractile response in rat vas deferens (Diamond and Janis, 1978). The same protocol as for the 0.1 mM SNP study was followed and the results are shown in Table 5. As was the case with 0.1 mM SNP, 5.0 mM SNP had no relaxant effect on PE-induced contractions (Table 5). The higher concentration of SNP produced a greater elevation of cGMP than that seen with 0.1 mM SNP. However, the basal cGMP control levels were higher than those seen in the previous study (1.91 vs. 0.62 pmol/g). It should be emphasized that these studies were done a year apart and on different batches of rats, so that this change in basal cGMP levels may simply reflect batch variability. Even with the higher basal activity, 5 mM SNP produced a greater increase in cGMP (10-fold) than that seen with 0.1 mM (7-fold). The higher concentration of SNP also markedly increased PKG activity ratio by 3.3-fold (double that seen with the lower concentration of SNP), but despite this increase no relaxation occurred. Total (+cGMP) PKG activity levels were at least 2-fold higher in this study than in the 0.1 mM SNP study. There was no difference in total PKG activity between control and treated strips. However, basal PKG activity was significantly elevated by 5.0 mM SNP (from 2.5 to 10.1 pmol/mg/min). In the vas deferens endogenous phosphorylation (i.e. non BPDEtide) accounted for approximately 50% of total activity. The average protein concentration in crude soluble extracts of vas deferens was 2.3 pg/pl. 86 The above results (Section 3.2.1. and 3.2.2.) clearly indicate that, while SNP induced relaxation in the rabbit aorta, it did not relax the vas deferens, despite elevating cGMP levels and increasing PKG activity ratio. A possible explanation as to why the vas deferens did not relax might be related to the relative distribution of PKG between the soluble and particulate fractions. It is conceivable that translocation of PKG from one pool to another (e.g. from the soluble to the particulate fraction) might be required for relaxation. Experiments were, therefore, initiated to attempt to compare the distribution of PKG in control and drug-treated rabbit aorta and rat vas deferens. 3.2.3. Effect of nitrovasodilators on PKG activity in soluble and particulate fractions in rabbit aorta and rat vas deferens. 3.2.3.1. Determination of optimal Triton X-100 concentration. Since PKG was determined initially in soluble fractions (the supernatant of the homogenized sample), in order to determine PKG activity in the particulate fractions it was necessary to extract the particulate bound PKG. Extraction of membrane-bound proteins was achieved by using Triton X-100, a non-ionic detergent which is used to separate membrane-bound PKA from membrane phospholipids (Findlay, 1989). A search of the published literature did not indicate what concentration of Triton X-100 should be used in these tissues. Therefore, experiments were conducted to determine optimal Triton-X100 concentrations for 87 extraction and measurement of particulate PKG activity in rabbit aorta and rat vas deferens. Rabbit aorta and rat vas deferens were homogenized and the supernatant (representing cytosolic PKG) assayed for total PKG activity. The pellet was resuspended in various Triton X-100 concentrations, as shown in Figure 12, for 30 min at 4° C, with 3 X 10 s mixing on a low speed vortex every 5 min. The re-suspended pellet was then centrifuged and the supernatant (now containing the solubilized particulate PKG) assayed for specific activity (phosphotransferase activity). Maximal extraction of PKG from aorta was found when the Triton X-100 concentration was 0.1% as shown by the maximum cpm values (Figure 12A). The cpm value is a reflection of how much PKG was extracted. In terms of pmol of phosphate transferred per mg of protein, this parameter decreased with increasing Triton X-100 concentration (Figure 12B) due to greater amounts of protein extracted at higher concentrations. The values shown for '0' Triton X-100 buffer presumably reflects some cytosolic PKG which was trapped in the pellet. In the vas deferens the maximal extraction (CPM) was observed when Triton X-100 concentration was 1% (Figure 12C). However, the difference between 0.1% and 1% was not large and, since the specific activity of PKG was higher at 0.1% (Figure 12D), the 0.1% concentration was used in all subsequent experiments. 88 Figure 12. Determination of optimal Triton X-100 concentration for PKG extraction. Extraction of PKG from particulate fractions of rabbit aorta (A and B) and rat vas deferens ( C and D) by different concentrations of Triton X-100. Panels A and C show the maximal extraction (CPM) as a function of Triton X-100 concentration. Panels B and D show PKG specific activity in extracts from rabbit aorta and rat vas deferens, respectively. The values shown for '0 ' Triton X-100 presumably reflects residual cytosolic PKG trapped in the pellet which was re-suspended in normal buffer. Each bar is the mean (± S.E.M.) of n = 3 experiments. 89 90 3.2.3.2 . Determination of soluble and particulate PKG activity in rabbit aorta and rat vas deferens. Rat vasa deferentia were treated with 5.0 mM SNP, frozen as outlined in Section 2.4.2. and PKG activity determined in the soluble fraction. The pellet was re-suspended in the same volume of homogenization buffer, now supplemented with Triton X-100 (0.1%). The pellet was suspended in this buffer for 30 min (4° C) with 3 X 10 s mixing on a low speed vortex mixer every 5 min. At the end of the 30 min period, the sample was centrifuged again and the supernatant assayed for PKG activity in the presence of 5 pM cGMP (+cGMP). For comparison, rabbit aorta strips treated with GTN (10 pM for 2 min) or SNP (10 pM for 2 min) were also subjected to the same protocol. In control rat vas deferens, when the total PKG (+cGMP) activity in the soluble fraction was compared to the sum of both the soluble and particulate PKG activities on a percentage basis, the soluble fraction contained 27.9 ± 3.1 % of the combined activity and the particulate fraction 72.1 ± 3.1% (n=5) of the combined activity (see Figure 13). This distribution pattern did not change in vas deferens treated with 5 mM SNP (28.4 ± 3.7% of PKG in soluble fraction) where cGMP levels were elevated and PKG activity ratios were increased. It should be emphasized that the PKG activities represent activity that was present in 20 pi of crude muscle extract. 91 In the rabbit aorta, the distribution of PKG was different, in that when the total activity in the soluble fraction was compared to the sum of both the soluble and particulate fractions on a percentage basis, the soluble fraction contained « 60 % of the combined activity (the vas deferens had 28% of the combined activity in the soluble fraction). This distribution pattern did not change in strips treated with either SNP or GTN (Figure 13). 3.2.4. Effect of ANF on contractility. PKG activity ratio and cGMP levels in rat colon. After setting up the colonic segments, consistent spontaneous contractions were normally observed after 20 min and were maintained for at least 60 min, by which time the treatment protocols were completed. The spontaneous contractions in the distal colon (Figure 14A and B) were generally lower in amplitude and higher in frequency than those in the proximal colon (Figure 14C and D). When 100 nM ANF was added to the proximal segment (Figure 14D), the spontaneous contractions were completely inhibited, for almost the full 2 min. In contrast, ANF did not have any inhibitory effect on the distal colon (Figure 14B). The effect of ANF on spontaneous contractions was determined visually and it is possible that small changes in contractility in the distal colon may have taken place which were not apparent visually. However Suthamnatpong et al., (1993b), using a quantitative approach did not detect any inhibition in the distal colon at the same 92 Figure 13. Distribution of PKG activity in rat vas deferens and rabbit aorta following treatment with SNP or GTN. PKG activity was determined in soluble and particulate fractions of smooth muscles in the presence of 5 pM cGMP. The data show PKG activity in the soluble (shaded bar) and particulate (open bar) fractions as a % of combined activity (soluble plus particulate). Rabbit aortae were treated with either GTN or SNP (10 pM/2 min) and vas deferens was treated with SNP (5.0 mM/ 90 s). Each bar is the mean (± S.E.M.) of n = 5 experiments. 93 SOLUBLE & PARTICULATE PKG ACTIVITY AS % OF TOTAL (SOL + PART) PKG ACTIVITY rO 4i>. Gi CO O O O O O O concentration of ANF. The vehicle (0.05 M acetic acid) had no effect on spontaneous contractions in either segment (Figure 14A and C). The inhibition of the contractions in the proximal colon was accompanied by significant increases in cGMP levels (3-fold) and PKG activity ratio (2.2-fold, Table 6). In the distal colon, ANF also significantly elevated cGMP levels (2.6-fold) and increased PKG activity ratio (2-fold), but despite this, no inhibition of the spontaneous contractions took place. ANF elevated basal (-cGMP) levels in both segments and there was no significant (p<0.05) difference in total (+cGMP) PKG activity levels. Phosphorylation of endogenous substrates (non BPDEtide) accounted for approximately 61% of total PKG activity (n = 6). The average protein concentrations in crude soluble extracts of rat proximal and distal colon was 5.3 pg/pl and 3.4 pg/pl respectively. 3.3. Comparision of total PKG activity in different smooth muscles. We considered the possibility that the failure of rat vas deferens and distal colon to relax, in spite of signifcant activation of PKG in these tissues, might be because the total amount of PKG present in these tissues was insufficient to cause relaxation even when activated. To test this possibility, total PKG activity (measured in the presence of 3 pM cGMP) was determined in control preparations from rabbit aorta, rat vas deferens and rat proximal and distal colon. In preliminary concentration-response curves to cGMP, it was found that a concentration of 3 pM 95 Figure 14. Effects of ANF on spontaneously contracting rat distal and proximal colon. Representative scanned images of spontaneously-contracting rat distal (A and B) and proximal (C and D) colon from one rat. After spontaneous contractions stabilized, colonic segments were assigned as control (A and C) or treated (B and D). Control segments received vehicle (0.05 M acetic acid) and were frozen 2 min later. Treated segments received 0.1 pM ANF as indicated for 2 min and were then frozen. 96 Distal Colon A B VEHICLE ANF 100 nM Proximal Colon C D VEHICLE ANF 100 nM 97 TABLE 6. EFFECT OF ATRIAL NATRIURETIC FACTOR ( 0.1 uM) ON PKG ACTIVITY, cGMP LEVELS AND CONTRACTILITY IN SPONTANEOUSLY CONTRACTING RAT PROXIMAL AND DISTAL COLON. cGMP (pmol/g tissue) pmol PO4/ min/ mg (-cGMP/+cGMP) Activity Ratio Inhibition of spontaneous contractions Distal Colon Control 0.42 ± 0.04 ANF 1.1 ± 0 . 1 * 0.91 ±0.21 /11 ± 2 1.40 ± 0.17/10 ± 1 0.07 ±0.01 0.14 ±0.01 NO Proximal Colon Control 0.47 ±0.03 0.82 ± 0.13 /10 ± 1 0.08 ± 0.01 ANF 1.5 ± 0 . 2 * 1.2 ± 0 . 2 / 7 . 5 ±0.9 0.18 ±0 .03* YES Note: Values are mean (± S.E.M.) for n = 7 . Means were compared to their controls by a paired Student's t test. * = Significant (p<0.05) difference from control. was sufficient to completely activate PKG in these tissues. As shown in Figure 15, total PKG activity in rabbit aorta (148 ± 17 pmol/mg protein/min) was about 9-fold higher than in the rat vas deferens (16.4 ± 1.6) and about 13-fold higher than in rat proximal (10.6 ± 0.8) and distal colon (11.4 ± 1.1). 3.4. Attempts to inhibit nitrovasodilator-induced relaxation in rabbit aorta with PKG inhibitors. Since Rp-8-pCPT-cGMPS had been found to inhibit smooth muscle PKG activity in vitro (see Section 3.1.6 above), it was decided to determine whether Rp-8-pCPT-cGMPS also inhibited SNP and GTN-induced relaxation in intact rabbit aorta. Initial studies showed that SNP-induced relaxation in aortic strips was not inhibited by 10 pM Rp-8-pCPT-cGMPS (30 min). This concentration of Rp-8-pCPT-cGMPS was chosen as it was 20 times its K, value and Nakazawa and Imai (1994) used a similar concentration ( 20 x K, ) of Rp-8-Br-cGMP to demonstrate inhibition of relaxation. Possible reasons for the lack of inhibition might be because (1) there was not enough time for Rp-8-pCPT-cGMPS to penetrate into the muscle cells, (2) the concentration was not sufficient to overcome the SNP-induced activation of PKG, or (3) in the intact aortic strips Rp-8-pCPT-cGMPS faced a greater barrier in getting into the vascular smooth muscle cells. 99 Figure 15. Total PKG activity levels in rabbit aorta, rat vas deferens and rat proximal and distal colon. Total PKG activity (pmol/ min/ mg protein) was determined in crude soluble fractions (20 pi) of smooth muscles in the presence of 3 pM cGMP. Each bar is the mean (± S.E.M.) of n = 5 - 9 observations. * = significant difference from mean values in aorta (see Section 3.3). 100 101 It was reasoned that aortic medial strips, which are very thin and consist almost exclusively of smooth muscle cells, might be more suitable for these studies than whole aortic strips. Medial strips were therefore set up for isometric tension measurement as described in Section 2.4.2. The effect of Rp-8-pCPT-cGMPS on GTN and SNP-induced relaxations is shown in Figure 16. Figure 16A shows that 30 pM (60 x K,) Rp-8-pCPT-cGMP (40 min) shifted the GTN-induced relaxation curve to the right by a factor of 7 (EC 5 0 0.7 ± 0.3 to 5.3 ± 3 nM). These results are consistent with a previous report where GTN-induced relaxation in rabbit aortic rings was inhibited by Rp-8-Br-cGMP, a less potent and less membrane permeant inhibitor (Nakazawa & Imai, 1994). However, there was no shift in the relaxation curves when SNP was used as the relaxant agent (Figure 16B) overcome the SNP-induced activation of PKG, or (3) in the intact aortic strips Rp-8-pCPT-cGMPS faced a greater barrier in getting into the vascular smooth muscle cells. The above result with SNP is in contrast to a preliminary study carried out using the rat tail artery. In that study, we found that 0.1 pM SNP relaxed PE (0.1 pM)-induced tonic contractions of rat tail artery by 56.7 ± 3.5% (n=4). However, in the presence of 10 pM Rp-8-pCPT-cGMPS (10 min), 0.1 pM SNP caused only a 21% inhibition of the contractions (i.e. SNP-induced relaxation was inhibited by 102 about 63 %). Thus, at least in this preparation, Rp-8-pCPT-cGMPS was able to inhibit SNP-induced relaxations. 103 Figure 16. Effect of Rp-8-pCPT- cGMPS on SNP-and GTN-induced relaxation in rabbit aorta medial strips. The effect of 30 uM Rp-8-pCPT- cGMPS (40 min) on relaxation induced by GTN (A) and SNP (B) was studied in rabbit aorta medial strips pre-contracted with phenylephrine (1.0 pM). Each point is the mean (± S.E.M.) of n = 3-4 medial strips. 104 Log (M) NITROGLYCERIN Log (M) SODIUM NITROPRUSSIDE 105 4.0. DISCUSSION A major objective of this thesis was to determine whether PKG could be activated in a number of smooth muscle preparations by nitrovasodilators which are known to increase intracellular cGMP levels. In the present study, SNP, GTN and ANF increased cGMP levels and activated PKG in all smooth muscle preparations studied. Cyclic nucleotide-dependent protein kinase activity is commonly expressed as an "activity ratio" which was first used in reporting PKA activity levels (Corbin et al., 1973) and subsequently used for measuring PKG activity in the rat heart (Lincoln and Keely, 1981). The PKG activity ratio is the PKG phosphotransferase activity in the absence of added cGMP divided by the PKG phosphotransferase activity in the presence of sufficient cGMP to completely activate the enzyme, and is considered to be a valid estimation of the activation state of the enzyme in the tissue (Beebe and Corbin, 1986). 4.1. Optimization of assay conditions. PKG phosphotransferase activity in the various smooth muscles was determined by measuring the ability of PKG in crude soluble fractions to transfer Pi from [ y-32P] ATP into a peptide substrate. The labeled substrate was separated from the unbound [y-32P] ATP using a P81 phosphocellulose cation-exchange paper method (Cook et al., 1982). This method requires the presence of at least 106 two basic amino acid residues in the substrate peptide or protein for binding to the P81 paper (Witt and Roskoski, 1975; Casnellie, 1991). The substrate used in this thesis, BPDEtide, has 4 basic residues in its sequence, and thus should be adequate for binding to P81 paper. 4.1.1. Choice of substrate. To study protein kinases, specificity and sensitivity of the substrate are important. In terms of measuring PKG activity, histones (especially H2B) have been the most widely used protein substrates (Lincoln and Keely, 1981., Fiscus et al., 1983, 1984, 1985). The use of histones is complicated by the fact that they increase cGMP binding to PKG (Tse et al., 1981). In addition, H2B has been reported to elevate basal PKG activity (Walton and Gill, 1981), thus contributing to endogenous (-cGMP) phosphotransferase activity. Protein substrates such as H2B are also phosphorylated by other kinases, which makes it more difficult to assay for the activity of a specific kinase. Synthetic peptides based on amino acid sequences found in endogenous substances suffer less from the drawbacks observed with H2B and should offer the greatest specificity (Glass, 1990). In terms of PKG, the most sensitive (as indicated by V M A X / K m values) substrate for PKG has recently been reported to be a peptide (GRTGRRNSI) derived from PKI (VM A X/Km values of 11,000 and 760 for PKA and PKG, respectively) (Mitchell et al., 1995). Other synthetic peptides include RKRSRKE, a 107 peptide derived from H2B (Glass and Krebs, 1979), GRRESLTSFG, a peptide derived from the amino acid sequence of the IP3 receptor (Komalavilas and Lincoln, 1994) and BPDEtide (Colbran et al., 1992). The last three exhibit similar sensitivities to phosphorylation by PKG, based on V M A X / K m values, but both RKRSRKE and GRRESLTSFG are equally well phosphorylated by PKA. When phosphorylation of BPDEtide by PKA and PKG was compared to RKRSRAE (an analog of RKRSRKE), it was found that BPDEtide was much more specific for PKG (16-fold) (Colbran et al., 1992). BPDEtide is a synthetic analog of the amino acid sequence around the phosphorylation site in bovine lung cGMP-dependent phosphodiesterase, and phosphorylation of this phosphodiesterase occurs only when cGMP is bound to it (Thomas et al., 1990), suggesting that the phosphodiesterase is a physiological substrate of PKG. Since the crude soluble fractions used in most of our assays would also contain other kinases, it was important to use a substrate such as BPDEtide which exhibited selectivity for PKG. The phosphorylation of endogenous substrates in the crude soluble extracts by PKG was addressed using non-BPDEtide blanks in an effort to ensure that only phosphorylation of BPDEtide was measured. Phosphorylation of endogenous substrates was substantial in the vas deferens (50 %) and rat proximal colon (60 %) but minimal (10%) in the rabbit aorta. This variation may simply reflect the fact that there was more PKG in aorta than in the other tissues, but about the same amount of phosphorylation of endogenous substrates. 108 The Km values for BPDEtide in the literature (Thomas et al., 1990; Colbran et al., 1992) were determined using purified bovine lung PKG. Since the studies in this thesis used PKG from crude soluble extracts, it was necessary to determine kinetic parameters to see whether they were appropriate for the studies proposed. Phosphotransferase activity was found to be linear over a 15 min time period (Figure 1), indicating that the substrate concentration used (150 pM) was not rate-limiting and that PKG activity could theoretically be measured for up to 15 min. Kinetic analysis revealed that the apparent Km (159 pM), in crude soluble fractions, for BPDEtide was similar to the concentration (150 pM) used in the assay. This value is higher than that observed (68 pM) with purified PKG (Colbran et al., 1992) and may be related to the presence of other kinases in the soluble fraction. A recent report (Bergh et al., 1995) has inferred that 150 pM BPDEtide may not offer enough sensitivity to measure endogenous activation of PKG. However, our studies clearly show that at 150 pM BPDEtide, the assay is sensitive enough to measure endogenous PKG activation, not in only in VSM, but also in non-VSM. The inability of Bergh et al. (1995) to measure endogenous activation, however, may stem from the fact that they used 150 pM BPDEtide, in spite of their observation that the concentration of BPDEtide which gave half maximal activity when phosphorylated by PKG in their crude fractions was 700 pM. While use of substrate concentrations much greater than Km (« 10Km ) would have resulted in near maximal rates, such high concentrations contribute to 109 high blank readings, thus reducing sensitivity, since maximum sensitivity is attained when the signal (product formed) to noise (blank) ratio is at a maximum (Oldham, 1993). If both factors are given the same weight, this condition is typically achieved at concentrations equal to K m (Oldham, 1993). Since, in our study, the Km was found to be approximately 150 pM, this concentration was used throughout the assays reported in this thesis. Furthermore, since the substrate is very expensive, using a higher concentration of the substrate would have increased the cost of the assays. 4.1.2. Is the PKG activity ratio a good indicator of in vivo activation? A major problem encountered during estimating PKG activity is dissociation of cGMP from PKG, following homogenization with subsequent deactivation of the enzyme. One way to minimize the dissociation is to run the assay at low temperatures, since dissociation of cGMP from PKG is slowed at low temperatures (McCune and Gill, 1979). If the temperature is kept at 0-4°C, dissociation is slowed considerably, but not completely (Lincoln & Keely, 1981; Fiscus et al., 1984). In preliminary experiments, we measured PKG activity ratios in rat vas deferens treated with 5 mM SNP. PKG activity was measured at 0°C and 30°C. The PKG activity ratio measured at 0°C (0.27) was significantly higher than that at 30°C (0.08) demonstrating that with higher temperatures dissociation was significant. How then does the dissociation affect activity ratio over time? In another 110 series of experiments, we studied PKG activity ratios in rabbit aorta treated with 130 pM GTN or 10 pM SNP as a function of assay incubation time (Figure 3). As expected, PKG activity ratios decreased from initial high values as the time of incubation was increased, suggesting that cGMP was dissociating from PKG even at 0-4°C. Despite this decrease, the ratios were higher in the drug treated muscles than those in control strips, indicating that despite some degree of dissociation activation of PKG was still measurable. The decrease slowed down after 5 min, after which time the rate of decrease was much slower. This type of dissociation has also been observed in canine trachea treated with SNP or methacholine, both of which elevate cGMP levels and increase PKG activity ratios (Fiscus et al., 1984). Based on this study, it would seem that measuring PKG activity ratios at shorter incubation time periods might be more appropriate, since measuring activity ratios at earlier time periods would have introduced more variability as the variation in PKG activity ratio was greater (larger S.E.M.) at the early time periods (1-5 min) compared to the later periods (Figure 3). It was decided that a reasonable compromise would be to use a 10 min incubation time for our standard assay. Therefore, based on the results in Figure 3, the activity ratios reported in this thesis are probably underestimates of the actual PKG activation. 4.1.3. Is the measured phosphotransferase activity due to PKG? Since crude soluble fractions were used in this thesis, it was essential to establish conditions to ensure that the phosphotransferase activity was solely due 111 to PKG. The first point to consider was interference from other kinases. One of the kinases which could also phosphorylate BPDEtide is PKA: Both PKA (Lincoln et al., 1989) and PKG (Lincoln and Corbin, 1977.; Walter, 1981) are found in VSM in appreciable amounts and their respective levels are similar (Francis et al., 1988). PKG and PKA are homologous enzymes (Takio et al., 1984) and appear to have a common ancestor (Lincoln and Corbin, 1977). Not surprisingly, PKG and PKA have similar substrate specificities: the sequence RRXS(P)X being sufficient for substrate recognition (Lincoln and Corbin, 1977) by both enzymes. The synthetic PKA inhibitor (PKI), which has been shown to inhibit PKA activity by 97% (Hei at al., 1990), was used routinely in our assay to block any effect PKA may have had on phosphorylation of BPDEtide. The extent of PKA-mediated phosphotransferase activity in our assay system was studied by measuring such activity in the absence and presence of PKI (Figure 5). The results indicate that PKA-mediated phosphorylation of BPDEtide was not substantial under the conditions of the assay. These results are consistent with those of Colbran et al. (1992) who found that PKI did not affect total phosphorylation of BPDEtide. It is apparent from these studies that BPDEtide may be used to determine PKG activity ratios in crude soluble extracts with little interference from PKA. Another kinase which could theoretically phosphorylate BPDEtide is casein kinase II (CKII). CKN's canonical sequence for recognizing amino acid sequences as a phosphoacceptor site is (-Ser[P]-X-X-Glu-X). where the G]u can be replaced 112 by any other acidic residue (Pinna et al., 1992). The amino acid sequence of BPDEtide (NHj-Arg-Lys-lle-SeAfPJ-yA/a-Ser-G/ty-Phe-Asp-Arg-Pro-Leu-Arg) contains a sequence (in italics) which is the same as that which is recognized by CKII as a potential phosphoacceptor site. In preliminary experiments, we studied the effect of heparin, an inhibitor of CKII (Hathaway and Traugh, 1982), on PKG activity ratios in nitrovasodilator-treated rabbit aorta. Heparin had no significant effect on activity ratios, suggesting that CKII was not contributing to measured phosphotransferase activity under our assay conditions. In subsequent experiments, heparin was not added to the assay reaction cocktail. To further determine whether the activity we measured was due to PKG activity, we used anion exchange column chromatography to resolve PKG and PKA activities in the smooth muscle preparations. These results clearly showed that the assay conditions used to detect PKG activity were selective for PKG, since there was no cGMP-dependent activity at peaks where only Type I PKA activity (-PKI) was present (Figure 6 and 7). Although we did not have direct proof, based on the similar elution profiles, the two PKA activity peaks observed in our experiments most likely represnt the two types of PKA isozymes as observed by others (Silver et al., 1982; Hei et al., 1990). Type II PKA was found in all the tissues studied, except the rabbit aorta. It could be argued that the "PKA" activity around fraction 28 (rabbit aorta) could be Type II PKA and that it co-eluted with PKG. However, since the profile of "PKA" activity exactly matched that of PKG in these fractions, it was probably due to cAMP activating PKG rather than PKA. In support of this 113 explanation, similar "PKA" activity profiles are seen in the other three tissues which parallel PKG activity in addition to the well characterized Type I and Type II PKA peaks. That the activity seen in the fractions was due to PKG was confirmed by Western Blot immunoanalysis using a PKG antibody provided by Dr. Steven Pelech. Strong immunoreactive bands were seen at « 75 kDa in fractions containing maximal PKG activity. These bands were in the same position as a strong immunoreactive band in the lane containing the PKG la standard holoenzyme. This is a strong indication that the activity measured under our assay conditions was due to PKG and not some other kinase. The nature of the PKG isozyme in these fractions could not be identified since the antibody used was not specific for either Type la or Type ip PKG, as it was raised against a peptide sequence from the C-terminus of PKG, in which there is 100% identity between the isozymes (Wolfe et al., 1989b; Wernet et al., 1989). However, since the molecular mass of the PKG la subunit is about 74 - 78 kDa (Wolfe et al., 1989b; Butt et al., 1993), the blots seen here may represent PKG la (since the immunoreactive bands were at the same MW position), although the presence of PKG ip cannot be discounted. The distribution of these two PKG isozymes in VSM has been reported to be about equal (Lincoln and Cornwell, 1988; Francis et al., 1988; Wolfe et al., 1989b: Shekhar et al., 1992). A more definitive identification of the isozyme present in our tissues might be obtained using a specific PKG ip antibody as described by Keilbach etal. (1992). 114 The slight immunoreactivity observed in fractions containing Type II PKA activity was most likely due to the presence of PKG in these fractions, since PKG and Type II PKA elute very close to one another (particularly in the colon). This is shown by the fact that in fractions in which maximal Type II PKA activity was observed, PKG activity was substantial (« 35 % of maximal PKG activity) (Figure 7). The PKG immunoreactivity in vas deferens in fraction 34 (which has maximal Type II PKA activity) was most probably also due to the higher PKG activity than in earlier fractions (1 - 22). In addition to the main PKG peak, a smaller PKG peak, which eluted before the main PKG peak, was observed in fraction 22 from rabbit aorta. This peak was again observed when PKG activity was resolved in a second batch of rabbit aorta tissue (Figure 8). It is tempting to speculate that this may represent Type II PKG activity since, (a) the common PKG I antibody (common against la and l(3 isozyme) did not recognize this PKG activity, at least at 77-81 kDa (Figure 8, lower panel), and (b) Type II PKG is less electronegative than Type I PKG (Francis et al., 1994), hence the earlier elution. However, this view is unlikely for the following reasons. No similar activity was observed in the rat proximal colon, a tissue which is recognized to contain Type II PKG (Markert et al., 1995). Thus, if the activity represents Type II PKG activity, a similar peak should also have been seen in the proximal colon. The fact that there is weak immunoreactivity in fractions 22 and 23, at 62 kDa, also argues against the activity being due to Type II PKG. This 115 immunoreactivity probably reflects a proteolytic fragment of Type I PKG (hence the immunoreactivity at a lower molecular mass position). The fact that it elutes earlier suggests that the fragment is less charged than the intact PKG. In fractions 27 - 31 two immunoreactive bands were seen (Figure 8, lower panel) and the lower band most probably represents a proteolytic fragment of PKG. The site of proteolysis would be expected to be at the Ay-terminal since the antigenic epitope corresponds to the C-terminal region of PKG. It has been reported that the A/-terminal of PKG contains a number of proteolytic sites (Francis and Corbin, 1994). Additional proof that the activity we measured was due to PKG-mediated phosphorylation of BPDEtide was provided by the studies with the PKG inhibitor Rp-8-pCPT-cGMPS. If the activity measured was due to PKG, Rp-8-pCPT-cGMPS should have inhibited its activity in a concentration-dependent manner. This is exactly what we found in crude homogenates from all the tissues studied (Figure 4). Since the inhibitor was relatively selective for PKG (Butt et al., 1994), these results provide further evidence that the activity measured was due to PKG-mediated phosphotransferase activity. In summary, the studies described above indicate that the phosphotransferase activity measured under our assay conditions was due to PKG. However, the activity ratios were most likely underestimates of actual in vivo activation states. 116 4.2. PKG activity in smooth muscle preparations. As mentioned at the beginning of the discussion, one of the objectives of this thesis was to determine whether PKG was activated by nitrovasodilators in rabbit aorta and if there was a correlation between activation of PKG, elevation of cGMP and relaxation. The results presented here generally support the notion that drug-induced relaxation correlates with elevations in cGMP and activation of PKG. 4.2.1. Studies with SNP and GTN in rabbit aorta. We chose to induce relaxation of rabbit aorta, which contains PKG (Ives et al., 1980), by SNP and GTN, compounds which are known to activate soluble guanylyl cyclase and elevate cGMP levels (Diamond and Blisard, 1976; Katsuki et al., 1977; Schultz et al., 1977) and to induce relaxation in a range of VSM preparations (Ignarro and Kadowitz, 1985; Waldman and Murad, 1987). Previous work on the effect of SNP on PKG activity ratios in VSM has been limited to two reports (Fiscus et al., 1983; Jiang et al., 1992). In the first of these reports, SNP (50 nM) significantly increased the PKG activity ratio in rat aorta. No concentration-dependent study was carried out, nor were cGMP levels reported, or relaxation monitored. Jiang et al. (1992) reported that SNP increased PKG activity ratios in pig coronary artery in a concentration-dependent manner. Fiscus et al. (1985) have also reported concentration- and time-dependent elevation of cGMP, activation of PKG and relaxation in rat aorta induced by another vasodilator, ANF, 117 which elevates cGMP by activating the particulate form of guanylyl cyclase (Waldman, et al., 1984; Winquist et al., 1984). Even in the studies in which concentration- and time-response data were reported (e.g. Fiscus et al., 1985), the relaxation and biochemical parameters were determined in separate preparations. In the present experiments, all three parameters were determined in the same strip of VSM and we feel that this should allow a better comparison of the changes of the three parameters. In addition, these early studies had used assay conditions that may not have been appropriate. For example, H2B was used as the substrate and, as mentioned in Section 4.1.1, H2B increases cGMP binding to PKG and elevates basal PKG activity. Furthermore high levels of 3 2 P (4000 - 5000 cpm/pmol ATP) were required in the assay (Fiscus et al., 1983, 1984). Finally, a recent study has indicated that H2B is a less specific substrate for PKG compared to BPDEtide (Colbran et al., 1992). We believe that the assay we have used represents a significant improvement over the previous assays mainly due to the PKG-specific substrate, BPDEtide. Our studies (a) demonstrate that SNP induces relaxation in rabbit aorta, and that this is accompanied by increased cGMP levels and activation of PKG, and (b) show a reasonably strong correlation between the first two parameters and increases in PKG activity ratios. The data show that the assay we used was sensitive enough to measure endogenous PKG activation. For example, a 2.8-fold increase in cGMP caused by a low concentration of SNP (0.1 pM) resulted in a 118 significant (1.3-fold) activation of PKG (Table 1) which was accompanied by a 14 % relaxation of PE-contracted rabbit aorta. A similar (1.4-fold) activation of PKG by 0.1 pM SNP was reported by Fiscus et al. (1983) in rat aorta using H2B as the substrate and a shorter (2.5 min) incubation time. Thus, even with a longer incubation time, when dissociation of bound cGMP would be greater than at 2.5 min, we were able to measure comparable increases in PKG activity ratios. If the PKG activity was solely due to the effect of SNP, only the endogenous (-cGMP) phosphorylation levels should increase, since this phosphorylation is thought to reflect endogenous (drug-induced) activation of PKG. In the present experiments, only endogenous phosphorylation (-cGMP) was increased: total activity remained essentially unchanged (Table 1). Since the ratios were measured in crude muscle extracts, it is not unreasonable to consider the possibility that the increase in endogenous phosphorylation could have been due to cyclic nucleotide-independent phosphorylation of BPDEtide. However, this does not appear to be the case, since such an effect should also have been seen in total PKG activity levels and no such effect was evident (i.e. total phosphorylation did not change significantly). In the study reported by Jiang et al. (1992), 0.1 pM SNP had no effect on PKG activity ratios in pig coronary artery. However, at higher concentrations of SNP, Jiang et al. (1992) found somewhat higher PKG activity ratios than were found in our studies. Control PKG activity ratios were similar in the two studies. However, as noted above, Jiang er al. (1992) did not measure relaxation. 119 When we studied the temporal effect of 10 pM SNP, we observed that, as early as 10 s after addition of SNP, cGMP levels were significantly elevated and this was accompanied by a significant increase in PKG activity ratio and relaxation of the aorta. To our knowledge, this is the first demonstration of PKG activation at this early time after SNP. It should be noted that the PKG activation reached a maximum at 30 s after SNP administration, even though cGMP levels and the degree of relaxation continued to increase with time. The fact that the muscles continued to relax after maximal PKG activation was not surprising. If it is assumed that PKG acts by phosphorylating a protein (or proteins) involved in lowering cytoplasmic calcium, as has been suggested (Cornwell and Lincoln, 1989), it seems reasonable to expect that some time would be required to lower the calcium levels sufficiently to relax the PE-contracted muscles. It is more difficult to explain why the PKG activity ratios are correlated with cGMP levels only at earlier times. It appears that maximal activation of the kinase occurs at less than maximal cGMP levels and that no further activation of the kinase occurs as cGMP levels continue to increase. The fact that the kinase is not fully activated is consistent with the suggestion made earlier that our kinase activity ratios may be underestimates of endogenous activity. As discussed by Fiscus et al. (1983), two distinct binding sites have been reported for cGMP binding to PKG, with full activation of the enzyme requiring occupancy of both sites. Binding at one of the sites occurs at high concentrations of cGMP and dissociation of cGMP from this site is more rapid than from the other. 120 Therefore, it is possible that underestimation of PKG may be greater in samples with very high cGMP levels. Since we did not measure the various parameters at periods earlier than 10 s, we cannot say with certainty whether PKG activation preceded relaxation. Previous studies in rabbit aorta (see e.g. Diamond & Chu, 1983; Brien et al., 1988) have shown that cGMP elevation precedes relaxation. Nevertheless, there appears to be a reasonable temporal correlation between PKG activation and relaxation in our studies since PKG was activated at the earliest time point studied, which was the earliest time point at which we were positive that the preparations were relaxed by SNP Temporal studies with a high concentration (10 pM) of GTN also resulted in significant elevation of cGMP and activation of PKG at an early time point coinciding with the onset of relaxation. However, in contrast to the SNP study, both cGMP levels and PKG activity ratios fell after attaining maximal values at approximately 2 min, although relaxation was maintained for the duration of the experiment (10 min). Similar patterns of transient cGMP changes and sustained relaxations have been reported in a number of VSM preparations, including rat aorta (Keith et al., 1982), bovine mesenteric (Axelsson et al., 1979) and coronary artery (Kukovetz et al., 1979). It would appear that only a small degree of elevation of cGMP and activation of PKG is required for maintenance of relaxation once relaxation has been attained. 121 In the concentration-dependent study, GTN-induced elevation of cGMP levels, activation of PKG activity, and relaxation showed a strong correlation. Basal cGMP levels in this study were much higher than those in the SNP study. This difference in cGMP control levels was most likely due to the fact that the two studies were done more than a year apart and probably represents simple animal batch variation. Large animal to animal variations in cGMP levels have been previously reported (see e.g. Katsuki and Murad, 1977; Diamond, 1977; Keith et al., 1982). In the majority of our experiments, increases in PKG activity ratios were due to elevation of basal (-cGMP) kinase activity and total (+cGMP) kinase activity was unchanged. As noted by Fiscus et al. (1983), this suggests that the activation of the enzyme observed in our experiments following administration of SNP and GTN was due to endogenous activation of PKG, and not to activation of a cyclic nucleotide-independent protein kinase. If there was activation of a cyclic nucleotide-independent protein kinase, there would have been an increase in activity both with and without added cGMP. 122 4.2.2. Studies with SNP in selected non-vascular smooth muscles. In contrast to VSM in which there was generally a strong correlation between cGMP elevation, activation of PKG and relaxation (above), the evidence for a role for cGMP in relaxation in some types of non-VSM was not strong. For example, in the rat vas deferens, SNP significantly elevated cGMP levels but did not inhibit PE-induced contraction (Diamond, 1983; Diamond and Janis, 1978). In another non-vascular smooth muscle, the guinea pig taenia coli, SNP elevated cGMP levels and induced relaxation (Janis and Diamond, 1979). If the effect of SNP is mediated via activation of PKG, PKG in the guinea pig taenia coli should be activated by SNP, and in the vas deferens there should be no activation. If lack of activation of PKG in vas deferens could be demonstrated, this would explain why SNP failed to induce relaxation in this tissue. Early attempts to measure activation of PKG in both vas deferens and guinea pig taenia coli were unsuccessful (Diamond et al., 1983). One possible reason for this may have been due to cGMP dissociating from PKG during the assay, which was conducted at 30°C. In our opinion, the techniques available for measuring PKG activation at that time were inadequate for studies of this type. The availability of an improved assay for PKG (Jiang et al., 1992) has now enabled us to re-investigate this problem. In the present experiments, a lower concentration of PE (3 pM, E C 4 0 ) was used, and the reason for choosing such a sub-maximal concentration of PE was the assumption that at this concentration the response would be more sensitive to 123 SNP, since the concentration of a contractile agent affects the degree of relaxation by nitrovasodilators (Nakatsu and Diamond, 1989). Diamond (1983) had used a PE concentration of 30 uM, which is equivalent to E C g o and it is possible that at the higher concentration of PE the SNP-generated cGMP was not sufficient to inhibit the contraction. In the present studies, 0.1 mM SNP elevated cGMP levels but did not inhibit PE-induced contractions, consistent with the previous report by Diamond (1983). Surprisingly, PKG activity ratios were also significantly elevated (1.5-fold). Despite this increase in activity ratio, no relaxation took place and this is in stark contrast to the rabbit aorta study (above) in which a SNP-induced increase in PKG activity ratio of 1.3-fold was accompanied by a significant relaxation. If activation of PKG is responsible for SNP-induced relaxation of rabbit aorta, as suggested by the data shown earlier, why is there is no relaxation in the vas deferens? Possible explanations could include the following: (1) the magnitude of PKG activation obtained in vas deferens is not sufficient to overcome the PE-induced contraction (i.e. a greater activation of the PKG is required to induce relaxation), (2) the activation of PKG is not physiologically significant (i.e. it cannot cause relaxation) because the absolute amount of PKG present in the vas deferens is much less than in vascular tissues or (3) the PKG activity was measured in the soluble fraction of the tissues and it may be that activation of PKG in a particulate fraction or translocation of the enzyme from one fraction to the other may be more important (i.e. activation of particulate PKG or translocation of PKG may occur in muscles relaxed by SNP but not in non-responsive tissues). 124 To address the first of the above possibilities, we repeated the vas deferens study with a higher concentration of SNP (5 mM). This concentration of SNP was chosen as it had been shown to increase cGMP levels in rat vas deferens by 16-fold (Diamond & Janis, 1978), which is more than double that seen with the 0.1 mM concentration. It was reasoned that this greater increase of cGMP could further activate PKG to such an extent that inhibition might result. Five mM SNP significantly increased PKG activity ratios (by 3-fold) compared to control. Despite this 3-fold increase in activity ratio, the PE-induced contraction was not inhibited. This degree of activation of PKG (i.e. 3-fold increase in activity ratio) in our rabbit aorta experiment was accompanied by marked relaxation of the muscles. This study clearly indicates that SNP can induce marked activation of PKG in rat vas deferens but this does not result in inhibition of contraction. Accordingly, a previous suggestion as to why SNP-induced increases in cGMP in vas deferens are not accompanied by inhibition, namely that SNP-induced increases in cGMP do not have access to PKG (Diamond et al., 1983), is not tenable. Similar results were also obtained using another model system, the rat proximal and distal colon. Recent reports from Suthamnatpong et al. (1993a, 1993b) and Maehara et al. (1994) indicated that spontaneous contractions in the proximal colon were inhibited by cGMP-elevating agents, such as ANF and NO, whereas in the distal colon no such inhibition was evident when exposed to these agents. Thus, another pair of smooth muscles consisting of a "responder" and a "non-responder" was available for comparative studies such as those described 125 above for rabbit aorta and rat vas deferens. However, in this case it might be expected that the two muscles might be more similar with respect to functional and biochemical characteristics. Again, it was of interest to determine whether or not the reported lack of inhibition to cGMP-elevating agents in the distal colon could be explained by failure of the agents to activate PKG in that tissue. To this end, the effects of ANF on spontaneous contractions and PKG activity were compared in segments of rat proximal and distal colon. As noted earlier, ANF is known to selectively activate the particulate form of guanylate cyclase whereas SNP, which was used in the vas deferens, activates the soluble form of the enzyme. In confirmation of the report by Suthamnatpong et al. (1993b), ANF inhibited spontaneous contractions in the proximal but not in the distal colon. PKG activity ratios, however, were significantly increased in both proximal and distal colon. While the increase in cGMP levels and PKG activity ratio in the proximal colon would be consistent with PKG mediating inhibition, the activation of PKG in the distal colon, which does not translate into inhibition despite a similar magnitude of PKG activation, was not consistent. Thus, as was the case with the vas deferens, the lack of inhibition in the distal colon cannot be explained by a lack of PKG activation. As mentioned above, another possible reason why there was no inhibition of contraction in non-responding muscles may be due to low total levels of PKG in these tissues. The next set of experiments was conducted to determine what the 126 total PKG levels were in crude soluble fractions of the various smooth muscle preparations and whether low levels of PKG activity in the non-responding muscles could explain why cGMP-elevating agents could not relax these preparations. The first preparations analyzed were the rabbit aorta and rat vas deferens and it was found that PKG levels were about 10-fold higher in rabbit aorta than in the rat vas deferens. This would be consistent with the above suggestion, since only the aorta was relaxed by SNP (the assumption being that the levels of PKG in the aorta, but not in the vas deferens, were sufficient, when activated, to overcome the contractile effect of phenylephrine). However, this conclusion was not supported by the results of the experiments with the rat proximal and distal colon. Total PKG levels in both of these tissues were very low (lower than those in the vas deferens) and yet spontaneous contractions in the proximal colon, but not distal colon, were inhibited by ANF. Total PKG levels were almost identical in these two tissues and, as noted earlier, the degree of activation of PKG by ANF was almost the same. Similar levels of activity (in terms of fold differences) were found when crude muscle extracts were resolved for PKG activity, by MonoQ chromatography, thus confirming the data from the tissue enzyme assays. No attempt was made to determine the relative contribution of the two types of PKG I isoforms to the total activity. Taken together, these data suggest that the ability of some smooth muscles, but not pthers, to relax in response to cGMP-elevating agents cannot be explained solely on the basis of differences in total PKG activity in the muscles. 127 Finally, we considered the possibility that a translocation of PKG may be more important in determining whether or not relaxation takes place. The rationale behind this was very simple: it is known that certain enzymes translocate when stimulated by an agonist. A well known example is the translocation of protein kinase C from the soluble to the particulate fraction in smooth muscle preparations stimulated by agonists (Haller et al., 1990; Langlands et al., 1992). However, in the present experiments, when PKG activity levels were determined in soluble and Triton X-100 extracted particulate fractions, there was no difference in PKG activities (+cGMP) in the two fractions from rabbit aorta treated with GTN and SNP compared to control strips. In vas deferens, as with the rabbit aorta, there was no difference in the distribution of PKG activity between the particulate and soluble fractions in SNP-treated and control tissues. Thus, the lack of relaxation in the rat vas deferens cannot be explained by a lack of translocation of PKG, since there was apparently no such translocation in the rabbit aorta, a tissue in which relaxation does occur. These experiments must be considered as preliminary, and it is difficult to draw any firm conclusions from these data. It is possible that the conditions used in our initial studies were not adequate for demonstrating translocation of PKG even if it did occur. For example, Lincoln and coworkers (Cornwell et al., 1991) have recently suggested that phospholamban is an important substrate for PKG and have proposed that co-localization of PKG with sarcoplasmic reticulum (SR) is 128 an important factor for cGMP-dependent phosphorylation of this protein. In view of this suggestion, our experiments should be repeated using a more rigorous differential centrifugation technique which can separate the SR from the soluble fraction. With the technique used in our initial studies, the SR would still be present in the "soluble" fraction. Until such studies are done, we cannot rule out the possibility that some form of translocation plays an important role in PKG-mediated smooth muscle relaxation. 4.2.3. Effect of Rp-8-pCPT-cGMPS on smooth muscle relaxation. In the experiments described above (Section 4.2.1), a reasonably strong correlation was found between activation of PKG and relaxation of rabbit aorta by GTN and SNP. If the relationship is a causal one, i.e. if activation of PKG is responsible for relaxation, if activation of the kinase could be prevented, relaxation should also be prevented. As shown in the Results (Section 3.1.6), Rp-8-pCPTcGMPS, a competitive inhibitor of PKG, was able to inhibit PKG in homogenates prepared from all smooth muscles tested in the present study under in vitro conditions. Since this inhibitor is membrane permeant and resistant to hydrolysis by phosphodiesterases (Butt et al., 1994), it was expected that exposure of intact muscle to the inhibitor would prevent or inhibit the activation of PKG by SNP and GTN and therefore would prevent or inhibit the relaxation normally caused by these agents. In agreement with this 129 prediction, 30 pM Rp-8-pCPTcGMPS shifted the GTN relaxation curve to the right in medial strips of rabbit aorta, which is consistent with the assumption that GTN-induced relaxation is mediated by PKG activation. A similar rightward shift of GTN-and 8-Br-cGMP-induced relaxation in intact rabbit aorta has been reported by Nakazawa and Imai (1994). Surprisingly, in the present study, Rp-8-pCPTcGMPS had no effect on SNP-induced relaxation of rabbit aorta. This is difficult to explain since the relaxant effects of both GTN and SNP are believed to be mediated by increases in cGMP levels, presumably acting via activation of PKG. Why there should be a differential sensitivity of the two vasodilators to the effects of a PKG inhibitor is not clear. It should be noted that, in the rat tail artery, SNP-induced relaxation of PE-induced contractions was partially inhibited by Rp-8-pCPTcGMPS (Results, Section 3.5.2). Thus, although in some experiments the relaxant effects of nitrovasodilators are attenuated by a PKG inhibitor, as would be expected, in other experiments this was not the case. It is possible that even higher concentrations of Rp-8-pCPTcGMPS might be able to block the relaxant effects of SNP in the rabbit aorta. However, the selectivity of the inhibitor for PKG may be lost at higher concentrations. We do not know what concentration of the inhibitor is attained inside the muscle cells when intact tissues are exposed to the inhibitor. In the absence of direct measurements of PKG activity after treatment of intact muscle with Rp-8-pCPTcGMPS, we cannot be certain that the kinase was actually inhibited under these conditions. Since, at the present time, these experiments are not feasible, no firm conclusion can be drawn from these studies at this time. 130 4.3. Future Directions. The studies described above have provided direct evidence for a correlation between activation of PKG and relaxation of a vascular smooth muscle, the rabbit aorta, by nitrovasodilators. They have also demonstrated that activation of PKG in some types of smooth muscle is not accompanied by relaxation. In these tissues, which include rat vas deferens and distal colon, SNP markedly increased cGMP levels and activated PKG without causing relaxation. The reason for this apparent dissociation between PKG activation and relaxation is not clear. Several possible explanations for this dissociation were investigated, as described in the previous section. These were concerned with variations in total amounts of PKG present in the different tissues and with possible compartmentalization of PKG within the muscle cells. Although we cannot rule out compartmentalization as an explanation, we were unable to provide any direct evidence for either of these possibilities. It is possible that the reason for lack of relaxation was downstream to elevation of cGMP and activation of PKG. Since a number of substrates have been shown to be phosphorylated in vitro by PKG (Casnellie at al., 1980; Raeymakers et al., 1988; Vrolix et al., 1988; Yoshida et al., 1991), it is logical to postulate that a lack/defect of phosphorylation of a critical endogenous substrate (or substrates) may be responsible for the lack of relaxation in non-responders. A recent publication (Bergh et al., 1995) used this approach to identify whether or not certain proteins were phosphorylated. These authors reported that SNP induced a 131 near maximal relaxation in 5HT-contracted bovine carotid artery and phosphorylated two proteins as determined by isoelectric focusing and SDS-PAGE. However, in human umbilical artery, in which SNP had no relaxant effect, the two proteins were not phosphorylated. Therefore, differences between different tissues in patterns of protein phosphorylation could theoretically indicate whether or not there is a lack/defect in phosphorylation of such proteins. If we could demonstrate phosphorylation of specific proteins in tissues which are relaxed, and a lack thereof in tissues not relaxed by the drug, we might be closer to identifying the putative substrate or substrates responsible for cGMP/PKG-induced relaxation. The second area of research which might be fruitful would be to determine whether PKG is co-localized with a target substrate protein in smooth muscle cells. The rationale for this is that even if PKG levels are low, if the PKG is localized with or near a substrate protein, upon activation by cGMP, phosphorylation will take place. Cornwell et al. (1991) have shown that in rabbit aorta cells, SNP and PKG induce phosphorylation of a subunit of phospholamban. Furthermore, PKG was localized to the same area as phospholamban. In broken cell preparations cGMP, but not cAMP, led to phosphorylation of membrane-associated proteins and phospholamban, which indicated that only PKG was associated with the protein substrates. In another report, Wyatt et al. (1991) have found that in neutrophils, PKG, whose concentration is about 1/100 of that found in some smooth muscles (Pryzwansky et al., 1990), redistributed from the cytoplasm to intermediate filaments which project into areas of the plasma membrane upon stimulation by 132 formyl-methionyl-leucyl-phenylalanine. This redistribution resulted in phosphorylation of vimentin, the endogenous intermediate filament protein. However, 8-Br-cGMP did not cause phosphorylation in unstimulated neutrophils. Both these studies suggested that it is the co-localization of PKG (compartmentalization) with putative substrates which may be more critical in determining whether phosphorylation takes place or not. In the context of this thesis, this would determine whether or not a tissue relaxes. Studies along these lines in responsive and non-responsive smooth muscles might yield useful information regarding the role of PKG in smooth muscle relaxation. 4.4. Summary. 1. Assay conditions required to measure PKG activity in smooth muscle preparations were optimized. The assay procedure involved the use of BPDEtide, a recently described peptide substrate exhibiting high specificity for PKG. The specificity of the assay for PKG was demonstrated by the fact that under the assay conditions used, no phosphotransferase activity was observed in MonoQ-resolved fractions containing only PKA activity. Immunoblotting with antibodies to PKG also confirmed that the phosphotransferase activity measured in the MonoQ fractions was due to PKG and not to other kinases. In addition, a PKG antagonist inhibited, in a concentration-dependent manner, phosphorylation of BPDEtide by soluble fractions from all smooth muscles tested. The assay was sensitive and reliable enough to demonstrate small increases in activity ratios in a variety of smooth 133 muscle preparations. 2. In rabbit aorta, a good correlation between SNP- and GTN -induced elevation of cGMP, activation of PKG and relaxation was demonstrated. Both concentration-dependent and temporal correlations were found. 3. In the rat vas deferens, SNP caused a marked increase in cGMP levels, and a significant activation of PKG, but did not inhibit PE-induced contractions in the vas deferens. Similar results were obtained in the rat distal colon, where ANF increased cGMP levels and activated PKG but did not inhibit spontaneous contractions. These results demonstrate that activation of PKG in smooth muscle preparations was not necessarily accompanied by relaxation. 4. The total amount of PKG present in a tissue did not appear to determine whether the tissue relaxes or not when PKG was activated. This possibility was suggested by the fact that the rabbit aorta, which was relaxed by SNP, contained about 10 times as much PKG as the rat vas deferens, where PE-induced contractions were not inhibited by SNP. However, in the rat proximal colon, which had significantly lower levels of PKG than the vas deferens, spontaneous contractions were inhibited when PKG was activated by ANF. A similar amount of PKG was present in the distal colon, but activation of the enzyme by ANF was not accompanied by inhibition of contraction of that tissue. 5. The lack of inhibition in response to SNP in rat vas deferens could not be 134 explained by a lack of translocation of PKG within the cells. Translocation from a soluble to a particulate fraction has previously been reported to be involved in activation of some protein kinases. However, no such translocation of PKG was evident in preliminary experiments in rabbit aorta, where activation of the enzyme was accompanied by relaxation. The fact that no translocation could be demonstrated in the vas deferens, therefore, cannot explain the lack of inhibition seen in that tissue. 6 . Other possible explanations for the lack of inhibition seen in response to PKG activation in the rat vas deferens and distal colon could include a) co-localization of PKG and its substrate(s) in responsive but not in non-responsive muscles and/or b) absence of a required protein substrate from non-responsive cells. 135 5.0. BIBLIOGRAPHY Alioua, A., Higgins, J.P., and Rousseau, E. (1995) PKG-1a phosphorylates the a-subunit and upregulates reconstituted GK C a channels from tracheal smooth muscle. Am. J. Physiol. 268: L1057-L1063. Amer, M.S., and McKinney, G.R. (1973) The possible role of cGMP in gall bladder contraction and gastric acid secretion in rabbits. Pharmacologist. 15: 157-159. Anderson, T.J., Meredith, IT., Ganz, P., Selwyn, A.P., Yeung, A.C. (1994) Nitric oxide and nitrovasodilators: similarities, differences and potential interactions. J. Amer. Coll. Cardiol. 24: 555-566. Arnold, W.P., Mittal, C.K., Katsuki, S., and Murad, F. (1977) Nitric oxide activates guanylate cyclase and increases guanosine 3'-5'-cyclic monophosphate levels in various tissue preparations. Proc. Natl. Acad. Sci. USA 74: 3203-3207. Ashman, D.F., Lipton, R., Melicow, M.M., and Price, T. D. (1963) Isolation of adenosine 3'-5'-monophosphate and guanosine 3'-5'-monophosphate from rat urine. Biochem. Biophys. Res. Commun. 11: 330-334. Axelsson, K.L., Wikberg, J.E.S., and Andersson, R.G.G. (1979) Relationship between nitroglycerin, cyclic GMP and relaxation of vascular smooth muscle. Life Sci. 24: 1779-1786. Bates, JN., Baker, M.T., Guerra, R Jr., and Harrison D.G. (1991) Nitric oxie generation from nitroprusside by vascular tissue: Evidence that reduction of the nitroprusside biologic activity and cyanide release. Biochem. Pharmacol. 42:5157-5165. Beavo, J.A. (1988) Multiple isozymes of cyclic nucleotide phosphodiesterase. Adv. Second Messenger Phosphoprotein Res. 22: 1-38. Beavo, J.A., Hardman, J.G., and Sutherland, E. W. (1970) Stimulation of adenosine 3'-5'-monophosphate hydrolysis by guanosine 3'-5'-monophosphate. J. Biol. Chem. 245: 5469-5655. Beebe, S., and Corbin, J.D. (1986) Cyclic nucleotide-dependent protein kinases. In Boyer, P.D and Krebs, E.G., eds. The Enzymes., Vol XVII. Academic Press Inc, pp 43-111. 136 Bergh, CM., Brophy, CM., Dransfield, D.T., Lincoln, T., Goldenberg., J.R., and Rasmussen, H. (1995) Impaired cyclic-nucleotide-dependent vasorelaxation in human umbilical artery smooth muscle. Am. J. Physiol. 268: H202-H212. Berridge, M.J. (1993) Inositol trisphosphate and calcium signalling. Nature. 361: 315-325. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248-254. Brien, J.F., McLaughlin, B.E., Kobus, S.M., Kawamoto, J.H., Nakatsu, K., and Marks, G.S. (1988) Mechanism of glyceryl trinitrate-induced vasodilation. 1. Relationship between drug biotransformation, tissue cyclic GMP elevation and relaxation of rabbit aorta. J. Pharmacol. Exp. Ther. 244: 322-327. Breuchler, W.A., Nakane, M., and Murad, F. (1991) Expression of soluble guanylyl cyclase activity requires both enzyme subunits. Biochem. Biophys. Res. Commun. 174: 351-357. Butt, E., Eigenthaller, M., and Genieser, H-G. (1994) (Rp)-8-pCPT-cGMPS, a novel cGMP-dependent protein kinase inhibitor. Eur. J. Pharmacol. 269: 265-268. Butt, E., van Bremmelen, M., Fischer, L , Walter, U., and Jastorf, B. (1990) Inhibition of cGMP-dependent protein kinase by (Rp)-guanosine 3',5'-monophosphothioates. FEBS Lett. 263: 47-50. Butt, E., Geiger, J., Jarchau, T., Lohmann, S.M., and Walter, U. (1993) The cGMP-dependent protein kinase-gene, protein, and function. Neurochem. Res. 18: 27-42. Casnellie, J.E. (1991) Assay of protein kinases using peptides with basic residues for phosphocellulose binding. Methods Enzymol. 200:115-120. Casnellie, J.E., Ives, H.E., Jamieson, J.D., and Greengard, P. (1980) Cyclic GMP-dependent phosphorylation in intact medial tissue and isolated cells from vascular smooth muscle. J. Biol. Chem. 255: 3770-3776. Cheng, H.C., Kemp, B.E., Pearson, R.B., Misconi, L., Van Pattens, S.M., and Walsh, D.A. (1986) A potent synthetic peptide inhibitor of the cAMP-dependent protein kinase. J. Biol. Chem. 261: 989-992. Chinkers, M., Garber, D.L., Chang, M.S., Lowe, D.G., Chin, H., Goeddel, D.V., and Schultz, S. (1989) A transmembrane form of guanylyl cyclase is an atrial natriuretic receptor. Nature. 338: 78-83. 137 Colbran, J.L., Francis, S.H., Leach, A.B., Thomas, M.K., Jiang, H., McAllister, L.M., and Corbin, J. (1992) A phenylalanine in peptide substrates provides for selectivity between cGMP- and cAMP- dependent protein kinases. J. Biol. Chem. 267: 9589-9594. Cook, P.F., Neville, M.E., Vrana, K.E., Hartl, FT., and Roskoski, R. (1982) Adenosine cyclic 3'-5'-monophosphate dependent protein kinase: Kinetic mechanism for the bovine skeletal muscle catalytic subunit. Biochemistry. 21: 5794-5799. Corbin, J.D.,and Doskeland S.O. (1983) Studies of two intrachain cGMP-binding sites of cGMP-dependent protein kinase. J. Biol. Chem. 258: 11391-11397. Corbin, J.D., Soderling, T.R., and Park, C.R. (1973) Regulation of adenosine 3'-5'-monophosphate-dependent protein kinase. J. Biol. Chem. 248: 1813-1821. Cornwell, T.M., and Lincoln, T. (1989) Regulation of intracellular Ca 2 + levels in cultured vascular smooth muscle cells. J. Biol. Chem. 264: 1146-1155. Cornwell, T.M., Przwansky, K., Wyatt, T, and Lincoln, T. (1991) Regulation of sarcoplasmic reticulum protein phosphorylation by localized cGMP-dependent protein kinase in vascular smooth muscle cells. Mol. Pharmacol. 40:923-931. Craven, P.A., and DeRubertis, F.R. (1978) Restoration of the responsiveness of purified guanylate cyclase to nitrosoguanidine, nitric oxide, and related activators by heme and heme proteins. Evidence for the involvement of the paramagnetic nitrosyl-heme complex in enzyme activation. J. Biol. Chem. 253: 8433-8443. DeJonge, H.R. (1976) Cyclic nucleotide-dependent phosphorylation in intestinal epithelium. Nature. 262: 590-593. DeJonge, H.R. (1981) Cyclic GMP-dependent protein kinase in intestinal brushborders. Adv. Cycl. Nucl. Res. 14: 315-333. DeRubertis, F.R., and Craven, P.A. (1976) Calcium-independent modulation of cyclic GMP and activation of guanylate cyclase by nitrosoamines. Science. 193: 897-899. Diamond, J. (1983) Lack of correlation between cyclic GMP elevation and relaxation of non-vascular smooth muscle by nitroglycerin, nitroprusside and sodium azide. J. Pharmacol. Exp. Ther. 225: 422-426. 138 Diamond, J. (1987) Effects of LY83583, nordihydroguaiaretic acid, and qiunacrine on cyclic GMP elevation and inhibition of tension by muscarinic agonists in rabbit aorta and left atria. Can. J. Physiol. Pharmacol. 65: 1913-1917. Diamond, J., and Blisard, K.S. (1976) Effects of stimulant and relaxant drugs on tension and cyclic nucleotide levels in canine femoral artery. Mol. Pharmacol. 12: 688-692. Diamond, J., and Chu, E. (1983) Possible role for cGMP in endothelial-dependent relaxation of rabbit aorta by acetylcholine. Comparision with nitroglycerin. Res. Commun. Chem. Pathol. Pharmacol. 41: 369-381. Diamond , J., and Hartle, D.K. (1976) Cyclic nucleotide levels during carbachol-induced smooth muscle contractions. J. Cycl. Nucl. Res. 2: 179-188. Diamond, J., and Holmes, T.G. (1975) Effects of potassium chloride and smooth muscle relaxants on tension and cyclic nucleotide levels in rat myometrium. Can. J. Physiol. Pharmacol. 53: 1099-1107. Diamond, J., and Janis, R. A. (1978) Increases in cGMP levels may not mediate relaxant effects of sodium nitroprusside, verapamil and hydralazine in rat vas deferens. Nature. 271: 472-473. Dunham, E.W., Haddox, M.K., and Goldberg, N.D. (1974) Alteration of vein cyclic 3'-5'-nucleotide concentrations during changes in contractility. Proc. Natl. Acad. Sci. USA. 71: 815-819. Eigenthaller, M.,, Nolte, C , Halbrugge, M., and Walter, U. (1992) Concentration and regulation of cyclic nucleotides, cyclic nucleotide-dependent protein kinases and one of their major substrates in human platelets. Estimating the rate of cAMP regulated and cGMP regulated protein phosphorylation in intact cells. Eur. J. Pharmacol. 205:471-481. Fallier-Becker, P., Rupp, J., Fingerle, J., and Betz, E. (1990) Smooth muscle cells from rabbit aorta. In Piper.H.M. ed. Cell culture techniques in heart and vessel research. Berlin: Springer-Verlag., pp 248-270. Feelisch, M., and Noack, E. (1987) Nitric oxide formation from nitrovasodilators occurs independently of hemoglobin or non-iron heme. Eur. J. Pharmacol. 142: 465-469. Fesenko, E.E., Kolesnikov, S.S., and Lyubarsky, A.L. (1985) Induction by cyclic GMP of cationic conductance in plasma membrane of retinal rod outer segment. Nature. 313: 310-333. 139 Findlay, J.B.C. (1989) Purification of membrane proteins. In Harris, E., and Angal, S. eds. Protein purification: A Practical Approach. Oxford. IRL Press., pp 59-82. Fiscus, R.R., and Murad, F. (1988) cGMP-dependent protein kinase activation in intact tissues. Methods Enzymol. 159: 1150-1158. Fiscus, R.R., Rapoport, R.M., and Murad, F. (1983) Endothelium dependent and nitrovasodilator-induced activation of cyclic GMP-dependent protein kinase in rat aorta. J. Cycl. Nucl. Res. 9(6): 415-426. Fiscus, R.R., Rapoport, R.M., Waldman, S., and Murad, F. (1985) Atriopeptin II elevates cyclic GMP, activates cyclic GMP-dependent protein kinase and causes relaxation in rat thoracic aorta. Biochim. Biophys. Acta. 846: 179-184. Fiscus, R.R., Torphy, T.J., and Mayer, S.E. (1984) Cyclic GMP-dependent protein kinase activation in canine tracheal smooth muscle by methacholine and sodium nitroprusside. Biochim. Biophys. Acta. 805: 382-392. Flockhart, D.A., and Corbin, J.D. (1982) Regulatory mechanisms in the control of protein kinases. CRC Crit. Rev. Biochem. 12(2): 133-186. Forte, L.R., Thorne, P.K., Eber, S.L., Krause, W.J., Freeman, R.H., Francis, S.H., and Corbin, J.D. (1992) Stimulation of intestinal Cl- transport by heat stable enterotoxin: Activation of cAMP-dependent protein kinase by cGMP. Am. J. Physiol. 263: C607-C615. Francis, S.H., and Corbin, J.D. (1994) Progress in understanding the mechanism and function of cyclic-GMP-dependent protein kinase. Adv. Pharmacol. 26: 115-170. Francis, S.H., Noblett, B.D., Todd, B.W., Wells, J.N., and Corbin, J.D. (1988) Relaxation of vascular and tracheal smooth muscle by cyclic nucleotide analogs that preferentially activate purified cGMP-dependent protein kinase. Mol. Pharmacol. 34: 506-517. Furchgott, F. (1984) The role of endothelium in the responses of vascular smooth muscle to drugs. Ann. Rev. Pharmacol. Toxicol. 24: 175-197. Furchgott, F. (1988) The basis for the proposal that the acid-activatable factor from bovine retractor penis is inorganic nitrite and the endothelium-derived relaxing factor is nitric-oxide. In: Vanhoutte, P.M. ed. Mechanisms of vasodilation. New York: Raven Press, pp 401-414. 140 Furchgott, R.F., and Zawadzki, J.V. (1980) The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature. 288: 373-376. Furchgott, R.F., Zawadzki, J.V., and Cherry, P.D. (1981) Role of endothelium in the vasodilator response to acetylcholine. In: Vanhoutte, P.M., and Leusen, I. eds. Vasodilation. New York: Raven Press, pp 49-66. Gamm, D.M., Francis, S.H., Angelotti, T.P., Corbin, J.D., and Uhler, M.D. (1995) The Type II isoform of cGMP-dependent protein kinase is dimeric and possesses regulatory and catalytic properties distinct from the Type I isoform. J. Biol. Chem. 270: 27380-27388. Gill, G.N., Holdy, K.E., Walton, G.M., and Kanstein, C.B. (1976) Purification and characterization of cGMP-dependent protein kinase. Proc. Natl. Acad. Sci. 73:3918-3922. Glass, D.B., (1990) Substrate specificity of cyclic GMP-dependent protein kinase. In: Kemp, B.E., ed. Peptides and protein phosphorylation. Boca Raton, FL: CRC press, pp 209-239. Glass, D.B., and Krebs, E.G. (1982) Phosphorylation by cGMP-dependent protein kinase of synthetic peptide analogs of a site phosphorylated in histone 2B. J. Biol. Chem. 257: 1196-1200. Goldberg, N.D., Haddox, M.K., Hartle, D.K., and Hadden, J.W. (1973) The biological role of cyclic 3'-5'-guanosine monophosphate. Pharmacology and the future of man. Proc. Int. Congr. Pharmacol. 5:146-169. Goy, M.F. (1991) cGMP: the wayward child of the cyclic nucleotide family. Trends Neurosci. 14: 293-299. Gruetter, C.A., Barry, B.K., McNamara, D.B., Kadowitz, P.J., and Ignarro, L.J. (1979) Relaxation of bovine coronary artery and activation of coronary artery guanylate cyclase by nitric oxide, nitroprusside and a carcinogenic nitrosoamine. J. Cycl. Nucl. Res. 5: 211-224. Gruetter, C.A., Gruetter, D.Y., Lyon, J.E., Kadowitz, P.J., and Ignarro, L. (1981a) Relationship between cyclic guanosine 3'-5'-monophosphate formation and relaxation of coronary arterial smooth muscle by glyceryl trinitrate, nitroprusside, nitrite and nitric oxide: Effects of methylene blue and methehemoglobin. J. Pharmacol. Expt. Therap. 219:181-186. 141 Gruetter, C.A., Kadowitz, P.J., and Ignarro, L. (1981b) Methylene blue inhibits coronary arterial relaxation and guanylate cyclase activation by nitroglycerin, sodium nitrite, and amyl nitrite. Can. J. Physiol. Pharmacol. 59: 150-156. Haller, H., Smalwood, J.I., and Rasmussen, H. (1990) Protein kinase C translocation in intact vascular smooth muscle. Biochem. J. 270: 375-381. Hamilton, T.C., Weir, S.W., and Weston, T.H. (1986) Comparision of the effects of BRL 34915 and verapamil on electrical and mechanical activity in rat portal vein. Br. J. Pharmacol. 88: 103-111. Harper, J.F and Brooker, G. (1975) Femtomole sensitive radioimmunoassay for cyclic AMP and cyclic GMP after 2'0 acetylation by acetic anhydride in aqeous solution. J. Cycl. Nucl. Res. 1: 207-218. Hathaway, G.M., and Traugh, JA. (1982) Casein kinases: multipotential protein kinases. Curr. Top. Cell. Regul. 21: 101-127. Hei, Y., MacDonell, K.L., McNeill, J.H., and Diamond, J. (1990) Lack of correlation between activation of cyclic AMP-dependent protein kinase and inhibition of contraction of rat vas deferens by cAMP analogs. Mol. Pharmacol. 39: 233-238. Hirata, M., Koshe, K.P., Chang, C.H., Ikebe, T., and Murad, F. (1990) Mechanism of cyclic-GMP inhibition of inositol phosphate formation in rat aorta segments and cultured bovine aortic smooth muscle cells. J. Biol. Chem. 265: 1268-1273. Holzman, S. (1982) Endothelium-dependent relaxation by acetylcholine associated with larger rises in cyclic GMP in coronary arterial strips. J. Cycl. Nucl. Prot. Phosph. Res. 8: 409-419. Huggins, J.P., Cook, E.A., Piggott, J.R., Mattinsley, T.J., and England, P.J. (1989) Phospholamban is a good substrate for cyclic GMP-dependent protein kinase in vitro, but not in intact cardiac or smooth muscle. Biochem. J. 260: 829-835. Ignarro, L.J. (1989) Biological actions and properties of endothelium-derived nitric oxide formed and released from artery and vein. Circ. Res. 65: 1-23. Ignarro, L.J. (1990) Biosynthesis and metabolism of endothelium-derived nitric oxide. Ann. Rev. Pharmacol. Toxicol. 30: 535-560. Ignarro, L.J. (1994) Regulation of cytosolic guanyl cyclase by porphyrins and metalloporphyrins. Adv. Pharmacol. 26: 35-65. 142 Ignarro, L, Barry, B.K., Gruetter, D.Y., Edwards, J.V., Ohlstein, O.H., Gruetter, C.A., and Baricos, W.H. (1980a) Guanylate cyclase activation by nitroprusside and nitrosoguanidine is related to formation of S-nitrosothiol intermediates. Biochem. Biophys. Res. Commun. 94: 93-100. Ignarro, L , Buga, G.M., Wood, K.S., Byrns, R.E., and Chaudhari, G. (1987a) Endothelium derived relaxing factor produced and released from artery and vein is nitric oxide. Proc. Natl. Acad. Sci. USA. 84: 9265-9269. Ignarro, L.J., Burke, T.M., Woods, K.S., Wolin, M.S., and Kadowitz, P.J. (1984) Association between cyclic GMP accumulation and acetylcholine-elicited relaxation of bovine pulmonary artery. J. Pharmacol. Exp. Ther. 228: 682-690. Ignarro, L , Byrns, R.E., and Wood, K.S. (1987b) Endothelium-dpendent modulation of cGMP levels and intrinsic smooth muscle tone in isolated bovine intrapulmonary artery and vein. Circ. Res. 60: 82-92. Ignarro, L, Byrns, R.E., and Wood, K.S. (1988) Biochemical and pharmacological properties of endothelium derived relaxing factor and its similarity to nitric oxide radical. In: Vanhoutte, P.M. ed. Mechanisms of vasodilation. New York: Raven Press, pp427-435. Ignarro, L, Edwards, J.C., Gruetter, D.Y., Barry, B.K., and Gruetter, CA. (1980b) Possible involvement of S-nitrosothiols in the activation of guanylate cyclase by nitroso compounds. FEBS Lett. 110: 275-278. Ignarro, L.J., Harbison, R.G., Wood, M.S., and Kadowitz, P.J. (1986) Activation of purified soluble guanylate cyclase by endothelium-derived relaxing factor from intrapulmonary artery and vein: Stimulation by acetylcholine, bradykinin and arachidonic acid. J. Pharmacol. Exp. Ther. 237: 893-900. Ignarro, L., and Kadowitz, P.J. (1985) The pharmacological and physiological role of cyclic GMP in vascular smooth muscle relaxation. Ann. Rev. Pharmacol. Toxicol. 25: 171-191. Ignarro, L. J., Lippton, H., Edwards, J.C., Baricos, W.H., Hyman, A.L., Kadowitz, P.J., and Gruetter, CA. (1981) Mechanism of vascular smooth muscle relaxation by organic nitrates, nitrites, nitroprusside and nitric oxide: Evidence for the involvement of S-nitrosothiols as active intermediates. J. Pharmacol. Exp. Ther. 218: 739-749. Ignarro, L.J., Wood, K.S., and Wolin, M.S. (1984) Regulation of purified soluble gunaylate cyclase by porphyrins and metallo porphyrins: A unifying concept. Adv. Cyclic. Nucl. Prot. Phosph. Res. 17: 267-274. 143 Ishikawa, E., Ishikawa, S., Davis, J.W., and Sutherland, E.W. (1969) Determination of guanosine 3'-5'-monophosphate in tissues and of guanylyl cyclase in rat intestine. J. Biol. Chem. 244: 6371-6376. Ives, H.E., Casnellie, J.E., Greengard, P., and Jamieson, J.D. (1980) Subcellular localization of cyclic GMP-dependent protein kinase and its substrates in vascular smooth muscle. J. Biol. Chem. 255:3777-3785. Janis, R., and Diamond, J. (1979) Relationship berween cyclic nucleotide levels and drug-induced relaxation of smooth muscle. J. Pharmacol. Exp. Ther. 211: 480-484. Janis, R., and Diamond, J. (1981) Relationship between contraction and cyclic GMP levels in guinea pig taenia coli. Eur. J. Pharmacol. 70: 149-156. Jiang, H., Colbran, J.L., Francis, S.H., and Corbin, J., (1992) Direct evidence for cross activation of cGMP-dependent protein kinase by cAMP in pig coronary arteries. J. Biol. Chem. 267(2): 1015-1019. Joyce, N.C., DeCamilli, P., Lohmann, S.M., and Walter, U. (1986) cGMP-dependent protein kinase is present in high concentrations in contractile cells of the kidney vasculature. J. Cycl. Nucl. Res. 11: 194-198. Karaki, H., Sato, K., Ozaki, H., and Murakami, K. (1988) Effects of sodium nitroprusside on cytosolic calcium level in vascular smooth muscle. Eur. J. Pharmacol. 156: 259-266. Katsuki, S., Arnold, W.P., Mittal, C.K., and Murad, F. (1977a) Stimulation of guanylate cyclase by sodium nitroprusside, nitroglycerin and nitric oxide in various tissue preparations and comparison to the effects of sodium azide and hydroxylamine. J. Cycl. Nucl. Res. 3: 23-35.. Katsuki, S., Arnold, W.P., and Murad, F. (1977b) Effects of sodium nitroprusside, nitroglycerin, and sodium azide on levels of cyclic nucleotides and mechanical activity of various tissues. J. Cycl. Nucl. Res. 3: 239-247. Kauffman, R.F., Schenk, K.W., Utterback, G.B., Crowe, V.G., and Cohen, M.L. (1987) In vitro vascular relaxation by new inotropic reagents: relationship to phosphodiesterase inhibition and cyclic nucleotides. J. Pharmacol. Exp. Ther. 242:864-871. Kaupp, U.B. (1991) The cyclic nucleotide gated channels of vertebrate photoreceptors and olfactory epithelium. Trends Neurosci. 14:150-157. 144 Keilbach, A., Ruth, P., and Hofmann, F. (1992) Detection of cGMP dependent protein kinase isozymes by specific antibodies. Eur. J. Biochem. 208: 467-473. Keith, R.A., Burkman, A.M., Sokoloski, T.D., and Fertel, R.H. (1982) Vascular tolerance to nitroglycerin and cyclic GMP generation in rat aortic smooth muscle. J. Pharmacol. Exp. Ther. 221: 525-531. Keith, R.A., Burkman, A.M., Sokoloski, T.D., and Fertel, R.H. (1983) Nitroglycerin tolerance and cyclic GMP generation in the longitudinal smooth muscle of the guinea pig ileum. J. Pharmacol. Exp. Ther. 225: 29-34. Kobayashi, S., Kanaide, H., and Nakamura, M. (1985) Cytosolic free calcium transients in cultured vascular smooth muscle cells: microflourometric measurements. Science. 229: 553-556. Komalavillas, P., and Lincoln, T. (1994) Phoshorylation of the inositol 1,4,5-triphosphate receptor by cyclic GMP-dependent protein kinase. J. Biol. Chem. 269:8701-8707. Kramer, G.L., and Wells, J.N. (1979) Effects of phosphodiesterase inhibitors on cyclic nucleotide levels and relaxation of pig coronary arteries. Mol. Pharmacol. 16: 813-822. Kukovetz, W.R., Holzmann, S., Wurm, A., and Poch, G. (1979) Evidence for cyclic GMP-mediated relaxant effects of nitro-compounds in coronary smooth muscle. Naunyn-Schmid. Arch. Pharmacol. 310: 129-138. Kukovetz, W.R., Holzmann, S., and Poch, G. (1982) Function of cyclic GMP in acetylcholine-induced contraction of coronary smooth muscle. Naunyn-Schmid. Arch. Pharmacol. 319: 29-33. Kuo, J.F., and Greengard, P. (1970) Cyclic nucleotide-dependent protein kinase VI. Isolation and partial purification of a protein kinase activated by guanosine 3'-5'-monophosphate. J. Biol. Chem. 245: 2493-2498. Kuo, J.F., Shoji, M., and Kuo, W.N. (1978) Molecular and physiopathologic aspects of mammalian cyclic GMP-dependent protein kinase. Ann. Rev. Pharmacol. 18: 341-355. Laemmli, R., (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 227: 680-685. 145 Langdarf, W., Hullin, R., Gobel, C , and Hoffman, F (1986) Phosphorylation of cGMP-dependent protein kinase increases the affinity for cyclic AMP. Eur. J. Biochem. 154: 113-117. Langlands, J., and Diamond, J. (1992) Translocation of protein kinase C in bovine tracheal smooth muscle strips: the effect of methacholine and isoprenaline. Eur. J. Pharmacol. 227: 131-138. Lau, D.T.W., Chan, E.K., and Benet, L.Z. (1992) Glutathione-S-transferase-mediated metabolism of glyceryl trinitrate in subcellular fractions of bovine coronary arteries. Pharm. Res. 9:1460-1464. Lee, T.P., Kuo, J.F., and Greengard, P. (1972) Role of muscarinic cholinergic receptors in regulation of guanosine 3':5' - cyclic monophosphate content in mammalian brain, heart and intestinal smooth muscle. Proc.Natl. Acad. Sci. USA. 69: 3287-3291. Lincoln, T. (1983) Effects of nitroprusside and 8-Br-cGMP on the contractile activity of the rat aorta. J. Pharmacol. Exp. Ther. 224: 100-107. Lincoln, T. (1989) Cyclic GMP and mechanisms of vasodilation. Pharmacol. Ther. 41: 479-502. Lincoln, T.M., and Corbin, J. (1977) Adenosine 3':5'-cyclic monophosphate and guanosine 3':5'-cyclic monophosphate-dependent protein kinases: possible homologous proteins. Proc. Natl. Acad. Sci. USA. 74: 3239-3244. Lincoln, T.M., and Corbin, J. (1983) Characterization and biological role of the cGMP-dependent protein kinases. Adv. Cycl. Nucl. Res. 15: 139-192. Lincoln, T., and Cornwell, T.L. (1991) Towards an understanding of the mechanism of action of cyclic AMP and cyclic GMP in smooth muscle relaxation. Blood Vessels. 28: 129-137. Lincoln, T.M., Dills, W.L., and Corbin, J. (1977) Purification and subunit composition of cGMP-dependent protein kinase from bovine lung. J. Biol. Chem. 252: 4269-4275. Lincoln, T. M., Hall, C.L, Park, C.R., and Corbin, J.D. (1976) Guanosine 3';5'-cyclic monophosphate binding proteins in rat tissue. Proc. Natl. Acad. Sci. USA. 73:2559-2563. Lincoln, T., Komalavilas, P., and Cornwell, T.L. (1994) Pleiotropic regulation of vascular smooth muscle tone by cyclic GMP-dependent protein kinase. Hypertension. 23[Part2]: 1141-1147. 146 Lincoln, T.M., Thompson, M., and Cornwell, T.L. (1989) Purification and characterization of two forms of cGMP-dependent protein kinase from bovine aorta. J. Biol. Chem. 263: 235-242. Lohmann, S.M., Walter, U., Miller, P.E., Greengard, P., and DeCamilli, P. (1981) Immunohistological localization of cyclic GMP-dependent protein kinase in mammalian brain. Proc. Natl. Acad. Sci. USA. 73: 2559-2563 Luo, D-L., Nakazawa, M., Ishibashi, T., Kato, K., and Imai, S. (1993) Putative selective inhibitors of sarcoplasmic reticulum Ca 2 + pump ATPase inhibit relaxation by nitroglycerin and atrial natriuretic factor of the rabbit aorta contracted by phenylephrine. J. Pharmacol. Exp. Ther. 265: 1187-1192 Maehara, T., Fujita, A., Suthamnatpong, N., Takeuchi, T., and Hata, F. (1994) Differences in relaxant effects of cGMP on skinned muscle preparations from the proximal and distal colon of rats. Eur. J. Pharmacol. 261: 163-170. Marker! T., Vaandrager, A.B., Gambaryan, S., Pohler, D., Hausler, C , Walter, U., De Jonge, H.R., Jarchau, T., and Lohmann, S.M. (1995) Endogenous expression of type II cGMP-dependent protein kinase mRNA and protein in rat intestine. J. Clin. Invest. 96: 822-830. Martin, W., Villani, G., Jothianandan, D., and Furchgott, R.F. (1985) Selective blockade of endothelium-dependent and glyceryl trinitrate-induced relaxation by hemoglobin and by methylene blue in the rabbit aorta. J. Pharmacol. Exp. Ther. 232: 708-716. Meisheri, K.D., Taylor, C.J., and Saneii, H. (1986) Synthetic atrial peptide inhibits intracellular calcium release in smooth muscle. Am. J. Physiol. 250: C171-C174. McCune, R.W., and Gill, G.N. (1979) Positive cooperativity in guanosine 3':5'-monophosphate binding to guanosine 3':5'-monophosphate dependent nase. J. Biol. Chem. 254: 5083-5091. McDonald, B.J., and Bennet, B.M. (1990) Cytochrome P-450-mediated biotransformation of organic nitrates. Can. J. Physiol. Pharmacol. 68: 561-566. Mitchell, R.D., Glass, D.B., Wong, C-W., Angelos, K.L., and Walsh, D.A. (1995) Heat stable protein derived peptide substrate analogs: phosphorylation by cAMP-dependent and cGMP-dependent protein kinase. Biochemistry. 34: 528-534. 147 Morgan, J.P., and Morgan, K.G. (1984) Alteration of cytoplasmic ionized calcium levels in smooth muscle by vasodilators in the ferret. J. Physiol. 357: 539-551. Morgan, K.G., and Suematsu, E. (1990) Effects of calcium on vascular smooth muscle tone. Am. J. Hypertension. 3: 291S-298S. Murad, F., Mittal, C.K., Arnold, W.P., Katsuki, S., and Kimura, H. (1978) Guanylate cyclase: Activation by azide, nitro compounds, nitric oxide, and hydroxylamine radical and inhibition by hemoglobin and myoglobin. Adv. Cycl. Nucl. Res. 9: 145-158. Nakatsu, K., and Diamond, J. (1989) Role of cGMP in relaxation of vascular and other smooth muscle. Can. J. Physiol. Pharmacol. 67: 251-262. Nakane, M., and Murad, F. (1994) Cloning of guanylyl cyclase isoforms. Adv. Pharmacol. 26: 7-18. Nakazawa, M., and Imai, S. (1994) Rp-8-Br-guanosine-3',5'-cyclic monophosphorothioate inhibits relaxation elicited by nitroglycerin in rabbit aorta. Eur. J. Pharmacol. 253: 179-181. Napoli, S.A., Gruetter, C.A., Ignarro, L.J., and Kadowitz, P. (1980) Relaxation of bovine coronary arterial smooth muscle by cyclic GMP and analogs. J. Pharmacol. Expt. Ther.212: 469-473. Nelson, M.T., Ptalak, J.B., Worley, J.F. and Standen, N.B. (1990) Calcium channels, potassium channels, and voltage dependence of arterial smooth muscle tone. Am. J. Physiol. 259: C3 - C18. Oldham, K.G. (1990) Radiometric assays. In Eisenthal, R., and Danson, M.J. eds. Enzyme Assays. New York: IRL Press., pp 93-122. Palmer, R.M.J., Ferrige, A.G., and Moncada, S. (1987) Nitric oxide release accounts for the biological activity in endothelium derived relaxing factor. Nature. 327: 524-526. Pinna, L.A., Meggio, F., Marin, O., Perich, J.W., Boldyreff, B., and Issinger, O-G. (1992) Casein kinases: an atypical class of ubiquitous pleiotropic protein kinases. In: Papa, S., Azzi, A., and Tager, J.M. eds. Adenine nucleotides in cellular energy transfer and signal transduction. Basel: Birkhauser Verlag, pp 269-280. 148 Raeymaekers, L, Hofmann, F. and Casteels, R. (1988) cGMP-dependent protein kinase phosphorylates phospholamban in isolated sarcoplasmic reticulum from cardiac and smooth muscle. Biochem. J. 252: 269-273. Rashatwar, S.S., Cornwell, T.L., and Lincoln, T.M. (1987) Effects of 8-Bromo-cGMP on Ca 2 + levels in vascular smooth muscle cells: possible regulation of Ca 2 + - ATPase by cGMP-dependent protein, kinase. Proc. Natl. Acad. Sci. USA. 84: 5685-5689. Rapoport, R.M., Draznin, M.B., and Murad, F. (1982) Sodium nitroprusside-induced protein phosphorylation in intact rat is mimicked by 8-bromo cyclic GMP. Proc. Natl. Acad. Sci. USA. 79: 6470 - 6474. Rapoport, R,M., Draznin, M.B., and Murad, F. (1983) Endothelial-dependent relaxation in rat aorta may be mediated through cGMP-dependent protein phosphorylation. Nature. 306: 174 - 176 . Robertson, B.E., Schubert, R., Hescheler, J., and Nelson, M.T. (1993) cGMP-dependent protein kinase activates Ca-activated K channels in cerebral artery smooth muscle cells. Am. J. Physiol. 265: C299-C303. Robison, A, G., Butcher, R.W., and Sutherland, E.W. In Robison, A. G., Butcher, R.W., and Sutherland, E.W. eds. Cyclic AMP. New York: Academic Press., pp 17-47. Rogers, J., Hughes, R.G., and Mathews, E.K. (1988) Cyclic GMP inhibits protein kinase C-mediated secretion in rat pancreatic acini. J. Biol. Chem. 263: 3713-3719. Ruth, P., Langdraf, W., Keilbach, A., May, B., Egleme, C , and Hoffman, F. (1991) The activation of expressed cGMP-dependent protein kinase isozymes 1a and ip is determined by the different amino-termini. Eur. J. Biochem. 202: 1339-1344. Sandberg, M., Natarajan, V., Ronander, I., Kalderon, D., Walter, U., Lohmann, S., and Jahnsen, T. (1989) Molecular cloning and predicted full-length amino acid sequence of the type 1p isozyme of cGMP-dependent protein kinase from human placenta: Tissue distribution and developmental changes in rat. FEBSLett. 255:321-329 Saracevic, B., Brookes, V., Martin, T.J., Kemp, B.E., and Robinson, P.J. (1989) Atrial natriuretic peptide-dependent phosphorylation of smooth muscle cell particulate fraction proteins is mediated by cGMP-dependent protein kinase. J. Biol. Chem. 264: 20648-20654. 149 Schmidt, H.H.H.W., Lohman, S.M., and Walker, U. (1993) The NO and cGMP signal transduction system: regulation and mechanism of action. Biochim. Biophys. Acta. 1178:153-175. Schmidt, H.H., Warner, T.D., Ishii, K., Sheng, H., and Murad, F. (1992) Insulin secretion from pancreatic p cells caused by L-arginine-derived nitrogen oxides. Science. 255: 721. Schultz, G., Hardman, J.G., Schultz, K., Baird, C.E.,and Sutherland, E.W. (1973) The importance of calcium ions for the regulation of guanosine 3'-5'-cyclic monophosphate levels. Proc. Natl. Acad. Sci. USA. 70: 3889-3893. Schultz, K-D., Bohme, E., Kreye, V.A., and Schultz, G. (1979) Relaxation of hormonally-stimulated smooth muscular tissues by the 8-bromo derivative of cGMP. Naunyn-Schmid. Arch. Pharmacol. 306:1-9. Schultz, K-D., Schultz, K, and Schultz, G. (1977) Sodium nitroprusside and other smooth muscle relaxants increase cGMP levels in rat ductus deferens. Nature. 265:750-751. Schuman, E.M., and Madison, D.V. (1991) The intracelluar messenger nitric oxide is required for long term potentiation. Science. 254: 1503. Shekhar, K.R., Hatchett, J.B., Shabb, J.B., Wolfe, L , Francis, S.H., Wells, J.N., Jastroff, B., Butt, E., Chakinala, M.M., and Corbin, J.D. (1992) Relaxation of pig coronary arteries by new and potent cGMP analogs that selectively activate Type 1a, compared with Type ip, cGMP-dependent protein kinase. J. Pharmacol. Exp. Ther. 42: 103-108. Silver, P.J., Schmidt-Silver, C , and DiSalvo, J. (1982) p-adrenergic relaxation and cAMP kinase activation in coronary arterial smooth muscle. Am. J. Physiol. 242: H177-H184. Smith, J.A., Francis, S.H., and Corbin, J.D. (1992) Activation of the type ip isozyme of cGMP-dependent protein kinase by pre-incubation with MgATP and cAMP. FASEB. J. 6: A315 (Abstract). Smith, J.A., Francis, S.H., and Corbin, J.D. (1993) Autophosphorylation: a salient feature of protein kinases. Mol. Cell. Biochem. 127/128: 51-70. Somlyo, A.P., and Himpens, B. (1989) Cell calcium and its regulation in smooth muscle. FASEB J. 3: 2266-2276. 150 Suthamnatpong, N., Hata., F., Kanada, A., Takeuchi, T., and Yagasaki, O. (1993a) Mediators of nonadrenergic, noncholinergie inhibition in proximal, middle and distal regions of rat colon. Br. J. Pharmacol. 108: 348-355. Suthamnatpong, N., Maehara, T., Kanada, A., Takeushi, T., and Hata, F. (1993b) Dissociation of cyclic GMP level from relaxation of the distal, but not proximal colon of rats. Jpn. J. Pharmacol. 62: 387-393. Sutherland, E.,W., and Rail, T.N. (1957) The properties of an adenine ribonucleotide produced with cellular particles, ATP, Mg 2 + and epinephrine or glucagon. J. Am. Chem. Soc. 79: 3608. Takai, Y., Nakaya, S., Inoue, M., Kishimoto, A., Nishiyama, K., Yamamura, H., and Nishizuka, Y. (1976) Comparision of mode of activation of guanosine 3':5'-monophosphate-dependent and adenosine 3':5'-monophosphate protein kinases from silkworm. J. Biol. Chem. 251:1481-1487. Takio, K., Wade, R.D., Smith, S.B., Krebs, E.G., Walsh, K.A., and Titani, K. (1984) Guanosine cyclic 3',5'-phosphate dependent protein kinase, a chimeric protein homologous with two separate protein families. Biochemistry. 23: 4207-4218. Tanaguchi, J., Furukawa, K-l., and Shigekawa, M. (1993) Maxi K channels are stimulated by cyclic guanosine monophosphate-dependent protein kinase in canine coronary artery smooth muscle cells. Pflugers. Arch. 423: 167-172. Thomas, M.K., Francis, S.H., and Corbin, J.D. (1990) Substrate and kinase directed regulation of phosphorylation of a cGMP-binding phosphodiesterase by cGMP. J. Biol. Chem. 265: 14971-14978. Tobwin, H., Staehlin, T., and Gordon, J. (1979) Electrophoretic transfer of proteins from acrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Natl. Acad. Sci. USA. 76: 4350-4354. Triner, L, Nahas, G.G., Vulliemoz, Y., Overweg, N.I.A., Verosky, M., Habif, D.V., and Ngri, S.H. (1971) Cyclic AMP and smooth muscle function. Ann. N.Y. Acad. Sci. 185: 458-476. Tse, J., Mackenzie, III, C.W., and Donelly, T.E. (1981) A sensitive binding assay for cyclic GMP-dependent and cyclic AMP-dependent protein kinases in rat liver. Int. J. Biochem. 13: 1071-1079. Twort, C.H.C., and van Breemen, C. (1988) Cyclic guanosine monophosphate enhanced sequestration of Ca 2 + by sarcoplasmic reticulum in vascular smooth muscle. Circ. Res. 62: 961-964. 151 Vrolix, M., Raeymaekers, L, Wuytack, FD., Hoffman, F., and Casteels, R. (1988) Cyclic GMP-dependent protein kinase stimulates the plasmalemmal Ca 2 + pump of smooth muscle via phosphorylation of phosphatidylcholine. Biochem. J. 255: 855-863. Yoshida, Y., Sun, H-T, Cai, J-Q., and Imai, S. (1991) Cyclic GMP-dependent protein kinase stimulates the plasma membrane Ca 2 + -ATPase pump of vascular smooth muscle via phosphorylation of a 240-kDa protein. J. Biol. Chem. 266: 19819-19825. Uhler, M. (1983) Cloning and expression of a novel cyclic GMP-dependent protein kinase from mouse brain. J. Biol. Chem. 268: 13586-13591. Waldman, S.A., and Murad, F. (1987) Cyclic GMP synthesis and function. Pharmacol. Rev. 39: 89-92. Waldman, S.A., Rapoport, R.M., and Murad, F. (1984) Atrial natriuretic factor selectively activates particulate guanylate cyclase and elevates cyclic GMP in rat tissues. J. Biol. Chem. 259: 14332-14334. Walter, U. (1981) Distribution of cyclic GMP-dependent protein kinase in various rat tissues and cell lines determined by a sensitive and specific radioimmunoassay. Eur. J. Biochem. 118: 339-346. Walter, U. (1989) Physiological role of cGMP and cGMP-dependent protein kinase in the cardiovascular system. Rev. Physiol. Biochem. Pharmacol. 113: 42-88. Walton, G.M., and Gill, G.N. (1981) Regulation of cyclic nucleotide-dependent protein kinase activity by histones and poly (L-arginine). J. Biol. Chem. 256: 1681-1688. Wernert, W., Flockerzi, V., and Hoffman, F. (1989) The cDNA of the two isoforms of bovine cGMP-dependent protein kinase. FEBS. Lett. 251:191-196. Wolfe, L., Corbin, J.D., and Francis, S.H. (1989a) Characterization of a novel isozyme of cGMP-dependent protein kinase from bovine aorta. J. Biol. Chem. 264:7734-7741. Wolfe, L , Francis, S.H., and Corbin, J.D., (1989b) Properties of a cGMP-dependent monomeric protein kinase from bovine aorta. J. Biol. Chem. 264: 4157-4162. 152 Wikberg, J.E.S., and Andersson, R.G.G. (1978) A promoting action of cyclic GMP on contractions of guinea-pig vas deferens. Experientia. 34: 727-739. Winquist, R.J., Faison, E.P., Waldman, S.A., Schwartz, K., Murad, F., and Rapoport, R.A. (1984) Atrial natriuretic factor elicits an endothelium-independent relaxation and activates particulate cyclase in vascular smooth muscle. Proc.Natl. Acad. Sci. 81:7661-7664. Witt, J.J., and Roskoski Jr., R. (1975) Rapid protein kinase assay using phosphocellulose-paper absorption. Anal. Biochem. 66:253-258. Zhou, H., and Torphy, J.T. (1991) Relationship between cyclic guanosine monophosphate accumulation and relaxation of canine trachealis induced by nitrovasodilators. J. Pharmacol. Exp. Ther. 258: 972-978 153 

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0087242/manifest

Comment

Related Items