Search

Soluble guanylate cyclase catalyzes the formation of cyclic GMP using GTP as substrate. It is now well established that soluble guanylate cyclase is highly activated by nitric oxide, and that many of the effects of nitric oxide on various cells and tissues are mediated through increased production of cyclic GMP. This review discusses the evidence for the presence of soluble guanylate cyclases in different classes of cells in vertebrate retina and the role of these enzymes in retinal physiology. It is concluded that the enzyme is present in nearly every class of cells in the retina and that it may be involved in signal transmission between some cells and in the modulation of signal transmission between others.

Guanylate cyclase-activating proteins (GCAPs) control the activity of membrane bound guanylate cyclases in vertebrate photoreceptor cells. They form a permanent complex with guanylate cyclase 1 (ROS-GC1) at low and high Ca2+-concentrations. Five different target regions of GCAP-1 have been identified in ROS-GC1 at rather distant sites. These findings could indicate a multipoint attachment site for GCAP-1 or, alternatively, the presence of transient binding sites with short contact to GCAP-1. In addition some data are consistent with the operation of one or more transducer units, that represent regulatory regions without being direct binding sites. A permanent ROS-GCI/GCAP-1 complex is physiologically significant, since it allows a very short response time of cyclase activity when the intracellular Ca2+-concentration changes. Thereby, activation of cyclase participates in speeding up the recovery of the photoresponse after illumination and restores the circulating dark current.

By the formation of cGMP the NO-sensitive guanylyl cyclase plays a key role within the NO/cGMP signaling cascade involved in vascular regulation and neurotransmission. The prosthetic heme group of the enzyme acts as the NO sensor, and binding of NO induces conformational changes leading to an up to 200-fold activation of the enzyme. The unexpected fast dissociation half-life of NO of a few seconds is fast enough to account for the deactivation of the enzyme in biological systems. YC-1 and its analogues acting as NO sensitizers uncovered a new pharmacologically and conceivably physiologically relevant regulatory principle of the enzyme.

Calcium concentration in the dark-adapted retinal rod outer segment is in the 200 to 600 nM range, and the guanylate cyclase of rod outer segments is thought to be activated in response to a fall in calcium concentration triggered by light. Calcium-binding proteins that mediate such activation, i.e., activation in the absence of or presence of low nanomolar calcium concentrations, have been identified and termed GCAPs (Guanaylate Cyclase Activating Proteins). In the course of our search for GCAP-like proteins in bovine retina, we isolated a protein fraction that stimulated rod outer segment cyclase activity at calcium concentrations higher than those in dark-adapted rod outer segments. We purified the protein responsible for this calcium-dependent stimulation of cyclase activity and found it to be of 6-7 kDa molecular weight as judged by electrophoresis under denaturing conditions and about 40 kDa by gel filtration analysis. Maximum stimulation of cyclase activity was observed at 3-4 micromolar concentration of the protein. It required about 1.5 micromolar free calcium concentration for half-maximal activation of the enzyme. Partial amino acid sequencing of peptide fragments of the activator suggested that the protein was identical with S100b, a previously described calcium-binding protein. Further characterization with antibody specific for S100b supported this possibility. However, the protein isolated in our laboratory and termed CD-GCAP (Calcium-Dependent Guanylate Cyclase Activator Protein) was found to differ significantly from commercially available S100b in the magnitude and calcium dependence of cyclase activation. It was also found to be inactivated by hydroxylamine while S100b was resistant. Investigation into these differences showed that purification methods had a significant influence on the properties of the activator, producing a less active (S 100b) or more active (CD-GCAP) protein, but that it was, otherwise, one and the same protein. We conclude from this study that rod outer segment guanylate cyclase, unlike any cyclase known so far, is capable of activation by two different types of calcium-binding proteins, one that activates in response to a decrease in calcium concentration, and the other, described here, which activates in response to an increase in calcium-concentration. We hypothesize that this cyclase and others like it will be colocalized with one or the other type of activator depending upon the physiological requirement, i.e., activation in response to decreasing or increasing calcium concentration.

Membrane-bound guanylate cyclase activity was detected by ultracytochemistry at the electron microscope level in several mammalian tissues. The technique used in these studies allows the detection of active enzyme at the membrane site where it is located. In a few cases, such as normal and regenerating peripheral nerves and placenta, membrane-bound guanylate cyclase could be detected in the absence of stimulators of enzyme activity. However, in the majority of these studies membrane-bound guanylate cyclase was investigated following stimulation with natriuretic peptides, guanylin, or the Ca2+ sensor proteins, S100B and S100A1. In general, membrane-bound guanylate cyclase was localized to plasma membranes, in accordance with the functional role of this enzyme. Yet, in secretory cells the enzyme activity was localized on intracellular membranes, suggesting a role of membrane-bound guanylate cyclase in secretory processes. Finally, S100B and S100A1 were found to colocalize with membrane-bound guanylate cyclase on photoreceptor disc membranes and to stimulate enzyme activity at these sites in dark-adapted retinas in a Ca2+-dependent manner. The results of these analyses are discussed in relation to the proposed functional role(s) of this enzyme.

Nitric oxide (NO) is an important mediator molecule synthesized by several cell types including vascular smooth muscle cells (VSMC; Busse and Mulsch, 1990) and endothelial cells (EC; Rubanyi et al., 1986). It is generated by the conversion of L-arginine to L-citrulline, a reaction that is catalyzed by the enzyme nitric oxide synthase (NOS). NOS exists in at least three isoforms mainly classified to be either constitutive (cNOS) or inducible (iNOS). cNOS, as present in EC, is dependent on an increase in intracellular calcium concentration [Ca2+]i and produces NO in the picomolar range. In contrast, iNOS, as present in VSMC, is independent of an increase in [Ca2+]i and produces NO in the nanomolar range upon activation with endotoxins and cytokines. Recent studies have reported a growth inhibitory effect of nitric oxide in EC (Yang et al., 1994), cerebellar glial cells (Garg et al., 1992), and VSMC (Scott-Burden et al., 1993). Although it is not known whether endothelial derived NO exhibits any effects on VSMC growth in vivo, nitric oxide may play a role under conditions associated with endothelial cell damage and excessive VSMC proliferation, a feature characteristic of atherosclerotic blood vessel disease (Ross, 1993). A similar phenomenon also occurs after angioplasty as well as during hypertension and diabetes (Moncada et al., 1991). Hypertrophy of VSMC might at least in part be mediated by an inappropriate production and/or release of NO. The aim of the present study was to investigate the possible mechanism of NO-dependent inhibition of VSMC growth in vitro.

Cyclic GMP was discovered in rat urine (Ashman et al., 1963) at almost the same time the remarkable discovery of cyclic AMP was made (Sutherland and Rall, 1960). Mimicking the conceptual framework formulated for the second messenger role of cyclic AMP, cyclic GMP was also proposed to act as a second messenger of certain hormonal signals (reviewed in: Goldberg and Haddox, 1977). Great excitement was generated in the signal transduction field with the “Ying Yang” hypothesis, which predicted that the biological regulation of a cell is governed by the opposing biological activities of cyclic AMP and cyclic GMP (Goldberg and Haddox 1975). However, this and the general concept bearing upon the second messenger role of cyclic GMP in receptor-mediated signal transduction processes became seriously compromised because a) the attempts to demonstrate a hormonally dependent guanylate cyclase failed in every tested system (Goldberg and Haddox, 1977); in contrast, the guanylate cyclase activity was nonspecifically stimulated by a variety of agents such as polyunsaturated fatty acids, peroxides, hydroperoxides, free radicals, ascorbic acid, sodium nitroprusside, and other agents that presumably affected the oxidation-reduction potential of the biochemical reactions (Murad et al., 1979); and b) there was a general consensus that cyclic GMP-dependent protein kinase—at the time the only known effector enzyme of cyclic GMP signal transduction—does not phosphorylate a specific protein (substrate) clearly distinct from that of the cyclic AMP-dependent protein kinase (Gill and McCune, 1979). Despite the fact that the very initial studies had suggested the existence of distinct forms of soluble and particulate guanylate cyclase which could be differentiated from each other by their physical and certain kinetic properties (Kimura and Murad, 1975a, 1975b; Nakazawa et al., 1976; Siegel et al., 1976), the mere fact that guanylate cyclase could be nonspecifically stimulated by non-hormonal agents made these original observations on the independent existence of hormonally dependent guanylate cyclase suspect.

Certain 1,4-dihydropyridines (DHPs) interact with the ‘L’-type voltage-operated Ca2+ channel (VOCC) and thereby modulate Ca2+ entry into a variety of excitable tissues. Radioligand binding data indicates that the DHP nifedipine binds with high affinity to the VOCC and electrophysiological data indicates that a Ca2+ current can be modulated by such compounds1. The high affinity that many DHPs show for the VOCC is suggestive that other potential sites of action are of secondary importance with respect to their effects on smooth muscle function. It is thus of interest to note that comparatively low affinity sites have been described for the organic Ca2+ channel modulators2. Although such sites are of low affinity it must be emphasized that since the organic Ca2+ channel antagonists are hydrophobic drugs, they are capable of accumulating within vascular smooth muscle cells at tissue:medium ratios of 200–500:13.

This brief article will review some of the studies, from this laboratory and others, that have led to our present understanding of the synthesis of cyclic GMP and its role in smooth muscle relaxation and more specifically its role to mediate the effects of several classes of vasodilators. The regulation of the synthesis of cyclic GMP and some of its effects in several model systems have recently been reviewed in some detail. Readers are referred to these reviews (Waldman and Murad, 1987; Murad, 1986; Leitman and Murad, 1987) and the references therein for a more detailed discussion.

Cyclic GMP was discovered in biological material in 19631, and guanylate cyclase, the enzyme that catalyzes the formation of the cyclic nucleotide from GTP, was first described in 19692–5. Because of its chemical similarity to cyclic AMP, cyclic GMP has long been assumed to be involved in cellular regulation. This assumption has been reinforced by observations that cellular levels of cyclic GMP rapidly increase in response to agonists that affect such processes as myocardial and smooth muscle contraction, cell division, exocrine and endocrine secretion, platelet aggregation and neuronal activity (see reference 6 for a thorough review of this area). The role of cyclic GMP in these processes, however, is still obscure.

Mitogen activation of human peripheral blood lymphocytes (HPBL) occurs as a consequence of several early membrane events. One of the earliest is the Ca2+ -dependent stimulation of guanylate cyclase (Coffey et al., 1981). Experiments with the T-cell mitogens phytohemagglutinin (PHA), concanavalin A (Con A), and phorbol myristate acetate (PMA) suggested that phospholipase and lipoxygenase activation are important in the sequence of early membrane events leading to guanylate cyclase stimulation, cGMP accumulation, and DNA synthesis (Coffey and Hadden,1981, 1983a,b). Cyclooxygenase (Coffey et al., 1981) and thromboxane synthetase activities (Gordon et al., 1981) were not found to be essential for mitogenesis.

Cyclic GMP in an intact cell system was reported to increase by various biologically active compounds, including muscarinic cholinergic agonists and excitatory amino acids, and supposed to act as intracellular messenger of the various neurotransmitters and hormones. Guanylate cyclase, a cyclic GMP-generating enzyme, in a cell free system can not be stimulated by agents increasing cyclic GMP contents. This suggests that the cyclase is activated by a substance generated or a process taking place within cells upon hormone stimulation. Thus cellular components such as Ca2+, fatty acids and their metabolites that are intracellularly liberated on the stimulation have been supposed to be possible physiological activators of guanylate cyclase in intact cells.

In light of recent results emphasizing the primary importance of capillary endothelium in the maintenance of the blood-brain barrier (4, 24, 41, 43), it has become important to perform biochemical studies on a subcellular fraction enriched in brain capillaries. A procedure for the isolation of brain capillaries was originally developed in our laboratory (26) and has been modified by several authors (5, 15, 32, 47). The availability of the micromethod has promoted different studies aiming at the better elucidation of regulatory mechanisms in relation to the adaptations of cerebral blood flow and the blood-brain barrier.

The lymphocyte has provided an important model for the study of the early biochemical events associated with the initiation of cellular proliferation. Resting lymphocytes (G0 or restricted G1 phase), as obtained from human peripheral blood, can be induced in large numbers to undergo clonal proliferation using a variety of mitogens. The two most commonly employed plant lectin mitogens are phytohem-agglutinin (PHA) and concanavalin A (Con A). The initiation process is characterized by rapid, saturable binding of the lectin to cellular receptors (