Increase of cyclic GMP induced in murine thymocytes by thymosin fraction 5

Using a radioimmunoassay (RIA) for the determination of adenosine 3'5' cyclic monophosphate (cAMP) and an acetylation-RIA procedure to measure guanosine 3'5' cyclic monophosphate (cGMP), we observed that cGMP levels, but not cAMP levels, were significantly elevated in murine thymocytes which had been incubated with preparations containing the thymic hormone, thymosin. Stimulation of intracellular cGMP levels was seen as early as 1 minute after incubation with thymosin fraction 5 and was maximal at approximately 10 minutes. Dose response studies indicated an optimum stimulation of cGMP with a thymosin concentration of 100 microg/ml. A control spleen fraction prepared by an identical procedure as fraction 5 did not affect the levels of either cyclic nucleotide.     

Effects of oxytocin and methacholine on cyclic nucleotide levels of rabbit myometrium

The effects of oxytocin and methacholine on cyclic nucleotide levels in estrogen-primed rabbit myometrium were studied in the presence and absence of 1-methyl-3-isobutyl xanthine (MIX), a phosphodiesterase inhibitor. In the absence of MIX, methacholine increased guanosine 3',5'-cyclic monophosphate (cGMP) levels at a time when contraction was decreasing, but had no influence on adenosine 3',5'-cyclic monophosphate (cAMP) levels. In contrast, oxytocin did not elevate cGMP, but rapidly decreased cAMP levels. MIX (1 mM) increased both cAMP and cGMP levels. Oxytocin or methacholine further increased cGMP, indicating activation of guanylate cyclase. Oxytocin- but not methacholine-induced stimulation of guanylate cyclase was abolished in Ca2+-free solution. Oxytocin increased cAMP over the levels produced by MIX alone, whereas methacholine decreased cAMP below the MIX control values; these effects were insensitive to indomethacin. Tissue levels of cGMP and cAMP did not directly correlate with isometric tension. The results also indicate that both oxytocin and methacholine stimulate guanylate cyclase but have opposing effects on adenylate cyclase of rabbit myometrium.     

Increased cyclic GMP levels lead to a stimulation of elastin production in ligament fibroblasts that is reversed by cyclic AMP

The effects of cyclic nucleotides on elastin synthesis were studied in ligamentum nuchae fibroblasts by adding exogenous cyclic nucleotide derivatives or beta-adrenergic agents to cell culture medium. Elastin synthesis was enhanced (approximately 80%) by dibutyryl cGMP (Bt2cGMP) in concentrations ranging from 0.01 to 100 nM. Two other cGMP derivatives, 8-bromoguanosine 3':5'-cyclic monophosphate (8-Br-cGMP) and 2'-deoxy-cGMP, were also potent stimulators of elastin synthesis. In the absence of calcium, basal elastin production was substantially decreased (40% of control) and cGMP analogs no longer stimulated elastin synthesis, suggesting a role for calcium in the cGMP response. Bt2cAMP had no demonstrable effect on elastin production except at high concentrations which produced a nonspecific decrease equivalent to the decrease in total protein synthesis. Similarly, elevation of endogenous cellular cAMP levels by beta-adrenergic stimulation produced no change in elastin production. When 8-Br-cGMP was added to cells together with Bt2cAMP, cGMP-dependent stimulation of elastin production was abolished by cAMP in a dose-dependent fashion. These results suggest a coordinated means by which elastin production is controlled in ligament cells, i.e. increased cGMP levels lead to a stimulation of elastin production that is reversed by cAMP.     

Plasma atrial natriuretic peptide and cyclic nucleotide levels before and after a marathon

Plasma alpha-atrial natriuretic peptide (alpha-ANP) concentration and levels of cyclic nucleotides [guanosine 3',5'-cyclic monophosphate (cGMP) and adenosine 3',5'-cyclic monophosphate (cAMP)] were studied in 23 runners before and after a marathon race. Blood samples were drawn from an antecubital vein the morning before the race (base line), at 3 P.M. (i.e., 2 h before the start), on arrival, and 12 and 36 h and 7 days later. Compared with the base-line values of plasma alpha-ANP (5 pmol/l), cGMP (3.8 nmol/l), and cAMP (15.8 nmol/l), the plasma levels of alpha-ANP, cGMP, and cAMP were increased immediately after the marathon, respectively, to 12.0 pmol/l, 12.7 nmol/l, and 50.5 nmol/l. The increase in the plasma alpha-ANP concentration was related (r = 0.85; P less than 0.001) to the changes in plasma cGMP, plasma lactate, hematocrit, and body weight. The plasma cGMP and cAMP concentrations had returned to the prerace levels 12 h after the marathon, whereas the plasma alpha-ANP concentration was significantly lower (3.1 pmol/l) than the base-line values and increased above the prerace values 36 h (7.5 pmol/l) and 7 days (6.8 pmol/l) after the marathon. The plasma cGMP level was also higher 36 h (5.4 nmol/l) and 7 days (5.0 nmol/l) after the marathon race.     

cGMP and cyclic nucleotide-gated channels participate in mouse sperm capacitation

During capacitation of mammalian sperm intracellular [Ca(2+)] and cyclic nucleotides increase, suggesting that CNG channels play a role in the physiology of sperm. Here we study the effect of capacitation, 8Br-cAMP (8-bromoadenosine 3',5'-cyclic monophosphate) and 8Br-cGMP (8-bromoguanosine 3',5'-cyclic monophosphate) on the macroscopic ionic currents of mouse sperm, finding the existence of different populations of sperm, in terms of the recorded current and its response to cyclic nucleotides. Our results show that capacitation and cyclic nucleotides increase the ionic current, having a differential sensitivity to cGMP (cyclic guanosine monophosphate) and cAMP (cyclic adenosine monophosphate). Using a specific inhibitor we determine the contribution of CNG channels to macroscopic current and capacitation.     

The effect in vivo of alterations in gonadotropins and cyclic nucleotides on oocyte maturation in the hamster

The temporal relationship of changes in cAMP and cGMP to oocyte maturation was examined in proestrous hamsters (day 4). The first series of experiments showed, in normal cycling hamsters, an increase in cAMP and a decrease in cGMP at 1400 h shortly after the rise in LH with oocyte maturation beginning at 1800 h. When a second group of animals was injected with phenobarbital at 1200 h to block the LH surge, no significant change occurred in either cyclic nucleotide and oocyte maturation was prevented. In the second series of experiments single injections of either saline, hCG (30 IU), LH (10 micrograms) or FSH (10 micrograms) were given each to a group of animals at 0900 h on day 4. Animals were killed at five time intervals between 15 min and 3 h following the injection. LH and hCG stimulated a simultaneous increase in cAMP and decline in cGMP. The injection of FSH, however, did not cause an increase in cAMP but still produced a sharp decline in cGMP. Oocyte maturation occurred at 3 h in those animals injected with gonadotropins. Animals injected with saline showed neither cyclic nucleotide changes nor oocyte maturation. When cAMP and cGMP levels were expressed as a ratio (cAMP/cGMP) a significant increase occurred in the normal cycling animals and in those injected at 0900 h with gonadotropins. Phenobarbital and saline injected control animals showed no significant increase in the cAMP/cGMP ratio and no oocyte maturation. The results of these experiments and previous studies by this investigator indicate that cGMP may play an important role in oocyte maturation in the hamster prior to the LH surge. Since, in the presence of gonadotropins, the cAMP/cGMP ratio increases prior to oocyte maturation, it may be that the cyclic nucleotide ratio is also of importance in this process. Previous work by Hubbard and Terranova (1) has shown that guanosine 3':5' cyclic monophosphate (cGMP), can inhibit spontaneous maturation of hamster oocytes in vitro. This inhibitory action was dose dependent and overcome by LH. The cGMP-mediated inhibition occurred only in cumulus-enclosed oocytes, while adenosine 3':5' cyclic monophosphate (cAMP) inhibited spontaneous maturation in both cumulus-enclosed and denuded oocytes. The results of this study suggested that cGMP may play a role in inhibiting oocyte maturation prior to the LH surge. LH, the initiator of oocyte maturation, has also been shown in the intact proestrous rat and hamster to cause a decrease in cGMP at the same time that cAMP is rising (2,3).(ABSTRACT TRUNCATED AT 400 WORDS)     

Levels of cyclic GMP and cyclic AMP in regenerating forelimbs of adult newts following denervation

The effect of short-term denervation (0, 12, 24, and 72 hours) on the levels of cyclic 3'5'-guanosine monophosphate (cGMP) and cyclic 3'5'-adenosine monophosphate (cAMP) in adult newt (Notophthalmus viridescens) forelimbs at 15, 22, and 35 days of regeneration was investigated. Regenerate blastema and stump cyclic nucleotide levels were compared with those of the contralateral intact forelimb and hindlimb, with levels in the normally regenerating blastema, and with levels measured in the forelimbs of intact, nonoperated animals. Variations in cyclic nucleotide levels occurred according to regeneration stage and tissue type. Changes in level were noted immediately upon denervation and subsequently at other sample times in all regenerate and control series. Parallel fluctuations occurred in regenerate stump and contralateral intact forelimbs. Our results from nonamputated denervated and sham-denervated animals indicate that short-term, denervation-associated cyclic nucleotide fluctuations cannot be attributed solely to the loss of innervation.     

Cyclic AMP-stimulated chloride fluxes in dialyzed barnacle muscle fibers

Unidirectional chloride efflux and influx were studied in giant barnacle muscle fibers that were internally dialyzed. When cyclic 3'5'- adenosine monophosphate (cAMP) was included in the dialysis fluid, both unidirectional fluxes were stimulated by about the same amount. This stimulation was not associated with measurable changes either in membrane electrical conductance or with net movements of chloride. The stimulation required the trans-side presence of chloride. The stimulated flux was inhibited by the sulfonic acid stilbene derivatives 4-acetamido-4'-isothiocyanostilbene-2',2'-disulfonate (SITS) and 4,4'- diisothiocyanostilbene-2,2'-disulfonate (DIDS) or by furosemide. When cAMP was presented in high concentrations (10-5 M), the effect on chloride fluxes was characterized by a desensitization phenomenon. This desensitization was not the result of an increased amount of phosphodiesterase activity, but may be related to ATP and/or intracellular calcium levels. These results further support the hypothesis that the barnacle sarcolemma possesses a specialized chloride transport mechanism that largely engages in Cl-Cl exchange under conditions of normal intracellular pH.     

Immunofluorescent localization of cGMP,cGMP-dependent protein kinase,calmodulin and cAMP in the rat uterus

Cyclic guanosine 3',5' monophosphate (cGMP), cGMP-dependent protein kinase, calmodulin and cyclic adenosine 3',5' monophosphate (cAMP) were localized in the uterus of the immature rat by an indirect immunofluorescence technique. cGMP, cGMP-dependent protein kinase and calmodulin were detected predominantly along epithelial and myometrial plasma membranes and in the adjacent cytoplasm. In contrast, cAMP immunoreactive material was found principally in the cytoplasm of connective tissue. After administration of 17 beta-estradiol, similar time-dependent changes were observed in the localization of cGMP, cGMP-dependent protein kinase and calmodulin in all uterine cell types. For the three compounds, nucleolar-like distribution of the immunofluorescence appeared approximately 12 h after treatment. A more dispersed, reticular distribution of the nuclear fluorescent staining was observed 20-24 h after hormonal treatment. Estrogen did not affect the localization of cAMP. The simultaneous mobilization of cGMP, cGMP-dependent protein kinase and calmodulin towards the same nuclear loci suggests concerted roles for these three molecules in nuclear metabolic processes during the development of the uterotrophic action of estrogens.     

Hyperkalemia, hyperglycemia, and plasma levels of cyclic AMP and cyclic GMP induced by portal vein injection of cyclic nucleotides and their butyryl derivatives into dogs

The levels of serum potassium, blood glucose, and plasma adenosine cyclic 3':5'-monophosphate (cAMP) and guanosine cyclic 3':5'-monophosphate (cGMP) were studied after the portal vein injection of cyclic nucleotides and their derivatives, (cAMP, cGMP, N6, O2'-dibutyryl adenosine 3':5'-monophosphate (DBcAMP), N6-monobutyryl adenosine cyclic 3':5'-monophosphate (NMBcAMP), and O2'-monobutyryl adenosine cyclic 3':5'-monophosphate (OMBcAMP), into dogs. Dose-related hyperglycemic responses were observed after the injection of DBcAMP (1-8 mg/kg). Transient and prominent hyperkalemia and hyperglycemia were caused by the injection of DBcAMP, NMBcAMP, and OMBcAMP (4 mg/kg). The hyperkalemic response was highest with NMBcAMP (1.22 mequiv./L), followed by OMBcAMP (0.64), DBcAMP (0.54), cGMP (0.47), and cAMP (0.41), whereas the hyperglycemic response was highest with NMBcAMP (146 mg/100 mL), followed by DBcAMP (93.6), OMBcAMP (77.1), and cAMP (56.0), and there was only a slight change with cGMP (28.4) compared with the control. The plasma level of cAMP was maximal with DBcAMP (1.92 nmol/mL), followed by NMBcAMP (1.28) and OMBcAMP (0.76), whereas the plasma levels of cGMP showed no evident change, except that caused by DBcAMP (0.27). Of the cyclic nucleotides tested, NMBcAMP was found to be most potent in causing both hyperkalemia and hyperglycemia. Based on these results, possible correlations between hyperkalemia, hyperglycemia, and plasma levels of cAMP and cGMP are discussed.     

Cyclic nucleotide phosphodiesterases in somatic and germ cells of mouse seminiferous tubules

The distribution of phosphodiesterase forms in somatic and germ cells, and their variations during testicular development and germ cell differentiation have been investigated. Seminiferous tubules from immature mice and Sertoli cells in culture possessed two enzyme activities which were comparable to forms described for different tissues and species: (a) a calcium-calmodulin-dependent enzyme with high affinity for guanosine 3',5'-(cyclic)-monophosphate (cGMP), and (b) a calcium-calmodulin-independent enzyme with high affinity for adenosine 3',5'-(cyclic)-monophosphate (cAMP) the activity of which increased in cultured Sertoli cells after treatment with FSH or dibutyryl cAMP. Seminiferous tubules from adult animals and germ cells at the meiotic and post-meiotic stage of differentiation possessed two enzyme forms that could be distinguished from those present in somatic cells of the seminiferous tubules: (a) a calcium-calmodulin-dependent form with high affinity for both cAMP and cGMP, similar to forms described in other tissues from different species, and (b) a calcium-calmodulin-independent phosphodiesterase with high affinity for cAMP and present only in post-meiotic cells, previously identified also in germ cells of the rat.     

Exogenous cAMP and cGMP modulate branching in fusarium graminearum.

A study was made of the effects of adenosine 3',5'-cyclic monophosphate (cAMP), guanosine 3',5'-cyclic monophosphate (cGMP) and choline on the morphology and growth of a wild-type strain (A 3/5) and a highly branched, 'colonial' mutant strain (C106) of Fusarium graminearum. Addition of up to 50 mM-cAMP or cGMP to the medium had no effect on the specific growth rate of strain A 3/5. For strain A 3/5, but not for strain C106, exogenous cAMP caused significant decreases in both mean hyphal extension rate (E) and hyphal growth unit length (G), i.e. cAMP caused mycelia of strain A 3/5 to branch profusely. By contrast, for both strains, cGMP caused significant increases in both E and G, i.e. exogenous cGMP caused mycelia to branch more sparsely. The effects of exogenous cGMP and choline in increasing E and G were synergistic, but the effects of cGMP and choline counteracted the effect of cAMP. The mutant phenotype of strain C106 was not correlated with altered levels of endogenous cAMP or cGMP.     

Effects of intracellular cyclic AMP and cyclic GMP levels on DNA synthesis of young-adult rat hepatocytes in primary culture.

Possible roles of dibutyryladenosine 3',5'-cyclic monophosphate (cAMP) and dibutyryl-guanosine 3',5'-cyclic monophosphate (cGMP) in regulation of hepatocyte DNA synthesis were examined using primary cultures of young-adult rat hepatocytes maintained in arginine-free medium. Throughout the experimental period, nonparenchymal cells were hardly observed in the selective medium. When epidermal growth factor (EGF) was added to the cultures, a transient increase in the intracellular cAMP level preceded the elevation of hepatocyte DNA synthesis. EGF-stimulated hepatocyte DNA synthesis was remarkably enhanced by the elevation of the intracellular cAMP level induced by treatment with cAMP alone or a combination of cAMP and theophylline, an inhibitor of cyclic nucleotide phosphodiesterase. Furthermore, the early elevation of intracellular cAMP alone, which was induced by treatment with the combination of cAMP and theophylline, caused a remarkable increase in hepatocyte DNA synthesis. On the other hand, addition of EGF to the cultures caused a rapid decrease in the intracellular cGMP level followed by an increase in hepatocyte DNA synthesis. EGF-stimulated hepatocyte DNA synthesis was severely suppressed or completely inhibited by the elevation of the intracellular cGMP level induced by treatment with cGMP alone or a combination of cGMP and dipyridamole, a specific inhibitor of cGMP phosphodiesterase. These findings indicate that cAMP and cGMP act oppositely on the regulation of DNA synthesis of young-adult rat hepatocytes in primary culture: cAMP plays a positive role, whereas cGMP plays a negative role. Also it is strongly suggested that an early elevation of the intracellular cAMP level is essential for the onset of DNA synthesis in hepatocyte primary cultures.     

Effect of short-chain fatty acids on cyclic 3',5'-guanosine monophosphate-mediated colonic secretion

Short chain fatty acids (SCFA) prevent and reverse cyclic 3',5'-adenosine monophosphate (cAMP) but not Ca(2+)-mediated Cl- secretion. Mucosal [HCO3-]i has an opposite effect on these secretagogues. We examined whether SCFA and [HCO3-]i affect cyclic 3',5'-guanosine monophosphate (cGMP)-induced secretion. Stripped segments of male Sprague-Dawley rat (Rattus norvegicus) proximal and distal colon, and cultured T84 cells were studied in Using chambers, and pHi and [HCO3-]i were determined. Mucosal [cGMP] was measured in proximal colon. In T84 cells, the increase in Cl- secretion (measured as Isc) induced by mucosal 0.25 microM Escherichia coli heat-stable enterotoxin (STa) was prevented/reversed by bilateral 50 mM Na+ butyrate (71%/73%), acetate (58%/76%), propionate (68%/73%) and (poorly metabolized) isobutyrate (80%/79%). In proximal colon in HCO3- Ringer, basal Cl- secretion was not affected by [HCO3-]i or 25 mM butyrate. Mucosal 0.25 microM STa decreased net Na+ and Cl- absorption. Bilateral but not mucosal 25 mM SCFA reversed STa-induced effects on Na+ absorption and Cl- secretion. Bilateral and mucosal 25 mM SCFA but not [HCO3-]i prevented STa-induced Cl- secretion and increases in mucosal [cGMP]. STa did not produce Cl- secretion in distal colon. It was concluded that SCFA but not [HCO3-]i can prevent and reverse cGMP-induced colonic Cl- secretion.     

A cell-based PDE4 assay in 1536-well plate format for high-throughput screening

The cyclic nucleotide phosphodiesterases (PDEs) are intracellular enzymes that catalyze the hydrolysis of 3,'5'-cyclic nucleotides, such as cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP), to their corresponding 5'nucleotide monophosphates. These enzymes play an important role in controlling cellular concentrations of cyclic nucleotides and thus regulate a variety of cellular signaling events. PDEs are emerging as drug targets for several diseases, including asthma, cardiovascular disease, attention-deficit hyperactivity disorder, Parkinson's disease, and Alzheimer's disease. Although biochemical assays with purified recombinant PDE enzymes and cAMP or cGMP substrate are commonly used for compound screening, cell-based assays would provide a better assessment of compound activity in a more physiological context. The authors report the development and validation of a new cell-based PDE4 assay using a constitutively active G-protein-coupled receptor as a driving force for cAMP production and a cyclic nucleotide-gated cation channel as a biosensor in 1536-well plates.     

Direct regulation of smooth muscle contractile elements by second messengers

The effects of adenosine 3',5'-cyclic monophosphate (cAMP), guanosine 3',5'-cyclic monophosphate (cGMP) and phorbol 12,13 dibutyrate (PDBu) on the Ca2+ sensitivity of the contractile elements in the rat mesenteric artery were investigated, using a method of permeabilizing smooth muscle with Staphylococcal alpha-toxin. Both cAMP and cGMP relaxed the permeabilized rat mesenteric artery at the intracellular Ca2+ concentrations [( Ca2+]i) held constant with Ca2+ EGTA buffer and Ca2+ ionophore, ionomycin. In addition, forskolin and sodium nitroprusside which activate adenylate and guanylate cyclases, respectively, also induced relaxation at a fixed [Ca2+]i. In contrast PDBu which stimulates protein kinase C caused an increase in force at a constant [Ca2+]i which could be partially reversed by cAMP or cGMP. These results indicate that second messengers exert direct control over smooth muscle Ca2+ sensitivity of the contractile elements, which is of physiologic and pharmacologic importance.     

PKA from Saccharomyces cerevisiae can be activated by cyclic AMP and cyclic GMP.

Analysis of Saccharomyces cerevisiae genome revealed no sequence homologous to cyclic GMP (cGMP) dependent protein kinase from other organisms. Here we demonstrate that cyclic AMP (cAMP) dependent protein kinase purified from S. cerevisiae was almost equally activated by cAMP and cGMP in 3 x 10(-6) M concentrations of either nucleotide in the presence of Mg2+ ions. Interestingly, if Mn2+ ions were used instead of Mg2+, cGMP was only 30% as effective as cAMP in the activation of cAMP-dependent protein kinase. Analogs of cAMP such as 8-chloro-cAMP and 3':5'-cyclic monophosphate of ribofuranosylbenzimidazole were as potent as cAMP in the enzyme activation, while N6,2'-O-dibutyryl-cAMP activated the enzyme to a lower extent. It was also found that yeast cAMP-dependent protein kinase can be activated by limited proteolytic digestion. The results presented were obtained with protamine and ribosomal protein S10 used as phosphorylation substrates.     

Optimization and validation of a reporter gene assay for screening of phosphodiesterase inhibitors in a high throughput system

Phosphodiesterases (PDEs) hydrolyze cyclic nucleotides, cyclic adenosine monophosphate (cAMP) and guanosine monophosphate (cGMP) into inactive 5' monophosphates, and exist as 11 families. Inhibitors of PDEs allow the elevation of cAMP and cGMP, which leads to a variety of cellular effects including airway smooth muscle relaxation and inhibition of cellular inflammation or of immune responses. PDE4 inhibitors specifically prevent the hydrolysis of cAMP. We have validated the manually developed reporter gene assay in a high-throughput screening format that allows for fast and cost-effective identification of potential inhibitors of PDE4 isozymes. The assay is sensitive and robust, with a Z' value of >0.5. The assay is also amenable to 384-well format.     

Resting and concanavalin-A stimulated levels of cyclic nucleotides in splenic cells of aging mice with spontaneous cancers.

The resting levels of cyclic 3′, 5′ -adenosine monophosphate (cAMP) and cyclic 3′, 5′ -guanosine monophosphate (cGMP) in splenic lymphoid cells of 25 aged (C57BL/10 × C3H)F₁ hybrid mice with spontaneous tumors, including 5 with hepatoma, 10 with lung tumor, 2 with lymphoma, and 8 with several varieties of tumor, as well as in 18 young and 13 tumor-free aging mice, were measured. The alterations in cyclic nucleotide levels in spleen cells characteristic of normal aging in tumor-free animals may be additionally influenced by the occurrence of spontaneous neoplasia. Furthermore, the levels may vary with different types of late-life tumors. For example, levels of cAMP in resting spleen cells of old mice with hepatomas were not different than in age-matched controls, whereas spleen, cells from old mice with lung tumors showed exceedingly high levels of resting cAMP. Upon $\text{in vitro}$ stimulation by Con-A, the splenic lymphoid cells from mice bearing spontaneous late-life lung and liver tumors displayed different kinetic patterns of percent changes in cAMP, cGMP and cAMP/cGMP ratios when compared to either young or age-matched tumor-free controls. Thus, both resting and Con-A stimulated levels of cAMP and cGMP and their ratios in splenic lymphoid cells may be affected by spontaneous cancer elsewhere in the body, including cancer of non-lymphoid type and origin. These findings plus the known functional decline in immune response capacity and the increase in spontaneous tumor incidence with age may suggest the existence of a complex relationship among cyclic nucleotide levels, immunity, aging, and cancer.     

Immunocytochemical localization of cAMP and cGMP in cells of the rat carotid body following natural and pharmacological stimulation.

Although the chemoreceptive function of the carotid body has been known for many decades, the cellular mechanisms of sensory transduction in this organ remain obscure. Common elements in the transductive processes of many cells are the cyclic nucleotide second messengers, cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP). Studies from our laboratory have revealed stimulus-induced changes in cyclic nucleotide levels in the carotid body as measured by RIA, but such changes in second messenger levels have not been localized to specific cellular elements in the organ. The present immunocytochemical study utilized the avidin-biotin-peroxidase method to investigate the distribution of cAMP and cGMP in the rat carotid body and to assess changes in the intensity of immunostaining following in vitro stimulation by hypoxia, forskolin, sodium nitroprusside, high potassium, and atrial natriuretic peptide. Both cAMP and cGMP immunoreactivity were localized to type I cells of organs maintained in vivo and fixed by perfusion. Organs exposed to 100% O2-equilibrated media in vitro produced low but visible levels of cAMP immunoreactivity in a majority of type I cells; hypoxia (5% O2-equilibrated media) for 10 min moderately increased the level of immunoreactivity; forskolin (10(-5) M), or forskolin combined with hypoxia, dramatically increased cAMP levels in virtually all cells. Moderate levels of cGMP immunoreactivity in control carotid bodies in vitro were strikingly reduced by hypoxia; a significant increase in cGMP levels occurred following incubation in high potassium (100 mM), and under these conditions, the decrease in cGMP immunoreactivity with hypoxia was much more pronounced.(ABSTRACT TRUNCATED AT 250 WORDS)