Isoproterenol-induced Cyclic AMP Accumulation Research Articles

Interferon beta-1a counteracts effects of activation on the expression of G-protein-coupled receptor kinases 2 and 3, β-arrestin-1, and regulators of G-protein signalling 2 and 16 in human mononuclear leukocytes

Activation regulates the responsiveness of G-protein-coupled receptors (GPCRs) on T cells, and modifications in the activity of GPCRs characterize lymphocytes from some immune disorders such as multiple sclerosis (MS) and rheumatoid arthritis (RA). Some lines of evidence suggest that such an effect is connected with the altered expression of some GPCRs regulatory proteins. Herein we demonstrate that phitoemagglutinin (PHA)-induced activation leads to differential expression of G-protein-coupled receptor kinase (GRK) 2, GRK3, β-arrestin-1, regulators of G-protein signalling (RGS) 2, and RGS16 and decreases responsiveness of mononuclear leukocytes (MNL) to the β-adrenergic agonist isoproterenol. Interferon beta-1a (IFNβ-1a), which is known to ameliorate the course of MS, counteracts the activation-induced effects on the expression of these GPCR regulatory proteins in MNL. Furthermore, IFNβ-1a quenches the effects of PHA on the isoproterenol-induced accumulation of cyclic AMP (cAMP). We suggest that regulation of GPCRs responsiveness may be a relevant property of IFNβ-1a in MS.

Two subtypes of G protein‐coupled nucleotide receptors, P2Y1 and P2Y2 are involved in calcium signalling in glioma C6 cells

1. In glioma C6 cells, the stimulation of P2Y receptors by ADP, ATP and UTP initiated an increase in the intracellular Ca2+ concentration, in a process that involved the release of Ca2+ from InsP(3)-sensitive store and the capacitative, extracellular Ca2+ entry. The presence of external Ca2+ was not necessary to elevate Ca(2+). 2. The rank order of potencies of nucleotide analogues in stimulating [Ca2+](i) was: 2MeSADP > ADP > 2MeSATP = 2ClATP > ATP > UTP. alpha,beta-Methylene ATP, adenosine and AMP were ineffective. 3. ADP and UTP effects were additive, while actions of ATP and UTP were not additive on [Ca2+](i) increase. Similarly, cross-desensitization between ATP and UTP but not between ADP and UTP occurred. 4. Suramin, a non-specific nucleotide receptors inhibitor, antagonized ATP-, UTP- and ADP-evoked Ca2+ responses. PPADS, a selective antagonist of the P2Y(1) receptor-generated InsP(3) accumulation, decreased ADP-initiated Ca2+ response with no effect on ATP and UTP. 5. Pertussis toxin (PTX) reduced ADP- and ATP-induced Ca2+ increases. Short-term treatment with TPA, inhibited both ATP and ADP stimulatory effects on [Ca2+](i). 6. ADP inhibited isoproterenol-induced cyclic AMP accumulation. PTX blocked this effect, but PPADS did not. 7. RT - PCR analysis revealed the molecular identity of P2Y receptors expressed by glioma C6 cells to be both P2Y(1) and P2Y(2). 8. It is concluded that both P2Y(1) and P2Y(2) receptors co-exist in glioma C6 cells. ADP acts as agonist of the first, and ATP and UTP of the second one. Both receptors are linked to phospholipase C (PLC).

Anandamide and WIN 55212-2 inhibit cyclic AMP formation through G-protein-coupled receptors distinct from CB1 cannabinoid receptors in cultured astrocytes.

The effects of anandamide and the cannabinoid receptor agonists WIN 55212-2 and CP 55940 on the evoked formation of cyclic AMP were compared in cultured neurons and astrocytes from the cerebral cortex and striatum of mouse embryos. The three compounds inhibited the isoproterenol-induced accumulation of cyclic AMP in neuronal cells, and these responses were blocked by the selective CB1 receptor antagonist SR 141716A. The three agonists were more potent in cortical than striatal neurons. Interestingly, WIN 55212-2, CP 55940 and anandamide also inhibited the isoproterenol-evoked accumulation of cyclic AMP in astrocytes but, in contrast to WIN 55212-2 and CP 55940, anandamide was much more potent in striatal than cortical astrocytes. Inhibition was prevented by pertussis toxin pretreatment, but not blocked by SR 141716A. Therefore, G-protein-coupled receptors, distinct from CB1 receptors, are involved in these astrocytic responses. Moreover, specific binding sites for [3H]-SR 141716A were found in neurons but not astrocytes. Furthermore, using a polyclonal CB1 receptor antibody, staining was observed in striatal and cortical neurons, but not in striatal and cortical astrocytes. Taken together, these results suggest that glial cells possess G-protein-coupled receptors activated by cannabinoids distinct from the neuronal CB1 receptor, and that glial cells responses must be taken into account when assessing central effects of cannabinoids.

Altered Expression of Cyclic Nucleotide Phosphodiesterase Isozymes during Culture of Aortic Endothelial Cells

Primary cultures of bovine aortic endothelial cells (BAEC) express cyclic nucleotide phosphodiesterase (CN PDE) isozymes of the PDE2, PDE4 and PDE5 gene families. We report here that the isozyme profiles of CN PDE and the amounts of each vary with the passage number of BAEC cultures. Characterization by anion-exchange chromatography and pharmacological criteria were used to study CN PDE in early (4–6), intermediate (6–10), and late (>17) passages of purified BAEC. PDE2 and a minor fraction of PDE5 accounted for cyclic GMP hydrolysis in early passages, but both isozymes were lost with cell passage. Cyclic AMP was hydrolyzed by both PDE2 and PDE4 isozymes in early passage endothelial cells, but PDE4 was increased dramatically in higher passage cells. Also appearing in the higher passage cells were prominent PDE1 and minor PDE3 activities. The ratios of cytosolic to particulate activities were similar at all passages. BAEC PDE isoforms in intact cells assessed by [ 3H]-adenine prelabeling showed that atriopeptin II decreased isoproterenol-induced cyclic AMP accumulation in early but not later passage cells, consistent with the loss of PDE2 expression. Enhancement of isoproterenol-induced cyclic AMP accumulation by rolipram, a PDE4 inhibitor, was also greatly diminished during culture passages. Changes in CN PDE isoform expression and consequent cyclic AMP turnover validate the importance of considering cell passage number when cultures of BAEC are used to study the regulation of endothelial cell cyclic nucleotide metabolism and processes mediated by cyclic nucleotides in this model system.

Modulation of airway smooth muscle β-adrenoceptor function by a muscarinic agonist

We have investigated the effect of a muscarinic agonist and protein kinase C (PKC) activation on β-adrenoceptors and their coupling to adenylyl cyclase in bovine tracheal smooth muscle. There was a significant reduction in maximum binding capacity of [ 125I]iodocyanopindolol ([ 125I]ICYP) after exposure to carbachol and 4β-phorbol 12β-myristate 13α-acetate (PMA). Similarly both carbachol and PMA inhibited the 5′-guanylylimidodiphosphate-induced shift in [ 125I]ICYP binding by isoproterenol and significantly decreased isoproterenol-induced cyclic AMP accumulation. A phorbol ester, 4α-phorbol 12,13-didecanoate which does not activate PKC had no effect on β-receptor binding or coupling. These results suggest that PKC activation directly via a phorbol ester and indirectly via muscarinic receptor stimulation may lead to reduction and uncoupling of β-receptors in airway smooth muscle. We suggest that this mechanism may be relevant to the reduction in β-receptor coupling in asthmatic airways as an effect of PKC activation by inflammatory mediators and neurotransmitters.

Effects of DS-4574, a New Orally Active Antiallergic Agent, in Experimental Allergic and Inflammatory Models

To determine the effects of the Kampo drugs Saiboku-to, Bakumondo-to and Orengedoku-to on beta-adrenergic function in airway smooth muscle, we studied isolated canine bronchial segments under isometric conditions in vitro. Incubation of tissues with these drugs did not alter the contractile responses to acetylcholine and histamine. The relaxation of tissues precontracted with acetylcholine induced by isoproterenol was augmented by Saiboku-to and Bakumondo-to, so that the concentration of isoproterenol required to produce a half-maximal effect (IC50) was decreased from 4.6 +/- 0.5 x 10(-8) M to 1.9 +/- 1.0 x 10(-8) M (P < 0.05) and to 1.0 +/- 0.8 x 10(-8) M (P < 0.01), respectively, whereas Orengedoku-to had no effect. These effects were concentration-dependent with the threshold concentrations being 0.3 mg/ml for Saiboku-to and 0.1 mg/ml for Bakumondo-to. Saiboku-to and Bakumondo-to did not affect cyclic AMP levels in airway smooth muscle per se but potentiated the isoproterenol-induced cyclic AMP accumulation. These results suggest that Saiboku-to and Bakumondo-to but not Orengedoku-to potentiate beta-adrenergic function in airway smooth muscle, which may reflect the efficacy of these drugs on airway hyperresponsiveness and asthma.

Idazoxan down-regulates β-adrenoceptors on C6 glioma cells in vitro

Incubation of the C6 cells with 10 μM idazoxan (an α 2-adrenoceptor antagonist and putative antideprssant for 5 days in vitro resulted in a 23% reduction of β-adrenoceptor number and a 37% decrease in isoproterenol-induced cyclic AMP accumulation. In contrast, post-receptor stimulated cyclic AMP accumulation (by the use of forskolin or cholera toxin) was unaffected. The desensitization of the β-adrenoceptor was accompanied by an increase in the K L/K H ratio for this receptor. Chronic in vitro treatment of C6 glioma cells with idazoxan did not significantly affect cholera or pertussis toxin catalyzed ribosylation of Gs and Gi/Go in these cells. Similarly, idazoxan did not alter either the basal levels of protein kinase C (PKC) α, or its cytoplasm to membrane translocation. These results suggest that idazoxan may have direct postsynaptic effects, the site of which may be at the level of receptor/G protein interaction.

Influence of ß-Adrenergic Receptor Function during Terbutaline Treatment on Allergen Sensitivity and Bronchodilator Response to Terbutaline in Asthmatic Subjects

Nine asthmatic patients with an allergy to birch or timothy underwent bronchial allergen provocations on three different trial days, with intervals of 2 to 5 wk. Two weeks prior to one of the provocations, no medication was allowed. Before the other two provocations the patients had been on continuous treatment with oral terbutaline (7.5-mg slow-release pill bid) for 2 wk, which was discontinued 12 or 48 h before the allergen provocation. After allergen challenges, terbutaline was inhaled in increasing doses (0.5 mg, 1.0 mg, and 2.0 mg), and pulmonary function was measured after each dose. Before each allergen provocation, blood samples were drawn for measurements of catecholamine and terbutaline concentrations and for in vitro measurements of beta-adrenergic receptor function on lymphocytes (isoproterenol-induced accumulation of cyclic AMP). Beta-adrenergic receptor function on blood lymphocytes was impaired after the two treatment periods, compared with the drug-free period, and was significantly more depressed at 12 h than 48 h after dosing. The bronchial responsiveness to allergen, defined as PC20PEF (median values), was 1,700 biologic units (BU) after the period of no treatment and 220 BU and 445 BU at 12 and 48 h after discontinuation of the terbutaline treatment (p less than 0.1 after 48 h). Five of the nine patients exhibited increased bronchial responsiveness 48 h after treatment, compared to results without treatment. The responsiveness was similar on all occasions in three patients. The bronchodilator response to inhaled terbutaline after allergen-induced bronchoconstriction was attenuated (p less than 0.01) at both 12 and 48 h after terbutaline, compared to results without treatment, indicating desensitization also of the bronchial beta-adrenergic receptors. We conclude that the early bronchial responsiveness to allergen is increased following a period of continuous treatment with a beta-adrenergic receptor agonist in some asthmatic patients and that the capability of a beta-agonist to reverse allergen-induced bronchoconstriction is attenuated after beta-agonist treatment.

Serotonin enhances the β-adrenergic response in rat brain cortical slices

Serotonin increased the isoproterenol-induced cyclic AMP accumulation in rat brain cortical slices but did not itself stimulate the production of cyclic AMP in this area. This effect appeared to be inhibited by the 5-HT 2 receptor antagonist, ritanserin. This potentiation was not observed in the hippocampus. These data reinforce the hypothesis of a coupling between the noradrenergic and serotonergic systems, which may be involved in the mechanism of action of antidepressant drugs.

Differential effect of pertussis toxin on adenosine and muscarinic inhibition of cyclic AMP accumulation in canine ventricular myocytes

Cyclic AMP regulation by muscarinic and adenosine receptors was investigated in isolated canine ventricular myocytes. Both the muscarinic receptor agonist, carbachol, and the adenosine receptor agonist, phenylisopropyladenosine, decreased isoproterenol-stimulated cyclic AMP accumulation in a concentration-dependent manner. Carbachol was more potent than phenylisopropyladenosine and had a greater inhibitory effect. At 10 −6 m, carbachol reduced isoproterenol-stimulated cyclic AMP by 73 ± 5% while 10 −3 m phenylisopropyladenosine was required to decrease cyclic AMP accumulation by 54 ± 8%. Pretreatment of myocytes with pertussis toxin to inactivate the inhibitory guanine nucleotide binding protein, G i, completely abolished the effect of phenylisopropyladenosine to reduce cyclic AMP stimulation. In comparison, pertussis toxin treatment blunted the response to carbachol and shifted the dose-effect curve to the right but did not eliminate the inhibitory action of carbachol. In toxin-treated myocytes, 10 −3 m carbachol produced a 26 ± 6% reduction of isoproterenol-induced cyclic AMP accumulation. This pertussis toxin-insensitive action of carbachol was antagonized by atropine and pirenzepine and was prevented when either of two different phosphodiesterase inhibitors, RO-20-1724 or isobutylmethylxanthine, was included in the incubation medium. The results indicate that adenosine receptor-mediated inhibition of hormone-stimulated cyclic AMP accumulation in ventricular myocytes occurs by a single, G i-dependent mechanism while muscarinic inhibition appears to involve both G i-dependent and G i-independent mechanisms. The G i-independent mechanism may reflect enhanced phosphodiesterase activity which results from the activation of muscarinic receptors.

Enhancement in β-Adrenergic Responsiveness of Adenylate Cyclase in Rat Liver after Galactosamine Administration

In rat liver at 48 hours after galactosamine (GalN) treatment, isoproterenol-induced accumulation of cyclic AMP and activation of adenylate cyclase (AC) were significantly enhanced. AC activity stimulated with GppNHp or NaF was also augmented in membranes from GalN-trcated rats. However, the MnC12-stimulated or forskolin-stimulated activity in GalN-treated rats did not differ from the control. These data suggest that the β-adrenergic responsiveness of AC in rat liver is potentiated, possibly due to the enhanced function of the Gs protein, during the process of regeneration after GalN treatment.

Enhancement in beta-adrenergic responsiveness of adenylate cyclase in rat liver after galactosamine administration.

In rat liver at 48 hours after galactosamine (GalN) treatment, isoproterenol-induced accumulation of cyclic AMP and activation of adenylate cyclase (AC) were significantly enhanced. AC activity stimulated with GppNHp or NaF was also augmented in membranes from GalN-treated rats. However, the MnCl2-stimulated or forskolin-stimulated activity in GalN-treated rats did not differ from the control. These data suggest that the beta-adrenergic responsiveness of AC in rat liver is potentiated, possibly due to the enhanced function of the Gs protein, during the process of regeneration after GalN treatment.

Involvement of protein kinase c in the mechanism of in vitro effects of imipramine on generation of second messengers by noradrenaline in cerebral cortical slices of the rat

Imipramine did not significantly inhibit the noradrenaline or isoproterenol-induced cyclic AMP accumulation in rat cerebral cortical slices, but inhibited the potentiation of this response by protein kinase C activator, a phorbol ester12- O-tetradecanoyl-phorbol 13-acetate. In low concentrations (0.1–1 μM) it prevented the inhibitory effect of the phorbol ester on accumulation of inositol phosphate induced by noradrenaline, while in higher concentrations it inhibited the response by itself. Imipramine did not bind to β-adrenoceptors but was an effective blocking agent of α 1-adrenoceptors ( K 1 = 38.1 nM). The data suggest that imipramine acts within the noradrenergic cyclic AMP generating system on two targets: inhibiting protein kinase C and blocking the α 1-adrenoceptor; both actions may reduce the α-adrenoceptor potentiation of β-adrenoceptor-mediated cyclic AMP generation.

Interactions between cyclic AMP and inositol phosphate transduction systems in astrocytes in primary culture

Astroglial cells in primary culture possess receptors with cyclic AMP and inositol phosphates (IP) as second messengers. The beta-receptor agonist, isoproterenol induces an increase in the accumulation of cyclic AMP, the alpha 2-receptor agonist clonidine inhibits the isoproterenol-induced accumulation of cyclic AMP, while the alpha 1-receptor agonist phenylephrine acts only on the inositol phosphate system. 5-Hydroxytryptamine (5-HT) stimulates, the formation of inositol phosphate, while isoproterenol and clonidine per se do not affect the inositol phosphate system. In the present paper the possibility of interactions between the cyclic AMP and the inositol phosphate transduction systems were investigated. In the presence of 10(-5) M 5-HT, in itself ineffective on the formation of cyclic AMP, isoproterenol stimulated the accumulation of cyclic AMP far more than in the absence of 5-HT. The potentiation was blocked by the 5-HT2 receptor antagonist ketanserin. On the other hand, there were no indications for a beta-receptor influence on the 5-HT-induced inositol phosphate formation. Stimulation of the alpha 2-receptor did not induce accumulation of inositol phosphate but significantly potentiated 5-HT2-receptor transduction, as measured by hydrolysis of phosphoinositide and formation of inositol phosphate. Stimulation by 5-HT also increased the formation of inositol phosphate after adrenergic stimulation and this effect was found to be synergistic at certain concentrations of adrenergic agonists. In addition, there was a statistically significant accumulation of cyclic AMP in the presence of both 5-HT and phenylephrine, none of which stimulated cyclic AMP alone. The results suggest specific interactions between the cyclic AMP and inositol phosphate systems on cultured astroglial cells.

Age-related differences in cyclic AMP metabolism and their consequences on relaxation induced by isoproterenol and phosphodiesterase inhibitors in rat isolated aorta

Experiments were designed to investigate the respective roles of adenylate cyclase and cyclic nucleotide phosphodiesterase in the alterations with age of cyclic AMP metabolism-dependent relaxation in rat aorta. Neither basal nor stimulated aortic adenylate cyclase activities from 7- and 18-week-old rats differed significantly. The maximal cyclic AMP accumulation induced by isoproterenol in the aorta decreased with the age of the rat as did the maximal relaxation produced by the β-adrenergic agonist in pre-contracted aorta strips. The age-related difference in cyclic AMP accumulation was abolished when the experiment was performed in the presence of a high concentration (500 μM) of the phosphodiesterase inhibitor, 3-isobutyl-1-methylxanthine (IBMX). In addition, four phosphodiesterase inhibitors, IBMX, rolipram, Ro 20-1724 and zaprinast relaxed aorta strips pre-contracted by serotonin with concentration-effect curves which were displaced to the right in the older rats, the maximal effects being unchanged. The results suggest that the previously reported increased phosphodiesterase activity could be responsible for (1) the diminution with age of isoproterenol-induced cyclic AMP accumulation and relaxation in rat aorta; (2) the age-related decrease in potency of phosphodiesterase inhibitors as relaxing agents.

Protein kinase A and/or C inhibitors potentiate isoproterenol-induced cyclic AMP accumulation in intact human lymphocytes

The objective of the present study was to investigate the roles of protein kinase A and/or C in agonist-induced beta adrenoceptor activation in intact human lymphocytes. Lymphocytes from healthy subjects were incubated with isoproterenol and phophodiesterase inhibitor (IBMX, 1.0 mM) after 20 minutes of preincubation with (or without) various compounds possessing protein kinase A and/or C inhibitory activities. These compounds included the relatively selective protein kinase C (PK-C) inhibitors (W-7, calmidazolium, polymyxin B, neomycin, tamoxifen and clomiphene), purified protein inhibitors of protein kinase A (PK-A) (obtained synthetically, or purified from bovine hearts and porcine hearts) and the two compounds (H-7, H-9), which have been found to inhibit both PK-A and PK-C. The results showed that all PK-C inhibitors alone decreased cellular basal cAMP levels while inhibitors of PK-A as well as both H-7 and H-9 increased basal cAMP levels in a dose dependent manner at certain concentrations. All inhibitors studied potentiated isoproterenol-induced cAMP accumulation. The protein kinase A and C inhibitor, H-7, also potentiated PGE 1 (but not forskolin)-induced cAMP accumulation. In contrast, the protein kinase C activator, PMA, inhibited isoproterenol- and PGE 1- (but not forskolin) induced cAMP accumulation. These data suggest that the potentiating effects of PK-A and/or C inhibitors may be related to the inhibition of PK-A and/or PK-C, both of which have been shown to be involved in beta 2 adrenoceptor desensitization and phosphorylation.

Interaction between catecholamines and vasoactive intestinal peptide in cultured astrocytes

Receptors for vasoactive intestinal peptide (VIP) were demonstrated with cyclic AMP as the second messenger on astroglial cells cultured from the cerebral cortex, striatum, hippocampus and brain stem of the newborn rat. Vasoactive intestinal peptide produced increased accumulations of cyclic AMP in nM and μM concentrations, the former more pronounced in the cultures of brain stem, while the latter was more pronounced in the other cultures studied, indicating regional differences in the activation of cyclic AMP of VIP receptors. Vasoactive intestinal peptide inhibited isoproterenol-induced accumulation of cyclic AMP dose-dependently, with some regional differences, suggesting interactions between the second messenger systems of β- and VIP receptors. On the other hand, there was no inhibition of the NA-induced accumulation of cyclic AMP in the presence of VIP, which is in agreement with an interaction between the second messenger systems for VIP and the α-adrenoceptor. The data support interactions between the second messenger systems for VIP and α- and β-adrenoceptors on cultured astrocytes. The functional implications are at present unknown, but it might be that the peptide can modulate the response of the second messenger to catecholamines on astroglial cells through intramembrane mechanisms.

Protein kinase C potentiates isoproterenol‐mediated cyclic AMP production without modifying the homologous desensitization process in J774 cells

The J774 murine macrophage cells possess a beta 2-adrenergic receptor coupled to adenylate cyclase, which can be regulated by homologous desensitization. Stimulation of protein kinase C by phorbol esters or oleoyl acetyl glycerol potentiates two-to-threefold the isoproterenol-induced cyclic AMP accumulation. These promoters act at a post-receptor level, since the number and affinity of the beta-adrenergic receptors, measured by use of the hydrophilic ligand [3H]CGP-12177, are not modified. In addition, the effect of cholera toxin is similarly increased and pretreatment of the cells with pertussis toxin prevents the action of phorbol esters. On the other hand, these promoters are ineffective on isoproterenol-induced desensitization and the rates of receptor segregation and recovery remain unchanged. Therefore, protein kinase C modulates the isoproterenol-stimulated adenylate cyclase, whereas it is inactive on the homologous desensitization process.

Beta-adrenoceptor function in white blood cells from newborn infants: no relation to plasma catecholamine levels.

The maturity of beta-adrenoceptors in newborn infants was studied in relation to the catecholamine surge during labor. Umbilical blood was collected at birth from 12 infants delivered vaginally and 13 infants delivered by elective cesarean section. Granulocytes and lymphocytes were isolated. Receptor numbers and binding affinity were determined in the granulocytes by incubation with 125I-iodohydroxybenzylpindolol. Receptor responsiveness was tested by assessing isoproterenol-induced cyclic AMP accumulation in lymphocytes. Significantly higher plasma noradrenaline, adrenaline, and dopamine concentrations were found in infants born vaginally (108; 8.9; 0.9 nmol/liter, liter, respectively, median values) as compared with those delivered by cesarean section (11.0; 2.4; 0.2 nmol/liter). No significant differences in beta-adrenoceptor binding sites (receptor number: 39.2 +/- 2.6 versus 44.7 +/- 5.9 fmol/mg protein and binding affinity: 66.6 +/- 7.8 versus 65.0 +/- 6.2 pM) or responsiveness (maximal isoprenaline induced cAMP formation 52.4 +/- 10.3 versus 40.6 +/- 8.9 pmol/10(6) cells) were found between the two groups of infants. Lymphocyte beta-adrenoceptor sensitivity was similar to that found in adults. The beta-adrenoceptors on whole blood cells seem to be mature at birth and have the same responsiveness as in adults. The higher catecholamine surge during vaginal delivery as compared to elective cesarean section does not seem to affect beta-adrenoceptor function. Our results do not support the idea that reduced beta-adrenoceptor function is the cause of the previously observed inappropriately small cardiovascular and metabolic responses to the exceptionally high plasma catecholamine concentrations at birth.

Role of adrenal hormones and prostaglandins in the control of mouse thymocytes lysis

The cytolytic actions of glucocorticoids and of agents increasing cyclic AMP were studied in vitro in thymocyte suspensions isolated from adrenalectomized or hydrocortisone-treated mice. Although considered as corticoresistant cells, the thymocytes isolated from hydrocortisone-treated mice were lysed to the same extent although more slowly in vitro by dexamethasone than whole thymocyte populations (i.e. corticosensitive cells). Moreover, these two cell populations were shown to contain comparable amounts of glucocorticoid receptors and to be almost equally sensitive to the metabolic effects of glucocorticoids when measured by inhibition of RNA and DNA synthesis. Studies performed with corticosensitive cells showed that prostaglandin E 2, isoproterenol and dibutyrilcyclic AMP were also able to induce cell lysis and that, isoproterenol and dexamethasone exerted additive cytolytic action in vitro. In vivo experiments showed also an additive effect of steroids and isoproterenol on thymus atrophy. In contrast, cells isolated from hydrocortisone-treated animals were not sensitive to the cytotoxic action of prostaglandin E 2, isoproterenol and dibutyril cyclic AMP. This difference between the two populations was not associated with any difference in the responsiveness of adenylate cyclase as determined following isoproterenol-induced accumulation of cyclic AMP. The cytolytic action of dexamethasone but also that of prostaglandin E 2 and isoproterenol, could be blocked in the presence of cycloheximide, an inhibitor of protein synthesis, thus suggesting that glucocorticoids and agents increasing cyclic AMP control the synthesis of some proteins involved in the triggering of cell lysis. Among the hypotheses proposed to explain the differences between in vitro and in vivo sensitivity of lymphoid cell to glucocorticoids, it was suggested that the drug may in vivo indirectly control the viability or the proliferation of thymocytes through the release of other mediators. We have shown that in vivo injection of hydrocortisone induces an accumulation of fatty acids in the whole thymus gland but not in the isolated thymocytes. Since exogenous fatty acids exert cytolytic actions on isolated thymocytes, we suggest that glucocorticoids may exert in vivo an indirect toxic action by promoting the release of fatty acids from adipose tissue or other sources.