Anti-inflammatory effect by lentiviral-mediated overexpression of IL-10 or IL-1 receptor antagonist in rat glial cells and macrophages.

Neuroinflammation, as defined by activation of local glial cells and production of various inflammatory mediators, is an important feature of many neurological disorders. Expression of pro-inflammatory mediators produced by glial cells in the central nervous system (CNS) is considered to contribute to the neuropathology observed in those diseases. To diminish the production or action of pro-inflammatory mediators, we have used lentiviral (LV) vector-mediated encoding rat interleukin-10 (rIL-10) or rat interleukin-1 receptor antagonist (rIL-1ra) to direct the local, long-term expression of these anti-inflammatory cytokines in the CNS. We have shown that cultured macrophages or astroglia transduced with LV-rIL-10 or LV-rIL-1ra produced far less tumor necrosis factor (TNF)alpha or IL-6, respectively in response to pro-inflammatory stimuli. Moreover, intracerebroventricular (i.c.v.) administration of LV-rIL-10 or LV-rIL-1ra resulted in transduction of glial cells and macrophages and, subsequently reduced TNFalpha, IL-6 and inducible nitric oxide synthase (iNOS) expression in various brain regions induced by inflammatory stimuli, whereas peripheral expression of these mediators remained unaffected. In addition, expression levels of the anti-inflammatory cytokines IL-4 and transforming growth factor-beta were not altered in either brain or pituitary gland. Furthermore, i.c.v. administration of LV-rIL-10 or LV-rIL-1ra given during the remission phase of chronic-relapsing experimental autoimmune encephalomyelitis, an animal model of multiple sclerosis, improved the clinical outcome of the relapse phase. Thus, local application of LV vectors expressing anti-inflammatory cytokines could be of therapeutic interest to counteract pro-inflammatory processes in the brain without interfering with the peripheral production of inflammatory mediators.

Single administration of interleukin-1 increased corticotropin releasing hormone and corticotropin releasing hormone-receptor mRNA in the hypothalamic paraventricular nucleus which paralleled long-lasting (weeks) sensitization to emotional stressors.

Single exposure to the proinflammatory cytokine interleukin-1 induces sensitization of the adrenocorticotropin hormone and corticosterone responses to stressors weeks later (hypothalamus-pituitary-adrenal sensitization). Hypothalamus-pituitary-adrenal responses are controlled by corticotropin-releasing hormone and arginine-vasopressin secreted from parvocellular corticotropin-releasing hormone neurons of the hypothalamic paraventricular nucleus and may involve autoexcitatory feedback mechanisms. Therefore, we studied the temporal relationship between resting levels of corticotropin-releasing hormone, corticotropin-releasing hormone-R1 and arginine-vasopressin receptor (V1a, V1b) mRNAs in the paraventricular nucleus and the development of hypothalamus-pituitary-adrenal sensitization to an emotional stressor (novelty). The adrenocorticotropin hormone precursor molecule proopiomelanocortin hnRNA in the pituitary gland served as an index for acute activation. Single administration of interleukin-1 induced sensitization of the hypothalamus-pituitary-adrenal to novelty from 3 to 22 days later, but not after 42 days. Single administration of interleukin-1 induced biphasic increases in corticotropin-releasing hormone and corticotropin-releasing hormone-R1 mRNAs in the paraventricular nucleus: an early peak within 24 h, followed by a delayed (>7 days) increase that peaked after 22 days. Hypothalamic V1a and V1b mRNA levels were unaffected. In contrast, in the pituitary gland, there was an early decrease in corticotropin-releasing hormone-R1 mRNA (from 10.5 to 3 h after interleukin-1) and V1b receptor mRNA (3 to 6 h), which returned to control levels from 24 h onwards. Thus, interleukin-1-induced long-lasting hypothalamus-pituitary-adrenal sensitizations associated with prolonged activation of corticotropin-releasing hormone and corticotropin-releasing hormone-R1 mRNA expression in the paraventricular nucleus, but not with changes in the expression of proopiomelanocortin hnRNA or V1b receptor or corticotropin-releasing hormone R1 mRNAs in the pituitary gland. We propose that transient exposure to immune events can induce long-lasting hypothalamus-pituitary-adrenal sensitization, which at least in part involves long-term hypothalamic adaptations that enhance central corticotropin-releasing hormone signaling.

A single administration of interleukin-1 or amphetamine induces long-lasting increases in evoked noradrenaline release in the hypothalamus and sensitization of ACTH and corticosterone responses in rats.

Single administration of the cytokine interleukin-1beta (IL-1) or the psychostimulant amphetamine causes long-term sensitization of the hypothalamus pituitary adrenal (HPA) axis, i.e. enhanced adrenocorticotropine hormone (ACTH) and corticosterone responses weeks later. HPA responses to these stimuli involve activation of hypothalamic corticotropin-releasing hormone (CRH) neurons by noradrenergic projections to the paraventricular nucleus (PVN). In search of the underlying mechanisms, we studied the temporal pattern of HPA sensitization in relation to (1) the reactivity of noradrenergic projections to the PVN and (2) altered secretagogue production in hypothalamic CRH neurons. Single exposure to IL-1 or amphetamine induced cross-sensitization of ACTH and corticosterone responses 11 and 22 days later, but not after 42 days. Amphetamine-induced HPA sensitization was not accompanied by increased costorage of arginine vasopressin (AVP) in CRH terminals, as found previously after IL-1 pretreatment. The reactivity of noradrenergic terminals was assessed by measuring the electrically evoked release of [3H]-noradrenaline from superfused PVN slices. Single administration of amphetamine and IL-1 induced a long-lasting (up to 22 days) increase (up to 165%) of evoked noradrenaline release. This indicates that single exposure to psychostimulants or to cytokines can induce a long-lasting increase in stimulus-secretion coupling in brainstem noradrenergic neurons that innervate the PVN. This common, long-lasting functional change may underlie, at least in part, IL-1- and amphetamine-induced HPA cross-sensitization. In addition, increased AVP signalling by hypothalamic CRH neurons appears to play a role in IL-1-induced, but not in amphetamine-induced, HPA sensitization.

Priming with interleukin-1beta suppresses experimental allergic encephalomyelitis in the Lewis rat.

Lewis rats exhibit multiple defects in their hypothalamus-pituitary-adrenal (HPA) system that are considered to play a causal role in the susceptibility of this strain to autoimmune diseases, i.e. experimental allergic encephalomyelitis (EAE). In the present study, we aimed to modulate the HPA response of the Lewis rat and establish its consequences for the susceptibility to EAE. Because in Wistar rats, single administration of interleukin (IL)-beta (priming) is known to induce long-lasting (weeks) sensitization of HPA responses to stressors and immune stimuli, Lewis rats were given a single dose of hIL-1beta or vehicle 1 week prior to induction of EAE by immunization with myelin basic protein (MBP). Subsequently, neurological deficits were monitored once daily. The results show that IL-1 priming markedly suppresses the neurological symptoms of EAE, without affecting the onset or duration of the disease. Measurement of vasopressin and corticotropin releasing hormone (CRH) in the external zone of the median eminence revealed that, as compared to Wistar rats, Lewis rats exhibit low vasopressin but identical CRH, and that IL-1 priming increases (0.001) vasopressin without affecting CRH stores, which is consistent with a shift to vasopressin-dominated control of adrenocorticotropic hormone (ACTH) secretion as described in Wistar rats under conditions of HPA hyper(re)activity. However, IL-1 priming did not affect a.m. corticosterone levels following immunization with MBP or during the clinical phase of EAE. IL-1 priming of Lewis rats attenuated the ACTH responses to an IL-1 challenge 11 days later, which may relate to an increase in resting corticosterone levels. Thus, the mechanisms underlying IL-1 induced suppression of EAE are not related to enhanced HPA responses. In addition, we did not find IL-1 priming-induced alterations in MBP-specific immunoglobulin (Ig)M, IgG1, IgGa and IgGb plasma titres, or gross alterations in T cell activation as reflected in spontaneous or concanavalin-induced T cell proliferation. We therefore speculate that IL-1-induced elevation of resting corticosterone levels may influence the development of EAE.

Stressor- or drug-induced sensitization of the corticosterone response is not critically involved in the long-term expression of behavioural sensitization to amphetamine.

Repeated exposure to drugs of abuse induces long-lasting behavioural sensitization, which is thought to play a role in the persistence of drug-seeking behaviour. Recently, we showed that repeated exposure of rats to cocaine resulted in a long-lasting (weeks) sensitization of the hypothalamus-pituitary-adrenal axis, i.e. hypersecretion of adrenocorticotropic hormone and of the glucocorticoid corticosterone. Moreover, we found that the administration of a glucocorticoid receptor antagonist abolished the expression of psychostimulant-induced behavioural sensitization. In the present study we tested whether stressor- or drug-induced long-term hypersecretion of corticosterone is associated with the long-term expression of behavioural sensitization to psychostimulant drugs. To that end, groups of male Wistar rats were exposed once to interleukin-1beta or to footshocks, treatments that are known to induce long-term sensitization of the hypothalamus-pituitary-adrenal axis, or were treated with amphetamine or morphine, according to protocols known to induce long-lasting behavioural (locomotor) sensitization. Three weeks later, the groups and their controls were challenged with amphetamine or vehicle. Previous exposure to interleukin-1beta or footshocks enhanced adrenocorticotropic hormone and corticosterone responses, but did not affect the long-term locomotor sensitization to amphetamine. Prior amphetamine treatment enhanced the locomotor response and the adrenocorticotropic hormone and corticosterone responses to amphetamine. Prior morphine treatment resulted in long-term locomotor sensitization, whereas the adrenocorticotropic hormone and corticosterone responses to amphetamine were decreased. From these findings and the absence of within-group correlation between corticosterone and locomotor responses in interleukin-1beta and morphine-pretreated rats, we conclude that there is no correlation between sensitization of the corticosterone response and behavioural sensitization to amphetamine. Apparently, sensitization of the corticosterone response is not a prerequisite for the long-term expression of behavioural sensitization, which suggests that drug-induced long-term behavioural sensitization may involve corticosteroid receptor-dependent (central) mechanisms that occur independent of hypothalamus-pituitary-adrenal axis responsiveness.

Dopaminergic mechanisms mediating the incentive to seek cocaine and heroin following long-term withdrawal of IV drug self-administration.

RATIONALE: The neurobiological mechanisms underlying the persistence of drug craving in detoxified addicts are still poorly understood. OBJECTIVE: The present study was designed to evaluate dopaminergic mechanisms in drug-seeking behaviour following long-term (>3 weeks) extinction of IV drug self-administration in rats. METHODS: To that end, we studied the effects of direct and indirect dopamine (DA) agonists on reinstatement of previously extinguished responding for heroin (50 microg/kg per injection; 14-15 daily 3-h sessions) and cocaine (500 microg/kg per injection; 10-11 daily 2-h sessions). RESULTS: In animals with a cocaine history, priming with cocaine, the selective DA reuptake inhibitor GBR-12909 and the DA D2 receptor agonist quinpirole resulted in robust and selective reinstatement of non-reinforced nose poking behaviour in the previously drug-paired hole. In contrast, the D1 agonist SKF-82958 failed to reinstate responding and the non-selective DA agonist apomorphine even suppressed responding in these animals. In heroin-trained rats, heroin and GBR-12909 strongly reinstated responding, whereas all direct DA agonists were ineffective. Again, the two highest doses of apomorphine decreased responding in these animals. In a parallel study, the ability of DA ligands to express behavioural sensitization in animals pretreated with amphetamine or morphine was evaluated. Interestingly, all agonists that reinstated responding in the present study caused expression of locomotor sensitization and vice versa. CONCLUSIONS: The differences between direct and indirect agonists indicate a clear, but complex, involvement of DA in drug-seeking behaviour long after detoxification. Moreover, the results show an important role of D2 receptor activation in the persistence of cocaine- but not heroin-seeking behaviour. Finally, the results from both studies suggest a relationship between drug-induced reinstatement and drug hyperresponsiveness in long-term abstinent rats.

Drug-induced reinstatement of heroin- and cocaine-seeking behaviour following long-term extinction is associated with expression of behavioural sensitization.

The present study was designed to evaluate the relationship between reinstatement of drug-seeking behaviour following long-term extinction of intravenous (i.v.) drug self-administration (an animal model for craving) and long-term behavioural sensitization. Rats were allowed to self-administer heroin (50 microg/kg per inj., 14 daily sessions), cocaine (500 microg/kg per inj., 10 daily sessions) or saline. Following a 3-week extinction period, reinstatement tests were performed to evaluate priming effects of amphetamine, cocaine and heroin on nonreinforced drug-seeking behaviour. In addition, the occurrence of long-term behavioural sensitization in rats with a history of heroin or cocaine self-administration was determined. Heroin-seeking behaviour was reinstated by heroin (0.25 mg/kg), amphetamine (1.0 mg/kg) and cocaine (10 mg/kg). In addition, animals with a history of heroin self-administration displayed locomotor sensitization to both heroin and amphetamine. Cocaine-seeking behaviour was reinstated by cocaine and amphetamine, but not by heroin. Interestingly, locomotor sensitization to amphetamine, but not heroin, was observed in animals with a history of cocaine self-administration. In other words, the induction of drug-seeking behaviour following a prolonged drug-free period was found to be associated with the expression of long-term behavioural sensitization. These data provide experimental evidence for a role of behavioural sensitization in the incentive motivation underlying drug-seeking behaviour. If drug hyperresponsiveness would indeed be a crucial factor in drug-induced craving in human addicts, pharmacological readjustment of the neuroadaptations underlying drug sensitization may prevent relapse to drug use long after detoxification.

MK-801 reinstates drug-seeking behaviour in cocaine-trained rats.

We evaluated the incentive motivational properties of MK-801 by determining its priming effect on drug-seeking behaviour following extinction of cocaine self-administration, an animal model for drug craving. Rats were allowed to self-administrate cocaine (0.5 mg/kg) or saline during 10 daily sessions. MK-801 (0.1-0.25 mg/kg, i.p.) dose-dependently reinstated responding for cocaine following an extinction period of 3 weeks. Responding was selectively enhanced in the previously drug-paired hole and was completely absent in rats with a history of saline self-administration. These data provide evidence for a possible role of NMDA receptors in the incentive motivation underlying cocaine-seeking behaviour. In addition, the ability of MK-801 to elicit drug-seeking behaviour may prove to be a serious drawback for the proposed use of NMDA antagonists in the treatment of drug addiction.

Transient suppression of resting corticosterone levels induces sustained increase of AVP stores in hypothalamic CRH-neurons of rats.

Previous studies showed that various stressors can induce delayed (days) and long-lasting (weeks) increases of vasopressin (AVP) stores in terminals of CRH neurons in the external zone of the median eminence (ZEME) in adult rats. Here we tested whether this long-lasting neuroplastic change can be induced by mechanisms other than stressor provoked transsynaptic activation of CRH neurons. Single i.v. administration of a CRH antibody to adult rats causes a delayed (at least 1 day) and long-lasting (3 weeks) increase (2-3 fold) of AVP stores in the ZEME without affecting CRH stores. It suppresses ether-induced ACTH-responses for at least 8 days. In contrast, resting pm levels of ACTH and corticosterone (CORT) were suppressed only during the first 2 days. Suppletion of CORT levels on day 1 and 2, attenuates the antibody induced AVP-increase by 57%. CRH-immunoneutralization did not affect the AVP stores in CORT supplemented ADX rats. Thus, long-term increases of AVP stores induced by CRH-immunoneutralization largely depend on short-term suppression of pm CORT levels. Accordingly, single administration of metyrapone, which causes a transient suppression of pm CORT levels, increases AVP (1.5 fold) but not CRH stores one week later. We conclude that transient activation of hypothalamic CRH neurons results in long-lasting increases in AVP co-expression irrespective of the nature of the activating stimulus.

Short stressor induced long-lasting increases of vasopressin stores in hypothalamic corticotropin-releasing hormone (CRH) neurons in adult rats.

Recently, we demonstrated that single administration of interleukin-1 beta (IL-1) to adult rats induces a long-lasting (weeks) increase of vasopressin (AVP) stores in terminals of CRH neurons in the external zone of the median eminence (ZEME). This is accompanied by hypersecretion of AVP into the pituitary portal circulation and long-lasting hyperresponsiveness of the hypothalamo-pituitary-adrenal (HPA) axis to stressors. Here, we determine whether this form of plasticity of hypothalamic CRH neurons is specific for IL-1 or represents a general response to a stressor. Single exposure of rats to lipopolysaccharide (LPS), IL-1, brain surgery or electric footshocks increases the AVP stores in the ZEME 7 and 11 days later. Exposure to insulin or ether does not affect the AVP stores. The stressors have little or no effect on the CRH stores in the ZEME. The amplitude of the increase in AVP as measured 7-11 days after stimulation correlates with the overall ACTH response to the stressor (area under curve, r = 0.89, P < 0.0001), with the peak ACTH levels (r = 0.52, P < 0.05), but not with the duration of the ACTH response nor with any parameter of the corticosterone response. Administration of ACTH or corticosterone at doses that mimic stress-induced plasma levels does not increase AVP stores 7 days later. We conclude that long-lasting increases of AVP stores in CRH terminals in the ZEME can be induced by various stressors and postulate that the amplitude of such increases depends on the degree of activation of the CRH neurons by the stressor. (NWO grant: 900-543-101.)

Diurnal variation in resting levels of corticosterone is not mediated by variation in adrenal responsiveness to adrenocorticotropin but involves splanchnic nerve integrity.

To study possible mechanisms controlling diurnal changes in corticosterone (CORT) levels, we tested the CORT responses to ACTH in the morning (AM) and evening (PM) in male Wistar and Sprague-Dawley rats. Rat ACTH-(1-39) or human ACTH-(1-24) (3.75-15 ng/rat) was given as an iv bolus or an intraarterial infusion to (un)anesthetized rats treated with dexamethasone (0.1-0.5 mg/kg, 2-6 h before ACTH). In all conditions studied, no AM/PM differences in CORT responses were found when ACTH was given in vehicle (pH 4.3-7). This contrasts with earlier studies in which ACTH was given in a strongly acid vehicle (pH 1-1.9). Administration of ACTH in such acid vehicle confirmed the reported AM/PM differences in CORT responses (P < 0.05). Because alterations in splanchnic nerve activity can modulate ACTH-induced CORT secretion, we studied the effect of splanchnic nerve transection (SPLNX) on the diurnal change in resting CORT levels in unilaterally adrenalectomized rats. SPLNX reduced resting CORT concentrations in the PM (approximately 50%; P < 0.05), but not in the AM. SPLNX did not abolish the acid vehicle-associated AM/PM differences in CORT responses to ACTH. We conclude that 1) diurnal variation in adrenal responsiveness to ACTH per se does not exist in this rat model; 2) the strongly acid vehicle interacts with ACTH-induced CORT secretion; 3) the PM rise in plasma CORT depends on the integrity of the sympathetic neural input to the adrenal gland.

Intermittent cocaine exposure causes delayed and long-lasting sensitization of cocaine-induced ACTH secretion in rats.

In view of the possible role of the hypothalamus-pituitary-adrenal axis in the long-term effects of drugs of abuse, we studied the response of the hypothalamus-pituitary-adrenal axis to cocaine challenges 3 and 14 days after cocaine withdrawal. Three days after intermittent cocaine exposure, the cocaine-induced increase of plasma adrenocorticotropic hormone (ACTH) is unchanged, whereas after 14 days the ACTH response is enhanced 2-fold. The cocaine-induced increase of plasma corticosterone is enhanced approximately 1.5-fold both 3 and 14 days after cocaine withdrawal. Apparently, prior cocaine treatment causes a delayed sensitization of cocaine-induced ACTH secretion and long-lasting corticosterone hyper-responsiveness. We propose that the long-lasting changes in the hypothalamus-pituitary-adrenal axis may facilitate drug-induced long-term behavioral sensitization.

Effects of monoclonal antibodies to specific epitopes of rat interleukin-1 beta (IL-1 beta) on IL-1 beta-induced ACTH, corticosterone and IL-6 responses in rats.

Recently, we developed a panel of monoclonal antibodies (MoAbs) to rat IL-1 beta and found that MoAbs binding to the aminoacid sequences 66-85 and 123-143 of mature rIL-1 beta inhibited the binding of rIL-1 beta to murine EL4 cells. Here we study whether MoAbs to these and other domains of IL-1 interfere with the biological effects of rIL-1 beta in adult male rats in vivo. Administration of rIL-1 beta (1 or 5 micrograms/kg i.v.) enhanced the plasma concentrations of ACTH, corticosterone (CORT) and of IL-6 in a time- (0.5-4 h) and dose-dependent manner. Because 2 h after 5 micrograms/kg i.v., all three parameters were consistently elevated, this dose and time interval was used for further studies. Prior to injection, rIL-1 beta was incubated alone or in the presence of a MoAb (10 mg/kg) for 30 min at 37 degrees C or at 4 degrees C. Plasma ACTH, CORT and IL-6 responses to these mixtures are compared to those obtained after preincubation of rIL-1 beta with a non-IL-1 binding MoAb (PEN7). SILK 3, a MoAb that binds to the 66-85 domain of rIL-1 beta, reduced the ACTH and IL-6 responses by 48 and 45% respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Chronic stress enhances vasopressin but not corticotropin-releasing factor secretion during hypoglycemia.

Quantitative immunocytochemistry was used to investigate the effect of chronic intermittent stress on the storage and release of corticotropin-releasing factor (CRF) and arginine vasopressin (AVP) in the zona externa of the median eminence (ZEME). Wistar rats were subjected to repeated once daily insulin stress and studied 24 h after the last episode of hypoglycemia. Chronic intermittent hypoglycemia differentially affected CRF immuno-staining (CRFi) and AVPi stores in the ZEME, characterized by a transitory fall in CRFi (7 days) and a gradual increase in AVPi (11 days). The acute hypothalamic response to stress was studied by measuring insulin-induced depletion of CRFi and AVPi in the ZEME after blockade of fast axonal transport with colchicine (5 micrograms/rat ic). Three hours after insulin, CRFi was reduced by 24 and 28% in naive and repeatedly (11 times) stressed animals, respectively. In contrast, insulin did not affect CRFi in rats exposed to hypoglycemia for 7 days. In addition, hypoglycemia reduced AVPi in the ZEME by 37, 42, and 54% in naive and repeatedly (7 and 11 times) stressed rats, respectively. We conclude that chronic intermittent hypoglycemia induces a shift in hypothalamic signals for adrenocorticotropic hormone release in favor of AVP. This may result in altered sensitivity of the corticotrophs for hypophysiotropic factors and corticosteroids that are of physiological importance for the pituitary response to subsequent stressors.

Physiological role of corticotropin-releasing factor in the control of adrenocorticotropin-mediated corticosterone release from the rat adrenal gland.

Marked fluctuations in adrenal sensitivity to ACTH have been reported under both physiological (e.g. diurnal) and experimental conditions. Recently, we reported that immunoneutralization of CRF reduces resting corticosterone (cort) levels in rats without inducing concomitant reductions in plasma ACTH. We postulated an endogenous CRF mechanism that controls the adrenal sensitivity to ACTH. In the present study, this hypothesis was tested by iv infusion of human ACTH (0, 1, 3, and 10 ng/kg.min for 60 min) into dexamethasone-treated anaesthetized male Wistar rats. Serial blood samples were taken for the determination of ACTH and corticosterone by RIA (ACTHi, corti). Infusion of ACTH resulted in dose-dependent steady state plasma ACTHi levels, ranging from 50-600 pg/ml, which were not affected by prior administration of a rat monoclonal antibody to rat CRF (PFU 83). As expected, infusion of ACTH resulted in a dose-dependent increase in plasma corti. In PFU 83-treated rats, preinfusion plasma corti levels were reduced compared to those of rat immunoglobulin G-treated controls (7.8 +/- 1.2 vs 25.3 +/- 3.2 ng/ml). In addition, the corti responses to infusion of 1 and 3 ng/kg.min ACTH were suppressed by PFU 83. However, at a (near) maximally effective dose of ACTH (10 ng/kg.min), no differences in plasma corti were found between PFU 83 and immunoglobulin G-treated rats. These findings suggest that immunoneutralization of endogenous CRF results in a 3-fold reduction of the adrenal sensitivity to ACTH. Subsequently, we studied the possible effects of exogenous CRF on the isolated perfused adrenal gland in situ. In this preparation, CRF alone (1-100 pmol) or ACTH alone (5 fmol) did not affect the corti secretion rate or the flow rate of the perfusion medium through the gland. However, when given together a marked (up to 3.2 times) CRF dose-dependent stimulation of corti secretion and an increase (up to 1.7 times) in adrenal flow rate were obtained. In experiments with freshly dispersed adrenal cells in vitro, PFU 83 (1 microM) or CRF (0.1-10 nM) did not influence corti secretion when given alone and did not affect ACTH-induced corti secretion. It is unlikely, therefore, that CRF acts directly on the steroid-producing cells.(ABSTRACT TRUNCATED AT 400 WORDS)