Inhibition of pituitary-adrenal secretion by a corticotropin releasing hormone antagonist in humans

Abstract

Corticotropin releasing hormone (CRH) is the primary modulator of ACTH release from the pituitary, and a neuromodulator in limbic and autonomic brain regions. Dysfunction of CRH-mediated neurotransmission is emerging as a critical mechanism in several disorders. Therefore, modulation of CRH availability at receptor sites is a potentially powerful therapeutic tool. Inhibitory analogues of CRH have been tested in rodents and primates, but their safety and hormonal effects in humans are unknown. We administered a CRH-antagonist, α-helical-CRH-(9–41) to six individuals. Each received two intravenous infusions: 50 μg kg^–1 on day 1, and 100 μg kg^–1 on the following morning. These doses block both endocrine and central effects of CRH in experimental animals. ACTH, cortisol, electrolytes, glucose and autonomic parameters were monitored in comparison with control values. Infusion of CRH antagonist did not alter heart rate, blood pressure, temperature or plasma electrolytes and glucose. Pre-infusion plasma ACTH levels averaged 26.8 ± 6.7 pg ml^–1 on day 1, and 29.0 ± 5.8 pg ml^–1 on day 2. Post-infusion values were 11.8 ± 2 and 11.5 ± 2.4 pg ml^–1, significantly lower than pre-infusion levels. Plasma cortisol levels, which averaged 21.4 ± 4 μg dl^–1 on the first morning and 22.9 ± 4.2 on the second, also decreased significantly after CRH antagonist infusions (to 14.0 ± 2.9 μg dl^–1 on day 1, and 13.9 ± 3.0 μg dl^–1 on day 2). Hormonal changes were transient, and circadian rhythm was not affected. Though not measured formally, euphoria, anxiety or somnolence were not observed. In conclusion, CRH antagonist administration to adults reduces hormonal secretion by pituitary corticotrophs, with resulting decrease in plasma ACTH and cortisol.

Introduction

Corticotropin releasing hormone (CRH) is a neuropeptide with both neuroendocrine and neurotransmitter properties. The peptide, isolated originally from the hypothalamus¹ is the primary modulator of the release of ACTH from the pituitary in response to stress.^1–4 CRH functions as a neurotransmitter in a number of limbic and autonomic brain circuits.^5–9 Both the peptide and CRH receptors are widely but specifically distributed in the central nervous system (CNS).^2,10–14 For example, in the human brain, CRH is found in cortical interneurons of layers II and III while receptors predominate in layers I and IV.^15,16 Abnormal CRH levels in brain and spinal fluid, and of CRH receptors in the CNS, have been demonstrated in Alzheimer disease,¹⁶ and abnormal CRH activity has been implicated in anxiety, depression and epilepsy.^3,16–18

CRH neurons impinge on postsynaptic receptors.^11–14 Two members of the CRH receptor family are currently known, and consist of membrane-spanning G-protein-coupled molecules.^12–14 The first, CRF₁, is found in the CNS, immune cells and the pituitary, and is thus the candidate mediator of the endocrine effects of CRH. A second receptor, CRF₂, is found primarily in the CNS. Its unique distribution such as high levels in the ventral medial hypothalamus suggests a role in mediating the anorexic effects of CRH.¹⁴

A growing body of evidence (recently reviewed in Ref. 19) strongly implicates abnormal CRH-mediated neurotransmission in several prevalent human disorders. In the adult, ‘excessive’ CRH activity may play a causative role in anxiety and depression. During development, upregulation of CRH-mediated neuronal excitation may underlie some age-specific seizure disorders¹⁸ and non-organic failure to thrive. Therefore, modulation of CRH-receptor activation offers powerful diagnostic and potentially therapeutic tools for a number of cardinal human disorders.^16–19 Inhibitory peptide analogues of CRH have been developed and tested in rodents and non-human primates.^20–22

The goal of this study was to assess the safety and hormonal efficacy of the first and best characterized CRH-antagonist, alpha-helical CRH-(9–41), in normal human adults.

Material and methods

A phase I study was approved by the Food and Drug Administration (FDA, IND 45969), and by the institutional review board of Childrens Hospital Los Angeles (CHLA). Subjects were recruited via electronic mail advertisement, and consented to participate. Six individuals were enrolled, three men and three women, and are described in Table 1. Four were Caucasian, one African-American and one Asian. All were healthy, and on no anti-hypertensive or corticosteroid therapy (one woman was on contraceptives, a second took ibuprofen for premenstrual cramps).

Table 1.

Characteristics of study subjects

No.	Age /sex	Ethnicity	Medications	Preinfusion			During infusiona
				BPb	HR	Temp	BP	HR	Temp
1	30/F	Caucasian	None	107/58	52	36.5	97/67–108/68	50–60	36–37
2	29/M	Caucasian	None	138/77	70	37.4	113/65–135/82	60–81	37–37,4
3	24/M	Caucasian	None	144/64	62	36	102/53–122/74	50–61	36–36.8
4	37/F	Af/Amerc	Ibuprofen	121/76	7a	37.3	106/71–126/77	73–81	36.7–37.2
5	19/M	Caucasian	None	136/69	60	36	114/59–129/77	52–65	36–36.7
6	28/F	Japanese	BCPd	143/102	68	37,1	133/94–152/110	66–79	36.6–37.2

^aParameters were measured every 5–15 min; the range for each subject is provided.

^bBP = blood pressure; HR = heart rate; Temp = oral temperature (°C).

^cAf/Ainer = African-American.

^dBCP = birth control pill.

Subjects were admitted to the Clinical Research Center (GCRC) at 7 am on the first study day, and infusions were carried out starting at 7.30–8.30 on two consecutive days. Vital signs and baseline plasma samples were obtained supine or sitting. Blood pressure, heart rate and core temperature were monitored at the onset of the infusions, and at 15-min intervals until their completion. These vital signs were measured hourly for the following 6 h. Subjects were fasted on both mornings until the end of each infusion, and permitted to ambulate 6 h after termination of the infusion. Plasma samples were also drawn 12 h after the end of the first infusion (evening nadir of ACTH, cortisol), and in the afternoon of the second day. Subjects were released on completion of a neurological examination performed 6–8 h after termination of the second infusion. Two (fasting) plasma samples, 2 h apart, were drawn on a follow-up visit, at 7.30 and 9.30 am. These were analyzed for ACTH and cortisol, to control for circadian fluctuations of these hormones.

The CRH antagonist, alpha-helical CRH-(9–41), was manufactured according to Good Manufacturing Practice by Bachem (Torrance, CA, USA). A sterile solution of 50 microgram (μg) per ml was infused over 2 h, to provide 50 μg kg^–1 on the first study day, and 100 μg kg^–1 on the second study day.

Plasma hormone levels were analyzed using commercial radioimmunoassays (Endocrine Sciences, Calabasas Hills, CA, USA). Sensitivity of ACTH assay was 5 pg ml^–1, and that of the cortisol assay was 1 μg dl^–1. Several samples were repeated, and interassay variabili ty averaged 8%. The significance of differences between groups was determined using the two-tailed paired Wilcoxon signed rank test without assumptions regarding value distribution.

Results

Infusion of CRH antagonist was not associated with significant alterations of blood pressure, heart rate or temperature (Table 1). Plasma glucose and electrolytes were also not affected.

Plasma ACTH and cortisol levels for individual subjects at the onset and at the completion of each infusion are shown in Figure 1. Prior to the first infusion, ACTH levels averaged 26.8 ± 6.7 pg ml^–1; mean plasma ACTH 2 h later, at the end of CRH antagonist infusion, was 11.8 ± 2 pg ml^–1. Cortisol levels at the end of the first infusion averaged 14.0 ± 2.9 μg dl^–1, compared with 21.4 ± 4.0 μg dl^–1 at its onset. For both hormones, post-infusion values were significantly lower (P < 0.05) than pre-infusion ones. A similar reduction in ACTH and cortisol levels was evident with the second infusion: ACTH levels were 29 ± 5.8 and 11.5 ± 2.4 pg ml^–1 prior to and after the infusion, respectively (P = 0.06); cortisol levels decreased from 22.9 ± 4.2 to 13.9 ± 3.0 μg dl^–1 (P = 0.03). Plasma ACTH levels measured at 7.30 am on the follow-up morning averaged 20.4 ± 5.1 pg ml^–1. Two hours later (with no antagonist infusion), values for ACTH, 14.0 ± 3.1 pg ml^–1, were not significantly different (P = 0.19, Wilcoxon signed rank sum test). Corresponding cortisol levels were 29.2 ± 9.0 and 20.4 ± 7.6 μg dl^–1, with no significant difference between time-points.

Figure 1.

Plasma levels of ACTH (a) and cortisol (b) in each individual subject before and immediately after infusions of alpha-helical CRH (9–41). The first infusion provided 50 μg kg^–1 and the second 100 μg kg^–1 of the CRH antagonist. Hormones were measured by radioimmunoassay as described in the text.

Afternoon plasma ACTH levels (3–4 pm, 6 h after the second infusion) averaged 15.5 ± 2.7 pg ml^–1. The evening levels of plasma ACTH, measured 12 h post-infusion (9–10 pm) on the first study day averaged 8.8 ± 1.1 pg ml^–1. These values suggest that CRH antagonist infusion did not disrupt the circadian rhythm of this hormone. Concomitant cortisol levels, 15.1 ± 3.0 μg dl^–1 in the afternoon and 5.9 ± 1.6 μg dl^–1 at night, were also consistent with a preserved circadian fluctuation.

Formal cognitive and emotional testing of the subjects was not a goal of this study. However, subjects were monitored continuously during infusions, engaged in conversation and encouraged to express concerns and feeling. No overt differences in behavior or cognition were noted between pre- and post-infusion interviews. All subjects had normal neurological examinations—including a Mini-Mental Status battery and measures of balance, coordination and dexterity—prior to discharge.

Discussion

CRH itself has been given to humans, and has profound effects on autonomic functions and hormonal out put.^17,21–23 CRH elevates plasma ACTH and cortisol, increases the respiratory drive,²² and results in hypotension and flushing.²³ Currently, the clinical uses of CRH are limited to diagnosis, and, rarely, to respiratory stimulation. CRH antagonists should be expected to exert effects secondary to unavailability of CRH at its receptors. Outside the CNS, lower ACTH and cortisol, and potentially, secondary effects of the lack of these hormones (such as hypoglycemia, electrolyte imbalance) may be expected. Although transient reductions of ACTH and cortisol were observed at the end of CRH antagonist infusion, no secondary effects on blood pressure, glucose or electrolytes were observed in this study.

The doses used in this study (50 and 100 μg kg^–1), were clearly sufficient to block pituitary CRH receptors and decrease plasma ACTH and, in turn, cortisol levels. On both days, post-infusion reductions of hormone levels were statistically significant. As infusions were given in the morning, at time of diurnal peak of ACTH and cortisol levels, it may be argued that the reduction simply represents the normal circadian fluctuation. However, when samples were obtained from subjects at the same hours but without alpha-helical CRH-(9–41) infusion, no significant reductions of plasma ACTH and cortisol levels were evident. This is consistent with a direct role for the antagonist in decreasing ACTH release. Further, an ‘artificial’ elevation of pre-infusion ACTH and cortisol levels due to the stress of hospitalization and blood drawing is unlikely. On the second morning of the study, subjects were comfortable and relaxed, and samples were drawn via an existing catheter.

No obvious anxiolytic or euphoric effects of alpha-helical CRH-(9–41) infusion were observed. The antagonist is a peptide, and is not expected to penetrate the intact blood-brain barrier. However, alpha-helical CRH (9–41) interacts avidly with the plasma CRH-binding protein, and active uptake of the complex cannot be excluded. The scope of this study precluded formal assessment of potential CNS effects of CRH antagonist infusion. Recent placebo-controlled studies have revealed placebo effects in up to 40% of subjects for measures of ‘dizziness’ or ‘anxiety’.²⁴ Subjects in this study were hospitalized and subjected to the stressful procedures of intravenous line insertion and blood drawing. Furthermore, placebo infusions were not administered. Therefore, any conclusions regarding cognitive effects of the CRH antagonist in this study would be tentative.

CRH-dysfunction outside the CNS may underlie some forms of Cushing's syndrome and immuno-inflammatory abnormalities.^3,21 Nevertheless, the majority of disorders in which CRH dysregulation may play a key role are those affecting the brain, such as depression, anorexia nervosa, anxiety and some developmental epilepsies.^3,17–19,25 Therefore, therapeutic use of CRH receptor blockers should require compounds that can cross the blood–brain barrier and alter the availability of CRH at receptor sites in the CNS. The cloning of two members of the CRH receptor family^12–14 has permitted the development of screening methods for non-peptide compounds which should be able to penetrate the blood–brain barrier and reach critical CRH receptors in limbic and cortical brain regions. The alterations of ACTH and cortisol levels seen in this study, using relatively low doses of a CRH antagonist, suggest that depression of pituitary-adrenal function poses a potentially important side effect of CRH antagonists considered for clinical research trials for neurological and psychiatric disorders.

In summary, the administration of an inhibitory analogue of CRH to humans is reported. The lack of observed side effects, in conjunction with evidence of efficacy of the antagonist on hormonal parameters, provides strong incentive for developing CRH antagonists as therapeutic agents in several critically important human disorders.