ACTH precursors: biological significance and clinical relevance

ACTH is synthesized as part of the precursor proopiomelanocortin or POMC ( Fig. 1). In the human pituitary POMC is enzymatically processed to pro-ACTH, which is then cleaved to give ACTH. Processing of POMC in the pituitary takes place within the regulated secretory pathway and is accomplished by the protease PC1 located within mature dense core granules ( Benjannet et al., 1991 ; Smeekens et al., 1991 ). Peptides derived from POMC in the human pituitary gland. β-LPH, β-lipotrophin; N-POC, N-terminal proopiomelanocortin; JP, joining peptide; γ-LPH, γ-lipotrophin; β-EP, β-endorphin. Y: Binding sites of monoclonal antibodies used in the ACTH precursor IRMA and the ACTH IRMA. The presence of ACTH precursors in plasma was first demonstrated in normal subjects after stimulation with metyrapone ( Yalow & Berson, 1971) and was later observed after insulin-induced hypoglycaemia ( Hale et al., 1986 ). However, complex chromatographic techniques were required to separate ACTH precursors from ACTH. Clearly, this approach could not be used for large numbers of patient samples and would not provide a quantitative assessment of the concentrations of the ACTH precursors in plasma. We have subsequently quantified the concentrations of ACTH precursors in the human circulation under basal conditions. Direct measurement of these peptides was made possible by the development of a two-site immunoradiometric assay for the ACTH precursors, POMC and pro-ACTH ( Crosby et al., 1988 ). The assay is based on a labelled monoclonal antibody which binds within the ACTH region of POMC and a solid phase antibody which recognizes N-terminal POMC (N-POC, Fig. 1). As binding of both antibodies is required to generate a signal, the assay does not detect ACTH. Using this assay, the concentrations of ACTH precursors in normal subjects were found to be 5-40 pmol/l which is equivalent to or greater than those of ACTH, N-POC, β-lipotrophin (β-LPH) and β-endorphin ( Gibson et al., 1994 ). The levels of ACTH precursors and ACTH in the circulation at any given time are dependent on regulatory mechanisms influencing expression of the POMC gene. However, the relative concentrations of ACTH precursors and ACTH will depend on precursor processing and mechanisms of secretion from the corticotroph cells. Evidence from studies with the mouse corticotroph adenoma cell line, AtT20, suggests that in the absence of stimulation, corticotroph cells ‘leak’ newly synthesized POMC ( Kelly, 1985). Furthermore, results we have obtained in normal subjects and from clinical studies indicate that acute regulation of secretion of ACTH and ACTH precursors differs (see below). The development of sensitive immunoradiometric assays for ACTH has improved the reliability of ACTH measurement. This is particularly important for the differentiation of ACTH dependent and independent Cushing's syndrome. However, ACTH measurement rarely distinguishes between pituitary and ectopic sources of ACTH in Cushing's syndrome because of the considerable overlap in ACTH concentrations in these two patient groups. Approximately 15-20% of cases of ACTH-dependent Cushing's syndrome are traced to ectopic tumours and small cell lung carcinoma (SCLC) is the most common cause. These tumours are almost always visible on plain chest X-ray but can also be distinguished from pituitary tumours by the short history, weight loss, severe myopathy, pigmentation and hypokalaemia ( Wajchenberg et al., 1994 ; Trainer & Grossman, 1991). However, small ‘occult’ ectopic ACTH secreting tumours such as carcinoids frequently present a clinical picture identical to that seen with pituitary-dependent Cushing's disease and can be difficult to diagnose. Chromatographic studies performed as early as 1971 suggested that high molecular weight forms of ACTH were present in extracts from ectopic tumours ( Yalow & Berson, 1971). However, the relevance of this to circulating hormone levels and their action on distant tissues was uncertain. Subsequently, Rees and colleagues demonstrated the presence of POMC and pro-ACTH in plasma obtained from 7 patients with the ectopic ACTH syndrome ( Ratter et al., 1983 ; Hale et al., 1986 ). In a study comparing patients with pituitary and ectopic tumours causing Cushing's syndrome, we quantified ACTH precursor concentrations and compared the molar ratio of ACTH precursors to ACTH. This provided clear evidence that ACTH precursor concentrations were elevated in all of the 15 patients with ectopic tumours when compared with patients with pituitary microadenomas. In comparison, ACTH levels in the two patient groups did not discriminate because they showed significant overlap ( Stewart et al., 1994 ). The measurement of ACTH precursors requires only a single sample. This is in contrast to the extensive and costly procedures required for inferior petrosal sinus sampling, now commonly used to differentiate between pituitary and ectopic causes of Cushing's syndrome ( Oldfield et al, 1991 ). In our study, the majority of patients with the ectopic ACTH syndrome had clinically obvious small cell carcinoma of the lung. Clearly, the value of ACTH precursor measurement depends on establishing the differential diagnosis between patients with a small corticotroph adenoma and patients with small and slow-growing carcinoid tumours. The most common of these are bronchial and thymic carcinoids which can both give misleading results in dynamic tests. To date we have studied a group of 48 patients histologically proven to have the ectopic ACTH syndrome. Of these, 12 patients had carcinoid tumours with ACTH precursor concentrations of 151-1300 pmol/l (median, 544 pmol/l) and corresponding ACTH results of 7-44 pmol/l (median, 17.5 pmol/l) ( Table 1). In the remaining 36 patients ACTH precursors ranged from 139-18000 pmol/l (median 1911 pmol/l) and ACTH from 1-564 pmol/l (median 50 pmol/l). These results suggest that the levels of ACTH precursors and ACTH are lower in carcinoid tumours than in other ectopic tumours. However, ACTH precursor measurement still differentiated between carcinoid tumours and pituitary microadenomas ( Fig. 2). Comparison of ACTH precursor levels in patients with the ectopic ACTH syndrome and patients with Cushing's disease (excluding those patients with macroadenomas). A recent report indicated that bronchial carcinoids were not easily identified by an assay which detects POMC but not pro-ACTH ( Raffin-Sanson et al., 1996 ). In view of this, it is clearly important to measure both POMC and pro-ACTH in clinical samples. Further support for this is given by the detection of the enzyme PC1 responsible for processing POMC to pro-ACTH in bronchial carcinoid tumours (n = 7) and in carcinoid tumours of the GI tract (n = 26) ( Scopsi et al., 1995 ). However, PC1 has been shown to process pro-ACTH to ACTH. Therefore for pro-ACTH to be elevated this mechanism must be inefficient in these tumours. In addition, expression of the processing enzyme PC2 was noted in 76% of the carcinoid tumours studied suggesting that some carcinoid tumours may process POMC to the smaller derivatives such as the C-terminal fragment of ACTH, corticotrophin-like intermediate lobe peptide (CLIP). Indeed, the presence of CLIP has been described in some carcinoids, but in tumour extracts rather than in the circulation ( Vieau et al., 1989 ). Expression of PC1 and PC2 has been demonstrated in only a minority of SCLC tumours and in a study of 13 SCLC cell lines expression of PC1 was restricted ( Creemers et al, 1992 ). This data complements our work with a panel of SCLC cell lines which secrete high levels of ACTH precursors but have undetectable ACTH in chromatographic studies ( Stewart et al., 1989 ; Crosby et al., 1990 ). While we and others have shown that patients with pituitary corticotroph microadenomas have low levels of ACTH precursors, patients with large invasive pituitary tumours exhibit impaired processing resulting in high ACTH precursor levels ( Gibson et al., 1996 ; Raffin-Sanson et al., 1996 ). In a series of 13 such patients, ACTH precursor concentrations were 138-5640 pmol/l (median, 1096 pmol/l) while ACTH levels were 4-66 pmol/l (median, 20 pmol/l). In some cases the ACTH levels could be explained by high levels of ACTH precursors which can cross-react up to 10% in the ACTH assay. We have used this characteristic secretion of ACTH precursors to monitor for tumour recurrence after surgery or treatment. Given that concentrations of ACTH precursors in the circulation are at least as high as those of ACTH, it is important to determine how they may be regulated. As indicated previously, where regulation is at the level of the POMC gene the effects should be seen on both ACTH and ACTH precursors. However, in inferior petrosal sinus sampling (IPSS), CRH-stimulated ACTH concentrations are seen within 1-2 min, representing release of stored ACTH. Our data on responsiveness of ACTH precursors to CRH comes from patients with Cushing's disease studied during IPSS. Concentrations of both ACTH precursors and ACTH were increased in the petrosal sinus after CRH administration. However, the magnitude of response was very different. The median increment (% increase above basal) in ACTH levels after CRH was 459% compared to only 96% for ACTH precursors ( Gibson et al., 1994 ). These data emphasize the high degree of processing to ACTH which occurs in secretory granules. The regulation of ACTH and ACTH precursor secretion by glucocorticoids is rather more complicated. In normal subjects, ACTH concentrations were undetectable after an overnight dexamethasone suppression test while ACTH precursor levels remained within the normal range ( Gibson et al., 1994 ). This may to some extent reflect differences in plasma half-lives of the ACTH precursors and ACTH. Following prolonged treatment with glucocorticoids, we would expect ACTH precursors to suppress due to down-regulation of the POMC gene. Our results in a patient with isolated ACTH deficiency would seem to confirm this view. The patient's initial ACTH precursor levels were 129 pmol/l, but reduced to within the normal range during hydrocortisone replacement ( Nussey et al., 1993 ). The existence of distinct constitutive and regulated pathways for ACTH secretion in corticotrophs has been proposed based on studies in the mouse corticotroph adenoma cell line AtT20 and in rat anterior pituitary cells. In the constitutive pathway, unprocessed POMC would be continuously released without becoming subject to regulatory events affecting secretory vesicles. The apparent lack of responsiveness of ACTH precursors to acute regulation by CRH and dexamethasone suggests that precursors may be released in this manner. The clinical significance of ACTH precursors is dependent on their biological activity. However, it has proved difficult to get a clear indication of precursor bioactivity because of problems in obtaining pure preparations of the peptides and the limitations of available bioassays. POMC itself is thought to have little biological activity ( Odell, 1991) while pro-ACTH was shown to be equipotent with ACTH in a rat adrenal cell bioassay (A.C. Hale et al., personal communication) or 8-33% as potent in a cytochemical ACTH bioassay ( Ratter et al., 1983 ). Nothing is currently known about the binding of POMC and pro-ACTH to the ACTH receptor (MC2-R) and the MSH receptors (MC1-R, MC3-R, MC4-R, MC5-R) (see Cone et al, 1996 ). It will be important to examine the agonist/antagonist activity of the precursors at the human receptors since precursors could preferentially bind MSH receptors in certain tissues, for example MC1-R in the skin. In a study of 24 patients with pigmentation related to postadrenalectomy Cushing's disease, concentrations of both ACTH and ACTH precursors correlated with pigmentation score ( Ray et al., 1995 ). The interaction of ACTH precursor peptides with the recently cloned receptor MC4-R found exclusively in the brain may also be of interest as we have demonstrated that concentrations of ACTH precursors in CSF are 100-fold those of ACTH (414 ± 83 pmol/l vs 3.2 ± 0.6 pmol/l, n = 26) ( Tsigos et al, 1993 ). Some information regarding in vivo POMC bioactivity can be gained from clinical studies. If patients with the ectopic ACTH syndrome due to SCLC produce POMC alone (as suggested by the evidence above) then it must either have biological activity when present at very high levels or be cleaved to ACTH in the periphery. Our study in patients with invasive pituitary macroadenomas indicates that ACTH precursors are of lower biological activity than ACTH, as two patients had normal cortisol levels in the presence of up to 2000 pmol/l ACTH precursors. In a case of isolated congenital ACTH deficiency in which the patient had ACTH precursor levels of 39-57 pmol/l, there was no authentic ACTH 1-39 and undetectable cortisol, inferring that these concentrations of precursors had no biological activity ( Nussey et al., 1993 ). How well current ACTH assays recognize the ACTH precursors is an important consideration for clinicians. In recent years there has been a shift towards the measurement of ACTH by two-site immunometric assays in preference to RIA. This change provides many benefits including speed, reproducibility, sensitivity and parallel results. The high sample throughput and wide working range of these assays makes them ideal for measuring samples taken during inferior petrosal sinus sampling, a test which is fast becoming an important component in the diagnosis of pituitary tumours secreting ACTH. However, we must recognize that there are clinical situations in which the use of an assay which is highly specific for ACTH 1-39 may be insufficient or misleading. In one patient with the ectopic ACTH syndrome shown by chromatography to be producing high molecular weight ACTH precursors, the ACTH was very low when measured by immunoradiometric assay ( Raff & Findling, 1989). To ensure that patients with the ectopic ACTH syndrome are flagged up by an ACTH assay, it is important that there is a high degree of cross-reactivity of the ACTH precursors in the ACTH assay or that a separate specific assay for ACTH precursors is available. Since the initial observations in the early 1970s of the presence of ACTH precursors in the circulation, much has been learnt about their existence in a range of ACTH-related disorders. This review has considered situations where precursors are in marked excess of ACTH including ectopic ACTH syndrome, pituitary macroadenomas, Nelson's syndrome, a case of isolated ACTH deficiency and in cerebro-spinal fluid. Important questions remain to be answered not least of which is how POMC and pro-ACTH bind to and activate the recently cloned ACTH and MSH receptors. Does the folding of POMC specify whether it binds through its ACTH, βMSH or γMSH sequences? Could POMC have biological activity distinct from its role as a precursor of ACTH by binding through its βMSH or γMSH sequences? The in vivo evidence of the relative biological activity of ACTH precursors and ACTH is sparse and limited to cortisol levels. Clearly further studies are required on the role and regulation of ACTH precursors before we fully understand their biological significance and clinical relevance.