Significant advances in health and social wellbeing have led to linear gains in life expectancy and an accompanying increase in the burden imposed by age-related morbidities. Complex alterations in hormonal networks which regulate homeostasis and survival may underlie this poor adaptation to later life, as exemplified by an increased fracture risk amongst post-menopausal women. Beyond overt under- or overactivity of hormonal axes, changes in the concentrations of regulatory hormones may also impact on health and disease. Subclinical hyperthyroidism, a disorder characterised by normal thyroxine levels in the presence of decreased thyroid-stimulating hormone, is, for instance, independently associated with an increased risk of atrial fibrillation amongst elderly populations. Both the menopause and subclinical thyroid disease demonstrate the difficulty in reversing endocrine changes in later life, with minimal impact from thyroxine therapy in subclinical hypothyroidism and multiple reports of harm resulting from hormone replacement therapy in peri- and post-menopausal women. Given these findings, strategies to locally regulate hormone bioavailability by altering pre-receptor metabolism may offer greater therapeutic potential in the fight against age-related disease. This review aims to provide an overview of the ageing endocrine system and its potential impact on health and disease in the elderly. It will postulate that strategies to coordinate pre-receptor hormone metabolism and a greater understanding of putative hormonal longevity pathways may offer key new drug targets in the fight against ageing, and will argue against applying the conventional endocrine maxim of ‘block and replace' to hormonal changes seen during ageing.
Despite the contribution of sustained improvements in health and social wellbeing to linear gains in life expectancy within the developed world, much of older age is impaired by detrimental changes in body composition and function . Complex alterations in hormonal networks which maintain homeostasis and regulate reproduction, metabolism, nutrition and growth may underlie this poor adaptation to later life. The secretion of hormones decreases within most axes, the impact of which is augmented by a reduction in the sensitivity of tissues to their action, and normal circadian rhythms are lost. Endocrine axes manifest these changes with clinically identifiable losses of function such as those seen in the ageing of the reproductive system (menopause and andropause), the growth axis (somatopause) and axes involving the adrenal gland (adrenopause).
The clinical sequelae of these changes are variable but include reductions in bone, skin and skeletal muscle mass and strength, derangement of insulin signalling, increases in adipose tissue and effects on immune function. Consequently, a number of studies have been carried out to assess the benefits of hormonal supplementation in the elderly, but the efficacy of these interventions remains relatively unclear. It is also increasingly recognised that endocrine adaptation during earlier life may affect longevity and health in older age . Relative decreases in early nutritional intake have, for instance, found associations with both prolonged lifespan and reduced incidence of pathologies such as diabetes and cardiovascular disease via effects mediated by the insulin and insulin-like growth factor-I (IGF-I) pathways .
This article will highlight pathways linked to both longevity and variable endocrine function in the elderly as a means to delineating the myriad changes in hormonal signalling seen during the ageing of an individual. It will subsequently evaluate the clinical impact of these changes before arguing that targeted strategies to locally modulate dysregulated hormonal signalling during ageing may be of potential therapeutic benefit.
Endocrine Regulation of the Ageing Process
The recognition that food restriction in animals alters biomarkers associated with longevity may afford the pituitary gland and its downstream pathways additional significance in the control of ageing [2,3]. The existence of these evolutionarily conserved pituitary-derived longevity pathways is supported by studies undertaken in mice and a raft of simpler organisms. It is recognised, for instance, that mutations of the IGF-I receptor homologue DAF2 which interfere with the functioning of the IGF-I/insulin signalling pathway in Caenorhabditis elegans and Drosophila melanogaster extend lifespan, as do reductions in signalling through the separate growth hormone (GH)/IGF-I and insulin pathways in mice [4,5].
The potential role of insulin and GH/IGF-I pathways in controlling longevity is exemplified by the Ames dwarf mouse, which lacks the PROP1 transcription factor. Deletion of this factor, which is crucial to the development of the pituitary, leads to an absence of GH, prolactin and thyroid-stimulating hormone (TSH), in addition to decreased circulating levels of IGF-I . A similar picture is found with mutations of the pituitary-forming transcription factor PIT1 in Snell dwarf mice, yet in both instances, mice live longer even with amelioration of their hypothyroidism by thyroxine replacement [6,7]. Given that deletions of the GH receptor and GH-releasing hormone (GHRH) receptor have similar consequences and that female mice with cell autonomous IGF-I receptor mutations live longer than their wild-type counterparts, it follows that the GH/IGF-I axis at least partly determines longevity . In these global knockout models, cell autonomous reduced signalling through the insulin and IGF-I pathways leads to a reduction in the phosphorylation of their downstream intracellular targets AKT/protein kinase B (PKB) and serum/glucocorticoid-regulated kinase (SGK). This, in turn, reduces the phosphorylation of the forkhead box transcription factors (FOXO), resulting in their nuclear translocation and both the subsequent transcription of longevity genes and inhibition of their pro-ageing counterparts (fig. 1) [2,6,7,8].
Alternatively, numerous studies focussed on the highly conserved insulin/IGF-I signalling pathway have raised the exciting potential of additional non-cell autonomous endocrine regulation of ageing which is succinct from pituitary-derived longevity pathways. It is recognised in a number of species of both flies and worms that control of putative longevity factors such as the insulin/IGF-I pathway may originate at the level of cells such as adipocytes and neurones. In these instances, a mutant cellular genotype is able to exert cell non-autonomous control of the phenotype of other non-mutated tissues by as yet undetermined endocrine factors in order to alter their phenotype to one compatible with a longer lifespan . Whilst it is not yet clear whether these findings are applicable to the mammalian IGF-I receptor, the fat-specific insulin receptor knockout (FIRKO) mouse provides evidence for the existence of mammalian cell non-autonomous control of ageing. In this mouse, specific knockout of the insulin receptor within adipocytes leads to a significantly increased lifespan and phenotypic change within a raft of tissues not carrying the knockout mutation .
This raises the prospect of adipocytes producing as yet undetermined humoral factors which trigger the ageing of multiple different tissues. Similarly, the discovery in the human circulation of fragments of the transmembrane protein Klotho, which favours longevity when expressed, lends more evidence to the existence of humoral ‘longevity factors', as do a multitude of lipophilic hormone pathways associated with ageing regulation within simple organisms. Should these processes be effectively characterised, novel therapeutic targets to modulate ageing may be on the horizon.
Ageing of the Endocrine System
Given that the most reproducible factor affecting the function of key endocrine regulators of homeostasis, the hypothalamus, is ageing, much of elderly morbidity may be attributed to more overt and potentially replaceable hormone imbalances. By better understanding the changes undergone by the ageing endocrine hypothalamus and both its downstream pathways and counterpart regulators, the potential utility of hormone supplementation in the elderly will be determined.
The Hypothalamic-Pituitary-Gonadal Axis
The decline in the reproductive capacity of women in the late fourth to fifth decade of life is accompanied by diverse sequelae, including an increased risk of osteoporosis, psychogenic disturbance and a greater burden of cardiovascular and cerebrovascular disease . Attempts to ameliorate losses in oestrogen through hormone replacement therapy (HRT) have, however, courted significant controversy relating both to increased risks of malignancy and vascular events, thereby standing as an exemplar for the difficulty in managing hormone balance during ageing [10,11]. Recognised changes in female sex steroids are matched by the andropause, a relatively poorly defined process characterised by a progressive age-dependent loss of the anabolic androgen testosterone within males . The reported positive and negative health outcomes associated with reproductive HRTs are summarised in table 1.
Within reproductive life, oestrogen production by the ovaries is tightly controlled by negative feedback mechanisms with limited influence exerted by extra-gonadal synthesis. However, approximately 90% of circulating oestrogen is lost at the time of ovarian failure, during the fourth and fifth decade of life, prompting extra-glandular formation of oestrogens by aromatase expression within adipose and skin to become the predominant source of sex steroid generation, with additional contribution from the adrenals.
Unlike the dramatic changes seen in women, testosterone loss seen during the andropause is both insidious and more contentious. There is no specific age at which this process starts and cross-sectional studies have suggested that it is driven by global reductions in activity across the hypothalamic-pituitary-gonadal axis, though reduced GnRH neuronal release is postulated to be the principal change with primary testicular failure as a contributing factor [8,13]. Sex hormone-binding globulin (SHBG) levels increase during ageing, meaning that the andropause is biochemically manifest by a relatively greater decrease in free testosterone than its bound counterpart [8,13].
In both men and women, loss of sex hormones during ageing contributes to changes in body mass, musculoskeletal integrity, sexual dysfunction and long-term risks of health and disease. Interestingly, decreased libido in post-menopausal women may be attributed to reductions in testosterone, the additional impact of which in men may be erectile dysfunction . The literature concerning the role of sex hormone changes in a raft of additional cognitive changes within ageing men and postmenopausal women is more disparate, with inconsistent studies within both males and females . Persistently elevated FSH within postmenopausal women has, however, been linked to Alzheimer's disease .
Perhaps the most recognised change resulting from the menopause is that of osteoporosis. This appears to be driven by reductions in metabolically active trabecular bone occurring due to uncoupling of the bone remodelling cycle secondary to oestrogen loss and leads to a marked increase in fracture risk amongst elderly women . Whilst men are relatively more protected from the effects of decreased bone mineral density due to the attainment of a higher peak bone mass, the prevalence of osteoporosis is known to increase amongst ageing males . A number of studies have reported both maintenance of bone mineral density following testosterone supplementation during ageing and a positive relationship between endogenous androgen levels and bone mineral density, yet men with defunctioning oestrogen receptor mutations and aromatase deficiency have been identified to have lower bone mineral density [18,19,20,21]. This, and research identifying that serum oestrogen correlates more closely with peak bone mass than testosterone, indicates that both male and female sex steroids are required for the maintenance of bone mineral density during ageing. Further, it is not yet known whether testosterone supplementation affords any reduction in the frequency of fractures amongst ageing men.
The significant increase in the risk of heart disease in women undergoing the menopause, coupled with the relatively reduced burden of heart disease amongst premenopausal women when compared to age-matched males, indicates a cardioprotective role for oestrogen which is lost in later age . The relationship between testosterone and heart disease in both males and females is less clear. In women, a pooled meta-analysis has suggested that, whilst beneficial for sexual function, the addition of testosterone to HRT may result in an adverse increase in high-density lipoprotein cholesterol. This study was, however, unable to confirm longer-term effects on morbidity. In contrast, whilst studies focussing on testosterone supplementation amongst males have extolled short-term gains in sexual function, strength and well-being, a number of studies have identified significant increases in cardiovascular adverse outcomes amongst men with late-onset hypogonadism who receive testosterone supplementation [22,23,24]. A less significant but similar effect for testosterone has been recognised amongst ageing men . Although the manner in which sex hormones modulate cardiovascular risk is unclear, it is postulated to include changes in cholesterol and both high- and low-density lipoproteins. It is likely to be additionally, albeit indirectly, related to increases in insulin resistance which follow falling sex hormone concentrations . Testosterone may additionally directly affect the myocardium via androgen receptors.
Despite similarly reduced sex hormone concentrations in ageing men and women, the menopause is set apart from the andropause by the extent to which it induces reproducible symptoms which significantly impair quality of life. It is accompanied by a cacophony of urogenital symptoms and changes, including urinary frequency, dysuria, incontinence and vaginal atrophy, all of which carry significant burden for postmenopausal women . These symptoms are in turn accompanied by those resulting from alterations in the set point of the hypothalamus, including recurrent hot flushes recognised to follow luteinizing hormone (LH) surges, and changes in serotonin levels .
Despite these many changes, HRT has met with considerable controversy in females. Its use for the primary or secondary prevention of cardiovascular disease has, for instance, been warranted both ineffective and harmful, at least in its combined continuous form . Analysing thirteen trials incorporating a total of 38,171 women, a recent Cochrane review identified an increased risk of stroke, venous thromboembolic events and pulmonary embolism in patients treated with HRT, revealing relatively low numbers needed to harm . Similarly, although initially considered to be of some excitement as a ‘rejuvenating hormone', testosterone therapy remains highly contentious within men. It is on the basis of current evidence entirely unclear as to whether androgen replacement affords any significant advantage to men with testosterone within normal parameters for their age .
The balance of risks and benefits for HRT in women is more complex and the British Menopause Society has published extensive guidance relating to its use in peri- and post-menopausal women with or without a uterus . In summary, it appears clear at present that HRT offers support with treating vasomotor, urogenital and some psychogenic symptoms related to the menopause. Whilst it decreases fracture incidence, its risks mean that alternative therapies such as bisphosphonates are preferred for the long-term prophylaxis and treatment of osteoporosis [25,27,28]. It has no role in the prevention or treatment of cardiovascular disease and previous links to a significantly reduced risk of colorectal cancer incidence have not been supported by strong evidence . There is, however, a clear increase in the risk of breast cancer, coronary events, venous thromboembolism, gallbladder disease and, putatively, ovarian cancer in women taking HRT, meaning that its use should ideally be curtailed to short-term therapy only [25,27,28].
The Hypothalamic-Pituitary-Adrenal Axis
The hypothalamic-pituitary-adrenal axis sits as an adaptive regulator of stress response and is integral to homeostasis. It is perhaps the adrenal steroids DHEA and DHEA-S that have evoked the greatest attention in the search for effective anti-ageing hormones . Arguably poorly understood, DHEA and DHEA-S are the most abundant of the steroid hormones and may act as neurosteroids, with postulated immunoprotective, anti-cancer and anti-ageing effects [8,30]. Unlike other adrenal hormones, the production of DHEA is closely tied to ageing so that secretion is greatest at the beginning of the second decade of life, falling thereafter. This decrease, which results in DHEA levels falling longitudinally by approximately 2-3% each year, is thought to follow involution of the zona reticularis of the adrenal gland [30,31].
In a large prospective observational study of DHEA, healthy persons were noted to have significantly lower DHEA and DHEA-S concentrations than their more poorly ageing counterparts . Similarly, a number of cross-sectional studies have revealed a close, albeit circumstantial, correlation between DHEA-S levels and a raft of morbidities, including depression, type 2 diabetes and Alzheimer's disease . Whether this relationship is causal remains to be determined, with a large double-blind randomized controlled study revealing only increased libido and selectively improved skin status and bone turnover in elderly women, and not men, treated with DHEA supplementation . A more recent double-blinded study, which reviewed the effect of DHEA replacement in patients with low levels of the hormone, recognised no beneficial effects from DHEA supplementation on physiological parameters of health. This apparent sexual dimorphism of adrenal hormone release may have greater significance in the pathogenesis of cardiovascular disease and cognitive dysfunction, particularly given suggestions that ageing women exhibit lower DHEA(S) and higher cortisol levels, and thus a higher cortisol/DHEAS ratio, than men .
Unlike many other hormones, concentrations of both cortisol and its binding globulin remain relatively consistent throughout ageing [35,36]. There is, however, some evidence to suggest diminished negative feedback on cortisol production, which may lead to a greater response to corticotropin-releasing hormone (CRH) challenge, particularly in women . Similarly, the mineralocorticosteroid aldosterone demonstrates a moderate decrease during life, and both this and an associated decrease in plasma renin activity are considered to impact minimally on physiological processes in the elderly [35,36].
The Hypothalamic-Pituitary-Thyroid Axis
Dysfunction of the thyroid axis is common in the general population and even more prevalent in the elderly, with an increased incidence of overt thyroid under- or overactivity . This is augmented by significant numbers of patients with subclinical thyroid pathology, which is found in more than 10% of patients aged over 80 years .
The burden imposed by disorders of the thyroid in the elderly is, however, unclear. Confounding factors include the ‘nonthyroidal illness syndrome', characterised by decreased serum free triiodothyronine (T3), low or normal TSH, high reverse T3 and relatively normal free thyroxine (T4) concentrations. It is important to make a differential diagnosis between this sick euthyroid syndrome and subclinical hyperthyroidism. The failure of the serum TSH to rise in response to low circulating thyroid hormone concentrations in critical illness arises from a degree of central hypothyroidism caused by alterations in the set point of the hypothalamo-pituitary-thyroid axis . A similar pattern may also be found during treatment with certain drugs including corticosteroids and dopamine agonists.
Primary hypothyroidism, the most common pathological hormone deficiency, occurs more often in women than in men and increases in incidence with age . Like hyperthyroidism, the presentation of hypothyroidism may be overt or subclinical (raised TSH with normal free T4 and free T3 concentrations). Its presentation is both insidious and non-specific, with psychiatric symptoms such as depression particularly common in the elderly, in addition to hair loss, dry skin, cold intolerance, reduced appetite, fatigue and weakness . Within elderly populations, Hashimoto's thyroiditis is the most common cause of hypothyroidism, although the use of radiotherapy, surgery and iodine for thyroid overactivity or malignancy is a significant contributing factor [39,40]. The increasing use of tyrosine kinase inhibitors (TKIs), such as sunitinib, may also impact on the prevalence of this disorder .
Subclinical hypothyroidism, on the other hand, reflects early thyroid failure and its accurate diagnosis depends on the recognition of the physiological shift in the upper limit of TSH towards higher concentrations, which occurs during ageing . This age-specific TSH trend necessitates the use of age-appropriate reference ranges when evaluating thyroid dysfunction in the elderly, although it is broadly classed as mild (serum TSH 4.5-10.0 mIU/l) or severe (TSH >10.0 mIU/l) . The diagnosis of subclinical hypothyroidism is additionally complicated by its paucity of objectively definable symptoms. Two large cross-sectional studies have, for example, failed to demonstrate any association with cognitive dysfunction, anxiety or depression in elderly patients, with an additional study suggesting that walking speed and physical function may in fact be better in patients with subclinical hypothyroidism .
There are a number of important postulated long-term impacts of both overt and subclinical hypothyroidism [37,38,41]. Overt hypothyroidism is associated with dyslipidaemia and, though the relationship between serum lipid concentrations and subclinical hypothyroidism is more controversial, a recent systematic review demonstrated that levothyroxine replacement results in small beneficial effects on total and low-density lipoprotein cholesterol . An increased risk of heart failure has also been demonstrated in patients with subclinical hypothyroidism aged over 65 years . Despite this, the Leiden study of persons aged over 85 years has associated raised TSH with decreased risk of death from cardiovascular disease and increased longevity, although TSH >10.0 mIU/l has been linked to an increased risk of coronary heart disease events and related mortality in all ages .
The prevalence of thyrotoxicosis increases with age and is highest within white populations and in iodine-deficient areas . A reduction in the frequency of classically described symptoms such as heat intolerance, tremor and anxiety is observed in elderly people who often present in an asymptomatic manner . Weight loss and shortness of breath are more commonly reported, and atrial fibrillation is more frequent in people aged more than 60 years . Thyrotoxicosis may be associated with hyperthyroidism when there is increased production and secretion of thyroid hormones or may be due to the release of pre-formed thyroid hormones into the circulation. Whilst Graves' disease accounts for the majority of hyperthyroidism in the general population, the prevalence of toxic nodular hyperthyroidism is disproportionately high in the elderly and may be the principal underlying aetiology . Causes of ‘leakage hyperthyroidism' include subacute and drug-induced thyroiditis. Amiodarone may also cause thyrotoxicosis in 6-10% of patients who take it, both by inducing hyperthyroidism (type 1) and by causing thyroiditis (type 2) .
The long-term sequelae of subclinical hyperthyroidism are somewhat more contested but summarised within table 2 [37,43]. Data from three meta-analyses have shown no increase in cardiovascular mortality or all-cause mortality in patients with subclinical hyperthyroidism, although a further three have demonstrated an association . Intriguingly, one meta-analysis highlighted that mild hyperthyroidism was associated with increased cardiovascular mortality only in those with other morbidities, perhaps instead identifying a link with non-thyroidal illness syndrome. Both overt and subclinical hyperthyroidism are additionally associated with reduced bone mineral density, whilst low TSH levels have also been linked to poor cognitive function and dementia, yet underlying mechanisms remain unclear [39,42,43].
Overt hypothyroidism is treated with levothyroxine monotherapy and hyperthyroidism is best managed with radioiodine, especially if the underlying aetiology is toxic nodular disease, although thyroidectomy (following induction of euthyroidism) or long-term low-dose anti-thyroid drugs may be preferred in certain patients. The treatment of subclinical thyroid dysfunction is more controversial and guidelines for treatment in such patients are based on the degree of abnormality in TSH concentrations, the presence of comorbidities and patients' symptom profiles [41,43]. There is an urgent need for well-conducted clinical trials to indicate the need for intervention in patients with mild thyroid dysfunction and the effects of these therapies on the long-term consequences of subclinical thyroid disease.
Pituitary: GH, IGF and Prolactin
Levels of both GH and IGF-I fall in concert during ageing and may more than halve between early and later life in a process termed the somatopause. Its postulated consequences are extensive and, given that the GH/IGF-I axis represents the principal anabolic hormone axis in females, may differ between sexes. Given that a number of factors are known to differentially stimulate GH/IGF-I action and activity, it is perhaps unsurprising that the clinical impact of altered GH/IGF-I levels in older age remains unclear . In isolation, GH and IGF-I supplementation has been inconsistently linked to improvements in strength training in older individuals .
Hormonal links between the ageing pituitary and cognitive, rather than physical, changes have been afforded by prolactin, a hormone secreted by the anterior pituitary lactotrophs . Peak concentrations of this hormone, which occur shortly prior to the onset of sleep, have been demonstrated to negatively correlate with age [46,47]. This, and reduced pulse amplitude during sleep in older age, indicates decreased dopaminergic inhibitory action on prolactin production in the elderly during the sleep period [15,26]. This may be restorable with dopaminergic agents such as metoclopramide, potentially assisting in the control of age-related disordered sleep patterns such as early sleep onset and awakening.
The Gut Hormones
Perhaps the most significant endocrine organ to impact on age-related morbidity is the endocrine pancreas. Insulin secretion is known to be affected by ageing, falling with advancing age by as much as 0.5% per year, independent of its action in peripheral tissues . A decline in both the number of pancreatic islet β cells and their function, occurring independently of peripheral resistance to insulin, is thought to underlie this and may contribute to the elderly population's susceptibility to developing diabetes .
The decrease in islet cell function appears to be multifactorial in aetiology, with postulated contributing factors including a loss of expression of β-adrenergic receptors and orphan G-protein-coupled receptors encoded by GPR39, with a consequent reduction in adrenergic stimulation of insulin secretion and pancreatic function . Reduced circulating vitamin D and impaired incretin sensitivity have also been linked to cellular dysfunction . There appears to be an additional age-related decline in cell cycle entry of β cells and a reduction in their regeneration capacity which contributes to a decrease in pancreatic islet β-cell mass .
Hormones involved in governing satiety and informing energy expenditure, such as adiponectin, ghrelin and leptin, also undergo a number of age-related changes which are yet to be accurately characterised .
Vitamin D has been linked to a premature ageing phenotype in mouse models with hypervitaminosis D. Seemingly conflictingly, severe vitamin D deficiency demonstrates a similar phenotype. These data demonstrate that vitamin D is linked to ageing, that it may contribute to health and disease in elderly individuals and that there appears to be a U-shaped dependence on its concentration . It should be noted, however, that reports of a U-shaped response to vitamin D levels have recently been extensively critiqued .
Pre-Receptor Hormone Metabolism
The activity of a number of hormones is determined by the degree to which they are available to bind at intracellular receptors. For the majority of hormones, the effect of this is masked by significant changes in the humoral level of the hormone during ageing, yet for hormones such as cortisol, the circulating level of which remains relatively stable during ageing, pre-receptor regulation may contribute to significant morbidity in the elderly.
It is now well documented that the glucocorticosteroid-activating enzyme 11β-hydroxysteroid dehydrogenase type 1 (11βHSD1), which converts the inactive steroid cortisone to its active catabolic product cortisol, increases in activity and expression during the ageing of bone, skeletal muscle, skin and the central nervous system [53,54]. It is postulated that this local glucocorticoid excess, occurring despite normal circulating concentrations of cortisol, could underlie conditions including osteoporosis, sarcopenia, skin atrophy and cognitive decline. Whilst work continues to focus on postulated regulatory pathways for this increased enzymatic activity, including those important in the concept of ‘inflammaging', specific inhibitors of 11βHSD1 have been developed and are entering study in phase 2 clinical trials following exciting results in mouse models in vivo. Replicating this approach for other hormones may provide yet more targeted avenues for regulating the impact of hormone imbalance on age-related morbidity.
Ageing is associated with significant changes in the regulation of hormonal axes and the downstream availability of their active products. Many of these changes are well characterised, but their precise impact on health and disease remains unclear. Similarly, whilst hormonal deficit and excess has been linked to pathological changes in ageing individuals, the harm imposed by inappropriate HRT administration demonstrates that the conventional endocrine maxim of ‘block or replace' is too simple a means to addressing age-related morbidity. Mechanisms to control pre-receptor hormone metabolism may afford significant future therapeutic potential, as may control of longevity pathways through as yet largely undetermined hormonal mediators of cell non-autonomous signalling.
The authors confirm that they have no conflicts of interest relating to this article.