Growth hormone and IGF-1 actions in the brain and neuropsychiatric diseases

Abstract

Growth hormone (GH) is secreted by the anterior pituitary gland under the control of hypothalamic neuroendocrine neurons that express somatostatin or growth hormone-releasing hormone (GHRH). Ghrelin, originating primarily in the stomach, is also an important GH secretagogue. GH stimulates the hepatic secretion of insulin-like growth factor-1 (IGF-1) and the expression of IGF-1 in extra-hepatic tissues, including the brain. Many regions of the brain express receptors for GH, IGF-1, and ghrelin. In recent decades, evidence from both human and animal studies has indicated that GH, IGF-1, and ghrelin regulate numerous brain functions. Alterations in the secretion or sensitivity to these hormones may represent risk factors for developing neurodegenerative diseases (e.g., Alzheimer’s and Parkinson’s) and neuropsychiatric conditions (such as depression, anxiety, posttraumatic stress disorder, schizophrenia, and bipolar disorder). Additionally, classical neurodevelopmental disorders such as autism spectrum and attention-deficit hyperactivity disorder may also be influenced by somatotropic hormones. This review aims to summarize and discuss the emerging role of GH and IGF-1 in influencing brain function and the predisposition to brain diseases and neuropsychiatric disorders.

Introduction

The somatotropic system, composed of extra-pituitary circuits and the somatotropic axis, ensures a balance between form and function in all organisms. While the ability to grow is a characteristic of all living beings, the acquisition of a spinal column and joints led to the emergence of vertebrates (1). The development of the appendicular skeleton, jaw and teeth, and skull and brain enabled vertebrates to perform deliberate postures and movements, secure their feeding and nutrition, and enhance cognitive function (2, 3). Invertebrates grow using a variety of extra-pituitary circuits, such as insulin and growth factors. Vertebrates have developed a somatotropic axis, which involves the production of pituitary growth hormone (GH), whose secretion is controlled by hypothalamic stimulatory and inhibitory neuropeptides, including growth hormone-releasing hormone (GHRH) and somatostatin (4, 5). Ghrelin, originating primarily in the stomach, is also an important GH secretagogue (4, 5). The pattern of GH secretion regulates the production of insulin-like growth factor-1 (IGF-1). GH can stimulate IGF-1 production in various tissues, although the liver is responsible for most circulating IGF-1 (6). This achievement coupled the body growth of vertebrates with an enhanced ability to obtain food and reproduce. GH and IGF-1 not only influence body growth, but also various brain cells express receptors for these hormones. So, the GH/IGF-1 axis can regulate numerous brain functions at different periods of development. Conditions that affect the secretion or sensitivity to GH or IGF-1 may influence the progression of different brain or mental disorders. This manuscript will discuss the potential role of GH and IGF-1 in the context of brain and neuropsychiatric diseases.

The brain is a direct target of GH

It has long been known that the brain expresses the GH receptor (GHR) (7). However, a detailed characterization of the distribution of GH-responsive cells in the brain was not possible until recently due to difficulties in identifying GHR-expressing cells. These difficulties arise from either the lack of antibodies that can accurately detect GHR histologically or the low expression of GHR in many neuronal populations. Therefore, some studies have utilized the capability of exogenous GH administration to induce the phosphorylation of signal transducer and activator of transcription 5 (pSTAT5) as a marker for GH-responsive neurons (8, 9). The binding of GH and the consequent activation of GHR recruit Janus kinase proteins, which then phosphorylate STAT5 (10). Subsequently, pSTAT5 is translocated to the cell nucleus and acts as a crucial transcription factor mediating the major cellular effects of GHR signaling (10). Cells that express functional GHRs also express GH-induced pSTAT5. It is important to note that when GH is administered systemically (typically via intraperitoneal or subcutaneous injections), this method requires the transport of GH from the systemic circulation to the brain. A recent study in mice has demonstrated that circulating GH quickly crosses the blood-brain barrier (BBB) and activates GHR in all brain regions. However, the time required to reach each area varies based on its distance from the ventricles (11). Using this approach, the distribution of GH-responsive neurons was described in the brains of both mice (8) and rats (9). Recently, a GHR-reporter mouse model was developed, showing a generally similar distribution of GHR fluorescent cells in the brain compared to GH-induced pSTAT5 (12). Remarkably, GH-responsive neurons are widely distributed across numerous brain areas, including the hippocampus, bed nucleus of the stria terminalis (BNST), amygdala, cerebral cortex, cerebellum, specific thalamic regions, and several brainstem nuclei (8, 9, 12). However, the greatest concentration of GH-responsive cells is found in the hypothalamus, a critical brain structure that regulates visceral functions and body homeostasis.

In recent years, several genetically modified mouse models have been developed to inactivate GHR in large brain areas (e.g., nestin-derived cells or glutamatergic and GABAergic neurons) or specific neuronal populations (13, 14). These studies have demonstrated that brain GHR signaling regulates energy balance, glucose homeostasis, hormone secretion, and responses to metabolic stress situations (15–23). This manuscript will not discuss these functions since more information can be found in recent reviews (13, 14). In the following sections, we will discuss the potential importance of GH and IGF-1 action in the brain in regulating cognitive and neurodevelopmental functions, as well as the association between the somatotropic axis and neurological and psychiatric disorders.

IGF-1 plays a crucial role in brain development and function

IGF-1 receptors (IGF1Rs) are strongly expressed in the brain (24), including developing neurons (25). Additionally, IGF-1 can be produced locally within the nervous system and is developmentally regulated in specific brain regions (26). Both GH and IGF-1 are linked to neurogenesis (27). The crucial role of IGF-1 in brain development is evident in mouse models displaying either overexpression or null mutations in the IGF-1 system (Figure 1). For instance, IGF-1 overexpression leads to brain overgrowth by increasing the number of neurons and oligodendrocytes, while loss-of-function mutations in IGF-1 or IGF1Rs result in brain growth retardation (28) (Figure 2). Therefore, IGF-1 impacts the development and function of neurons and glial cells, including myelination (29, 30). Additionally, IGF-1 may indirectly influence brain function and development by regulating nutrient and blood supply, as local IGF-1 production governs brain glucose uptake and metabolism (31). Furthermore, IGF-1 affects blood vessel remodeling in the adult brain, potentially mediating the angiogenic effects of physical exercise or injury (32).

Figure 1.

Animal studies show evidence indicating that GH, IGF-1, and ghrelin have various effects on the nervous system, significantly impacting brain health and contributing to the development of neurological and psychiatric disorders. Created in BioRender.

Figure 2.

Summary of the key findings from various studies suggesting a possible link between alterations in the GH/IGF-1 axis and neurodegenerative, neuropsychiatric, and neurodevelopmental disorders. Created in BioRender.

IGF-1 regulates the expression and activity of glutamate receptor subunits in the hippocampus, which are essential proteins for memory formation (33, 34). Thus, IGF-1 can mediate the beneficial effects of GH treatment on hippocampal plasticity and learning (35, 36) (Figure 1). Although brain-specific STAT5 knockout mice exhibit memory deficits (37), suggesting a direct effect of the GHR-STAT5 signaling pathway in regulating memory (Figure 2), it is possible that GHR signaling stimulates local IGF-1 production. Indeed, brain-specific STAT5 knockout mice show reduced IGF-1 expression in the brain (37). Due to the effects of GH/IGF-1 treatment on hippocampal function, some animal studies suggested the potential of GH or IGF-1 treatment to mitigate age-related cognitive impairments (34). However, GH- or GHR-deficient mice are remarkably protected against age-induced cognitive decline (38, 39). Considering the importance of GH and IGF-1 for brain functions and development, this apparent paradoxical protection may be explained by increased insulin sensitivity (40) and the prevention of age-related activation of the inflammasome and neuroinflammation (41, 42) observed in these mouse models (Figure 2). Thus, potential deficits caused by GH or IGF-1 deficiency are compensated for by the positive effects of higher insulin sensitivity and reduced systemic and brain inflammation (43). Both GH and IGF-1 treatments are also associated with protective and regenerative effects in various brain injury protocols (44–48). In this context, while chronic GH oversecretion causes neuroinflammation (49), GH treatment following brain injury exhibits an anti-inflammatory effect on the damaged tissue, promoting a regenerative response (47, 48).

The role of GH/IGF-1 in brain disorders: lessons from animal studies

Neurodegenerative diseases

Several studies provide evidence linking the GH/IGF-1 system to the predisposition for neurodegenerative diseases (50–52). In the context of Parkinson’s disease, not only can circulating (liver-derived) IGF-1 affect midbrain dopamine neurons, which express IGF-1 receptors, but midbrain dopaminergic neurons also produce IGF-1 following depolarization (53). IGF-1 regulates dopamine release from these neurons, and the absence of IGF-1 in dopaminergic neurons decreases dopamine levels in the striatum, leading to deficits in dopamine neuron firing, spontaneous locomotion, and exploratory and learning behaviors (53). Additionally, IGF-1 can protect dopamine neurons from oxidative stress (54). A recent study demonstrated that IGF-1 gene therapy prevents cognitive deficits and enhances tyrosine hydroxylase levels in the striatum in a rat model of Parkinson’s disease (55). These effects correlated with an improved anti-inflammatory profile of microglial cells and astrocytes in the striatum (55).

Ghrelin is a potent GH secretagogue, and ghrelin therapy has been proposed for neurodegenerative diseases (56). Consequently, ghrelin receptor (known as GH secretagogue receptor – GHSR) knockout mice exhibit greater susceptibility to dopaminergic neuron degeneration due to impaired autophagy in a Parkinson’s disease experimental model (57). Another study found that treatment with GH for 21 days improved motor function and dendrite morphology of dopaminergic neurons in an experimental model of Parkinson’s disease in rats (58). Low serum IGF-1 is associated with elevated brain amyloid-beta levels, a hallmark of the pathogenesis of Alzheimer’s disease (52) (Figure 2). Moreover, Alzheimer’s patients show insulin and IGF-1 resistance in the brain (50). Thus, impaired brain IGF-1 action appears to be a potential causative factor common to several neurodegenerative diseases.

Depression and anxiety disorder

Mice lacking the Ghrh gene show reduced GH and IGF-1 secretion. Notably, these animals display decreased anxiety and depression-related behaviors (59, 60) (Figure 2). Another mouse model of mild hypopituitarism (nestin-cre mice) also shows reduced hypothalamic GHRH expression, GH secretion, and anxiety (61). However, these models do not allow for the differentiation of effects produced by GH deficiency from those caused by the absence of GHRH signaling. Both GHRH and central somatostatin are associated with mood disorders. Somatostatin exhibits anxiolytic and antidepressant effects when centrally infused in rodents (62, 63). These effects are mediated by the somatostatin-2 receptor (63, 64).

Early life experiences (e.g., enriched environments) reduce anxiety-like behavior in adult rats through IGF-1-dependent mechanisms (65). Another study has shown that GH treatment alleviates anxiety and depression-like behavior in rats subjected to total sleep deprivation for 21 days (66) (Figure 1). Numerous studies indicate that ghrelin diminishes anxiety-like behavior in rodents (67, 68) (Figure 2). Administration of ghrelin, whether central or peripheral, reduces anxiety and depression-like behaviors resulting from chronic, unpredictable, mild stress or forced swimming (68). In agreement, mice that lack ghrelin exhibit increased anxiety-like behaviors (67). The antidepressant-like effects of paeoniflorin depend on the expression of GHSR (69). The anxiolytic effects induced by ghrelin likely involve the expression of ghrelin receptors in the nucleus accumbens and lateral septum, as ghrelin infusion into these brain regions decreases anxiety in mice and rats (70, 71). Simultaneously, blocking GHSR in these nuclei prevents ghrelin’s anxiolytic effects (70, 71). Importantly, previous studies have demonstrated that activation of ghrelin receptors in specific brain regions triggers local GH production (72, 73). Therefore, it remains unclear whether GH serves as a mediator of ghrelin’s effects on anxiety and depression-related behaviors.

Approximately 60% of somatostatin-expressing neurons in the BNST and central nucleus of the amygdala (CEA) respond to a systemic GH injection (74). This finding is noteworthy because somatostatin neurons in the BNST/CEA complex, also known as the extended amygdala, are directly involved in the stress response and anxiety regulation (75–78). To investigate whether GH modulates the function of this somatostatinergic circuitry, mice lacking GHR exclusively in somatostatin-expressing cells were produced. Male knockouts exhibited increased anxiety-like behavior in the open field, elevated plus maze, and light-dark box tests, suggesting that the absence of GHR signaling in somatostatin cells leads to an anxiogenic phenotype in male mice, but not in females (74) (Figure 2). Of note, GHR ablation in somatostatin neurons caused no changes in GH or IGF-1 secretion, body growth, or metabolism (74, 79). Thus, GH appears to have direct anxiolytic effects via somatostatin-expressing neurons. Given that ghrelin is a GH-secretagogue and also exhibits anxiolytic properties, further studies are warranted to determine whether ghrelin-induced GH secretion contributes to the effects of ghrelin on mental health.

Posttraumatic stress disorder

Fear memory is a key feature of posttraumatic stress disorder. The formation and expression of specific contextual fear memory depend on several brain structures, including the hippocampus and some subdivisions of the amygdala. Chronic immobilization stress decreases hippocampal GH expression and impairs hippocampus-dependent behavioral tasks, including contextual fear conditioning (80). The stress-induced defects in hippocampal function are reversed in rats that received an intrahippocampal virus inducing GH overexpression (80).

GHSR is expressed in the hippocampal dentate gyrus (81). Calorie restriction enhances adult hippocampal neurogenesis and fear memory in wild-type animals but not in GHSR knockout mice (81). GHSR is also widely expressed in the amygdala, especially in the basolateral nucleus (82). Repeated stress can serve as a rodent model of posttraumatic stress disorder, which induces increases in circulating ghrelin (72). Meyer et al. (72) observed that ghrelin acts in the amygdala to exacerbate fear learning (Figure 2). Additionally, chronic stress leads to GH overexpression in the amygdala. Pharmacological GHR antagonism blocks the fear-enhancing effects of repeated ghrelin receptor stimulation (72). Virus-induced GH overexpression in the amygdala increases the number of amygdala cells activated by fear memory formation and enhances dendritic spine density in GH-overexpressing neurons (73). Furthermore, virus-induced IGF-1 overexpression in the amygdala promotes the formation and expression of fear memory in rats, causing structural plasticity modifications (83).

As we demonstrated earlier, somatostatin neurons in the extended amygdala respond to GH (74) and are crucial in regulating anxiety (75–78). Additionally, somatostatin neurons in the BNST/CEA complex are implicated in the formation and expression of fear memories (84–87). Thus, fear memory was evaluated in mice with GHR ablation in somatostatin-expressing cells. During an auditory Pavlovian fear conditioning test, both male and female conditional knockouts showed reduced fear memory (74). This finding suggests that GH modulates fear memory by acting directly on specific neuronal populations, including somatostatin-expressing neurons in the extended amygdala (Figure 2). Moreover, ghrelin’s influence on fear memory may also be related to its GH-secretagogue function. Therefore, a ghrelin-GH-IGF-1 axis is activated by chronic stress, enhancing fear memory via amygdala neurons. The overactivation of this circuit could lead to posttraumatic stress disorder.

Association between the somatotropic axis and neurodegenerative diseases in humans

The reduced activity of the somatotropic axis in the elderly, often referred to as somatopause, coincides with the neurocognitive decline associated with aging. This has led to the hypothesis that decreased GH or IGF-1 secretion may contribute to the neurological impairments of aging, though this has never been definitively proven. Several components of the somatotropic system exhibit pleiotropic and neuroprotective effects. GH administration enhances cognitive performance in individuals with GH deficiency (GHD) (88), and GHRH modulates neuronal exosome biomarkers in cases of mild cognitive impairment (89).

Regarding potential therapeutic approaches, the GHRH agonist tesamorelin appears to enhance cognitive function in individuals with mild cognitive impairment (90). Similarly, GHRH agonist MIA-690 demonstrated antioxidant and neuroprotective properties in human cortical cells exposed to amyloid 1 beta (Aβ)1-42 peptide, leading to improved cognitive performance in a mouse model of Alzheimer’s disease (AD) (91). However, intranasal administration of GHRH(1-44)NH2 in humans hindered hippocampal memory learning by suppressing endogenous GHRH (92). In addition to stimulating pituitary GH secretion, GHRH may function similarly to ghrelin, which binds to hippocampal neurons while increasing spine synapse density and enhancing spatial memory (93). Conversely, a ghrelin agonist (MK-677) showed no clinical effect on AD progression (94). These findings suggest the potential therapeutic application of GHRH agonists in neurodegenerative diseases and GHRH antagonists in mood disorders and cognitive dysfunction, including posttraumatic stress disorders (95).

Rivastigmine, a potent oral inhibitor of cerebral acetylcholinesterase, enhances GH release following repeated GHRH stimulation in healthy elderly human subjects (96). GHRH stimulates GH secretion from the pituitary gland by binding to and activating the pituitary type of GHRH receptor (GHRH-R). Splice variants (SVs) of GHRH-R found in extra-pituitary tissues have been identified; among them, splice variant 1 (SV1) exhibits the greatest similarity to GHRH-R and remains functional upon GHRH stimulation, eliciting cAMP signaling and mitogenic activity (97). Phenol-lipoyl hybrids (SV1-13) containing antioxidant moieties and anti-amyloid properties have recently been produced, with SV5 showing moderate antioxidant activity and good neuroprotective effects, warranting further investigation in future experiments for the treatment of neurodegenerative diseases such as Alzheimer’s or Parkinson’s (98).

A study showed that although individuals with Parkinson’s disease have similar IGF-1 levels to healthy controls, those patients in the lowest IGF-1 quartile had lower performance on cognitive tasks assessing executive function, attention, and verbal memory (99). However, recent studies revealed a significant and positive relationship between increased IGF-1 levels and a higher risk of Parkinson’s disease (100–103). Altogether, changes in components of the somatotropic axis likely contribute to the development and progression of neurodegenerative diseases.

Association between the somatotropic axis and neuropsychiatric diseases in humans

Both GH deficiency and excessive GH secretion have been linked to a higher prevalence of depression and anxiety (104–107) (Figure 2). Thus, optimal activity of the somatotropic axis appears to protect against mood disorders. GH oversecretion (acromegaly) is linked to a higher incidence of neuropsychiatric diseases in humans (108–110). The reduction in GH levels following surgical resection of the pituitary tumor in acromegalic patients correlates with improvements in depressive symptoms (107). Other studies investigated the relationship between serum IGF-1 levels and the prevalence of depressive disorders. Women with low IGF-1 (below the 10th percentile) and men with levels above the 90th percentile faced a greater risk of depressive disorder (106). Another study indicated that patients with major depressive disorder exhibit higher serum IGF-1 levels compared to healthy controls, and the pharmacological treatment of depression reduces IGF-1 levels (111). Two meta-analyses assessing this relationship across 9 or 14 studies observed similar results (112, 113). Furthermore, increased diurnal GH secretion was noted in men suffering from major depressive disorder (114). Antidepressant treatments restore GH secretion levels during wakefulness to normal values (115).

A high prevalence of mood disorders is also observed in individuals with GH deficiency (105). Interestingly, Mahajan et al. (104) demonstrated that adult-onset, rather than childhood-onset, GH deficiency is associated with a higher prevalence of atypical depression. Furthermore, two months of GH therapy are sufficient to significantly improve the scores on the depression rating scale, as well as emotional reactions and social isolation (104). GH deficiency is the most common neuroendocrine disorder resulting from mild traumatic brain injuries in soldiers during military service (116, 117). These individuals present a high prevalence of posttraumatic stress disorder symptoms, including cognitive deficits, mood and anxiety disorders, sleep problems, and reduced quality of life (116, 117) (Figure 2). GH replacement therapy effectively improves verbal learning, attention, short-term memory, and symptoms of posttraumatic stress disorder in a small group of individuals with acquired GH deficiency caused by traumatic brain injury (118).

Elevated IGF-1 levels are observed in patients experiencing their first episode of schizophrenia (119, 120) (Figure 2). Furthermore, the severity of symptoms correlates with IGF-1 levels (120). In experiments using human SH-SY5Y neuroblastoma cells, IGF-1 can modulate the expression of genes that are altered in schizophrenia, normalizing their expression levels (121). Additionally, IGF-1 levels are increased in patients with bipolar disorder (112, 113, 122–124) (Figure 2). In contrast, reduced cerebrospinal fluid GH levels are noted in bipolar disorder patients compared to healthy controls (125). It is suggested that elevated circulating IGF-1 levels in individuals with neuropsychiatric diseases might represent a compensatory mechanism to offset potentially low brain IGF-1 expression or a state of central IGF-1 resistance, or even serve to produce IGF-1-induced neuroprotection and angiogenesis in these pathological conditions (123, 124, 126).

Cognitive alterations and neuropsychiatric diseases in lifetime untreated GH-deficient individuals

Individuals with GHD, often related to hypopituitarism, experience a higher prevalence of depression, emotional lability, and decreased energy levels. Importantly, GH therapy provides benefits in quality of life measures (127). However, the impairment of psychological parameters appears to depend on the etiology of GHD and concomitant deficiency of other pituitary hormones, which affect neural cells. Cognitive dysfunction is likely linked to GHD, while diminished emotional well-being and motor performance seem to correlate with other pituitary hormone deficiencies (105, 128, 129). Patients with acquired GHD may face several confounding factors, including varied etiologies of GHD, pituitary surgery, irradiation, anticonvulsant use, and deficiencies of other pituitary hormones coupled with inadequate replacement therapies (130, 131). Isolated GHD (IGHD) is rare and is often treated with GH replacement therapy during childhood. In the Brazilian city of Itabaianinha, a substantial cohort of individuals with severe IGHD resulting from a homozygous (c.57+1G→A) mutation in the GHRHR gene (OMIM n.618157) have been studied (132). Most affected adults have not received GH replacement and demonstrate a normal lifespan (133), with an extended health span (2), indicated by increased insulin sensitivity (134) and elevated levels of glucagon-like peptide-1 (135). Additionally, these individuals exhibit a favorable longevity miRNA profile, specifically an up-regulation of miR-100-5p, miR-195-5p, miR-1981B-5p, and miR-30E-5p, linked to in vitro reductions in the expression of target age-related genes (mTOR, AKT, NFκB, and IRS1) (136).

In this cohort of individuals with lifetime untreated IGHD, those over 50 exhibited similar total cognitive performance but better attention and executive function than controls, as assessed by a battery of psychological tests, specifically the Independent Cognitive Assessment of Literacy (LICA) (137, 138). Additionally, there have been no documented cases of dementia in nearly 30 years of follow-up for this cohort, nor any mention of dementia as a cause of death on the death certificates of IGHD subjects born since 1892 (133) (Figure 2).

We also recently performed brain morphometry and assessed the aging brain using MRI with a trained neural network in these IGHD subjects (139). Most absolute values of cortical thickness and regional brain volumes of the IGHD subjects were similar to those of controls. However, the normalized volumes relative to the intracranial volume of the insula, frontal cortical pole, and caudate are larger in IGHD than in controls (Figure 2), suggesting that the thalamus and putamen may also be larger. These data indicate that the development or maintenance of these regions is less dependent on the function of the somatotropic axis, which is severely blunted in these IGHD subjects. Conversely, the insula forms networks with the prefrontal cortex and the striatum (composed of the caudate nucleus and the lentiform nucleus), specializing in creating and organizing interoceptive representations that support habitual, model-based, and exploratory control of visceral organs and physiological processes, serving a wide variety of functions in humans, from sensory and affective processing to advanced cognition (140). The frontal lobe regulates executive function, which is crucial for the cognitive control of behavior, physical and mental health, and academic and professional performance (141). Interestingly, the normalized volume of the frontal pole is increased in IGHD individuals (139). Additionally, the hypothalamus and frontal and striatal regions are particularly sensitive to insulin, modulating memory, hippocampal, and visual brain areas (142). The increased insulin sensitivity of these IGHD subjects (133), similar to that seen in the GH-resistant Ecuadorian cohort, may contribute to the elevated normalized volumes of these regions in both cohorts (139, 143). Furthermore, IGHD does not seem to impair resilience to brain aging (139).

Somatotropic axis and genetic conditions associated with neurodevelopmental disorders and autism

In 1985, a genetic neurological condition linked to the monosomy of the distal long arm of chromosome 22 was described. This syndrome was later recognized as 22q13.3 deletion syndrome, or Phelan-McDermid syndrome (PMS), which is associated with haploinsufficiency of the SHANK3 gene (144). PMS is a rare condition characterized by hypotonia, developmental delay, absent or severely delayed speech, moderate to severe intellectual disability, dysmorphic features, a higher prevalence of autism, and a wide range of systemic abnormalities (144, 145).

SHANK is a family of essential scaffolding proteins located at the postsynaptic density of glutamatergic neurons, regulating synaptic formation, development, and plasticity (146). Reduced expression of SHANK3 is associated with impaired synapse formation and synaptic integrity. Currently, there is no specific treatment to restore synaptic function or address the neurological manifestations of PMS. As previously mentioned, IGF-1 is a potent growth factor in the central nervous system that significantly influences synapse formation and proliferation in glutamatergic neurons (33, 34). The IGF-1 receptor is expressed in neural cells throughout life, with IGF-binding proteins (IGFBPs) elevated in the developing brain (25, 27, 45). Preclinical studies indicate that recombinant human IGF-1 (rhIGF-1) positively affects synaptic restoration, suggesting that IGF-1 may alter the disease course in individuals with PMS.

Controlled clinical trials lasting 12 weeks with commercially available rhIGF-1 have shown safety and demonstrated benefits for PMS symptoms, improving social withdrawal, repetitive behaviors, hyperactivity, and sensory reactivity (147, 148) (Figure 2). However, rhIGF-1 necessitates inconvenient subcutaneous administration twice daily, close to meals, and involves glucose monitoring. Therefore, recombinant human GH (rhGH) appears to be an alternative to rhIGF-1. rhGH has been utilized for approximately 40 years with daily injections for approved indications and has an excellent safety profile. Open-label clinical trials have shown that rhGH treatment in children with PMS raised serum IGF-1 levels by at least 2 SD from baseline without serious adverse events and produced effects similar to rhIGF-1 on social isolation and other central manifestations of PMS syndrome (149, 150). Clinical trials for rhGH concerning PMS and idiopathic autism are still ongoing (ClinicalTrials.gov, accessed September 17, 2024). A new drug, NNZ-2591, is being developed for PMS. Orally administered NNZ-2591 can enhance IGF-1 function in neurons by competitively binding to IGFBP-3 and normalizing IGF-1 levels in the brain (ClinicalTrials.gov, accessed September 17, 2024) (151). Recently, trofinetide, an analog of IGF-1 that is more resistant to degradation, was approved by the US Food and Drug Administration to treat Rett syndrome, another neurodevelopmental disorder, underscoring the benefits of IGF-1 on neuronal function. Trofinetide appears to promote synaptic maturation and restore dendritic morphology, neuronal signaling, and synaptic protein synthesis to normal (152).

Down syndrome (DS), a genetic condition caused by trisomy 21, is the most common chromosomal disorder, with an incidence of 1-10 in 1,000 live births worldwide, according to the WHO. Individuals with DS have a higher prevalence of autism than the general population (153). A recent review indicated that GH axis activity is impaired in children with DS, and that 87% of them have IGF-1 levels below the 50th centile, while 41% could be considered IGF-1 deficient (154) (Figure 2). The authors suggested that IGF-1 deficiency was not associated with classical GH deficiency but occurred downstream of GH secretion, probably related to inflammatory conditions involving tumor necrosis factor-α. Notably, long-term IGF-1 deficiency was considered the most critical biosignature associated with neurodegeneration and neuroinflammation in individuals with DS.

Therefore, although children with various genetic conditions and neurodevelopmental disorders usually have normal GH secretion as assessed by the classical criteria for diagnosing GH deficiency (155), altered IGF-1 levels or actions in the various compartments of the central nervous system of these patients can significantly influence brain development and function. Thus, administration of GH, IGF-1, or IGF-1 analogs represents a promising therapeutic approach to alleviate the symptoms of some conditions associated with neurodevelopmental disorders and autism spectrum disorder (126).

Insights from human genome variability

Genome-wide association studies (GWAS) identify genetic variants, particularly single-nucleotide polymorphisms (SNPs), associated with traits or diseases by comparing the genomes of thousands of individuals (156, 157). This hypothesis-free approach enables the discovery of previously unknown genetic influences on both case-control outcomes (e.g., neuropsychiatric disorders) and continuous traits (e.g., serum hormone levels) (158). GWAS have revolutionized our understanding of the pathophysiological underpinnings of common disorders of the brain by mapping hundreds of genes involved in these phenotypes (159–163).

GWAS efficiently explores biological associations between traits by utilizing global and local genetic correlation analyses to identify shared biology (i.e., pleiotropy) across different traits. In contrast, Mendelian randomization (MR) methods assess causal relationships by employing genetic variants as proxies for risk factors (instrumental variables). These complementary approaches provide valuable insights into the genetic influences on physiology and complex traits (156, 158). Notably, GWAS have substantial potential to offer new insights into the nature of the connections, whether causal or associative, between components of the GH/IGF-1 axis and neurological and neuropsychiatric phenotypes, helping to elucidate how these hormones may influence mental health.

Circulating IGF-1 concentration is a highly polygenic trait, with thousands of genetic variants contributing to its variance across the genome. This results in an SNP-based heritability of 0.22 (the proportion of phenotypic variance in a trait explained by the additive effects of common genetic variants) (164). The most comprehensive study on this topic examined genetic correlations and potential causal relationships between IGF-1 and 48 other blood-based biomarkers, along with ten psychiatric disorders and related traits (165). These disorders included attention-deficit hyperactivity disorder (ADHD), major depressive disorder, posttraumatic stress disorder, autism spectrum disorder, Tourette syndrome, schizophrenia, bipolar disorder, obsessive-compulsive disorder, and anorexia nervosa. Reay’s study utilized data from an IGF-1 GWAS in the UK Biobank, which comprised a sample of 342,439 individuals (165). Among the neuropsychiatric phenotypes, only ADHD demonstrated a significant negative genetic correlation with IGF-1 (Figure 2 and Figure 3). IGF-1 also exhibited a nominal (significant without correction) non-zero negative correlation with major depressive disorder and posttraumatic stress disorder, along with a positive correlation with obsessive-compulsive disorder. Importantly, the study found no evidence of a causal relationship between IGF-1 and the neuropsychiatric traits investigated, as estimated using latent causal variable (LCV) models. These findings align with IGF-1 measurements in a case-control study of ADHD (166) and further suggest that lower IGF-1 levels observed in ADHD cases are likely a correlate of the disorder rather than a causal factor. Nonetheless, pleiotropic effects of genetic variants on other traits may mediate this relationship, and associations between IGF-1 and other mediating factors could still indicate evidence of causality rather than mere correlation.

Figure 3.

Genetic correlations between IGF-1 levels and psychiatric conditions and related traits. These data were extracted from the supplementary tables in Reay et al. Science Advances (2022 Apr 8;8(14):eabj8969). The analysis included posttraumatic stress disorder (PTSD), attention-deficit/hyperactivity disorder (ADHD), major depressive disorder (MDD), tourette syndrome (TS), autism spectrum disorder (ASD), bipolar disorder (BIP), general cognitive ability (GCA), anorexia nervosa (AN), schizophrenia (SZ), and obsessive-compulsive disorder (OCD). Significant genetic correlations were observed for ADHD (p = 5.07 × 10⁻⁵), PTSD (p = 0.0043), MDD (p = 0.0027), and OCD (p = 0.0024). However, after applying a Bonferroni correction, only the correlation with ADHD remained significant (p = 0.0248).

Another study utilized MR models to investigate the causal relationship between genetically determined variability in IGF-1 levels and AD. While most analyses found no causal association, one MR model (the weighted median) indicated an inverse association between IGF-1 and clinical AD solely in women (167). Although the results were not robust in this instance, they underscore the necessity of exploring sex-specific effects in the relationship between IGF-1 and neurological and neuropsychiatric disorders. Furthermore, as GWAS is a rapidly evolving field, increasing sample sizes in biobank datasets will enhance the power for localized analyses. This encompasses local genetic correlations or causality models that distinguish between horizontal pleiotropy (non-causal relationships) and vertical pleiotropy (causal relationships) effects.

Concluding remarks

In conclusion, numerous interactions exist among the components of the somatotropic system and the nervous system. Neuropsychiatric diseases represent a serious global health issue. The development of new molecules capable of interacting with the various pathways involved in the progression of these disorders poses a significant challenge. In the current manuscript, we provide evidence that GH and IGF-1 can influence the central nervous system in different life course stages. Alterations in the secretion and/or sensitivity to GH and IGF-1 may serve as risk factors for the development of several brain diseases or as their correlates. In this regard, optimal activity of the somatotropic axis is crucial, as both deficiency and excess of these hormones can negatively impact brain functions. However, some consequences may be obscured by indirect effects, such as changes in insulin sensitivity. Additionally, at least part of central ghrelin’s effects can be mediated by GH/IGF-1, as ghrelin stimulates systemic GH secretion and local GH expression in various brain regions. Thus, the brain is a significant target of GH and IGF-1, and changes in the somatotropic system are fundamental to brain health or disease.