Abstract
Adenylate cyclase 3 (ADCY3) gene alterations have been reported to be associated with obesity. However, few patients with homozygous mutations have been described to date and the follow-up procedure and treatment options are unclear. A 10-month-old female presented with increased appetite and weight gain. She was born from a consanguineous marriage. Weight, height, and head circumference measurements and standard deviation scores (SDS) were 19 kg (+6.98 SDS), 82 cm (+3.53 SDS), and 49 cm (+3.07 SDS), respectively. Laboratory tests revealed a fasting glucose level of 103 mg/dL (5.7 mmol/L), insulin level of 25.39 µIU/mL, and homeostatic model assessment for insulin resistance value of 6.43. Whole-exome sequencing revealed a novel, homozygous c.1102G>A (p.Asp368Asn) variant in ADCY3. Her parents and healthy sister were heterozygous for the variant. At the age of 2.5 years, neurodevelopmental delay was observed. At the age of 3.5 years, the patient’s weight, height, and body mass index values were 49.5 kg (+8.16 SDS), 111 cm (+2.59 SDS), and 40.18 kg/m2 (+6.48 SDS), respectively. Signs of Blount disease and acanthosis nigricans were evident, and she had hyperphagia. She was undergoing speech therapy. Homozygous ADCY3 variants may present with early onset, severe obesity, insulin resistance, and neurodevelopmental delay in children. Severe complications may occur, even at young ages. More data in terms of the optimal treatment and follow-up process of these patients are needed.
What is already known on this topic?
Adenylate cyclase 3 (ADCY3) gene alterations have been previously found to be associated with obesity. However, only a small number of cases with homozygous mutations have been reported. Besides early-onset severe obesity, hyperphagia, insulin resistance, hyperlipidemia, anosmia/hypo-osmia and intellectual disability may occur. The follow up and treatment options, especially in affected young children, are still unclear.
What this study adds?
In patients with homozygous ADCY3 mutations, severe obesity and insulin resistance may occur from infancy. These cases should be followed and supported in term of neuromotor developmental delay. Moreover, serious complications of obesity may be exhibited at very young ages. The treatment is challenging, especially in young children, and so more data is needed.
Introduction
Monogenic obesity (MO) is a severe, early onset form of obesity caused by a single gene mutation that leads to dysfunction in the leptin-melanocortin pathway controlling energy balance (1). During the past 30 years, more than a dozen genes have been identified in the leptin-melanocortin pathway and the tyrosine kinase receptor B-brain-derived neurotrophic factor signalling system. However, alterations in previously defined genes account for only 5% to 30% of cases of MO, and melanocortin 4 receptor (MC4R) deficiency is the most common cause of MO (1, 2).
ADCY3 gene alterations were recently found to be associated with severe obesity (1, 3-14). The ADCY3 gene (OMIM*600291) is located on the short arm of chromosome 2 (2p23.3) and encodes the adenylate cyclase 3 enzyme (ADCY3) (13). This protein has a pseudosymmetric structure of two transmembrane and two cytoplasmic domains. Nine isoforms of ADCY3 are expressed in various human tissues, such as adipose tissue and the hypothalamus. ADCY3 catalyses the synthesis of cyclic adenosine monophosphate (cAMP), which plays a role in intracellular signal transduction. In the paraventricular nucleus of hypothalamus, ADCY3 co-localizes with MC4R and inhibition of signaling at the primary cilia of these neurons results in increased body weight (1, 13). ADCY3-cAMP signaling also controls the metabolic processes of carbohydrates and lipids; and appears to regulate the proliferation and differentiation of adipocytes (10). Some anorexigenic peptides, such as glucagon-like peptide-1 (GLP-1), act centrally to control appetite by upregulating cAMP formation (1, 13, 15, 16, 17). Apart from its effects on appetite and body weight, ADCY3 seems to be linked to olfactory signal transduction based on the finding that disruption of ADCY3 causes peripheral and behavioral anosmia (15). To date, only 12 patients with homozygous ADCY3 alterations have been reported in large cohorts, and severe obesity, anosmia or hypo-osmia, hyperlipidemia, and insulin resistance were common features in these individuals (Table 1) (7, 10).
Nevertheless, data on long-term follow-up and treatment strategies in these patients remain limited, despite the well-established link between ADCY3 deficiency and obesity. Here, we report a child with early-onset and severe obesity caused by a novel homozygous ADCY3 mutation. We present follow-up data and discuss potential treatment approaches, with a particular focus on management in young children.
Case Report
A 10-month-old female infant presented with increased appetite and weight gain. She had been born from a consanguineous marriage at 38 weeks of gestation, and her birth weight was 3050 g with a standard deviation score (SDS) of -0.08. Her mother had been diagnosed with insulin resistance and required dietary intervention during pregnancy. The infant’s medical records revealed that her weight at the age of two months and seven months were 6.5 kg (+2.03 SDS) and 14 kg (+5.06 SDS), respectively. She was still breastfeeding and had not yet been successfully transitioned to complementary feeding. No steroids or other medications were being used. Physical examination at the age of 10 months showed a weight, height, and head circumference of 19 kg (+6.98 SDS), 82 cm (+3.53 SDS), and 49 cm (+3.07 SDS), respectively. Her weight age was 5.3 years, weight-for-height centile was 213%, and body mass index (BMI) was 28.26 kg/m2 (+4.81 SDS) (18). Her target height was 155 cm (-1.38 SDS). She was at Tanner stage 1. She was able to sit up without support. The patient’s appearance, together with height, weight, and BMI measurements on a growth chart are shown in Figure 1a, 1b. Laboratory testing found a fasting glucose level of 103 mg/dL (5.7 mmol/L), insulin concentration of 25.39 µIU/mL, and homeostatic model assessment for insulin resistance (HOMA-IR) value of 6.43 (>2.22) (calculated as fasting blood glucose × fasting insulin / 22.5) (19). Thyroid function test results were within the reference ranges. The insulin-like growth factor-1 level was 44 ng/mL (reference range: 40.8-93.6 ng/mL) (20). The basal cortisol level was 5.9 µg/dL, and the peak level stimulated by a low-dose adrenocorticotrophic hormone (ACTH) stimulation test with 1 μg/kg intravenous cosyntropin was 18.4 µg/dL, with an ACTH level of 20.1 pg/mL. A leptin level test could not be performed. The ophthalmologic examination and echocardiography findings were normal.
The patient was diagnosed with MO based on the presence of early onset and severe obesity, hyperphagia, normal height without dysmorphic/syndromic features, birth from a consanguineous marriage, and a family history of insulin resistance and obesity. After an MO panel for known common obesity-related genes was reported as normal, whole-exome sequencing analysis was performed and a novel homozygous c.1102G>A (p.Asp368Asn) variant in ADCY3 was found (Figure 2). This variant had not been previously reported and was classified as a variant of unknown significance with a high pathogenicity score according to the American College of Medical Genetics classification (21).
The patient’s pedigree is shown in Figure 2. Her mother was heterozygous for the same variant c.1102G>A (p.Asp368Asn) and had obesity and insulin resistance. Her father was also a heterozygous carrier of the mutation but was of normal weight. The patient’s brother and sister had been diagnosed with ulcerative colitis, but only the sister was a heterozygous carrier of this variant (Figure 2).
At the age of 2.5 years, the patient’s weight, height, and BMI were 34 kg (+8.42 SDS), 103 cm (+3.06 SDS), and 32.05 kg/m2 (+6 SDS), respectively. Her weight age was 10.3 years (18). She had hyperphagia. She was unable to say two consecutive words. The Denver II Developmental Screening Test revealed that she was delayed in two domains: personal-social (13.5 months) and language (8.5 months). Her audiometry results were within the normal range. She was referred for speech therapy.
One year later at the age of 3.5 years, the patient’s weight, height, and BMI were 49.5 kg (+8.16 SDS), 111 cm (+2.59 SDS), and 40.18 kg/m2 (+6.48 SDS), respectively. Her weight age was 12.9 years (18). Signs of Blount disease and acanthosis nigricans were evident. She continued to be hyperphagic. The patient’s clinical features and height, weight, and BMI measurements are also shown on the growth chart shown in Figure 1. Laboratory testing showed that her fasting glucose level was 86 mg/dL (4.8 mmol/L), fasting insulin level was 20.9 µIU/mL, c-peptide level was 3.27 ng/mL, and HOMA-IR value was 4.44 (19). Thyroid function test results, uric acid and alanine aminotransferase levels, and a lipid profile were within normal ranges. Metformin was initiated at a dose of 250 mg/day, but was promptly discontinued as it had no significant effect on insulin levels or insulin resistance. An operation for Blount disease was planned. Her parents reported that she was able to smell and react to pleasant and unpleasant odors. An odor identification test was planned, to rule out hypo-osmia but was postponed because of speech delay.
The patient’s parents provided written informed consent for publication.
Discussion
ADCY3 catalyses the formation of cAMP and mediates GS signalling from G protein-coupled receptors. It co-localises in the primary cilia of the paraventricular nucleus neurons with MC4R (a type of G protein-coupled receptor), which transduces anorexigenic signals. In addition, cAMP seems to be involved in intracellular signaling of anorexigenic peptides such as GLP-1, and GLP-1 upregulates ADCY3 (1, 3-14).
Specific inhibition of ADCY3 activity in MC4R-expressing neurons was found to be associated with obesity in mice (22). Wang et al. (22) showed that ADCY3-knockout mice may exhibit both severe obesity and hyperphagia, low locomotor activity, and hypogonadism. Several studies in different populations have also supported the association between ADCY3 and severe obesity in humans (3, 4, 5, 7, 8, 9, 11, 12, 13). Three population-based genetic studies have demonstrated that ADCY3 variants are also associated with type 2 diabetes (3, 7, 11). In 2018, patients with homozygous variants in the ADCY3 gene were first reported. Grarup et al. (7) identified seven patients with homozygous c.2433-1G>A variant in a large cohort of the Greenland population (n=4,217). The affected individuals had a 7.3-kg/m2 higher BMI, an 8.1% higher body fat mass, a 17-cm larger waist circumference, and higher fasting glucose and 2-hour plasma glucose concentrations than the rest of the study group. Saeed et al. (10) reported three homozygous mutations in four patients with severe obesity from three unrelated Pakistani families, as well as a compound heterozygous mutation in a Euro-American child. The mutations and phenotypes of previously reported patients and the presented case are summarized in Table 1. The girl presented here had more severe obesity with a higher BMI than previously described patients. Moreover, she had prominent insulin resistance defined at a younger age than two patients reported by Saeed et al. (10) (6 and 15 years of age). Hyperphagia was present in all previously reported patients, as in the current patient. Anosmia was reported in three patients and two others had hypo-osmia (10). The patient described in this report was able to smell but hypo-osmia could not be ruled out because the patient’s age and speech delay prevented the performance of an odor identification test. While neurodevelopmental delay is relatively common in monogenic forms of obesity (MO) (1), it has not been previously documented in association with ADCY3 gene variants. Intellectual disability was reported in two earlier cases; however, detailed information regarding their neurodevelopmental progress was not provided. The association between neurological development and ADCY3 gene alterations is not well defined but an animal model showed that loss of type 3 adenylyl cyclase in mice led to decreased neuronal activity, altered sleep pattern, and depression-like behavior (23). Of note, although the currently presented patient exhibited complications of obesity, including Blount disease at a young age, such findings have not been described in the previously reported cases (10).
No clear follow-up procedure for patients with homozygous ADCY3 mutations has been established. Insulin resistance and complications of obesity, including Blount deformity, were exhibited early in our patient. Furthermore, speech delay was marked and neuromotor developmental delay was a major problem for the patient and her family. Another important feature reported in an animal study was hypogonadism (22). Saeed et al. (10) reported secondary amenorrhoea in the adolescent proband with a normal profile of serum gonadotropins and oestradiol. We suggest close follow-up of these children in terms of obesity and its complications, neurological development, puberty, and other additional features not previously detected.
The mother, father, and sister of the index case were heterozygous carriers of the c.1102G>A (p.Asp368Asn) mutation in ADCY3. The mother had obesity and insulin resistance, but the father and the sister were of normal weight. Some heterozygous carriers of ADCY3 gene variants have previously been reported to not exhibit obesity or insulin resistance (10). Interestingly, the brother and sister of our patient had been diagnosed with ulcerative colitis. Inflammatory bowel diseases have been associated with ADCY3 variants (24). However, only the sister was a carrier. Therefore, we plan to perform further genetic analyses for ulcerative colitis in the brother and sister of our patient.
Although there are previously reported treatments for children with MO, the treatment in patients with ADCY3 variants are not established. Setmelanotide is an MC4R agonist that is 20 times more potent than endogenous melanocortin-stimulating hormone. It was approved in 2020 for the treatment of MO syndromes affecting the proximal leptin signal pathway in adults and children aged ≥6 years (25). The ADCY3 protein plays a role in correct function of MC4R. As ADCY3 does not function correctly in patients with MO due to variants in ADCY3, whether setmelanotide would be effective in this condition remains questionable. However, setmelanotide increases MC4R activity. The MC4R pathway appears to be a modifiable system, and another question is therefore whether it can be effectively used for overcoming and improving ADCY3 dysfunction by regulating food intake and preference. Liraglutide is a GLP-1 analogue approved in children aged >12 years with type 2 diabetes and obesity and is also the subject of ongoing clinical trials in children aged 7-12 years (26). Liraglutide was found to be effective in increasing hepatic AC3 mRNA and protein levels, and serum AC3 levels, in mice, as well as upregulating ADCY3 (17). Liraglutide has also been useful for regulating appetite in diabetic patients with hypothalamic hyperphagia and obesity (16). In one study of adolescents with obesity, the use of liraglutide with lifestyle therapy reportedly led to a significantly greater reduction in the BMI SDS. However, this drug has also been found to be associated with some serious complications, such as pancreatitis (26). Unfortunately, the treatment options for MO are limited and the presented patient was younger than other children treated with these agents. ADCY3 has recently been proposed as a target for anti-obesity agents (27). The main safety concern regarding such agents is the risk of malignancy because the upregulation of ADCY3 may lead to an increase in the tumorigenic potential of cells via activation of the cAMP-response element binding protein pathway (28).
Conclusion
Homozygous ADCY3 variants may lead to very early onset severe obesity, insulin resistance, and neurodevelopmental delay in children. Severe complications may occur in the early stages. More evidence is needed to determine optimal management of these patients beginning at young ages.