Can Digenic, Tri-Allelic Inheritance of Variants in STAR and CYP11A1 Give Rise to Primary Adrenal Insufficiency? A Case Report

An eight-year old South Asian boy presenting with progressive hyperpigmentation was found to have primary adrenal insufficiency (PAI) in the form of isolated glucocorticoid deficiency. Follow up of this boy for nine years, until the age of 17 years showed normal pubertal onset and progression. Molecular evaluation, by targeted next generation sequencing of candidate genes linked to PAI revealed changes in two genes that are intricately linked in the early stages of steroid biosynthesis: compound heterozygous variants in STAR, c.465+1G>A and p.(E99K), plus a heterozygous rs6161 change in CYP11A1. No variants in other known causal genes were detected. The proband's mother was heterozygous for the c.465+1G>A STAR and rs6161 CYP11A1 variants, while the father was homozygous for the p.(E99K) alteration in STAR but wild-type for CYP11A1. Both parents had normal adrenal cortical function as revealed by short Synacthen tests. The STAR variant c.465+1G>A will lead to abnormal splicing of exon 4 in mRNA and the addition of the p.(E99K) variant, predicted damaging by SIFT and CADD, may be sufficient to cause PAI but this is by no means certain given that the unaffected father is homozygous for the latter change. The rs6161 CYP11A1 variant [c.940G>A, p.(E314K)] has recently been demonstrated to cause PAI in conjunction with a severe rare disruptive change on the other allele, however sequencing of the coding region of CYP11A1 revealed no further changes in this subject. We wondered whether the phenotype of isolated glucocorticoid deficiency had arisen in this child due to tri-allelic inheritance of a heterozygous CYP11A1 change along with the two STAR variants each of which contribute a partial loss-of-function burden that, when combined, is sufficient to cause PAI or if the loss-of-function c.465+1G>A combined with the presumed partial loss-of-function p.(E99K) in STAR could be causative.

In Vitro Splicing Assay Proves the Pathogenicity of Intronic Variants in MRAP

Introduction: Familial glucocorticoid deficiency (FGD) is characterised by isolated glucocorticoid deficiency in a patient who retains normal mineralocorticoid production. FGD causing mutations in the MC2R accessory protein, MRAP, are often splice-site or nonsense mutations resulting in a truncated protein. Many of these mutations occur at the canonical donor splice-site of intron 3, where it has been shown previously that c.106 + 2_3dupTA, for example, results in skipping of the first coding exon with unknown consequences at the protein level. Patients and methods: DNA was isolated from three consanguineous individuals diagnosed with early onset FGD (0 - 13 months) with high ACTH and/or low cortisol levels and underwent whole exome sequencing. The proband in family 1 (P1) presented at 13 months and had a hyperpigmented sibling who died in neonatal period due to adrenal failure. Patient 2 (P2), who also had a family history of adrenal insufficiency, was noted to be hyperpigmented at birth with markedly raised ACTH, patient 3 (P3) was noted to have diffuse hyperpigmentation in the early neonatal period and on formal testing at 16m was found to have low serum cortisol. Variants were confirmed using Sanger sequencing and predicted splice-site mutations were investigated using an in vitro splicing assay. Results: Homozygous mutations in MRAP were identified in all three cases which were heterozygous in their parents. Previously described mutations, c.106 + 1delG (chr21:33671388delG; rs1476574441; CD050155) in P1 and c.106 + 2dupT (Chr21: 33671390_91insT; rs761576317; CI118288) in P2 at the canonical donor splice-site of intron 3, were identified, with the former predicted to destroy the splice site and the latter to weaken it. These mutations in vitro resulted in the complete skipping of exon 3, which contains the translational start site, and presumably result in no protein product. A novel homozygous mutation in intron 4, c.206 + 5G>T; (chr21:33679055G>T rs1064796398) was identified in P3, but was not predicted to alter splicing. In vitro, this mutation negates the canonical donor splice site and creates two different alternative sites, both resulting in frameshifts and predicted early termination of the protein (p.Val44fs*50, p.Pro72fs*90). Conclusion: All mutations reported here are predicted to produce no protein, either because the start site is excluded (for c.106 + 1delG and c.106 + 2dupT) or because the transcripts are likely to undergo nonsense mediated decay (for c.206 + 5G>T), resulting in the early onset FGD seen in the patients. Splice prediction protocols, although effective for variants within 2bp of exon/intron boundaries may not predict the true outcome of a base change whereas the splice assay conclusively revealed the effect of all three variants allowing us to assign pathogenicity to them.

Sphingosine 1- Phosphate Lyase Insufficiency Syndrome (SPLIS); A Role in Multiple Endocrinopathies

Sphingosine 1-phosphate lyase insufficiency syndrome (SPLIS) was described in 2017 as a novel condition affecting sphingolipid metabolism. There is a multisystemic phenotype including nephrotic syndrome and primary adrenal insufficiency (PAI) and to a lesser extent ichthyosis, neurological disease and lymphopenia. A proportion of patients also presented with hypothyroidism and hypogonadism. To interrogate the endocrine aspect of the syndrome we reviewed clinical data within our patient cohort with SPLIS and those within the wider literature. To date there have been 45 patients identified with SPLIS with significant associated mortality (n=23/45, 51%; 4 of these in utero). There is no clear genotype-phenotype correlation. Whilst nephrotic syndrome is most prevalent (n=34/45; 76%), a significant proportion of patients (n=27/45, 60%) also presented with glucocorticoid deficiency, some with additional mineralocorticoid deficiency (n=7/27). Five further patients were noted to have adrenal calcifications though biochemistry was not undertaken. Most patients presented with PAI in the first 2 years of life (n=21/27), with the oldest presentation being 11 years of age. Adrenal calcifications are a common finding in those who had documented imaging (n=13/15, 87%). Primary gonadal failure has been reported in 8 male cases, all with concomitant PAI. Presenting features included microphallus (n=7/8) and cryptorchidism (n=8/8), indicating reduced in utero androgen exposure. All who had biochemical evaluation demonstrated raised basal LH and FSH/ exaggerated response to LHRH stimulation, a lack of testosterone response to HCG stimulation and low antimullerian hormone (AMH) levels. To date there are no reports of pubertal delay in female patients, and those of age within our cohort have normal ovarian reserve as evidenced by AMH levels (n=2). Primary hypothyroidism, with mildly raised TSH and low Free T4 is reported in 12 patients. Most did not have goiters and had concomitant PAI and nephrotic syndrome (n=11/12). SPLIS is unique amongst sphingolipid disorders in presenting with significant endocrinopathy. This may be the consequence of the particular sphingolipid signature of the disease and the pathogenic mechanisms need to be explored further. It is clear that endocrine dysfunction needs to be considered at diagnosis and surveillance undertaken to detect evolving disease which could have a significant impact on morbidity and mortality.