Clinical Endocrinology (2014)

doi: 10.1111/cen.12637

ORIGINAL ARTICLE

The role of the sonic hedgehog signalling pathway in patients with midline defects and congenital hypopituitarism L.C. Gregory*, C. Gaston-Massuet†,1, C.L. Andoniadou‡,1, G. Carreno§, E.A. Webb*, D. Kelberman¶, M.J. McCabe*, L. Panagiotakopoulos*, J.W. Saldanha**, H.A. Spoudeas††, J. Torpiano‡‡, M. Rossi§§, J. Raine§§, N. Canham¶¶, J.P. Martinez-Barbera§ and M.T. Dattani* *Genetics and Epigenetics in Health and Disease Section, Genetics and Genomic Medicine Programme, UCL Institute of Child Health, †Centre for Endocrinology, William Harvey Research Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, John Vane Science Centre, ‡Craniofacial Development and Stem Cell Biology Division, King College London, §Neural Developmental Unit, UCL Institute of Child Health, ¶Birth Defects Research Centre, UCL Institute of Child Health, **Division of Mathematical Biology, National Institute for Medical Research, ††Department of Endocrinology, Great Ormond Street Hospital for Children, London, ‡‡Department of Paediatrics, Mater Dei Hospital, Msida, Malta, §§Department of Endocrinology, The Whittington Hospital, London, and ¶¶North West Thames Regional Genetics Service, Northwick Park Hospital, Harrow, UK

Summary Introduction The Gli family of zinc finger (GLI) transcription factors mediates the sonic hedgehog signalling pathway (HH) essential for CNS, early pituitary and ventral forebrain development in mice. Human mutations in this pathway have been described in patients with holoprosencephaly (HPE), isolated congenital hypopituitarism (CH) and cranial/midline facial abnormalities. Mutations in Sonic hedgehog (SHH) have been associated with HPE but not CH, despite murine studies indicating involvement in pituitary development. Objectives/Methods We aimed to establish the role of the HH pathway in the aetiology of hypothalamo-pituitary disorders by screening our cohort of patients with midline defects and/or CH for mutations in SHH, GLI2, Shh brain enhancer 2 (SBE2) and growth-arrest specific 1 (GAS1). Results Two variants and a deletion of GLI2 were identified in three patients. A novel variant at a highly conserved residue in the zinc finger DNA-binding domain, c.1552G > A [pE518K], was identified in a patient with growth hormone deficiency and low normal free T4. A nonsynonymous variant, c.2159G > A [p.R720H], was identified in a patient with a short neck, cleft palate and hypogonadotrophic hypogonadism. A 26
6 Mb deletion, 2q12
3-q21
3, encompassing GLI2 and 77 other genes, was identified in a patient with short stature and impaired growth.

Correspondence: Mehul T. Dattani, Genetics and Epigenetics in Health and Disease Section, Genetics and Genomic Medicine Programme, UCL Institute of Child Health, 30 Guilford Street, London WC1N 1EH, UK. Tel.: +44 207 905 2657; Fax: +44 207 404 6191; E-mail: [email protected] 1

These authors contributed equally to this study.

© 2014 John Wiley & Sons Ltd

Human embryonic expression studies and molecular characterisation of the GLI2 mutant p.E518K support the potential pathogenicity of GLI2 mutations. No mutations were identified in GAS1 or SBE2. A novel SHH variant, c.1295T>A [p.I432N], was identified in two siblings with variable midline defects but normal pituitary function. Conclusions Our data suggest that mutations in SHH, GAS1 and SBE2 are not associated with hypopituitarism, although GLI2 is an important candidate for CH. (Received 23 July 2014; returned for revision 22 August 2014; finally revised 11 September 2014; accepted 13 October 2014)

Introduction The sonic hedgehog signalling pathway (HH) is critical for embryonic development of the CNS. Sonic hedgehog (Shh) binds to its receptor Patched, leading to the activation/repression of target genes via the Gli family of zinc finger transcription factors.1 GLI family zinc finger 2 (Gli2) is an obligate mediator of Shh signal transduction, essential for early pituitary and ventral forebrain development in mice.2 Holoprosencephaly (HPE) is a common malformation of the human forebrain3 characterized by its failure to divide, with 80% of patients having associated midline facial abnormalities that in turn correlate to the severity of brain malformation.4,5 Various forms of HPE have been identified: alobar, semi-lobar and lobar, decreasing in severity, with alobar often resulting in major deformities such as cyclopia and lobar often presenting near normal phenotypes.5 Studies have shown that Gli2/Shh signalling in the ventral diencephalon regulates genetic cascades critical for patterning/ growth during early pituitary development, consistent with 1

2 L. C. Gregory et al. mutations in Shh/Gli2 potentially leading to disordered pituitary development.6 GLI2 mutations are associated with variable HPE phenotypes, including bilateral cleft lip/palate, microcephaly, single central incisor, postaxial polydactyly, optic nerve hypoplasia, and an absent/hypoplastic pituitary with panhypopituitarism or isolated growth hormone deficiency (GHD).7 Mouse and human studies have identified numerous transcription factors/signalling molecules that, when mutated, disrupt the development of the primordial Rathke’s pouch during early embryogenesis and may be associated with congenital hypopituitarism (CH) leading to variable hormone deficiencies in human patients.8 In addition to HPE, GLI2 mutations have been implicated in CH in the presence/ absence of cranial/midline facial abnormalities, with digital anomalies often present.9,10 Previous studies have shown that patients specifically with truncating GLI2 mutations frequently had both pituitary anomalies and polydactyly with a common facial phenotype (midface hypoplasia, cleft lip/palate and hypotelorism).11 No SHH mutations have been reported in CH despite murine studies implicating Shh in early hypothalamo-pituitary development.12 Shh is expressed throughout the oral ectoderm but not Rathke’s pouch (RP), and Shh-null mutants manifest loss of expression of ventral transcription factors leading to pituitary hypoplasia, with the absence of thyroid-stimulating hormone (TSH) and GH-producing cells.12 A highly conserved regulatory region, Shh brain enhancer 2 (SBE2), 460 kb upstream of SHH, regulates Shh expression. A pathogenic nucleotide substitution in SBE2 was previously identified in a cohort of patients with midline defects (n = 474), resulting in a loss of Shh expression in the hypothalamus of transgenic mice.13 These data suggest a role for SBE2 in the aetiology of other midline craniofacial/hypothalamo-pituitary disorders, for example septo-optic dysplasia (SOD). Growth-arrest-specific 1 (GAS1) is a membrane-bound glycoprotein with antagonistic effect on Shh signalling. GAS1 acts as a cell membrane receptor and positive regulator of SHH and controls multiple events during development and homoeostasis by direct interaction with SHH.14 HPE-associated phenotypes are induced in Gas1/ mice, characterized by mid-facial hypoplasia, premaxillary incisor fusion, severe ear defects and cleft palate.15 Variants in GAS1 have been identified previously in variable HPE patients with/without SHH mutations.16 Other studies have demonstrated that GAS1 sequence variants identified in HPE patients can impair its physical interaction with SHH.14 We aimed to investigate the role of SHH signalling in our cohort with complex midline craniofacial/hypothalamo-pituitary disorders including HPE-like phenotypes, by screening GLI2, SHH, SBE2 and GAS1 for mutations. We identified two heterozygous variants and a deletion in GLI2 in three unrelated patients with variable phenotypes, and a heterozygous variant in SHH in two siblings with HPE-related phenotypes, but normal hypothalamo-pituitary function.

Materials and methods Patients All patients underwent pituitary function assessment by baseline measurement of early morning cortisol, free thyroxine (FT4) and TSH, prolactin, insulin-like growth factor-1 (IGF-1), and insulin-like growth factor binding protein-3 (IGFBP3); glucagon stimulation testing and GnRH testing was performed as clinically indicated. Peak GH response to glucagon testing ≥6
7 lg/l was defined as normal. Hormone concentrations were measured using standard immunoassays (TSH, FT4, prolactin, cortisol: chemiluminescent microparticle immunoassay; IGF-1, IGFBP3 and GH: Solid-phase, enzyme labelled chemiluminescent immunometric assay). Standard magnetic resonance imaging (MRI) of the brain (including 3-mm slices through the hypothalamopituitary axis) was performed. Height velocity was calculated at baseline, over a period of at least 4 months, to determine annual growth rate in cm/year. Cohorts DNA was extracted from blood samples taken from a total of 450 patients (M:F 1
1:1) with SOD/HPE/CH, recruited from national/international centres between 1998 and 2011: HPE n = 62, severe SOD (characterized by the classical triad of midline forebrain, eye and pituitary defects) n = 36, mild SOD (two of the three triad features present) n = 310, and isolated CH n = 42. Ethical committee approval was obtained from the UCL Institute of Child Health/Great Ormond Street Hospital for Children Joint Research Ethics Committee, and informed written consent was obtained from patients and/or parents. Direct sequencing analysis Ninety-six patients (severe SOD n = 36, HPE n = 18 and isolated CH n = 42) were screened for SHH (ENST00000297261), GLI2 (ENST00000361492) and GAS1 (ENST00000298743) mutations. An additional 44 HPE patients were screened for SHH mutations. The large SOD cohort n = 346 (including the 36 severe cases) was screened for SBE2 mutations. Detailed PCR and sequencing conditions are available upon request. For any mutations identified, control databases were consulted including 1000 Genomes (www.1000genomes.org), dbSNP (www.ncbi.nlm. nih.gov/SNP/) and Exome Variant Server (EVS) evs.gs.washington. edu/EVS/). A CGH-microarray, using Agilent 8x60k 60mer oligo ISCA design (024612) analysis, was conducted on DNA from patient 3. Functional studies Cell culture. NIH3T3 cells (ATCC) were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% calf serum (PAA) in the presence of penicillin/ streptomycin (Invitrogen, Life Technologies Ltd, Paisley, UK). Cells were passaged at 80% confluency. © 2014 John Wiley & Sons Ltd Clinical Endocrinology (2014), 0, 1–11

SHH pathway in congenital hypopituitarism 3 Preparation of constructs for qualitative analysis. Mutations were introduced into full length GLI2 in the pENTR223.1 vector (OHS5893-101547024; Invitrogen) by site-directed mutagenesis. Mutant and wtGLI2 coding regions were cloned into mammalian expression vector pCMV-Sport6 (Invitrogen) and verified by direct sequencing analysis. Cell transfection and luciferase assays. NIH3T3 cells were seeded onto 24-well plates at 4 9 104 cells/well, 24 h prior to transfection. Transient transfection with 100 ng of mutant and wtGLI2 constructs, respectively, alongside GBS-Luc, a firefly luciferase reporter construct containing eight repeats of the Glibinding sequence,17 was conducted using FuGene6 transfection reagent (Roche, Promega Ltd, Southampton, UK). The cells were cotransfected with a Renilla luciferase reporter to normalize (a)

(b)

(e)

(c)

transfection efficiency. Cells were lysed after 24 h and luciferase activities measured using the Dual Luciferase Reporter Assay System (Promega). Results are shown as means  SD of three independent experiments in triplicate (Fig. 2f). Western blot analysis. Western blot analysis was performed as previously described18 using protein derived from cell lysates, rabbit polyclonal Gli2 antibody (VWR International Ltd, Lutterworth, UK), goat anti-rabbit IgG HRP-linked secondary antibody (Cell signalling) and green fluorescent protein (GFP) as a negative control. In vitro translation and electrophoretic mobility shift assay (EMSA). Three separate vector in vitro translation (IVT) mixes were generated containing 500 ng of empty vector, wtGli2 or (d)

(f)

(g)

Fig. 1 (a–c) Mutations in the SHH pathway associated with midline forebrain defects. Ninety-six hypopituitary patients were screened for GLI2 mutations. A novel heterozygous missense mutation (c.1552G>A) was identified in exon 10 in a Caucasian female patient (patient 1) (a – shown by ‘N’), and a heterozygous missense mutation, (c.2159G>A), was found in exon 12 in a female patient of Maltese origin (patient 2) (b – shown by ‘N’). The 96 hypopituitary patients and an additional 44 holoprosencephaly patients were screened for SHH mutations. A novel heterozygous missense mutation (c.1295T>A) was identified in exon 3 in two Caucasian siblings (Patients 4 and 5) with variants of HPE (c – shown by ‘N’). (d–f) The conservation of GLI2 and SHH variants. (d) The glutamic acid residue (represented by the green ‘E’) at location p.E518 is highly conserved between multiple species and is substituted by lysine in Patient 1. (e) The amino acid arginine at location p.R720 (represented by the green ‘R’) is highly conserved between multiple species and is substituted by histidine in Patient 2. (f) The isoleucine amino acid (represented by the green ‘I’) at position p.I432 is highly conserved between multiple species and is substituted by asparagine in Patient 3. (g) A schematic diagram of known GLI2 mutations and their location on the gene. The locations of known GLI2 mutations are shown by arrows. Our two newly identified variants are shown in red. © 2014 John Wiley & Sons Ltd Clinical Endocrinology (2014), 0, 1–11

SHH pathway in congenital hypopituitarism 5 Table 1. Clinical phenotype of affected individuals Patient

1

2

3

4

5

Variant Sex Birthweight in kg (SDS) Age at presentation (years) Height in cm at presentation (SDS)

GLI2 (p.R720H) Female 2
45 (2
2) 16
5 135 (4
6)

2q12
3-q21
3 Female 2
14 (1
9) 2
2 73 (4
6)

SHH (p.I432N) Male 1
93 (4
3) 11 125
2 (2
7)

SHH (p.I432N) Male 2
33 (2
2) 2
9 95
6 (0
3)

0
5 25
8 (6
3)

0
5 9
02 (3)

0
5 18
5 (5
1)

0
5 13
6 (0
5)

Growth velocity in cm/yr (SDS) Hearing

GLI2 (p.E518K) Female 3
87 (+1) 2
5 (Re-presented 14
4) 74
5 (4
4) (Represented: 135
8 (3
3)) 0
5 9
8 (2
4) (Represented: 50
1 (0
2)) 5
8 (1
8) Normal

4
9 (2
4) Normal

1
7 (4
5 SDS) Normal

6
8 (0
6) Normal

Vision Additional features

Normal None

3
1 (4
5) Bilateral sensorineural hearing loss Normal Cleft palate, short neck, cervical vertebral fusion

Normal Cleft lip and palate, developmental delay

Normal Solitary median maxillary central incisor

Current age (years) Current height in cm (SDS)

22 152 (1
9)

29
5 142
9 (3
4)

18
5 169
6 (1
1)

12
5 152
8 (0
2)

Fasting glucose (mmol/l) Thyroxine (normal: 9–19 pmol/l) TSH (normal: 0
35–4
94 mIU/l) Prolactin (73
0–557
0 mIU/l) Sodium (mmol/l) Plasma osmolality (mosm/kg) Urine osmolality (mosm/kg) Oestradiol

2
2 9
8 1
4 167 142 N/A N/A 230

2
8 18
9 0
6 101 137 292 1258 N/A

N/A 17
7 (N 12–32) N/A 270 N/A N/A N/A N/A

IGF-1 ng/ml (NR) IGFBP-3 lg/ml (NR) 0800 Cortisol/peak on glucagon stimulation (nmol/l) Basal LH (IU/l) Basal FSH (IU/l) Peak LH (IU/l) Peak FSH (IU/l) GH peak (lg/l)

12
2 nmol/l (15
1–46
5) N/A 696

4
17 14
2 (NR 10
3–24
5) 2
34 (NR 0
4–4
0) 426 (NR 40–530) 140 295 N/A