A recessive mutation resulting in a disabling amino acid substitution (T194R) in the LHX3 homeodomain causes combined pituitary hormone deficiency

Abstract

Background/aims: Recessive mutations in the LHX3 homeodomain transcription factor gene are associated with developmental disorders affecting the pituitary and nervous system. We describe pediatric patients with combined pituitary hormone deficiency (CPHD) who harbor a novel mutation in LHX3.

Methods: Two female siblings from related parents were examined. Both patients had neonatal complications. The index patient had CPHD featuring deficiencies of GH, LH, FSH, PRL, and TSH, with later onset of ACTH deficiency. She also had a hypoplastic anterior pituitary, respiratory distress, hearing impairment, and limited neck rotation. The LHX3 gene was sequenced and the biochemical properties of the predicted altered proteins were characterized.

Results: A novel homozygous mutation predicted to change amino acid 194 from threonine to arginine (T194R) was detected in both patients. This amino acid is conserved in the DNA-binding homeodomain. Computer modeling predicted that the T194R change would alter the homeodomain structure. The T194R protein did not bind tested LHX3 DNA recognition sites and did not activate the α-glycoprotein and PRL target genes.

Conclusion: The T194R mutation affects a critical residue in the LHX3 protein. This study extends our understanding of the phenotypic features, molecular mechanism, and developmental course associated with mutations in the LHX3 gene.

Fig. 1

DNA analysis reveals a homozygous mutation in the LHX3 gene in patients with CPHD. a Pedigree of the affected family. Filled symbols indicate patients with homozygous genotype. Half-filled symbols denote heterozygous individuals. A line through a symbol indicates that the subject is deceased. The dotted line indicates a family connection four or five generations back from the index patient (IV4). b A 3-Tesla MRI of patient IV4 reveals an absent or severely hypoplastic anterior pituitary gland. The anterior part of the sella is filled with a cystic structure with high signal intensity in non-enhanced T₁-weighted sequence (open arrow) and a reduced signal intensity in T₂-weighted sequences (not shown), most likely corresponding to a cyst with high protein content. The posterior pituitary gland is found in an orthotopic location (arrow). c Coronal view. d, e A T₂-weighted sequence of the cervical spine demonstrates a reduced cervical lordosis with slight kyphosis of the vertebral bodies in segment C2 to C5. f-h DNA sequencing reveals a C to G mutation in the LHX3 gene altering a threonine (Thr) to an arginine (Arg) (f control; g III4 (mother of patient IV4 with heterozygous genotype); h patient IV4).

Fig. 2

A point mutation in the LHX3 gene causes a T194R amino acid substitution in the homeodomain. a Diagram of the key domains of the 397-amino-acid human LHX3a protein. N-terminus = Amino terminus; C-terminus = carboxyl terminus. b Sequence analyses demonstrate that T194 is conserved in LHX3/LIM3 proteins. Alignments of the homeodomain primary amino acid sequence are shown. Dashes denote gaps introduced to optimize alignments; dots denote amino acid identity at that position; the box shows the equivalent amino acids for T194. Comparisons are of human (HsLHX3) to chimpanzee (PtLHX3), rhesus macaque (MmLHX3), rat (RnLHX3), dog (CfLHX3), mouse (MuLHX3), pig (SsLHX3), cow (BtLHX3), opossum (DmLHX3), chicken (GgLIM3), Xenopus laevis (XlLIM3), Xenopus tropicalis (XtLIM3), zebrafish (DrLIM3), Drosophila melanogaster (DmLIM3), and the patient. c The threonine at position 194 in LHX3 is conserved in human LIM-HD class transcription factor proteins. Alignments of the homeodomain primary amino acid sequence are shown. Dashes denote gaps introduced to optimize alignments; dots denote amino acid identity at that position; the box shows the equivalent amino acids for T194. d Generation of radiolabeled WT and T194R LHX3 proteins from cDNA expression vectors by in vitro transcription/translation using rabbit reticulocyte lysates in the presence of ³⁵S-methionine. Labeled products were separated by SDS electrophoresis and dried gels were visualized by fluorography. The migration positions of protein standards (in kDa) are shown. Control = Reaction from a lysate programmed with empty vector (negative control).

Fig. 3

Structural prediction of LHX3 homeodomain/DNA interaction. A model of LHX3 interaction with target DNA was performed using the engrailed homeodomain/DNA complex as a template [22]. a Ribbon model of WT LHX3 homeodomain/DNA showing the three α-helices in orange (helix 1), green (helix 2) and purple (helix 3). b Predicted WT LHX3 homeo-domain with a threonine at position 194 (Thr194). c, d Predicted T194R LHX3 homeodomain with an arginine at position 194 (Arg194). Twenty-six different rotamers for this residue are possible. In all cases, except rotamer 21, these result in steric hindrances with other amino acids (mainly with glutamine 170 [Gln170] and leucine 172 [Leu172]). The figure shows two of the R194 rotamers (green side chains) that destabilize the tertiary structure by causing steric hindrance (purple dotted lines) with glutamine 170 and leucine 172 (orange side chains).

Fig. 4

The LHX3 T194R protein does not bind to LHX3 DNA recognition elements. EMSA experiments were performed using the WT or T194R LHX3 proteins translated in vitro in rabbit reticulocyte lysates and radiolabeled probes representing the LHX3 consensus-binding site [23] (a) or theαGSU promoter pituitary glycoprotein-binding element (b). Unprogrammed lysates were used as negative controls (Control). The LHX3/DNA complexes are denoted by arrowheads. F = Free probe.

Fig. 5

The LHX3 T194R protein is unable to activate pituitary target genes. a Expression vectors for WT and T194R LHX3 proteins were transiently cotransfected into mouse pituitary GHFT1 cells with a luciferase reporter gene under the control of theαGSU promoter. Promoter activity was assayed by measuring luciferase activity 48 h after transfection. Negative controls (Control) received equivalent amounts of empty expression vector plasmid. Activities are mean (light units/10 s/μg total protein) of triplicate assays ± SEM. A representative experiment of at least three experiments is depicted. b Similar experiments to those depicted in a were performed using a prolactin (PRL) promoter/enhancer reporter gene.