Novel and recurrent PITX3 mutations in Belgian families with autosomal dominant congenital cataract and anterior segment dysgenesis have similar phenotypic and functional characteristics

Abstract

Background: Congenital cataracts are clinically and genetically heterogeneous with more than 45 known loci and 38 identified genes. They can occur as isolated defects or in association with anterior segment developmental anomalies. One of the disease genes for congenital cataract with or without anterior segment dysgenesis (ASD) is PITX3, encoding a transcription factor with a crucial role in lens and anterior segment development. Only five unique PITX3 mutations have been described, of which the 17-bp duplication c.640_656dup, p.(Gly220Profs*95), is the most common one and the only one known to cause cataract with ASD. The aim of this study was to perform a genetic study of the PITX3 gene in five probands with autosomal dominant congenital cataract (ADCC) and ASD, to compare their clinical presentations to previously reported PITX3-associated phenotypes and to functionally evaluate the PITX3 mutations found.

Methods: Sanger sequencing of the coding region and targeted exons of PITX3 was performed in probands and family members respectively. Transactivation, DNA-binding and subcellular localization assays were performed for the PITX3 mutations found. Ophthalmological examinations included visual acuity measurement, slit-lamp biomicroscopy, tonometry and fundoscopy.

Results: In four Belgian families with ADCC and ASD the recurrent 17-bp duplication c.640_656dup, p.(Gly220Profs*95), was found in a heterozygous state. A novel PITX3 mutation c.573del, p.(Ser192Alafs*117), was identified in heterozygous state in a Belgo-Romanian family with a similar phenotype. Functional assays showed that this novel mutation retains its nuclear localization but results in decreased DNA-binding and transactivation activity, similar to the recurrent duplication.

Conclusions: Our study identified a second PITX3 mutation leading to congenital cataract with ASD. The similarity in phenotypic expression was substantiated by our in vitro functional studies which demonstrated comparable molecular consequences for the novel p.(Ser192Alafs*117) and the recurrent p.(Gly220Profs*95) mutations.

Figure 1

Sequence electropherograms of the identified PITX3 mutations. A. Sequence electropherograms of the four index patients with the recurrent c.640_656dup PITX3 mutation. B. Sequence electropherogram of the proband with the novel PITX3 mutation, c.573del. The reference sequence is displayed in blue on the top of each panel. Sequence electropherograms of the wild-type are shown at the bottom of each panel. The mutations are indicated with a red box.

Figure 2

Schematic representation of mutated PITX3 proteins. The top diagram represents the wild-type PITX3 protein. The green box represents the homeodomain of 60 amino acids and the blue box the OAR (named after otp, aristaless and rax) domain of 14 amino acids. The middle and bottom diagram represent the mutant PITX3 proteins p.(Gly220Profs*95) (recurrent) and p.(Ser192Alafs*117) (novel), respectively. The positions of the mutations are indicated with a red line and the resulting aberrant protein segments are highlighted by a red box.

Figure 3

Pedigrees of Families 1 (A), 2 (B), 3 (C), 4 (D) and 5 (E). Affected individuals are indicated by filled symbols and index patients by an arrow. In the genotype M stands for the mutant allele (c.640_656dup in Families 1–4 and c.573del in Family 5) and + for the wild-type allele.

Figure 4

Clinical pictures. In this figure, anterior segment pictures are presented for both right (RE) and left eyes (LE). All eyes shown are pseudophakic, except for family 4 (panel D). A. Both eyes of the father (Family 1, II:1) show corneas with small diameter (10 mm) and normal transparency. An anterior chamber lens implant is observed with iris atrophy and iatrogenic correctopia. The index patient (III:1) underwent a rotational corneal autograft in the RE. Corneal clouding, iridocorneal adhesions and iris atrophy can be seen in the LE. Both eyes of the brother (III:2) show inferior corneal clouding with endothelial haze. B. In both eyes of the index patient (Family 2, III:1) corneal opacities can be observed. C. In the RE of the index patient (Family 3, IV:2) optical iridectomy was performed because of severe central corneal clouding. In the LE corectopia and corneal haze is observed. D. Iridocorneal adhesions and iris hypoplasia are observed in both eyes of the index patient (Family 4, III:1) although more pronounced in the RE. E. In the index patient (Family 5, III:1) and her brother (III:2) microcornea (6 mm) and severe corneal opacities are observed.

Figure 5

Subcellular localization (A), DNA-binding (B) and transactivation activity (C) of the novel PITX3 mutation in comparison with the recurrent mutation. A. Immunocytochemistry of the PITX3 wild-type and mutant proteins transfected in B3 lens epithelial cells. Cells were stained with monoclonal anti-FLAG M2 primary antibody and Alexa Fluor 568 donkey anti-mouse IgG as a secondary antibody (red); Hoechst 33342 was used as a nuclear counterstain (blue). All three proteins localize predominantly to the nucleus. B. Results of EMSA experiments demonstrating the DNA-binding capacity of the PITX3 wild-type and mutant proteins. The profile of the wild-type PITX3 protein (third from the left) displays two major bands likely representing DNA interactions with PITX3 homo- or hetero-dimers (upper band; arrow) or PITX3 monomers (lower band; dashed arrow). The profile of the mutant PITX3 proteins only shows the lower band. Addition of anti-PITX3 antibody results in the appearance of a supershifted band (asterisk) and disappearance of the other bands. Western blot (bottom) shows similar levels of PITX3 wild-type and mutant proteins in the nuclear extracts used for EMSA experiments. C. Luciferase assay results for wild-type and mutant PITX3 co-transfected with the bcd-TK-luc or MIP656-bcd1,2-pGL3 reporters in human lens epithelial cells. The values are reported as fold changes of luciferase activity in comparison to an empty vector (pcDNA3.1). All luciferase activities were normalized to β-galactosidase activity and error bars indicate the standard deviation of two independent experiments performed in triplicate. The asterisk indicates the statistically significant differences in fold change (p < 0.001).