Molecular Vision 2017; 23:179-184
<http://www.molvis.org/molvis/v23/179>
Received 13 December 2016 |
Accepted 28 March 2017 |
Published 30 March 2017
Allie Simpson,1,2 Armin Avdic,3 Ben R. Roos,1,2 Adam DeLuca,1,2 Kathy Miller,1,2 Michael J. Schnieders,2,3 Todd E. Scheetz,1,2 Wallace L.M. Alward,1,2 John H. Fingert1,2
1Department of Ophthalmology, Carver College of Medicine, University of Iowa, Iowa City, IA; 2Stephen A. Wynn Institute for Vision Research, University of Iowa, Iowa City, IA; 3Department of Biochemistry, Carver College of Medicine, University of Iowa, Iowa City, IA
Correspondence to: John H. Fingert, Department of Ophthalmology and Visual Sciences Carver College of Medicine, University of Iowa, 3111B MERF, 375 Newton Road, Iowa City, IA 52242: Phone: (319) 335 7508; FAX: (877) 434 9041; email: john-fingert@uiowa.edu
Purpose: Lacrimo-auriculo-dento-digital (LADD) syndrome is an autosomal dominant disorder displaying variable expression of multiple congenital anomalies including hypoplasia or aplasia of the lacrimal and salivary systems causing abnormal tearing and dry mouth. Mutations in the FGF10, FGFR2, and FGFR3 genes were found to cause some cases of LADD syndrome in prior genetic studies. The goal of this study is to identify the genetic basis of a case of LADD syndrome with glaucoma and thin central corneal thickness (CCT).
Methods: Whole exome sequencing was performed, and previously described disease-causing genes (FGF10, FGFR2, and FGFR3) were first evaluated for mutations. Fifty-eight additional prioritized candidate genes were identified by searching gene annotations for features of LADD syndrome. The potential pathogenicity of the identified mutations was assessed by determining their frequency in large public exome databases; through sequence analysis using the Blosum62 matrix, PolyPhen2, and SIFT algorithms; and through homology analyses. A structural analysis of the effects of the top candidate mutation in tumor protein 63 (TP63) was also conducted by superimposing the mutation over the solved crystal structure.
Results: No mutations were detected in FGF10, FGFR2, or FGFR3. The LADD syndrome patient’s exome data was searched for mutations in the 58 candidate genes and only one mutation was detected, an Arg343Trp mutation in the tumor protein 63 (TP63) gene. This TP63 mutation is absent from the gnomAD sequence database. Analysis of the Arg343Trp mutation with Blosum62, PolyPhen2, and SIFT all suggest it is pathogenic. This arginine residue is highly conserved in orthologous genes. Finally, crystal structure analysis showed that the Arg343Trp mutation causes a significant alteration in the structure of TP63’s DNA binding domain.
Conclusions: We report a patient with no mutations in known LADD syndrome genes (FGF10, FGFR2, and FGFR3). Our analysis provides strong evidence that the Arg343Trp mutation in TP63 caused LADD syndrome in our patient and that TP63 is a fourth gene contributing to this condition. TP63 encodes a transcription factor involved in the development and differentiation of tissues affected by LADD syndrome. These data suggest that TP63 is a novel LADD syndrome gene and may also influence corneal thickness and risk for open-angle glaucoma.
Lacrimo-auriculo-dento-digital (LADD) syndrome (OMIM 149730), also known as Levy-Hollister syndrome, is an autosomal dominant disorder displaying variable expression of multiple congenital anomalies. Levy first described LADD syndrome in 1967 as an isolated case with an absent lacrimal system, dry mouth, malformed ears, and dental, arm, and digital abnormalities [1]. Tearing is a central feature in LADD syndrome due to abnormalities of the lacrimal system that may range from hypoplasia to aplasia of puncta, ducts, and glands. Hypoplasia or aplasia of the salivary glands and ducts may also cause dry mouth. Patients with LADD syndrome have external ear abnormalities that include low-set, cup-shaped auricles. Sensorineural, conductive, and mixed hearing deficits may also be present [2]. Dental anomalies in LADD syndrome include microdontia, hypodontia, and enamel dysplasia. Finally, patients with LADD syndrome may have a variety of distal limb and digit abnormalities, such as hypoplastic thumbs and radii, lateral or medial bending of the digits (clinodactyly), and webbing or fusion of the digits (syndactyly) [3,4]. Most LADD cases have ocular involvement. Lacrimal system defects are present in 71% of patients with LADD syndrome and subsequent tear deficiency and dry eye may cause chronic keratoconjunctivitis or corneal ulcerations in 64% of cases [5]. Limbal stem cell deficiency has also been associated with LADD syndrome, which may predispose the patient to corneal epithelial erosions, neovascularization, and hypoesthesia [5,6].
We recently described a 30-year-old male patient that had typical features of LADD syndrome as well as thin corneas and open angle glaucoma [7]. He had classic abnormalities of his lacrimal system, ears, teeth, and fingers. He also had limbal corneal stem cell deficiency as has been previously reported [5]. This patient had additional corneal abnormalities and open angle glaucoma, which have not been previously identified as part of LADD syndrome. His central corneas were very thin at 387 μm right eye (OD) and 334 μm left eye (OS) and he also had severe irregular astigmatism and corneal neovascularization. Moreover, this LADD syndrome patient had classic signs of open angle glaucoma including glaucomatous optic neuropathy and corresponding visual field defects [7].
Genetic studies of several LADD syndrome pedigrees led to the identification of disease-causing mutations in the fibroblast growth factor receptor 2 (FGFR2) gene [2]. Linkage analysis of these families mapped a LADD syndrome gene to chromosome 10q26, which encompassed the FGFR2 gene. DNA sequencing subsequently detected FGFR2 mutations in three LADD syndrome pedigrees and in one sporadic LADD syndrome case. However, no FGFR2 mutations were identified with genetic studies of two additional LADD syndrome pedigrees, indicating that this condition is heterogeneous and may be caused by more than one gene. These pedigrees were consequently tested for mutations in other genes in the FGFR gene family that are related to FGFR2. A mutation in the fibroblast growth factor receptor 3 (FGFR3) gene was discovered in one pedigree, while a mutation in the fibroblast growth factor 10 (FGF10) gene was detected in the other pedigree [2]. Mutations in FGFR2, FGFR3, and FGF10 are plausible causes of LADD syndrome because defects in these genes have been previously reported in humans and transgenic mice with similar abnormalities of the lacrimal and salivary glands as well as limb and digit anomalies [8-10]. These data indicate that the cluster of mutations in three members of the fibroblast growth factor (FGF) signaling pathway have central importance in the pathogenesis of LADD syndrome. In this report, we describe research to identify the genetic cause of our patient’s case of LADD syndrome that includes thin corneas and open angle glaucoma among its clinical features.
The study was approved by the Institutional Review Board at the University of Iowa and adhered to the tenets of the Declaration of Helsinki and the ARVO statement on human subjects. Clinical assessments of this patient were previously reported [7]. After the patient was enrolled in the study, a blood sample was drawn and a skin biopsy was obtained as previously described [11,12]. DNA was isolated from a blood sample using the QIAamp Blood Maxi Kit (Qiagen, Valencia, CA) and fibroblast cells were isolated and cultured from the skin biopsy using standard techniques [13].
Whole exome sequencing was performed on the LADD syndrome patient’s DNA with the Agilent v5 kit (Santa Clara, CA) and an Illumina Hiseq2500 next-generation DNA sequencer (San Diego, CA) using the manufacturers’ protocols. The Burrows-Wheeler aligner (BWA) was used to align DNA sequencing reads to the human reference genome [14]. DNA sequence variations were identified using the Genome Analysis Toolkit (GATK) [15], CoNIFER [16], and a custom sequence analysis and annotation pipeline (Wynn Institute for Vision Research, Iowa City, IA).
First, DNA variations in the coding sequences or the canonical splicing sites of known LADD genes (FGFR2, FGFR3, and FGF10) were identified and analyzed. Next, variations present in the entire exome were filtered and prioritized using previously described methods [17,18]. Variants were filtered to identify those that altered the coding sequences or canonical splicing sites of known genes. Variants that were previously reported in the public database of the Exome Sequencing Project (ESP) [19] were judged to be too common to cause LADD syndrome and were excluded. Candidate mutations and genes were further filtered based on their annotation and reported function using an advanced search of the GeneCards Human Gene Database [20] with input keywords “lacrimal,” “auricular,” “dental,” “digital,” “salivary,” and “morphogenesis.” Top candidate mutations were further evaluated by investigating their prevalence in the public database curated by the Genome Aggregation Database (gnomAD) [21]. The pathogenicity of these candidate mutations was also estimated using the BLOSUM62 matrix [22], PolyPhen2 [23], and SIFT [24] analyses. For the top candidate, TP63, the Arg343Trp mutation was assessed for alteration of conserved protein sequences by constructing homology tables using the UCSC browser (genome.ucsc.edu) [25]. The likely effects of the Arg343Trp mutation on TP63 protein structure were investigated based on a wild-type structure determined by X-ray crystallography (3US2) [26]. The analysis began by mutating the Arg343 residue of the crystal structure to Trp, followed by repacking of nearby residues using a rotamer optimization algorithm [27] and a potential energy function defined by the polarizable atomic multipole AMOEBA force field [28] in the program Force Field X [29].
We recently reported a patient with LADD syndrome that included early-onset open angle glaucoma and remarkably thin corneas (central corneal thickness of 387 µm OD and 334 µm OS) [7]. This patient was of non-Hispanic Caucasian and Asian heritage. Reportedly, neither the patient’s parents nor other family members are affected by LADD syndrome, but were unavailable for examination or inclusion in our genetic study.
In this report, we searched for the genetic cause of our patient’s LADD syndrome using whole exome sequencing. In the first stage of our analysis, we searched our patient’s exome data for mutations in the three known LADD syndrome genes, FGFR2, FGFR3, and FGF10. A heterozygous, synonymous variant, Leu703Leu, was discovered in FGFR2. However, no plausible disease-causing mutations were detected in these genes.
As a second stage, we identified genes previously annotated to influence the structures altered in LADD syndrome using the keywords “lacrimal,” “auricular,” “dental,” “digital,” “salivary,” and “morphogenesis” as potential causes of this disease using the online resource GeneCards [20]. A total of 58 candidate genes were prioritized for analysis (Appendix 1). Three of the top four candidates on this gene list were the previously identified LADD syndrome genes—FGFR2, FGFR3, and FGF10—providing some validation for our prioritization approach. The fourth top candidate on the list was tumor protein 63 (TP63).
When we examined our patient’s exome, we detected a plausible disease-causing mutation in only one of the 58 candidate genes. A missense mutation, Arg343Trp, was identified in TP63. This mutation is not present in the exomes of more than 126,000 humans (gnomAD database) [21], including over 24,000 Asian and 56,000 Non-Finnish Europeans who have the most relevant heritage for use as a control data set for this study’s LADD syndrome patient. These data suggest that the TP63 Arg343Trp mutation is extremely rare, as would be expected for a mutation thought to cause LADD syndrome. The potential pathogenicity of the TP63 mutation was further assessed with several mutation analysis algorithms (Table 1). The Arg343Trp mutation generated a Blosum62 score of −3, a PolyPhen2 score of “probably damaging,” and a SIFT score of “deleterious.” Moreover, the Arg343Trp mutation alters an amino acid that has been highly conserved in the encoded TP63 protein across many diverse species (Figure 1). Finally, molecular modeling of the TP63 protein based on its solved crystal structure was conducted both with and without the Arg343Trp mutation using structural refinement algorithms based on the polarizable AMOEBA force field. These analyses showed that the Arg343Trp mutation caused a dramatic conformational alteration of TP63 that would likely be deleterious to its DNA-binding domain (Figure 2).
Our genetic studies of a patient with LADD syndrome, glaucoma, and thin corneas detected no mutations in the previously reported LADD syndrome genes, FGF10, FGFR2, and FGFR3. Subsequently, whole exome sequencing identified a novel Arg343Trp mutation in the TP63 gene as a potential cause of LADD syndrome. This TP63 mutation is exceedingly rare and was not detected in over 126,000 individuals in the gnomAD database. Moreover, a range of mutation analysis approaches (Blosum62, SIFT, PolyPhen2), homology studies, and protein modeling all suggest that the Arg343Trp mutation is likely pathogenic. Additional support for the pathogenicity of the TP63 mutation might be obtained by examining our patient’s family with genetic tests. Since both of his parents are reportedly unaffected by LADD syndrome, one might expect them to test negative for the Arg343Trp mutation. Such test results would indicate that our patient carries a de novo TP63 mutation and would be further evidence that the Arg343Trp mutation is disease-causing. Unfortunately, neither parents were available for examination or enrollment in our study.
TP63 encodes a transcription factor that is expressed in ectodermal tissue during development, and mutation of this gene is a highly plausible cause of the congenital features of LADD syndrome. Mutation of TP63 has been previously associated with the spectrum of developmental diseases termed ectrodactyly, ectodermal dysplasia, and cleft lip/palate syndrome (EEC, OMIM 604292) [30]. The features of EEC overlap significantly with LADD syndrome. Both syndromes may include lacrimal duct abnormalities, hearing abnormalities, digital abnormalities (syndactyly), and dental abnormalities (hypodontia). There are even families in which some members have LADD syndrome while others have cleft lip and palate and were diagnosed with EEC [31]. The overlap of the clinical features of LADD syndrome and ECC may be due in part to common genetic bases, that is, that some mutations in TP63 may be more likely to cause LADD syndrome while others may be more likely to cause ECC. Together these data and observations provide additional support for our hypothesis that mutation of TP63 may cause LADD syndrome. We further suggest the possibility that TP63 may also be associated with risk for thin central corneal thickness and for glaucoma. Overall, the evidence is compelling that mutation of TP63 plays a role in the pathogenesis of LADD syndrome, however, further investigation with functional studies is warranted. Additional support for the pathogenicity of the TP63 mutation might be obtained by examining our patient’s family with genetic tests. Since both of his parents are reportedly unaffected by LADD syndrome, one might expect them to test negative for the Arg343Trp mutation. Such test results would indicate that our patient carries a de novo TP63 mutation and would be further evidence that the Arg343Trp mutation is disease-causing. Unfortunately, neither parents were available for examination or enrollment in our study.
This research was funded in part by NIH grant R01EY023512, the Research to Prevent Blindness Physician Scientist Award, and Dr. Albert and Evelyn Mintzer.