Identification of a Novel Homozygous IHH Variant Causing Novel Acromesomelic Maroteaux-Type Skeletal Dysplasia in a Pakistani Family

Family was recruited and samples obtained with informed written consent under a protocol approved by the Institutional Ethical Review Board of the Abdul Wali Khan University Mardan (Pakistan). A four-generation pedigree was constructed after consultation with family elders (Figure 1a). Affected members (IV-1, IV-3, IV-4) underwent clinical examination at local government hospitals. The extraction of Genomic DNA was performed using standard methods (Sambrook et al. 1989). Whole-exome sequencing (WES) was performed using the Illumina HiSeq: Agilent Sureselect Whole Exome v6 targeting was used for exome capture, read alignment was done using BWA-MEM (v0.7.17) which was followed by mate-pairs fixed and duplicates removal by Picard v2.15.0 was used, followed by InDel realignment and base quality recalibration with GATK v3.7.0, and subsequent detection of single-nucleotide variants (SNVs) and InDel by GATK Haplotype Caller, and annotation by Alamut v1.8. Sequencing depth was assessed with GATK DepthOfCoverage. Based on the most likely inheritance pattern inferred from pedigree analysis, homozygous, compound heterozygous, and heterozygous variants were prioritized. Selection criteria included a minor allele frequency ≤ 0.005 in gnomAD, a CADD-Phred score ≥ 15, exonic or splice-site variants (±12 bp), and phenotypic similarity of candidate genes to the family phenotype. Segregation of selected variants was confirmed by Sanger sequencing in all available family members. PCR primers were designed using the Primer3 tool (https://primer3.ut.ee) while the reference gene sequences were retrieved from the UCSC Genome Browser (https://genome.ucsc.edu/).

Figure 1.

(a) Pedigree of family exhibiting autosomal recessive pattern. (b) The father (III-7), of average height along with the affected males (IV-4, IV-1) who are short statured with normal trunk length and markedly shortened arms and legs. (c, d) Hands of affected males (IV-1 and IV-4) are short and broad. Clinodactyly is seen. Nails are dystrophic, with irregular curvature and thickening. (e, f) Clinical Photographs of feet of IV-1and IV-4 demonstrate broad, flat feet with disproportionately short and stubby toes. Toenails are dystrophic, with bilateral hyponychia of the great toes.

I-Tasser server was used to build three-dimensional structure of human HH C-terminal auto-processing domain of wildtype IHH protein (IHH^WT) through homology modelling approach [22] using 1at.0 PDB molecule as template with 36.36% identity. Next, Insilco mutagenesis approach mutant three-dimensional structure of IHH(IHH^Val340Met) protein was built through I-Tasser server [22] using IHH^WT protein model as template. Predicted model were refined by WinCoot [23]and stereochemistry and validity of constructed 3D structure of human IHH^WT protein (C-terminal auto-processing domain) and mutant HH^Val340Met was assessed by Ramachandran plot [24] PROCHECK [25] and verify 3D [26]. To measure the effect of identified mutation on protein structure HOPE server [27] was used.

Comparative MD simulations of IHH^WT and IHH^Val-340Met (c-terminal auto processing domain) were performed to evaluate the conformational alterations, stability, and dynamic properties of the IHH protein. GROMACS 4.5 package was used to do simulations experiments by using Amber03 force field [28]and solvated using the TIP4P water model [29] within a periodic box. Next, Na⁺ and Clˉ counter ions were added to both (IHH^WT and IHH^Val340Met) systems to neutralize them. Simulations were carried out for a duration of 20ns time scale under constant conditions of temperature (300K) and pressure (1 atm) for the IHH^WT and IHH^Val340Met systems. The stability of the secondary structure elements stability along with their interactions and conformational changes were assessed using UCSF Chimera.

The affected siblings, IV-1 (male, 28 years), IV-3 (female, 26 years), IV-4 (male 22 years) were born from a consanguineous union following full-term pregnancies without complications. At the time of evaluation IV-1 measured 116.84 cm in height (−9.6 SD), weighed 38.5 kg, hand length of 40.64cm and wrist-to-finger length of 15.24cm; IV-3 had a height of 99.06cm(−9.7 SD), weighed 32 kg, hand 38.1cm, wrist to fingers 15.24cm; IV-4 had a height of 121.92cm(−10.4 SD), weighed 38 kg, ,upper limb length (shoulder to fingertip) of 41.14cm and wrist to finger length of 15.66cm. Head circumference of the three affected (IV-1, IV-3, IV-4) was 54.61,50.8,53.84cm, respectively. Father of the affected individuals (55 years old) was of average height 160.02 cm (−1.54SD) and weight of 65 kg (Figure 1b). Head circumference and hand length were 54.61 cm and 66.04cm respectively. Based on the available information, the mother was reported to be clinically normal with no skeletal abnormalities. All the three affected individuals had very short stature with normal trunk length, severely shortened arms and legs (Figure 1b). Hands were short and broad. Fingers, especially the thumbs and index were short relative to the hand size. The 5th fingers on both hands showed inward deviation toward the 4th fingers. Nails appeared dystrophic, with irregular curvature and thickening (Figure 1c, d). Feet were broad and flat, with disproportionately short, broad, and stubby toes. Toes nails were dystrophic with bilateral hyponychia in big toes. Bowed legs (genu varum) were also evident in the standing image (Figure 1e, f) (Table 1).

Lateral and anterior posterior radiographs of the skull of the affected members (IV-1 and IV-4) showed normal craniofacial proportions but increased radiodensity of the calvarium. No midface hypoplasia and frontal bossing was observed (Figure 2a–d). The upper limb X-rays of the same individuals revealed mildly shortened humerus, bilateral shortened radius and ulna with thickened cortices and narrow medullary canals, more pronounced in the distal radius. There was also volar and ulnar bowing of the distal radius, along with dorsal subluxation of the distal ulna and increased radial inclination supporting the diagnosis of Madelung deformity (Figure 2e–h). Chest radiographs of the affected individual IV-1 and IV-4 were normal while the chest and upper limb radiographs of the affected individual IV-3 showed Madelung deformity characterized by radial bowing with increased cortical density of the long bones, and normal thorax and vertebral column (Figure 2i–k). Radiographs of the hands of the individuals IV-1, IV-3 and IV-4 showed severe brachydactyly with generalized cortical sclerosis of the metacarpals and phalanges. The 4th and 5th metacarpals were disproportionately shortened and appeared stubbier compared to other metacarpals. The phalanges, particularly the proximal were mildly shortened with relatively preserved alignment and morphology. In addition, the carpal bones appeared to form a V-shaped (triangular) configuration at the proximal carpal row. No cone-shaped epiphyses, synostosis, or overt joint dislocations were identified in the visualized phalanges (Figure 2l–n). Pelvis radiographs of the affected individuals IV-1, IV-3 and IV-4 revealed short femoral necks with decreased neck-shaft angles consistent with coxa vara. The femoral shafts showed cortical thickening with narrowing of medullary canal. Hip and knee joint spaces were preserved (Figure 2o–q). Lower limb radiographs showed shortened and broad tibia and fibula with genu varum (bowing of the legs). There was flaring and irregularity of the metaphysis at the distal femora and proximal tibiae. The cortices appeared thickened, and the medullary canals were narrowed. Additionally multiple well-defined calcified loose bodies were seen in the left suprapatellar pouch and juxtaarticular region (Figure 2r–t). The X-rays of the feet showed short and broad tarsals, metatarsals and phalangeal bones with bony fusion (synostosis) of the distal phalanges of the fifth toes bilaterally. The epiphyses, particularly of the distal metatarsals and proximal phalanges, appeared broad and slightly squared. There was evidence of fixed plantar flexion of the forefeet and equinus deformity at the ankle joints (Figure 2u, v) (Table 1).

Figure 2.

(a–d) Lateral and anteroposterior (AP) skull radiographs of affected individual IV-1 and IV-4 show generalized calvarial thickening with absence of frontal bossing and no evidence of macrocephaly. (e–h) Left and right upper limb radiograph of affected individuals (IV-1, IV-4) showing bilateral shortening of the radius and ulna, with cortical thickening and narrowed medullary canals, most pronounced in the distal radius. Features of true Madelung deformity, including volar-ulnar bowing of the distal radius, dorsal subluxation of the distal ulna and increased radial inclination are also observed. (i,j,k) Chest radiographs of individuals IV-1, IV-3 and IV-4 reveals normal rib and clavicular structure (l–n) Hand radiographs of all three affected siblings demonstrate severe brachydactyly. A V-shaped configuration of the proximal carpal row is also observed in the individual IV-3. (o–q) AP pelvic radiographs of the affected individuals (IV-1, IV-3 and IV-4) demonstrate bilateral coxa vara, and femoral bowing. (r–t) Lower limb radiographs of the individuals IV-1, IV-3 and IV-4 show shortened and broadened tibia and fibula with genu varum. Flared and irregular metaphyses of the distal femora and proximal tibiae, cortical thickening with narrowed medullary canals is noted. Multiple well-defined calcified loose bodies are seen in the left suprapatellar pouch and juxta-articular region in the affected individual IV-1. (u–v) Foot radiographs of the individuals IV-1, IV-4 depict short and broad tarsals, metatarsal, and phalangeal bones with bilateral fusion of the distal phalanges of the fifth toes. The distal metatarsal and proximal phalangeal epiphyses appear broad and squared. Fixed plantar flexion of the forefeet and equinus deformity at the ankle joints are also evident.

All the patients of the family had normal intellect and central nervous system. Blood sugar levels, kidneys, heart and vision were normal. No central facial dysmorphic features were detected. Heterozygous carriers exhibited normal hands and feet with no other anomalies were noted.

In the affected individual IV-1, WES analysis revealed a homozygous missense variant [(c.1018G>A; p. Val340Met) in the exon 3 of IHH gene, reporting the most significant variant. Segregation of the variant was tested using available DNA samples from the family members (III-7, IV-2, IV-3, IV-4) (Figure 3a). No homozygous occurrence of the variant was found in online population databases (EVC, ExAC, gnomAD, 1000genomes) or among 135 in-house Pakistani control exomes. Conservation analysis showed the wild type residue to be highly conserved among different species (Fig. 3B). The variant was estimated to be damaging and disease causing via different online software’s: Mutation Taster, Versome, PolyPhen-2, Provean and SIFT FATHMM-MKL (Table 2). Based on ACMG criteria the sequence variant was classified as likely pathogenic (Table S1) [30].

Figure 3.

(a) Sanger electrograms of the variant (c.1018G>A; p. Val340Met) identified in an affected (upper panel), a carrier (middle panel) and a normal (lower panel) (b) Showing conservation of valine amino acid across several species. Secondary structure of (c)IHH^WT and (d) IHH^Val340Met. Structure validation through Ramachandran plot (e) IHH^WT and (f) IHH^Val340Met (g) and (h) represents ribbon form of wild type and mutant tertiary structures respectively. (i) Superimposition of mutant and wild type IHH auto processing domain.

Understanding secondary structure elements is essential for acquiring deeper insights into the conformational changes at the 3D level. The Val340Met substitution occurs in the C-terminal auto processing domain of the IHH protein. The two-dimensional structural of IHH^WT comprised 10 β-sheets, a single helix, 10 beta turns and 4 hairpins loops. In contrast, IHH^Val340Met, exhibited changes in β-sheets lengths unlike RPTN^WT (Figure 3c and d).

To assess the structural impact of the identified mutation, wild type and mutant protein models of IHH (C-terminal auto processing domain) were generated and analysed. Ramachandran plots analysis of the modelled IHH^WT and IHH^Val340Met showed that over 95% residues were present in the sterically allowed region (Figure 3e, f, g and h). Both modelled structures displayed acceptable stereochemical quality with parameters including peptide bond planarity, non-bonded interactions, Cα tetrahedral distortion, main chain Hydrogen bond energy and G-factor within the favoured regions. Superimposition of IHH^WT and IHH^Val340Met produced an overall RMSD of Q score of 0.741(Figure 3i).

Due to their distinct chemical characteristics (hydrophobic versus amphipathic) and differences in side-chain size, the substitution of the wild-type valine residue with methionine may lead to several consequences. Firstly, the mutated residue lies within a domain critical for processing and secretion of the IHH protein and this substitution may disrupt interactions with binding partners thereby affect the function of the protein. Secondly the wild-type valine residue is buried within the protein core due to its hydrophobic nature, whereas the bulkier methionine side chain may not fit into protein core, disrupting the core packing and altering the overall protein conformation (Figure 4a and b).

Figure 4.

(a) Ribbon representation of protein in grey colour with mutant residue in magenta colour. In close up view side chain of both wild type and mutant residues are shown in green and red colour respectively. (b) schematic structures of the original (left) and the mutant (right) amino acid. The backbone, which is the same for each amino acid, is coloured red. The side chain, unique for each amino acid, is coloured black. RMSF analysis (c, d) simulation trajectories and (e) C-α distance restraints analysis.

The results of MDs were consistent with our structure analysis of the IHH protein. RMSF analysis derived from atom trajectories revealed high fluctuations in the region harbouring the p. Val340Met substitution (Figure 4c and d) which was further validated by the C-α distance restraints analysis (Figure 4e). Simulations results suggests that p. Val340Met substitution resulted in loosely packed c-terminal auto processing domain of the IHH protein and significant structural alterations may contribute to protein dysfunction leading to disease onset. So, the identified substitution is thought to reduce the compactness and ultimately decrease the stability of the IHH protein.