Spondylocostal dysostosis (SCD) is a group of rare orthopedic disorders defined by multiple vertebral segmentation defects and rib deformities. Six genes are known to cause this disease, namely, DLL3 (1), MESP2 (2), LFNG (3), HES7 (4), TBX6 (5), and RIPPLY2 (6). Recently, two other genes DMRT2 and SLC35A3 (7) have been associated with SCD-like form (8).

Affected individuals have a short trunk and a short neck with limited mobility. Spine deformity occurs in 10 contiguous vertebrae with mild and often non-progressive scoliosis. There is also a deformity of the ribs which can be fused, missing, or fanning out of the spine in a “crab-like” fashion. Respiratory complications are among the most common causes of morbidity and mortality in affected individuals. Deformity of the rib cage affects lung development, which causes lung infections and respiratory failure (9).

The association of SCD with fusion of the spinal processes of the vertebrae and multiple pterygium has been rarely reported and called spondylospinal thoracic dysostosis (SSTD) (10). SSTD seems to represent a separate clinical entity with overlapping findings with SCD. To date, no study reported cases of SSTD in North Africa or investigated the genetic etiology of this particular syndrome.

Several variations in distinct genes might occasionally result in a more severe phenotype that affects how the disease manifests itself in Mendelian disorders; this phenomenon is defined as oligogenic inheritance. Many Mendelian disorders, such as Gaucher’s disease, have been linked to oligogenic inheritance (11, 12).

In this work, we describe a patient with a complex phenotype suggestive of SSTD for whom whole exome sequencing (WES) revealed the presence of the pathogenic TPM2 variant. We further analyzed WES data using bioinformatic tools for investigating the variant with oligogenic inheritance. Our findings point to a list of possible modifiers genes that might influence the patient's phenotype.

The clinical phenotype of the patient in our study consists of multiple deformities of the axial skeleton including a defect of segmentation of nearly all the vertebrae with multiple irregularly shaped vertebral bodies, bloc vertebrae, rachischisis, and multiple hemivertebrae. These deformities are associated with a severe lordosis of the entire spine with fusion of the spinous processes that impair the sitting ability of the child. Another axial anomaly is represented by the deformity of the rib cage with a fanning out of the ribs from the costovertebral junction in a “crab-like” fashion (20). These clinical features are highly consistent with the diagnosis of SCD (9). The latter are a heterogenous group of rare genetic disorders also known as spondylothoracic dysplasia, costovertebral dysplasia, or Jarcho–Levin syndrome transmitted on an autosomal-recessive or an autosomal-dominant way (21). These diseases are caused by mutation in DLL3, MESP2, LFNG, HES7, TBX6, RIPPLY2, DMRT2, and SLC35A3 genes (21). The clinical and radiological distinction between the different types of SCD is based on differences in the vertebral and rib anomalies with different degrees of thoracic insufficiency that can be lethal in extreme cases (22). But none of the genetically identified types of SCD include multiple pterygia and arthrogryposis as it has been observed in our patient, which also cannot be explained by the multiple pterygium syndrome (MPS) because of the absence of akinesia and variants in the genes related to this disorder.

The association of vertebral segmentation defects and rib anomalies similar to what has been described in SCD, with fusion of the spinal processes and severe hyperlordosis of the entire spine and multiple pterygia with arthrogryposis, seems to be a separate entity. Johnson et al. termed this association SSTD stressing the fusion of the spinal processes and the resulted hyperlordosis (10). Becerra-Solano et al. named this particular entity the Turkel–Chen–Johnson syndrome after the first three authors who described the association of an SCD phenotype with extended fusion of the spinal processes, multiple pterygia, and arthrogryposis (23). To the best of our knowledge, only seven cases of SSTD have been reported previously (10, 23, 24). Table 2 compares the clinical and radiological manifestation found in SCD SSTD, and our patient. SSTD was considered lethal by most authors as the six first reported cases died at an early age due to thoracic insufficiency. This was also the case in our patient. However, Becerra-Solano et al. reported the case of a healthy 16-year-old Mexican girl, who goes to school without respiratory failure (23). As no genetic diagnosis for this association was known, the authors suggested to perform WES to sort out SSTD from SCD which gives very similar vertebral segmentation defects and rib anomalies. The genetic investigation carried out in this study suggests that Tropomyosin gene (TPM) mutations are associated with SSTD. Mutations in the tropomyosin-2 gene (TPM2) have been linked to different clinical manifestations ranging from pure distal arthrogryposis to cap disease and nemaline myopathy (NM) (25, 26), with more than 35 disease-causing mutations (26). The latter has been ruled out by histological analysis. Among these diseases, the non-lethal Escobar variant of MPS is characterized by multiple contractures of the joints (pterygia), webbing of the skin (syndactyly), and anomalies throughout the body and shows marked phenotypic variability (18, 27–29). Multiple pterygium, dwarfism, and facial dysmorphism are among the principal signs of the non-lethal form of MPS. Vertebral segmentation defects can also be seen but are usually localized and do not span the entire spine associated with rib fusions (30). Apart from SCD, the clinical phenotype of our patient is also consistent with the diagnosis of the Escobar syndrome although she lacked the characteristic facial features of the disease (31). In our study, the investigation of the genetic data obtained following whole exome sequencing and validation by Sanger sequencing identified the variation c.628 C>T in the TPM2 gene that results in the emergence of a premature stop codon. This variation has already been reported in four Algerian patients with Escobar syndrome associated with NM (27). As in our patient, this was also reported in two other Tunisian patients with Escobar syndrome who did not exhibit myopathy (18). We therefore presume the absence of a clear phenotype–genotype correlation and the implication of other modifier genes in these complex syndromes as suggested by Najjar et al. (18). It is also important to note that since we used WES for genetic investigation, it is impossible to exclude other genetic factors responsible for the vertebral phenotype separately, in the setting of consanguinity that is known to be high in Tunisian population (32). As further analysis such as whole genome sequencing has not been performed, non-coding variants or del-dups in other genes have not been investigated but may be involved.

The WES analysis also revealed mutations in the SYNE2, ADCY6, and CAP2 genes that can be involved in the clinical manifestations in the patient. But none of the diseases caused by these mutations correspond to what was observed in our patient. American college of medical genetics and genomics (ACMG) classification and in silico prediction further delineate SYNE2, ADCY6, and CAP2 as they display a potential role of modifier genes by playing a role of actin binding proteins (33–35). Indeed, we found a heterozygous variant in SYNE2 gene, and this gene encodes for the Nesprin 2 protein, which is highly expressed in cardiac and skeletal muscles (36, 37). Previous studies suggest that mutations in this gene are associated with autosomal-dominant Emery–Dreifuss muscular dystrophy phenotypes (38–40). These myopathies are typically characterized by early contractures of the elbow flexors and Achilles tendons, spine rigidity, gradual progressive muscle wasting, weakness that begin at the humeroperoneal stage, and cardiopathy (39, 41). However, histological analysis of muscle biopsy and EMG revealed that this variant did not cause muscle pathology in our case. In addition, Sanger sequencing revealed that it was a heterozygous variant inherited from the healthy mother, with the lowest score with RegulomeDB and could be, therefore, the least implicated in the clinical manifestations of our patient.

The second variant that could play a role in the clinical particularities of this patient was observed also at a heterozygous state. Indeed, CAP2, which encodes the cyclase-associated protein (CAP), an enzyme that regulates the actin cytoskeleton, is critical for cell polarity and development (42). CAP2 is mostly expressed in the heart, skeletal muscle, brain, and skin (43). It is related to autosomal-recessive dilated cardiomyopathy (44) and more recently to NM (45).

Through oligogenic analysis, we have also highlighted the possible involvement of a variant in ADCY6 (Adenylate Cyclase 6) gene. This gene encodes for an enzyme from the adenylate cyclase family that is involved in the generation of cAMP implicated in myelinating signals in Schwann cells throughout development (46). It was previously associated with autosomal-recessive forms of arthrogryposis multiplex congenita (47). More recently, it has been also linked to recessive form of lethal congenital contracture syndrome 8, which consists of multiple joint contractures at birth, hypertonia, decreased tendon reflexes, and respiratory insufficiency (48). However, its presence in a heterozygous state as in our case could play a role in the particular clinical manifestation of our patient.

It is worth noting that the combination of the heterozygous variants in CAP2, ADCY6, and SYNE2, associated with the loss of function in TPM2, could also be related to cardiomyopathy. As both parents were apparently healthy and did not exhibit any clinical problems, we cannot, however, exclude any cardiac complication in the patient who died young during the time of investigation and did not undergo deep clinical examination. The same consideration applies for muscle atrophy, which was not observed following histological analysis, but cannot be ruled out due to her sudden death with a pulmonary infection.

SSTD is characterized by an overlap between SCD and Escobar syndrome. It is transmitted in an autosomal-recessive inheritance. In our patient, the sole mutation that can explain the clinical manifestations is the variant in the TPM2 gene that we found. However, this mutation by itself cannot account for all of the disease variability giving the fact that it has been related to several syndromes (27). A contribution of other unknown modifier genes, environmental factors, or their combination is highly suspected (49). This study supports also the utility of the WES approach, particularly in patients who do not exhibit clear phenotypic feature related to one specific syndrome, and point out the need to do further investigations about the functionalities of TPM2 in muscle and orthopedic disorders.

Through this report, we listed a possible variant that may be involved in disease severity based on the clinical manifestations observed in our case study. However, a larger cohort and further functional analysis could provide more information about gene interaction effects on the entire genome. As a result, many disease-related genetic connections remain yet to be discovered, and this report would provide elements of comparison for other similar clinical cases in order to define this syndrome with more precision.

Our findings could also raise concerns about the clinical concept of spondylospinal thoracic dysostosis as a separate entity disorder, as some of the known patients with SSTD may be cases of Escobar syndrome, clinically different due to genetic heterogeneity and due to the possible implication of modifier genes.