Pseudoachondroplasia (PSACH) is a rare skeletal dysplasia. This disorder is inherited in an autosomal dominant manner, which means that inheriting a copy of the altered gene from a single parent is sufficient to cause the disorder. The condition is recognized by the Online Mendelian Inheritance in Man (OMIM) database with the MIM number 177170. PSACH is typically characterized by short-limb short stature, brachydactyly, loose joints, joint stiffness, and pain (1). People with PSACH typically have normal growth and development during the first 2 years of life, but then their growth rate slows down significantly and falls below the standard growth curve. Other characteristic features of PSACH include a waddling gait, genu varus, limited elbow and hip range of motion, scoliosis, lumbar lordosis, and spinal stenosis. However, it does not affect craniofacial appearance or intelligence. The severity of these symptoms can vary among affected individuals.

PSACH is caused by mutations in the COMP gene (COMP; 600310) (2). COMP is a member of the thrombospondin gene family and encodes cartilage oligomeric matrix protein, a family of secreted matricellular proteins that modulate cell behavior and ECM organization. (3–5). The COMP protein is composed of multiple domains, including an N-terminal region, a central domain composed of two types of repeats: type II epidermal growth-like repeats (T2 repeats) and type III calmodulin-like repeats (T3 repeats), which are responsible for oligomerization of the protein and its interaction with other ECM proteins, and a C-terminal globular domain (CTD) (6, 7). The T2 repeats, T3 repeats, and CTD can bind calcium ions, which is essential for the correct folding of the protein, and enables it to function properly in cartilage and other connective tissues (8, 9). The loss of tertiary structure and misfolding of the protein due to COMP mutations cause substantial retention of COMP in the endoplasmic reticulum (ER), resulting in skeletal abnormalities (such as PSACH or multiple epiphyseal dysplasia) (10, 11).

In PSACH, mutations are mostly found in the T3 repeats whereas mutations in the CTD are less common (6, 7). Interestingly, the position of the pathogenic variant may cause different levels of short stature. Research shows that when the variant occurs in the T3 repeats rather than in other domains, patients may suffer more (12). Specific missense variants in the C-type motif of the T3 repeats result in PSACH, whereas missense variants in the N-type motif generally result in MED (multiple epiphyseal dysplasia), a milder skeletal dysplasia (13). It is worth noting that not all missense variants in the C-type or N-type motifs lead to PSACH or MED. Different mutations that occur at the same site can also lead to different severity of the disease. Individuals carrying p. Asp469del appear extremely short (12), whereas p. Asp469dup results in mild MED (13–15).

In this report, we describe two families with PSACH. All the affected patients had a typical clinical phenotype and were subsequently diagnosed by X-ray. Whole-exome sequencing (WES) identified two novel missense variants in the COMP gene. Following the ACMG guidelines, the two variants were interpreted as likely pathogenic. Our findings offer valuable insights into the molecular and clinical characteristics of PSACH among Chinese patients, which can facilitate the early detection of the condition and improve treatment outcomes for individual patients.

In this study, we present two families diagnosed with PSACH, which were confirmed through exome sequencing and further validated using Sanger sequencing. Through WES, we identified two novel missense variants (c.1319G>T and c.1304A>T).

The pathogenicity of c.1319G>T (p.G440V) is supported by various lines of evidence. Firstly, the variation occurs within the functional thrombospondin type 3 repeat domain (PM1). Secondly, the frequency of this variant is not present in the gnomAD database (PM2_supporting). Thirdly, the variant occurs at the same position as other known pathogenic missense changes, such as c.1318G>A (p. Gly440Arg), c.1318G>C (p. Gly440Arg), and c.1319G>A (p. Gly440Glu) (PM5). Fourthly, the variant co-segregates with PSACH in the affected families (PP1). Lastly, multiple bioinformatics tools predict deleterious effects of the mutation on COMP (PP3). Considering these factors, this variant was interpreted as a likely pathogenic variant (PM1+PM2-supporting+PM5+PP1+PP3) following the ACMG guidelines (16).

In the second case, the variant c.1304A>T was derived from the mosaic father of the proband. Several factors were taken into consideration to assess its pathogenicity: (1) the variation occurred within the functional thrombospondin type 3 repeat domain (PM1); (2) the absence of this variant in various reference population databases, including 1000 Genomes Project, CNGBdb, ExAC, and gnomAD, supports its rarity (PM2_supporting); (3) the variant co-segregated with the disease in affected family members (PP1); (4) multiple bioinformatics tools predicted deleterious effects on COMP (PP3); (5) the observed phenotype in this case aligns with the known disease association (PP4). Based on these considerations, this variant was interpreted as a likely pathogenic variant following the ACMG guidelines (PM1+PM2_supporting+PP1+PP3+PP4) (16).

A comprehensive study conducted by Briggs et al., which examined 300 COMP pathogenic variants associated with pseudohydroplasia and/or autosomal dominant multiple epiphyseal dysplasia (MED), established a robust genotype–phenotype correlation. Pathogenic missense variants predominantly associated with the pseudoachondroplasia phenotype were located in the nucleotides encoding the sixth to eighth type III calcium-binding repeats, specifically T3₆, T3₇, and T3₈. In this study, two novel variants (c.1319G>T and c.1304A>T) were identified in the sixth type III calcium-binding repeat (T3₆) of COMP. Moreover, our investigation unveiled the co-segregation of the variant with PSACH, as evidenced by the presence of the heterozygous COMP mutation in all affected family members, whereas unaffected individuals did not harbor this mutation. The clinical phenotype observed in affected individuals is consistent with the specific features of PSACH, including brachydactyly, a waddling gait, genu varum, mild thoracic deformity, and restricted extension at the elbows in the upper limbs. Additionally, in the second family, the proband’s father manifested a milder clinical phenotype, with only a joint effusion, and WES analysis revealed the presence of the c.1304A>T mutation in 15% of the reads. Collectively, the evidence presented in our study strongly supports the concordance between the two mutations and the well-established genotype–phenotype correlation in PSACH.

Cartilage oligomeric matrix protein (COMP) is a multimeric glycoprotein composed of five monomers which are held together by a coil–coil domain to form a pentameric structure (Posey et al.,2018). Interestingly, knockout of the COMP gene in mice does not result in any noticeable phenotypic changes (17). This suggests that it is the dysfunction of mutated forms of COMP, rather than a reduction in protein levels, that leads to PSACH. The concept of dominant negative was proposed by Ira Herskovitz in 1987 as an approach to studying gene function (18). In cases where a protein functions as a multimer, mutated forms of the protein may exert inhibitory effects by interacting with wild-type peptide chains, thereby resulting in the formation of non-functional multimers. Notably, Maddox et al. observed the accumulation of COMP and type IX collagen proteins on expanded rough endoplasmic reticulum vesicles in cartilage from PSACH patients, and Posey et al. simulated this process at the cellular level using a mouse model(19). The classic mutant Asp469del COMP has been well-documented to be retained in the endoplasmic reticulum (ER)(20). Similarly, the D475N mutant COMP, also located in the T3 repeat region, has been reported to exhibit ER retention (21). In contrast, mutant proteins situated in the C-terminal region, such as COMP H587R, T585M, and several other mutations within the C-terminal domain (CTD), exhibited efficient secretion into the culture media of diverse cell models. Despite their efficient secretion, these mutants continued to exert detrimental effects on cellular viability and disrupt the extracellular matrix (21, 22). Additionally, research conducted by Bell et al. has provided valuable insights into the pathogenic mechanisms. They demonstrated that mutations in matrilin-3 (Matn3 V194D) or COMP (Comp T585M and Comp DelD469) can alter the extractability of other cartilage proteins, ultimately leading to the disruption of cartilage tissue integrity. Notably, this disruption appears to be genotype-specific (23). Collectively, these findings underscore both the consistency and diversity of pathogenic mechanisms.

The Gly440 residue is originally situated within a bend curvature in the protein structure. Glycine, characterized by its small side chain, exhibits greater flexibility compared with other amino acids, allowing it to adopt a wide range of phi/psi backbone dihedral angles. This flexibility makes glycine commonly found in loops and regions where the polypeptide chain undergoes sharp turns. The absence of a bulky side chain enables glycine to occupy compact spaces and execute tight turns without encountering steric clashes. When any other amino acid replaces glycine at position 440, it can lead to an increase in energy required for conformational changes and may impact the overall stability of the protein. Consequently, such substitutions are considered damaging.

The Asp435 residue is situated within the T3 repeats, with each repeat sequence containing two DxDxDGxxDxxD motifs (11). The Asp residues within these motifs carry a negative charge, enabling them to bind to calcium ions and contribute to the structural stability of the protein. Specifically, Asp435 interacts with calcium ion CA 817 in conjunction with Asp437, Asp439, His441, and Asp446. The substitution of Asp435 with Valine (D435V) results in the elimination of the negative charge, disrupting the interaction between Asp435 and CA 817. This disruption ultimately destabilizes the protein. Maddox et al. revealed that COMP3-Mu (D446N) peptides bound less than half the amount of calcium when compared with the COMP3-WT peptides. This indicates that a single amino acid mutation at a calcium-binding site could significantly impact binding at other sites (24). Additionally, Asp435 is involved in forming hydrogen bonds with glycine (Gly) residues at positions 438 and 440. While the D435V mutation does not affect the hydrogen bond with Gly438, it does disturb the hydrogen bond with Gly440, potentially leading to structural and functional alterations in the protein. The D435V mutation not only breaks this hydrogen bond but also affects its interaction with calcium ions (CA 817). This may explain why the D435V mutation is associated with more severe clinical symptoms compared with G440V.

Taken together, it is highly probable that these two missense mutations induce endoplasmic reticulum retention, exerting a dominant negative influence on the normal function of wild-type COMP and other cartilage proteins. The inclusion of in vitro and animal experiments may provide valuable insights to address this question more definitively.

In summary, we have identified two novel heterozygous likely pathogenic mutations (c.1319G>T and c.1304A>T) in the COMP gene, which contribute to our knowledge of the genetic basis of PSACH. Our findings underscore the clinical significance of T3 repeats and calcium ions in the functioning of COMP proteins. The addition of these two novel variants to the existing spectrum of COMP mutations enhances our understanding of the genetic diversity underlying this disease. This information can be valuable for diagnostic testing, genetic counseling, and the development of targeted therapies for affected individuals and families.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The human samples used in this study were acquired from primarily isolated as part of our previous study for which ethical approval was obtained.

LZ: Data curation, Writing – original draft, Writing – review & editing. JC: Resources, Writing – original draft. QL: Software, Writing – original draft. SY: Formal Analysis, Writing – original draft. WX: Formal Analysis, Writing – review & editing. YP: Project administration, Supervision, Writing – review & editing.