X-linked retinoschisis (XLRS, MIM 312700) is a hereditary retinal disease characterized by a splitting of the neurosensory retina, with a prevalence of 1:5,000 to 1:25,000 males worldwide (1). Typical fundus changes include radiating cysteic maculopathy in most cases and peripheral retinoschisis in half of the cases (2). However, the disease has a high degree of phenotypic variability (3–6), in which genetic testing is of value in confirming the diagnosis (4).

XLRS accounts for most congenital retinoschisis (2,7) and is due to mutations in the retinoschisin gene (RS1, OMIM 312700) localized on Xp22.13 (8,9). The encoded protein, retinoschisin, is secreted from photoreceptors and bipolar cells as a functional homo-octameric complex that is thought to play a role in cellular adhesion and cell-to-cell interaction (10).

Gene transference to mouse models of X-linked juvenile retinoschisis, which suggest gene replacement may be a possible future therapy for patients (11–13). Genetic diagnosis is the basis for gene transference in the future. Therefore, we have to fully understand the molecular basis of XLRS. To date, more than 160 different RS1 mutations have been identified in patients with XLRS (http://www.dmd.nl/rs), including small intragenic deletions, nonsense and missense mutations, frame shift insertions and deletions, and splice site mutations. However, there are still some RS1 mutations that remain unknown.

In this study, we analyzed the coding exons and the adjacent regions of RS1 in patients from 20 unrelated Chinese families with XLRS. Ten hemizygous mutations, including 4 novel mutations, were detected in 14 families.

Probands with XLRS from 20 unrelated families were enrolled in this study. Written informed consent was obtained from the participating individuals or their guardians prior to the collection of clinical data and genomic samples. This study was approved by the Internal Review Board of the Zhongshan Ophthalmic Center.

In this study, ten different hemizygous mutations in RS1 were identified in 14 families with XLRS. These mutations are predicted to be pathogenic. All patients with mutations demonstrated typical signs of XLRS. The ten mutations affected different domains of retinoschisin, including the RS1 domain (1 mutation), discoidin domain (8 mutations) and C-terminal segment (1 mutation). These mutations were not randomly distributed over the gene (Fig. 3) because 80% of mutations were clustered in the discoidin domain (16). The two novel mutations, Tyr177X and Pro203Ala in the discoidin domain, may cause a shorter retinoschisin form or protein misfolding (13). The cysteine mutations in the RS1 domain (Cys59Tyr) and C-terminal segment (Cys223Tyr) may cause failure of the discoidin domain to assemble into a normal multisubunit complex (17,18).

Most of RS1 mutation loci were hot mutation spots, while the Cys59, Glu72, Arg102, Glu146, Arg182, Arg200, Pro203, Glu215 and Cys223 could be substituted by 1–2 other kinds of amino acids and be reported more frequently (19–30). However, the mutations in the present study also differed from those reported previously. The RS1 mutations accounts for 70% of the Chinese retinoschisis (14/20) cases in our study. The Cys59Tyr, Tyr177X, Pro203Ala, Glu215Val and Cys223Tyr mutations only are present in the Chinese population (31), and the Cys59Tyr mutation was more common (10% frequency in our retinoschisis cases). The Glu72Lys mutation is the most common among Chinese (15%) as well as other populations (19,32), while another very common mutation, Pro192Ser (33), which was reported from people of different ethnic backgrounds was not found. We do not know whether the spectrum and frequency of RS1 gene in the Chinese is different from others. Our study contributes to the current state of knowledge.

In summary, we identified ten mutations in 14 of 20 families with XLRS. Our results expand the mutation spectrum of RS1 that might enrich our understanding of the molecular basis of XLRS in the Chinese population.