To the Editor:

X-linked forms of retinitis pigmentosa (XLRP) are a genetically heterogeneous group of retinal dystrophies that result in relatively severe clinical manifestations (Bird Bird, 1975; for a review, see Aldred et al. Aldred et al., 1994). The two major XLRP loci, RP2 (MIM 312600) and RP3 (MIM 312610), have been mapped to Xp11.32-11.23 and Xp21.1, respectively (for a review see Aldred et al. Aldred et al., 1994; Fujita et al. Fujita et al., 1996; Fujita and Swaroop Fujita and Swaroop, 1996; Thiselton et al. Thiselton et al., 1996). The RP15 locus (MIM 300029) has been mapped to Xp22.13-22.11 in a single family with retinal degeneration (McGuire et al. McGuire et al., 1995), and some evidence exists for a fourth locus, RP6 (MIM 312612), at Xp21.3 (Musarella et al. Musarella et al., 1990; Ott et al. Ott et al., 1990). We recently localized another genetic locus, RP24 (MIM 300155), at Xq26-27 by using linkage analysis in an XLRP family (Gieser et al. Gieser et al., 1998). In addition, the disease in some retinitis pigmentosa (RP) families with apparently X-linked inheritance does not seem to be linked to markers in the region of mapped XLRP loci (Teague et al. Teague et al., 1994; L. Gieser, R. Fujita, and A. Swaroop, unpublished data). It therefore appears that mutations in several genes on the X chromosome may lead to RP.

The first XLRP gene, RPGR (retinitis pigmentosa GTPase regulator), was isolated from the RP3 region (Meindl et al. Meindl et al., 1996; Roepman et al. Roepman et al., 1996). Genetic analysis has suggested that RP3 accounts for 70% of XLRP (Ott et al. Ott et al., 1990; Teague et al. Teague et al., 1994; Fujita et al. Fujita et al., 1997). However, RPGR mutations are detected in only 20% of XLRP (and genetically defined RP3) families (Buraczynska et al. Buraczynska et al., 1997; Fujita et al. Fujita et al., 1997; M. Guevara-Fujita, S. Fahrner, and A. Swaroop, unpublished data). The RP2 gene has recently been isolated by a positional cloning strategy (Schwahn et al. Schwahn et al., 1998) and is predicted to encode a protein of 350 amino acids with homology to cofactor C, which is involved in folding of β-tubulin (Tian et al. Tian et al., 1996). The RP2 locus is believed to represent 20%–30% of XLRP in Europe (Ott et al. Ott et al., 1990; Teague et al. Teague et al., 1994), but little or no genetic evidence exists for an RP2 subtype in the XLRP families from North America (Musarella et al. Musarella et al., 1990; Ott et al. Ott et al., 1990). Because our haplotype analysis provided suggestive evidence for RP2 in two North American families (R. Fujita, L. Gieser, S. G. Jacobson, P. A. Sieving, and A. Swaroop, unpublished data), we examined the genomic DNA from our cohort of XLRP patients for causative mutations in the RP2 gene.

The procedures for clinical ascertainment of patients, obtaining blood samples, and preparation of genomic DNA have been reported elsewhere (Fujita et al. Fujita et al., 1997). The families included in the present study showed an apparent X-linked inheritance and no male-to-male transmission. Affected male individuals had a clinical diagnosis of RP. Initially, one affected male each from 51 XLRP families was included in the RP2 screening project. This cohort did not include families with a causative RPGR mutation or those in which the disease was genetically mapped to the RP3 locus (see Buraczynska et al. Buraczynska et al., 1997 and Fujita et al. Fujita et al., 1997). Oligonucleotide primers flanking each of the five RP2 exons (Schwahn et al. Schwahn et al., 1998) were used to amplify products from genomic DNA. PCR products were sequenced with various primers (Schwahn et al. Schwahn et al., 1998), either directly or after gel purification, by means of the ³³P-Thermosequenase cycle-sequencing kit (Amersham Life Science). The composite nucleotide sequence of the RP2 exons and at the exon-intron boundaries is shown in figure 1. The derived sequence of RP2 polypeptide was identical to that reported elsewhere (Schwahn et al. Schwahn et al., 1998).

Display large version of this figure Display high quality version of this figure

Figure 1

Composite nucleotide sequence showing RP2 exons, including the coding region, and the exon-intron boundaries. The numbers on the right refer to the amino acid residues of the predicted RP2 protein.

The complete sequencing of RP2 exons and their corresponding exon-intron junction regions in 51 North American XLRP patients revealed sequence changes in five individuals (fig. 2 and Table 1). All of the alterations were identified in the coding region: a 2-bp insertion in exon 1, a 13-bp deletion in exon 2, a nonsense mutation in exon 2, a 7-bp insertion in exon 2, and a 2-bp insertion in exon 4. Except for the C→T change at nucleotide 358 (arginine codon 120 in exon 2), resulting in a nonsense codon, the remaining four changes are deletions or insertions that would cause a frameshift. Therefore, all changes are predicted to result in a truncated RP2 protein. One of the patients (A1137) has an additional sequence alteration (T→G at nucleotide 322, leading to a Cys108Gly change); however, because this individual also has a 13-bp deletion nearby, we did not determine whether the T→G alteration may represent a disease-causing substitution. Each sequence change segregated in complete concordance with the disease in the respective family members that were available for the study (Table 1). We suggest, on the basis of the nature of mutations and their cosegregation in respective families, that these sequence changes are causative RP2 mutations.

Display large version of this figure Display high quality version of this figure

Figure 2

Representative sequencing gels showing two of the RP2 mutations identified in this report. Sequences in the region of causative mutations are shown. The boxed sequence indicates the 2-bp insertion in patient A514. The location of the 13-bp deletion in patient A1137 is indicated by the horizontal bar. This patient also has a nucleotide substitution, indicated by an asterisk (*).

This is the first report demonstrating the presence of the RP2 subtype in North American families with XLRP. In addition to reporting five novel RP2 mutations, our study addresses several significant issues:

Although much of the genetic and phenotypic complexities of XLRP have yet to be resolved, the cloning of RPGR and RP2 genes represents a milestone in RP research. Identification of mutations in these two genes in many XLRP families provides renewed hope for more-precise diagnosis and better genetic counseling for this devastating disease.