Members of this large French non-consanguineous family (RP470) were identified with CRD which segregated as a dominant trait (adCRD). There were 9 affected patients out of 22 in 4 generations (Figure 1). Examination included assessment of visual acuity, slit lamp biomicroscopy, direct funduscopy and full field electroretinography. There was no evidence of extra-ocular signs of disease indicating that CRD was non syndromic.

Informed written consent and peripheral blood samples were obtained from 22 examined family members. The investigators followed the tenets of the Declaration of Helsinki. Genomic DNA was isolated from 10 ml peripheral blood leucocytes using standard salting out procedure 13 . The DNA samples were quantified by a spectrophotometer and diluted to 25 ng/μl for PCR amplification. PCR was carried out in a 25 μl final volume containing 50 ng genomic DNA, 5 picomoles of each primer, 0.2 mM dNTPs (MP Biochemicals), 2 mM MgCl₂, PCR buffer and 1 unit of DNA polymerase (AmpliTaq Gold; Applied Biosystems, Foster city, CA). Initial denaturation at 95°C for 10 minutes was followed by 35 cycles of denaturation at 94°C for 30 seconds, specific annealing temperature for 30 seconds, and extension at 72°C for 1 minute. A final extension step was performed at 72°C for 10 minutes. The PCR products were diluted and mixed with Genescan 400HD ROX size standard and subsequently analysed on an Applied Biosystems 3130xL genetic analyser (Applied Biosystems, Foster city, CA).

Genotyping was performed using 2 to 3 polymorphic commercially available microsatellite markers from ABI PRISM Linkage Mapping Set version 2.5 (Applied Biosystems, Foster city, CA), within or contiguous to known adRP and adCRD genes, and within the locus CORD12. Results were analysed with GeneMapper software (version 4.0, Applied Biosystems, Foster city, CA). Segregation of the markers among the family members was examined.

Two-point LOD scores were calculated with Superlink-online http://bioinfo.cs.technion.ac.il/superlink-online/. The phenotype was analyzed as an autosomal dominant and fully penetrant trait with an affected allele frequency of 0.0001. Family and haplotype data were generated using Cyrillic software (version 2.1.3; Cherwell Scientific, Oxford, UK).

To map the disease locus, a genome-wide scan was performed by the Centre National de Génotypage (CNG, http://www.cng.fr) by genotyping 262,264 SNPs (GeneChip Mapping 250 K Nsp Array, Affymetrix, Santa Clara, CA). Results were analyzed using TASE (Transmitted Allele Search Engine) a home-made software which compared every SNP between each individuals in the family.

The first test, named Common Allele to All Affected individuals (C3A), highlighted the common allele to all affected patients within the family. The second test, Transmitted Allele to All Children (TAAC), estimated the specific allele carried by the affected parent in a nuclear family (parents + child) and transmitted to the affected child. Two consecutive mismatched SNPs limited the size of the locus. Only the regions longer than 1 Mb were considered.

Coding exons and adjacent intronic sequences of candidate genes were sequenced with an Applied Biosystems 3130xL genetic analyser (Applied Biosystems, Foster city, CA) using BigDye Terminator cycle sequencing ready reaction kit V3.1 (Applied Biosystems, Foster city, CA) following manufacturer's instructions. Primer pairs and PCR conditions are available on request. Sequence analysis and mutation identification were performed using Collection and Sequence Analysis software package (Applied Biosystems, Foster city, CA).

The pedigree of the four generations family is shown in Figure 1. The 9 affected patients revealed features of adCRD with intra-familial variable phenotype including progressive loss of the visual acuity, typical bone spicule-shaped pigmentary deposits in the macular area or macular atrophy, moderate night blindness and reduced electroretinogram (ERG) responses (Table 1). The proband (III:1) showed patches of atrophy in the macular area with a few pigment deposits, attenuation of retinal arterioles and temporal pallor of the optic disc (Figure 2).

Microsatellite markers for the 21 adRP genes http://www.sph.uth.tmc.edu/Retnet, the 3 most frequent adCRD genes (CRX, GUCY2D and PRPH2/RDS) 7 8 9 10 and a fourth adCRD gene, GUCA1A, were used to genotype family members, and to search for co-segregation of the markers with the disease phenotype. All these candidate genes were excluded. We then performed a genome wide scan using Affymetrix 250 K microarrays and genotypes were analysed with the TASE software. No linkage was found for most chromosomal regions except for a large region located on chromosome 2q24.2-2q33.1. The boundaries of the locus were determined by SNP exclusion between SNPs rs174240 and rs4619591 and encompassed a 41.3-Mb region (Figure 3).

Microsatellite markers were then used to confirm linkage with the locus. We genotyped all 22 members of the family with 8 microsatellite markers located on 2q24.2-2q33.1 (Figure 1). All affected patients had a common haplotype and the boundaries of the region were determined by recombination events that occurred in affected individuals III:1, III:4, III:9, IV:4, IV:9 and healthy individual III:7. The proximal boundary was defined by the recombination event between markers D2S142 and D2S306 in affected patient IV:9 and distal boundary by the recombination event between markers D2S117 and D2S325 in healthy individual III:7 (Figure 3). Using Superlink software, we found a maximum LOD score of 2.86 at θ = 0 for the marker D2S118, defining a new locus named CORD12. The markers D2S142, D2S325, D2S2361 and D2S126, outside the locus, gave negative LOD scores (Table 2).

The CORD12 41.3-Mb interval contains 280 genes. None of them were previously reported in adCRD or adRP. However the interval does contain two previously described autosomal recessive RP genes, namely CERKL and BBS5, which cause autosomal recessive RP and Bardet-Biedl syndrome, respectively 14 15 . All exons and flanking intron regions were sequenced but no mutation was found. Within the CORD12 locus, three other candidate genes were also sequenced. KLHL23 has strong similarities with the recently described gene KLHL7 responsible for adRP 16 . NEUROD1 regulates development and maintenance in the visual system 17 . SF3B1 is a splicing factor 18 . Other essential components of the spliceosome, PRPF31, PRPF3, PRPF8, PAP1 and SNRNP200, have been associated with adRP 19 20 21 22 . No disease causing mutations were detected in KLHL7, NEUROD1 and SF3B1.