Introduction

Nance-Horan syndrome (NHS) is a rare X-linked disorder1,2,3 characterised by (1) bilateral congenital cataracts associated with microcornea and usually responsible for severe visual impairment; (2) distinctive dental anomalies, mainly unusual dental morphology; (3) characteristic facial features; and (4) mental impairment in 30% of cases.4 Heterozygous females show milder signs of the disease, typically posterior Y-sutural lens opacities and/or slight dental anomalies. Linkage studies have localised the NHS gene to Xp22.25,6 and more recent unpublished data have placed the gene in a 2 cM interval between markers DXS43 and DXS999 (Javed et al, personal communication). Francis et al.7 recently identified a locus for isolated X-linked cataract (without the other features of NHS but with cardiac abnormalities in some patients) mapping to Xp22 (Zmax for DXS8036), within a 3 cM interval flanked by DXS9902 and DXS999 and encompassing the NHS interval. This suggests allelic heterogeneity and strongly supports the existence of a gene for congenital cataract in this Xp region. Several genes have been identified in this region and we previously reported the exclusion of RAI2 as the causative gene for NHS.6 Here we report the results of linkage studies performed in 11 multiplex families refining the localisation of the gene on the telomeric side. We also report the results of sequencing and SSCP analysis of five genes, SCML1, SCML2, STK9, RS1 and PPEF1, considered as candidates on the basis of their position in the critical interval.

Patients, materials and methods

Thirteen independent multiplex NHS families and three sporadic cases (two males and one symptomatic heterozygous female) were recruited from various countries in the world, including previously reported cases.3,8 Male patients and heterozygous females were diagnosed according to precise clinical criteria as previously reported.5

DNA was extracted from blood lymphocytes of each consenting family member according to standard procedures. Microsatellite marker assay and linkage analysis were performed as previously described.5 Linkage analysis was performed in 11 multiplex families, including the four families tested in an earlier study5 in which additional markers and additional individuals (III-5 in family 3 and III-11 in family 4) were analysed (Figure 1).

Figure 1
figure 1

Pedigrees of the eleven families included in the linkage analysis. Filled squares denote affected males and dotted circles denote heterozygous females. Female carrier status was assigned from family history (obligate carriers) or from clinical signs (lens opacities and/or typical dental anomalies, in most cases limited to infraclinical findings).

Full size image

Mutation screening in all coding exons of the five candidate genes was performed in 12 unrelated NHS patients. DNA was amplified by polymerase chain reaction (PCR) by using 250 ng template DNA, 50 pmoles gene-specific primers (primer sequences are available upon request) and 2.5 U Taq DNA polymerase (Promega). The PCR products were analysed by SSCP analysis (GenePhor system, Amersham-Pharmacia) for STK9 (20 exons) and by sequencing using a PRISM BigDye Terminator Cycle sequencing kit and an ABI 377 automated DNA sequencer (Perkin-Elmer Applied Biosystems) for SCML1 (6 exons), SCML2 (15 exons), RS1 (6 exons) and PPEF1 (15 exons).

Results and discussion

Linkage analysis showed that the disease-causing gene mapped to the Xp22 region in all families tested and a combined maximum two-point lod score of 9.94 at θ=0 was obtained at the RS1 locus (Table 1). A recombination with the marker DXS1195 was observed in two independent families, refining the localisation of the NHS gene on the telomeric side. These results combined with the previous data reduce the NHS gene interval to around 1 Mb between DXS1195 on the telomeric side and DXS999 on the centromeric side (Figure 2). In view of the recent knowledge on the human genome sequence,9 the NHS gene is now placed in the Xp22.13 region.

Table 1 Results of pairwise Lod scores
Full size table
Figure 2
figure 2

Schematic map of the Xp22.13 region, linkage intervals and schematic position of the candidate genes.

Full size image

Six genes (SCML1, RAI2, SCML2, STK9, RS1 and PPEF1) have already been identified in this newly defined interval and were considered as candidates for NHS by virtue of their position (Figure 2). Although RS1 is expressed exclusively in the sensory neuroretina and is responsible for X-linked juvenile retinoschisis,10 an ocular condition different from the eye problems observed in NHS, it was considered as a possible candidate because the maximum lod score was obtained at this locus. None of the other five genes located in the NHS critical interval has so far been associated with a human condition. We previously reported that no mutations were found in RAI2 (retinoic-acid-induced 2) in nine NHS patients.6 SCML1 and SCML2 (sex comb on midleg-like-1 and 2), which are mammalian homologues to Scm, a member of the Drosophila Polycomb group of genes,11,12 are probably involved in transcriptional repression of HOX genes, and mutations in these genes may cause developmental malformations (though not similar to the spectrum of anomalies seen in NHS). However, their precise role in human embryo development remains to be elucidated. STK9 (serine–threonine kinase 9) encodes a serine–threonine protein kinase expressed in a variety of tissues,13 and there are several examples, such as Coffin–Lowry syndrome, showing that serine–threonine kinases may be involved in the pathogenesis of genetic disorders. The PPEF1 (protein phosphatase with E-F hand motif) gene is highly homologous to the Drosophila retinal degeneration C (rdgC) gene and encodes a serine-threonine phosphatase which is selectively expressed in sensory neurons of the dorsal root ganglia and neural crest-derived cranial ganglia.14 The coding sequence of SCML1, SCML2, STK9, RS1 and PPEF1 was analysed in 12 unrelated NHS patients. No mutations or polymorphic sequence changes were identified in any of the patients tested. On the basis of these results we conclude that it is very unlikely that any of these five genes is involved in NHS and further analysis will be necessary to identify the molecular basis of this condition. Similarly, SCML1, SCML2, PPEF1 and STK9 remain candidate genes for other diseases mapping to Xp22.13.