Fine mapping and clinical reevaluation of a Brazilian pedigree with a severe form of X‐linked mental retardation associated with other neurological dysfunction

X-linked mental retardation (XLMR) is one of the most common human birth defects, occurring with an incidence of approximately 1.8 per thousand [Herbst and Miller, 1980; Crow and Tolmie, 1998]. It is divided into syndromic and nonsyndromic varieties depending upon the presence or absence of characteristic physical or biochemical abnormalities [Toniolo, 2000]. To date, at least 23 genes responsible for the syndromic forms and more than 12 genes responsible for nonsyndromic forms of XLMR have been described [Chelly and Mandel, 2001]. However, there are likely more than 100 different loci involved inXLMR, so thatmany forms remain yet to have the responsible genes identified [Toniolo, 2000] and possibly mutations in some of these loci will cause both syndromic and non-syndromic forms of XLMR [Billuart et al., 1998; Bergmann et al., 2003; Philip et al., 2003]. In 1993, we reported linkage analysis in a Brazilian family with a severe form of XLMR [Passos-Bueno et al., 1993; Fig. 1]. In our original work, RFLP mapping positioned the gene between DXS84 and DXYS1X, with a maximum LOD score of 3.66 at DXS14 (located in Xp11.21). In the present study, we decided to take advantage of higher precision microsatellite markers in order to narrow down the critical interval for the X chromosome gene responsible for the condition in this pedigree, as part of an effort to identify the causative mutation. In order to better understand the syndrome, we also undertook a more thorough clinical examination of the affected patients. Finally, we have started a candidate gene-based search in the critical interval for causative mutations. A total of 25 microsatellites mapped within Xp11 and Xq21 wereanalyzed onDNAsamples of all available familymembers (Fig. 1). Informative recombinations were noted in family members II-3 (betweenmarkersDXS8080andDXS988)and II5 (between markers DXS983 and DXS986; Fig. 1). Based upon these, wewere able to redefine the critical region of the disease gene segregating in this genealogy between the markers DXS8080 and DXS986, which are located in Xp11.2 and Xq13, respectively. This genetic interval of approximately 18.9 cM contains more than 220 genes based upon current human genome database information, including many uncharacterized EST clusters and computer predicted genes (NCBI). We have tested 18 candidate genes mapped within the critical interval by sequence analysis of exons and flanking intron sequences in an attempt to identify the causative mutation. Using information from publicly available genetic databases (NCBI, ENSEMBL), candidate genes were selected on the basis of expression in fetal and/or adult human brain, along with possible homology to genes already known to be responsible for either various forms of mental retardation associated or not with other neurological dysfunctions. The following genes were analyzed: AR, CHIC1, CITED1, Dach2, EphrinB1,ARHGEF9,GPR23,HCA1,HCA3, ITM2A,KLHL4, NE-DLG3, NLG3, SNX12, TAF9L, ZNF261, ZNF6, TM4SF2. No obvious pathologic mutations in regions important for either protein coding or intron splicing in any of the geneswere found. Therefore, it is unlikely that mutations in any of these genes are responsible for the phenotype unless there is an alteration in the promoter or other regulatory regions at one of these loci. The genetic interval for the disease gene in this family shows an overlap with the critical interval for Allan–Herndon– Dudley syndrome (AHDS; OMIM 309600) which has been mapped to Xq12-Xq13 [Schwartz et al., 1990; Bialer et al., 1992]. AHDS is characterized by mental retardation with spastic paraplegia, dysarthria, and muscular atrophy, along with normal longevity [Stevenson et al., 1990; Bialer et al., 1992]. Affected family members of the present family are characterized by severe mental retardation and muscle atrophy, normal lifespan, and urinary and fecal incontinence, in combination with normal testicular volume, normal head circumference, and the absence of any metabolic or hematologic abnormality [Passos-Bueno et al., 1993]. Although there is some phenotypic variability, all patients of this family share a very similar clinical picture andnone of themwere able towalk or learn how to speak. We recently reexamined two of the affected individuals (III-8 and III-9, ages 25 and 23; Table I). Other affected members of the family live far away from our center and they were not available for clinical reevaluation. The results of the clinical workup revealed the presence of spastic paraplegia besides other neurological dysfunctions not reported by us in our original work (Table I) and we now suggest that this pedigree might represent another familial case of AHDS. In this case, the candidate region for this disease gene would be restricted between DXS106 (Xq12) and DXS986 (Xq13), an interval of 3 cM (10Mb). While still a large genetic interval, this region is estimated to contain only 100 genes and is therefore more amenable to a candidate gene approach (NCBI). The identification of other AHDS families would also be useful to further narrow the genetic interval to a reasonable size. As an example, the family recently reported byClaes et al. Grant sponsor: FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo) (to M.R.P.B. and M.Z.); Grant sponsor: FAPESP-CEPID (Centro de Pesquisa, Inovação, e Difusão) (to M.R.P.B. andM.Z.); Grant sponsor: PRONEX (Programa de Apoio a Núcleo de Excelência) (to M.R.P.B. and M.Z.); Grant sponsor: CNPq (Conselho Nacional de Desenvolvimento Cientı́fico e Tecnológico) (to M.R.P.B. and M.Z.).