664 J Med Genet 1999;36:664–669 J Med Genet: first published as 10.1136/jmg.36.9.664 on 1 September 1999. Downloaded from Original articles A new gene (DYX3) for dyslexia is located on chromosome 2 Toril Fagerheim, Peter Raeymaekers, Finn Egil Tønnessen, Marit Pedersen, Lisbeth Tranebjærg, Herbert A Lubs Abstract skills, and adequate schooling.1 In 1950, Developmental dyslexia is a specific read- Hallgren2 reported that more than 80% of chil- ing disability aVecting children and adults dren with dyslexia in Stockholm schools had who otherwise possess normal intelli- other family members with dyslexia. Subse- gence, cognitive skills, and adequate quent studies have shown that there is both a schooling. DiYculties in spelling and physical and genetic basis for many children reading may persist through adult life. and adults with severe reading disability and Possible localisations of genes for dyslexia that the most frequent deficit is in phonological have been reported on chromosomes 15 coding.34This is manifested by decreased abil- (DYX1), 6p21.3-23 (DYX2), and 1p over the ity to pronounce letter strings (or pseudowords) last 15 years. Only the localisation to correctly that have not been encountered previ- 6p21.3-23 has been clearly confirmed and ously. Recent studies using new methods have a genome search has not previously been provided a clearer view of the diVerence in the carried out. We have investigated a large processing of written words in dyslexics and Norwegian family in which dyslexia is non-impaired readers. In none of the studies, inherited as an autosomal dominant trait. however, were the details of the family history A genome wide search for linkage with an known, and gene localisation studies were not average 20 cM marker density was initi- carried out. Using whole brain magnetoen- ated in 36 of the 80 family members. The cephalograpy,5 it was shown that in response to linkage analysis was performed under reading lists of real and non-words normal Department of three diVerent diagnostic models. Linkage readers first activated the posterior occipital http://jmg.bmj.com/ Medical Genetics, analysis in the family identified a region in cortex in 150 ms, then activated the left tempo- University Hospital of 2p15-p16 which cosegregated with dys- roparietal area by 184 ms. Six dyslexics, Tromsø, N-9038 lexia. Maximum lod scores of 3.54, 2.92, however, failed to activate the temporoparietal Tromsø, Norway and 4.32 for the three diVerent diagnostic region in the first 200 ms and instead activated T Fagerheim M Pedersen models were obtained. These results were the right or left inferior frontal gyrus. This L Tranebjærg confirmed by a non-parametric multi- region contains Broca’s area and its activation H A Lubs point GENEHUNTER analysis in which suggests the use of guessing as an alternative on September 24, 2021 by guest. Protected copyright. the most likely placement of the gene was processing strategy. Using functional magnetic Neurogenetics ina4cMinterval between markers resonance imaging (fMR)6 and a carefully con- Laboratory, D2S2352 and D2S1337. Localisation of a Department of structed, increasingly complex series of reading Biochemistry, Born gene for dyslexia to 2p15-16, together with tests, 29 dyslexics were found to show relatively Bunge Foundation, the confirmed linkage to 6p21.3-23, con- less posterior activation (including Wernicke’s University of Antwerp stitute strong evidence for genetic area, the angular gyrus, and striate cortex), and (UIA), Belgium heterogeneity in dyslexia. Since no gene increased activation anteriorly (inferior frontal P Raeymaekers for dyslexia has been isolated, little is gyrus), compared to non-impaired readers. known about the molecular processes Center For Reading These studies show the use of a diVerent Research, Stavanger, involved. The isolation and molecular processing pathway in response to phonological Norway characterisation of this newly reported tasks and the presence of an altered temporal F E Tønnessen gene on chromosome 2 (DYX3) and DYX1 response in dyslexics. In addition, impaired dis- will thus provide new and exciting insights crimination of both rapidly presented visual and Division of Genetics, into the processes involved in reading and Department of auditory non-verbal information have been 7–9 Pediatrics, Miami, FL, spelling. reported, and a more generalised defect in the USA (J Med Genet 1999;36:664–669) rapid processing of information may also play a H A Lubs Keywords: developmental dyslexia; reading disability; role in some or all instances of dyslexia. Many children under 2 years of age with Correspondence to: linkage analysis; chromosome 2 Dr Fagerheim. damage to the language areas of the left hemi- sphere can go on to develop satisfactory Revised version received Developmental dyslexia is a specific reading language ability. This observation implies that 28 April 1999 Accepted for publication disability aVecting children and adults who specialisation of the brain mechanism responsi- 5 May 1999 otherwise possess normal intelligence, cognitive ble for language function is not completed at A new gene for dyslexia is located on chromosome 2 665 J Med Genet: first published as 10.1136/jmg.36.9.664 on 1 September 1999. Downloaded from I ?? 81 79 79 70 77 76 76 74 70 64 II 1 2 43 65 7 8 91415 41 12 46 11 11 11 42 12 12 21 53 34 51 51 55 23 53 64 18 12 22 18 18 66 32 12 46 22 24 21 22 22 24 14 24 21 38 36 33 38 38 33 38 36 34 21 21 43 21 21 13 11 21 42 23 22 32 23 23 32 13 22 32 53 55 36 43 38 35 49 5047 39 44 47 41 36 35 34 32 30 III 1 34567 8 9 10 12 15 16 17 18 21 23 24 30 31 32 11 41 14 11 11 51 11 14 24 11 11 11 25 21 11 22 12 24 53 55 53 53 13 55 41 15 15 55 35 36 56 34 83 18 12 28 13 32 83 13 78 86 86 16 85 24 14 26 24 22 22 42 24 13 24 21 41 22 22 24 44 42 22 41 86 36 33 16 35 48 85 37 23 83 83 33 13 63 33 64 14 22 14 42 24 43 14 22 43 11 11 23 33 14 24 12 33 23 23 23 21 33 31 23 32 33 33 22 33 23 23 22 25 21 11 13 19 18 16 17 15 9 13 11 5 6 7 9 11 12 13 20 21 23 32 33 Marker order: 11 11 36 11 11 15 11 42 12 12 15 Telomere 54 54 54 55 33 35 53 54 54 53 53 38 38 32 18 32 13 12 37 17 18 68 D2S1352 IV 12 12 11 22 43 21 23 14 24 24 44 36 36 43 36 58 34 38 72 32 31 31 D2S2352 42 42 13 12 43 24 23 24 24 23 33 D2S378 33 33 32 23 13 23 23 33 23 23 23 D2S2279 D2S2183 Positive on tests for dyslexia + history of reading problems D2S1337 D2S393 Positive tests only Positive history only Centromere Figure 1 Pedigree showing only family members who were included in the linkage analysis. Pedigree numbers from the complete pedigree are retained. The numbers given above the symbol are the ages at the time the study was initiated. The proband (III.9) is indicated by an arrow. Boxed haplotypes indicate the haplotype cosegregating with dyslexia. Two critical recombinants were observed in adult patients (II.1, III.7) and their oVspring (III.3,IV.9) establishing the DYX3 candidate between markers D2S2352 and D2S1337. In generation IV,IV.7to IV.33,who were below the age of 20, were excluded from the model 3 linkage analysis. Chromosome analysis of III.31 was normal. that age and that the plasticity of these complex genes may be involved. In order to identify neural networks permits their adaptive reor- possible additional genes predisposing to ganisation. In contrast, the majority of teenag- dyslexia, we have carried out a genome search ers who experience comparable brain damage and performed linkage analysis in a large Nor- http://jmg.bmj.com/ never recover normal language function.10 A wegian family in which dyslexia is clearly molecular test, or tests, for dyslexia, which inherited. The pedigree is shown in fig 1. would result from cloning genes for dyslexia, would allow earlier diagnosis of children at Methods high risk for dyslexia. This in turn would TESTS FOR DYSLEXIA permit institution of therapy while the lan- The test instrument used to test for dyslexia in guage areas were at an earlier, more plastic the Norwegian population was developed at stage of development. the National Center for Reading Research. The on September 24, 2021 by guest. Protected copyright. The localisation of a susceptibility locus for assessments of both orthographic and phono- dyslexia to 6p21.311 has been confirmed in logical abilities are very similar to those used in three independent studies.12–14 Localisation to the genetic studies of Olson et al23 and the 6p, however, was not found in a large sib pair Woodcock Reading Mastery Tests.24 However, study,15 or in the large family studied by Sawyer a measure of reaction time was also included in et al.16 The linkage with chromosome 15p three of the tests. markers reported in 198317 was not initially The present dyslexia test battery was based confirmed,11 18 19 but two recent reports suggest on selected subtests from a series of tests given a localisation to D15S143, with lod scores of to Norwegian school children with possible 3.1512 and 1.38.20 This marker, however, is 49 reading problems to test their reading level and cM distal to the centromere and thus does not ability (KOAS25 and KOAP26).