Genetics of specific language impairment

Genetics of specific language impairment

Prostaglandins, Leukotrienes and Essential FattyAcids (2000) 63(1/2),101^107 & 2000 Harcourt Publishers Ltd doi:10.1054/plef.2000.0199, available onli...

162KB Sizes 1 Downloads 119 Views

Prostaglandins, Leukotrienes and Essential FattyAcids (2000) 63(1/2),101^107 & 2000 Harcourt Publishers Ltd doi:10.1054/plef.2000.0199, available online at http://www.idealibrary.com on

Genetics of specific language impairment J. Nasir,1 W. Cohen,1 H. Cowie,1 A. Maclean,1 J.Watson,2 J. Seckl,3 A. O'Hare2 1

Human Genetics Unit, Molecular Medicine Centre,Western General Hospital, Edinburgh, UK Royal Hospital for Sick Children, Edinburgh, UK 3 Molecular Endocrinology Section, Dept of Medical Sciences, University of Edinburgh, UK 2

INTRODUCTION Language acquisition is a remarkable feature of early childhood development, following a constant and predictable course and involving discrete regions of the brain including Broca's area and Wernicke's area. By the age of five, the fundamentals of grammar are already in place.1 Between 25 and 50% of receptive and expressive vocabulary is attributable to genetic factors.2±4 Specific language impairment (SLI) refers to a condition where a child is significantly behind in language acquisition and development, despite having normal hearing, normal IQ, and showing no evidence of any other clinical conditions.5 SLI is clinically distinct from other childhood conditions which can affect language, such as autism and dyslexia. It is estimated that 5.4% of children suffer from SLI,1 and the condition is prevalent across a wide range of social backgrounds including families with stable homes and adequate opportunities to develop language skills. The condition can resolve during normal development but in some cases it can persists through adolescence and into adulthood.6±8 Adolescents and adults not only suffer through failure to thrive in school,9 but SLI has also been associated with behavioural problems such as delinquency and truancy and it is suggested there is overrepresentation of SLI in the prison population. However, early therapeutic intervention during childhood can be effective.10±12 The aetiology of SL1 is poorly understood and a variety of explanations have been offered, including the possibility of impaired auditory processing.13 For example, children with SLI tend to perform poorly in simple tasks Received 7 June 2000 Accepted 14 June 2000

UNDERSTANDING SLI The ability to communicate requires the integration of a variety of components of language. For example, when a child is listening to speech he or she must first be able to hear the words and then process them in the brain. In its simplest form, a model for interpreting language would consist of an adequate hearing system, the ability to process the auditory signal, and an understanding of grammar and vocabulary. When producing speech, the child needs to be able to access the grammatical rules and vocabulary, process this information and produce a sound. However, problems can occur at different levels. For example, a child who knows a word (that is, he/she would be able to point to a correct object when required) may be unable to name it spontaneously. Such a problem can arise when a child has difficulty with lexical retrieval, commonly referred to as word finding difficulty. Alternatively, when the comprehension of lexical items is impaired the child may be unable to point to the correct object. In this case, the child may have failed to store an adequate representation of the word meaning, indicating a difficulty with the semantic level of language.

DIAGNOSIS OF SLI

Correspondence to: Jamal Nasir, Human Genetics Unit, Molecular Medicine Centre,Western General Hospital, Edinburgh, UK.Tel.: 0131 6511060; Fax: 0131 6511059; E-mail: [email protected]

& 2000 Harcourt Publishers Ltd

which require them to discriminate between brief tones or a series of sounds occurring in rapid progression.14 Recent studies support a genetic basis for SLI and molecular studies are underway to seek out genes which underlie this complex condition. This could eventually lead to improved diagnosis and help identify children at increased risk so therapeutic intervention can be offered.

SLI is a clinically heterogeneous condition with an increased prevalence in boys, and families of probands are approximately four times more likely to have SLI

Prostaglandins, Leukotrienes and Essential FattyAcids (2000) 63(1/2), 101^107

102

Nasir et al.

compared to the rest of the population.1,4 Variations in clinical standards of diagnosis reflect the complexity of this disorder and the diagnostic criteria which are applied.2,5,6 In order to diagnose SLI, the following criteria have to be met: (1) language development significantly lower than expected (2) language scores at least one standard deviations below the non-verbal IQ score (3) no other clinical symptoms such as poor hearing or autism. A non-verbal IQ of less than 85 is the most commonly used exclusion clause.1 According to the DSMIII-R (American Psychiatric Association) definition, a distinction is made between expressive and receptive disorders.6 This has led to much debate on other subtypes of SLI, including purely grammatical SLI.15±17 Others have argued that these are merely different components of the same condition with a continuum of clinical phenotypes. We and others have adopted the ICD-10 (International classification of diseases, WHO) criteria which impose a more stringent definition on SLI compared to DSMIII-R. A variety of tests have been developed to accommodate the range of phenotypes associated with SLI (Table 1), taking into consideration, age, non-verbal ability, vocabulary, grammar, comprehension, reading, writing, phonology, production of speech sounds (articulation) and a host of other parameters. For children who have improved scores as a direct consequence of speech and language intervention, the results of assessments administered when the children were younger are valid indicators of language difficulties. A uniform, clear and precise determination of the phenotype remains a vitally important challenge in understanding this complex condition. GENETICS OF SLI Common disorders are a result of both genetic and environmental contributions. Heritability is a measure of the proportion of the phenotypic variance which is due to genetic variance. Twin studies provide the ideal opportunity to distinguish between genetic and environmental factors. Studies of SLI twins consistently demonstrate a significantly greater concordance rate for monozygotic (which share the same genes) versus dizygotic twins (which share only half their genes),1,4 indicating a strong genetic predisposition for SLI. However, the concordance rates for MZ twins rarely each 100% due to the environmental component in SLI. In one study, Bishop and colleagues found 8 of 27 DZ twins were concordant for SLI compared to 34 of 63 MZ twins.2 In a more recent

Table 1 Tests for receptive and expressive language 1. Vocabulary a. First words and first sentences test18 Age range:18 months to 36 months b. British PictureVocabulary Scale19 Age range: 2 years to18 years c. Renfrew Language Scales20 Age range: 312 years to 812 years d. Reynell Developmental Language Scales21 Age range: 2 years to 7 years e. Subtests in theTest of Language Development: Primary, 3rd ed.22 Age range: 5 years to 9 years f. Subtests in theTest of Language Development:Intermediate, 3rd ed.23 Age range: 8 years to16 years g. Subtests in theTest of Word Knowledge24 Age range: 4 years to17 years 2. Grammar h. Renfrew Language Scales20 Age range: 312 years to 812 years i. Reynell Developmental Language Scales21 Age range: 2 years to 7 years j. Subtests in theTest of Language Development: Primary, 3rd ed.22 Age range: 5 years to 9 years k. Subtests in theTest of Language Development:Intermediate, 3rd ed.23 Age range: 8 years to16 years l. Subtests in theTest of Word Knowledge24 Age range: 4 years to17 years m. Subtestsin the Clinical Evaluation of Language FundamentalsRevised25 Age range: 5 years to16 years n. Test for Reception of Grammar26 Age range: 4 years to12 years 3. Speech production skills o. Edinburgh ArticulationTest27 Age range: 3 years to 6 years 4. Hearing and auditory processing skills p. Children'sTest of Non-Word Repetition28 Age range: 4 years to 9 years 5. Non-verbal IQ q. Raven's Progressive Matrices29 Age range:11years to adult r. Subtests of the British Ability Scales ^ 2nd ed.30 Age range: 4 years to adult 6. Pervasive difficulties s. Children's Communication Checklist13 Age range: 7 years to 9 years

study, language acquisition of 3000 twins born in 1994 was tested and it is estimated that up to 73% of the variation is due to genetic factors.31 Fisher and colleagues32 have described a large three generation family (KE family) with a severe speech and language disorder,33 which appears to be segregating as an autosomal dominant trait. These individuals have specific difficulties in applying grammatical rules and also suffer from severe orofacial dyspraxia. Twenty seven members of this extended pedigree were genotyped using microsatellite markers from across the genome. This

Prostaglandins, Leukotrienes and Essential FattyAcids (2000) 63(1/2), 101^107

& 2000 Harcourt Publishers Ltd

Genetics of specific language impairment 103

revealed linkage to chromosome 7q, close to the location of the cystic fibrosis gene, CFTR. Focusing on this region, the investigators have pinned down the location of the gene, dubbed SPCHI, to a 5.6 cM interval flanked by D7S643 and D7S2459. Fortunately, this is one of the best characterized regions of the genome and, it should not be long before the SPCHI gene is cloned. A number of genes of interest map to this interval, including WNT-2 and other signaling molecules. Intriguingly, the same chromosomal region has also been linked to autism. In a recent paper, Ashley-Koch and colleagues describe a family with an expressive language disorder and autism, associated with a cytogenetic inversion involving the chromosomal region 7q, including the region implicated in SLI.34 GENETIC ANALYSIS OF COMPLEX DISEASES Co-segregation of random markers with a disease in families has proved to be a successful strategy for linkage analysis of simple Mendelian disorders associated with a single gene, such as cystic fibrosis and Huntington's disease. With rare exceptions such as the KE family, the genetics of SLI is complex. Family and pedigree studies consistently reveal complex patterns of inheritance due to a variety of factors. Similar to other complex disease such as diabetes or hypertension, the inheritance patterns can be influenced by genetic heterogeneity (different genes contributing to the same phenotype in different families), incomplete penetrance (a genetic mutation may not result in a phenotype), phenocopies (where the disease arizes without a known genetic cause) and oligogenic effects (interactions of several genes leading to a phenotype). The bewildering complexity of genetic diseases is best illustrated by considering autism, where 91±93% of the

phenotypic variation is attributed to genetic factors. Yet, concordance rates for MZ twins are around 60% and much lower for DZ twins, and the search for susceptibility genes has remained elusive. This can partly be explained by reduced penetrance and oligogenic inheritance. It is suggested that genes are interacting multiplicatively rather than additively, thus drastically affecting disease susceptibility even when the contribution of individual genes is only modest. This is reflected in the risk to relatives, which drops off more rapidly than expected across different degrees of relatives. A common strategy for exploring the underlying genetics of complex diseases has been the affected sibpair method, which essentially examines the concordance between phenotypic variance and allele sharing.35,36 It involves examining the number of alleles (0, 1, 2) identical by descent (IBD) between 2 sibs at a particular locus (Fig. 1). Typically, a two-stage genome scan is performed using markers at 10 centimorgan intervals. Regions which show promise of linkage, giving LOD scores of above 2, are examined in further detail using a higher density of markers. The ability to detect a disease susceptibility locus using an affected sib-pair strategy depends upon ls, the relative risk to a sibling of an affected proband versus the population prevalence. ls can vary widely. For example, it is 500 for cystic fibrosis, 145 for autism, 15 for type 1 (insulin dependent, juvenile onset) diabetes, 8.6 for schizophrenia, 3.5 for type II diabetes (insulin independent, maturity onset) and around 7 for SLI. Genetic mapping is easier for traits with high ls (ls>10) than those with low ls (ls52). Generally speaking, the minimum number of sib-pairs required to detect linkage for a disease with a modest ls, is approximately 200.36,37 A genome wide search for human type 1 diabetes susceptibility genes was carried out on just 96 sib-pairs and revealed 18 different

Fig. 1 Allele-sharing identical by descent. Affected relatives inherit a region identical by descent (IBD) more often than expected under random Mendelian segregation.Two sibs can show IBD sharing for zero, one or two copies of any locus, with a 25%, 50%, 25% distribution expected under random segregation. Sib-pairs are expected to share 50% of their genes identical by descent from their parents. A significantly increased sharing ofgenes at a givenlocuswouldimplicatethat regionin disease susceptibility.For simplicity, it is assumed all four parentalalleles (A,B,C,D) can be distinguished.The offspring can share both parental alleles (AC), an allele from the father (A), an allele from the mother (C), or they can share no parental alleles in common. Father, mother and offspring are represented by squares, circles and diamonds, respectively.

& 2000 Harcourt Publishers Ltd

Prostaglandins, Leukotrienes and Essential FattyAcids (2000) 63(1/2), 101^107

104

Nasir et al.

chromosomal regions which showed evidence of linkage to disease.38 However, the power to detect linkage can be increased by increasing the number of sib-pairs. A number of other strategies can also be deployed to increase the power of detecting linkage. First of all, careful and thorough diagnosis of the phenotype is of vital importance. This allows distinct components of the phenotype to be analysed separately. For example, in one study where colon cancer was restricted to cases with extreme polyposis, the trait behaved as an autosomal dominant, allowing positional cloning of the gene.39 Similarly, breast cancer and Alzheimer's disease are rendered genetically more homogeneous by focusing on early-onset cases, and the relative risk from heart attacks is much greater for early onset cases (ls=7 for men and 15 for women) compared with late onset cases (ls=2). Furthermore, requiring the presence of at least three affected relatives within a family, can enhance the chances of finding linkage. Traditionally, linkage analysis of multiplex families, involving simple Mendelian traits, requires the mode of inheritance to be specified and some assumptions regarding allele frequencies and penetrance. However, where the mode of inheritance is unknown, as in SLI, nonparametric methods have been developed which require no assumptions regarding the mode of inheritance, but instead gain power through large sample sizes and allele sharing methods. These methods are based on the principle that sibs sharing more alleles identical by descent (IBD) than expected should be more similar in phenotype than those sharing fewer alleles at a locus influencing that trait. For simplicity, if we assume that all four parental alleles at a given locus can be scored, the probability of a sib-pair sharing 0, 1 and 2 alleles identical by descent is 25%, 50% and 25%, respectively (Fig. 1). Deviation from this Mendelian ratio would be indicative of a disease predisposing allele. This is evaluated by studying s2, which is defined as the variance of D (the difference between phenotypes of two siblings in a pair), and testing the hypothesis that s02 (the variance for sibs sharing 0 alleles IBD)4s12 (sibs sharing 1 allele)4s22 (sibs sharing 2 alleles). a commonly used application of this method involves regression of s2 against observed values of allele sharing.40

QTLs Some traits are purely qualitative and for a disease an individual is categorized as either affected or nonaffected. Such a distinction cannot be applied for traits like intelligence, weight, height and blood pressure, where it is harder to define the boundaries between affected and unaffected. These traits are quantitative in

nature, and measured on a continuous scale. They fail to show Mendelian patterns of segregation, and are under the influence of different loci, defined as quantitative trait loci (QTLs). For the analysis of such traits, including SLI, quantitative measures for a phenotype provide a more powerful strategy than affected sib-pair or affected versus non-affected sib-pair analyses. Moreover, the identification of sib-pairs with extreme discordance enhances the power of detecting a QTL.41,42 However, the utility of these methods is limited where the contributed genetic variance is less than 10%, unless very large sample sizes are employed. SLI lends itself particularly well to QTL analysis. For example, a total language score can be assessed as a quantitative variable. Alternatively, receptive language scores, expressive language scores or no-word repetition performance can be analysed separately. A similar approach applied to dyslexia revealed linkage to a phonological awareness phenotype on chromosome 6p23-21.3 and a single word reading phenotype to chromosome 15.43,44 QTL analysis has been successfully applied to a variety of other conditions, including alcoholism.

GENETICS OF SLI IN SCOTLAND The Scottish population is ideal for genetic studies. It represents a relatively homogeneous and stable population. In addition to the excellent clinical records, family histories are available. We have sought to undertake a study to identify the predisposing genes in SLI by searching through records of language units throughout central Scotland, particularly in the Lothian and Fife regions. We are beginning to select families, through careful selection of a proband, using established ICD-10 criteria. In order to meet ICD-10 criteria, language skills should be at least two standard deviations below a child's age and one standard deviation below their non-verbal IQ, as discussed earlier. We have collected 30 families and plan to collect another 50 families for QTL-based genetic analysis. Family EH010 includes two boys with SLI and a third child with speech and language delay (Fig. 2). The parents, especially the father, experienced reading and writing difficulties at school and a first cousin is suspected to have dyslexia. Our preliminary findings suggest SLI is common, as expected. Family EH010 is a typical of family. Many of the families have multiple affected sibs and often a history of speech and language related problems. It will be of particular interest to determine whether speech and language delay is distinct from SLI.

Prostaglandins, Leukotrienes and Essential FattyAcids (2000) 63(1/2), 101^107

& 2000 Harcourt Publishers Ltd

Genetics of specific language impairment 105

Fig. 2 Segregation of SLI in families. Family EH010 has two boys with SLI and a third child with language delay. In addition, both parents, especially the father, have a history of speech and language problems. A first cousin of the sibs is suspected to suffer from dyslexia.

STANDARDIZED TESTS FOR SLI A variety of standardized tests are available for scoring receptive and expressive language (Table 1). For example, the children's test of non-word repetition tests verbal short-term memory, requiring repetition of non-words such as `blonterstaping' or `confrantually'. This test is an accurate indicator of SLI in children.45 The test for reception of grammar (TROG), a measure of receptive language, is a multiple choice test in which the task is to select a picture to match a sentence. There is a strong correlation between test scores and heritability, based on the performance of MZ and DZ SLI twins on these tests (reviewed in Stromswold).4 For example, non-word repetition performance is 100% heritable (unpublished data), yet IQ is approximately only 50% heritable. Similarly, the Edinburgh Articulation Test, the British Picture Vocabulary Scale and CELF-R scores show strong evidence for heritability.

CLONING GENES FOR SLI Genome wide linkage analysis is proving to be a viable approach in identifying candidate genes for SLI, at least for large extended families such as the KE family, but & 2000 Harcourt Publishers Ltd

since the resolution of this analysis is low, ultimately association studies and linkage disequilibrium methods will be required in order to narrow the regions containing candidate genes and thus facilitate the cloning of these genes. This task will be made easier with the release of the complete human genome sequence shortly. Candidate genes will have to be thoroughly tested in cell culture assays and animal models to study mutations and functional polymorphisms. Genetic variants will also be tested to determine their relationship to less severe speech and language disorders. Genome wide linkage disequilibrium studies represent a powerful approach for mapping genes associated with complex diseases. However, these studies are currently restricted because linkage disequilibrium rarely extends beyond 3 kb, but new approaches will take advantage of the discovery of up to 500,000 single nucleotide polymorphisms (SNPs) and automated procedures for genotyping.46,47 Cloning the genes for SLI and identifying their normal function, will not only be important for SLI, but it would also lead to a better understanding of the fundamental properties of language itself. In the meantime, detailed and careful clinical assessment of SLI is likely to be important in mapping the genes.

Prostaglandins, Leukotrienes and Essential FattyAcids (2000) 63(1/2), 101^107

106

Nasir et al.

Identification of the SPCH1 gene is unlikely to provide a ready answer to how it leads to dyspraxia, but it could potentially pave the way for carrying out screening for mutations in this gene on at risk children. This would enable early medical intervention using well established and successful methods including counseling. Furthermore, functional neuroimaging techniques, including magnetic resonance imaging (MRI) and positron emission tomography (PET), which can provide images of the brain while the subjects are performing specific tasks, represent new ways of understanding this complex condition.48±52 Hopefully, greater resources will also be directed at other childhood conditions likely to be strongly influenced by genetic factors, including attention deficit and hyperactivity disorder (ADHD), autism and dyslexia.

15.

16.

17.

18. 19. 20. 21.

REFERENCES

22.

1. Rice M. L. (ed.) Towards a genetics of language. Lawrence Erlbaum Associates, 1996. 2. Bishop D. V. M., North T., Donlan C. Genetic basis of specific language impairment: Evidence from a twin study. Dev Med Child Neurol 1995; 37: 56±71. 3. Stromswold K. Genetics of spoken language disorders. Hum Biol 1998; 70(2): 297±324. 4. Stromswold K. The heritability of language: a review and metaanalysis of twin and adoption studies. Language (in press). 5. Bishop D. V. M. Is specific language impairment a valid diagnostic category? Genetic and psycholinguistic evidence. Phil Trans R Soc Lond 1994; 346: 105±111. 6. Bishop D. V. M., Edmundson A. Language impaired 4-year olds. Distinguished transient form persistent impairment. J Speech Hear Disord 1987a; 52: 156±173. 7. Bishop D. V. M., Edmundson A. Specific language impairment as a maturation lag: evidence from longitudinal data on language and motor development. Dev Med Child Neurol 1987b; 29: 442±449. 8. Stothard S. E., Snowling M. J., Bishop D. V. M., Chipchase B. B., Kaplan C. A. Language-impaired preschoolers: a follow up into adolescence. J Speech Lang Hear Res 1998; 41: 407±418. 9. Rutter M., Mawhood L. The long-term psychological sequelae of specific developmental disorders of speech and language. In: M. Rutter, P. Caesar, eds. Biological misfactors for psychological disorders. Cambridge University Press, 1991; 233±259. 10. Tallal P., Miller S. L., Bedi G., Byma G., Wang X., Nagarajam S. S., Shreiner C., Jenkins W. M., Merzenich M. N. Language comprehension in language impaired children improved with acoustically modified speech. Science 1996; 271: 81±84. 11. Law J. C., Boyle J. M., Harris F., Harkness A. D. Child health surveillance: an evaluation of screening for speech and language delay. UK. National Health Service Centre for Review Dissemination: University of York, 1998. 12. Almost D., Rosenbaum P. Effectiveness of speech intervention for phonological disorders: a randomized controlled trial. Dev Med Child Neurol 1998; 40: 319±325. 13. Bishop D. V. M., Bishop S. J., Bright P., James C., Delaney T., Tallal P. Different origin of auditory and phonological processing problems in children with language impairment: Evidence from a twin study. J Speech Lang Hear Res 1999; 42: 155±168. 14. Wright B. A., Lombardino L. J., King W. M., Puranik C. S., Leonard C. M., Merzenich M. M. Deficits in auditory temporal and

23.

24. 25.

26. 27.

28.

29. 30. 31.

32.

33.

34.

35. 36.

Prostaglandins, Leukotrienes and Essential FattyAcids (2000) 63(1/2), 101^107

spectral resolution in language impaired children. Nature 1997; 387: 176±178. van der Lely H. K. J., Stollewerk L. A grammatical specific language impairment in children: an autosomal dominant inheritance? Brain Lang 1996; 52: 484±504. van der Lely H. K. J., Stollwerck L. Binding theory and grammatical specific language impairment in children. Cognition 1997; 62: 245±290. van der Lely H. K. J., Rosen S., McClelland A. Evidence for a grammar-specific deficit in children. Curr Biol 1998; 8: 1253±1258. Gillham B., Boyle J., Smith N. First words and first sentences test. London: Hodder and Stoughton, 1998. Dunn L. M., Dunn L. M., Whetton C., Burley J. The British Picture Vocabulary Test II. Windsor: NFER-Nelson, 1997. Renfrew C. Renfrew language scales. England: Winslow Press, 1995. Edwards S., Fletcher P., Garman M., Hughes A., Letts C., Sinka I. Reynell Developmental Language Scale III (The Reading University Edition). Windsor: NFER-Nelson, 1997. Newcomer P., Hammill D. Test of language development ± primary, 3rd ed. England: The Psychological Corporation, 1998a. Newcomer P., Hammill D. Test of language development ± intermediate, 3rd ed. England: The Psychological Corporation, 1998b. Wiig E. H., Secord W. Test of word knowledge. Texas: The Psychological Corporation/Harcourt Brace, 1992. Semel E., Wiig E., Secord W. The clinical evaluation of language fundamentals ± revised. Texas: The Psychological Corporation/ Harcourt Brace, 1987. Bishop D. V. M. Test for reception of grammar, 2nd ed. London: MRC, 1989. Anthony A., Bogle D., Ingram T. T. S., McIsaac M. W. The edinburgh articulation test. Edinburgh: Churchill Livingstone, 1971. Gathercole S. E., Willis C. S., Baddeley A. D., Emslie H. The children's test of nonword repetition: a test of phonological working memory. Memory 1994; 2: 103±127. Raven J. C. Raven's progressive matrices and vocabulary scales. Windsor: NFER-Nelson, 1998. Elliot C. D. British Ability Scales, 2nd ed. Windsor: NFER-Nelson, 1996. Dale P. S., Simonoff E., Bishop D. V. M., Eley T. C., Oliver B., Price T. S., Purcell S., Stevenson J., Plomin R. Genetic influence on language delay in two-year old children. Nature Neuroscience 1998; 1: 324±328. Fisher S. E., Vargha-Khadem F., Watkins K. E., Monaco A. P., Pembrey M. E. Localisation of a gene implicated in a severe speech and language disorder. Nat Genet 1998; 18: 168±170. Vargha-Khadem F., Watkins K., Alcock K., Fletcher P., Passingham R. Praxic and nonverbal cognitive deficits in a large family with a genetically transmitted speech and language disorder. Proc Natl Acad Sci USA 1995; 92: 930±933. Ashley-Koch A., Wolpert C. M., Menold M. M., Zaeem L., Basu S., Donnelly S. L., Ravan S. A., Powell C. M., Qumsiyeh M. B., Aylsworth A. S., Vance J. M., Gilbert J. R., Wright H. H., Abramson R. K., DeLong G. R., Cuccaro M. L., Pericak-Vance M. A. Genetic studies of autistic disorder and chromosome 7. Genomics 1999; 61: 227±236. Lander E. S., Schork N. J. Genetic dissection of complex traits. Science 1994; 265: 2037±2048. Weeks D. E., Lathrop G. M. Polygenic disease: methods for mapping complex disease traits. Trends Genet 1995; 11: 513±519.

& 2000 Harcourt Publishers Ltd

Genetics of specific language impairment 107

37. Ott J. Invited editorial: cutting a guardian knot in the linkage analysis of complex human traits. Am J Hum Genet 1990; 46: 219±221. 38. Davies J. L., Kawaguchi Y., Bennett S. T. et al. A genome wide search for human type 1 diabetes susceptibility genes. Nature 1994; 371: 130±136. 39. Joslyn G., Carlson M., Thliveris A. et al. Identification of deletion mutations and three new genes at the familial polyposis locus. Cell 1991; 66: 601±613. 40. Haseman J. K., Elston R. C. The investigation of linkage between a quantitative trait and a marker locus. Behav Genet 1972; 2: 3±19. 41. Risch N., Zhang H. Extreme discordant sib pairs for mapping quantitative trait loci in humans. Science 1995; 268: 1584±1589. 42. Risch N. J., Zhang H. Mapping quantitative trait loci with extreme discordant sib pairs: sampling considerations. Am J Hum Genet 1996; 58: 836±843. 43. Grigorenko E. L., Wood F. B., Meyer M. S., Hart L. A., Speed W. C., Shuster A., Pauls D. L. Susceptibility loci for distinct components of developmental dyslexia on chromosomes 6 and 15. Am J Hum Genet 1997; 60: 27±39. 44. Morris D. W., Robinson L., Turic D., Duke M., Webb V., Milham C., Hopkin E., Pound K., Fernando S., Easton M., Hamshere M., Williams N., McGuffin P., Stevenson J., Krawczak M., Owen M. J., O'Donovan M. C., Williams J. Family-based association mapping

& 2000 Harcourt Publishers Ltd

45.

46. 47. 48. 49.

50. 51.

52.

provides evidence for a gene for reading disability on chromosome 15q. Hum Mol Genet 2000; 9: 843±848. Bishop D. V. M. Nonword repetition as a behavioural marker for inherited language impairment: evidence from a twin study. J Child Psychol Psychiat 1996; 37(4): 391±403. Kruglyak L. Genetic isolates: separate but equal? Proc Natl Acad Sci 1999a; 96: 1170±1172. Kruglyak L. Prospects for whole-genome linkage disequilibrium mapping of common disease genes. Nat Gen 1999b; 22: 139±144. Plante E. MRI findings in the parents and siblings of specifically language-impaired boys. Brain Lang 1991; 41: 67±80. Tzourio N., Heim A., Zilbovicius M., Gerard C., Mazoyer B. M. Abnormal regional CBF response in left hemisphere of dysphasic children during a language task. Pediatr Neurol 1994; 10(1): 20±26. Dronkers N. F. A new brain region for coordinating speech articulation. Nature 1996; 384: 159±161. Rae C., Lee M. A., Dixon R. M., Blamire A. M., Thompson C. H., Styles P., Talcott J., Richardson A. J., Stein J. F. Metabolic abnormalities in developmental dyslexia detected by 1H magnetic resonance spectroscopy. Lancet 1998; 351: 1849±1852. Wise R. J. S., Greene J., Buchel C., Scott S. K. Brain regions involved in articulation. Lancet 1999; 353: 1057±1061.

Prostaglandins, Leukotrienes and Essential FattyAcids (2000) 63(1/2), 101^107