مطالعه ژنتیکی لکنت زبان در یک جمعیت مؤسس
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|33501||2007||18 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Fluency Disorders, Volume 32, Issue 1, 2007, Pages 33–50
Genome-wide linkage and association analyses were conducted to identify genetic determinants of stuttering in a founder population in which 48 individuals affected with stuttering are connected in a single 232-person genealogy. A novel approach was devised to account for all necessary relationships to enable multipoint linkage analysis. Regions with nominal evidence for linkage were found on chromosomes 3 (P = 0.013, 208.8 centiMorgans (cM)), 13 (P = 0.012, 52.6 cM), and 15 (P = 0.02, 100 cM). Regions with nominal evidence for association with stuttering that overlapped with a linkage signal are located on chromosomes 3 (P = 0.0047, 195 cM), 9 (P = 0.0067, 46.5 cM), and 13 (P = 0.0055, 52.6 cM). We also conducted the first meta-analysis for stuttering using results from linkage studies in the Hutterites and The Illinois International Genetics of Stuttering Project and identified regions with nominal evidence for linkage on chromosomes 2 (P = 0.013, 180–195 cM) and 5 (P = 0.0051, 105–120 cM; P = 0.015, 120–135 cM). None of the linkage signals detected in the Hutterite sample alone, or in the meta-analysis, meet genome-wide criteria for significance, although some of the stronger signals overlap linkage mapping signals previously reported for other speech and language disorders.
Developmental stuttering is a common disorder of speech disfluency that affects 5% of children with an average population prevalence of 1% (Craig, Hancock, Tran, Craig, & Peters, 2002; Felsenfeld, 2002). The overt symptomatology of the disorder is characterized by excessive repetitions of sounds, syllables, and monosyllabic words, as well as sound prolongations and complete blockages of the vocal tract. Any of these characteristics may be accompanied by physical tension or movements, especially in the head and neck areas (Conture & Kelly, 1991; Wingate, 1964). Young children are often first diagnosed between ages 2 and 5, when they begin forming sentences and connecting thoughts verbally, with a higher occurrence in males than females at a ratio of 2:1. Nearly 80% of these affected children recover naturally from stuttering within one to four years of onset (Andrews & Harris, 1964; Mansson, 2000; Yairi & Ambrose, 1999). More females recover than males, resulting in a more skewed male-to-female ratio of 4:1 in older children and adults (Bloodstein, 1995; Buchel and Sommer, 2004; Felsenfeld, 2002; Yairi & Ambrose, 1999). Twin and family studies have indicated a strong genetic component to stuttering. Three twin studies showed considerably higher concordance levels of stuttering in monozygotic twins (20–90%) compared with dizygotic twins (3–19%) (Andrews, Morris-Yates, Howie, & Martin, 1991; Bloodstein, 1995, Felsenfeld et al., 2000 and Howie, 1981). When the monozygotic and dizygotic twin correlations are used to model the additive genetic and environmental components, both Andrews et al. (1991) and Felsenfeld et al. (2000) concluded that approximately 70% of the phenotypic variance is due to additive genetic effects and approximately 30% to non-shared effects. Several studies have shown a higher incidence of stuttering in first degree relatives (20–74%) than in the general population (1.3–42%) (Kidd, Heimbuch, & Records, 1981; Yairi, Ambrose, & Cox, 1996). Both the concordance of stuttering among monozygotic twin pairs and the familial aggregation of stuttering are consistent with a genetic component to stuttering. Several genetic models have been suggested for the inheritance of stuttering within families. Kidd, Kidd, and Records (1978) performed a segregation analysis (a statistical procedure that provides maximum likelihood values to allow testing of models of genetic transmission) in 511 families to identify the mode of inheritance that would account for the observed skewed sex ratio. They concluded that the model most consistent with the observed data was a sex-modified transmission model in which males and females have different genetic thresholds, with females requiring more susceptibility alleles than males to express a stuttering phenotype. This model was also proposed several years later in a study of a large Utah pedigree in which 38 individuals of a 269-member family stuttered (MacFarlane, Hanson, Walton, & Mellon, 1991). The sex-modified transmission of stuttering was consistent with both a multifactorial-polygenic model (many genes with small effect, as well as environmental components) and a single-locus genetic model (one gene with large effect, with numerous genes with small effects and environmental components) (Kidd, 1977). In 1993, Ambrose, Yairi, and Cox conducted a segregation analysis in 69 families in which at least one child stuttered and found that a single major genetic locus was the best explanation for the transmission of stuttering observed in these families, but that a polygenic-multifactorial model could not be rejected. Recently, Viswanath, Lee, and Chakraborty (2004) conducted a segregation analysis of 56 multigenerational pedigrees to assess if a major locus could account for the persistent stuttering observed in these families. They concluded that an autosomal dominant locus could explain the occurrence of stuttering in these pedigrees. Unfortunately, determining that the pattern of transmission of a trait is consistent with a major locus affecting susceptibility to a trait does not mean that there necessarily is a single major gene or that it will be simple to identify the relevant genetic variation. In fact, finding genes that influence stuttering, or any complex trait, has proven far more challenging than mapping genes for simple, Mendelian disorders (Botstein & Risch, 2003). Among the challenges that have made these studies difficult are etiologic and genetic heterogeneity, complex genetic models with many contributing loci of varying effects, gene by gene interaction, and gene by environment interaction (For a discussion of the challenges of mapping and identifying genetic loci for complex traits, see Risch, 2000). The accumulated findings of a genetic component to stuttering justified a move from statistical genetics into biological genetics. Typically, the first phase in such research is linkage analysis aimed at identifying the chromosomal location of genes underlying the disorder in question. To this end, DNA is extracted from blood or saliva. Then, genetic markers are identified on every chromosome, or only chromosomes of interest. When a genetic marker is co-inherited with stuttering (“linkage”), the indication is that the gene contributing to stuttering is on the same chromosome as the genetic marker; in fact, very close to it. Two measures are obtained: (a) the distance of the chromosomal location of the genetic marker in centiMorgans (cM) from the end of the upper arms of a chromosome, and (b) the LOD (log of the odds score), which is a measure indicating the likelihood that two genetic loci are physically near enough to each other to be linked, or inherited together. Tests of association compare allele frequency differences at genetic markers in affected and unaffected individuals. Association implies either the associated allele directly affects the phenotype or susceptibility to disease, or the associated allele is located sufficiently near, and is in linkage disequilibrium with, a susceptibility allele. That is, the susceptibility allele at the disease locus and an allele at the marker locus are found together on the same chromosome more often than expected based on their individual allele frequencies. Four genome-wide linkage investigations in families have identified several chromosomal regions that might harbor susceptibility genes for stuttering. A study in 68 Caucasian families identified a region on chromosome 18 centered at D18S976 that was suggestive of linkage to stuttering, but was not genome-wide significant (NPL = 1.51; non-parametric linkage (NPL)) (Shugart et al., 2004). Note that parametric linkage analyses require a genetic disease model to be specified to examine the evidence for linkage and a LOD greater than 3.0 for genome-wide significance. Non-parametric linkage analyses, on the other hand, do not require a genetic disease model to be determined to test for linkage of a phenotype to a genetic region. Evidence of linkage to chromosome 1 at the location of 1q21–1q22 (LOD = 2.27) was reported in a study of stuttering in a large Cameroon family (Levis, Ricci, Lukong, & Drayna, 2004). Genome-wide significance for linkage was found on chromosome 12q (NPLall = 4.61) using 44 inbred Pakistani families (Riaz et al., 2005). Most recently, the Illinois International Study of Stuttering Project identified possible regions increasing susceptibility for persistent stuttering on chromosomes 5 (LOD = 1.47), 13 (LOD = 1.72), and 15 (LOD = 1.98), and for ever stuttering on chromosomes 2 (LOD = 1.73), 7 (LOD = 1.69), and 9 (LOD = 2.28), in a genome-wide linkage study of 100 Caucasian families (Suresh et al., 2006). Unfortunately, there were no obvious overlap of chromosomal regions among the four studies in primary linkage analyses, a rather common outcome of linkage studies in many other complex disorders. One approach to reducing the challenges inherent in mapping a complex trait such as stuttering is to focus on isolated populations (Lander & Schork, 1994; Neel, 1970; Ober & Cox, 1998; Wright, Carothers, & Pirastu, 1999). Isolated populations include those that have expanded for a relatively small number of generations from a limited, and often well-defined, number of founders (Kruglyak, 1999). The small number of founders increases the likelihood that genetic heterogeneity will be reduced, which may also reduce the complexity of genetic models even for complex traits, because such populations may be segregating for only a portion of the genetic variation affecting a given trait. Our studies focus on the Hutterites, a religious isolate that left Europe in the late 1800s and settled in the northern United States. The Hutterites are ideal for genetic studies because of their large family sizes and communal lifestyle. The subjects included in this study are from 9 colonies in South Dakota. These individuals are related to each other within a 1,623-member, 13-generation pedigree traced back in time to 64 founders, some of whom may have been related (Abney, McPeek, & Ober, 2000). The small number of founders and the relatively rapid expansion of the population should enhance the power to map genes influencing susceptibility to a complex trait like stuttering. Although the Hutterites avoid close inbreeding, the relative isolation of the population has led to substantial relatedness among members of the group. The communal lifestyle, in which they are exposed to the same environmental conditions, eat the same food, work on the same farm, and attend the same schools and social events, etc., is expected to reduce the variability in the environmental factors that might affect risk of stuttering. This should, in turn, lead to additional improvement in the power to map susceptibility loci. On the other hand, a disadvantage of using large inbred pedigrees in genetic studies is the lack of computationally feasible approaches that allow such pedigrees to be analyzed as single, complete families (Ober & Cox, 1998). Consequently, we separated the 1,623-member Hutterite pedigree into four smaller sub-pedigrees, but preserved the relationships between family members and maintained a level of relatedness within each sub-pedigree consistent with that of the population. To increase our power to detect susceptibility loci regardless of allele frequency and risk, we carried out both genome-wide linkage and association studies in the Hutterite population. We also conducted a meta-analysis of stuttering, including the Hutterites and an outbred Caucasian population used in the Suresh et al. (2006) study.