Dominant influence of gamma-globin promoter polymorphisms on fetal haemoglobin expression in sickle cell disease.

Polymorphisms of multiple cis-acting elements in the beta-globin locus are associated with variable fetal haemoglobin (HbF) level in sickle cell disease. We developed a multiplex assay permitting simultaneous analysis of three polymorphic cis elements spanning 53 kb of the beta-globin locus. We identified concordance between polymorphic alleles in gamma- and beta-globin promoters however a significant number of betaS-chromosomes were identified with polymorphisms in hypersensitive site 2 (HS2) of the beta-globin locus control region juxtaposed to atypical cis alleles in the gamma-promoter. Analysis of an unusually large number of such hybrid haplotype chromosomes provided unique insight into HbF level associated with specific cis alleles. Associations between cis alleles and HbF level in patients were verified by in vitro functional analysis. Our findings indicate that compared to HS2, polymorphism in the gamma-promoter exerts a dominant influence on HbF level in sickle cell disease.

Ancestral origin of variation in the triosephosphate isomerase gene promoter

A high frequency of nucleotide substitutions -5A/G, -8G/A, -24T/G in the triosephosphate isomerase (TPI) gene promoter has been demonstrated in African-Americans. The biological significance of these promoter variants, two of which, -8G/A and -24T/G, occur within regulatory elements essential for transcription, is controversial. The geographical distribution and frequency of allelic variation in the TPI promoter was determined in 378 unrelated normal subjects from Sub-Saharan African (n = 103), Caribbean (n = 26), Northern European (n = 57), Mediterranean (n = 55), Middle Eastern (n = 42), Asian Indian (n = 48) and Oriental (n = 47) populations. Five haplotypes were identified: the common haplotype, -5A-8G-24T, -5G, -8A, -5G-8A, and -5G-8A-24G. All, with the exception of the -8A haplotype, were present in geographically dispersed populations. The -5G allele, which was found at varying frequency in the African, Caribbean and Oriental populations. Phylogenetic comparison suggests this may represent the ancestral promoter haplotype. Homozygosity for the -5G-8A haplotype identified in four subjects confirms that these variants are not responsible for a null allele as formerly postulated. Linkage disequilibrium between related TPI promoter haplotypes, -5G, -5G-8A and -5G-8A-24G, and a single nucleotide polymorphism at nt2262 of the TPI gene supports a single ancestral origin for these mutations which preceeds the separation of African populations.(Au)

Interethnic variation in genetic risk factors for vascular disease

Genetic predisposition to vascular disease has important implications for population screening and prevention. The most common hereditary cause of venous thrombosis is resistance to activated protein C caused by the G1691A point mutation in exon 10 of the factor V gene (1q21-25) which leads to the substitution of glutamine for arginine (factor V Leiden). A thermolabile variant of 5, 10-methylenetetrahydrofolate reductase (MTHFR) caused by the C677T mutation of the MTHFR gene (1p36.3) which substitutes valine for alanine is associated with the vascular disease risk factor hyperhomocysteinaemia. The possibility that these mutations may predispose individuals of African-American origin to thrombosis was investigated in 9 patients (6 male, 3 female) with sickle cell anaemia who had experienced a thrombotic episode. The frequency of the MTHFR C677T mutation was also determined in unrelated subjects from six different populations: African-Caribbean (50), Oriental (47), Asian Indian (21), Middle Eastern (24), Meditteranean (50) and Northen European (61). The MTHFR and factor gene regions of interest were amplified by the polymerase chain reaction method. Factor V Leiden was screened for by single strand conformation polymorphism analysis and the MTHFR C 677T mutation by Hinf 1 restriction. All patients were homozygous normal (G/G) for the factor V allele. This is consistent with population studies which failed to identify factor V Leiden in normal subjects of Sub-Saharan African populations and found a low frequency (0.65 percent in Black Americans. By contrasts, factor V Leiden was found to be most prevalent in European populations (from 1.4 percent in Finland to 7 percent in Greece). One patient was heterozygous (C/T) and 8 homozygous normal (C/C) for the MTHFR mutation. Population studies revealed the observed frequency of the mutant allele (T) to be lowest in African-Caribbean subjects (9 percent) of whom none were homozygous and only 18 percent heterozygous. The frequency was highest in the Meditteranean population (42 percent), followed by Middle Eastern (38 percent), Northern European (30 percent), Asian Indian (21 percent) and Oriental (19 percent). No deviation from Hardy-Weinberg equilibrium was detected. The proportion of subjects homozygous for the mutation (T/T) was 18 percent Meditteranean, 17 percent Middle East, 10 percent Northern European and Asian Indian and 2 percent Oriental. (AU)

The molecular basis for genetic variation in fetal haemoglobin level in sickle cell anaemia

Raised fetal haemoglobin levels ameliorate the clinical severity of sickle cell anaemia. This provides a rationale for therapy and signals the need to elucidate the molecular basis for the variability of HbF level in sickle cell anaemia. Polymorphism within regulatory sites of the globin locus alter the affinity with which transcription factors bind their cogante recognition sites thereby modulating gene expression. A novel chromosomal haplotype utilising polymorphic variation within two enhancers hypersensitive site 2 (HS2) of the locus control region and the pre g y framework and the silencer protein BP1 binding site that spans a 53 kb interval of the globin locus was determined in 205 patients with sickle cell anaemia from the UK and Jamaica. Multiplexed polymerase chain reactions developed to facilitate rapid analysis of polymorphisms within each site allowed individual haplotype construction in a single lane of a single strand conformation polymorphism (SSCP) electrophoresis gel. SSCP banding patterns for the combined polymorphic sites were confirmed as unique chromosomal haplotypes by DNA sequence anaylsis. Three hundred and ten chromosomes with sequence TA7 N 12 TA8, GA and AT(AT)8T4 were designated class 1. Twenty-five class II with sequence TA8 N10 TA11, GG, AC(AT)6T9; 17 class III with TA9 N10 TA10, AG, AC(AT)8T4; 7 class IV with TA10 N10 TA12, AG, AC(AT)9T5 and 13 class Ia haplotype with sequence TA9 N10 TA10, GA, AT(AT)8T4 were identified. The proportion of class I chromosomes in both groups (159/210 UK; 151/200 Jamaican) is identical, however, significantly more chromosomes with sub-classes I and II are present among the Jamaican sample. There is incomplete association between the functional haplotype classes defined and conventional haplotypes based on restriction fragment length polymorphisms. The level of Hbf is significantly higher in patients with functional haplotype classes III and IV compared to those with classes I and II. Both high HbF haplotypes share a high affinity binding motif for the transcription factor GATA-1. This novel approach allows the combined effets of genetic variation in regulatory sequences within the globin locus on HbF level to be defined. (AU)