Dr Reyhan Öner, Hacettepe University, Science Faculty, Department of Molecular Biology, 06532, Beytepe, Ankara, Turkey. E-mail: email@example.com
We provide the first description of a homozygote patient for the G→A substitution in the 5′ UTR of the β-globin gene. The proband was a 17-year-old girl with β-thalassaemia intermedia who had never received a blood transfusion. The physical examination revealed a well-developed women with no facial or bony abnormalities. There was mild paleness and mild splenomegaly which was 2 cm below the costal margin. The haemoglobin (Hb) was 7·6 g/dl, Hb A2 5·4% and Hb F 14·6% of the total Hb. The Hb A2 of both parents was 3·5%. The Hb F level in the mother and father were 0·9, 1·2% and the mean cell volume (MCV) value was 70 and 72 fl respectively. DNA analysis of the β-gene region of the propositus revealed homozygosity for a G→A substitution at nucleotide +22 relative to the β-gene cap site, within a functional downstream region that was referred to as the DCE (downstream core element). In addition to the data obtained previously from in vitro transcription assays, clinical findings and in vivo expression studies gave some valuable clues about the effect of +22 G→A mutation on the expression of β-gene. Phenotypic expression of this homozygous patient is highly suggestive that G→A substitution at nt +22 confers a relatively mild (silent) β+-thalassaemia phenotype.
The mature mRNA of the β-globin gene possesses the coding regions and additional sequences at both ends known as the 5′ and 3′ untranslated regions (UTRs). The 5′ UTR from the cap site to the initiation codon has 51 nucleotides in the β-gene and contains rather unique base sequences (Rosatelli et al, 1989). The cap site appears to be crucial for the initiation of the globin chain synthesis. An Asian Indian homozygous individual for the cap site +1 (A→C) point mutation was found to have a β-thal trait phenotype. A total of five mutations have been described in the initiation (ATG) Cd, and all were associated with a β0-thal phenotype (Wong et al, 1989). To date, five mutations in the segment of DNA between the cap site and initiation Cd have been reported. A C→G mutation in position +33 3′ to the cap site is the fifth mutation found in this region so far. The other four are well distributed along this segment. Two of them are 4 bp deletion (– AAAC) at position +40 to +43 and a –T at position +10 (Ho et al, 1999). A C→G substitution at position +20 relative to the cap site is polymorphic and is observed exclusively on a chromosome associated with IVS-II-745 mutation (Öner et al, 1991a). The other mutation in this region is a G→A mutation at position +22. This mutation was originally observed in two Turkish and one Bulgarian families (Öner et al, 1991a) and, later, one Italian family in heterozygous or in compound heterozygous states (Cai et al, 1992). The compound heterozygotes exhibited variable clinical pictures, changing from β+-thalassaemia major to β+-thalassaemia intermedia.
Several hypothesis have been put forward to explain the possible effect of this 5′ UTR mutation on the β-gene expression (Öner et al, 1991a; Cai et al, 1992). More recently, Lewis et al (2000) described the identification and characterization of a functional downstream element in the human adult β-globin promoter. Scanning mutagenesis from +10 to +45 indicated that this region contains a functional cis element(s) in vitro, and they designated this element the DCE (downstream core element). The DCE functions in concert with the β-globin CATA box and initiator element, as well as in a heterologous, TATA-less context. It was concluded that TFIID makes sequence-specific contacts to the DCE, and that transcription factor IID (TFIID), at least in part, is necessary for promoter function via the DCE. Incorporation of the β-thalassaemia mutations into promoter constructs led to decreases in transcription in vitro (Lewis et al, 2000). Because no data was available for the homozygosity status of the +22 mutation, we took the opportunity to demonstrate the first description of a homozygous patient for this alteration.
Patient and methods
The patient A 17-year-old female patient was referred to our hospital for evaluation of chronic mild anaemia. Her past history revealed that she had been pale since infancy and was prescribed iron treatment on several occasions without any benefit. She had never had a blood transfusion. The family history indicated that she was the first child of a first-cousin marriage. Her parents and the two siblings were apparently healthy. Physical examination of the propositus revealed a well developed (height, 1·80 cm; weight, 60 kg) pale girl without any facial stigmata or bony changes suggestive of β-thalassaemia. The liver and spleen were both palpable 3 and 6 cm below their respective costal margins.
The whole β-globin gene was amplified encompassing the region from the position −456 relative to the cap site to the position 80 nucleotides after the poly A signal in 3′ UTR. The amplified product was directly sequenced (Öner et al, 1991b). The average α:β mRNA ratio was characterized by reverse transcriptase–polymerase chain reaction (RT–PCR). Reticulocyte mRNA was extracted from packed red blood cells (RBC) using the method of Chomczynki & Sacchi (1987). Ten healthy individuals and 10 patients with β-thalassaemia with specific genotypes, namely three who were homozygous for IVS-I-110 (G→A), two who were homozygous for IVS-II-1 (G→A), three homozygosus for −30 (T→A), two homozygous for CD 39 (C→T), two homozygous for IVS-I-6 (T→C), six individuals who were compound heterozygotes for three of these five mutations and one patient who was compound heterozygote for CD 37/IVS-I-6 mutations were included in the mRNA analysis as controls.
Total RNA (1·5µg) was reverse transcribed into cDNA using an oligo dT 15 primer and avian myeloblastosis viral (AMV) reverse transcriptase according to the manufacturer' instructions (Promega, Southhampton, UK). The RT–PCR amplification primers and conditions were the same as those described by Smetanina et al (1997). The RT–PCR products were separated on a non-denaturing 6% polyacrylamide gel, and quantification of either the radiolabelled or silver-stained specific globin cDNA bands was performed by densitometric analysis as described by Lin et al (1994).
Northern-blot analysis was carried out according to the method of Maniatis et al (1982). In vitro chain synthesis analysis in reticulocytes was determined by the method of Huisman & Jonxis (1977). Apart from the homozygous patient and her immediate family members, two previously studied brothers aged 12 and 14 years old with compound heterozygosity for the +22 and Fsc CD 8 mutation were studied. α-Gene number and haplotypes of the β-globin gene cluster were determined by digestion with appropriate restriction endonucleases followed by Southern blot analysis and hybridization (Southern, 1975; Öner et al, 1995). Nomenclature used was the system introduced by Orkin et al (1982).
The result of the routine haematological studies of the patient, her parents and previously reported compound and simple heterozygotes are given in Table I and Table II for comparison. Two previously reported families and the family presented in this communication were not related. Furthermore, they originated from different geographical areas of Turkey.
Table I. Haematological data of the parents with +22 (G→A) heterozygosity.
Table II. Some of the haematological data of the presented patient and compound heterozygous patients.
Sequence analysis of the patient indicated homozygosity for a G→A mutation at position +22 relative to the cap site (Fig 1A). Although both parents were heterozygous, the two siblings did not carry this alteration. Haplotyping of the DNA from the patient and her parents identified the +22 β-thal chromosome as Mediterranean type II.
Sequence analysis of the Aγ-gene region revealed that the chromosome with G→A at +22 was associated with a T→C mutation at codon 75 of the Aγ-globin gene. The Xmn I polymorphic site 5′ to the Gγ-region was +/– for the mother and two siblings, and –/– for the patient and the father. All of the subjects had four α-genes.
The RT–PCR study indicated that the mean α:β mRNA ratios in normal adults was 3·1 ± 0·6. The reduction in mRNA transcript produced by the β-thalassaemia alleles was estimated and found to be in agreement with the spectrum of phenotypes as assessed by clinical status and Hb red blood cell indices. The average value of 9·55 was found in three patients with the IVS-I-II0 homozygosity. The α:β-mRNA ratios were dramatically increased in homozygotes for Cd 39 and IVS-II-1 defects. Trace amounts of β-cDNA were observed for the patients with these two mutations, as expected. A relatively low ratio of 4·10 was obtained for the IVS-I-110 (G→A) heterozygotes. The ratio was 5·55 for the IVS-I-6 homozygotes. Individuals bearing the IVS-I-6 and −30 mutations in the heterozygote state had average values of 4·10 and 4·85 respectively; these values were similar to those published by Lin et al (1994) and Smetanina et al (1997). Significant increases were observed in the α:β mRNA ratio in samples from the patients with β0-thal/β+-thal (Cd 37/IVS-I-6) and β+-thal (−30/−30) (17·2 and 7·7 respectively). However, a minor increase in α:β-mRNA ratio was noted in the patient with +22 G→A mutation compared with normal adults and β-thal patients (Fig 1B). The ratio was 3·30 for the patient and 3·05 and 3·1 for the mother and the father respectively.
Northern-blot analysis showed that there was no dramatic reduction of mRNA levels for the patient and her parents compared with the normal control group (data not shown). In vitro globin chain synthesis analysis indicated an α:β ratio of 1·10, 1·00, and 0·98 for the propositus, the mother and the father respectively. The ratios were 0·60 and 0. 62 for two brothers who were compound heterozygote for the +22 and the Fsc CD 8 mutations (Table I).
The compound heterozygosity for +22 G→A and one of the common β-thalassaemia mutations results in a wide spectrum of the disease from thalassaemia intermedia to thalassaemia major (Gürgey & Altay, 1993). Presence of a mild clinical condition together with mild elevation in Hb F and Hb A2 value in our patient shows close resemblance to the homozygous condition associated with IVS-I-6 mutation. Haematological findings of simple heterozygotes reported in this study and five previously reported subjects showed that mild microcytosis (MCV: 70–72) was present in five subjects; Hb A2 level was either normal or at the upper level of normal in four, and mildly elevated in three subjects (Table II). This observation indicates that the phenotypical expression of +22 mutation may vary from a common β-thal trait to a silent one.
The mutation at position +10 in compound heterozygosity with other β-thalassaemia mutations produces a phenotype of thalassaemia intermedia, whereas individuals heterozygous for this 5′ UTR mutation alone have normal Hb levels with minimal changes in red blood cell indices and Hb A2 levels. No clinical or haematological details were provided for the 4 bp deletion at position +40 to +43. Another mutation at +20 (C→G) in 5′ UTR is polymorphic and links to IVS-II-745 mutation (Öner et al, 1991a). A C→G mutation at position +33 3′ to the cap site is the fifth mutation found in this region so far. Compound heterozygotes for the β+33 C→G and a β+-thalassaemia allele were completely asymptomatic. Individuals heterozygous for the β+33 C→G mutation alone are clinically and haematologically silent, with normal red blood cell indices and normal levels of Hb A2 (Ho et al, 1999).
Several in vitro studies have been performed to explain the possible effect(s) of the 5′ UTR mutations on the β-gene expression. Ho et al (1999) showed moderate reduction of β-globin gene transcript by a +33 C→G mutation in the 5′ UTR region in a stably transfected MEL cell system. It is interesting to note that the –T deletion at position +10 resulted in a translational defect. Functional analysis of the 4 bp deletion at positions +40 to +43 in the 5′ UTR of the β-globin gene using reporter gene in transient expression assays failed to show any significant effect of this mutation (Ho et al, 1996). Recently, Lewis et al (2000) presented the detailed characterization of a functional downstream region within a TATA and Inr-dependent promoter, which they refer to as the DCE (downstream core element). The existence of the DCE was suggested by two mutations at +22 and +33 found in β-thalassaemia patients, and a third mutation at +13. Incorporation of the β-thalassaemia mutations into promoter constructs led to decreases in transcription in vitro. Expression levels of the +13, +22 and +33 mutants are at 0·42 ± 0·07, 0·58 ± 0·22 and 0·46 ± 0·12 of wild-type promoter activity respectively.
Phenotypical expression of the disease in the presented homozygous patient with the +22 mutation was more severe than those of β-thalassaemia trait. In support of the β+22 G→A alteration causing the above mentioned clinical picture, we performed globin chain synthesis, RT–PCR and Northern blot analysis to measure the deficit β-cDNA level in the homozygous patient and her heterozygous parents. However, our in vivo expression studies failed to show sufficiently reduced levels of β-cDNA. Because the RT–PCR and Northern blot techniques are known to give wide ranges of values and globin chain synthesis can also be temperamental, our results may not reflect the actual in vivo situation.
In vitro expression studies performed by Lewis et al (2000) indicated that the relative transcriptional level of the +22 mutant was at 58 ± 22% of wild-type promoter activity and was mildest when compared with the +33 and +13 defects. Hence, the possibility should be considered that the modest deficit in β-globin mRNA production could be insufficient for the detection of the actual amount of β-gene product at least in part in our measurements.
In conclusion, our clinical findings and phenotypical expression of the disease are in agreement with the results of transient heterologous expression assay which showed that the β-globin gene with the +22 5′ UTR mutation generates a modest deficit in β-globin mRNA production.
This study indicated that in patients with slightly elevated Hb A2 and Hb F associated with mild microcytic anaemia, homozygosity for the +22 G→A mutation should be kept in mind. Similarly, heterozygosity for the +22 5′ UTR mutation may mimic both the Hb A2 normal and silent β-thalassaemias.
This study was partly supported by the Turkish Administration of State Planning: DPT (to C.Ö) and TUBA (to C.Ö and Ç.A).