Molecular, haematological and clinical studies of the -101 C T substitution of the beta-globin gene promoter in 25 beta-thalassaemia intermedia patients and 45 heterozygotes

Molecular, haematological and clinical studiesof a silent b-gene C → G mutation at 6 bp 30 to thetermination codon (þ1480 C → G) in twelve Greek families

E. MARAGOUDAKI, C. VRE TTOU, E. KANAVAKIS, J. TRAEGER-SYNODINOS, A. METAXOTOU-MAVROMMATI AND

C. KATTAMIS First Department of Paediatrics, Athens University, St Sophia’s Children’s Hospital, Athens, Greece

Received 18 March 1998; accepted for publication 10 July 1998

Summary. We report the clinical, haematological, biosyn-thetic and molecular data of 12 b-thalassaemia intermediapatients and their heterozygous parents, all of whom carrieda rare C → G mutation at nucleotide position 6 30 to thetermination codon (term. cd þ6 C → G) in the 30 untrans-lated region (30 UTR) of the b-globin gene (þ1480 C → G).This mutation has been reported previously in a singleb-thalassaemia intermedia patient of Greek origin. The 12patients of the present study had the clinical phenotype ofmild non-transfusion-dependent thalassaemia intermedia,preserving haemoglobin levels around 9 g/dl and haemo-globin F levels <25%. All were compound heterozygotesfor the þ1480 C → G mutation and common severeb-thalassaemia mutations. The haematological parameters

of heterozygotes with this mutation were within the normalrange with the exception of a slightly raised a/non-a-globinchain synthesis (1·2–1·9).

mRNA analysis demonstrated a 20–34% reduction inmRNA levels associated with the þ1480 C → G mutationcompared to normal b-globin alleles. These findings confirmthat the C → G mutation at position 6 30 to the terminationcodon is a mild b-thalassaemia mutation causing slightreduction in b-globin mRNA levels and b-globin chainsynthesis. It becomes clinically relevant when co-inheritedwith a severe b-thalassaemia mutation in trans.

Keywords: b-thalassaemia intermedia, silent b-thalassaemia,mRNA quantitation, 30 UTR, prenatal diagnosis.

The b-thalassaemias are a heterogenous group of autosomalrecessive disorders characterized by reduced (bþ) or absent(b8) production of the b-globin chain by the affected allele(Weatherall & Clegg, 1981). The majority are caused bypoint mutations and more than 180 have been characterizedto date (Huisman & Carver, 1998). Most b-thalassaemiamutations cause a substantial reduction of b-chain synthesisand heterozygotes for such mutations have a distinctivehaematological phenotype with hypochromic, microcyticred blood cells and characteristically raised levels of HbA2

(Weatherall & Clegg, 1981). Exceptions to this include thesilent b-thalassaemia carrier state in which haematologicalindices, HbA2 and HbF levels are within the normal range.Silent b-thalassaemia heterozygotes are distinguished onlyby slightly unbalanced globin chain synthesis (a/non-a1·3–1·7) and may be missed by procedures commonly used

for the diagnosis of b-thalassaemia trait (Schwartz, 1969;Kattamis et al, 1979). The interaction of silent with classicb-thalassaemia results in the clinical phenotype of thalas-saemia intermedia (Kattamis et al, 1982).

Several mild b-globin gene mutations have been describedwith haematological phenotype of silent b-thalassaemia inthe heterozygous state (Gonzales-Redondo et al, 1989; Ristaldiet al, 1990; Jankovic et al, 1991; Murru et al, 1991;Athanassiadou et al, 1994; Rosatelli et al, 1994, 1995; Hoet al, 1996; Traeger-Synodinos et al, 1998). The mostcommon is the C → T substitution at position ¹101 from theCap site within the distal CACCC box (Gonzales-Redondo et al,1989; Ristaldi et al, 1990), although most silent mutationshave been observed in sporadic cases only.

This study includes 12 Greek thalassaemia intermediapatients, all compound heterozygotes for classic and silentb-thalassaemia. These patients and their silent b-thalassaemiacarrier parents were characterized with the C → G mutationat position 6 30 to the termination codon (þ1480 C → G). Thismutation has so far been reported in a single b-thalassaemiaintermedia patient of Greek origin ( Jankovic et al, 1991).

British Journal of Haematology, 1998, 103, 45–51

45q 1998 Blackwell Science Ltd

Correspondence: Dr C. Kattamis, First Department of Paediatrics,Athens University, St Sophia’s Children’s Hospital, Athens 11527,Greece.

Here we report clinical, haematological, biosynthetic andmolecular data associated with the þ1480 C → G mutation.

MATERIALS AND METHODS

SubjectsAmongst 36 mild thalassaemia intermediate patients com-pound heterozygous for typical and silent b-thalassaemia(bthal/bsil), the þ1480 C → G mutation was identified in 12patients and in their silent b-thalassaemia carrier parents.

The severity of clinical phenotype in all probands wasassessed using criteria which included type and levels ofhaemoglobin, degree of growth and bone impairment,hepatosplenomegaly and transfusion requirements (Kattamiset al, 1982).

Haematological analysis and globin biosynthesisBlood samples (10–20 ml) were collected in EDTA andhaematological parameters were measured using an auto-mated cell counter (Cell-Dyn 1700, Abbott Diagnostics). HbFwas quantitated by alkali denaturation (normal HbF<2%)and HbA2 by elution from cellulose acetate (normalHbA2<3·5%). Globin chain synthesis was measured by themethod of Clegg (1983) and ferritin levels using the ElisaFerritin Kit-IMX System (Abbott Diagnostics).

DNA analysisDNA isolation. Genomic DNA was isolated from white

blood cells by the salt extraction method (Miller et al, 1988).a and b globin gene analysis. The a-globin genes were

mapped using standard protocols (Kanavakis et al, 1986).Mutations in the b-globin gene were localized using Denatur-ing Gradient Gel Electrophoresis analysis (DGGE) (Losekoot etal, 1990) (Fig 1). Common Mediterranean b-thalassaemia

mutations were characterized by Amplification RefractoryMutation System-PCR (ARMS-PCR) (Old et al, 1990). Theþ1480 C → G mutation was initially identified using directsequencing (Thein & Hinton, 1991) (Fig 2) and subsequentlyby allele-specific oligonucleotide (ASO) hybridization (Tzetiset al, 1994) or a simple PCR-mediated primer directedmutagenesis assay (Friedman et al, 1991). For the assay, theforward mutagenesis primer (50 CACAAGTATCACTAAGCCC30) (mismatch base underlined) was paired with the reverseprimer (50 CAGTTTAGTAGTTGGACTTA 30) and the 86 bp PCRproduct was digested with the restriction enzyme HpaII whichcuts in the presence of the þ1480 C → G mutation. Therestriction products were analysed on a 3:1 Nuseive:Seakemagarose gel (FMC BioProducts).

b-globin gene cluster haplotypes. To define b-globin genecluster haplotypes, restriction enzyme digestion analysis ofseven b-gene cluster diallelic polymorphic sites (Orkin et al,1982) was based on previously described PCR protocols(Sutton et al, 1989; Old et al, 1990). The presence of the XmnIpolymorphism ¹158 bp to the Gg-globin gene was assessedby restriction endonuclease digestion of a PCR-generatedfragment of the Gg-gene promoter region (Sampietro et al,1992).

RNA analysisRNA isolation. Total RNA was isolated from a reticulocyte-

enriched fraction (0·5 ml) of freshly collected blood (5 ml),using RNAsolTM B solution (Wak-Chemie Medical GmbH)followed by extraction with chloroform and isopropanolprecipitation.

q 1998 Blackwell Science Ltd, British Journal of Haematology 103: 45–51

46 E. Maragoudaki et al

Fig 1. Denaturing gradient gel electrophoresis (DGGE) analysis.Following DGGE analysis, the gel was stained with ethidium bromideand viewed with UV. The analysis showed the following genotypes:lane 1: heterozygote bþ AATAAA → AATGAA; lanes 2 and 3: hetero-zygote bþ þ1480 C → G; lanes 4, 5 and 6: no mutation.

Fig 2. Direct sequencing of amplified genomic DNA of the b-globingene. The sequence from 1470 to 1487 bp from the CAP site whichincludes the termination codon is shown. The C → G mutation atposition þ1480 (nt 6 30 to the term. codon) of the coding strand isshown.

47Silent þ1480 C → G b-thalassaemia Mutation

q 1998 Blackwell Science Ltd, British Journal of Haematology 103: 45–51

Reverse-transcription PCR (RT-PCR). 1–2 mg total RNAwere reverse-transcribed into cDNA followed by specificPCR amplification of bcDNA using the Gene AMP RNA/PCR kit (Perkin Elmer Cetus, Norwalk, Conn., U.S.A.). Theb-gene-specific primers were: forward: 50-CACAACTGTGTT-CACTAGCAACC-30 (þ15 to þ38 in the 50 UTR) and reverse:50-AGCACACAGACCAGCACGTT-30 (þ1355 to þ1373), pro-ducing an amplified product of 380 bp.

Restriction enzyme digestion for mutant and normal b cDNAdifferentiation. The normal b allele carries an ApaLI restric-tion site in codon 2. In all samples of the study the codon 2C → T polymorphism was observed to be in linkagedisequilibrium with the þ1480 C → G mutation. Thispolymorphism abolishes the ApaLI restriction site(GTGCAC) and thus digestion with ApaLI (New EnglandBiolabs) of amplified b cDNA facilitated the differentiationbetween b cDNA derived from mutant alleles (no digestion:380 bp) and normal alleles (digestion: 335 bp plus 45 bp).

Quantitation of bthal cDNA. A modification of the RT-PCRprotocol was used to quantitate the mutant/normal (bthal/bN) mRNA levels by one-step extension of fluorescent primer.A 20 ml aliquot of amplified b cDNA was used as a templatefor a single extension of 20 pmol fluorescently labelledreverse primer (bromoacetyl Texas Red dye, Genset) bydenaturation at 958C for 5 min, annealing at 588C for2 min and extension at 728C for 12 min. The product wassubsequently digested with ApaLI and 2 ml loaded onto anon-denaturing 5% acrylamide gel in a Vistra automaticsequencer (Molecular Dynamics/Amersham Life Sciences).Relative amounts of the bN and bthal transcripts were sizedand quantitated using the Fragmentor software (Molecular

Dynamics/Amersham Life Sciences). Each sample wasanalysed at least twice. One-step primer extension avoidsformation of bN:bthal heteroduplexes that remain uncleavedby the restriction enzyme, leading to false ratios (Dimovskyet al, 1994). Validity of quantitation by ApaLI digestion wasmonitored by including normal (non-thalassaemic) controlseither homozygote, heterozygote or without the codon 2C → T polymorphism (Fig 3).

RESULTS

Compound heterozygotes for the b þ1480 C → Gand severe b-thalassaemia mutationsOf the 12 patients, nine were diagnosed during childhood(aged 2–8 years) and three in adolescence (aged 14–18years). Present ages range from 5 to 33 years and only fourare less than 15 years of age. Clinical phenotype has beenshown to be mild with anaemia, slight splenomegaly andmild jaundice but no transfusion dependence. In one patient(A-12, diagnosed at 4 years with Hb level 8·5 g/dl), regulartransfusions were initiated following repeated reduction ofHb levels during infectious episodes. Growth and develop-ment of all patients is satisfactory with mild thalassaemicappearance, minimal bone changes and moderate spleno-megaly. Two of the adult patients (A-2 and M-9) and twochildren (M-5 and K-11) have been splenectomized.

Relevant haematological parameters of the 12 patients atdiagnosis are summarized in Table I. All had moderateanaemia with mean Hb 9 6 0·9 g/dl (7·8–10·5 g/dl) andHbF <25% with a mean of 11·2 6 6·9% (2·6–25%). Red cellindices and morphology as well as the presence of inclusion

Fig 3. bthal c-DNA quantitation. (A) A 380 bp of b cDNA amplified. The Cd 2 C → T polymorphism, observed to be in linkage disequilibrium withthe þ1480 C → G mutation, abolishes the ApaLI restriction site and facilitated the differentiation between the mutant (no digestion; 380 bp) andnormal b cDNA (digestion: 335 bp). (B) The relative amounts of the bN and bthal transcripts were sized and quantitated using the Fragmentorsoftware (Molecular Dynamics/Amersham Life Sciences). Lane 1: normal individual; lane 2: normal (non-thalassaemic) homozygote for thecodon 2 polymorphism; lanes 3, 4 and 5: heterozygotes for the b þ1480 C → G mutation; lanes 6 and 7: normal heterozygotes for the codon 2C → T polymorphism.

bodies and rare normoblasts (1/100 to 10/100) wereconsistent with the haematological findings of b-thalassaemiaintermedia. Ferritin levels were generally <200 mg/l. Mildjaundice was present in all patients with bilirubin levels inthe range of 2·5–5·5 mg/dl.

DGGE and sequence analysis of the entire b-globin geneidentified a C → G mutation at position 6 30 to the terminationcodon, linked to the C → T polymorphism at codon 2 and thethree IVSII polymorphisms at nt 16 C → G, nt 74 C → T andnt 81 C → T, in all patients. In addition all carried a severeb-thalassaemia mutation in trans: five bþ IVSI nt 110 G → A,three b8 codon 39 C → T and four b8 IVSI nt 1 G → A (Table I).

Gene mapping showed normal a-globin genotype in allcases.

Eleven patients were homozygous for the absence of theGg XmnI polymorphism (¹/¹) and one was heterozygote(þ/¹). The b-gene cluster haplotype associated with theþ1480 C → G mutation was Mediterranean VI in 9/12 casesand Mediterranean VII in 2/12 cases. In one case familymembers were unavailable for study.

Heterozygotes for the þ1480 C → G mutationRelevant haematological and molecular data of the 12parents, heterozygotes for the þ1480 C → G mutation, are

q 1998 Blackwell Science Ltd, British Journal of Haematology 103: 45–51

48 E. Maragoudaki et alTable I. Clinical, haematological and molecular findings in the b þ1480 C → G/b thal compound heterozygotes.

Age (yr)Hb MCV MCH HbA2 HbF Retics b-Globin Ferritin

Pt At diagnosis Present (g/dl) (fl) (pg) (%) (%) (%) XmnI genotype (mg/l)

N-1 7 33 10·5 72·0 21·8 5·2 10·3 6·0 þ/¹ þ1480/110 533A-2* 9 29 9·0 80·0 20·5 6·1 8·2 10·0 ¹/¹ þ1480/cd39 150P-3 2 4 8·0 66·0 19·7 5·4 7·6 2·0 ¹/¹ þ1480/110 91K-4 3 7 8·2 60·0 19·0 5·9 25·0 ¹/¹ þ1480/cd39 48M-5* 3 20 8·8 5·0 15·0 6·6 ¹/¹ þ1480/1-1 312T-6 14 16 10·0 63·0 18·1 6·9 3·0 2·5 ¹/¹ þ1480/110 103B-7 8 10 9·4 63·0 20·4 4·0 21·6 7·0 ¹/¹ þ1480/1-1 40K-8 14 16 10·0 60·0 5·2 7·0 ¹/¹ þ1480/110 65M-9* 18 29 8·4 65·0 20·0 6·2 7·0 10·0 ¹/¹ þ1480/cd39 ¹

K-10 5 5 9·2 65·0 18·7 5·2 2·6 4·0 ¹/¹ þ1480/110 42K-11* 2 7 7·8 60·0 19·0 6·5 15·0 6·0 ¹/¹ þ1480/1-1 164A-12 4 18 8·5 63·0 19·0 6·0 12·5 5·0 ¹/¹ þ1480/1-1 ¹†

Mean 7·4 9·0 65·2 19·6 5·6 11·2 5·9SD 5·4 0·9 6·0 1·1 0·8 6·9 2·7

b-Thalassaemia mutations: 110 ¼ IVSI nt 110 G → A; cd39 ¼ codon 39 C → T; I-1 ¼ IVSI nt 1 G → A. Ferritin levels at present age.

* Splenectomized patients.† Transfused patient.

Table II. Haematological and biosynthetic findings in the b þ480 C → G heterozygotes.

Hb MCV MCH HbA2 HbF Osmotic b cDNAPt (g/dl) (fl) (pg) (%) (%) fragility Morphology a/non-a (%)

N-a 13·1 87·0 27·0 2·6 0·70 Normal Normal 1·60 80·2A-b 13·5 84·0 27·0 3·0 0·35 Normal Normal 1·50 66·6P-c 13·7 93·0 29·7 2·7 2·00 Normal Normal 1·97 72·5K-d 13·6 83·2 27·9 2·7 3·00 Normal Normal 1·20 –M-e 12·1 78·7 25·7 2·9 0·44 Normal Normal 1·38 –T-f 14·4 88·0 28·0 2·6 0·66 Normal Normal – –B-g 12·5 88·0 20·0 3·0 0·77 Normal Normal – –K-h 14·8 91·0 29·3 1·9 0·50 Normal Normal 1·70 –M-i 13·3 90·0 28·6 2·3 0·50 Normal Normal – –K-k 16·6 92·0 30·7 2·9 0·30 Normal Normal – –K-l 14·8 95·0 32·0 2·8 0·44 Normal Normal 1·50 –A-m 14·8 90·0 29·0 3·4 0·60 Normal Normal 1·62 –

Mean 13·9 88·3 27·9 2·7 0·9 1·6SD 1·2 4·6 3·0 0·4 0·8 0·2

b cDNA ¼bþ 1480/bN ratio.

49Silent þ1480 C → G b-thalassaemia Mutation

q 1998 Blackwell Science Ltd, British Journal of Haematology 103: 45–51

summarized in Table II. They had normal red cell indices,morphology and osmotic fragility. HbA2 and F values werewithin the normal range as were plasma ferritin levels.Globin-chain synthesis was slightly unbalanced (meana/non-a 1·6).

With one exception (case K-d, aa/-a3·7) gene mappingshowed normal a-globin genotypes in all heterozygotes(Table II).

Semiquantitation of bthal cDNA indicated that the þ1480C → G mutation was associated with reduced mRNA levelscompared to normal. The relative quantities of b cDNA inthree heterozygotes with this mutation, as quantitated afterone-step extension of the Texas Red forward primer (Fig 3B),showed that b mRNA derived from the mutant allele wasreduced to about 70–80% of the normal b-globin mRNAlevels (Table II).

DISCUSSION

The b þ1480 C → G mutation has as yet been reported inonly one patient. The patient, diagnosed with severeanaemia at the age of 6 months, had haematological indicesand clinical manifestations consistent with thalassaemiaintermedia and after childhood started regular transfusionsbecause of thalassaemic facies ( Jankovic et al, 1991). Inthe present study, only 6/12 patients were diagnosed before5 years of age (Table I); two were diagnosed incidentally ona routine laboratory check-up. Moderate anaemia, mildsplenomegaly and mild jaundice were the presentingsymptoms.

Haematological data at diagnosis was comparable amongstall patients and remained almost stable during follow-upwith Hb levels around 9 g/dl. HbF was <10% in 7/12 cases.HbA2 levels were increased (mean value 5·6%), consistentwith previous observations in mild b-thalassaemia inter-media patients with high HbA levels (unpublished obser-vation). Splenectomy ameliorated the haematologicalindices in only one of the four splenectomized patients; theremaining three preserved the same Hb levels pre- and post-splenectomy.

Most of the patients were followed-up for several yearsprior to definitive characterization of their b genotypes. Allpatients had normal growth and development and minimalbone changes. Furthermore, haemosiderosis was mild; ferritinlevels were generally <200 mg/l, although measurements insome adult patients indicated a tendency for these levels torise over time to as high as 500 mg/l (Table I). According tothe criteria used in our unit (Kattamis et al, 1982) thesepatients have mild non-transfusion-dependent b-thalassaemiaintermedia. The only regularly transfused patient (A-12),discontinued transfusions after characterization of thisb-globin genotype and since his last transfusion, 9 monthsago, has preserved Hb levels about 8 g/dl and HbF <20%.The clinical phenotype in all patients of our study are incontrast to the case of Jancovic et al (1991), despite com-parable b-thalassaemia genotypes (compound heterozygosityfor the þ1480 C → G and severe b-thalassaemia mutations).

Haematological parameters of the 12 heterozygotes for theþ1480 C → G mutation were within the normal range, apart

from a slightly imbalanced a/non-a-globin chain synthesis.The balanced a/non-a ratio of heterozygote K-d is consistentwith the interaction of the ¹a3·7 deletion. The osmoticfragility and red cell morphology were normal in all cases,thus confirming that this mutation is truly silent (Kattamis etal, 1979).

The families with the þ1480 C → G mutation originatedfrom various regions of the country. In nine families theþ1480 C → G mutation was associated with Mediterraneanhaplotype VI, the same as that found in the first reportedpatient, suggesting that the mutation is likely to be of thesame origin ( Jancovic et al, 1991). In two families themutation was associated with Mediterranean haplotype VIIwhich differs from haplotype VI 50 to the wb gene. These twofamilies, although apparently unrelated, originated from thesame village and the alternative haplotype could have arisenin a common ancestor through a recombination event onthe þ1480 C → G chromosome.

Two other mutations have been described in the 30 UTR ofthe b-globin gene; a 13 bp deletion from þ1565 to þ1577(Basak et al, 1993) associated with b thalassaemia and aT → C substitution at þ1570 (Cai et al, 1992) that appears tobe a polymorphism (Divoky et al, 1994). The function of theregion between the termination codon and the polyA signalof the b globin gene (30 UTR) is not well understood. Thisregion probably plays a role in stability, translation andtransport of the mRNA (Kronenberg et al, 1989; Jackson &Standart, 1990; Proudfoot, 1996). Although the mech-anism is not apparent, the þ1480 C → G mutation leadsto a reduction in b-globin mRNA levels, resulting in aslight reduction of b-globin chain synthesis, causing mildthalassaemia intermedia in compound heterozygotes for themutation and severe b-thalassaemia determinants, and atruly silent haematological phenotype in carriers of themutation.

All b-gene mutations underlying silent b-thalassaemiabecome phenotypically apparent when co-inherited with asevere b-thalassaemia mutation, and the majority of suchpatients have mild non-transfusion-dependent thalassaemiaintermedia (Gonzales-Redondo et al, 1989; Ristaldi et al,1990; Murru et al, 1991; Athanassiadou et al, 1994; Rosatelliet al, 1994, 1995; Ho et al, 1996; Traeger-Synodinos et al,1998). However, rare exceptions to this have been reported(Gonzales-Redondo et al, 1989; Jankovic et al, 1991).Amongst the silent b-thalassaemia mutations, the C → Tsubstitution at position ¹101 from the Cap site within thedistal CACCC box is considered the most common (Gonzales-Redondo et al, 1989; Ristaldi et al, 1990), although whenstrict criteria were used to assess haematology in carriers forthis mutation, only a third of them were completely silent(unpublished observation). In contrast to the ¹101 C → Tmutation, which is commonly detected in the generalpopulation in Greece, the þ1480 C → G mutation hasnever been detected in any of the suspected b-thalassaemiaheterozygotes with borderline/normal haematologicalvalues who are often referred to our laboratory for moleculardiagnosis, indicating that it is not even suspected by routinemass screening. In the b-silent parents of the 36 bthal/bsil

patients followed in our unit, molecular characterization was

necessary for definitive diagnosis of a b-thalassaemia carrierstate. It is interesting to note that, within this group, the¹101 C → T and þ1480 C → G mutations have a similarfrequency, indicating that at least in Greece the þ1480C → G mutation is a relatively common silent b-thalassaemiamutation.

Progress in molecular diagnostics has enabled the identi-fication of mutations underlying milder clinical conditions,creating at the same time the necessity to re-evaluate theethical relevance of using this information for reproductivegenetic services. The þ1480 C → G mutation is such anexample. This comprehensive study provides importantinformation concerning phenotype and genotype correla-tions in bthal/bsil patients, which is invaluable when offeringgenetic counselling in such couples. In our experience theclinical phenotype of double heterozygotes for severe b-genemutations with þ1480 C → G is mild. Therefore, whencounselling couples at risk, who as a rule already have anaffected child, we inform them of this fact. However, to solvethe dilemma as to when it is appropriate to carry out DNAanalysis in partners of b-thalassaemia carriers and prenataldiagnosis of mild/silent mutations, more patients need to beassessed.

ACKNOWLEDGMENTS

The authors thank S. Papadopoulou for technical assistance.Five of the 12 families of the study were under the care ofDrs M. Athanasiou, I. Georgiou, M. Karageorga and V. Ladis.

This study was partly supported by an EU FrameworkSupport Programme from the Greek Ministry of Industry,Research and Technology (Operational Programme forResearch and Technology no. 236).

REFERENCES

Athanassiadou, A., Papachatzopoulou, A., Zoumbos, N., Maniatis,G. & Gibbs, R. (1994) A novel b thalassaemia mutation in the 50

untranslated region of the b globin gene. British Journal ofHaematology, 88, 307–310.

Basak, A.N., Ozer, A., Kindar, B. & Akar, N. (1993) A novel 13 bpdeletion in the 30 UTR of the b globin gene causes b thalassemia ina Turkish patient. Hemoglobin, 17, 551–555.

Cai, S.P., Eng, B., Francombe, W.H., Kendall, A.G., Waye, J.S. &Chui, D.H.K. (1992) Two novel b thalassemia mutations in the50 and 30 noncoding regions of the b globin gene. Blood, 79,1342–1346.

Clegg, J.B. (1983) Haemoglobin synthesis. Methods in Haematology:the Thalassaemias (ed. by D. J. Weatherall), pp. 54–73. ChurchillLivingstone, London.

Dimovsky, A.J., Efremov, D.G., Gu, L.H. & Huisman, T.H.J. (1994) Therelative levels of bA and bS mRNAs in HbS heterozygotes and inpatients with HbS-bþ thalassaemia or HbS-bþ HPFH combina-tions. British Journal of Haematology, 87, 353–356.

Divoky, V., Baysal, E., Oner, R., Akif Curuk, M., Walker, E.L.D., III,Indrak, K. & Huisman, T.H.J. (1994) The T → C mutation at þ96of the untranslated region 30 to the termination codon of the b

globin gene is a rare polymorphism that does not cause a b

thalassemia as previously ascribed. Human Genetics, 93, 77–78.Friedman, K., Highsmith, W.E. & Silverman, L. (1991) Detecting

multiple cystic fibrosis mutations by polymerase chain reaction-

mediated site-directed mutagenesis. Clinical Chemistry, 37, 753–755.

Gonzales-Redondo, J.M., Stoming, T.A., Kutlar, A., Kutlar, F., Lanclos,K.D., Howard, E.F., Fei, Y.J., Aksoy, M., Altay, C., Gurgey, A., Basak,A.N., Efremov, G.D., Petkov, G. & Huisman, T.H.J. (1989) A C → Tsubstitution at nt ¹101 in a conserved DNA sequence of thepromoter region of the b globin gene is associated with ‘silent’ b

thalassemia. Blood, 73, 1705–1711.Ho, P.J., Rochette, J., Fisher, C.A., Wonke, B., Jarvis, M.K.,

Yardumian, A. & Thein, S.L. (1996) Moderate reduction of b

globin gene transcript by a novel mutation in the 50 untranslatedregion: a study of its interaction with other genotypes in twofamilies. Blood, 87, 1170–1178.

Huisman, T.H.J. & Carver, M. (1998) The b and d ThalassemiaRepository. Hemoglobin, 22, 169–195.

Jackson, R.J. & Standart, S. (1990) Do the polyA tail and 30

untranslated region control mRNA translation? Cell, 62, 15–24.Jankovic, L., Dimovski, A.J., Kollia, P., Karageorga, M., Loukopoulos,

D. & Huisman, T.H.J. (1991) A C → G mutation at nt position 6 30

to the terminating codon may be the cause of a silent b

thalassemia. International Journal of Hematology, 54, 289–293.Kanavakis, E., Tzotzos, S., Liapaki, A., Metoxotou-Mavrommati, A. &

Kattamis, C. (1986) Frequency of a thalassemia in Greece.American Journal of Hematology, 22, 225–232.

Kattamis, C., Metoxotou-Mavrommati, A., Ladis, V., Tsiarta, H.,Laskari, S. & Kanavakis, E. (1982) The clinical phenotype of b anddb thalassaemia in Greece. European Journal of Paediatrics, 139,135–138.

Kattamis, C., Metaxotou-Mavrommati, A., Wood, W.G., Nash, J.R. &Weatherall, D.J. (1979) The heterogeneity of normal HbA2-bthalassaemia in Greece. British Journal of Haematology, 42, 109–123.

Kronenberg, H.M., Roberts, B.E. & Efstratiadis, A. (1989) The 30

noncoding region of b globin mRNA is not essential for in vitrotranslation. Nucleic Acids Research, 6, 153–166.

Losekoot, M., Fodde, R., Harteveld, C.L., van Heeren, H., Giordani,P.C. & Bernini, L.F. (1990) Denaturing gradient gel electrophoresisand direct sequencing of PCR amplified genomic DNA: a rapid andreliable diagnostic approach to beta thalassaemia. British Journalof Haematology, 76, 269–274.

Miller, S.A., Dykes, D.D. & Polesky, H.F. (1988) A simple salting outprocedure for extracting DNA from human nucleated cells. NucleicAcids Research, 16, 1215.

Murru, S., Loudianos, G., Deiana, M., Camaschella, C., Sciarratta,G.V., Agosti, S., Parodi, M.I., Cerruti, P., Cao, A. & Pirastu, M.(1991) Molecular characterization of b thalassemia intermedia inpatients of Italian descent and identification of three novel b

thalassemia mutations. Blood, 77, 1342–1347.Old, J.M., Varawalla, N.Y. & Weatherall, D.J. (1990) Rapid detection

and prenatal diagnosis of b thalassaemia: studies in Indian andCypriot populations in UK. Lancet, ii, 834–837.

Orkin, S.H., Kazazian, H.H., Antonarakis, S.E., Goff, S.C., Bochm,C.D., Sexton, J.P., Waber, P.G. & Giardina, P.J.V. (1982) Linkage ofb thalassaemia mutations and b globin gene polymorphisms withDNA polymorphisms in human b globin gene cluster. Nature, 296,627–631.

Proudfoot, N. (1996) Ending the message is not so simple. Cell, 87,779–781.

Ristaldi, M.S., Murru, S., Loudianos, G., Casula, L., Porcu, S.,Pigheddu, D., Fanni, B., Sciarratta, G.V., Agosti, S., Parodi, M.I.,Leone, D., Camascella, C., Serra, A., Pirastu, M. & Cao, A. (1990)The C → T substitution in the distal CACCC box of the b globingene promoter is a common cause of silent b thalassaemia in theItalian population. British Journal of Haematology, 74, 480–486.

q 1998 Blackwell Science Ltd, British Journal of Haematology 103: 45–51

50 E. Maragoudaki et al

51Silent þ1480 C → G b-thalassaemia Mutation

q 1998 Blackwell Science Ltd, British Journal of Haematology 103: 45–51

Rosatelli, M.C., Faa, V., Meloni, A., Fiorenza, F., Galanello, R.,Gasperini, D., Amendola, G. & Cao, A. (1995) A promotermutation, C → T at position ¹92, leading to silent b thalassaemia.British Journal of Haematology, 90, 483–485.

Rosatelli, M.C., Pischedda, A., Meloni, A., Saba, L., Pomo, A., Travi,M., Fattore, S. & Cao, A. (1994) Homozygous b thalassaemiaresulting in the b thalassaemia carrier state phenotype. BritishJournal of Haematology, 88, 562–565.

Sampietro, M., Thein, S.L., Contreras, M. & Pazmany, L. (1992)Variation of HbF and F cell number with the Gg XmnI (C → T)polymorphism in normal individuals. Blood, 79, 832–833.

Schwartz, E. (1969) The silent carrier of beta thalassemia. NewEngland Journal of Medicine, 281, 1327–1333.

Sutton, M., Bouhassira, E.E. & Nagel, R.L. (1989) Polymerase chain

reaction applied to the determination of the b-like globin genecluster haplotypes. American Journal of Hematology, 32, 66–69.

Thein, S.L. & Hinton, J. (1991) A simple and rapid method of directsequencing using Dynabeads. British Journal of Haematology, 79,113–115.

Traeger-Synodinos, J., Maragoudaki, E., Vrettou, C., Kanavakis, E. &Kattamis, C. (1998) Rare b thalassemia alleles in the Greek andGreek Cypriot populations. Hemoglobin, 22, 89–94.

Tzetis, M., Traeger-Synodinos, J., Kanavakis, E., Metaxotou-Mavrommati, A. & Kattamis, C. (1994) The molecular basis ofnormal HbA2 (type 2) b thalassaemia in Greece. HematologicPathology, 8, 25–34.

Weatherall, D.J. & Clegg, J.B. (1981) The Thalassaemia Syndromes.Blackwell Scientific Publications, Oxford.