Molecular analysis of b-thalassaemia patientsin a high incidence area of southern Italy

L. RIGOLI*,A. MEO ,

M.R. MICELI ,

K. ALESSIO ,R.A. CARUSOà,

M.A. LA ROSA ,D.C. SALPIETRO*,

M. RICCA ,I. BARBERI*

*Department of Paediatrics, University of Messina, School of Medicine, Italy,

 Paediatric Department of Thalassaemia Ward, University of Messina,

School of Medicine, Italy and àDepartment of Human Pathology,

University of Messina, School of Medicine, Italy

Summary The prevalence of eight mutations in 84 patients with b-thalassaemia major and in

16 subjects with thalassaemia intermedia was investigated. All of the patients were

Italian, originating from Eastern Sicily (Messina area) and some Calabrian regions.

Genomic DNA was ampli®ed by polymerase chain reaction (PCR). DNA molecular

investigations were performed by allele-speci®c oligonucleotide (ASO) hybridization, to

identify the following b-thalassaemia mutations: CD39 (C-T), IVS1-110 (G-A), IVS1-6

(T-C), IVS1-1 (G-A), IVS2-745 (C-G), IVS2-1 (G-A), )87 (C-G), CD6 A ()A).

Our data underline that in thalassemia intermedia two mutations were statistically

prevalent: IVS1-6 T®C (P < 0.001) and CD 6-A (P < 0.05). CD 39 was statistically

prevalent in b-thalassaemia major patients (P < 0.01). The difference between the two

groups was not statistically signi®cant for all the other mutations. Five different gen-

otypes were recorded among thalassaemia intermedia and 15 among b-thalassaemia

major patients. Twenty-®ve percent of the intermedia patients and 4.5% of the major

patients had homozygosity for mild mutations (group I); 62.5% of the intermedia

patients and 26.2% of the major patients had combinations of mild/severe mutations

(group II). In addition, homozygosity or double heterozygosity for severe mutations

(group III) was found in 12.5% of the intermedia patients and 69% of the major

patients. Some genotypes were restricted to thalassaemia intermedia, including

heterozygosity )87/IVS1-6 and IVS1-6/CD 6-A. It is essential to understand the

distribution and frequency of the relevant mutations in each population where

b-thalassaemias exist. This is of particular importance for genotype±phenotype cor-

relation and for carrier detection, genetic counselling and prenatal diagnosis.

Keywords b-thalassaemia, mutations, PCR, ASO

Introduction

The b-thalassaemias are a heterogeneous group of inher-

ited disorders caused by mutations in and around the

structural gene of the adult haemoglobin (HbA) b-chain

(Weatherall & Clegg, 1981; Orkin & Kazazian, 1984).

Studies at the gene level have identi®ed a large number of

b-thalassaemia gene variations in different populations

(El-Hazmi et al., 1983; Wong et al., 1992). The clinical

severity of b-thalassaemia can be in¯uenced by genetic

determinants that are capable of reducing globin chain

imbalance. Clinically, b-thalassaemias may be mild or

silent conditions, or they may cause a severe disease,

leading to early transfusion dependence (Gelehrter &

Collins, 1990). In the latter case, affected individuals

Accepted for publication 15 August 2001

Correspondence: Dr Luciana Rigoli, Dipartimento di Scienze Pedia-

triche Mediche e Chirurgiche, Policlinico Universitario, 98100

Messina, Italy. Tel: 090 2213133; Fax: 090 696037; E-mail: luciana.

rigoli@unime.it

Clin. Lab. Haem.

2001, 23, 373±378

Ó 2001 Blackwell Science Limited

373

present with severe anaemia in early infancy, requiring

lifelong transfusion and iron chelation therapy (b-thalas-

saemia major) (Orkin & Kazazian, 1984). In a small

minority of cases, the inheritance of two b-thalassaemia

alleles results in a disease of moderate severity known as

b-thalassaemia intermedia. b-thalassaemia intermedia

may result from the inheritance of two b+-thalassaemia

alleles or the combination of a severe b°-thalassaemia

allele and a particularly mild or silent b+-thalassaemia

allele (Camaschella & Cappellini, 1995; Camaschella et al.,

1959). Sophisticated molecular techniques are required

for genotypic identi®cation. In this study, DNA molecular

investigations were performed along with conventional

laboratory investigations in a group of major and

intermedia b-thalassaemia patients.

Patients and methods

The study was approved by the Committee for Ethics in

Medical Research and was performed in accordance with

the ethical standards laid down in the 1964 Declaration of

Helsinki. All subjects gave their informed consent prior to

their inclusion in the study.

The study population included 84 patients with

b-thalassaemia major and 16 subjects with thalassaemia

intermedia. All of the patients were Italian, originating

from eastern Sicily (Messina area) and some Calabrian

regions, and attended our Paediatric Department Thalas-

saemia Ward. There were no interrelationships among the

patients. b-thalassaemia characterization studies included

family history, clinical and haematological investigations

and biosynthesis studies. All of the thalassaemia major

patients were on a regular transfusion regimen. Four of

the 16 thalassaemia intermedia patients (25%) had never

received blood transfusions. Ten patients (62.5%) had

received occasional transfusions, two patients (12.5%)

had received transfusions as adults.

The mean age of the b-thalassaemia major patients was

13 years (range 5±18 years), while the mean age of

the thalassaemia intermedia group was 27 years (range

11±53 years).

Blood samples for routine haematological analysis were

drawn in EDTA tubes. For haemoglobin phenotyping, a

sample of whole blood was analysed by electrophoresis at

alkaline pH. HbA2 (Marengo Rowe, 1965) and HbF (Betke

et al., 1959) were estimated and the a/b chain ratio was

determined (El).

Genomic DNA was extracted from blood samples (Sam-

brook, 1989) and ampli®ed by the polymerase chain

reaction (PCR) (Saiki et al., 1988; Cai et al., 1989). Allele-

speci®c oligonucleotide (ASO) hybridization (Old et al.,

1990) was used to determine the following b-thalassaemia

mutations: CD39 (C-T), IVSI-110 (G-A). IVSI-6 (T-C), IVS1-

1 (G-A), IVS2-745 (C-G), IVS2-1 (G-A), -87 (C-G), CD6-A

(Table 1). The b-thalassaemia mutations screened were

those commonly encountered in the Mediterranean popu-

lation (Di Marzo et al., 1990; Kattamis et al., 1990; Murru

et al., 1991). Oligonucleotide primers used for ASO hy-

bridization were purchased commercially. PCR was carried

out in a 2400 Perkin±Elmer Thermocycler. A denaturation

step at 94 °C for 1 min was followed by 34 cycles of

denaturation (94 °C, 45 s), annealing (62 °C, 15 s), and

primer extension (72 °C, 30 s). The ®nal extension was

performed at 72 °C for 7 min. The PCR products were

separated by electrophoresis on a 2.0% agarose gel and

visualized by ethidium bromide staining and ultraviolet

light illumination prior to ASO hybridization.

Results

Different gene mutations were detected in the two groups

of patients (b-thalassaemia major and intermedia), and at

different frequencies.

In the b-thalassaemia major patients the following

mutations were detected: IVS1-110 (34.5%), CD 39

(34.5%), IVS1-1 (10.7%), IVS1-6 (9.5%), IVS1-745

(6.0%), IVS1-1 (2.4%), -87 (2.4%) (Table 2). Several

patients (16/84) were homozygous for the IVS1-110

Location Mutation Effect b-thalassaemia Type

IVS1-110 G®A Internal IVS change a b+

CD 39 C®T Nonsense mutant b b°IVS1-1 G®A Splice junction change a b°CD 6 A )A Frameshift mutant b b°IVS1-1 G®A Splice junction change a b°IVS1-6 T®C Consensus change a b+

IVS1-745 C®G Internal IVS changes b+

±87 C®G Transcriptional mutant b+

a, RNA processing mutant; b, nonfunctional RNA.

Table 1. b-thalassaemia mutations

tested in Eastern Sicily and Calabria

regions

Molecular analysis of b-thalassaemia patients374

Ó 2001 Blackwell Science Ltd., Clin. Lab. Haem., 23, 373±378

mutation. In this group, the mean ferritin level was 1500

ng/ml, the mean age of ®rst transfusion was 14 months,

and the mean transfusion requirement was 17 per year.

CD 39 homozygosity was seen in 16 patients. In this

group, the mean ferritin level was 2750 ng/ml, the mean

age of ®rst transfusion was 13 months and the mean

transfusion requirement was 18 per year.

IVS1-6 homozygosity was seen in four patients. In this

group, the mean ferritin level was 1339 ng/ml, the mean

age of ®rst transfusion was four years and the mean

transfusion requirement was 14 per year.

IVS1-1 homozygosity was found in four patients. In this

group, the mean ferritin level was 800 ng/mg, the mean

age of ®rst transfusion was 15 months and the mean

transfusion requirement was 16 per year.

Twenty-four patients (28.5%) were IVS1-110 com-

pound heterozygosites for other mutations. Heterozygosity

for other mutations was found as follows: CD 39

heterozygosity in 20 patients, IVS1-1 heterozygosity in

seven patients, IVS1-745 heterozygosity in ®ve patients,

IVS1-6 heterozygosity in four patients, IVS1-1 heterozy-

gosity in three patients and -87 heterozygosity in one

patient.

In the thalassaemia intermedia patients, the most

frequent mutations were IVS1-6 (50%), IVS1-110

(31.3%), CD 6-A (12.5%) and -87 (6.2%) (Table 3).

Three of the patients (18.8%) were homozygous for the

IVS1-110 mutation. In this group, the mean ferritin level

was 1018 ng/ml, the mean age of ®rst transfusion was six

years and the mean transfusion requirement was 22 per

year.

In four patients with IVS1-6 homozygosity the mean

ferritin level was 354 ng/mg, the mean age of ®rst

transfusion was 10 years, and the mean transfusion

requirement was 11 per year.

In the remaining patients, heterozygosity for other

mutations was found as follows. Two patients with -87/

IVS1-6 had never received any blood transfusions. In four

patients with an IVS1-6/CD 6 genotype, the mean ferritin

level was 1100 ng/mg, the mean age of ®rst transfusion

was four years and the mean transfusion requirement was

19 per year. In three patients with an IVS1-6/IVS1-110

genotype the mean ferritin level was 1768 ng/ml, the

mean age of ®rst transfusion was eight years and the

mean transfusion requirement was 10 per year.

The differences in mutation distribution between the

major and intermedia groups were assessed statistically. In

thalassaemia intermedia, prevalance data for two muta-

tions were statistically signi®cant. These were IVS1-6

T®C (P < 0.001) and CD 6-A (P < 0.05). In the

b-thalassaemia major patients, the prevalence of CD 39

was statistically signi®cant (P < 0.01). The differences

between the two groups were not statistically signi®cant

for any other mutations.

According to these results, the IVS1-6 T®C and -87

C®G mutations may be considered as `mild'. Severe

mutations included b° CD 39 C®T, which was statistically

prevalent in the thalassaemia major group, IVS1-110

G®A and IVS2-745 C®G, which are severe in vitro (Bunn

and Forget, 1986) and all the mutations (IVS1-1 G®A,

IVS1-1 G®A) which cause b°-thalassaemia (Orkin &

Kazazian, 1984).

Genotypes

Five different genotypes were recorded among b-thalas-

saemia intermedia patients and 15 among b-thalassaemia

major patients. To simplify genotype±phenotype analysis,

the genotypes were classi®ed in three groups, according to

mutation severity (see above). Group I included patients

with two mild mutations, group II included patients with

combinations of mild/severe mutations and group III

included patients with two severe mutations (Table 4).

Twenty-®ve percent of the intermedia and 4.8% of the major

patients were included in group I; 62.5% of the intermedia

and 26.2% of the major patients were included in group II,

while group III included 12.5% of the intermedia and 69%

Table 2. Frequency of mutations in b-thalassaemia major

patients

Mutation location

Number of chromosomes

(n � 168)

Frequency

(%)

IVS1-110 58 34.5%

CD 39 58 34.5%

IVS1-6 16 9.5%

IVS1-1 18 10.7%

IVS1-745 10 6.0%

IVS1-1 4 2.4%

)87 4 2.4%

CD 6-A 0 0

Number of b-thalassaemia major patients � 84.

Table 3. Frequency of mutations in b-thalassaemia intermedia

patients

Mutation location

Number of chromosomes

(n � 32)

Frequency

(%)

IVS1-6 16 50%

IVS1-110 10 31.3%

CD-A 4 12.5%

)87 2 6.2%

Number of b-thalassaemia intermedia patients � 16.

L. Rigoli et al. 375

Ó 2001 Blackwell Science Ltd., Clin. Lab. Haem., 23, 373±378

of the major patients (Table 4). Some genotypes were

seen only in thalassaemia intermedia patients, including

heterozygosity ±87/IVS1-6 (two patients in group I) and

IVS1-6/CD 6-A (four patients in group II).

Discussion

The molecular defects in b-thalassaemia are heterogene-

ous, with over 150 b-thalassaemia mutations reported in

the literature (Weatherall & Clegg, 1981; Orkin et al.,

1984). Although population studies have revealed a wide

spectrum of molecular pathology in b-thalassaemia,

limited ethnic group-speci®c alleles account for over 90%

of the defective genes (Tamagnini et al., 1983; Galanello

et al., 1986; Wong et al., 1986; Di Marzo et al., 1988;

Galanello et al., 1989; Kattamis et al., 1990; Old et al.,

1990; Murru et al., 1991; Rosatelli et al., 1992; El-Hazmi

et al., 1995). It is therefore important to understand the

distribution and frequency of the relevant mutations in

each population in which b-thalassaemia occurs.

In the Mediterrean area, b-thalassemias are encoun-

tered at a relatively high frequency (Tamagnini et al.,

1983; Wong et al., 1986; Bunn et al., 1986; Di Marzo

et al., 1988; Galanello et al., 1989; Kattamis et al., 1990;

Old et al., 1990; Murru et al., 1991). Genetic counselling

of couples at risk for b-thalassaemia is complicated by the

variable relationship between genotype and phenotype

(Cao et al., 1994). This is particularly true of genotypes

involving mild b+-thalassaemia mutations such as IVS1-6

(T-C). For example, the IVS1-6 (T-C)/ codon 39 (C-T)

genotype has been reported in both b-thalassaemia major

and b-thalassaemia intermedia patients (Waye et al.,

1995).

In this report, we have described the range of pheno-

types associated with genotypes involving the most

common b-thalassaemia mutations in Eastern Sicily

(Messina area) and some Calabrian regions. Two muta-

tions were statistically prevalent in thalassaemia interme-

dia: IVS1:6 T®C (P<0.01) and CD 6:A (P<0.05). CD 39

was statistically prevalent in thalassaemia major

(P<0.01).

Classi®cation of our patients into three groups by b-

genotype identi®ed four homozygotes for mild mutations

(IVS1:6/IVS1:6) in group I who were major patients,

receiving their ®rst transfusions at a mean age of 8

months. Two other patients with the same genotype

suffered only with thalassaemia intermedia, receiving

their ®rst blood transfusions at a mean age of 10 years.

According to recent studies (Ragusa et al., 1992; Murru

et al., 1992; Efremov et al., 1994; Camaschella et al.,

1995), this phenotypic variation cannot be attributed to

sequences within the b-globin gene cluster. Moreover,

combinations of mild and severe mutations were present

both in the intermedia and in the major groups. In our

patients, IVS1 : 6 T®C in combination with severe

mutations such as CD 39 and IVS1 : 110 was associated

with a milder phenotype. It is well known that IVS1 : 6

C®T mutation in vitro creates an alternative splice site

which is rarely used, so that normal b chains are produced

in signi®cant amounts (Waye et al., 1995). There is

general agreement that this mutation is also mild in vivo.

Heterozygotes for IVS1 : 6 C®T have less severe red cell

changes (Efremov et al., 1994); in the homozygous state,

the phenotype is usually that of thalassaemia intermedia,

although some patients with thalassaemia major have

been reported (Chehab et al., 1989; Meloni et al., 1983).

This was also seen in our study.

Our study underlines the relative roles of b+ and b°genes in thalassaemia major homozygotes. It is known

that about half the b-thalassaemia alleles lead to complete

inactivation of the b-globin gene, causing b°-thalas-

saemia. The b+-thalassaemia mutations allow some

production of b-globin, but the output is reduced. The

reduction in b-globin output ranges from minimal to

almost complete absence. The severity of these b+

thalassaemia alleles can be correlated to the degree of

reduction in MCV in heterozygotes. Heterozygotes for such

mutations have normal Hb A2 levels and normal red cell

indices and are often referred to as `silent' carriers.

In our study, double heterozygotes for CD39/IVS1-6 or

IVS1-110/IVS1-6 could have either intermedia or major

disease. Differentiation of Cooley's disease from thalas-

saemia intermedia is imprecise. Thalassaemia intermedia

is a term applied to a disease characterized by a

transfusion-independent clinical course of intermediate

severity between thalassaemia major and the asympto-

matic carrier state. As the de®nition of thalassaemia

intermedia is relative, it is clinically and genetically

Group

I

Mild/Mild

II

Mild/Severe

III

Severe/Severe

Thalassaemia major 4/84 (4.8%) 22/84 (26.2%) 58/84 (69%)

Thalassaemia intermedia 4/16 (25%) 10/16 (62.5% 2/16 (12.5%)

Table 4. b-genotypes in the major and

intermedia groups b-genotypes

Molecular analysis of b-thalassaemia patients376

Ó 2001 Blackwell Science Ltd., Clin. Lab. Haem., 23, 373±378

hetereogeneous. The spectrum includes, at one end,

patients with haemoglobin levels of 6 g/dl, skeletal

abnormalities and signi®cant transfusion requirements

and, at the other end, asymptomatic subjects with mild

anaemia and splenomegaly diagnosed by chance or

during family studies.

Transfusion independence is an unsatisfactory criterion

for de®ning thalassaemia intermedia, as it depends to a

large extent on patient perceptions and medical trends. The

introduction of a policy requiring high-level transfusion

regimens in infancy for homozygous thalassaemia patients

may have led to the enrolment of some patients who would

otherwise have been classi®ed as intermedia. On the other

hand, as the prognosis of thalassaemia intermedia was

considered better than that of patients regularly trans-

fused, some major patients maintaining adequate Hb levels

by massive haemopoietic expansion, with associated bone

abnormalities, were not enrolled in such programs. New

clinical and molecular criteria are clearly required.

Our observations indicate that the clinical phenotype

associated with individual b-thalassaemia cases cannot be

predicted accurately based only on the b- and a-globin

genotypes. Nevertheless, it is possible that concomitant

a-thalassaemia may ameliorate the clinical severity of

b-thalassaemia when other genetic determinants are also

present. Such determinants may ameliorate the disease

severity by increasing b-globin gene expression or the level

of Hb F. It is not easy to predict whether an a-thalassaemia

gene would invariably ameliorate disease severity in

homozygous b°-thalassaemia.

Our data underline the importance of molecular

investigation to clarify the phenotypic variability of

thalassaemias in countries with a high prevalence of

haemoglobinopathies. Genotype±phenotype correlation is

useful for clari®cation of the clinical course and will

contribute to the eradication of thalassaemias in future

generations.

References

Betke K., Marti H.R. & Schlicht L. (1959) Estimation of small

percentage of foetal haemoglobin. Nature 184, 1877±1878.

Bunn H.F. & Forget B.G. (1986) The thalassaemias: molecular

pathogenesis. In: Hemoglobin: Molecular, Genetic and Clinical

Aspects. pp. 225±321.W.B. Saunders Co, Philadelphia.

Cai S.P., Chang C.A. & Zhang J.Z. (1989) Rapid prenatal diag-

nosis of b-thalassaemia using DNA ampli®cation and non

radioactive probe. Blood 73, 372±374.

Camaschella C. & Cappellini M.D. (1995) Thalassaemia inter-

media. Haematologica 80, 58±68.

Camaschella C., Mazza U. & Roetto A. (1995) Genetic interac-

tions in thalassaemia intermedia. analysis of b-mutations,

a-genotype, c-promoters and b-LCR hypersensitive site 2 and

4 in Italian patients. American Journal of Hematology 48,

82±87.

Cao A., Galanell R. & Rosatelli M.C. (1994) Genotype±phenotype

in b-thalassaemias. Blood Review 8 (1±6), 1±6.

Chehab F.F., der Kaloustian V. & Khouri F.P. (1987) The

molecular basis of b-thalassaemia in Lebanon: application to

prenatal diagnosis. Blood 69, 1141±1145.

Di Marzo R., Dowling C.E. & Wong C. (1988) The spectrum of

b-thalassaemia mutations in Sicily. British Journal of Haematology

69, 393±396.

Efremov D.G., Dimovski A.J. & Baysal E. (1994) Possible factors

in¯uencing the haemoglobin and fetal haemoglobin levels

in patients with b-thalassaemia due to a homozygosity for the

IVS-I-6 (T®C) mutation. British Journal of Haematology 86,

824±828.

El-Hazmi M.A.F., Al-Swailem A.R. & Warsy A.S. (1995)

Molecular defects in b-thalassaemias in the population of Saudi

Arabia. Human Hered 45, 278±285.

El-Hazmi M.A.F. & Lehmann H. (1976) Hemoglobin Riyadh

a2b2 (120 [GH3]Lys®Asn) -A new variant found in associ-

ation with a-thalassaemia and iron de®ciency. Hemoglobin 1,

59±74.

Galanello R., Dessi E. & Melis M.A. (1989) Molecular analy-

sis of b°-thalassaemia intermedia in Sardinia. Blood 74,

823±827.

Gelehrter T.D. & Collins F.S. (1990) Molecular genetics of human

disease: hemoglobinopathies. In: Principles of Medical Genetics.

pp. 97±121.Williams & Wilkins, Baltimore.

Kattamis C., Cheng H. & Hg Reese A.L. (1990) Molecular char-

acterization of b-thalassaemia in 174 Greek patients with

b-thalassaemia major. British Journal of Haematology 74,

342±345.

Marengo Rowe A.J. (1965) Rapid electrophoresis and quantita-

tion of haemoglobin on cellulose acetate. Journal of Clinical

Pathology 18, 790±792.

Meloni A., Rosatelli C. & FaaÁ V. (1992) Promoter mutations

producing mild b-thalassaemia in the Italian population.

British Journal of Haematology 80, 222±226.

Murru S., Loudianos G. & Deiana M. (1991) Molecular charac-

terization of b-thalassaemia intermedia in patients of Italian

decent and identi®cation of three novel b-thalassaemia muta-

tions. Blood 77, 1342±1347.

Murru S., Loudianos G. & Porcu S. (1992) A b-thalassaemia

genotype not linked to the b-globin cluster in. An Italian Family

81, 283±287.

Old J.M., Varawalla N.Y. & Weatherall D.J. (1990) Rapid detection

and prenatal diagnosis of b-thalassaemia. Studies in Indian and

Cypriot populations in the UK. Lancet ii, 834±837.

Orkin S.H. & Kazazian H.R. Jr (1984) The mutation and poly-

morphism of the human b-globin gene and its surrounding

DNA. Annual Review of Genetics 18, 131±171.

Ragusa A., Lombardo M. & Beljord C. (1992) Genetic epidemiol-

ogy of b-thalassaemia in Sicily: do sequences 5¢ to the

Gc gene and 5¢ to the b-gene interact to enhance HbF expression

in b-thalassaemia? American Journal of Hematology 40, 199±

202.

Rosatelli M.C., Oggiano L. & Leoni G.B. (1989) Thalassaemia

intermedia resulting from a mild b-thalassaemia mutation.

Blood 73, 601±605.

L. Rigoli et al. 377

Ó 2001 Blackwell Science Ltd., Clin. Lab. Haem., 23, 373±378

Rosatelli M.C., Tuveri T. & Scala M.T. (1992) Molecular screen-

ing and fetal diagnosis of b-thalassaemia in the Italian popu-

lation. Human Genetics 83, 590±592.

Saiki R.H., Gelfand D.H. & Stoeffel S. (1988) Primer-directed

enzymatic ampli®cation of DNA with a thermostable DNA

polymerase. Science 239, 487±489.

Sambrook-Fritsc-Maniatis (1989) Analysis and cloning of

eukaryotic genomic DNA. In: Molecular Cloning ± a Laboratory

Manual., 2 edn. pp. 17±18.Cold Spring. Harbor Laboratory

Press, USA.

Tamagnini G.P., Lopes M.C. & Castanheira M.E. (1983) b+

thalassaemia Portuguese type: clinical, haematological and

molecular studies of a newly de®ned form of b-thalassaemia.

British Journal of Haematology 54, 189±200.

Treisman R., Orkin S.H. & Maniatis T. (1983) Speci®c tran-

scription and RNA splicing defects in ®ve cloned b-thalas-

saemia genes. Nature 302, 591±596.

Waye J.S., Eng B. & Patterson M. (1995) Severity of b-thalas-

saemia due to genotypes involving the IVS-I-6 (T®C) muta-

tion. American Journal of Hematology 50, 15±19.

Weatherall D.J. & Clegg J.B. (1981) The Thalassaemia Syndrome,

3rd edn. Blackwell Scienti®c Publication, Oxford.

Wong C., Antonarakis S.E. & Goff S.C. (1986) On the origin and

spread of b-thalassaemia: recurrent observation of four muta-

tion in different ethnic groups. Proceedings of the National

Academy of Sciences of the USA 83, 6529±6532.

Molecular analysis of b-thalassaemia patients378

Ó 2001 Blackwell Science Ltd., Clin. Lab. Haem., 23, 373±378