Molecular analysis of the b-globin gene cluster haplotypes

in a Sudanese population with sickle cell anaemiaA. Y. ELDERDERY*†, J. MILLS*, B. A. MOHAMED‡, A. J. COOPER*, A. O. MOHAMMED†,

N. ELTIEB‡, J. OLD§

INTRODUCTION

Single-nucleotide changes are the most common type

of genetic difference among people. They may create

a new restriction enzyme recognition site or destroy a

naturally occurring one. Although many restriction

enzyme sites have been reported in the human b-glo-

bin gene cluster, seven are nonrandomly associated

with each other and form the basis of the b-globin

gene cluster haplotype. Five sites make up a 50

subhaplotype (from the e-globin gene to the d-globin

gene) and two sites a 30 subhaplotype (containing the

b-globin gene) (Antonarakis, Kazazian & Orkin,

1985).

*School of Pharmacy and

Biomedical Sciences, Portsmouth

University, Portsmouth, UK†Department of Pathology, Faculty

of Medicine, El University of Imam

El Mahdi, Sudan‡Sudan Medical Specialization

Board, Khartoum, Sudan§National Haemoglobinopathy

Reference Laboratory, Molecular

Haematology, John Radcliffe

Hospital, Oxford, UK

Correspondence:

AbozerElderdery, School of

Pharmacy and Biomedical Sciences,

New St Michael Building, University

of Portsmouth, New White Swan

Road, Portsmouth PO1 2DT, UK.

Tel.: 0044 2392 845609;

Fax: 0044 2392 843265;

E-mail: abozer904@hotmail.com

doi:10.1111/j.1751-553X.2011.01388.x

Received 2 June 2011; accepted

for publication 8 September 2011

Keywords

Sickle cell gene, b globin gene,

haplotype, population, Sudan

SUMMARY

Introduction: Sudan has a multiethnic population with a high fre-

quency of Hb S, but little is known about the bS haplotypes in this

population.

Methods: Blood samples from Sudanese Hb SS individuals were

taken at two locations. Family history, age, ethnicity and clinical

symptoms were recorded for each subject. Hb S was investigated

using cellulose acetate electrophoresis (CAE) and cation exchange–

high performance liquid chromatography. Dried blood samples from

93 individuals were used for bS haplotype identification based on

restriction fragment length polymorphism analysis for seven restric-

tion sites.

Results: Haplotypes could be assigned unequivocally to 143 chromo-

somes. Four of the five typical bS-globin haplotypes were identified.

The most frequent was the Cameroon (35.0%), followed by the

Benin (29.4%), the Senegal (18.2%) and the Bantu (2.8%). The

Indian-Arab haplotype was not observed. Three atypical haplotypes

were identified in 17 patients, occurring at a combined frequency of

14.6%. One of these, found at the high frequency of 11.8%, possibly

represented a new Sudan haplotype.

Conclusion: bS Haplotyes were demonstrated successfully from dried

blood samples. A new haplotype is apparent in Sudan, in addition to

the four African haplotypes.

ORIGINAL ARTICLE INTERNATIONAL JOURNAL OF LABORATORY HEMATOLOGY

262 � 2012 Blackwell Publishing Ltd, Int. Jnl. Lab. Hem. 2012, 34, 262–266

International Journal of Laboratory HematologyThe Official journal of the International Society for Laboratory Hematology

These haplotypes modify the severity of sickle cell

anaemia (SCA). In particular, the XmnI polymorphic

site, in the Gc-globin gene promoter region, associates

with increased Hb F during erythropoietic stress and

attenuates the severity of sickle cell disease (Haj Khelil

et al., 2011). This polymorphism is reported as present

in the Arab-Indian haplotype and absent in three of

the African haplotypes.

Five to 10% of patients carry atypical bS haplo-

types. These have previously been reported as a single

mutation at 50 end sequences of the b gene and are

only found in the Bantu (BA) haplotype and the

normal population (Srinivas et al., 1988).

Hb S is the variant of Hb most commonly found in

Sudan where SCA is a particular problem (Bayoumi,

Taha & Saha, 1985; Elderdery et al., 2008). Sudan is

multiethnic, with an admixture of Arab and African

lineages, which may be associated with variability in

Hb types including bS allele haplotypes (Hanchard

et al., 2007). This study has undertaken bS haplotype

analysis in a Sudanese population with Hb SS and

revealed the possible presence of new haplotypes.

MATERIAL AND METHODS

Initially, interviews and questionnaires were used to

collect demographic information: age, sex, ethnicity

(tribe) and family history. A total of 93 patients were

recruited, 42 (39.2%) from the Khartoum military

hospital and 51 (60.8%) from Kosti teaching hospital

in the White Nile state. Approval from the institu-

tional review board at our institution was obtained.

Venous blood samples (5 mL, anticoagulated with

K2-EDTA) were collected from each participant and a

full blood count and Hb electrophoresis performed as

follows. Dried specimens were prepared from the

blood samples as previously described (Kudzi, Dodoo

& Mills, 2009) using Guthrie and FTA cards (Whatman,

UK, whatmaninfo@ge.com). Guthrie card specimens

were stored at )20 �C with desiccant for Hb analysis

by cation exchange–high performance liquid chroma-

tography. For the analysis of bS haplotypes, FTA card

specimens were stored at room temperature. The dried

specimens were treated before amplification of geno-

mic DNA according to the manufacturer’s instructions

(Whatman).

Seven restriction fragment length polymorphism

(RFLP) regions in the b-gene cluster were amplified

using published primers and methods to analyse the

e-HindII, Gc-XmnI, Gc-HindIII, Ac-HindIII, 30wb-Hin-

dII, 50wb-AvaII and b-HinfI RFLP sites (Old, 1996;

Rahimi et al., 2003). Five of the RFLPs studied form

part of the standard b-globin gene haplotype, which is

defined by seven RFLPs: e-HindII, Gc-HindIII, Ac-Hin-

dIII, 50wb-HindII, 30wb-HindII (the 50 subhaplotype),

b-AvaII and b-Bam HI (equivalent to b-HinfI) (the 30

subhaplotype).

PCR amplification was carried out directly from

FTA cards (Kudzi, Dodoo & Mills, 2009). This is the

first demonstration of bS haplotypes from dried blood

samples. Amplified products were subjected to the

appropriate restriction enzyme digestion, separated by

agarose gel electrophoresis and the fragments visual-

ised as previously described (Old, 1996).

RESULTS

The average haematological values for 57 patients

diagnosed homozygous for either the Benin (BE),

Cameroon (CA), Senegal, BA and Sudan haplotypes

were as follows: Hb 7 g/dL (±1), PCV 21% (±2), MCV

99 fL (±4), MCH 26pg (±26). The Hb F level ranged

from 0.5% to 3.0% and was detected with low levels

among the majority of samples studied, apart from

those expressing the BA haplotype.

The bS haplotypes were constructed from the

absence ()) and presence (+) of each of the seven

restriction enzyme sites in 93 individuals with Hb SS.

Of the samples studied, bS haplotypes were identified

by the following algorithm. Only the Arab-Indian

haplotype is (+) for e-Hind II; only the BE is ()) forGc-HindIII; only the BA is ()) for 30wb-HindII and

only the CA is (+) for Ac-HindIII.

The haplotype results for each individual were clas-

sified as uncontested or contested depending upon

whether the segregation of the two haplotypes could

be unequivocally determined. Uncontested bS haplo-

types were identified in patients homozygous for each

RFLP site ()/) or +/+) and for those patients with just

one heterozygous RFLP site ()/+).

For those patients with heterozygous results for

two or more RFLP sites, the segregation of haplotypes

could not be identified because there are more than

two possible haplotypes for each chromosome. These

results were classed as contested and accounted for 43

of 186 chromosomes. Haplotype identification is

� 2012 Blackwell Publishing Ltd, Int. Jnl. Lab. Hem. 2012, 34, 262–266

A. Y. ELDERDERY ET AL. SICKLE CELL HAPLOTYPES IN SUDAN 263

impossible in these cases without further family studies

and deduction by linkage analysis.

Table 1 shows the relative frequencies for each of

the uncontested bS haplotypes found in the Sudanese

population. The CA haplotype was the most preva-

lent, with a frequency of 35% (on 50 of 143 uncon-

tested chromosomes), followed by the BE at 29.4%,

the Senegal at 18.2% and the BA at just 2.8%. The

Arab-Indian haplotype was not represented in this

Sudanese population. Forty of the patients were

homozygous for an uncontested haplotype, with the

CA haplotype being the most frequently detected in

17 patients, followed by the BE in 12 cases, the Sene-

gal in nine cases and the BA in two cases. The most

frequent compound heterozygous haplotypes were the

CA/BE combination in 12 cases, followed by BA/SE in

three cases and CA/SE in two cases.

The remaining uncontested haplotypes were

classed as atypical as they did not match the five

known typical bS haplotypes. Three different atypical

haplotypes were found, each differing from a typical

bS haplotype by the change of one polymorphic site

(Table 1). One was observed on 17 chromosomes, giving

a frequency of 11.8% compared with the other sickle

haplotypes. This haplotype, called atypical 1 in

Table 1, was observed in the homozygous form

in seven Hb SS individuals. The other two atypical

haplotypes were found, each in a single individual

who was homozygous for the haplotype.

Figure 1 shows the results obtained for one patient

after restriction enzyme digestion. The patient had a

()/)) genotype for e-HindII and Gc-XmnI, and a (+/+)

genotype for Gc-HindIII, Ac-HindII, 50wb-AvaII, 30wb-

HindII and b-HinfI RFLPs. Using the algorithm

described earlier, the results show that this patient

is homozygous for the CA haplotype.

DISCUSSION

Previously published studies indicate that Hb S is com-

mon among Sudanese tribes, particularly in the Wes-

tern regions (Kordofan and Darfur). Furthermore,

haemoglobinopathies are also known to be prevalent

Table 1. Prevalence of typical and atypical bS haplotypes in Sudanese Hb SS individuals

Haplotype (RFLPs*) Haplotype (Name) Chromosomes Frequency (%)

) ) + + + + + Cameroon 50 35.0

) ) ) ) + + + Benin 42 29.4

) ) + ) + + + Senegal 26 18.2

) ) + ) ) + + Bantu 4 2.8

) ) ) ) ) + + Atypical 1 – Sudan 17 11.8

) ) ) + + + + Atypical 2 2 1.4

) ) + + ) + + Atypical 3 2 1.4

Not definable Contested 43 23.1

RFLP, restriction fragment length polymorphism.

*The seven RFLP’s used to define the haplotype are in the following order: e-HindII, Gc-XmnI, Gc-HindIII, Ac-HindIII,

30wb-HindII, 50wb-AvaII and b-HinfI.

442

1 2 3 4 5 6 7 8760

650

235

98

327

308

352

480

434

219

108

1000

500

200

100

155

400

Figure 1. Results showing the digestion products for

each of the seven restriction fragment length

polymorphisms in DNA from an individual

homozygous for the Cameroon haplotype. Lane 1 is

the DNA that is (+/+) for b-Hinf. Lane 2 is the DNA

that is (+/+) for 30wb-HindII. Lane 3 is the DNA that

is (+/+) for 50wb-AvaII. Lane 4 is the DNA that is

(+/+) for Ac-HindIII. Lane 5 is the DNA that is (+/+)

for Gc-HindIII. Lane 6 is the DNA that is ()/)) for 50

to Gc-XmnI. Lane 7 is the DNA that is ()/)) for 50 to

e-HindII. Lane 8 shows 100-bp PCR DNA ladder.

� 2012 Blackwell Publishing Ltd, Int. Jnl. Lab. Hem. 2012, 34, 262–266

264 A. Y. ELDERDERY ET AL. SICKLE CELL HAPLOTYPES IN SUDAN

in the Blue Nile and Northern tribes (Elderdery et al.,

2008). Currently, little is known about the frequency

of the haplotype background of the bS mutation and

its effect on blood profiles among the Sudanese, in

part because of the scarcity of appropriate technology

and resources.

The CA haplotype was found to be the most

frequent at 35.0%, with the BE (29.4%) and the Sengal

at 18.2% also observed in many of the samples. This

result is in contrast to other studies in African popula-

tions, in which the CA haplotype has been reported at a

lower frequency in sickled patients (Pagnier et al.,

1984; Nagel et al., 1985). The present findings are in

agreement with the frequency of those haplotypes

compared with a previous study conducted among

Sudanese patients with the Hb S gene by Mohammed

et al. (2006). However, their methodology was unable

to distinguish between the uncontested and contested

haplotypes, and their study was unable to distinguish

between the Senegal and Arab-Indian haplotypes

(Mohammed et al., 2006). The Senegal haplotype was

detected at 18.2% in the current study using the RFLP

method and was not mentioned in the previous study.

After the CA haplotype, the BE haplotype was

found to be the next most prevalent in this study. In

contrast, its frequency was lower than that found in

North Africa, some Arab countries and other Mediter-

ranean countries (Abbes et al., 1991; Samarah et al.,

2009). The BA haplotype was found at the lowest fre-

quency at 2.8% in our study. These data reported

here for the Sudan is in marked contrast to studies

reported from the Central African Republic, where

over 93% of patients were found to be homozygous

for the BA type haplotype (Srinivas et al., 1988).

The Arab-Indian haplotype was not detected at all,

but the remaining four haplotypes (CA, BE, BA and

Senegal) were, which is surprising considering that

the Sudanese are a mix of Arab and African tribes.

The BE, BA and Senegal haplotypes are more com-

mon in Africans and African Americans (Powars &

Hiti, 1993; Zago et al., 2001), whereas the Arab-Indian

haplotype is more common in the Gulf region, Iran

and India (Padmos et al., 1991).

Sudan is multiethnic, with mixed Arab and African

ancestry, as discussed previously. The suggestion

is that this variety in the population will be associated

with variation in Hb types, as detected in multiethnic

groups with high frequency of the bS allele (Hanchard

et al., 2007). Therefore, as bS mutations have arisen

on various DNA backgrounds, a study is needed to

consider the possibility of the presence of haplotypes

other than those reported in the current literature.

We have identified three new atypical bS haplo-

types in the Sudanese population, because of the

observation of Hb SS individuals who were homozy-

gous for these haplotypes. It is possible that there

were other different atypical haplotypes in the 43

contested cases, as although the segregation of partic-

ular haplotypes could not be identified, in many of

these cases, if one chromosome was postulated to

have a typical bS haplotype, then the other chromo-

some was left with an atypical haplotype pattern.

Approximately 95% of bS haplotypes worldwide are

defined as ‘typical’ and the rest are, therefore, atypical

(Srinivas et al., 1988). It has been suggested that the

atypical haplotypes are caused by genetic recombina-

tion among the currently identified haplotypes

(Srinivas et al., 1988). Here the uncontested atypical

haplotypes were detected at a frequency among the

Hb SS patients of 14.6%, which was higher than was

reported previously in African and Arab populations

(Wood et al., 1980; Srinivas et al., 1988). This finding

was similar to the study by Steinberg and co-workers,

concerning the widespread presence of the atypical

b-globin gene in African Americans.

One of the three atypical haplotypes was found at

a much higher frequency than the other two

(Table 1). The haplotype called atypical 1 was found a

frequency of 11.8% in the Sudanese Hb SS patients,

which is higher than that of any of the atypical haplo-

types mentioned in the citations earlier. Therefore,

this may represent a newly defined bS haplotype, spe-

cifically for the population of Sudan. Our samples

were collected from two states, White Nile and Khar-

toum. We did not recruit unique ethnic groups, but

most of them from the Western tribes. A larger popu-

lation size may need to be studied to confirm the

assignment of a fifth ‘typical’ African bS haplotype to

a particular tribe or region of the Sudan.

The Gc-XmnI polymorphisms were included in this

study, because they are associated with a high level of

Hb F, giving patients milder clinical symptoms (Rahimi

et al., 2003, 2007). The XmnI (+) site is present in the

Arab-Indian haplotype, causing clinical symptoms

milder than the other African haplotypes (Rahimi

et al., 2003, 2007). The Sudan haplotype has ())

� 2012 Blackwell Publishing Ltd, Int. Jnl. Lab. Hem. 2012, 34, 262–266

A. Y. ELDERDERY ET AL. SICKLE CELL HAPLOTYPES IN SUDAN 265

Gc-XmnI RFLP and is, therefore, predicted to be associ-

ated with a severe phenotype. The seven patients who

were found to be homozygous for the Sudan haplo-

type all had a severe form of SCA, similar to the Suda-

nese Hb SS patients with the other haplotypes.

In conclusion, the RFLP results reported in this

study have established that bS haplotype detection is

technically feasible by the analysis of dried blood

samples on FTA cards. The findings revealed that four

of the African haplotypes (particularly the CA) were

commonly found among the Sudanese patients carry-

ing the Hb S gene, only the Arab-Indian haplotype

was not observed. The atypical haplotypes were

found at an elevated frequency, and a new haplotype

might be assigned to the Sudanese population. Fur-

ther testing is needed both to confirm these findings

– in particular, the new Sudanese haplotype, and also

to fill in detail of the partially characterised haplo-

types. The latter requires parental sampling, the for-

mer a larger sample base with the missing ethnicities

represented.

ACKNOWLEDGEMENTS

The authors thank Dr Mirgani Ali and staff of the

Haematology Department at the Military hospital in

Omdurman for their assistance with this study. Finan-

cial support from Faculty of Medicines of the Imam

Mahadi, Ministry of Higher education, Sudan, and

from the University of Portsmouth, School of Phar-

macy and Biomedical Sciences, UK, is greatly

acknowledged.

REFERENCES

Abbes S., Fattoum S., Vidaud M., Goossens M.

& Rosa J. (1991) Sickle cell anemia in the

Tunisian population: haplotyping and HB F

expression. Hemoglobin 15, 1–9.

Antonarakis S.E., Kazazian H.H. & Orkin S.H.

(1985) DNA polymorphism and molecular

pathology of the human globin gene clusters.

Human Genetics 69, 1–14.

Bayoumi R.A., Taha T.S. & Saha N. (1985)

A study of some genetic characteristics of the

Fur and Baggara tribes of the Sudan. American

Journal of Physical Anthropology 67, 363–370.

Elderdery A.Y., Mohamed B.A., Karsani M.E.,

Ahmed M.H., Knight G. & Cooper A.J.

(2008) Hemoglobinopathies in the Sudan.

Hemoglobin 32, 323–326.

Haj Khelil A., Moriniere M., Laradi S., Khelif

A., Perrin P., Ben Chibani J. & Baklouti F.

(2011) Xmn I polymorphism associated with

concomitant activation of Ggamma and

Agamma globin gene transcription on a

beta0-thalassemia chromosome. Blood Cells

Molecules and Diseases 46, 133–138.

Hanchard N., Elzein A., Trafford C., Rockett K.,

Pinder M., Jallow M., Harding R., Kwiatkow-

ski D. & McKenzie C. (2007) Classical sickle

beta-globin haplotypes exhibit a high degree

of long-range haplotype similarity in African

and Afro-Caribbean populations. BMC Genet-

ics 8, 52.

Kudzi W., Dodoo A.N. & Mills J.J. (2009) Char-

acterisation of CYP2C8, CYP2C9 and

CYP2C19 polymorphisms in a Ghanaian pop-

ulation. BMC Medical Genetics 10, 124.

Mohammed A.O., Attalla B., Bashir F.M.,

Ahmed F.E., El Hassan A.M., Ibnauf G., Jiang

W., Cavalli-Sforza L.L., Karrar Z.A. & Ibrahim

M.E. (2006) Relationship of the sickle cell

gene to the ethnic and geographic groups

populating the Sudan. Community Genetics

9, 113–120.

Nagel R.L., Fabry M.E., Pagnier J., Zohoun I.,

Wajcman H., Baudin V. & Labie D. (1985)

Hematologically and genetically distinct forms

of sickle cell anemia in Africa. The Senegal

type and the Benin type. New England Jour-

nal of Medicine 312, 880–884.

Old J.M. (1996) Hemoglobinopathies, commu-

nity clues to mutation detection. In: Methods

in Molecular Medicine: Molecular Diagnosis

of Genetic Diseases (ed. R. Elles), 160–183.

Humana Press Inc, Totowa, NJ.

Padmos M.A., Roberts G.T., Sackey K., Kulozik

A., Bail S., Morris J.S., Serjeant B.E. & Serjeant

G.R. (1991) Two different forms of homozy-

gous sickle cell disease occur in Saudi Arabia.

British Journal of Haematology 79, 93–98.

Pagnier J., Mears J.G., Dunda-Belkhodja O.,

Schaefer-Rego K.E., Beldjord C., Nagel R.L. &

Labie D. (1984) Evidence for the multicentric

origin of the sickle cell hemoglobin gene in

Africa. Proceedings of the National Academy

of Sciences USA 81, 1771–1773.

Powars D. & Hiti A. (1993) Sickle cell anemia.

Beta S gene cluster haplotypes as genetic

markers for severe disease expression. Ameri-

can Journal of Diseases of Children 147,

1197–1202.

Rahimi Z., Karimi M., Haghshenass M. & Merat

A. (2003) Beta-globin gene cluster haplotypes

in sickle cell patients from Southwest Iran.

American Journal of Hematology 74, 156–

160.

Rahimi Z., Merat A., Haghshenass M. & Rezaei

M. (2007) Level of hemoglobin F and Gg gene

expression in sickle cell disease and their asso-

ciation with haplotype and Xmni polymorphic

site in South of Iran. Iranian Journal of Medi-

cal Sciences 32, 234–239.

Samarah F., Ayesh S., Athanasiou M., Christakis

J. & Vavatsi N. (2009) Beta(S)-globin gene

cluster haplotypes in the West Bank of

Palestine. Hemoglobin 33, 143–149.

Srinivas R., Dunda O., Krishnamoorthy R.,

Fabry M.E., Georges A., Labie D. & Nagel

R.L. (1988) Atypical haplotypes linked to the

beta S gene in Africa are likely to be the

product of recombination. American Journal

of Hematology 29, 60–62.

Steinberg M.H., Lu Z.H., Nagel R.L., Venkatara-

mani S., Milner P.F., Huey L., Safaya S. &

Rieder R.F. (1988) Hematological effects of

atypical and Cameroon beta-globin gene

haplotypes in adult sickle cell anemia. Ameri-

can Journal of Hematology 59, 121–126.

Wood W.G., Pembrey M.E., Serjeant G.R., Per-

rine R.P. & Weatherall D.J. (1980) Hb F syn-

thesis in sickle cell anaemia: a comparison of

Saudi Arab cases with those of African origin.

British Journal of Haematology 45, 431–445.

Zago M.A., Silva W.A. Jr, Gualandro S., Yok-

omizu I.K., Araujo A.G., Tavela M.H., Gerard

N., Krishnamoorthy R. & Elion J. (2001)

Rearrangements of the beta-globin gene clus-

ter in apparently typical betas haplotypes.

Haematologica 86, 142–145.

� 2012 Blackwell Publishing Ltd, Int. Jnl. Lab. Hem. 2012, 34, 262–266

266 A. Y. ELDERDERY ET AL. SICKLE CELL HAPLOTYPES IN SUDAN