340

Short ReportTHE NATIONAL MEDICAL JOURNAL OF INDIA VOL. 14, No.6, 2001

Prenatal diagnosis ofhaemoglobinopathies

SADHNA ARORA, MADHULIKA KABRA,MANJULA MAHESHW ARI, SHIV ARAM SHASTRI,DA YINDER KAUR, DEEPIKA DEKA,ALKA KRIPLANI, P. S. N. MENON

ABSTRACTBilckground. Haemoglobinopathies constitute a major health

problem in the Indian subcontinent. In the absence of anymethod for achieving complete cure and treatment beingexpensive, prenatal diagnosis and selective tennination of anaffected foetus is a feasible option to decrease the disease load.We report our experience with prenatal diagnosis of haemoglo-binopathies over a two-and-a-half year period in 257 pregnan-cies.

Hethods. Amplification refractory mutation system (ARMS)was used to detect IHhalassaemia, haemoglobin E and sickle cellmutations.

Results. Five mutations in the B-globin gene which arecommon in the Indian population were detected in 92.3% ofmutant chromosomes, whereas 3.1 % of chromosomes carriedrare mutations followed by 0.8% haemoglobin E and 0.4%sickle cell mutations. Mutations in 3.3% chromosomes wereuncharacterized. The prenatal procedure, carried out early inpregnancy, was a chorionic villus sampling in most cases. Aconfirmed diagnosis based on ARMS-PCR was given in 241(93.8%) cases. In 10 cases (3.9%) Iinkageanalysiswasrequiredto confirm the foetal status, as mutations in both parents were notidentified or the chorionic villus sample carried the singleidentified mutation. Four families with haemoglobin E-Bthalassaemia and one family with sickle cell disease were alsoincluded. Ofthe study population, 91.25% ofthe couples hada previous child with haemoglobinopathy, whereas 8.75% of thecouples came before the birth of the first affected child.

Conclusion. We conclude that ARMS-PCR is a highlysensitive technique for detecting mutations in the B-globin geneand its efficacy in the prenatal diagnosis of haemoglobinopathiesis proven.Natl Med J India 2001; 14:340-2

INTRODUCTION

Haemoglobinopathies including B-thalassaemia, sickle cell dis-

All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110029,India

SADHNAARORA, MADHULIKA KABRA,MANJULA MAHESHW ARI, SHIV ARAM SHASTRI,DAVINDER KA UR, P. S. N. MENON Genetics Unit, Department of

PaediatricsDEEPlKA DEKA, ALKA KRIPLANI Department of Obstetrics and

Gynaecology

Correspondence to MADHULIKA KABRA; madhulikakabratehotmail.commadhu@medinst.ernet.in

© The National Medical Journal of India 2001

ease and haemoglobin E (HbE) are major public health problemsin the Indian subcontinent. The frequency of B-thalassaemiacarriers varies from 3% to 17% in different populations; thalas-saemia being the commonest single gene disorder in India. I About9000 homozygous babies are born every year in India.' Sickle celldisease is more common in some tribal populations, whereas HbE,the variant, is more prevalent in the eastern parts of India.v'Homozygous HbE status is asymptomatic but in combination withB-thalassaemia mutation, heterozygotes have manifestations simi-lar to those of B-thalassaemia major and intermedia.

Therapy for B-thalassaemia is expensive and there is no defi-nite cure except bone marrow transplantation.' One of the meansof reducing the burden of the disease is by prenatal diagnosis andselective termination of affected foetuses. Prenatal diagnosis of B-thalassaemia was initially done using foetal blood for globin chainsynthesis studies? and is still the method of choice when mutationsare not identified in the parents, or in cases of advanced gestationalage. Later, DNA-based techniques became more popular, thecommon ones being dot blot," reverse dot blot.!" PCR-basedamplification refractory mutation system (ARMS),1O denaturinggradient gel electrophoresis, II single strand conformation poly-morphism'? and the latter two followed by DNA sequencing.P:"Of these, ARMS-PCR is a simple and cost-effective method.DNA studies can use any foetal tissue such as amniotic fluid cellsor, more commonly, chorionic villus samples (CYS).

We report our experience of prenatal diagnosis ofhaemoglobinopathies in 257 pregnancies using CYS.

SUBJECTS AND METHODS

A total of 257 prenatal diagnostic tests were performed on 240pregnant women from January 1997 to September 1999. Ofthese,223 women had the test done once (including 2 twin pregnancies)and 17 women conceived twice during this period and henceunderwent prenatal diagnosis twice. CYS was repeated for onepatient because of an inadequate first sample.

Identification of mutation in parents and affected childrenPeripheral blood samples were collected in EDT A vacutainers fromparents and affected children. (Blood samples of the affectedchildren were taken at least three weeks after the last transfusion.)Genomic DNA was extracted by the standard phenol-chloroform(n=113 families)" or salt method (n=127 families)." B-thalassaemiamutations in the parents and affected children were identified usingthe ARMS- PCR technique.'? We first tested for the five mutationsreported commonly in the Indian population [IYS-l, nt5 (G~C),IVS-l, ntl (G~T), codon 8/9 (+G), codon 41142 (-CTIT) and619bp deletion] 18,19followed by 12 rare mutations [codon 16 (-C),Capsite +1 (A~C), codon 15 (G~A), -88 (C~T), codon 30(G~A), Codon 30 (G~C), codon 5 (-CT), codon 47/48 (+ATCT),IVS-2, nt837 (T ~G), IVS-l, ntl l O (G~A), codon 88 (+T), IVS-1 3' end del 25 bp]. The primers and protocols for the tests wereobtained from the published literature.lo,18,2o Sickle cell mutationand HbE were also tested using the ARMS technique.v-?

Identification of mutations in the foetusFoetal tissue used for prenatal diagnosis was obtained by chorion-ic villus sampling (CYS, usually 10-20 mg) in all except 1 case.

ARORA et al. : PRENATAL DIAGNOSIS OF HAEMOGLOBINOPATHIES

The route of CVS was either transcervical or transabdominal,based on the choice of the obstetrician. Chorionic villus tissue wasdissected under the inverted microscope to remove maternaldecidua and DNA extracted using a standard protocol."

The CVS was tested for parental mutation(s) to make a prenataldiagnosis. Wherever parental mutations were identical, normal aswell as mutant primers were used in separate reaction tubes todifferentiate between heterozygous and homozygous cases. Posi-tive and negative controls were run with each sample. Figure 1shows the results of prenatal diagnosis in one family. The resultsof the prenatal test were usually ready within a week. Restrictionfragment length polymorphism (RFLP) studies were done in theGenetics Laboratory at Sir Ganga Ram Hospital, New Delhi incases where mutations were not identified in the parents or insituations where the CVS carried the identified mutation. To ruleout maternal contamination, polymorphism of apolipoprotein B(Apo B) was examined by PCR whenever necessary." Maternalcontamination was not checked in all heterozygous foetusescarrying the maternal mutation but was used only in situationswhere CVS tissue was scanty and there was doubt about foetaltissue separation.

RESULTSDuring the study period, 257 prenatal diagnoses were offered in240 pregnant women for haemoglobinopathies. In 223 women,prenatal testing was done in one pregnancy whereas in 17 womentwo CV biopsies were done. The gestational age at the time ofsampling varied from 9 to 23 weeks (mean: 12.5 weeks). Fifty-six(21.8%) CVS were obtained by the transcervical route and 200(77.8%) by the transabdominal route. In one subject cord bloodwas used for diagnosis. Two hundred and nineteen couples had achild previously affected with the disease whereas 21 couplesrequested for a prenatal diagnosis before the birth of an affectedchild.

Out of 480 carrier parents, one of the five common B-thalassaemia mutations were identified in 443 (92.29%), rare B-thalassaemia mutations in 15 (3.1%), HbE in4 (0.8%) and HbS in2 (0.4%). Mutations in 16 individuals were not characterized(3.3%). The commonest B-thalassaemia mutation identified wasIVS-l, nt5 (G--7C) in 137 (28.5%), followed by 619bp del in 132(27.5%; TableI).

Mutations were identified in both partners in 241 out of 257pregnancies (93.8%). Both partners had the same mutations in 85families though only 9 couples were consanguineous. Amongthese 85 families, 70 couples had a common place of origin. Of the70 couples who originated from the same place, 35 were from

TABLE1. Distribution of various mutations

Type of mutation Number of cases positive (%)

Common (n=443)

IVS-I, nt5 (0-tC)619bp delIVS-I, nt l (G-tT)Codon 8/9 (+0)Codon 41142 (-CTIT)Rare (n=15)

Codon 16 (-C)Codon 30 (G-tC)Cap site + 1 (A-tC)Codon 88 (+T)Codon 30 (G-tA)

137 (28.5)132 (27.5)77 (16.0)59 (12.3)38 (7.9)

8 (1.7)3 (0.6)2 (0.4)I (0.2)1 (0.2)

341

FIG1. Electrophoregram showing the results of prenatal diagno-sis in one family

LANE 1: Molecular weight marker (<\lXI74, Bangalore genei,India)LANE 2: Mother positive for 619 bp deletionLANE 3: Father negative for 619 bp deletionlane 4 and 5: foetus positive for 619 bp deletion (in duplicate)lane 6: father positive for ivs-l , ntl (G--7T)lane 7: negative control for ivs-I, ntl (G--7T)lane 8 and 9: foetus positive for 619 bp deletion and ivs-L, ntl(G--7T) (affected with thalassaemia major)

regions now in Pakistan, 15 from Uttar Pradesh, 5 each fromPunjab and Haryana, 3 each from Bihar and Rajasthan, 2 fromMaharashtra, 1 each from Madhya Pradesh and Tamil Nadu.Fifteen couples, though carrying same mutations, originated fromdifferent places: 10from regions now in Pakistan, 8 from Haryana,3 from Rajasthan, 2 each from Punjab, Delhi and Bihar, and 1eachfrom Maharashtra, Madhya Pradesh and Jammu and Kashmir.

Among the 241 pregnancies in which the mutation was iden-tified in both partners, 67 (27.8%) foetuses were normal, 137carriers (56.8%) and 37 (15.4%) were affected. In the remaining16 couples, the mutation was identified in only one parent in 15families, whereas no mutations were identified in one couple. Ofthese 16 families, 5 foetuses did not carry the identified mutationso the foetus was reported to be 'normal' or 'carrier'. In theremaining 11, RFLP analysis for prenatal diagnosis was done atthe Genetics Laboratory of Sir Ganga Ram Hospital, New Delhi(n=lO) and one foetus was tested using the globin chain synthesisby cordocentesis at the Institute ofImmunohaematology, Mumbai,as the gestation was advanced. Of these 11 cases, a prenataldiagnosis was available in 7 (2 were affected and 5 were carriers).The remaining 4 families were lost to follow up. In 7 samples therewere problems in DNA amplification and the PCR signals werefaint. However, re-extraction of the DNA and repeat PCR solvedthe problem. One CVS procedure was repeated owing to inad-equate sample and maternal contamination.

Follow upFollow up of 180 foetuses (out of 204) who were diagnosed to be

342

normal carriers has shown that 150 of these babies are asympto-matic and> 1 year of age. Twenty-four babies are -cl year (mini-mum age: 6 months) and asymptomatic till date. Six foetuses wereaborted spontaneously; however, all abortions took place morethan four weeks after the procedure. Haemoglobin electrophoresiswas not done routinely. In 11families where it was done the resultsmatched the prenatal diagnosis. Mutation analysis was donepostnatally in 62 cases, which confirmed the prenatal diagnosis.Follow up for only 5 of the affected foetuses is available in whompost-termination samples could be obtained. These confirmed theprenatal diagnosis.

DISCUSSIONWe performed prenatal diagnosis for haemoglobinopathies using256 CVS and 1 cordocentesis, including B-thalassaemia in 252(98.05%), HbE in 4 (1.55%) and sickle cell disease in one family.Most of the CVS procedures were done early «20 weeks) whereasonly 4 0.55%) were done after 20 weeks of pregnancy. Thedelayed procedures were due to patients presenting late because oflack of awareness, difficulty in identifying the mutation in thefamily or due to the decision of the attending obstetrician.

Though there are various methods to detect mutations, ARMSis a simple and cost-effective way to identify mutations in at-riskcouples during the first trimester of pregnancy. The ARMStechnique gave us good results and we were able to detectmutations in both parents with a precise prenatal diagnosis in93.8% of foetuses at risk. This is in concordance with previouslyreported studies from India.25,26 In 7 families, there was difficultyin amplifying the DNA. In 4 of them the quality of DNA was notgood and this was rectified by re-extraction using phenol-chloroform. In 3 samples the non-amplification was probablydue to the degeneration of primers and freshly prepared workingsolutions allowed us to obtain the results. In ARMS-PCR, as theprimers are designed based on the mismatch only at the 3' endand there can be problems due to degeneration especially if theprimers are old. Only 3.9% of patients required RFLP analysisfor prenatal diagnosis. Fortunately, all these non-informativefamilies had affected children and RFLP studies were possible.

The same mutations were found in 35.4% of couples but only3.8% were consanguineous marriages. This may be explained bythe tradition of marriages in the same communitylcaste in India.It was interesting that 21 couples sought a prenatal diagnosisbefore the birth of an affected child because either there was anaffected child among the first-degree relatives or the wife had alow haemoglobin level which led to a subsequent revelation ofcarrier status of both husband and wife. This indicates an increas-ing awareness among doctors and the general public.

Maternal contamination is a problem when the CVS is <10 mgor has maternal mutations only. Hypervariable human DNAmarkers which show high levels of heterozygosity are useful inDNA typing, antenatal diagnosis, forensic medicine and linkagemapping." Foetal samples can be checked for maternal contamina-tion using these markers.":" In our study, this problem wasovercome by doing a PCR for Apo B polymorphism in selectedcases.

We conclude that prenatal diagnosis and selective terminationof affected foetuses is a feasible way to control haemoglobinopathiesin India. ARMS or any other technique based on allele-specificamplification is suitable. Till such time a formal campaign todetect high-risk couples is in place, we can at least help familieshaving one (or more) children affected with haemoglobinopathies.There is an urgent need to increase awareness among the masses

THE NATIONAL MEDICAL JOURNAL OF INDIA VOL. 14, NO. 6, 200)

as well as health care providers about the availability and utility ofgenetic services.

ACKNOWLEDGEMENTSWe are grateful to the Genetics Department, Sir Ganga Ram Hospital, NewDelhi and the Institute of Immunohaematology, Mumbai (rCMR) for doingprenatal testing in non-informative families. We also acknowledge the financialsupport received from the Department of Biotechnology. We gratefully ac-knowledge the cooperation extended by our patients in making our servicebetter.

REFERENCESModell B, Bulzhenkov V. Distribution and control of some genetic disorders. WorldHealth Stat Q 1988;41:209-18.

2 Choudhary VP, Kotwal J, Saxena R. Thalassemia screening and control programme.Paediatr Today 1998;1:283--6.Kaur M, Das GP, Verma IC. Sickle cell trait and disease among tribal communities inOrissa, Madhya Pradesh and Kerala.lndian} Med Res 1997;105: 111-16.

4 Pati AR, Bhargava M, Rath PK, Kochupillai V. Unusual features of haemoglobin Ethalassaemia.lndian} Med Res 1985;81:409-12.

5 Chandy M. Bone marrow transplantation forthalassemia. PaediatrToday 1998;1:399-412.

6 Kan YW, GolbusMS, KJein P, Dozy AM. Successful application in prenatal diagnosisofa pregnancy atrisk for homozygous beta-thalassemia. N Engl J Med 1975;292: 1099.

7 Saiki RK, Bugawan TL, Horn GT, Mullis KB, Erlich HA. Analysis of enzymaticallyamplified beta-globin gene a.nd HLA-DQ alpha DNA with allele- specific oligonucleotideprobes. Nature 1986;324: 163-6.

8 Saiki RK, Walsh PS, Levenson CH, Erlich HA. Genetic analysis of amplified DNAwith immobilized sequence-specific oligonucleotide probes. Proc Natl Acad Sci USA1989;86:6230-4.

9 CaiSP, WalIJ,Kan YW,ChehabFF. Reversedotblotprobesforthescreeningofbeta-thalassemia mutations in Asians and American blacks. Hum Mutat 1994;3:59--63.

10 OldJM, Varawalla NY, Weatherall OJ. Rapid detecuon and prenatal diagnosis of beta-thalassemia studies in Indian and Cypriot populations in the UK. Lancet 1990;336:834-7.

II Garewal G, Fearon CW, Warren TC, Marwaha N, MarwahaRK, MahadikC, etal. Themolecular basis ofbeta-thalassaemia in Punjabi and Maharashtrian Indians includes amulti locus aetiology involving triplicated alpha-globin loci. Br J Haematol1994;86:372--6.

12 Orita M,lwahana H, KanazawaH, Hayashi K, Sekiya T. Detection of polymorphismof human DNA by gel electrophoresis as single-strand conformation polymorphisms.Proc Natl Acad Sci USA 1989;86:2766-70.

13 Jain PK, Dozy Alv!, Verma IC, Chehab FF. A new frameshift mutation, insertion ofATCT, at codon 48 in the beta-globin gene causes beta-thalassemia in an Indianproband. Hum Mutat 1994;3:397-8.

14 Wong C, Dowling CE, Saiki RK, Higuchi RG, Erlich HA, Kazazian HH Jr.Characterization of beta-thaI assaemi a mutations using direct genomic sequencing ofamplified single copy DNA. Nature 1987;330:384-6.

15 Old JM. Gene analysis. In: Weatherall OJ (ed). Methods in haematology: Thethalassaemias. Edinburgh:Churchill Livingstone, 1982:74-102.

16 Miller SA, Dykes DO, Polesky HF. A simple salting out procedure for extracting DNAfrom human nucleated cells. Nucleic Acids Res 1988;16: 1215.

17 Newton CR, Graham A, Heptinstall LE, Powell SJ, Summers C, Kalsheker N, et al.Analysis of any point mutation in DNA. The amplification refractory mutation system(ARMS). Nucleic Acids Res 1989;17:2503-16.

18 VarawallaNY,OldJM,SarkarR, Venkatesan R, Weatherall OJ. The spectrum of beta-thalassaemia mutations on the Indian subcontinent: The basis for prenatal diagnosis.Br} Haematol1991 ;78:242-7.

19 Verma IC, Saxena R, Thomas E, Jain PK. Regional distribution of beta-thalassemiamutations in India. Hum Genet 1997;100: 109-13.

20 Varawalla NY, Old JM, Weatherall OJ. Rare beta-thalassaemia mutations in AsianIndians. Br J Haematol1991 ;79:640-4.

21 Agarwal S, Gulati R, Singh K. Hemoglobin E beta-thalassemia in Uttar Pradesh.Indian Pediatr 1997;34:287-92.

22 Wu DY, Ugozzoli L, Pal BK, Wallace RB. Allele-specific enzymatic amplification ofbeta-globin genomic DNA for diagnosis of sickle cell anemia. Proc NatlAcad Sci USA1989;86:2757--60.

23 Old JM. Fetal DNA analysis. In: Davies KE (ed). Human genetic diseases-Apractical approach. Oxford:lRL Press, 1993: 1-19.

24 Decorte R, Cuppens H, Marynen P, Cassiman JJ. Rapid detection of hyper variableregions by the polymerase chain reaction technique. DNA Cell Bioi 1990;9:461-9.

25 Saxena R, Jain RK, Thomas E, Verma Ie. Prenatal diagnosis of beta-thaI assaemi a:Experience in a developing country. Prenat Diagn 1998;18: 1-7.

26 OldJ, Petrou M, Varnavides L, Layton M, Modell B. Accuracy of prenatal diagnosisfor haemogloin disorders in the UK: 25 years' experience. Prenat Diagn 2000;20:986-99.

27 Jeffreys AJ, Wilson V, Thein SL. Hypervariable 'minisatellite' regions in humanDNA. Nature 1985;314:67-73.