Molecular Biology Reports

, Volume 39, Issue 7, pp 7365–7372

Novel and recurrent LDLR gene mutations in Pakistani hypercholesterolemia patients

  • Waqas Ahmed
  • Muhammad Ajmal
  • Ahmed Sadeque
  • Roslyn A. Whittall
  • Sobia Rafiq
  • Wendy Putt
  • Athar Khawaja
  • Fauzia Imtiaz
  • Nuzhat Ahmed
  • Maleeha Azam
  • Steve E. Humphries
  • Raheel Qamar
Article

Abstract

The majority of patients with the autosomal dominant disorder familial hypercholesterolemia (FH) carry novel mutations in the low density lipoprotein receptor (LDLR) that is involved in cholesterol regulation. In different populations the spectrum of mutations identified is quite different and to date there have been only a few reports of the spectrum of mutations in FH patients from Pakistan. In order to identify the causative LDLR variants the gene was sequenced in a Pakistani FH family, while high resolution melting analysis followed by sequencing was performed in a panel of 27 unrelated sporadic hypercholesterolemia patients. In the family a novel missense variant (c.1916T > G, p.(V639G)) in exon 13 of LDLR was identified in the proband. The segregation of the identified nucleotide change in the family and carrier status screening in a group of 100 healthy subjects was done using restriction fragment length polymorphism analysis. All affected members of the FH family carried the variant and none of the non-affected members nor any of the healthy subjects. In one of the sporadic cases, two sequence changes were detected in exon 9, one of these was a recurrent missense variant (c.1211C > T; p.T404I), while the other was a novel substitution mutation (c.1214 A > C; N405T). In order to define the allelic status of this double heterozygous individual, PCR amplified fragments were cloned and sequenced, which identified that both changes occurred on the same allele. In silico tools (PolyPhen and SIFT) were used to predict the effect of the variants on the protein structure, which predicted both of these variants to have deleterious effect. These findings support the view that there will be a novel spectrum of mutations causing FH in patients with hypercholesterolaemia from Pakistan.

Keywords

Hypercholesterolemia LDLR Mutation Pakistani 

Introduction

Familial hypercholesterolemia (FH; OMIM # 143890) is an inherited heterogeneous cardiovascular disorder that is manifested primarily due to elevated cholesterol levels in the blood. Clinically the disease is characterized by high blood cholesterol, tendon xanthomas, xanthelasmas, premature corneal arcus and atherosclerosis leading to coronary heart diseases [1, 2]. To date three FH causative genes have been identified, these are inherited as monogenic autosomal dominant hypercholesterolemia (ADH). These genes are the low density lipoprotein receptor (LDLR), apo-lipoprotein B-100 (ApoB-100) and proprotein convertase subtilisin/kexin type 9 (PCSK9) [3]. Recently a novel locus HCHOLA4 on 16q22.1 has also been identified to be associated with the disease [4]. Of all the associated genes, mutations in the LDLR gene account for more than 70% of all the FH cases and also result in elevated cholesterol levels in sporadic hypercholesterolemia patients, thus this gene could be used as a preliminary diagnostic marker for the screening of FH as well as sporadic hypercholesterolemia cases.

To date, besides our previous work [5, 6], no LDLR mutation spectrum data are available for the Pakistani population therefore in the present study a FH family and a panel of 27 unrelated hypercholesterolemia patients were studied to identify the causative mutations.

Materials and methods

Ethics declaration

The current study adhered to the tenants of the Declaration of Helsinki and was approved by the Ethics Committee and Institutional Review Board of Shifa College of Medicine and Shifa International Hospital, Islamabad. All patients and healthy individuals were informed about the study in their local language and informed written consent was also obtained from them prior to inclusion in the study.

Sample collection and DNA isolation

All the patients of the FH family (Fig. 1a) and 27 sporadic hypercholesterolemia cases were diagnosed on the basis of US Make Early Diagnosis Prevent Early Death (MEDPED) criteria that are routinely used to diagnose FH patients [7, 8]. In addition the lipid profile of all the participating individuals was determined to ascertain their clinical status. DNA extraction from whole blood and serum lipid profile determination was as described previously [5].
Fig. 1

Pedigree and restriction fragment length (RFLP) based segregation analysis of family. a In the pedigree squares represent male and circles represent females. The halffilledsymbols represent heterozygous individuals, whereas the filledsquare represents the homozygous variant proband (indicated by the arrow). Segregation of ancestral T and variant G alleles is also shown in the pedigree. b RFLP based segregation analysis of mutation (c.1916T > G) among FH-12 family members. The PCR amplified product of 596 bp was digested into two fragments of 129 bp (not shown) and 467 bp for ancestral homozygotes. Homozygous variant sequence remained un-digested while heterozygotes gave three fragments of 596, 467 and 129 bp

Candidate gene analysis

In the case of the proband of the affected family the promoter region and the eighteen coding exons of the LDLR gene including the exon–intron boundaries were sequenced using the primers and polymerase chain reaction (PCR) conditions as described previously [5].

Segregation analysis and mutation screening

Polymerase chain reaction-restriction fragment length polymorphism (PCR–RFLP) analysis was carried out to confirm the segregation of the identified variant in all family members (Fig. 1b), in addition unaffected random controls were also screened using this technique. The samples were PCR amplified with the primers that were used for sequencing above. The c.1916T > G variant results in the elimination of a HincII restriction site, thus for the homozygous wild type sequence the amplified product of exon 13 (596 bp) was digested into two fragments (467 and 129 bp), while the homozygous mutant sequence remained undigested. In the case of heterozygotes all three fragments were obtained after digestion (596, 467 and 129 bp).

High resolution melting (HRM) analysis of sporadic cases

A panel of 27 sporadic hypercholesterolemia patients were screened using HRM analysis, which uses melting temperature curves to differentiate between ancestral and variant alleles [9]. HRM was performed using Quanta Accumelt kit (Quanta BioSciences, Inc. Gaithersburg) as per the manufacturer’s instructions on a Corbett Rotor Gene 6000 system (Corbett Life Sciences, NSW, Australia).

Cloning to determine the allelic status of the double heterozygous mutations

Designing of oligonuclotides for cloning and PCR amplification

For cloning the fragments of the sporadic patient (HCS-39) containing the two mutations, oligonucletides containing SacI restriction site (underlined) were designed covering exon 9 as well as the exon–intron boundaries. The forward oligonucleotide sequence was: 5′-ATC GAG CTC TCC ATC GAC GGG TCC CCT CTG ACC C-3′ and the reverse was: 5′-ATC GAG CTC AGC CCT CAT CTC ACC TGC GGG CCA A-3′. PCR amplification of the 291 bp DNA fragment was carried out by G Strom Thermocycler (GS4) in 30 μl reaction volume using 60 ng template DNA, 0.2 mM dNTPs, 1× polymix, 4 pM of each primer, 2 mM MgCl2 and 1 U Taq DNA polymerase. The thermal profile was as follows, initial denaturation at 93°C for 4 min followed by 35 cycles of denaturation at 94°C for 20 s, annealing at 60°C for 30 s and elongation at 72°C for 45 s. The final elongation was carried out at 72°C for 7 min, the amplified product was then electrophoretically separated on 2% agarose gel at 120 V for 45 min.

Cloning into pGEM vector and sequencing of the cloned fragments

The PCR product and pGEM11 vector were digested with SacI at 37°C for 3 h followed by dephosphorylation of the sticky ends and ligation overnight at 16°C using DNA ligase. The ligation mixture was then used to transform competent E coli strain (DH5α) using a classical heat shock mechanism followed by growth of the cells for an hour at 37°C in Luria Broth (LB). The transformed cells were then plated on LB agar containing ampicillin as a selection marker. After an overnight incubation at 37°C the colonies from each plate were sub-cultured overnight with shaking at 37°C in LB. This was followed by plasmid DNA preparation using Miniprep Plasmid DNA extraction Kit (Qiagen, Hilden; Germany) and sequencing of the recombinant vectors to determine the allelic status of the cloned fragments.

Bioinformatics analysis

In order to determine the predicted effect of the variants in silico web based tool Polymorphism phenotyping (polyPhen; http://genetics.bwh.harvard.edu/pph/) was used, which predicts the possible impact of substituted amino acid residues on the structure and function of the protein, based upon physical and comparative considerations [10]. Moreover HOPE (Homotopy optimization using perturbations and ensembles) [11], tool was used to determine the effect of variation on the 3D structure of the protein.

Statistical analysis

Statistical analysis was carried out for the determination of the association of the identified variant with the disease in the FH family, for this Fisher Exact Probability test (http://faculty.vassar.edu/lowry/odds2x2.html) was used, where a p value of less than 0.05 was taken to be statistically significant.

Results

Analysis of FH family

The mode of inheritance in the clinically diagnosed FH family was autosomal dominant with 7 affected members (Fig. 1a). The proband (IV:5) had undergone coronary artery bypass graft (CABG) before the age of 45 years, in addition the affected members were on statin treatment at the time of blood collection and had no xanthomas when clinically assessed.

Mutation analysis of the LDLR gene promoter sequence, coding exons and splice exon–intron junctions of the proband from the FH family (Fig. 1a) resulted in the identification of a novel missense variant (c.1916T > G, p.(V639G)) in exon 13 in the family. The variant segregated with the disease in all the affected family members (Fig. 1b), the proband (IV: 5) was homozygous for c.1916T > G and had elevated cholesterol levels (372 mg/dL), whereas his sons (V:1 and V:2) along with the individuals III:4, IV:7, IV:9 and V:3 were heterozygous and had TC in the range of 222–280 mg/dL (Table 1). While six normolipidemic individuals (III:1, III:9, IV:1, IV:2, IV:3 and IV:6) were homozygous for the wild type sequence (Fig. 1a, b).
Table 1

The genotype and lipid profiles of the FH family members

ID

Age (years)

Sex

Genotype

Triglycerides

Cholesterol

LDL

HDL

III:1

50

F

TT

157

176

100

45

III:4

57

M

TG

108

239

182

42

III:9

29

M

TT

58

182

102

70

IV:1

30

M

TT

81

147

95

40

IV:2

19

M

TT

170

180

108

43

IV:3

19

M

TT

79

175

118

42

IV:5

51

M

GG

177

372

230

87

IV:6

44

F

TT

123

190

135

42

IV:7

54

M

TG

144

222

157

37

IV:9

20

M

TG

181

249

187

40

V:1

17

M

TG

194

280

209

41

V:2

20

M

TG

165

264

206

40

V:3

16

M

TG

115

255

179

48

Screening of 100 ethnically matched healthy control individuals for the c.1916T > G did not reveal any carriers. Fisher exact probability test revealed a strong association of the mutant G allele (c.1916T > G) with the disease, only for the dominant model (p < 0.001; Table 2), which is in concordance with the dominant mode of inheritance of FH in this family. Moreover the frequency of the G allele was significantly different in the patients of the family (p = 0.001), when compared to the normolipidemic individuals (Table 2).
Table 2

Fischer exact probability test for the association of mutant G allele of c.1916T > G mutation in the Familial Hypercholesterolemia family

Genotype

Normal Individuals

Affected individuals

p(z-test)

p(χ2)

p-value

N = 6 (%)

N = 7 (%)

T/T

6 (100%)

0 (0%)

<0.001(3.6)

 

DM< 0.001

T/G

0 (0%)

6 (86%)

<0.005(3.01)

0.001(13)

RM > 0.05

G/G

0 (0%)

1 (14%)

>0.05(0.96)

  

Allele

Normal Individuals

Affected individuals

p(χ2)

  

N = 6 (%)

N = 7 (%)

T

12 (100%)

6 (43%)

0.001(9.9)

  

G

0 (0%)

8 (57%)

   

DM Dominant model, RM Recessive model, p < 0.05 was taken as significant

Polyphen predicted the amino acid substitution from valine (Val) to glycine (Gly) in the LDLR to be “probably damaging” with a PSCI (position-specific independent counts) score of 2.176. In addition protein modeling showed this substitution to result in variation in the structure of the protein, with the hydrogen bonding potential of the side chain methyl group of Val or the alternate conserved isoleucine (Ile) (Fig. 2a) being lost in the mutated structure containing the Gly residue (Fig. 3), this loss of hydrogen bonding thus seems to have detrimental effects on the structure of LDLR.
Fig. 2

Amino acid conservation in different species. The conserved amino acids are given in whiteletters with black background, whereas the non-conserved residues are shown in blackletters. a Proband amino acid conservation p.V639G. b Sporadic sample HCS-39 amino acids conservation p.T404I_N405T

Fig. 3

The 3D structural effect of the LDL receptor mutation. a 3D structure of the protein with purplearea representing the site of mutation. b The wild type structure with the valine side chain. c The substitution of glycine results in the elimination of the side chain thereby affecting the hydrogen bonding of the amino acid. d The structural representation of valine and glycine

Analysis of sporadic hypercholesterolemia patients

The HRM based analysis followed by sequencing of the 27 unrelated sporadic hypercholesterolemia patients (TC = 220–712 mg/dL) revealed two missense variants, the first was a recurrent variant (c.1211C > T; p.T404I) and the second a novel mutation (c.1214 A > C; N405T) both occurring in exon 9 of one patient (Fig. 4b). In order to determine if both these mutations were on the same allele or different alleles the PCR amplified products were cloned and sequenced, which revealed that both the changes (c.1211 C > T, p.T404I and c.1214 A > C, p.N405T) were on the same allele (Fig. 4c). In silico analysis of these variants separately using Polyphen indicates that the p.T404I variant is Possibly Damaging (score 1.84) and Damaging using Sift, while for the p.N405T variant Polyphen indicates that the variant is Probably Damaging (score 2.31) and Damaging using Sift. Both these changes are highly conserved across different species (Fig. 2b).
Fig. 4

LDLR exon 9 sequencing for sample HCS-39; c.1211_1214delinsTCAC(p.T404I_N405T). a Wild type exon 9 sequence (appropriate bases arrowed). b Original HCS-39 sequencing showing that the sample is heterozygous for both variations (arrowed). c Cloned fragment from HCS-39 showing one allele with both mutant bases (arrowed)

Discussion

Under normal physiological conditions LDLR plays a crucial role in maintaining the cholesterol level in the body [12, 13]. LDLR is a cell surface receptor that is responsible for the uptake of LDL-C from the circulating blood. This uptake process is mediated by apolipoprotein B100 (ApoB), which is part of the LDL particle carrying ~65–70% of the cholesterol in the blood [14, 15]. The internalization of ApoB-LDL-LDLR complex is carried out through clathrin coated pit mediated endosome formation. The low pH inside the endosome results in release of LDL from the receptor, which is then recycled back to the outer membrane [15]. Most of the LDLR mutations identified so far in different world populations result in malfunctioning of the receptor. Defective LDLR usually leads to elevated TC and LDL-C levels, which is the major risk factor for the development of coronary artery disease (CAD) and its related complications [16].

The ligand binding domain of LDL receptor is located near the N-terminal and is encoded by exons 2–6 of the gene. This domain consists of seven LDL class A motifs that are responsible for electrostatic attachment of LDL-C-ApoB to LDLR via the receptor binding motif of ApoB. Adjacent to the ligand binding domain, is the epidermal growth factor precursor homology (EGFP) domain that is encoded by exons 7–14 and consists of three EFG-like motifs in which the last two have 5 YWTD repeats in between. These repeats along with two flanking EGF-like motifs form a six bladed β-propeller that is responsible for the release of LDL particle in the acidic environment of the endosome and also in recycling of the receptor [15].

Mutations in the β-propeller and C terminal region of the last EGF-like motif has been previously reported to cause hypercholesterolemia due to defects in recycling of the receptors because of disruption of the β-propeller structure [15, 17]. As a result the LDL particles are not released from the receptor in the endosomes, which result in the degradation of the LDLR-LDL complex inside the endosomes. This defective recycling may ultimately lower the quantity of cell surface LDLR, which subsequently results in elevation of TC and LDL-C in the blood of the patients.

The identified variants (p.V639G, p.T404I and p.N405T) in the present study are in close proximity of the YWTD repeats and are present at the C terminal region of the third EGF-like motif and are therefore highly likely to be responsible for higher LDL levels in the patients. The combined effect of the latter two variants is difficult to assess, as neither Polyphen nor SIFT allow the analysis of two variants next to each other on the same allele, but it seems unlikely that these two damaging variants would compensate for each other. The V639 in LDLR is at a common site for substitution mutations, Fouchier et al. [18] and Nauck et al. [19] have previously reported a missense substitution (c.1916T > A, p.(V639D)) at the same site in the Dutch and German populations. When we performed in silico analysis of the p.V639D variant, it revealed that both the changes i.e. V639G and V639D were comparable because both the changes affect hydrogen bonding in the 3D structure of the protein. The amino acid conservation of the V639 shows the Val to be conserved only in higher vertebrates, while in most of the vertebrates the V639 is often replaced with Ile (Fig. 2). We also performed protein modeling for this change, which revealed that although Ile is larger in size as compared to the Val but this change does not have any effect on the hydrogen bonding potential of the protein.

To date 1,122 mutations have been identified in the promotor, exons and intron–exon boundaries of the LDLR gene [20], where most of them result in elevated cholesterol levels, the novel c.1916T > G variant in exon 13 in the current Pakistani family is also believed to be the causative variant involved in the higher than normal levels of serum cholesterol in the patients, where the homozygous individual IV:5 had 372 mg/dL serum TC, whereas heterozygotes had levels that ranged between 222–280 mg/dL TC (Table 1). In the case of the sporadic patient (HCS-39) both the changes (p.T404I and p.N405T) were found to be on the same allele and as the cholesterol level (238 mg/dL) of this patient was similar to that of heterozygous cases, therefore this points to the possibility that both of the substitutions are acting as a single heterozygous mutant allele.

Despite the huge mutational data currently available for the LDLR gene there is a need to fully define all the pathogenic variants in the different world populations, especially Pakistan where there is no comprehensive data available for the disease. Among the identified variants most of the LDLR mutations are rare and occur in single families [21], as Pakistan is a multi-racial nation therefore this heterogeneous population is expected to carry a number of novel LDLR mutations. Recently we reported a triple nucleotide variant in the LDLR gene that results in the production of a truncated protein lacking the β-propeller region [6], which was predicted to cause the observed cardiac complications including premature coronary artery disease leading to myocardial infarction. Our current work further defines the LDLR mutation spectrum in sporadic patients and families, which hopefully will result in timely management of the disease and better understanding of the molecular mechanisms involved in pathogenesis.

Notes

Acknowledgments

This work was financially supported by grant No 934 from the Higher Education Commission of Pakistan, awarded to RQ. Part of this work was supported by Shifa College of Medicine through a core grant to RQ. WA was supported by an IRSIP grant from HEC for his studies at UCL. We would like to thank the FH families and healthy individuals for donating their blood. SEH, RW and WP are supported by the British Heart Foundation (RG008/08).

References

  1. 1.
    Rader DJ, Cohen J, Hobbs HH (2003) Monogenic hypercholesterolemia: new insights in pathogenesis and treatment. J Clin Invest 111:1795–1803PubMedGoogle Scholar
  2. 2.
    Zech LA Jr, Hoeg JM (2008) Correlating corneal arcus with atherosclerosis in familial hypercholesterolemia. Lipids Health Dis 7:7PubMedCrossRefGoogle Scholar
  3. 3.
    Garg A, Simha V (2007) Update: update on dyslipidemia. J Clin Endocrinol Metab 92:1581–1589PubMedCrossRefGoogle Scholar
  4. 4.
    Marques-Pinheiro A, Marduel M, Rabes JP, Devillers M, Villéger L, Allard D, Weissenbach J, Guerin M, Zair Y, Erlich D, Junien C, Munnich A, Krempf M, Abifadel M, Jaïs JP; French Research Network on ADH, Boileau C, Varret M (2010) A fourth locus for autosomal dominant hypercholesterolemia maps at 16q22.1. Eur J Hum Genet 18:1236–1242Google Scholar
  5. 5.
    Ajmal M, Ahmed W, Sadeque A, Ali SH, Bokhari SH, Ahmed N, Qamar R (2010) Identification of a recurrent insertion mutation in the LDLR gene in a Pakistani family with autosomal dominant hypercholesterolemia. Mol Biol Rep 37:3869–3875PubMedCrossRefGoogle Scholar
  6. 6.
    Ajmal M, Ahmed W, Akhter N, Sadeque A, Khalid A, Ali SHB, Ahmed N, Azam M, Qamar R (2011) The first report of a pathogenic nonsense triple nucleotide polymorphism in a gene. Genet Test Mol Biomarkers 15:601–606PubMedCrossRefGoogle Scholar
  7. 7.
    Marks D, Thorogood M, Neil HA, Humphries SE (2003) A review on the diagnosis, natural history, and treatment of familial hypercholesterolemia. Atherosclerosis 168:1–14PubMedCrossRefGoogle Scholar
  8. 8.
    Williams RR, Hunt SC, Schumacher MC, Hegele RA, Leppert MF, Ludwig EH, Hopkins PN (1993) Diagnosing heterozygous familial hypercholesterolemia using new practical criteria validated by molecular genetics. Am J Cardiol 72:171–176PubMedCrossRefGoogle Scholar
  9. 9.
    Whittall RA, Scartezini M, Li K, Hubbart C, Reiner Z, Abraha A, Neil HA, Dedoussis G, Humphries SE (2010) Development of a high-resolution melting method for mutation detection in familial hypercholesterolaemia patients. Ann Clin Biochem 47:44–55PubMedCrossRefGoogle Scholar
  10. 10.
    Ramensky V, Bork P, Sunyaev S (2002) Human non-synonymous SNPs: server and survey. Nucleic Acids Res 30:3894–3900PubMedCrossRefGoogle Scholar
  11. 11.
    Venselaar H, Beek TAH, Kuipers RKP, Hekkelman ML, Vriend G (2010) Protein structure analysis of mutations causing inheritable diseases. An e-science approach with life scientist friendly interfaces. BMC Bioinformatics 11:548PubMedCrossRefGoogle Scholar
  12. 12.
    Austin MA, Hutter CM, Zimmern RL, Humphries SE (2004) Genetic causes of monogenic heterozygous familial hypercholesterolemia: a HuGE prevalence review. Am J Epidemiol 160:407–420PubMedCrossRefGoogle Scholar
  13. 13.
    Snozek CL, Lagerstedt SA, Khoo TK, Rubenfire M, Isley WL, Train LJ, Baudhuin LM (2009) LDLR promoter variant and exon 14 mutation on the same chromosome are associated with an unusually severe FH phenotype and treatment resistance. Eur J Hum Gen 17:85–90CrossRefGoogle Scholar
  14. 14.
    Goldstein JL, Brown MS, Anderson RG, Russell DW, Schneider WJ (1985) Receptor-mediated endocytosis: concepts emerging from the LDL receptor system. Annu Rev Cell Biol 1:1–39PubMedCrossRefGoogle Scholar
  15. 15.
    Jeon H, Meng W, Takagi J, Eck MJ, Springer TA, Blacklow SC (2001) Implications for familial hypercholesterolemia from the structure of the LDL receptor YWTD-EGF domain pair. Nat Struct Biol 8:499–504PubMedCrossRefGoogle Scholar
  16. 16.
    Lusis AJ, Fogelman AM, Fonarow GC (2004) Genetic basis of atherosclerosis: part I new genes and pathways. Circulation 110:1868–1873PubMedCrossRefGoogle Scholar
  17. 17.
    Springer TA (1998) An extracellular b-propeller module predicted in lipoprotein and scavenger receptors, tyrosine kinases, epidermal growth factor precursor, and extracellular matrix components. J Mol Biol 283:837–862PubMedCrossRefGoogle Scholar
  18. 18.
    Fouchier SW, Kastelein JJ, Defesche JC (2005) Update of the molecular basis of familial hypercholesterolemia in The Netherlands. Hum Mutat 26:550–556PubMedCrossRefGoogle Scholar
  19. 19.
    Nauck MS, Koster W, Dorfer K, Eckes J, Scharnagl H, Gierens H, Nissen H, Nauck MA, Wieland H, Marz W (2001) Identification of recurrent and novel mutations in the LDL receptor gene in German patients with familial hypercholesterolemia. Hum Mutat 18:165–166PubMedCrossRefGoogle Scholar
  20. 20.
    Leigh SEA, Foster AH, Whittall RA, Hubbart CS, Humphries SE (2008) Update and analysis of the University College London low density lipoprotein receptor familial hypercholesterolemia database. Ann Hum Genet 72:485–498PubMedCrossRefGoogle Scholar
  21. 21.
    Bunn CF, Lintott CJ, Scott RS, George PM (2002) Comparison of SSCP and DHPLC for the detection of LDLR mutations in a New Zealand cohort. Hum Mutat 19:311PubMedCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media B.V. 2012

Authors and Affiliations

  • Waqas Ahmed
    • 1
    • 2
  • Muhammad Ajmal
    • 1
    • 3
  • Ahmed Sadeque
    • 1
  • Roslyn A. Whittall
    • 2
  • Sobia Rafiq
    • 4
  • Wendy Putt
    • 2
  • Athar Khawaja
    • 5
  • Fauzia Imtiaz
    • 6
  • Nuzhat Ahmed
    • 4
  • Maleeha Azam
    • 1
  • Steve E. Humphries
    • 2
  • Raheel Qamar
    • 1
    • 3
  1. 1.Department of Biosciences, Faculty of ScienceCOMSATS Institute of Information TechnologyIslamabadPakistan
  2. 2.Centre for Cardiovascular Genetics, Institute Cardiovascular SciencesUniversity College LondonLondonUK
  3. 3.Shifa College of MedicineIslamabadPakistan
  4. 4.Centre for Molecular GeneticsUniversity of KarachiKarachiPakistan
  5. 5.Shifa International HospitalIslamabadPakistan
  6. 6.Dow International Medical CollegeDow University of Health SciencesKarachiPakistan

Personalised recommendations