Blood Res 2021; 56(4):
Published online December 31, 2021
https://doi.org/10.5045/br.2021.2021016
© The Korean Society of Hematology
Correspondence to : Rudragouda Bulagouda, M.D.
Human Genetics Laboratory, Department of Anatomy, Shri B.M Patil Medical College, Hospital and Research Centre, BLDE (Deemed to be University), Vijayapura 586101, India
E-mail: drravisb2012@gmail.com
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Hemophilia B (HB) is an X-linked bleeding disorder resulting from coagulation factor IX defects. Over 3,000 pathogenic, HB-associated mutations in the F9 gene have been identified. We aimed to investigate the role of F9 variants in 150 HB patients using sequencing technology.
Methods
F9 gene sequences were amplified from peripheral blood-derived DNA and sequenced on an Applied Biosystems (ABI) 3500 Sanger sequencing platform. Functional and structural predictions of mutant FIX were analyzed.
Results
Among 150 HB patients, 102 (68%), 30 (20%), and 18 (12%) suffered from severe, moderate, and mild HB, respectively. Genetic analysis identified 16 mutations, including 3 novel mutations. Nine mutations (7 missense and 2 stop-gain) were found to be pathogenic. Only 3 mutations (c.127C>T, c.470G>A, and c.1070G>A) were associated with different severities. While 2 mutations were associated with mild HB cases (c.304C>T and c.580A>G), 2 (c.195G>A and c.1385A>G) and 3 mutations (c.223C>T, c.1187G>A, and c.1232G>A) resulted in moderate and severe disease, respectively. Additionally, 1 mutation each was associated with mild-moderate (c.*1110A>G) and mild-severe HB disease (c.197A>T), 4 mutations were associated with moderate-severe HB cases (c.314A>G, c.198A>T, c.676C>T, and c.1094C>A). FIX concentrations were lower in the mutated group (5.5±2.5% vs. 8.0±2.5%). Novel p.E66D and p.S365 mutations were predicted to be pathogenic based on changes in FIX structure and function.
Conclusion
Novel single nucleotide polymorphisms (SNPs) largely contributed to the pathogenesis of HB. Our study strongly suggests that population-based genetic screening will be particularly helpful to identify risk prediction and carrier detection tools for Indian HB patients.
Keywords Hemophilia B, F9 gene, Stop-gain mutation, Missense mutation, India
Hemophilia is an X-linked blood disorder that causes sustained bleeding after injuries or trauma. Symptoms mainly include bleeding in the joints and muscles. The term hemophilia was first used by Friedrich Hopff, University of Zurich, in 1828 [1, 2]. Hemophilia B (HB) is caused by defects in coagulation factor IX. It is also called “Christmas disorder”, after it was identified by Stephen Christmas in 1952 [3]. FIX activity can be classified as severely (<1%), moderately (1–5%), and mildly (5–30%) impaired [1, 4, 5]. HB is less common than hemophilia A (HA) (1 in 25,000 males vs. 1 in 5,000 males worldwide) [6, 7]. Approximately 14,000 people with hemophilia have been registered at the Haemophilia Federation of India; however, hemophilia remains under-diagnosed and many cases are thus not registered. HA occurs in 1 out of 10,000 male births, while HB occurs in 1 out of 30,000 male births [8].
The
Recently, more than 3,000 pathogenic mutations and neutral polymorphisms have been identified in the
This study was approved by the Institutional Ethical Committee of Shri B. M. Patil Medical College, Hospital and Research Centre, BLDE (Deemed to be University), Vijayapura (Ref. No.: BLDE (DU)/IEC/340/2018-19), and the SNMC Institutional Ethics Committee of Human Subject Research, Bagalkot (Ref. No.: SNMC/IECHSR/2018-19/A-B2/1.1). Written informed consent was obtained from all HB patients before blood sample collection. A total of 150 HB patients were included in this study, which were followed up at 12 different hemophilia societies across the Karnataka state of India. A detailed clinical history was obtained from all HB patients.
Peripheral blood from HB patients was collected in EDTA vacutainers (BD, Franklin Lakes, NJ, USA). Prior to DNA isolation, the FIX concentration was measured and inhibitor assay were performed. DNA was extracted from peripheral blood using a blood and tissue DNA extraction kit (QIAGEN, Hilden, Germany). All exonic regions were amplified and the products were sequenced on an Applied Biosystems (ABI) 3500 Sanger sequencing platform using the BigDye Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher Scientific, Waltham, MA, USA). Results were analyzed using DNA sequence analysis software v5.4.
The pathogenicity of the novel non-synonymous variants was analyzed using bioinformatics tools, such as PROVEAN (http://provean.jcvi.org/seq_submit.php), PolyPhen-2 (http://genetics.bwh.harvard.edu/pph2/index.shtml), PHD SNP (https://snps.biofold.org/phd-snp/phd-snp.html), SNPs & GO (https://snps.biofold.org/snps-and-go/snps-and-go.html), PANTHER (http://www.pantherdb.org/), and SNAP2 (https://www.rostlab.org/services/snap/). The conservation property of missense variants was investigated using the Clustal Omega multiple sequence alignment tool (https://www.ebi.ac.uk/Tools/msa/clustalo/). A 3D model of wild type and mutant FIX proteins was predicted using the Swiss model (https://swissmodel.expasy.org/), and the results were visualized and analyzed using the UCSF Chimera program.
Of the 150 HB patients included in this study, 102 (68%; FIX concentration, 0.6±0.2; age of onset, 2.0±1.0 y), 30 (20%; FIX concentration, 2.5±1.3; age of onset, 7.5±2.8 y), and 18 (12%; FIX concentration, 8.0±2.6; age of onset, 10.0±3.5y) suffered from severe, moderate, and mild HB, respectively. The detailed clinicopathological parameters are summarized in Table 1. In our study, we recorded 16 mutations. Of those, 1 was a synonymous mutation, 12 were missense mutations, 2 were stop-gain mutations, and 1 was a 3’ UTR variant. Notably, 13 (81.25%) mutations were previously reported, but 3 (18.75%) were novel mutations, which had not been entered into any of the human SNP databases. The majority of the mutations were found in exon 8 of the gene and largely comprised missense mutations. Exon 8 showed a high number of mutations compared to other exons of the
Table 1 Clinicopathological characteristics of the study population.
Clinicopathological parameters | Disease severity | ||||
---|---|---|---|---|---|
Severe | Moderate | Mild | |||
(Factor IX <1%), N=102 (68%) | (Factor IX 1–5%), N=30 (20%) | (Factor IX >5%), N=18 (12%) | |||
Mean factor IX levels (%) | 0.6±0.2 | 2.5±1.3 | 8.0±2.6 | ||
Family history of bleeding (N) | 80 | 10 | 7 | ||
Mean age of onset (yr) | 2.0±1.0 | 7.5±2.8 | 10.0±3.5 | ||
Mutation frequency (within group) | 48% (49/102) | 86.7% (26/30) | 83.3% (15/18) | ||
Inhibitor-positive (N) | 23 | 11 | 4 | ||
Inhibitor-negative (N) | 79 | 19 | 14 | ||
Hemoglobin concentration | Normal | Normal | Normal | ||
Platelet count | Normal | Normal | Normal | ||
Prothrombin time | Normal | Normal | Normal | ||
Activated partial thromboplastin time | High | High | High | ||
Bleeding sites | |||||
Joints | 55 (54%) | 19 (63.3%) | 12 (66.7%) | ||
Gum | 17 (16.7%) | 6 (20%) | 4 (22.2%) | ||
Muscle | 13 (12.7%) | 1 (3.3%) | 2 (11.1%) | ||
Epistaxis | 7 (6.9%) | 4 (13.3%) | 0 | ||
Petechiae | 10 (9.8%) | 0 | 0 |
Table 2 List of mutations recorded in the
Mutation type | Nucleotide change | cDNA position | Amino acid change | Exon | Status | Frequency, N (%) |
---|---|---|---|---|---|---|
Missense | g.11313C>T | c.127C>T | p.R43W | 2 | Reported (rs1603264205) | 9 (6.0%) |
Missense | g.11381G>A | c.195G>A | p.M65I | 2 | Reported (rs763568424) | 1 (0.66%) |
Missense | g.11383A>T | c.197A>T | p.E66V | 2 | Reported (CM940423) | 3 (2.0%) |
Missense | g.11384A>T | c.198A>T | p.E66D | 2 | Not reported | 12 (8.0%) |
Stop-gain | g.11409 C>T | c.223C>T | p.R75a) | 2 | Reported (rs137852227) | 2 (1.33%) |
Missense | g.15369 T>C | c.304C>T | p.C102R | 4 | Reported (CM960574) | 2 (1.33%) |
Missense | g.22664 T>C | c.314A>G | p.G143R | 5 | Reported (CM940499) | 4 (2.66%) |
Missense | g.22706 G>A | c.470G>A | p.C157Y | 5 | Reported (rs1367198680) | 17 (11.33%) |
Missense | g.25386 A>G | c.580A>G | p.T194A | 6 | Reported (rs6048) | 3 (2.0%) |
Missense | g.25482C>T | c.676C>T | p.R226W | 6 | Reported (rs137852240) | 6 (4.0%) |
Missense | g.36020 G>A | c.1070G>A | p.G357E | 8 | Reported (rs137852275) | 8 (5.33%) |
Stop-gain | g.36044C>A | c.1094C>A | p.S365a) | 8 | Not reported | 15 (10.0%) |
Missense | g.36137G>A | c.1187G>A | p.C396Y | 8 | Reported (rs137852273) | 1 (0.66%) |
Missense | g.36182 G>A | c.1232G>A | p.S411N | 8 | Reported (rs137852276) | 3 (2.0%) |
Synonymous | g.36335 A>G | c.1385A>G | p.Ter462= | 8 | Reported (rs561793582) | 1 (0.66%) |
3’ UTR | g.37446A>G | c.*1110A>G | ……… | 8 | Not reported | 3 (2.0%) |
a)NG_007994.1, NM_000133.4, and NP_000124.1 reference sequences wereused for genomic DNA variant nomenclature, coding region variant nomenclature, and protein variant nomenclature, respectively. All nomenclatures were made according to Human Genome Variation Society (HGVS) guidelines.
The novel missense mutation c.198A>T was found in the
Table 3 Genotypic and phenotypic associations of the mutations recorded in the study population.
Mutation type | cDNA position | Patients | Factor IX (IX) concentration level (%) | Age of onset | Inhibitor-positive (N) | Inhibitor-negative (N) | Clinical significance | ||
---|---|---|---|---|---|---|---|---|---|
m | M | S | |||||||
Missense | c.127C>T | 03 | 04 | 02 | 6.5, 8.0, 10 (m) 2.0, 2.3. 1.8, 2.0 (M) 0.8, 0.5 (S) | 10, 12, 8 (m) 7, 6, 9, 9 (M) 1, 1.5 (S) | 2 (m) 2 (M) - | 1 (m) 2 (M) 2 (S) | Pathogenic |
Missense | c.195G>A | 00 | 01 | 00 | 3.0 (M) | 7.5 (M) | 1 (M) | - | Likely pathogenic |
Missense | c.197A>T | 02 | 00 | 01 | 6.0, 9.0 (m) 0.3 (S) | 13, 11 (m) 1 (S) | - | 2 (m) 1 (S) | Likely benign |
Missense | c.198A>T | 00 | 04 | 07 | 1.5, 2.0. 2.5, 2.0 (M) 0.8, 0.5, 0.1, 0.6, 0.6, <0.01, 0.01 (S) | 6, 7, 7, 8 (M) 1, 2, 1, 1.5, 2, 2, 1.6 (S) | 4 (M) 4 (S) | - 3 (S) | Pathogenic |
Stop-gain | c.223C>T | 00 | 00 | 02 | <0.01, <0.01 (S) | 1, 1 (S) | 1 (S) | 1 (S) | Pathogenic |
Missense | c.304C>T | 02 | 00 | 00 | 12.0, 10.0 (m) | 13, 10 (m) | 2 (m) | - | Likely pathogenic |
Missense | c.314A>G | 00 | 01 | 03 | 2.0 (M) 0.2, 0.1, 0.01 (S) | 8 (M) 1, 2, 1 (S) | - | 1 (M) 3 (S) | Pathogenic |
Missense | c.470G>A | 03 | 05 | 09 | 5.0, 7.5, 10.3 (m) 2.0, 3.5, 2.5, 1.8, 4.0 (M) 0.8, 0.5, 0.01, 0.2, <0.01, <0.01, 0.4, 0.2, <0.01 (S) | 6.5, 10, 11 (m) 5, 7, 7, 9, 7 (M) 1, 2, 2, 3, 2, 2, 2, 1, 2 (S) | 3 (m) 1 (M) 2 (S) | - 4 (M) 7 (S) | Pathogenic |
Missense | c.580A>G | 03 | 00 | 00 | 10.7, 9.0, 9.0 (m) | 10, 8, 8 (m) | - | 3 (m) | Likely benign |
Missense | c.676C>T | 00 | 01 | 05 | 4.0 (M) 0.5, 0.1, <0.01, 0.01, <0.01 (S) | 5.5 (M) 3, 1, 2, 2, 1 (S) | 1 (M) 1 (S) | - 3 (S) | Pathogenic |
Missense | c.1070G>A | 01 | 03 | 04 | 6.5 (m) 2.0, 3.8, 2.5 (M) 0.01, 0.1, <0.01, <0.01 (S) | 12 (m) 6, 7, 7 (M) 2, 2, 2, 1 (S) | - 2 (M) 2 (S) | 1 (m) 1 (M) 1 (S) | Pathogenic |
Stop-gain | c.1094C>A | 00 | 03 | 12 | 2.6, 3.5, 1.8 (M) 0.01, 0.01, 0.1, 0.2, <0.01, <0.01, 0.1, 0.01, <0.01, 0.01, <0.01, <0.01 (S) | 7, 7, 6 (M) 1, 2, 2, 2, 2, 3, 1.5, 3, 2, 2, 2.5, 1 (S) | - 10 (S) | 3 (M) 2 (S) | Pathogenic |
Missense | c.1187G>A | 00 | 00 | 01 | 0.6 (S) | 2 (S) | - | 1 (S) | Benign |
Missense | c.1232G>A | 00 | 00 | 03 | 0.6, 0.3, 0.1 (S) | 2, 1, 1.5 (S) | 1 (S) | 2 (S) | Pathogenic |
Synonymous | c.1385A>G | 00 | 01 | 00 | 4.0 (M) | 10 (M) | 1 (M) | - | Benign |
3’ UTR | c.*1110A>G | 01 | 03 | 00 | 12.0 (m) 5.0, 4.0, 3.8 (M) | 13 (m) 7, 7.5, 9 (M) | 1 (m) 1 (M) | - 2 (M) | - |
Abbreviations: m, mild; M, moderate; S, severe.
The novel missense mutation p.E66D was shown to be harmful to FIX protein function by PROVEAN, SNAP2, PholyPhen2, SNP&GO, PHD-SNP, and PANTHER (Table 4). Multiple sequence alignment of this novel missense mutation indicated that it was present in the highly conserved residue of the FIX protein (Fig. 2). Homology modeling of the protein structure was conducted using the Swiss model server, which was then visualized and analyzed using chimera program. In p.E66D, the mutant residue is smaller than wild type residue, which might lead to the loss of interactions with the metal ion: “calcium 4 or magnesium 1; via 4-carboxyglutamate” (Fig. 3). The novel stop-gain mutation, p.S365*, results in a premature stop codon that leads to a truncated FIX variant of 365 amino acids, corresponding to a loss of approximately 21% of the wild type FIX protein (Fig. 3).
Table 4 Pathogenicity predictions of the p.E66D mutation.
Mutation | PROVEANa) | SNAP2b) | PolyPhen2c) | PHD-SNPd) | SNP&GOe) | PANTHER |
---|---|---|---|---|---|---|
E66D | Deleterious Score: -2.540 | Effect Score: 37 | Probably damaging Score: 0.999 | Disease Score: 4 | Disease Probability: 0.705 | Probably damaging |
a)PROVEAN: “Deleterious” if the prediction score was ≤2.5, “Neutral” if the prediction score was ≥2.5. b)NAP2: “Neutral” if the score ranged from 0 to -100. “Effect” if the score was between 0 and 100. c)PolyPhen2: “Probably damaging” is the most disease-causing ability, with a score near 1. “Possibly damaging” signifies less disease-causing ability with a score of 0.5–0.8. “Benign”, which does not alter protein function, with a score closer to 0. d)PHD-SNP: if the probability is >0.5, mutation is predicted as “Disease” and if less than <0.5, mutation is predicted to be “Neutral”. e)SNP & GO: Probability of >0.5 is predicted to be a disease-causing nsSNP.
HB is a bleeding disorder that causes abnormal or poor blood clotting. Pathogenic variants of the
In the present study, we recorded a total of 16 mutations, of which 15 (93.75%) were coding sequence variants. The majority of them (81.25%) were missense variants.
We recorded the stop-gain mutation p.R75W in 2 patients with severe HB, which have been described previously by Parrado Jara
The novel missense mutation p.E66D was recorded in 4 patients with moderate HB, with a mean FIX concentration of 2.0±0.5, and in 7 patients with severe HB, with a mean FIX concentration of 0.3±0.5. It was found to be harmful to the function of the FIX protein by
Of the 150 HB patients, 90 (60%) had mutations in the
Additionally, 2 stop-gain mutations and 7 missense mutations were major pathogenic variations or disease-causing mutations recorded in our study; patients who were carriers of these mutations showed more severe conditions compared to the other 2 groups (mild and moderate). Likely pathogenic and benign mutations were also missense mutations, and all the patients carrying these mutations were mildly to moderately affected by HB, except for 2 patients. Only 3 mutations (c.127C>T, c.470G>A, and c.1070G>A) were associated with different severities. Moreover, 2 mutations were only associated with mild HB (c.304C>T and c.580A>G), 2 mutations were associated with moderate HB (c.195G>A and c.1385A>G), and 3 mutations were only associated with severe HB (c.223C>T, c.1187G>A, and c.1232G>A). One mutation each was associated with mild-moderate (c.*1110A>G) and mild-severe HB (c.197A>T). In addition, 4 mutations were associated with moderate-severe HB (c.314A>G, c.198A>T, c.676C>T, and c.1094C>A). Among the patients with severe HB, 48% (49/102) featured mutations in the
In our study, we also observed clinical differences between the mutated and wild type
Our study strongly suggests that the majority of HB cases feature pathogenic single nucleotide variations, which may be novel or previously recorded. In many cases, novel single nucleotide variants are involved. Population-based screening of mutations will help establish inhibitor risk prediction and carrier detection strategies in India.
We thank all hemophilia societies and hemophilia B patients and their families for participating in this study. We thank the Karnataka Institute for DNA Research, Dharwad, India, for providing instrumentation support for the research work. We also thank Shri B. M. Patil Medical College, Hospital and Research Centre, BLDE (Deemed to be University), Vijayapura and S. Nijaliangappa Medical College, HSK Hospital and Research Center, Bagalkot, for their continuous support throughout this research project.
No potential conflicts of interest relevant to this article were reported.
Blood Res 2021; 56(4): 252-258
Published online December 31, 2021 https://doi.org/10.5045/br.2021.2021016
Copyright © The Korean Society of Hematology.
Sujayendra Kulkarni1,2, Rajat Hegde4, Smita Hegde4, Suyamindra S. Kulkarni4, Suresh Hanagvadi5, Kusal K. Das6, Sanjeev Kolagi3, Pramod B. Gai4, Rudragouda Bulagouda1
1Human Genetics Laboratory, Department of Anatomy, Shri B.M Patil Medical College, Hospital and Research Centre, BLDE (Deemed to be University), Vijayapura, 2Division of Human Genetics (Central Research Lab), 3Department of Anatomy, S. Nijaliangappa Medical College, HSK Hospital and Research Center, Bagalkot, 4Karnataka Institute for DNA Research (KIDNAR), Dharwad, 5Department of Pathology, J. J. M. Medical College, Davangere, Karnataka, 6Laboratory of Vascular Physiology and Medicine, Department of Physiology, Shri B. M. Patil Medical College, Hospital and Research Centre, BLDE (Deemed to be University), Vijayapura, India
Correspondence to:Rudragouda Bulagouda, M.D.
Human Genetics Laboratory, Department of Anatomy, Shri B.M Patil Medical College, Hospital and Research Centre, BLDE (Deemed to be University), Vijayapura 586101, India
E-mail: drravisb2012@gmail.com
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background
Hemophilia B (HB) is an X-linked bleeding disorder resulting from coagulation factor IX defects. Over 3,000 pathogenic, HB-associated mutations in the F9 gene have been identified. We aimed to investigate the role of F9 variants in 150 HB patients using sequencing technology.
Methods
F9 gene sequences were amplified from peripheral blood-derived DNA and sequenced on an Applied Biosystems (ABI) 3500 Sanger sequencing platform. Functional and structural predictions of mutant FIX were analyzed.
Results
Among 150 HB patients, 102 (68%), 30 (20%), and 18 (12%) suffered from severe, moderate, and mild HB, respectively. Genetic analysis identified 16 mutations, including 3 novel mutations. Nine mutations (7 missense and 2 stop-gain) were found to be pathogenic. Only 3 mutations (c.127C>T, c.470G>A, and c.1070G>A) were associated with different severities. While 2 mutations were associated with mild HB cases (c.304C>T and c.580A>G), 2 (c.195G>A and c.1385A>G) and 3 mutations (c.223C>T, c.1187G>A, and c.1232G>A) resulted in moderate and severe disease, respectively. Additionally, 1 mutation each was associated with mild-moderate (c.*1110A>G) and mild-severe HB disease (c.197A>T), 4 mutations were associated with moderate-severe HB cases (c.314A>G, c.198A>T, c.676C>T, and c.1094C>A). FIX concentrations were lower in the mutated group (5.5±2.5% vs. 8.0±2.5%). Novel p.E66D and p.S365 mutations were predicted to be pathogenic based on changes in FIX structure and function.
Conclusion
Novel single nucleotide polymorphisms (SNPs) largely contributed to the pathogenesis of HB. Our study strongly suggests that population-based genetic screening will be particularly helpful to identify risk prediction and carrier detection tools for Indian HB patients.
Keywords: Hemophilia B, F9 gene, Stop-gain mutation, Missense mutation, India
Hemophilia is an X-linked blood disorder that causes sustained bleeding after injuries or trauma. Symptoms mainly include bleeding in the joints and muscles. The term hemophilia was first used by Friedrich Hopff, University of Zurich, in 1828 [1, 2]. Hemophilia B (HB) is caused by defects in coagulation factor IX. It is also called “Christmas disorder”, after it was identified by Stephen Christmas in 1952 [3]. FIX activity can be classified as severely (<1%), moderately (1–5%), and mildly (5–30%) impaired [1, 4, 5]. HB is less common than hemophilia A (HA) (1 in 25,000 males vs. 1 in 5,000 males worldwide) [6, 7]. Approximately 14,000 people with hemophilia have been registered at the Haemophilia Federation of India; however, hemophilia remains under-diagnosed and many cases are thus not registered. HA occurs in 1 out of 10,000 male births, while HB occurs in 1 out of 30,000 male births [8].
The
Recently, more than 3,000 pathogenic mutations and neutral polymorphisms have been identified in the
This study was approved by the Institutional Ethical Committee of Shri B. M. Patil Medical College, Hospital and Research Centre, BLDE (Deemed to be University), Vijayapura (Ref. No.: BLDE (DU)/IEC/340/2018-19), and the SNMC Institutional Ethics Committee of Human Subject Research, Bagalkot (Ref. No.: SNMC/IECHSR/2018-19/A-B2/1.1). Written informed consent was obtained from all HB patients before blood sample collection. A total of 150 HB patients were included in this study, which were followed up at 12 different hemophilia societies across the Karnataka state of India. A detailed clinical history was obtained from all HB patients.
Peripheral blood from HB patients was collected in EDTA vacutainers (BD, Franklin Lakes, NJ, USA). Prior to DNA isolation, the FIX concentration was measured and inhibitor assay were performed. DNA was extracted from peripheral blood using a blood and tissue DNA extraction kit (QIAGEN, Hilden, Germany). All exonic regions were amplified and the products were sequenced on an Applied Biosystems (ABI) 3500 Sanger sequencing platform using the BigDye Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher Scientific, Waltham, MA, USA). Results were analyzed using DNA sequence analysis software v5.4.
The pathogenicity of the novel non-synonymous variants was analyzed using bioinformatics tools, such as PROVEAN (http://provean.jcvi.org/seq_submit.php), PolyPhen-2 (http://genetics.bwh.harvard.edu/pph2/index.shtml), PHD SNP (https://snps.biofold.org/phd-snp/phd-snp.html), SNPs & GO (https://snps.biofold.org/snps-and-go/snps-and-go.html), PANTHER (http://www.pantherdb.org/), and SNAP2 (https://www.rostlab.org/services/snap/). The conservation property of missense variants was investigated using the Clustal Omega multiple sequence alignment tool (https://www.ebi.ac.uk/Tools/msa/clustalo/). A 3D model of wild type and mutant FIX proteins was predicted using the Swiss model (https://swissmodel.expasy.org/), and the results were visualized and analyzed using the UCSF Chimera program.
Of the 150 HB patients included in this study, 102 (68%; FIX concentration, 0.6±0.2; age of onset, 2.0±1.0 y), 30 (20%; FIX concentration, 2.5±1.3; age of onset, 7.5±2.8 y), and 18 (12%; FIX concentration, 8.0±2.6; age of onset, 10.0±3.5y) suffered from severe, moderate, and mild HB, respectively. The detailed clinicopathological parameters are summarized in Table 1. In our study, we recorded 16 mutations. Of those, 1 was a synonymous mutation, 12 were missense mutations, 2 were stop-gain mutations, and 1 was a 3’ UTR variant. Notably, 13 (81.25%) mutations were previously reported, but 3 (18.75%) were novel mutations, which had not been entered into any of the human SNP databases. The majority of the mutations were found in exon 8 of the gene and largely comprised missense mutations. Exon 8 showed a high number of mutations compared to other exons of the
Table 1 . Clinicopathological characteristics of the study population..
Clinicopathological parameters | Disease severity | ||||
---|---|---|---|---|---|
Severe | Moderate | Mild | |||
(Factor IX <1%), N=102 (68%) | (Factor IX 1–5%), N=30 (20%) | (Factor IX >5%), N=18 (12%) | |||
Mean factor IX levels (%) | 0.6±0.2 | 2.5±1.3 | 8.0±2.6 | ||
Family history of bleeding (N) | 80 | 10 | 7 | ||
Mean age of onset (yr) | 2.0±1.0 | 7.5±2.8 | 10.0±3.5 | ||
Mutation frequency (within group) | 48% (49/102) | 86.7% (26/30) | 83.3% (15/18) | ||
Inhibitor-positive (N) | 23 | 11 | 4 | ||
Inhibitor-negative (N) | 79 | 19 | 14 | ||
Hemoglobin concentration | Normal | Normal | Normal | ||
Platelet count | Normal | Normal | Normal | ||
Prothrombin time | Normal | Normal | Normal | ||
Activated partial thromboplastin time | High | High | High | ||
Bleeding sites | |||||
Joints | 55 (54%) | 19 (63.3%) | 12 (66.7%) | ||
Gum | 17 (16.7%) | 6 (20%) | 4 (22.2%) | ||
Muscle | 13 (12.7%) | 1 (3.3%) | 2 (11.1%) | ||
Epistaxis | 7 (6.9%) | 4 (13.3%) | 0 | ||
Petechiae | 10 (9.8%) | 0 | 0 |
Table 2 . List of mutations recorded in the
Mutation type | Nucleotide change | cDNA position | Amino acid change | Exon | Status | Frequency, N (%) |
---|---|---|---|---|---|---|
Missense | g.11313C>T | c.127C>T | p.R43W | 2 | Reported (rs1603264205) | 9 (6.0%) |
Missense | g.11381G>A | c.195G>A | p.M65I | 2 | Reported (rs763568424) | 1 (0.66%) |
Missense | g.11383A>T | c.197A>T | p.E66V | 2 | Reported (CM940423) | 3 (2.0%) |
Missense | g.11384A>T | c.198A>T | p.E66D | 2 | Not reported | 12 (8.0%) |
Stop-gain | g.11409 C>T | c.223C>T | p.R75a) | 2 | Reported (rs137852227) | 2 (1.33%) |
Missense | g.15369 T>C | c.304C>T | p.C102R | 4 | Reported (CM960574) | 2 (1.33%) |
Missense | g.22664 T>C | c.314A>G | p.G143R | 5 | Reported (CM940499) | 4 (2.66%) |
Missense | g.22706 G>A | c.470G>A | p.C157Y | 5 | Reported (rs1367198680) | 17 (11.33%) |
Missense | g.25386 A>G | c.580A>G | p.T194A | 6 | Reported (rs6048) | 3 (2.0%) |
Missense | g.25482C>T | c.676C>T | p.R226W | 6 | Reported (rs137852240) | 6 (4.0%) |
Missense | g.36020 G>A | c.1070G>A | p.G357E | 8 | Reported (rs137852275) | 8 (5.33%) |
Stop-gain | g.36044C>A | c.1094C>A | p.S365a) | 8 | Not reported | 15 (10.0%) |
Missense | g.36137G>A | c.1187G>A | p.C396Y | 8 | Reported (rs137852273) | 1 (0.66%) |
Missense | g.36182 G>A | c.1232G>A | p.S411N | 8 | Reported (rs137852276) | 3 (2.0%) |
Synonymous | g.36335 A>G | c.1385A>G | p.Ter462= | 8 | Reported (rs561793582) | 1 (0.66%) |
3’ UTR | g.37446A>G | c.*1110A>G | ……… | 8 | Not reported | 3 (2.0%) |
a)NG_007994.1, NM_000133.4, and NP_000124.1 reference sequences wereused for genomic DNA variant nomenclature, coding region variant nomenclature, and protein variant nomenclature, respectively. All nomenclatures were made according to Human Genome Variation Society (HGVS) guidelines..
The novel missense mutation c.198A>T was found in the
Table 3 . Genotypic and phenotypic associations of the mutations recorded in the study population..
Mutation type | cDNA position | Patients | Factor IX (IX) concentration level (%) | Age of onset | Inhibitor-positive (N) | Inhibitor-negative (N) | Clinical significance | ||
---|---|---|---|---|---|---|---|---|---|
m | M | S | |||||||
Missense | c.127C>T | 03 | 04 | 02 | 6.5, 8.0, 10 (m) 2.0, 2.3. 1.8, 2.0 (M) 0.8, 0.5 (S) | 10, 12, 8 (m) 7, 6, 9, 9 (M) 1, 1.5 (S) | 2 (m) 2 (M) - | 1 (m) 2 (M) 2 (S) | Pathogenic |
Missense | c.195G>A | 00 | 01 | 00 | 3.0 (M) | 7.5 (M) | 1 (M) | - | Likely pathogenic |
Missense | c.197A>T | 02 | 00 | 01 | 6.0, 9.0 (m) 0.3 (S) | 13, 11 (m) 1 (S) | - | 2 (m) 1 (S) | Likely benign |
Missense | c.198A>T | 00 | 04 | 07 | 1.5, 2.0. 2.5, 2.0 (M) 0.8, 0.5, 0.1, 0.6, 0.6, <0.01, 0.01 (S) | 6, 7, 7, 8 (M) 1, 2, 1, 1.5, 2, 2, 1.6 (S) | 4 (M) 4 (S) | - 3 (S) | Pathogenic |
Stop-gain | c.223C>T | 00 | 00 | 02 | <0.01, <0.01 (S) | 1, 1 (S) | 1 (S) | 1 (S) | Pathogenic |
Missense | c.304C>T | 02 | 00 | 00 | 12.0, 10.0 (m) | 13, 10 (m) | 2 (m) | - | Likely pathogenic |
Missense | c.314A>G | 00 | 01 | 03 | 2.0 (M) 0.2, 0.1, 0.01 (S) | 8 (M) 1, 2, 1 (S) | - | 1 (M) 3 (S) | Pathogenic |
Missense | c.470G>A | 03 | 05 | 09 | 5.0, 7.5, 10.3 (m) 2.0, 3.5, 2.5, 1.8, 4.0 (M) 0.8, 0.5, 0.01, 0.2, <0.01, <0.01, 0.4, 0.2, <0.01 (S) | 6.5, 10, 11 (m) 5, 7, 7, 9, 7 (M) 1, 2, 2, 3, 2, 2, 2, 1, 2 (S) | 3 (m) 1 (M) 2 (S) | - 4 (M) 7 (S) | Pathogenic |
Missense | c.580A>G | 03 | 00 | 00 | 10.7, 9.0, 9.0 (m) | 10, 8, 8 (m) | - | 3 (m) | Likely benign |
Missense | c.676C>T | 00 | 01 | 05 | 4.0 (M) 0.5, 0.1, <0.01, 0.01, <0.01 (S) | 5.5 (M) 3, 1, 2, 2, 1 (S) | 1 (M) 1 (S) | - 3 (S) | Pathogenic |
Missense | c.1070G>A | 01 | 03 | 04 | 6.5 (m) 2.0, 3.8, 2.5 (M) 0.01, 0.1, <0.01, <0.01 (S) | 12 (m) 6, 7, 7 (M) 2, 2, 2, 1 (S) | - 2 (M) 2 (S) | 1 (m) 1 (M) 1 (S) | Pathogenic |
Stop-gain | c.1094C>A | 00 | 03 | 12 | 2.6, 3.5, 1.8 (M) 0.01, 0.01, 0.1, 0.2, <0.01, <0.01, 0.1, 0.01, <0.01, 0.01, <0.01, <0.01 (S) | 7, 7, 6 (M) 1, 2, 2, 2, 2, 3, 1.5, 3, 2, 2, 2.5, 1 (S) | - 10 (S) | 3 (M) 2 (S) | Pathogenic |
Missense | c.1187G>A | 00 | 00 | 01 | 0.6 (S) | 2 (S) | - | 1 (S) | Benign |
Missense | c.1232G>A | 00 | 00 | 03 | 0.6, 0.3, 0.1 (S) | 2, 1, 1.5 (S) | 1 (S) | 2 (S) | Pathogenic |
Synonymous | c.1385A>G | 00 | 01 | 00 | 4.0 (M) | 10 (M) | 1 (M) | - | Benign |
3’ UTR | c.*1110A>G | 01 | 03 | 00 | 12.0 (m) 5.0, 4.0, 3.8 (M) | 13 (m) 7, 7.5, 9 (M) | 1 (m) 1 (M) | - 2 (M) | - |
Abbreviations: m, mild; M, moderate; S, severe..
The novel missense mutation p.E66D was shown to be harmful to FIX protein function by PROVEAN, SNAP2, PholyPhen2, SNP&GO, PHD-SNP, and PANTHER (Table 4). Multiple sequence alignment of this novel missense mutation indicated that it was present in the highly conserved residue of the FIX protein (Fig. 2). Homology modeling of the protein structure was conducted using the Swiss model server, which was then visualized and analyzed using chimera program. In p.E66D, the mutant residue is smaller than wild type residue, which might lead to the loss of interactions with the metal ion: “calcium 4 or magnesium 1; via 4-carboxyglutamate” (Fig. 3). The novel stop-gain mutation, p.S365*, results in a premature stop codon that leads to a truncated FIX variant of 365 amino acids, corresponding to a loss of approximately 21% of the wild type FIX protein (Fig. 3).
Table 4 . Pathogenicity predictions of the p.E66D mutation..
Mutation | PROVEANa) | SNAP2b) | PolyPhen2c) | PHD-SNPd) | SNP&GOe) | PANTHER |
---|---|---|---|---|---|---|
E66D | Deleterious Score: -2.540 | Effect Score: 37 | Probably damaging Score: 0.999 | Disease Score: 4 | Disease Probability: 0.705 | Probably damaging |
a)PROVEAN: “Deleterious” if the prediction score was ≤2.5, “Neutral” if the prediction score was ≥2.5. b)NAP2: “Neutral” if the score ranged from 0 to -100. “Effect” if the score was between 0 and 100. c)PolyPhen2: “Probably damaging” is the most disease-causing ability, with a score near 1. “Possibly damaging” signifies less disease-causing ability with a score of 0.5–0.8. “Benign”, which does not alter protein function, with a score closer to 0. d)PHD-SNP: if the probability is >0.5, mutation is predicted as “Disease” and if less than <0.5, mutation is predicted to be “Neutral”. e)SNP & GO: Probability of >0.5 is predicted to be a disease-causing nsSNP..
HB is a bleeding disorder that causes abnormal or poor blood clotting. Pathogenic variants of the
In the present study, we recorded a total of 16 mutations, of which 15 (93.75%) were coding sequence variants. The majority of them (81.25%) were missense variants.
We recorded the stop-gain mutation p.R75W in 2 patients with severe HB, which have been described previously by Parrado Jara
The novel missense mutation p.E66D was recorded in 4 patients with moderate HB, with a mean FIX concentration of 2.0±0.5, and in 7 patients with severe HB, with a mean FIX concentration of 0.3±0.5. It was found to be harmful to the function of the FIX protein by
Of the 150 HB patients, 90 (60%) had mutations in the
Additionally, 2 stop-gain mutations and 7 missense mutations were major pathogenic variations or disease-causing mutations recorded in our study; patients who were carriers of these mutations showed more severe conditions compared to the other 2 groups (mild and moderate). Likely pathogenic and benign mutations were also missense mutations, and all the patients carrying these mutations were mildly to moderately affected by HB, except for 2 patients. Only 3 mutations (c.127C>T, c.470G>A, and c.1070G>A) were associated with different severities. Moreover, 2 mutations were only associated with mild HB (c.304C>T and c.580A>G), 2 mutations were associated with moderate HB (c.195G>A and c.1385A>G), and 3 mutations were only associated with severe HB (c.223C>T, c.1187G>A, and c.1232G>A). One mutation each was associated with mild-moderate (c.*1110A>G) and mild-severe HB (c.197A>T). In addition, 4 mutations were associated with moderate-severe HB (c.314A>G, c.198A>T, c.676C>T, and c.1094C>A). Among the patients with severe HB, 48% (49/102) featured mutations in the
In our study, we also observed clinical differences between the mutated and wild type
Our study strongly suggests that the majority of HB cases feature pathogenic single nucleotide variations, which may be novel or previously recorded. In many cases, novel single nucleotide variants are involved. Population-based screening of mutations will help establish inhibitor risk prediction and carrier detection strategies in India.
We thank all hemophilia societies and hemophilia B patients and their families for participating in this study. We thank the Karnataka Institute for DNA Research, Dharwad, India, for providing instrumentation support for the research work. We also thank Shri B. M. Patil Medical College, Hospital and Research Centre, BLDE (Deemed to be University), Vijayapura and S. Nijaliangappa Medical College, HSK Hospital and Research Center, Bagalkot, for their continuous support throughout this research project.
No potential conflicts of interest relevant to this article were reported.
Table 1 . Clinicopathological characteristics of the study population..
Clinicopathological parameters | Disease severity | ||||
---|---|---|---|---|---|
Severe | Moderate | Mild | |||
(Factor IX <1%), N=102 (68%) | (Factor IX 1–5%), N=30 (20%) | (Factor IX >5%), N=18 (12%) | |||
Mean factor IX levels (%) | 0.6±0.2 | 2.5±1.3 | 8.0±2.6 | ||
Family history of bleeding (N) | 80 | 10 | 7 | ||
Mean age of onset (yr) | 2.0±1.0 | 7.5±2.8 | 10.0±3.5 | ||
Mutation frequency (within group) | 48% (49/102) | 86.7% (26/30) | 83.3% (15/18) | ||
Inhibitor-positive (N) | 23 | 11 | 4 | ||
Inhibitor-negative (N) | 79 | 19 | 14 | ||
Hemoglobin concentration | Normal | Normal | Normal | ||
Platelet count | Normal | Normal | Normal | ||
Prothrombin time | Normal | Normal | Normal | ||
Activated partial thromboplastin time | High | High | High | ||
Bleeding sites | |||||
Joints | 55 (54%) | 19 (63.3%) | 12 (66.7%) | ||
Gum | 17 (16.7%) | 6 (20%) | 4 (22.2%) | ||
Muscle | 13 (12.7%) | 1 (3.3%) | 2 (11.1%) | ||
Epistaxis | 7 (6.9%) | 4 (13.3%) | 0 | ||
Petechiae | 10 (9.8%) | 0 | 0 |
Table 2 . List of mutations recorded in the
Mutation type | Nucleotide change | cDNA position | Amino acid change | Exon | Status | Frequency, N (%) |
---|---|---|---|---|---|---|
Missense | g.11313C>T | c.127C>T | p.R43W | 2 | Reported (rs1603264205) | 9 (6.0%) |
Missense | g.11381G>A | c.195G>A | p.M65I | 2 | Reported (rs763568424) | 1 (0.66%) |
Missense | g.11383A>T | c.197A>T | p.E66V | 2 | Reported (CM940423) | 3 (2.0%) |
Missense | g.11384A>T | c.198A>T | p.E66D | 2 | Not reported | 12 (8.0%) |
Stop-gain | g.11409 C>T | c.223C>T | p.R75a) | 2 | Reported (rs137852227) | 2 (1.33%) |
Missense | g.15369 T>C | c.304C>T | p.C102R | 4 | Reported (CM960574) | 2 (1.33%) |
Missense | g.22664 T>C | c.314A>G | p.G143R | 5 | Reported (CM940499) | 4 (2.66%) |
Missense | g.22706 G>A | c.470G>A | p.C157Y | 5 | Reported (rs1367198680) | 17 (11.33%) |
Missense | g.25386 A>G | c.580A>G | p.T194A | 6 | Reported (rs6048) | 3 (2.0%) |
Missense | g.25482C>T | c.676C>T | p.R226W | 6 | Reported (rs137852240) | 6 (4.0%) |
Missense | g.36020 G>A | c.1070G>A | p.G357E | 8 | Reported (rs137852275) | 8 (5.33%) |
Stop-gain | g.36044C>A | c.1094C>A | p.S365a) | 8 | Not reported | 15 (10.0%) |
Missense | g.36137G>A | c.1187G>A | p.C396Y | 8 | Reported (rs137852273) | 1 (0.66%) |
Missense | g.36182 G>A | c.1232G>A | p.S411N | 8 | Reported (rs137852276) | 3 (2.0%) |
Synonymous | g.36335 A>G | c.1385A>G | p.Ter462= | 8 | Reported (rs561793582) | 1 (0.66%) |
3’ UTR | g.37446A>G | c.*1110A>G | ……… | 8 | Not reported | 3 (2.0%) |
a)NG_007994.1, NM_000133.4, and NP_000124.1 reference sequences wereused for genomic DNA variant nomenclature, coding region variant nomenclature, and protein variant nomenclature, respectively. All nomenclatures were made according to Human Genome Variation Society (HGVS) guidelines..
Table 3 . Genotypic and phenotypic associations of the mutations recorded in the study population..
Mutation type | cDNA position | Patients | Factor IX (IX) concentration level (%) | Age of onset | Inhibitor-positive (N) | Inhibitor-negative (N) | Clinical significance | ||
---|---|---|---|---|---|---|---|---|---|
m | M | S | |||||||
Missense | c.127C>T | 03 | 04 | 02 | 6.5, 8.0, 10 (m) 2.0, 2.3. 1.8, 2.0 (M) 0.8, 0.5 (S) | 10, 12, 8 (m) 7, 6, 9, 9 (M) 1, 1.5 (S) | 2 (m) 2 (M) - | 1 (m) 2 (M) 2 (S) | Pathogenic |
Missense | c.195G>A | 00 | 01 | 00 | 3.0 (M) | 7.5 (M) | 1 (M) | - | Likely pathogenic |
Missense | c.197A>T | 02 | 00 | 01 | 6.0, 9.0 (m) 0.3 (S) | 13, 11 (m) 1 (S) | - | 2 (m) 1 (S) | Likely benign |
Missense | c.198A>T | 00 | 04 | 07 | 1.5, 2.0. 2.5, 2.0 (M) 0.8, 0.5, 0.1, 0.6, 0.6, <0.01, 0.01 (S) | 6, 7, 7, 8 (M) 1, 2, 1, 1.5, 2, 2, 1.6 (S) | 4 (M) 4 (S) | - 3 (S) | Pathogenic |
Stop-gain | c.223C>T | 00 | 00 | 02 | <0.01, <0.01 (S) | 1, 1 (S) | 1 (S) | 1 (S) | Pathogenic |
Missense | c.304C>T | 02 | 00 | 00 | 12.0, 10.0 (m) | 13, 10 (m) | 2 (m) | - | Likely pathogenic |
Missense | c.314A>G | 00 | 01 | 03 | 2.0 (M) 0.2, 0.1, 0.01 (S) | 8 (M) 1, 2, 1 (S) | - | 1 (M) 3 (S) | Pathogenic |
Missense | c.470G>A | 03 | 05 | 09 | 5.0, 7.5, 10.3 (m) 2.0, 3.5, 2.5, 1.8, 4.0 (M) 0.8, 0.5, 0.01, 0.2, <0.01, <0.01, 0.4, 0.2, <0.01 (S) | 6.5, 10, 11 (m) 5, 7, 7, 9, 7 (M) 1, 2, 2, 3, 2, 2, 2, 1, 2 (S) | 3 (m) 1 (M) 2 (S) | - 4 (M) 7 (S) | Pathogenic |
Missense | c.580A>G | 03 | 00 | 00 | 10.7, 9.0, 9.0 (m) | 10, 8, 8 (m) | - | 3 (m) | Likely benign |
Missense | c.676C>T | 00 | 01 | 05 | 4.0 (M) 0.5, 0.1, <0.01, 0.01, <0.01 (S) | 5.5 (M) 3, 1, 2, 2, 1 (S) | 1 (M) 1 (S) | - 3 (S) | Pathogenic |
Missense | c.1070G>A | 01 | 03 | 04 | 6.5 (m) 2.0, 3.8, 2.5 (M) 0.01, 0.1, <0.01, <0.01 (S) | 12 (m) 6, 7, 7 (M) 2, 2, 2, 1 (S) | - 2 (M) 2 (S) | 1 (m) 1 (M) 1 (S) | Pathogenic |
Stop-gain | c.1094C>A | 00 | 03 | 12 | 2.6, 3.5, 1.8 (M) 0.01, 0.01, 0.1, 0.2, <0.01, <0.01, 0.1, 0.01, <0.01, 0.01, <0.01, <0.01 (S) | 7, 7, 6 (M) 1, 2, 2, 2, 2, 3, 1.5, 3, 2, 2, 2.5, 1 (S) | - 10 (S) | 3 (M) 2 (S) | Pathogenic |
Missense | c.1187G>A | 00 | 00 | 01 | 0.6 (S) | 2 (S) | - | 1 (S) | Benign |
Missense | c.1232G>A | 00 | 00 | 03 | 0.6, 0.3, 0.1 (S) | 2, 1, 1.5 (S) | 1 (S) | 2 (S) | Pathogenic |
Synonymous | c.1385A>G | 00 | 01 | 00 | 4.0 (M) | 10 (M) | 1 (M) | - | Benign |
3’ UTR | c.*1110A>G | 01 | 03 | 00 | 12.0 (m) 5.0, 4.0, 3.8 (M) | 13 (m) 7, 7.5, 9 (M) | 1 (m) 1 (M) | - 2 (M) | - |
Abbreviations: m, mild; M, moderate; S, severe..
Table 4 . Pathogenicity predictions of the p.E66D mutation..
Mutation | PROVEANa) | SNAP2b) | PolyPhen2c) | PHD-SNPd) | SNP&GOe) | PANTHER |
---|---|---|---|---|---|---|
E66D | Deleterious Score: -2.540 | Effect Score: 37 | Probably damaging Score: 0.999 | Disease Score: 4 | Disease Probability: 0.705 | Probably damaging |
a)PROVEAN: “Deleterious” if the prediction score was ≤2.5, “Neutral” if the prediction score was ≥2.5. b)NAP2: “Neutral” if the score ranged from 0 to -100. “Effect” if the score was between 0 and 100. c)PolyPhen2: “Probably damaging” is the most disease-causing ability, with a score near 1. “Possibly damaging” signifies less disease-causing ability with a score of 0.5–0.8. “Benign”, which does not alter protein function, with a score closer to 0. d)PHD-SNP: if the probability is >0.5, mutation is predicted as “Disease” and if less than <0.5, mutation is predicted to be “Neutral”. e)SNP & GO: Probability of >0.5 is predicted to be a disease-causing nsSNP..
Eun Jin Choi, Tai ju Hwang, Yong Mook Choi, Hugh Chul Kim, Myung Chul Yoo, Haylee Song, Kayode Badejo
Blood Res 2020; 55(4): 246-252Ki-Young Yoo, Soo-Young Jung, Sung-Ho Hwang, Su-Min Lee, Jong-Ho Park, and Hyun-Ja Nam
Blood Res 2018; 53(1): 41-48Sj A Shin, Young Dai Kim, Hahng Lee
Korean J Hematol 2003; 38(3): 188-190