- Email: [email protected]
Rapid genotyping of two common G6PD variants, African (A-) and Mediterranean, by high-resolution melting analysis
Available online at www.sciencedirect.com
Clinical Biochemistry 43 (2010) 193 – 197
Rapid genotyping of two common G6PD variants, African (A-) and Mediterranean, by high-resolution melting analysis Philippe Joly a,b , Philippe Lacan a , Caroline Garcia a , Cyril Martin b , Alain Francina a,⁎ a
Unité de Pathologie Moléculaire du Globule Rouge, Fédération de Biochimie et de Biologie Spécialisée, Hôpital Edouard Herriot, Hospices Civils and Université Claude Bernard-Lyon 1, Lyon, France b Centre de Recherche et d'Innovation sur le Sport (EA 647), Université Claude Bernard Lyon I, Lyon, France Received 19 May 2009; received in revised form 12 August 2009; accepted 10 September 2009 Available online 24 September 2009
Abstract Objectives: The Mediterranean and A(-) G6PD variants are particularly prevalent in Africa and Southern Europe. Our study was aimed to develop an assay for the rapid genotyping of these two variants by HRM. Methods: After PCR reactions corresponding to the G6PD Mediterranean (exon 6), G6PD (A-) (exon 4) and G6PD (A-) (exon 5) mutations, amplicons were submitted to HRM. This protocol was applied to a cohort of 132 patients suffering from sickle cell disease. Results: Wild, homozygous or hemizygous and heterozygous states were fully discriminated by HRM for all three mutations. HRM results were in total accordance with DNA sequencing for 22 patients of our cohort with a ‘A’ genotype: presence of the (A-) (exon 5) mutation but absence of the (A-) (exon 4) mutation. Conclusions: Our HRM protocols allow a rapid, simple and cost-effective screening of G6PD deficiency in patients originating from the Mediterranean and the African areas. © 2009 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. Keywords: G6PD genotyping; HRM; Mediterranean mutation; African (A-) mutation
Introduction Glucose-6-phosphate dehydrogenase (G6PD) deficiency is the most common enzymatic defect in the world [1–4]. As it is an X-linked genetic disorder, red blood cells (RBC) enzymatic activities are significantly decreased in hemizygous males and homozygous females for most of G6PD variants. In such cases, the main clinical presentations can be a neonatal jaundice, an acute hemolytic anemia or a chronic non-spherocytic hemolytic anemia depending on the type of G6PD variant (Classes I to IV of the WHO classification) [1–4]. For heterozygous females, the clinical manifestations are less frequent because RBC enzymatic activities can be highly variable due to a possible bias in the X chromosome inactivation [4]. More than 160 mutations or combinations of mutations in the G6PD gene have been ⁎ Corresponding author. Unité de Pathologie Moléculaire du Globule Rouge, Fédération de Biochimie et de Biologie Spécialisée, Hôpital Edouard Herriot, 5, Place d'Arsonval, 69437 Lyon Cedex 03, France. Fax: +33 4 72 11 05 98. E-mail address: [email protected] (A. Francina).
characterized at the DNA level (available from: The Human Gene Mutation Database at the Institute of Medical Genetics in Cardiff (UK); http://www.hgmd.cf.ac.uk/ac/index.php, July 31, 2009). Nevertheless, the direct sequencing of the entire G6PD gene (13 exons) is time-consuming and quite expensive, particularly to perform mass screening studies. In our clinical experience, as well as in litterature data [1–4], two G6PD variants, Mediterranean and (A-), are particularly prevalent in Africa and in Southern Europe. The Mediterranean variant results from a single amino-acid substitution Ser188Phe on exon 6 (563 C→T in cDNA; rs5030868)[1–5]. The G6PD African (A-) variant results from two amino acid substitutions. The first one is the Asn126Asp substitution on exon 5 (376 A→G in cDNA; rs1050829) which has no direct consequence by itself on G6PD activity [5]. The resulting G6PD variant is called (A). But, in association with another mutation, it can give rise to ‘abnormal’ (A-) variants with lower G6PD activity. The most frequent one is the Val68Met on exon 4 (202 G→A in cDNA; rs1050828) while the two others (Arg227Leu and Leu323Pro) are much less frequent [1–5].
0009-9120/$ - see front matter © 2009 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.clinbiochem.2009.09.012
194
P. Joly et al. / Clinical Biochemistry 43 (2010) 193–197
The present study was aimed to develop an assay for the rapid genotyping by high-resolution melting (HRM) of these two common G6PD variants. This technique has been further applied for validation to a large cohort of patients suffering from sickle cell disease. Materials and methods Patient samples and DNA isolation The DNA samples for the preliminary tests were obtained from patients with a (A-) or a Mediterranean G6PD variant, previously characterized by direct sequencing (data not shown) after written consent for genetic analysis. Genomic DNA was extracted from 200 μL of whole EDTA blood using the DNA Qiacube extractor (Qiagen, Courtaboeuf, France). The final elution was made in Tris-EDTA buffer before storage at -20 °C. After the development assays, our technique was evaluated on a cohort of 132 patients suffering from sickle cell disease (SCD) and originating from Africa or French West Indies. This validation study was a part of a larger study conducted on this cohort for which a written consent for genetic analysis had been previously obtained. Selection of primers All primers (GenBank Accession Number × 55448) used for HRM genotyping were designed with the Primer 3 software. For the G6PD (A-) (exon 4) mutation, PCR primers were 5′CTTCTGCCCGAAAACACCTT-3′ (forward- at coordinates: 13479–13498) and 5′-ACACCACCCACCTTGAAGAAG-3′ (reverse- at coordinates: 13559–13579). The PCR product was 101-bp length. For the G6PD (A-) (exon 5) mutation, PCR primers were 5′-TCTTTGCCCGCAACTCCTAT-3′ (forwardat coordinates: 14152–14171) and 5′-GCAAGGCCAGGTAGAAGAGG-3′ (reverse- at coordinates: 14258–14277). The PCR product was 126-bp length. For the G6PD Mediterranean (exon 6) mutation, PCR primers were: 5′-TCTGACCGGCTGTCCAAC-3′ (forward- at coordinates: 15059–15076) and 5′GTCGATGCGGTAGATCTGGT-3′ (reverse- at coordinates: 15102–15121). The PCR product was 63 bp length. PCR and high-resolution melting Two distinct steps were performed: (i) a specific PCR reaction with a standard thermocycler (with the double strand DNA dye included in the Master Mix); (ii) a HRM analysis on the resulting amplicon with the Rotor Gene Instrument (RotorGene 6000, Qiagen, Courtaboeuf, France). The search for the G6PD (A-)(exon 5) and for the G6PD Mediterranean mutations was carried out simultaneously by duplex PCR, while the G6PD (A-)(exon 4) mutation was studied independently. For the duplex PCR reaction, each tube contained, in a final volume of 30 μL, 50 ng DNA template, 1 × PCR buffer (QBiogen, Illkirch, France), 10 mM of each dNTP, 100 ng of each primer, 1.5 U of Taq DNA polymerase (Q-Biogen), 2.5 mM of Syto®9 (Invitrogen, Cergy-Pontoise, France) and 1.5 mM
MgCl2. PCR conditions were as follows: 10 min denaturation at 95 °C; 35 cycles with denaturation 30 s at 95 °C, hybridization 30 s at 59 °C and elongation 1 min at 72 °C and final elongation 5 min at 72 °C. The HRM protocol was performed with 15 μL of PCR product: 10 min at 95 °C; 1 min at 50 °C (to facilitate the formation of heteroduplexes) and a shift from 70 to 90 °C (0.1 °C/s) with a continuous fluorescence acquisition (green channel). For the G6PD (A-)(exon 4) PCR reaction, the PCR composition was exactly the same. PCR conditions were also identical except for the hybridization temperature (58 °C). The HRM protocol was performed with 10 μL of PCR product: 10 min at 95 °C; 1 min at 50 °C and a shift from 75 to 85 °C (0.1 °C/s) with a continuous fluorescence acquisition (green channel). Results Duplex PCR for G6PD (A-)(exon 5–376 A→G)+ G6PD Mediterranean (exon 6–563 C→T) The duplex PCR gives three products. Two of them are the expected amplicons: 63 pb for the G6PD Mediterranean and 126 pb for the G6PD (A-)(exon 5). The third PCR product is 969-bp length and results from a PCR reaction using the G6PD (A-)(exon 5) forward primer and the G6PD Mediterranean reverse primer. Derivative plots (Fig. 1) show that these three amplicons have distinct denaturation intervals: approximately 76–82 °C for G6PD Mediterranean, 82–86 °C for G6PD (A-) (exon 5) and 86–92 °C for the 969-bp unexpected amplicon. As they do not overlap each other, the G6PD Mediterranean and the G6PD (A-) (exon 5) amplicons can be analyzed separately. In each case, the three possible genotypes are clearly distinguished (Figs. 2a and b): homozygous wild (or hemizygous wild for male subjects), heterozygous (only possible for females) and homozygous mutant (or hemizygous mutant for male subjects). G6PD (A-) exon 4 (202 G→A) The HRM analysis, between 75 and 85 °C, of the G6PD (A-) (exon 4) PCR product also allows to discriminate unambiguously between the three possible genotypes (Fig. 3). Cohort of SCD patients Among the 132 SCD patients of our cohort, the Mediterranean variant was found at the hemizygous state for one male and at the heterozygous state for three family-related females. The African (A-) genotype (association of the exon 4 and of the exon 5 mutations) was found at the hemizygous state for 15 males, at the homozygous state for one female and at the heterozygous state for 14 females (one was homozygous for the exon 5 mutation). We also found 22 patients without the exon 4 mutation but with the exon 5 mutation (7 hemizygous males, 14 heterozygous females and 1 homozygous female). For these patients, a direct sequencing was performed to confirm the
P. Joly et al. / Clinical Biochemistry 43 (2010) 193–197
results and to search one of the two others A(-) mutations (Arg227Leu on exon 7 and Leu323Pro on exon 9). HRM results were confirmed for all patients and no additional A(-) mutation was found on exons 7 and 9. Discussion Several molecular methods have been reported for the genotyping of G6PD variants, e.g., DNA sequencing, allelespecific oligoprobe hybridization, PCR restriction analysis, reverse dot blot assay, allele-specific PCR, denaturing HPLC or real-time PCR using FRET probes [6–10]. However, these methods are time-consuming, expensive and include many steps. A recent paper from Farez-Vidal et al. describes a multiplex multicolour assay using fluorescence detection and capillary electrophoresis for the simultaneous detection of 15 G6PD variants [11]. This method is rapid and very cost-saving compared to others but requires four fluorescently labeled ddNTPs and multiple primers pairs. Comparatively, using the present HRM protocols, we carried out a simple, rapid and very cost-effective strategy for the molecular screening of G6PD deficiency. The duplex reaction is performed in a first step and allows the detection of the Mediterranean variant and of the (A-)(exon 5) mutation. The second HRM analysis needs to be performed only on samples carrying the (A-)(exon 5) mutation. The presence of the Asn126Asp mutation on exon 5 and the absence of the Val68Met substitution on exon 4 require an additional DNA sequencing to eventually detect another ‘(A-)’ mutation. This
195
strategy is sufficient for a first screening of G6PD deficiency. A negative result must be checked by direct sequencing of the entire gene if a G6PD deficiency has been biochemically documented. The main limitation of all HRM genotyping protocols is the possibility that the single nucleotide polymorphism (SNP) of interest is replaced by another unexpected SNP. That is why we have designed amplicons as short as possible to limit the probability of such an event. Nevertheless, a few G6PD variants have been described in close proximity of the three mutations searched here (Available from: The Human Gene Mutation Database at the Institute of Medical Genetics in Cardiff (UK); http://www.hgmd.cf.ac.uk/ac/index.php, July 31, 2009) but their frequencies are negligible compared to the Mediterranean and A(-) variants. To avoid a misdiagnosis, a direct sequencing should be performed in two main situations: (i) if the HRM curve of an unknown patient is not perfectly matched with one of the DNA controls in a repeatedly way and (ii) if there is a genotype / phenotype (i.e., G6PD activity) discordance. The screening of the G6PD Mediterranean and (A-) variants seems particularly interesting in SCD patients. Indeed, Bernaudin et al have shown that a G6PD deficiency independently increases the risk of cerebral vasculopathy in SCD patients [12]. As these patients frequently receive blood transfusions, the enzymatic G6PD activity could be potentially false and a genotyping appears more appropriate. Results obtained for our 132 SCD patients cohort seem to validate the diagnostic strategy presented above as the direct genotyping of the 22 patients with ‘unexpected’ results (dissociation between
Fig. 1. Derivative plots obtained during the HRM analysis following the Multiplex PCR. The recordings of the derivative values of fluorescence vs. temperature (dF/dT) show three distinct intervals: (A) This 76–82 °C temperature interval corresponds to the entire denaturation of the ‘G6PD Med’ PCR product. (B) This 82–86 °C temperature interval corresponds to the entire denaturation of the ‘G6PD (A-) mutation (exon 5)’ PCR product. (C) This 86–90 °C temperature interval corresponds to the entire denaturation of the 969 bp nonexpected PCR product. This product results from a PCR reaction using the G6PD (A-)(exon 5) forward primer and the G6PD Mediterranean reverse primer.
196
P. Joly et al. / Clinical Biochemistry 43 (2010) 193–197
Fig. 2. (a) Normalized HRM curves of the three possible genotypes for the G6PD Med mutation. 1, homozygous or hemizygous wild; 2, heterozygous (note the two inflexion points on the graph); 3, homozygous or hemizygous mutant. At 76 °C, the denaturation of the 63 bp amplicon has not begun yet (100% fluorescence). At 82 °C, the denaturation is complete (0% fluorescence). (b) Normalized HRM curves of the three possible genotypes for the G6PD (A-) Exon 5 mutation. 1, homozygous or hemizygous wild; 2, heterozygous (note the two inflexion points on the graph); 3, homozygous or hemizygous mutant. At 82 °C, the denaturation of the 126 bp amplicon has not begun yet (100% fluorescence). At 86 °C, the denaturation is complete (0% fluorescence).
P. Joly et al. / Clinical Biochemistry 43 (2010) 193–197
197
Fig. 3. Normalized HRM curves of the three possible genotypes for the G6PD (A-) Exon 4 mutation. 1, homozygous or hemizygous wild; 2, heterozygous (note the two inflexion points on the graph); 3, homozygous or hemizygous mutant. At 78.5 °C, the denaturation of the 126 bp amplicon has not begun yet (100% fluorescence). At 84 °C, the denaturation is complete (0% fluorescence).
the A(-)(exon 5) and the A(-)(exon 4) mutations) has confirmed HRM results. The fact that neither the exon 7 nor the exon 9 A(-) mutations was found for these patients could lead to suppress this last step of direct sequencing, except in a case of a biochemically documented G6PD deficiency. In conclusion, the HRM diagnostic strategy presented here is perfectly adapted for a rapid screening of G6PD deficiency on patients originating from the Mediterranean and the African areas. For populations (i.e., South-East Asia) where the prevalent G6PD variants are different [7–9], this protocol will not be sufficient. But, it is perfectly possible to adapt similar HRM protocols to any mutation of interest. The strategy described here could be theoretically even more simplified if the three PCR products were amplified and analyzed in a single tube. This implies no overlapping of the three denaturation intervals. To achieve such a result, the addition of AT- or GCrich tails at the 5′-end of each primer seems to be the more promising way [13,14]. References [1] Beutler E. Glucose-6-phosphate dehydrogenase deficiency: a historical perspective. Blood 2008;111:16–24. [2] Cappellini M, Fiorelli G. Glucose-6-phosphate dehydrogenase deficiency. Lancet 2008;371:64–74. [3] Mason PJ, Bautista JM, Gilsanz F. G6PD deficiency : the genotypephenotype association. Blood Rev 2007;21:267–83. [4] Beutler E. G6PD deficiency. Blood 1994;84:3613–36. [5] Vulliamy TJ, D'Urso M, Battistuzzi G, et al. Diverse point mutations in the human glucose-6-phosphate dehydrogenase gene cause enzyme deficiency
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
and mild severe hemolytic anemia. Proc Natl Acad Sci U S A 1988;85: 5171–5. Di Montemuros FM, Dotti C, Tavazzi D, Fiorelli G, Cappellini MD. Molecular heterogeneity of glucose-6-phosphate dehydrogenase (G6PD) variants in Italy. Hematologica 1997;82:440–5. Zhang DT, Hu LH, Yang YZ. Detection of three common G6PD gene mutations in Chinese individuals by probe melting curves. Clin Biochem 2005;38:390–4. Wu G, Liang WH, Zhu J, et al. Rapid, simultaneous genotyping of 10 Southeast Asian glucose-6-phosphate dehydrogenase deficiency causing mutations and a silent polymorphism by multiplex-primer extension/ denaturing HPLC. Clin Chem 2005;51:1288–91. Li L, Zhou YQ, Xiao QZ, Yan TZ, Xu XM. Development and evaluation of a reverse dot blot assay for the simultaneous detection of six common Chinese G6PD mutations and one polymorphism. Blood Cells Mol Dis 2008;41:17–21. Minucci A, Delibato E, Castagnola M, et al. Identification of RFLP G6PD mutations by using microcapillary electrophoretic chips (Experion™). J Sep Sci 2008;31:2694–700. Farez-Vidal ME, Gandia-Pla S, Blanco S, Gómez-Llorente C, GómezCapilla A. Multi-mutational analysis of fifteen common mutations of the glucose 6-phosphate dehydrogenase gene in the Mediterranean population. Clin Chim Acta 2008;395:94–8. Bernaudin F, Verlhac S, Chevret S, et al. G6PD deficiency, absence of αthalassemia, and hemolytic rate at baseline are significant independent risk factors for abnorally high cerebral velocities in patients with sickle cell anemia. Blood 2008;112:4314–7. Vossen RH, Aten E, Roos A, Den Dunnen JT. High-resolution melting analysis (HRMA) more than just sequence variant screening. Hum Mutation 2009 published Online 18 March 2009. Tindall EA, Petersen DC, Woodbridge P, Schipany K, Hayes VM. Assessing high-resolution melt curve analysis for accurate detection of gene variants in complex DNA fragments. Hum Mutation 2009 published Online 11 March 2009.
Clinical Biochemistry 43 (2010) 193 – 197
Rapid genotyping of two common G6PD variants, African (A-) and Mediterranean, by high-resolution melting analysis Philippe Joly a,b , Philippe Lacan a , Caroline Garcia a , Cyril Martin b , Alain Francina a,⁎ a
Unité de Pathologie Moléculaire du Globule Rouge, Fédération de Biochimie et de Biologie Spécialisée, Hôpital Edouard Herriot, Hospices Civils and Université Claude Bernard-Lyon 1, Lyon, France b Centre de Recherche et d'Innovation sur le Sport (EA 647), Université Claude Bernard Lyon I, Lyon, France Received 19 May 2009; received in revised form 12 August 2009; accepted 10 September 2009 Available online 24 September 2009
Abstract Objectives: The Mediterranean and A(-) G6PD variants are particularly prevalent in Africa and Southern Europe. Our study was aimed to develop an assay for the rapid genotyping of these two variants by HRM. Methods: After PCR reactions corresponding to the G6PD Mediterranean (exon 6), G6PD (A-) (exon 4) and G6PD (A-) (exon 5) mutations, amplicons were submitted to HRM. This protocol was applied to a cohort of 132 patients suffering from sickle cell disease. Results: Wild, homozygous or hemizygous and heterozygous states were fully discriminated by HRM for all three mutations. HRM results were in total accordance with DNA sequencing for 22 patients of our cohort with a ‘A’ genotype: presence of the (A-) (exon 5) mutation but absence of the (A-) (exon 4) mutation. Conclusions: Our HRM protocols allow a rapid, simple and cost-effective screening of G6PD deficiency in patients originating from the Mediterranean and the African areas. © 2009 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. Keywords: G6PD genotyping; HRM; Mediterranean mutation; African (A-) mutation
Introduction Glucose-6-phosphate dehydrogenase (G6PD) deficiency is the most common enzymatic defect in the world [1–4]. As it is an X-linked genetic disorder, red blood cells (RBC) enzymatic activities are significantly decreased in hemizygous males and homozygous females for most of G6PD variants. In such cases, the main clinical presentations can be a neonatal jaundice, an acute hemolytic anemia or a chronic non-spherocytic hemolytic anemia depending on the type of G6PD variant (Classes I to IV of the WHO classification) [1–4]. For heterozygous females, the clinical manifestations are less frequent because RBC enzymatic activities can be highly variable due to a possible bias in the X chromosome inactivation [4]. More than 160 mutations or combinations of mutations in the G6PD gene have been ⁎ Corresponding author. Unité de Pathologie Moléculaire du Globule Rouge, Fédération de Biochimie et de Biologie Spécialisée, Hôpital Edouard Herriot, 5, Place d'Arsonval, 69437 Lyon Cedex 03, France. Fax: +33 4 72 11 05 98. E-mail address: [email protected] (A. Francina).
characterized at the DNA level (available from: The Human Gene Mutation Database at the Institute of Medical Genetics in Cardiff (UK); http://www.hgmd.cf.ac.uk/ac/index.php, July 31, 2009). Nevertheless, the direct sequencing of the entire G6PD gene (13 exons) is time-consuming and quite expensive, particularly to perform mass screening studies. In our clinical experience, as well as in litterature data [1–4], two G6PD variants, Mediterranean and (A-), are particularly prevalent in Africa and in Southern Europe. The Mediterranean variant results from a single amino-acid substitution Ser188Phe on exon 6 (563 C→T in cDNA; rs5030868)[1–5]. The G6PD African (A-) variant results from two amino acid substitutions. The first one is the Asn126Asp substitution on exon 5 (376 A→G in cDNA; rs1050829) which has no direct consequence by itself on G6PD activity [5]. The resulting G6PD variant is called (A). But, in association with another mutation, it can give rise to ‘abnormal’ (A-) variants with lower G6PD activity. The most frequent one is the Val68Met on exon 4 (202 G→A in cDNA; rs1050828) while the two others (Arg227Leu and Leu323Pro) are much less frequent [1–5].
0009-9120/$ - see front matter © 2009 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.clinbiochem.2009.09.012
194
P. Joly et al. / Clinical Biochemistry 43 (2010) 193–197
The present study was aimed to develop an assay for the rapid genotyping by high-resolution melting (HRM) of these two common G6PD variants. This technique has been further applied for validation to a large cohort of patients suffering from sickle cell disease. Materials and methods Patient samples and DNA isolation The DNA samples for the preliminary tests were obtained from patients with a (A-) or a Mediterranean G6PD variant, previously characterized by direct sequencing (data not shown) after written consent for genetic analysis. Genomic DNA was extracted from 200 μL of whole EDTA blood using the DNA Qiacube extractor (Qiagen, Courtaboeuf, France). The final elution was made in Tris-EDTA buffer before storage at -20 °C. After the development assays, our technique was evaluated on a cohort of 132 patients suffering from sickle cell disease (SCD) and originating from Africa or French West Indies. This validation study was a part of a larger study conducted on this cohort for which a written consent for genetic analysis had been previously obtained. Selection of primers All primers (GenBank Accession Number × 55448) used for HRM genotyping were designed with the Primer 3 software. For the G6PD (A-) (exon 4) mutation, PCR primers were 5′CTTCTGCCCGAAAACACCTT-3′ (forward- at coordinates: 13479–13498) and 5′-ACACCACCCACCTTGAAGAAG-3′ (reverse- at coordinates: 13559–13579). The PCR product was 101-bp length. For the G6PD (A-) (exon 5) mutation, PCR primers were 5′-TCTTTGCCCGCAACTCCTAT-3′ (forwardat coordinates: 14152–14171) and 5′-GCAAGGCCAGGTAGAAGAGG-3′ (reverse- at coordinates: 14258–14277). The PCR product was 126-bp length. For the G6PD Mediterranean (exon 6) mutation, PCR primers were: 5′-TCTGACCGGCTGTCCAAC-3′ (forward- at coordinates: 15059–15076) and 5′GTCGATGCGGTAGATCTGGT-3′ (reverse- at coordinates: 15102–15121). The PCR product was 63 bp length. PCR and high-resolution melting Two distinct steps were performed: (i) a specific PCR reaction with a standard thermocycler (with the double strand DNA dye included in the Master Mix); (ii) a HRM analysis on the resulting amplicon with the Rotor Gene Instrument (RotorGene 6000, Qiagen, Courtaboeuf, France). The search for the G6PD (A-)(exon 5) and for the G6PD Mediterranean mutations was carried out simultaneously by duplex PCR, while the G6PD (A-)(exon 4) mutation was studied independently. For the duplex PCR reaction, each tube contained, in a final volume of 30 μL, 50 ng DNA template, 1 × PCR buffer (QBiogen, Illkirch, France), 10 mM of each dNTP, 100 ng of each primer, 1.5 U of Taq DNA polymerase (Q-Biogen), 2.5 mM of Syto®9 (Invitrogen, Cergy-Pontoise, France) and 1.5 mM
MgCl2. PCR conditions were as follows: 10 min denaturation at 95 °C; 35 cycles with denaturation 30 s at 95 °C, hybridization 30 s at 59 °C and elongation 1 min at 72 °C and final elongation 5 min at 72 °C. The HRM protocol was performed with 15 μL of PCR product: 10 min at 95 °C; 1 min at 50 °C (to facilitate the formation of heteroduplexes) and a shift from 70 to 90 °C (0.1 °C/s) with a continuous fluorescence acquisition (green channel). For the G6PD (A-)(exon 4) PCR reaction, the PCR composition was exactly the same. PCR conditions were also identical except for the hybridization temperature (58 °C). The HRM protocol was performed with 10 μL of PCR product: 10 min at 95 °C; 1 min at 50 °C and a shift from 75 to 85 °C (0.1 °C/s) with a continuous fluorescence acquisition (green channel). Results Duplex PCR for G6PD (A-)(exon 5–376 A→G)+ G6PD Mediterranean (exon 6–563 C→T) The duplex PCR gives three products. Two of them are the expected amplicons: 63 pb for the G6PD Mediterranean and 126 pb for the G6PD (A-)(exon 5). The third PCR product is 969-bp length and results from a PCR reaction using the G6PD (A-)(exon 5) forward primer and the G6PD Mediterranean reverse primer. Derivative plots (Fig. 1) show that these three amplicons have distinct denaturation intervals: approximately 76–82 °C for G6PD Mediterranean, 82–86 °C for G6PD (A-) (exon 5) and 86–92 °C for the 969-bp unexpected amplicon. As they do not overlap each other, the G6PD Mediterranean and the G6PD (A-) (exon 5) amplicons can be analyzed separately. In each case, the three possible genotypes are clearly distinguished (Figs. 2a and b): homozygous wild (or hemizygous wild for male subjects), heterozygous (only possible for females) and homozygous mutant (or hemizygous mutant for male subjects). G6PD (A-) exon 4 (202 G→A) The HRM analysis, between 75 and 85 °C, of the G6PD (A-) (exon 4) PCR product also allows to discriminate unambiguously between the three possible genotypes (Fig. 3). Cohort of SCD patients Among the 132 SCD patients of our cohort, the Mediterranean variant was found at the hemizygous state for one male and at the heterozygous state for three family-related females. The African (A-) genotype (association of the exon 4 and of the exon 5 mutations) was found at the hemizygous state for 15 males, at the homozygous state for one female and at the heterozygous state for 14 females (one was homozygous for the exon 5 mutation). We also found 22 patients without the exon 4 mutation but with the exon 5 mutation (7 hemizygous males, 14 heterozygous females and 1 homozygous female). For these patients, a direct sequencing was performed to confirm the
P. Joly et al. / Clinical Biochemistry 43 (2010) 193–197
results and to search one of the two others A(-) mutations (Arg227Leu on exon 7 and Leu323Pro on exon 9). HRM results were confirmed for all patients and no additional A(-) mutation was found on exons 7 and 9. Discussion Several molecular methods have been reported for the genotyping of G6PD variants, e.g., DNA sequencing, allelespecific oligoprobe hybridization, PCR restriction analysis, reverse dot blot assay, allele-specific PCR, denaturing HPLC or real-time PCR using FRET probes [6–10]. However, these methods are time-consuming, expensive and include many steps. A recent paper from Farez-Vidal et al. describes a multiplex multicolour assay using fluorescence detection and capillary electrophoresis for the simultaneous detection of 15 G6PD variants [11]. This method is rapid and very cost-saving compared to others but requires four fluorescently labeled ddNTPs and multiple primers pairs. Comparatively, using the present HRM protocols, we carried out a simple, rapid and very cost-effective strategy for the molecular screening of G6PD deficiency. The duplex reaction is performed in a first step and allows the detection of the Mediterranean variant and of the (A-)(exon 5) mutation. The second HRM analysis needs to be performed only on samples carrying the (A-)(exon 5) mutation. The presence of the Asn126Asp mutation on exon 5 and the absence of the Val68Met substitution on exon 4 require an additional DNA sequencing to eventually detect another ‘(A-)’ mutation. This
195
strategy is sufficient for a first screening of G6PD deficiency. A negative result must be checked by direct sequencing of the entire gene if a G6PD deficiency has been biochemically documented. The main limitation of all HRM genotyping protocols is the possibility that the single nucleotide polymorphism (SNP) of interest is replaced by another unexpected SNP. That is why we have designed amplicons as short as possible to limit the probability of such an event. Nevertheless, a few G6PD variants have been described in close proximity of the three mutations searched here (Available from: The Human Gene Mutation Database at the Institute of Medical Genetics in Cardiff (UK); http://www.hgmd.cf.ac.uk/ac/index.php, July 31, 2009) but their frequencies are negligible compared to the Mediterranean and A(-) variants. To avoid a misdiagnosis, a direct sequencing should be performed in two main situations: (i) if the HRM curve of an unknown patient is not perfectly matched with one of the DNA controls in a repeatedly way and (ii) if there is a genotype / phenotype (i.e., G6PD activity) discordance. The screening of the G6PD Mediterranean and (A-) variants seems particularly interesting in SCD patients. Indeed, Bernaudin et al have shown that a G6PD deficiency independently increases the risk of cerebral vasculopathy in SCD patients [12]. As these patients frequently receive blood transfusions, the enzymatic G6PD activity could be potentially false and a genotyping appears more appropriate. Results obtained for our 132 SCD patients cohort seem to validate the diagnostic strategy presented above as the direct genotyping of the 22 patients with ‘unexpected’ results (dissociation between
Fig. 1. Derivative plots obtained during the HRM analysis following the Multiplex PCR. The recordings of the derivative values of fluorescence vs. temperature (dF/dT) show three distinct intervals: (A) This 76–82 °C temperature interval corresponds to the entire denaturation of the ‘G6PD Med’ PCR product. (B) This 82–86 °C temperature interval corresponds to the entire denaturation of the ‘G6PD (A-) mutation (exon 5)’ PCR product. (C) This 86–90 °C temperature interval corresponds to the entire denaturation of the 969 bp nonexpected PCR product. This product results from a PCR reaction using the G6PD (A-)(exon 5) forward primer and the G6PD Mediterranean reverse primer.
196
P. Joly et al. / Clinical Biochemistry 43 (2010) 193–197
Fig. 2. (a) Normalized HRM curves of the three possible genotypes for the G6PD Med mutation. 1, homozygous or hemizygous wild; 2, heterozygous (note the two inflexion points on the graph); 3, homozygous or hemizygous mutant. At 76 °C, the denaturation of the 63 bp amplicon has not begun yet (100% fluorescence). At 82 °C, the denaturation is complete (0% fluorescence). (b) Normalized HRM curves of the three possible genotypes for the G6PD (A-) Exon 5 mutation. 1, homozygous or hemizygous wild; 2, heterozygous (note the two inflexion points on the graph); 3, homozygous or hemizygous mutant. At 82 °C, the denaturation of the 126 bp amplicon has not begun yet (100% fluorescence). At 86 °C, the denaturation is complete (0% fluorescence).
P. Joly et al. / Clinical Biochemistry 43 (2010) 193–197
197
Fig. 3. Normalized HRM curves of the three possible genotypes for the G6PD (A-) Exon 4 mutation. 1, homozygous or hemizygous wild; 2, heterozygous (note the two inflexion points on the graph); 3, homozygous or hemizygous mutant. At 78.5 °C, the denaturation of the 126 bp amplicon has not begun yet (100% fluorescence). At 84 °C, the denaturation is complete (0% fluorescence).
the A(-)(exon 5) and the A(-)(exon 4) mutations) has confirmed HRM results. The fact that neither the exon 7 nor the exon 9 A(-) mutations was found for these patients could lead to suppress this last step of direct sequencing, except in a case of a biochemically documented G6PD deficiency. In conclusion, the HRM diagnostic strategy presented here is perfectly adapted for a rapid screening of G6PD deficiency on patients originating from the Mediterranean and the African areas. For populations (i.e., South-East Asia) where the prevalent G6PD variants are different [7–9], this protocol will not be sufficient. But, it is perfectly possible to adapt similar HRM protocols to any mutation of interest. The strategy described here could be theoretically even more simplified if the three PCR products were amplified and analyzed in a single tube. This implies no overlapping of the three denaturation intervals. To achieve such a result, the addition of AT- or GCrich tails at the 5′-end of each primer seems to be the more promising way [13,14]. References [1] Beutler E. Glucose-6-phosphate dehydrogenase deficiency: a historical perspective. Blood 2008;111:16–24. [2] Cappellini M, Fiorelli G. Glucose-6-phosphate dehydrogenase deficiency. Lancet 2008;371:64–74. [3] Mason PJ, Bautista JM, Gilsanz F. G6PD deficiency : the genotypephenotype association. Blood Rev 2007;21:267–83. [4] Beutler E. G6PD deficiency. Blood 1994;84:3613–36. [5] Vulliamy TJ, D'Urso M, Battistuzzi G, et al. Diverse point mutations in the human glucose-6-phosphate dehydrogenase gene cause enzyme deficiency
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
and mild severe hemolytic anemia. Proc Natl Acad Sci U S A 1988;85: 5171–5. Di Montemuros FM, Dotti C, Tavazzi D, Fiorelli G, Cappellini MD. Molecular heterogeneity of glucose-6-phosphate dehydrogenase (G6PD) variants in Italy. Hematologica 1997;82:440–5. Zhang DT, Hu LH, Yang YZ. Detection of three common G6PD gene mutations in Chinese individuals by probe melting curves. Clin Biochem 2005;38:390–4. Wu G, Liang WH, Zhu J, et al. Rapid, simultaneous genotyping of 10 Southeast Asian glucose-6-phosphate dehydrogenase deficiency causing mutations and a silent polymorphism by multiplex-primer extension/ denaturing HPLC. Clin Chem 2005;51:1288–91. Li L, Zhou YQ, Xiao QZ, Yan TZ, Xu XM. Development and evaluation of a reverse dot blot assay for the simultaneous detection of six common Chinese G6PD mutations and one polymorphism. Blood Cells Mol Dis 2008;41:17–21. Minucci A, Delibato E, Castagnola M, et al. Identification of RFLP G6PD mutations by using microcapillary electrophoretic chips (Experion™). J Sep Sci 2008;31:2694–700. Farez-Vidal ME, Gandia-Pla S, Blanco S, Gómez-Llorente C, GómezCapilla A. Multi-mutational analysis of fifteen common mutations of the glucose 6-phosphate dehydrogenase gene in the Mediterranean population. Clin Chim Acta 2008;395:94–8. Bernaudin F, Verlhac S, Chevret S, et al. G6PD deficiency, absence of αthalassemia, and hemolytic rate at baseline are significant independent risk factors for abnorally high cerebral velocities in patients with sickle cell anemia. Blood 2008;112:4314–7. Vossen RH, Aten E, Roos A, Den Dunnen JT. High-resolution melting analysis (HRMA) more than just sequence variant screening. Hum Mutation 2009 published Online 18 March 2009. Tindall EA, Petersen DC, Woodbridge P, Schipany K, Hayes VM. Assessing high-resolution melt curve analysis for accurate detection of gene variants in complex DNA fragments. Hum Mutation 2009 published Online 11 March 2009.