Clinica Chimica Acta 418 (2013) 45–49

Contents lists available at SciVerse ScienceDirect

Clinica Chimica Acta

j ourna l homepage: www.e lsev ie r .com/ locate /c l inch im

A reverse dot blot assay for the expanded screening of eleven ChineseG6PD mutations

Xiaomei Lu a,⁎,1, Liang Hua b,1, Ting Zhang c, Siping Li a, Xuejin Fan a, Qi Peng a, Wenrui Li a, Junqin Ye a,Jianling Long a, Xiaoguang He a

a Dongguan Institute of Pediatrics, Guangdong Medical Collage Affiliated Shilong Boai Hospital, 68 Xihu Third Road, Dongguan, Guangdong, Chinab Guangzhou Women and Children Medical Center, 318 Renmin Middle Road, Guangzhou, Guangdong, Chinac Department of Genetics, School of Medicine, Zhejiang University, 866 Yuhangtang Road, Hangzhou, China

⁎ Corresponding author. Tel.: +86 86186505; fax: +E-mail address: lxm020@126.com (X. Lu).

1 These two authors contributed equally in this study

0009-8981/$ – see front matter © 2013 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.cca.2012.12.023

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 14 November 2012Received in revised form 9 December 2012Accepted 18 December 2012Available online 8 January 2013

Keywords:Glucose 6-phosphate dehydrogenasedeficiency (G6PD)Reverse dot blot (RDB)Multiplex polymerase reaction (M-PCR)Point mutationGenetic screening

Background: Glucose-6-phosphate dehydrogenase (G6PD) deficiency is a multiethnic inherited disease with aparticularly high prevalence in tropical and subtropical regions including southern China. A convenient andreliable method is required to detect common G6PD mutations in the Chinese population.Methods: We developed a reverse dot blot (RDB) assay for the expanded screening of eleven mutations(c.95A>G, c.392G>T, c.871G>A, c.1004C>T, c.1004C>A, c.1024C>T, c.1360C>T, c.1376G>T, c.1381G>A,c.1387C>T, c.1388G>A). The method consists of a single-tube multiplex PCR amplification of four fragmentsin the G6PD target sequence of wild-type and mutant genomic DNA samples followed by hybridization to atest strip containing allele-specific oligonucleotide probes.We applied ourmethod to a group of 213 unrelatedChinese patients.Results: The test had a detection rate of 95.8%, validated by direct sequencing in a blind study with 100%concordance.Conclusions: The results demonstrate that our reverse dot blot assay is an easy, reliable, high-yield and cost-

effective method for genetic screening to identify G6PD patients and carriers among the Chinese population.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

Glucose 6-phosphate dehydrogenase (G6PD) deficiency is one ofthe most common X-linked enzymopathies, affecting more than400 million people worldwide [1,2]. G6PD deficiency manifests in dif-ferent clinical phenotypes, including neonatal jaundice and hemolysisinduced by ingestion of fava beans or oxidant drugs. The G6PD gene,mutations in which are the molecular genetic basis of the disorder,is located at Xq28, consists of 13 exons and encodes a 515 aminoacid polypeptide. To date, more than 180 G6PD mutations have beendocumented [3], and each ethnic population presents a characteristicmutation profile. There are at least 21 distinct point mutationsreported in the Chinese population, eight of which (c.95A>G, c.392G>T,c.871G>A, c.1004C>T, c.1024C>T, c.1360C>T, c.1376G>T, c.1388G>A)account for more than 80% of cases (Table 1) [4–7].

Presently, different methods are used to detect G6PD mutations,including classic polymerase chain reaction/restriction enzyme analy-sis (PCR-RFLP) [8], amplification refractory mutation system (ARMS)[9] and denaturing high-performance liquid chromatography (DHPLC)[10]. However, for convenience and technical simplicity, reverse dot

86 86118287.

.

rights reserved.

blot assay has been developed and applied in several genetic disordersincluding G6PD deficiency [11].

In this study, we employed a reverse dot blot assay for the ex-panded screening of the eight most common G6PD mutations andthree adjacent mutations (c.1004C>A, c.1381G>A and c.1387C>T)in 213 unrelated Chinese patients. All results were confirmed by directsequencing in a blind study.

2. Materials and methods

2.1. Subjects

A total of 243 unrelated samples (80 males and 163 females) com-posed of 213 G6PD deficient cases and 30 normal controls were inves-tigated and used to test the specificity and accuracy of the assay. G6PDdeficient cases ranged from newborn to 37 years of age. Samples werecollected at Shilong Boai Hospital, affiliated with Guangdong MedicalCollege. The quantitative assay of G6PD activity was measured withG6PD/6GPD ratios (GuangzhouMiji Company). Informedwritten con-sent was obtained from each patient's parent or guardian. The studywas approved by the Institutional Ethnics Committee of Shilong BoaiHospital.

Positive controls for the establishment of the assay were clini-cal blood samples with seven common G6PD mutations (c.95A>G,

Table 1The frequency of the eleven G6PD mutations reported in Chinese population both in our experiment and literature.

Mutation types Hemizygote Heterozygote Homozygote Total Frequency % (experiment) Frequency % (literature [12])

G1376T 28 47 2 77 36.2 3.6–53.0G1388A 24 38 0 62 29.1 10.5–64.0A95G 13 18 0 31 14.6 2.2–26.8G871A 6 8 0 14 6.6 2.8–11.1G392T 3 6 0 9 4.2 0.8–8.3C1024T 1 4 0 5 2.3 1.0–28.6C1004T 1 0 0 1 0.5 0.8–1.23G1381A 1 0 0 1 0.5C1004A 0 0 0 0 0C1360T 0 0 0 0 0 0.8–4.6C1387T 0 0 0 0 0G1376T/G1388A 0 2 0 2 0.9G1376T/A95G 0 1 0 1 0.5G1388A/G871A 0 1 0 1 0.5Mutation unknown 0 0 0 9 4.2 3.5–54.1

46 X. Lu et al. / Clinica Chimica Acta 418 (2013) 45–49

c.392G>T, c.871G>A, c.1004C>T, c.1024C>T, c.1376G>T andc.1388G>A) and one adjacent mutation c.1381G>A, previously identi-fied by DNA sequencing. The remaining three (c.1004C>A, c.1360C>Tand c.1387C>T)were obtained by site-directedmutagenesis (ShanghaiShanjing Biological Company).

2.2. Site-directed mutagenesis

PCR reaction using primers of 3F/3R and 4F/4R, to amplify thesegments containing the specific sites (c.1004C>A, c.1360C>T andc.1387C>T), was performed with normal sample. Ligature of the PCRproduct and pGEM-T Easy vector was transformed to JM109 compe-tent cells. The white clones were selected from plates containing am-picillin, X-gal and IPTG, and gained the desired plasmids as confirmedby sequencing. These correct plasmids were amplified with one ofthe three sets of allele-specific primers (Table 2) and the remainingtemplate DNA was eliminated by DpnI. Finally, the PCR productswere treated with the same process, including ligation, transforma-tion and selection, and the plasmid of interest was selected and vali-dated again by direct-sequencing.

2.3. Genomic DNA extraction

Genomic DNAwas extracted from peripheral whole blood sampleswith the QIAamp DNA Blood Mini kit (QIAGEN) according to the in-structions of manufacturer. The qualitative estimation of DNA wascarried out by spectrophotometry (Nanodrop).

2.4. Design of primers and probes

Four sets of primers were designed for multiplex polymerasechain reaction (M-PCR) to amplify exons 2, 5, 9, 11 and 12 of theG6PD gene spanning the 10 mutation sites (Table 3). The 5′ end ofprimers was labeled with biotin. A total of 19 probes including 8 nor-mal and 11 mutant probes covering the point mutation sites weredesigned (Table 4). Three of the normal probes were used as controls

Table 2Three sets of allele-specific primers for site-directed mutagenesis.

Mutation Primers Sequence of primers (5′–3′)

C1004A 1004A-P1 gcgggtccaccaccgAacacttttgcagccgtc1004A-P2 gacggctgcaaaagtgTcggtggtggacccgc

C1360T 1360T-P1 ccagatgcacttcgtgTgcaggtgaggcccagc1360T-P2 gctgggcctcacctgcAcacgaagtgcatctgg

C1387T 1387T-P1 gctccgtgaggcctggAgtattttcaccccactgc1387T-P2 gcagtggggtgaaaatacTccaggcctcacggagc

a The capital letters indicate the mutagenesis of the base.

for six mutant probes: 1004N for C1004T and C1004A; 1376N forG1376T and G1381A; and 1388N for G1388A and C1387T. In orderto monitor the performance of the color development system, an ad-ditional probe PC was synthesized as an indicator.

2.5. Multiplex PCR amplification

Multiplex PCR amplification was performed on an Mx3000p PCRmachine (Stratagene) in a final volume of 50 μL containing 1×PCRbuffer (MgCl2 plus), 1×Q buffer, 0.2 mmol/L of each dNTPs (Promega),0.2 μmol/L each of the eight primers, 0.1 U/μL HotStarTaq DNA Poly-merase (QIAGEN) and 1 μg genome DNA. PCR conditions were per themanufacturer's instructions: pre-denaturation at 96 °C for 15 min;35 cycles of denaturation at 95 °C for 30 s, annealing at 55 °C for 30 s,and extension at 72 °C for 30 s; and a final extension at 72 °C for5 min. The products were subsequently visualized by electrophoresison a 1.5% agarose gel.

2.6. Reverse dot blot assay

All 20 probes (mentioned above) were fixed to nylon strips (PALLBiodyne C) arrayed as indicated in Fig 1. The PCR products wereheated to 95 °C or higher then immediately cooled to 0 °C. Eachstrip and 8 ml hybridization solution A (2×SSC, 0.1% SDS, pH7.4)was pre-heated to 45 °C after which the denatured PCR productswere added and incubated at 45 °C for 2 h in a screw-top tube. Sub-sequently, the strips were washed with wash solution B (0.5×SSC,0.1% SDS, pH7.4) at 45 °C for 10 min. Afterwards the strips weretransferred to a hybridization solution A diluted mixture containing0.125 U/mL Streptavidin-POD conjugate (Roche, Mannheim, Germany)and incubated at room temperature for 30 min. Excess conjugate wasremoved with another two washes of solution A. Finally, the color de-veloping solution composed of pH5.4 0.1 mg/mL TMB, 0.015‰ H2O2

and 0.1 mol/L sodium citrate was added and the color reaction devel-oped for 20 min. Blue dots indicated the positive results of detection.

Table 3Multiplex PCR primers.

Sequences of primer (5′–3′) Mutations Size (bp)

Exon 2 1F ACAGCGTCATGGCAGAGCAG1R GGGCGACCAGAGCAAAACT

95 354

Exon 5 2F TGCCCGCAACTCCTATGTGG2R AGGACTCGTGAATGTTCTTGGTGA

392 167

Exon 9 3F GTCATCCCTGCACCCCAACTC3R GCCGCAGCGCAGGATGAAG

871, 1004, 1024 414

Exon 11/12 4F TGGTGGCAGGCAGTGGCATCA4R CGTGGCGGGGGTGGAGGTG

1360, 1376, 1381,1387, 1388

538

Table 4G6PD gene mutation detection probes.

Name of probe Mutation detection probe (5′→3′) Name of probe Normal probe (5′→3′)

A95G NH2-GGATACACGCATATTCATCA 95N NH2-GGATACACACATATTCATCAG392T NH2-CTCCACCTGGTGTCACAG 392N NH2-CTCCACCTGGGGTCACAGG871A NH2-AGGTCAAGATGTTGAAATG 871N NH2-AGGTCAAGGTGTTGAAATGC1004T NH2-CACCACCGTCACTTTTGCA 1004N NH2-CACCACCGCCACTTTTGCAC1004A NH2-CACCACCGACACTTTTGCAC1024T NH2-AGCCGTCGTCTTCTATGT 1024N NH2-AGCCGTCGTCCTCTATGTC1360T NH2-GCACTTCGTGTGCAGGTGA 1360N NH2-GCACTTCGTGCGCAGGTGAG1376T NH2-GAGCTCCTTGAGGCCTGG 1376/1381N NH2-GAGCTCCGTGAGGCCTGGG1381A NH2-GAGCTCCGTGAGACCTGGC1387T NH2-CCTGGTGTATTTTCACCCC 1387/1388N NH2-CCTGGCGTATTTTCACCCCG1388A NH2-CCTGGCATATTTTCACCCCPC bio-ATGCATGCATGCATGCATGC-NH2

47X. Lu et al. / Clinica Chimica Acta 418 (2013) 45–49

All samples were analyzed independently by DNA sequencing toconfirm the accuracy of the assay.

3. Results

3.1. Establishment of the assay

The assay was established using DNA samples from eight male pa-tients and three plasmid samples with the remaining three knownmutations introduced by site-directed mutagenesis. When analyzingthe results, observation of color development of the PC spot servedas the primary positive control. An absence of color development ofall or the majority of control probes served as a negative control.Since the G6PD gene is X-linked, gender was confirmed to match thenumber of alleles observed in each analysis.

In this assay, the eleven positive control samples tested showedexpected results. Normal DNA samples showed positive blue coloredspots for the PC spot and all the control probes, and negative whitecolored spots for the mutation probes. Homozygous and hemizygousDNA samples showed blue colored spots for the appropriate mutantprobes as well as negatively colored white spots for the correspond-ing probes (Fig. 2, Nos. 1–8). Heterozygous and compound heterozy-gous DNA samples showed blue colored spots for one or two mutantprobes, respectively, and the corresponding white colored controlprobe spots (Fig. 2, Nos. 12–14, 18–20, 15–17).

3.2. Validation of the assay

Among the 213 G6PD deficient cases assayed, mutations in 204patients were detected giving a mutation detection frequency of95.8%. The three most common mutations in this patient group werec.1376G>T, c.1388G>A and c.95A>G. Mutations c.1004C>A, c.1360C>Tand c.1387C>T were not detected in any of the samples (Table 1).All samples were subsequently examined by DNA sequencing anddemonstrated 100% concordance between the two methodologies(Fig. 3).

Fig. 1. Layout of the probes' dotting in the nylon strip designed for the reverse dot blotassay. N denotes the wild-type probes corresponding to each mutant probes (denotedas each mutation). PC is located at the lower right corner of the strip.

4. Discussion

G6PD deficiency is endemic in certain parts of the world, particu-larly in the tropics and subtropics, including southern China.

Traditional laboratory diagnosis for G6PD deficiency has involvedboth protein-based and DNA-based testing. Routine enzyme activitydetection methodology has included methemoglobin reduction test-ing, fluorescence spots methods, nitrocellulose tetrazole blue (NBT)slip methods, and G6P/6PG ratio methods, among others. However,few reliably identify G6PD deficiency heterozygous females, whichis a critical aspect of disease prevention. As a classic monogenic dis-ease, simple molecular genetic diagnosis is the most effective meansof detecting both affected males and carrier females.

Studies of G6PD using WHO standardized procedure have docu-mented more than 400 biochemical enzyme variants, and more than180 G6PD mutations have been documented. Most of the mutationsreported are single-base substitutions that lead to amino acid ex-changes (missense mutations), and no large-deletions or entire genedeletions have been reported. Mutations are distributed throughoutthe 12 exons except exon 1. Thus, a reliable, rapid and inexpensivemethod for detecting G6PD point mutations would be helpful to pa-tients, their families, the physicians that treat them, and the laborato-ries responsible for testing.

To date, common methods used to detect G6PD mutations (ARMS,PCR/RE) have presented variable challenges such as insufficient accura-cy, insufficient simplicity to run, and a limitation of mutations detectedin a single run.

Successful reverse dot blot assay has been reported in thegenotyping of six common Chinese G6PDmutations and one polymor-phism [11]. Our study expands the numbers of mutations screenedand is designed based on the recently updated G6PD mutation data-base. Our search of the literature between 2006 and 2012 pertainingto the Chinese population led us to add c.392G>T and c.1360C>T tothe common mutation screening panel. Notably, c.392G>T has beendescribed as one of the six most frequent mutations both by T. Yanet al. [5] and J.B. Yan et al. [6] and is thus an important addition, asour study confirmed. The three mutations (c.1004C>A, c.1381G>A,c.1387C>T), are adjacent to c.1004C>T and c.1376G>T, therefore en-abling us to amplify one fragment spanning several mutations, andeven to make the same normal probes for two neighboringmutations.

Furthermore, in this reverse dot blot assay, we have modifiedsome steps, such as the addition of a PC probe, to increase the accura-cy of the assay. Our results for 243 unrelated G6PD samples showed100% concordance with independent direct sequencing. Our reversedot blot assay can simultaneously genotype the eleven most commonChinese G6PD mutations for hemizygous, heterozygous and com-pound heterozygous individuals in one single test strip with a detec-tion rate of more than 95%. Given that our patient group is derivedfrom a very typical south China population, we expect our test stripwill be useful with a good detection rate in most of south China.

Fig. 2. Representative results of genotyping eleven G6PD deficiency mutations using reverse dot blot assay. Hemizygote: (1) G1376T; (2) G1388A; (3) A95G; (4) G392T; (5) G871A;(6) C1024T; (7) G1381A; and (9) C1004T. Vectors obtained by site-directed mutagenesis: (8) C1387T; (10) C1004A; and (11) C1360T. Heterozygotes and homozygote(12) G1376T/N; (13) G1388A/N; (14) G871A/N; (15) G1376T/A95G; (16) G1376T/G1388A; (17) G1388A/G871A; (18) A95G/N; (19) C1024T/N; and (20) G392T/N.

Fig. 3. Representative sequencing results of eleven G6PD deficiency mutations. The mutation sites are indicated by arrows. (A) A95G/N; (B) A95G hemizygote; (C) G392T/N;(D) G871A/N; (E) G1024T/N; (F) G1376T hemizygote; (G) G1376T/N; and (H) G1388A hemizygote.

48 X. Lu et al. / Clinica Chimica Acta 418 (2013) 45–49

49X. Lu et al. / Clinica Chimica Acta 418 (2013) 45–49

This method is easily performed with common equipment, with lowcost and can be completed in 6 h. It is one of the most ideal methodsfor smaller clinics and laboratories with limited resources to performfront-line testing to detect G6PD without having to resort to directDNA sequencing.

Conflict of interest

None declared.

Acknowledgments

We are grateful for the participation of all the patients and theirfamilies. We thank Dr. Michael Raff for his meticulous editing of themanuscript. This work was partially supported by the Dongguan Bu-reau of Science and Technology for the City Key Program of Scienceand Technology (Project Number: 2011105102017).

References

[1] Beutler E. G6PD: population genetics and clinicalmanifestations. Blood Rev 1996;10:45–52.

[2] Beutler E. Glucose-6-phosphate dehydrogenase deficiency: a historical perspec-tive. Blood 2008;111:16–24.

[3] Minucci A, Moradkhani K, Hwang MJ, Zuppi C, Giardina B, Capoluongo E.Glucose-6-phosphate dehydrogenase (G6PD) mutations database: review of the“old” and update of the new mutations. Blood Cells Mol Dis 2012;48:154–65.

[4] Du CS, Xu YK, Hua XY, Wu QL, Liu LB. Glucose-6-phosphate dehydrogenase vari-ants and their frequency in Guangdong, China. Hum Genet 1988;80:385–8.

[5] Yan T, Cai R, Mo O, et al. Incidence and complete molecular characterization ofglucose-6-phosphate dehydrogenase deficiency in the Guangxi Zhuang autonomousregion of southern China: description of four novel mutations. Haematologica2006;91:1321–8.

[6] Yan JB, Xu HP, Xiong C, et al. Rapid and reliable detection of glucose-6-phosphatedehydrogenase (G6PD) gene mutations in Han Chinese using high-resolutionmelting analysis. J Mol Diagn 2010;12:305–11.

[7] Jiang W, Yu G, Liu P, et al. Structure and function of glucose-6-phosphatedehydrogenase-deficient variants in Chinese population. Hum Genet 2006;119:463–78.

[8] Tang TK, Huang CS, Huang MJ, Tam KB, Yeh CH, Tang CJ. Diverse point mutationsresult in glucose-6-phosphate dehydrogenase (G6PD) polymorphism in Taiwan.Blood 1992;79:2135–40.

[9] Maffi D, Pasquino MT, Caprari P, et al. Identification of G6PD Mediterranean mu-tation by amplification refractory mutation system. Clin Chim Acta 2002;321:43–7.

[10] Wu G, Liang WH, Zhu J, et al. Rapid, simultaneous genotyping of 10 SoutheastAsian glucose-6-phosphate dehydrogenase deficiency-causing mutations and asilent polymorphism by multiplex primer extension/denaturing HPLC assay. ClinChem 2005;51:1288–91.

[11] Li L, Zhou YQ, Xiao QZ, Yan TZ, Xu XM. Development and evaluation of a reversedot blot assay for the simultaneous detection of six common Chinese G6PD muta-tions and one polymorphism. Blood Cells Mol Dis 2008;41:17–21.

[12] Pan SL. G6PD deficiency: distribution in East and Southeast Asia and positive se-lection by malaria. Chin J Health Birth Child Care 2007;13:42–52.