Vox Sanguinis (2014) 106, 227–233 © 2013 International Society of Blood Transfusion DOI: 10.1111/vox.12096
ORIGINAL PAPER
Glucose-6-phosphate dehydrogenase deficiency in Italian blood donors: prevalence and molecular defect characterization D. Maffi,1 M. T. Pasquino,1 L. Mandarino,1 P. Tortora,1 G. Girelli,2 D. Meo,2 G. Grazzini3 & P. Caprari1 1
Department of Hematology Oncology and Molecular Medicine, Istituto Superiore di Sanit a, Rome, Italy UOC Immunoematologia e Medicina Trasfusionale, Azienda Policlinico Umberto I, Rome, Italy 3 Italian National Blood Centre, Istituto Superiore di Sanita, Rome, Italy 2
Background In the countries with high G6PD deficiency prevalence, blood donors are not routinely screened for this genetic defect. G6PD deficiency is often asymptomatic, blood donors may be carriers of the deficiency without being aware of it. The aim of the study was to evaluate the prevalence of G6PD deficiency among the Italian blood donors. Design and Methods From October 2009 to April 2011, 3004 blood donors from a large hospital transfusion centre were screened for G6PD deficiency using differential pH-metry and the characterization of G6PD mutations was performed on G6PD-deficient subjects. The haematological features of G6PD-deficient and normal donors were also compared. Results Thirty-three subjects (25 men and 8 women) with low G6PD activity were identified, corresponding to 11% of the examined blood donor population. The frequencies of class II severe alleles (Mediterranean, Valladolid, Chatham and Cassano) and class III mild alleles (Seattle, A- and Neapolis) were 48% and 43%, respectively. The haematological parameters of G6PD- donors were within normal range; however, the comparison between normal and G6PD- class II donors showed significant differences.
Received: 22 March 2013, revised 14 August 2013, accepted 21 August 2013, published online 18 October 2013
Conclusion In Italy, the presence of blood donors with G6PD deficiency is not a rare event and the class II severe variants are frequent. The identification of G6PD-deficient donors and the characterization of the molecular variants would prevent the use of G6PD-deficient RBC units when the haemolytic complications could be relevant especially for high risk patients as premature infants and neonates and patients with sickle cell disease submitted to multiple transfusions. Key words: blood, donation testing, donors, red cell components, transfusion medicine (in general).
Introduction Glucose-6-phosphate dehydrogenase (G6PD E.C. 1.1.1.49) deficiency is a hereditary enzymatic disease in humans, due to mutations in the housekeeping X-linked Gd gene that encodes the glucose-6-phosphate dehydrogenase Correspondence: Patrizia Caprari, Istituto Superiore di Sanita, Viale Regina Elena 299, 00161 Rome, Italy E-mail: [email protected]
(G6PD) producing NADPH. The NADPH, preserving the reduced form of glutathione (GSH) and stabilizing catalase, counterbalances the oxidative stress triggered in all cells by oxidant agents. Red blood cells that lack the nucleus and mitochondria are unable to generate NADPH in any other way than in the pentose phosphate pathway and are more susceptible to oxidative damage than any other cells. The majority of G6PD-deficient subjects are asymptomatic; however, they may develop severe acute
227
228 D. Maffi et al.
haemolytic anaemia triggered by oxidative stress in the extracellular environment, mainly from oxidative drugs, infections and fava beans (favism) [1–3]. The clinical manifestations are variable and correlate with the type of molecular variant. To date, the analysis of the Gd gene has led to the identification of a number of mutations, mostly missense mutations, scattered throughout the entire coding region, underlying the phenotypic heterogeneity of G6PD deficiency. Recently, an update of G6PD mutations database has been published [4] reporting 189 G6PD variants characterized at the DNA level. Based on the levels of enzyme activity and severity of clinical manifestations, the G6PD variants were grouped into five classes: class I, characterized by severe deficiency (2700) was calculated by the formula: n ¼ ðP ð1 PÞ 2 582 Þ=e2 where the value of the expected prevalence P was 096%, derived from a previous newborn screening for G6PD deficiency in Latium [18], and the accepted error e for sample estimates was given as 05% (99% confidence interval). From October 2009 to April 2011, a screening for erythrocyte G6PD deficiency was performed on 3004 blood donors from the Transfusion Centre of Azienda Policlinico Umberto I, Rome, a big hospital transfusion service having about 20 000 blood donors and an average number of donations for donor (donation index) of 12/year. All donors were randomly selected, and they completed a predonation questionnaire stating that if they had ever suffered from favism, but this was not considered an exclusion criterion. The blood donor mean age (mean–SD) was 43 – 10 years in men, and 40 – 11 years in women, 72% and 28% of the screened people were periodic and occasional donors, respectively.
Blood samples Venous blood samples were collected in ethylene diamine tetraacetic acid (EDTA) tubes, kept at 4°C and processed within 4 h after collection. All samples were tested using an Advia 120 analyzer (Siemens – USA) to determine the haematological parameters: red blood cells (RBC), haemoglobin (Hb), haematocrit (HCT), mean cell volume (MCV) and mean cell haemoglobin concentration (MCHC).
Quantitative G6PD determination As a screening test to detect G6PD deficiency, the quantitative determination of G6PD activity on whole blood was performed by differential double starter pH-metry. Twenty-five ll of each whole blood sample was analysed on the differential PH-meter CL10 PLUS (Biocontrol Systems – Italy) using the EC-Bio G6PD/6PGD kit (Eurachem, Rome, Italy) at 37°C. The differential double starter pH-metry, like the spectrophotometric methods, has high diagnostic sensitivity and specificity (100%) for the © 2013 International Society of Blood Transfusion Vox Sanguinis (2014) 106, 227–233
G6PD deficiency in Italian blood donors 229
detection of severe deficient hemizygous males and homozygous females, but a lower diagnostic sensitivity in detecting the G6PD heterozygous females. The differential double starter pH-metry allows the determination of G6PD activity both as IU/gHb and as G6PD/6PGD ratio, so increasing the diagnostic sensitivity of the G6PD activity alone in detecting the G6PD Mediterranean heterozygotes from 63,3% up to 853% with a diagnostic specificity of 974% [19].
DNA analysis For G6PD-deficient (G6PD-) blood donors, the DNA analysis was performed to identify G6PD gene mutations. Genomic DNA was extracted from peripheral blood leucocytes (Qiagen, Dusseldorf, Germany). To characterize the G6PD molecular variants, such as G6PD Mediterranean 563T, G6PD A-202A, 376G, G6PD Seattle 844C and G6PD Union 1370T, which were more frequent in Italy, restriction fragment length polymorphism (RFLP) analysis and allele refractory mutation system (ARMS) analysis were used, in accordance with Martinez di Montemuros, Maffi and Liese [8, 20, 21]. On the negative DNA samples for RFLP or ARMS, the entire coding region (12 exons) was amplified by PCR according to Poggi et al. [22] and a nucleotidic sequence analysis was performed. The NC-000023 genetic reference was employed for the primer design and the sequence alignment.
Statistical analyses The Kolmogorov–Smirnov normality test and Q-Q plot were used to determine whether G6PD, G6PD/6PGD ratio and haematological parameters were normally distributed. Statistical analyses of haematological parameters were performed using the unpaired Student’s t-test and oneway ANOVA.
Results Reference intervals for the G6PD activity were determined on a reference population consisting of 200 male and 200 female donors with normal haematological parameters (Table 1), which were normally distributed. The G6PD
activity and G6PD/6PGD ratio values of the reference population resulted as being normally distributed; therefore, the reference intervals were calculated as parametric centiles 25–975 (Table 1). For G6PD deficiency, 3004 blood donors were screened: 2342 (78%) male donors and 662 (22%) female donors. The high percentage of male donors is a characteristic of the Italian blood donor population. In 2012, the national percentages of male and female donors were 70% and 30%, respectively. All donors declared in the questionnaire that they had never suffered from favism; however, 33 subjects (25 men and 8 women) with low G6PD activity were identified, corresponding to 11% of analysed population sample. None of the 33 subjects was aware of being G6PD deficiency carriers, even if 25 of these were periodic donors (76%). Among the G6PD-deficient donors, the values for G6PD activities for men and women were 027–349 and 310–598 U/g Hb, respectively; and G6PD/6PGD ratios for men and women were 003–063 and 039–083 U/ gHb, respectively (Table 2). The DNA analysis of the G6PD- donors led to the identification of the following class II G6PD molecular variants: (i) Mediterranean 563T in 13 subjects, 9 males and 4 heterozygous females; (ii) Valladolid 406 T in a male subject; (iii) Chatham 1003A in a male donor; iv) Cassano 1347C in an heterozygous female subject. Furthermore, 14 donors were carriers of class III G6PD molecular variants: (i) Seattle 844C variant in 12 subjects, 11 men and a women; (ii) A-202A,376G 1 heterozygous female donor; (iii) Neapolis 1400G, 1 male subject. No G6PD- blood donor carried double mutations in the Gd gene. In 2 subjects, the nucleotide sequence analysis of the Gd gene coding region did not identify any mutation, and in a female donor, it was not possible to perform the DNA analysis. As far as the allelic frequency is concerned, the frequency of class II severe alleles in donors was 48%, while the frequency of class III mild alleles was 43% (Table 2). The G6PD activities of different variants confirmed the very low activity values in male donors with class II variants (T
1003G>A 406C>T 1347G>C
M F M M F
844G>C
202G>A 376A>G 1400C>G
N.I. N.D.
M F F M M F
Alleles
Allelic frequency (%)
G6PD U/gHb
G6PD/6PGD
16 13 9 4 1 1 1 14 12 11 1 1 1 2 1
48 39 27 12 3 3 3 43 36 33 3 3 3 6 3
027–488
003–075
027–076 310–488 061 073 347 152–598
003–010 039–075 008 007 050 024–083
152–325 598 417 333 085/349 518
024–033 083 062 048 012/063 065
N.I., not identified; N.D., DNA analysis not performed.
15–325 U/gHb) and in all heterozygous females (310– 598 U/gHb) (Table 2). The gender differences in G6PD enzyme activity in male versus female carriers of the same mutation are due to the Mendelian X-linked inheritance of G6PD gene. Males are hemizygous for the G6PD gene mutation, and the enzyme activity value is always low. Females may be homozygous or heterozygous for the G6PD gene mutation. Homozygous females have low enzyme activity comparable with the values of hemizygous males, while heterozygous females, due to the random inactivation of one X chromosome (Lyonization phenomenon), are genetic mosaics and can have values of G6PD activity normal, intermediate or deficient depending on the ratio between normal and G6PD-deficient red cells. Accordingly, the G6PD activity values of heterozygous females may be variable ranging from very low to normal values. The heterozygous female donors showed intermediate G6PD activity values (310–598 U/ gHb) (Table 2). The haematological parameters of G6PD- donors were within normal ranges (Table 3); however, the comparison between normal and G6PD- donors showed highly significant differences between the two groups. In G6PD- male donors, increased values in MCV (P < 0001) were found with also decreased values in RBC (P < 0001), Hb (P = 004) and MCHC (P = 001). When the multiple comparison was made between normal donors and G6PDdonors with class II (G6PD
ORIGINAL PAPER
Glucose-6-phosphate dehydrogenase deficiency in Italian blood donors: prevalence and molecular defect characterization D. Maffi,1 M. T. Pasquino,1 L. Mandarino,1 P. Tortora,1 G. Girelli,2 D. Meo,2 G. Grazzini3 & P. Caprari1 1
Department of Hematology Oncology and Molecular Medicine, Istituto Superiore di Sanit a, Rome, Italy UOC Immunoematologia e Medicina Trasfusionale, Azienda Policlinico Umberto I, Rome, Italy 3 Italian National Blood Centre, Istituto Superiore di Sanita, Rome, Italy 2
Background In the countries with high G6PD deficiency prevalence, blood donors are not routinely screened for this genetic defect. G6PD deficiency is often asymptomatic, blood donors may be carriers of the deficiency without being aware of it. The aim of the study was to evaluate the prevalence of G6PD deficiency among the Italian blood donors. Design and Methods From October 2009 to April 2011, 3004 blood donors from a large hospital transfusion centre were screened for G6PD deficiency using differential pH-metry and the characterization of G6PD mutations was performed on G6PD-deficient subjects. The haematological features of G6PD-deficient and normal donors were also compared. Results Thirty-three subjects (25 men and 8 women) with low G6PD activity were identified, corresponding to 11% of the examined blood donor population. The frequencies of class II severe alleles (Mediterranean, Valladolid, Chatham and Cassano) and class III mild alleles (Seattle, A- and Neapolis) were 48% and 43%, respectively. The haematological parameters of G6PD- donors were within normal range; however, the comparison between normal and G6PD- class II donors showed significant differences.
Received: 22 March 2013, revised 14 August 2013, accepted 21 August 2013, published online 18 October 2013
Conclusion In Italy, the presence of blood donors with G6PD deficiency is not a rare event and the class II severe variants are frequent. The identification of G6PD-deficient donors and the characterization of the molecular variants would prevent the use of G6PD-deficient RBC units when the haemolytic complications could be relevant especially for high risk patients as premature infants and neonates and patients with sickle cell disease submitted to multiple transfusions. Key words: blood, donation testing, donors, red cell components, transfusion medicine (in general).
Introduction Glucose-6-phosphate dehydrogenase (G6PD E.C. 1.1.1.49) deficiency is a hereditary enzymatic disease in humans, due to mutations in the housekeeping X-linked Gd gene that encodes the glucose-6-phosphate dehydrogenase Correspondence: Patrizia Caprari, Istituto Superiore di Sanita, Viale Regina Elena 299, 00161 Rome, Italy E-mail: [email protected]
(G6PD) producing NADPH. The NADPH, preserving the reduced form of glutathione (GSH) and stabilizing catalase, counterbalances the oxidative stress triggered in all cells by oxidant agents. Red blood cells that lack the nucleus and mitochondria are unable to generate NADPH in any other way than in the pentose phosphate pathway and are more susceptible to oxidative damage than any other cells. The majority of G6PD-deficient subjects are asymptomatic; however, they may develop severe acute
227
228 D. Maffi et al.
haemolytic anaemia triggered by oxidative stress in the extracellular environment, mainly from oxidative drugs, infections and fava beans (favism) [1–3]. The clinical manifestations are variable and correlate with the type of molecular variant. To date, the analysis of the Gd gene has led to the identification of a number of mutations, mostly missense mutations, scattered throughout the entire coding region, underlying the phenotypic heterogeneity of G6PD deficiency. Recently, an update of G6PD mutations database has been published [4] reporting 189 G6PD variants characterized at the DNA level. Based on the levels of enzyme activity and severity of clinical manifestations, the G6PD variants were grouped into five classes: class I, characterized by severe deficiency (2700) was calculated by the formula: n ¼ ðP ð1 PÞ 2 582 Þ=e2 where the value of the expected prevalence P was 096%, derived from a previous newborn screening for G6PD deficiency in Latium [18], and the accepted error e for sample estimates was given as 05% (99% confidence interval). From October 2009 to April 2011, a screening for erythrocyte G6PD deficiency was performed on 3004 blood donors from the Transfusion Centre of Azienda Policlinico Umberto I, Rome, a big hospital transfusion service having about 20 000 blood donors and an average number of donations for donor (donation index) of 12/year. All donors were randomly selected, and they completed a predonation questionnaire stating that if they had ever suffered from favism, but this was not considered an exclusion criterion. The blood donor mean age (mean–SD) was 43 – 10 years in men, and 40 – 11 years in women, 72% and 28% of the screened people were periodic and occasional donors, respectively.
Blood samples Venous blood samples were collected in ethylene diamine tetraacetic acid (EDTA) tubes, kept at 4°C and processed within 4 h after collection. All samples were tested using an Advia 120 analyzer (Siemens – USA) to determine the haematological parameters: red blood cells (RBC), haemoglobin (Hb), haematocrit (HCT), mean cell volume (MCV) and mean cell haemoglobin concentration (MCHC).
Quantitative G6PD determination As a screening test to detect G6PD deficiency, the quantitative determination of G6PD activity on whole blood was performed by differential double starter pH-metry. Twenty-five ll of each whole blood sample was analysed on the differential PH-meter CL10 PLUS (Biocontrol Systems – Italy) using the EC-Bio G6PD/6PGD kit (Eurachem, Rome, Italy) at 37°C. The differential double starter pH-metry, like the spectrophotometric methods, has high diagnostic sensitivity and specificity (100%) for the © 2013 International Society of Blood Transfusion Vox Sanguinis (2014) 106, 227–233
G6PD deficiency in Italian blood donors 229
detection of severe deficient hemizygous males and homozygous females, but a lower diagnostic sensitivity in detecting the G6PD heterozygous females. The differential double starter pH-metry allows the determination of G6PD activity both as IU/gHb and as G6PD/6PGD ratio, so increasing the diagnostic sensitivity of the G6PD activity alone in detecting the G6PD Mediterranean heterozygotes from 63,3% up to 853% with a diagnostic specificity of 974% [19].
DNA analysis For G6PD-deficient (G6PD-) blood donors, the DNA analysis was performed to identify G6PD gene mutations. Genomic DNA was extracted from peripheral blood leucocytes (Qiagen, Dusseldorf, Germany). To characterize the G6PD molecular variants, such as G6PD Mediterranean 563T, G6PD A-202A, 376G, G6PD Seattle 844C and G6PD Union 1370T, which were more frequent in Italy, restriction fragment length polymorphism (RFLP) analysis and allele refractory mutation system (ARMS) analysis were used, in accordance with Martinez di Montemuros, Maffi and Liese [8, 20, 21]. On the negative DNA samples for RFLP or ARMS, the entire coding region (12 exons) was amplified by PCR according to Poggi et al. [22] and a nucleotidic sequence analysis was performed. The NC-000023 genetic reference was employed for the primer design and the sequence alignment.
Statistical analyses The Kolmogorov–Smirnov normality test and Q-Q plot were used to determine whether G6PD, G6PD/6PGD ratio and haematological parameters were normally distributed. Statistical analyses of haematological parameters were performed using the unpaired Student’s t-test and oneway ANOVA.
Results Reference intervals for the G6PD activity were determined on a reference population consisting of 200 male and 200 female donors with normal haematological parameters (Table 1), which were normally distributed. The G6PD
activity and G6PD/6PGD ratio values of the reference population resulted as being normally distributed; therefore, the reference intervals were calculated as parametric centiles 25–975 (Table 1). For G6PD deficiency, 3004 blood donors were screened: 2342 (78%) male donors and 662 (22%) female donors. The high percentage of male donors is a characteristic of the Italian blood donor population. In 2012, the national percentages of male and female donors were 70% and 30%, respectively. All donors declared in the questionnaire that they had never suffered from favism; however, 33 subjects (25 men and 8 women) with low G6PD activity were identified, corresponding to 11% of analysed population sample. None of the 33 subjects was aware of being G6PD deficiency carriers, even if 25 of these were periodic donors (76%). Among the G6PD-deficient donors, the values for G6PD activities for men and women were 027–349 and 310–598 U/g Hb, respectively; and G6PD/6PGD ratios for men and women were 003–063 and 039–083 U/ gHb, respectively (Table 2). The DNA analysis of the G6PD- donors led to the identification of the following class II G6PD molecular variants: (i) Mediterranean 563T in 13 subjects, 9 males and 4 heterozygous females; (ii) Valladolid 406 T in a male subject; (iii) Chatham 1003A in a male donor; iv) Cassano 1347C in an heterozygous female subject. Furthermore, 14 donors were carriers of class III G6PD molecular variants: (i) Seattle 844C variant in 12 subjects, 11 men and a women; (ii) A-202A,376G 1 heterozygous female donor; (iii) Neapolis 1400G, 1 male subject. No G6PD- blood donor carried double mutations in the Gd gene. In 2 subjects, the nucleotide sequence analysis of the Gd gene coding region did not identify any mutation, and in a female donor, it was not possible to perform the DNA analysis. As far as the allelic frequency is concerned, the frequency of class II severe alleles in donors was 48%, while the frequency of class III mild alleles was 43% (Table 2). The G6PD activities of different variants confirmed the very low activity values in male donors with class II variants (T
1003G>A 406C>T 1347G>C
M F M M F
844G>C
202G>A 376A>G 1400C>G
N.I. N.D.
M F F M M F
Alleles
Allelic frequency (%)
G6PD U/gHb
G6PD/6PGD
16 13 9 4 1 1 1 14 12 11 1 1 1 2 1
48 39 27 12 3 3 3 43 36 33 3 3 3 6 3
027–488
003–075
027–076 310–488 061 073 347 152–598
003–010 039–075 008 007 050 024–083
152–325 598 417 333 085/349 518
024–033 083 062 048 012/063 065
N.I., not identified; N.D., DNA analysis not performed.
15–325 U/gHb) and in all heterozygous females (310– 598 U/gHb) (Table 2). The gender differences in G6PD enzyme activity in male versus female carriers of the same mutation are due to the Mendelian X-linked inheritance of G6PD gene. Males are hemizygous for the G6PD gene mutation, and the enzyme activity value is always low. Females may be homozygous or heterozygous for the G6PD gene mutation. Homozygous females have low enzyme activity comparable with the values of hemizygous males, while heterozygous females, due to the random inactivation of one X chromosome (Lyonization phenomenon), are genetic mosaics and can have values of G6PD activity normal, intermediate or deficient depending on the ratio between normal and G6PD-deficient red cells. Accordingly, the G6PD activity values of heterozygous females may be variable ranging from very low to normal values. The heterozygous female donors showed intermediate G6PD activity values (310–598 U/ gHb) (Table 2). The haematological parameters of G6PD- donors were within normal ranges (Table 3); however, the comparison between normal and G6PD- donors showed highly significant differences between the two groups. In G6PD- male donors, increased values in MCV (P < 0001) were found with also decreased values in RBC (P < 0001), Hb (P = 004) and MCHC (P = 001). When the multiple comparison was made between normal donors and G6PDdonors with class II (G6PD
