Available online at www.annclinlabsci.org Annals of Clinical & Laboratory Science, vol. 44, no. 4, 2014 437 Implication of Gene Expression, F and Hemoglobin E Levels on β-/Hb E Disease Severity

Suwimol Siriworadechkul1, Sumalee Jindadamrongwech1, Suporn Chuncharunee2, and Saranya Aupparakkitanon1

1Department of Pathology and 2Department of Medicine, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok, Thailand

Abstract. One of the factors affecting the degree of severity in β-thalassemia disease is the presence of un- matched α-hemoglobin chains. Thus, the expression levels of globin genes in reticulocytes of β-thalassemia subjects were measured using quantitative RT-PCR, demonstrating that α/β globin mRNA ratio, as well as levels of γ-globin mRNA and Hb F, increased with progressing degree of β globin synthesis defect. The levels of γ-globin mRNA and Hb F could not be directly correlated with severity of β-thalassemia/Hb E disease due to a low statistical power of this analysis. Higher levels of Hb E were present, however, in clini- cally mild patients, as compared to moderately severe β-thalassemia/Hb E subjects. This suggests that in β-thalassemia/Hb E disease, elevation of Hb E level through enhancing correctly spliced βE-globin mRNA offers another approach in ameliorating disease severity. In addition, co-inheritance of α-thalassemia 2 trait in β-thalassemia/Hb E subjects was associated with milder outcome compared with those with the same β-thalassemia genotypes, confirming the notion of the beneficial effect of a more balanced α:β-globin chain ratio.

Key words: Globin gene expression, Hemoglobin E, Hemoglobin F, β-thalassemia, modifying factor.

Introduction Pathogenesis of thalassemia is mainly due to an im- balance of α-and β- globin chain synthesis. With β-Thalassemia is an important autosomal recessive β-thalassemia, the presence of unpaired inherited disease in regions where is or has α-hemoglobin chains results in their precipitation been endemic, causing anemia of variable degrees onto red blood cell membranes, causing oxidative of severity [1,2,3]. Most defects are due to point damage leading to ineffective erythropoiesis and re- or small deletions leading to reduction duced lifespan of circulating effete red cells [11-13]. in or absence of β-globin chain synthesis [4,5]. Heterogeneity in disease severity of β-thal/Hb E Heterozygotes have only a few numbers of micro- disease has been attributed to the types of β- thalas- cytic hypochromic red blood cells but no clinical semia mutations, co-inheritance α-thalassemia, symptoms. The homozygous state causes and persistence of Hb F production [14-17]. In the β-thalassemia major, and compound heterozygosi- expected severe β0-thal/Hb E genotypes, milder ty of β-thalassemia and Hb E (β26 Glu>Lys)(β-thal/ forms have been reported to be associated with Hb E) can result in mild to severe anemia. The lat- more elevated levels of Hb E [15]. However, a sig- ter of which is often comparable to that of nificant negative correlation of age of onset with β-thalassemia major [6-8]. This type of Hb E was determined, but no correlation with Hb β-thalassemia is a major health problem in regions F was found [14]. Although a wide range in α/non where both genotypes are prevalent, such as certain α-globin mRNA ratios was found in β-thal/Hb E regions of , including Thailand patients, there is no difference between mild and [9,10]. severe phenotypes, but interestingly, elevated cor- rectly spliced βE mRNA was detected in one-third Address correspondence to Assist. Prof. Dr. Sumalee Jindadamrongwech, Department of Pathology, Faculty of Medicine, of mild patients, suggesting that proper splicing of Ramathibodi Hospital, Bangkok 10400, Thailand; phone: +662 E 2011076; fax: +662 2011445; e mail: [email protected] β mRNA may contribute to disease outcome [18].

0091-7370/14/0400-437. © 2014 by the Association of Clinical Scientists, Inc. 438 Annals of Clinical & Laboratory Science, vol. 44, no. 4, 2014

Table 1. Primers and probes used in qRT-PCR.

Gene Primera Sequence (5’-3’)b

α-globin HBA1:F GGAGGCCCTGGAGAGGAT HBA1:R CGTGGCTCAGGTCGAAGTG HBA1:P FAM-TGTCCTTCCCCACCACCAAGACCT-BBQ β-globin HBB:F CTGACACAACTGTGTTCACTAGC HBB:R GGTAGACCACCAGCAGCCT HBB:P YAK-CCCACAGGGCAGTAACGGCAGACT-BBQ βE-globin HBE:R AGCCTGCCCAGGGCCTT γ-globin HBG:F GACTTCCTTGGGAGATGCCAC HBG:R ATTTCCCAGGAGCTTGAAGTTCT HBG:P LC610-AGCTTGTCACAGTGCAGTTCACTCAGC-BBQ GAPD GAPD:F GAAGGTGAAGGTCGGAGTC GAPD:R GAAGATGGTGATGGGATTTC GAPD:P LC670-CAAGCTTCCCGTTCTCAGCCT-BBQ

aF: forward primer, R: reverse primer, P: probe. bBBQ: Black Berry quencher, FAM: 6-carboxy-fluorescein, YAK: Yakima yellow, LC610: Light Cycler-Red 610, LC670: Light Cycler-Red 670. GAPD, glyceraldehydes-3-phosphate dehydrogenase.

The involvements of β-thalassemia genotypes, Human Rights Related to Research Involving Human α-thalassemia co-inheritance, imbalance of globin Subjects of Ramathibodi Hospital, and written informed chains synthesis, and Hb F and Hb E levels in consents were obtained from subjects prior to collecting β-thalassemia trait (β-thal trait) and β-thal/Hb E specimens. disease were explored in this study. The α/β ratio, γ α- and β- thalassemia genotyping. DNA was extracted (both Aγ and Gγ), and correctly spliced βE globin from blood leukocytes. β-thalassemia mutations were mRNA levels were determined by quantitative re- identified using multiplex ARMS-PCR as previously de- verse-transcriptase PCR (qRT-PCR). scribed [19]. α- (– – SEA,––THAI,––FIL, –α3.7,–α4.2 deletions and non deletions Hb Constant Materials and Methods Spring and Hb Pakse) were detected using multiplex Gap-PCR and multiplex ARMS-PCR as described pre- Subjects and measurement of hematological parame- viously [20-22]. ters. EDTA-whole blood samples were obtained from the Department of Medicine and Department of Globin mRNA quantitation. Total RNA was extracted Pathology, Faculty of Medicine Ramathibodi Hospital, from reticulocyte fraction separated from blood samples Mahidol University, Bangkok, Thailand. A total of 48 using TRIzol® Reagent (Invitrogen, California, U.S.A.) thalassemia subjects, comprising 28 β-thal/Hb E disease [23]. Contaminating DNA was removed by treatment and 20 β-thal trait, and 10 normal controls were en- with RNase-free DNase (Promega, Madison, WI, USA). rolled. Blood samples were investigated for hematologi- Quantity and purity of RNA were measured spectropho- cal parameters and hemoglobin typing using Sysmex XS- tometrically (WPA Biowave DNA-Isogen Life Science 1000i blood cell counter (Kobe, Japan) and capillary instrument). Expression of α-, β-, βE-, and γ-globin electrophoresis (Sebia, Lisses, France) respectively. mRNAs were measured using a one-step qRT–PCR Hemoglobin typings obtained in addition to hemato- method with TaqMan™ probe system [24,25]. Each re- logical parameters were compatible with their diagnostic action of 20 µl contained 40 ng of total RNA, 0.30-0.38 phenotypes: namely, normal (A2A), β-thalassemia trait µM each primer, 0.1-0.125 µM each probe, and (A2A with Hb A2>3.5%), and β-thalassemia disease (EF, LightCycler® 480 RNA Master Hydrolysis Probes EFA, A2F, or A2FA). The percentage of Hb F or HbE (Roche, Berlin, Germany). The primers and probes used from the Hb typing result was calculated with total he- are shown in Table 1. qRT-PCR was performed in a moglobin to yield the absolute amount in g/dl. The LightCycler® 480 Real-Time PCR System (Roche, study protocol was approved by the Committee on Berlin, Germany) using the following thermocycling Role of Globin Gene Expression, Hb F and E on β-Thalassemia Severity 439

Table 2. Hb typing and β-thalassemia mutations of 48 β-thalassemia subjects.

Subject (n) Hb typing β-thalassemia

β0-thal/Hb E (19)* EF cd41/42 (-TTCT) (n=4), cd17 (A>T ) (n=13), IVSI-1 (G>T ) (n=2) β+(severe)-thal/Hb E (6) EF(A) IVSII-654 (C>T ) β+-thal/Hb E (3) EFA nt-28 (A>G) β-thalassemia trait (20) A2A cd41/42 (-TTCT) (n=8), cd17 (A>T ) (n=2), IVSI-1 (G>T ) (n=1), (high A2) cd27/28 (+C) (n=1), IVSII-654 (C>T ) (n=4), IVSI-5 (G>C) (n=1), nt-28 (A>G) (n=3) *Four cases (3 with cd17 and 1 IVSI-1) co-inherited α-thalassemia 2 trait (3.7-kb deletion). cd, codon. conditions: 3 min at 63°C for reverse transcription, and genotypes of β-thal/Hb E subjects. A remarkably 40 cycles of 15 s at 95°C, 1 min at 60°C, 1 s at 72°C. All higher γ-globin mRNA expression was observed in samples were determined in triplicate, and mRNA levels β+(severe)- (median 119.2, range 39.5-144.9) com- -∆∆CT were quantitated by 2 method relative to glyceral- pared to β+-thalassemia (median 8.9, range 7.0- dehyde-3-phosphate dehydrogenase (GAPD) mRNA 27.1) (p=0.025). However, in β0-thal/Hb E sub- [26,27]. jects the lower median γ-globin mRNA level Statistical analysis. Kruskal-Wallis and Mann-Whitney (median=68.6, with a wide range=3.2-404.6) may U tests were used to evaluate the differences among account for the overall lack of significant differ- groups. Correlations among variables were analyzed by ences among the β-thal/Hb E genotypes. These val- Spearman's correlation analysis, with a p-value <0.05 ues correlated with Hb F amounts (median 4.9, considered statistically significant. range 2.7-5.7 g/dl and median 1.2, range 1.0-1.5 g/dl) for β+(severe)- and β+-thalassemia respectively Results (p<0.01) (Figure 1C).

β-Thalassemia genotypes identified by PCR-based Expression of βE mRNAs (relative to GAPD methods were as follows: β0-thal/Hb E (19 sub- mRNA) showed no significant differences among jects), β+(severe)-thal/Hb E (6), β+-thal/Hb E (3), β0-, β+(severe)- and β+-thal/Hb E groups (Figure andβ-thalassemiatrait(20)(Table2). Concomitant 1D). Amounts of Hb E were significantly welo r in inheritance of heterozygous α-thalassemia 2 (–α3.7) β+(severe)- than in β+-thal/Hb E subjects, but there was found in 4 cases of β0-thal/Hb E, 3 with were not significant differences between β0-thal/ βcodon 17- and 1 with βIVSI-1-globin mutations. Hb E and other β-thal/Hb E genotypes (Figure 1E). α/β mRNA ratios in each genotype of β thal/Hb E were significantly higher than those of normal con- In 4 cases of β0-thal/Hb E with co-inheritance of trols (median of 1.1, range of 0.6-3.7) (Figure 1A). α-thalassemia 2 trait, the range of α/β globin Similar ratios were observed among β-thal/Hb E mRNA ratio (0.8-1.3) was lower than those with- subjects (median 2.3, range 0.8-11.2) and among out α-thalassemia (1.0-11.2). Similarly, the range β-thal trait (median 2.1, range 1.0-4.6). Significant of Hb F amounts was lower in β0-thal/Hb E cases differences of α/β mRNA ratios were apparent with α-thalassemia 2 trait (1.3-2.8 g/dl) than in among β0-, β+(severe)- and β+-thal/Hb E groups. those without α-thalassemia (1.5-5.8 g/dl) (p = 0.016), whereas the range of Hb E amounts was γ-Globin mRNA expression levels (relative to higher in cases with α-thalassemia 2 trait (5.4-7.5 GAPD mRNA) were higher in β-thal/Hb E (me- g/dl) than in those without (1.7-5.9 g/dl) (p<0.01). dian 66.3, range 3.2-404.6) than in normal con- trols (median 0.3, range 0.2-2.9) (Figure 1B), but Clinical severity of β-thal/Hb E cases was classified there is no significant difference between that of using such clinical data including age of first trans- β-thal trait and normal or among the different fusion, transfusion requirement, spleen size, and 440 Annals of Clinical & Laboratory Science, vol. 44, no. 4, 2014 Discussion

Many factors can affect the severity of thalassemia dis- ease [14,15,29]. Imbalance in glo- bin chain synthe- sis is the major cause of thalasse- mic red blood cell pathology, partic- ularly the pres- ence of un- matched α-hemoglobin chains in β-thalassemia. Measurement of the levels of mRNA of each type of globin chains reflects, to some extent, the degree of globin chain production.

Hereditary persis- tence of adult γ-globin produc- tion is able to compensate for Figure 1. Plot of globin gene mRNA expression levels, and Hb E and Hb F amounts in 48 β-thal/ the deficiency in Hb E and β-thal trait and in 10 normal control subjects. (A) α/β-globin mRNA, (B) γ-globin/ β-globin synthesis GAPDH mRNA, (C) Hb F amount, (D) βE-globin/GAPDH mRNA, (E) Hb E amount. Bar in of β-thalassemia box is median; box represents 25% and 75% interval; and the upper and lower bars represent maxi- mum and minimum values. GAPD, glceraldehyde-3-phosphate dehydrogenase. to some extent, thereby ameliorat- ing disease severi- splenectomy [28]. Only 17 cases with adequate ty [30,31]. In β-thal/Hb E disease, the levels of Hb clinical data were available. These cases were divid- E arising from correctly spliced βE-globin mRNA ed into 2 groups, namely, clinically mild (n=6) and also play an important role in governing disease se- moderately severe (11). There were no statistical verity [15,18]. Our study postulated that both Hb differences in the values of α/β- and γ-, βE-globin F and Hb E may co-contribute to decreasing disease mRNAs and in the amounts of Hb F between these severity in β thal/Hb E disease. 2 groups, but the amount of Hb E in the mild group was significantly higher than in the moder- Using qRT-PCR, we examined levels of globin gene ately severe group (p<0.01) (Table 3). However, to mRNA expression in 28 β-thalassemia disease sub- confirm this observation, a higher number of sam- jects, 20 with β-thalassemia trait, and 10 normal ples will be needed for further study of the impact controls. The α/β globin mRNA ratios of β-thal/ of Hb F and Hb E on clinical severity. Hb E and β-thal trait were both higher than those Role of Globin Gene Expression, Hb F and E on β-Thalassemia Severity 441

In summary, among a cohort of 28 β-thal/Hb E Table 3. Median (min-max) of Hb F and Hb E amounts subjects, clinical data were available to classify 17 in 17 clinically mild and moderately severe β-thal/Hb E disease subjects. cases into mild and moderately severe groups, in which the amounts of Hb E, but not Hb F, ap- Median (min.- max.) peared to be a major factor affecting disease severity. Clinical severity (n) Hb F (g/dl) Hb E (g/dl) Whether the latter group would benefit from the Mild (6) 1.4 (1.0-5.7) 5.1 (3.6-7.3) induction of Hb E production remains to be Moderately severe (11) 2.7 (1.5-4.0) 2.7 (1.7-7.5) tested.

Acknowledgements of normal subjects, in agreement with other studies [32,33]. However, among patients with β-thal/Hb This study was supported by a research grant from the Faculty of Medicine Ramathibodi Hospital, Mahidol University. The E syndromes, α/β globin mRNA ratios differed be- authors thank Prof. Prapon Wilairat for correcting the English tween those with different β-thalassemia genotypes, of the manuscript. contrary to previous findings [32]. The median of α/β globin mRNA ratio in β0 genotype was pecu- References liarly lower than those of β+ genotype. This discrep- ancy might be due to differences in turnover rates 1. Rund D, Rachmilewitz E. β-Thalassemia. N Engl J Med 2005;353:1135-1146. of the mutant β-globin mRNA transcripts under 2. Olivieri NF. The β-thalassemias. N Engl J Med the surveillance of nonsense-mediated mRNA de- 1999;341:99-109. 3. Higgs DR, Engel JD, Stamatoyannopoulos G. Thalassemia. cay mechanism [34-37]. For instance, a mutation Lancet 2012;379:373-383. in codon 17 may produce transcripts that can es- 4. Thein SL. The molecular basis of β-thalassemia. Cold Spring cape the NMD mechanism and be detected, thus Harb Perspect Med 2013;3(a011700). 5. Orkin SH. The mutation and polymorphism of the human resulting in an aberrant α/β globin mRNA ratio. β-globin gene and its surrounding DNA. Ann Rev Genet 1984;18:131-171. γ-Globin mRNA levels and Hb F amounts in each 6. Olivieri N, Pakbaz Z, Vichinsky E. Hb E/beta-thalassemia: a common & clinically diverse disorder. Indian J Med Res β-thal/Hb E genotype were changed in a concomi- 2011;134:522-531. tant manner, with no statistical significance. 7. Fuchareon S, Weatherall DJ. The hemoglobin E thalassaemias. + Cold Spring Harb Perspect Med 2012;2(a011734). However, the amount of Hb E was higher in β - 8. Vichinsky E. Hemoglobin E syndromes. Hematology Am Soc thal/Hb E than in β+(severe)/Hb E cases, suggesting Hematol Educ Program 2007:79-83. a more significant role of Hb E over that of Hb F in 9. Fucharoen S, Weatherall DJ. The Hemoglobin E Thalassemias. Cold Spring Harb Perspect Med 2012;2(8). impacting disease severity, at least among the sub- 10. Fuchareon S, Winichagoon P. Clinical and hematologic as- jects in this study. A suppression of γ globin syn- pects of hemoglobin E beta-thalassemia. Curr Opin Hematol thesis and concurrent increase in βE globin levels 2000;7:106-112. 11. Rivella S. Ineffective erythropoiesis and thalassemias. Curr have been reported previously in heavily transfused Opin Hematol 2009;16:187-194. patients [31]. Thus, the unexpectedlywe lo r Hb F 12. Schrier SL. Pathophysiology of thalassemia. Curr Opin 0 Hematol 2002;9:123-126. levels found in β -thal/Hb E genotypes may reflect 13. Schrier SL, Centis F, Verneris M, Ma L, Angelucci E. The role previous history of frequent blood transfusions. of oxidant injury in the pathophysiology of human thalassemi- Additionally induction of Hb F has been used to as. Redox Rep 2003;8:241-245. 14. Panigrahi I, Agarwal S, Gupta T, Singhal P, Pradhan M. ameliorate disease severity in severe β-thalassemia Hemoglobin E-: Factors affecting phenotype. disease [38]. Indian Pediatr 2005;42:357-362. 15. Winichagoon P, Fucharoen S, Chen P, Wasi P. Genetic factors affecting clinical severity in β-thalassemia syndromes.J Pediatr Co-inheritance of an α-thalassemia trait was pres- Hematol/Oncol 2000;22:573-580. ent in 4 cases of β0-thal/Hb E disease. As expected, 16. Oberoi S, Das R, Panigrahi I, Kaur J, Marwaha RK. Xmn1-G gamma polymorphism and clinical predictors of severity of dis- this resulted in a milder phenotype, compared to ease in beta-thalassemia intermedia. Pediatr Blood Cancer individuals with the same β-thalassemia genotypes 2011;57:1025-1028. absent of α-thalassemia. This supports the notion 17. Cao A, Moi P. Genetic modifying factors in beta-thalassemia. Clin Chem Lab Med 2000;38:123-132. that a more balanced globin chain ratio provides a 18. Tubsuwan A, Munkongdee T, Jearawiriyapaisarn N, Boonchoy better benefit to the affected individual than theab- C, Winichagoon P, Fucharoen S, Svasti S. Molecular analysis solute amount of intracellular red cell Hb. of globin gene expression in different thalassaemia disorders: 442 Annals of Clinical & Laboratory Science, vol. 44, no. 4, 2014

individual variation of βE pre-mRNA splicing determine dis- LA. β-globin gene cluster polymorphisms are strongly associ- ease severity. Br J Haematol 2011;154:635-643. ated with severity of Hb E/β0-thalassemia. Clin Genet 19. Thedsawad A, Jindadamrongwech S, Chuncharunee S, 2007;72:497-505. Butthep P. Multiplex ARMS-PCR analysis for nineteen 30. El-Beshlawy A, Hamdy M, El Ghamrawy M. Fetal globin in- β-thalassemia mutations. J Hematol Transfus Med duction in beta-thalassemia. Hemoglobin 2009;33(Suppl 2012;22:31-40. 1):S197-203. 20. Chong SS, Boehm CD, Higgs DR, Cutting GR. Single-tube 31. Rees DC, Porter JB, Clegg JB, Weatherall DJ. Why are hemo- multiplex-PCR screen for common deletional determinants of globin F levels increased in Hb E/β thalassemia? Blood alpha-thalassemia. Blood 2000;95:360-362. 1999;94:3199-3204. 21. Fucharoen S, Sanchaisuriya K, Fucharoen G, Panyasai S, 32. Watanapokasin Y, Winichagoon P, Fuchareon S, Wilairat P. Devenish R, Luy L. Interaction of hemoglobin E and several Relative quantification of mRNA in β-thalassemia/Hb E using forms of alpha-thalassemia in Cambodian families. real-time polymerase chain reaction. Hemoglobin Haematologica 2003;88:1092-1098. 2000;24:105-116. 22. Fucharoen S, Fucharoen G, Fukumaki Y. Simple non-radioac- 33. Chaisue C, Kitcharoen S, Wilairat P, Jetsrisuparb A, Fucharoen tive method for detecting haemoglobin Constant Spring gene. G, Fucharoen S. α/β-Globin mRNA ratio determination by Lancet 1990;335:1527. multiplex quantitative real-time reverse transcription-poly- 23. Chomczynski PA.A reagent for the single-step simultaneous merase chain reaction as an indicator of globin gene function. isolation of RNA, DNA and from cell and tissue sam- Clin Biochem 2007;40:1373-1377. ples. Biotechniques 1993;15:532-537. 34. Peixeiro I, Silva AL, Romao L. Control of human β-globin 24. Caplin BE, Rasmussen RP, Bernard PS, Wittwer CT. The most mRNA stability and its impact on beta-thalassemia phenotype. direct way to monitor PCR amplification for quantification Haematologica 2011;96:905-913. and mutation detection. Biochemica 1999;1:5-8. 35. Romao L, Inacio A, Santos S, Avila M, Faustino P, Pacheco P, 25. Arya M, Shergill IS, Williamson M, Gommersall L, Arya N, Lavinha J. Nonsense mutations in the human β-globin gene Patel HRH. Basic principles of real-time quantitative PCR. lead to unexpected levels of cytoplasmic mRNA accumulation. Expert Rev Mol Diagn 2005;5(2):1-11. Blood 2000;96:2895-2901. 26. Silver N, Best S, Jiang J, Thein SL. Selection of housekeeping 36. Neu-Yilik G, Beate Amthor B, Gehring NH, Bahri S, Paidassi genes for gene expression studies in human reticulocytes using H, Hentze MW, Kulozik AE. Mechanism of escape from non- real-time PCR. BMC Mol Biol 2006;7:33. sense-mediated mRNA decay of human β-globin transcripts 27. Bianchi N, Borgatti M, Gambari R. Quantitative RT-PCR for with nonsense mutations in the first exon. RNA the analysis of expression of α-, β- and γ-globin genes in ery- 2011;17:843-854. throid cells. Minerva Biotec 2003;15:137-144. 37. Danckwardt S, Neu-Yilik G, Thermann R, Frede U, Hentze 28. Sripichai O, Makarasara W, Munkongdee T, Kumkhaek C, MW, Kulozik AE. Abnormally spliced β-globin mRNAs: a Nuchprayoon I, Chuansumrit A, Chuncharunee S, single point mutation generates transcripts sensitive and insen- Chantrakoon N, Boonmongkol P, Winichagoon P, Fucharoen sitive to nonsense-mediated mRNAdecay. Blood S. A scoring system for the classification of β-thalassemia/HbE 2002;99:1811-1816. disease severity. Am J Hematol 2008;83:482–484. 38. Musallam KM, Taher AT, Cappellini MD, Sankaran VG. 29. Ma Q, Abel K, Sripichai O, Whitacre J, Angkachatchai V, Clinical experience with induction therapy in Makarasara W, Winichagoon P, Fuchareon S, Braun A, Farrer patients with β-thalassemia. Blood 2013;121:2199-212.