Candidate gene mutation analysis in idiopathic acquired sideroblastic anemia (refractory anemia with ringed sideroblasts)

Leukemia Research 31 (2007) 623–628

Candidate gene mutation analysis in idiopathic acquired sideroblastic anemia (refractory anemia with ringed sideroblasts) David P. Steensma ∗ , Kathleen A. Hecksel, Julie C. Porcher, Terra L. Lasho Division of Hematology, Department of Medicine, Mayo Clinic, Rochester, Minnesota, USA Received 17 May 2006; received in revised form 2 June 2006; accepted 9 June 2006 Available online 25 July 2006

Abstract Background: For most cases of idiopathic acquired sideroblastic anemia (IASA), the molecular pathogenesis is unknown, despite the consistent morphological signature of abundant pathological ringed sideroblasts with their characteristic iron-engorged mitochondria. Moderately elevated free erythrocyte protoporphyrin (FEP) levels have been described in IASA, suggesting that the activity of ferrochelatase, the enzyme that catalyzes the final step in heme biosynthesis (incorporation of ferrous iron into protoporphyrin), might be diminished in erythroid progenitor cells from IASA patients. Methods: We confirmed FEP elevation in IASA, then pursued a candidate gene approach that included screening the gene encoding ferrochelatase, FECH, for promoter and coding region mutations and mRNA expression changes in bone marrow from 37 patients with IASA. Results: The analytical techniques employed detected mutations in a test cohort of previously undiagnosed patients with biochemical evidence for erythropoietic protoporphyria, a condition resulting from germline mutations in FECH, but somatic missense mutations of FECH and its promoter were not observed in IASA patients. FECH was modestly overexpressed in progenitor cells from patients with IASA, compared with MDS patients without sideroblasts and healthy controls. In addition, we analyzed ABCB7 and PUS1, genes implicated in congenital sideroblastic anemia syndromes, but again found no coding mutations in acquired cases. Conclusion: We conclude that acquired mutations in the factors currently known to cause inherited sideroblastic anemias are uncommon in IASA. © 2006 Elsevier Ltd. All rights reserved. Keywords: Sideroblastic anemia; Ferrochelatase; Erythropoietic protoporphyria; Mutation analysis

1. Introduction The molecular etiology of the common adult-onset forms of refractory anemia associated with pathological ringed sideroblasts (RARS) remains obscure [1]. These idiopathic acquired sideroblastic anemias (IASA), first clearly distinguished by Bjorkman in 1956 [2], are now considered a form of myelodysplastic syndrome (MDS)—a diagnostic label that generally implies clonal chromosomal instability and a risk of transformation to leukaemia [3–6]. However, ∗ Corresponding author at: Mayo Clinic and Mayo Clinic College of Medicine, 200 First Street SW, Rochester, MN 55905, USA. Tel.: +1 507 284 3805; fax: +1 507 266 9277. E-mail address: [email protected] (D.P. Steensma).

0145-2126/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.leukres.2006.06.005

patients with ringed sideroblasts who have a normal marrow karyotype with morphological abnormalities limited to the erythroid lineage (i.e. “pure” sideroblastic anemia, where >15% of erythroid precursors are ringed sideroblasts, and without marked dysplasia involving non-erythroid lineages) have a more indolent course and considerably longer median survival than patients with most other forms of MDS, and they also progress to acute myeloid leukemia much less commonly [7,8]. This distinction suggests the possibility of a unique pathobiology for IASA/RARS, compared to MDS more generally. In contrast to IASA/RARS, the molecular causes of some of the major congenital sideroblastic anemia syndromes are well characterized. Although many cases remain unexplained, inherited sideroblastic conditions usually result

624

D.P. Steensma et al. / Leukemia Research 31 (2007) 623–628

from germline mutations in factors implicated either directly or indirectly in iron–sulfur cluster formation, mitochondrial oxidative phosphorylation, and heme biosynthesis [9,10]. For instance, X-linked sideroblastic anemia (XLSA, Mendelian Inheritance in Man (MIM) #301300) is caused by germline point mutations in ALAS2 (␦-aminolevulinate synthase 2) at Xq11.2, and rare late-onset presentations of XLSA can mimic MDS [11–17]. Other genes mutated in rare inherited sideroblastic anemia syndromes include ABCB7 (ATP-binding cassette, subfamily B, member 7) at Xq13.1 in the X-linked sideroblastic anemia associated with spinocerebellar ataxia (MIM #301310), mitochondrialencoded genes in Pearson marrow-pancreas syndrome (MIM #557000), and PUS1 (pseudouridine synthase-1) at 12q24.33 in mitochondrial myopathy with lactic acidosis and sideroblastic anemia (MIM #600462) [10,18]. These observations, and potential parallels with other MDSassociated erythrocyte phenotypes such as acquired thalassemia [19], raise the possibility that IASA/RARS could be associated with somatic mutations in the same pathways where germline mutations cause inherited sideroblastic anemia. Ferrochelatase (E.C. 4.99.1.1) is of special interest in sideroblastic anemias because it is the enzyme that catalyzes the final step in the heme biosynthetic pathway: insertion of ferrous (Fe2+ ) iron into the macrocyclic protoporphyrin IX ring [20]. Germline mutations in the gene that encodes ferrochelatase, FECH at chromosome 18q21.3, are associated with the inherited disease erythropoietic protoporphyria (EPP; MIM + 177000). EPP was first described by Magnus in 1961 [21], and is characterized by cutaneous photosensitivity beginning in childhood, as well as progressive hepatic injury that can be fatal in some patients [22,23]. Red cell levels of total protoporphyrins and of free erythrocyte protoporphyrin (FEP) are typically markedly elevated in EPP, often >1000 mcg/dL packed red cells (normal range for FEP, 1–10 mcg/dL packed red cells). Occasional patients with EPP have ringed sideroblasts detectable in the bone marrow [24], raising the possibility that FECH might be involved in IASA/RARS. Rare cases of chronic myeloid disorders have also exhibited extremely elevated FEP levels in a range characteristic of EPP, usually with a phenotype of acquired late-onset photosensitivity (i.e. after age 40) [25–30]. More modest FEP elevations without photosensitivity have been reported in several forms of sideroblastic anemias, including IASA/RARS [31–34]. In two unusual cases of chronic myeloid neoplasia associated with very high total plasma protoporphyrins (>1000 mcg/dL), photosensitivity, and (in one case) acute liver injury, a somatic deletion of one allele of the FECH gene was detected [25,29]. The etiology of more modest elevations in FEP in IASA/RARS remains unknown. Enzymatic activity assays of heme biosynthetic factors in IASA/RARS have been unrevealing [35]; however, to our knowledge, genetic screening analyses have not been performed in this group of patients.

We sought to determine whether cases of IASA/RARS are characterized by somatic mutations in the promoter or coding region of FECH. In addition, we screened IASA/RARS patient DNA for acquired mutations in other genes (ABCB7 and PUS1) linked to congenital sideroblastic anemias.

2. Methods 2.1. Patients and DNA/RNA source The study was approved by the Mayo Clinic Institutional Review Board, and patients consented to analysis of their biological material. Governmental guidelines on medical records privacy were followed. We analyzed bone marrow and blood specimens from 37 patients with IASA/RARS, and blood specimens from an analytical test group of four patients with suspected EPP and without a prior molecular diagnosis in their pedigree. Early MDS patients (i.e. refractory anemia without ringed sideroblasts) and healthy donors, including those undergoing hip arthroplasty, served as control group for FECH expression analysis. Mayo Clinic hematopathologists rendered the diagnoses of IASA/RARS using published World Health Organization (WHO) diagnostic criteria [36]. Heavy metal poisoning and exposure toxins and drugs known to be associated with non-clonal sideroblastic anemia (isoniazid, chloramphenicol, cycloserine, ethanol, etc.) were excluded, in so far as possible. IASA patients included both “pure” sideroblastic anemia (RARS), as well as ringed sideroblasts arising in the context of refractory cytopenias with multilineage dysplasia (RCMDRS in the 2001 WHO classification of hematologic neoplasia [4]). All patients had >15% ringed sideroblasts on marrow Prussian blue staining. None of the IASA/RARS patients had chromosome 18q deletions or translocations on routine marrow karyotyping. EPP patients had cutaneous photosensitivity and markedly elevated FEP levels (>10 times the upper limit of the reference range), and were referred to our laboratory by the Mayo Clinic Biochemical Genetics Laboratory. EPP patients additionally were seen by a genetic counselor before molecular analysis, because of the possibility of discovering a germline mutation. In a subset of IASA/RARS patients designated for expression analysis, erythroid precursors were separated from marrow by flow cytometry in a core facility using antibodies to glycophorin A and CD71 (transferrin receptor) conjugated to fluorescein isothiocyanate and phycoerythrin, respectively (Becton Dickinson, San Jose, CA). After ammonium chloride lysis of erythrocytes, genomic DNA was obtained from clinical samples using a resinbased DNA extraction kit (HighPure® DNA Template Preparation Kit, Roche Diagnostics, Mannheim, Germany), and RNA isolated with RNEasy® Mini Kit (QIAgen, Venlo, The Netherlands). We confirmed RNA quality with an Agi-

D.P. Steensma et al. / Leukemia Research 31 (2007) 623–628

lent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA), and generated cDNA using a SuperScript III® RTS First-Strand cDNA Synthesis Kit (Invitrogen, Carlsbad, CA). 2.2. Mutation analysis of FECH, ABCB7, and PUS1 Genomic DNA and cDNA were amplified by polymerase chain reaction (PCR) in a 50 ␮L volume on a PTC-200 Peltier thermocycler (MJ Research, Waltham, MA), and products confirmed by 1% agarose gel electrophoresis. Reagents included GeneAmp® PCR Buffer II (Applied Biosystems, Foster City, CA), at least 1.5 mM MgCl2 (Applied Biosystems), 200 ␮M dNTPs (Roche, Mannheim, Germany), forward and reverse oligonucleotide primers (list available on request; IDT DNA, Coralville, IA), 100 ng of template DNA, and a 5:1 ratio of AmpliTaq Gold® DNA polymerase (total, 1 U; Applied Biosystems) to Pwo DNA polymerase (total, 0.2 U; Roche). For FECH mutation analysis, amplicons covered the entire protein coding region and canonical splice donor–acceptor sites of the FECH gene; oligonucleotide primers for genomic DNA amplification were modified from those used in the Deybach laboratory [37], which were originally developed for denaturing gradient gel electrophoresis. In addition, we analyzed the promoter region of FECH as described [38]. We also sequenced FECH cDNA from a subset of EPP and IASA patients using nested PCR based on the primer sets of Gouya [39] and Bloomer and co-workers [23,40], to rule out the possibility of aberrant spliceoforms due to alterations in distant cis-acting splicing regulatory elements or trans-acting factors. For ABCB7 and PUS1, only genomic DNA was analyzed; amplicons covered the entire protein coding region and canonical splice donor–acceptor sites. We screened amplicons first by denaturing high performance liquid chromatography (DHPLC), with two exceptions: FECH exons 7 and 9, which frequently displayed DHPLC heteroduplex peaks due to the high prevalence of single nucleotide polymorphisms NCBI dbSNP rs536765 and rs536560 at nucleotides 827 and 950, respectively (GenBank RefSeq NM 000140), were taken directly to sequencing. For DHPLC, we used the WAVETM 3500HT DNA Fragment Analysis System with DNASepTM C18 polystyrene-divinylbenzene copolymer chromatography column (Transgenomic, Omaha, NE). PCR products were warmed to 95 ◦ C, then gradient-cooled to room temperature over 30 min to promote heteroduplex formation. As in our previous successful DHPLC analysis of the ATRX gene in MDS [41], amplified DNA was not mixed with wildtype DNA before heteroduplexing because of the admixture of mutant and wild-type clones characteristically found in hematopoietic tissues from patients with chronic myeloid disorders, including IASA/RARS [42]. The expected sensitivity of this analytical method is <5% mutant DNA in a wildtype background [41]. Each amplified sample was injected

625

into the DHPLC column and eluted through a 260 nM photodetector; the concentrations of triethylammonium acetate and acetonitrile buffers were adjusted automatically, as calculated by the WAVE NavigatorTM software package (Transgenomic). All samples were run on at least two different oven and column temperatures, designed to maximize the extent of the amplicon analyzed under partially denatured conditions. Subcloning of PCR products with DHPLC patterns suggestive of heteroduplexed DNA was performed using the pGEM® -T Easy Vector System (Promega, Madison, WI) and DH5␣ competent cells (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions [41]. For each sample with an apparent heteroduplex peak where a mutation was not obvious on initial bidirectional fluorescent dye-chemistry sequencing, we sequenced plasmid DNA inserts from at least 24 blue-white selected subclones, as well as an independently amplified unfractionated amplicon. Sequencing was performed in a core laboratory with the ABI PRISM® BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems) and an ABI PRISM® 3730xl Genetic Analyzer (Applied Biosystems). We analyzed sequence data with SequencherTM v4.5 (Gene Codes, Ann Arbor, MI) and MacVectorTM v8.0 (Accelrys, San Diego, CA). Wherever possible, we confirmed identified mutations with restriction endonuclease digestion of amplicons according to the enzyme manufacturer’s recommendations. 2.3. FECH expression analysis For five patients with IASA, four patients with nonsideroblastic MDS, and five healthy controls, real-time PCR (RT-PCR) was performed using TaqMan® Universal Master Mix, an ABI 7900HT FastTM RT-PCR system, the Hs00164616 m1 FECH FAM multiplex primer-probe set (interrogates exons 1–2 of FECH cDNA), and TaqMan® glyceraldehyde-3-phosphate dehydrogenase (GAPDH) control reagent with JOE probe (all Applied Biosystems). Assays were performed in triplicate, and expression ratios calculated using the 2∧ -CT method [43]. 2.4. Quantification of erythrocyte protoporphyrins Total, zinc-complexed, and free erythrocyte protoporphyrins were quantified for the four EPP patients and the subset of five IASA patients who also underwent FECH expression analysis, using a clinically validated protocol in a Clinical Laboratory Improvement Amendments (CLIA)approved laboratory. Briefly, organic solvent-based extraction of protoporphyrins was followed by fractionation by standard HPLC, and zinc-complexed protoporphyrins (which are elevated in association with iron deficiency and lead poisoning) and free protoporphyrins were separately quantified, as described [44].

626

D.P. Steensma et al. / Leukemia Research 31 (2007) 623–628

3. Results All four of the test group of four suspected EPP patients with cutaneous photosensitivity and very high FEP levels (all >10 times the upper limit of the reference range) were heterozygous for the IVS3 -48C/T FECH polymorphism (rs2272783). This polymorphism is associated with reduced expression of the wild-type FECH allele and with penetrance of the clinical EPP phenotype when co-inherited with a coding mutation, and is present in 8–13% of healthy Western European populations and in most patients with clinical EPP [45–47]. In three of the four EPP patients, we identified germline mutations in FECH that had previously been associated with EPP (Table 1). The fourth patient with clinical signs of EPP, in whom an FECH mutation was not identified, had markedly reduced expression of FECH mRNA (10% of the mean for healthy controls) and was heterozygous for IVS3 -48C/T, but no coding mutation was found; an independent outside genetic laboratory was also unable to detect a coding mutation in this patient. FEP levels were above the normal range on repeated testing in all five patients with RARS/RCMD-RS analyzed (range, 28–115 mcg/dL packed red cells, median 65 mcg/dL; normal range in Mayo Medical Laboratories, 1–10 mcg/dL). (These patients were not receiving regular red cell transfusions.) However, in these 5 patients, as well as 32 others with IASA/RARS where FEP was not measured, we did not identify any mutations in the coding region or the promoter of FECH. Zinc protoporphyrin levels were within the reference range, or only very slightly above. Synonymous SNPs in exons 7 (rs536765) and 9 (rs536560) of FECH were detected in 19% and 44% of the IASA/RARS patients (NCBI dbSNP values for heterozygosity in the general population, 39.8% and 40.2%, respectively; p-values for Fisher’s exact test of the differences, not significant). The prevalence of the IVS3 -48C/T polymorphism (rs2272783) in the IASA/RARS patients was 9%, exactly as expected for this group of patients, almost all of whom were of European descent. Three apparently novel intronic SNPs were identified in locations well outside of the canonical splicing recognition sites: IVS8 +34C/T (6%), IVS8

Table 1 Mutations in FECH detected in patients with erythropoietic protoporphyria (EPP) Gene mutation (GenBank accession NM 000140)

Predicted protein consequence (NP 000131)

Reference

c.429delA (exon 4) c.69delG (exon 1) Del(18q1218q23) with loss of promoter and exons 1–3

p.I134Y fs144X p.A14P fs72X Null

[56] [57] Similar to [25]

All patients had IVS3 -48C/T heterozygosity. In a fourth patient with biochemical evidence of EPP, no mutation was identified. Nomenclature used to describe mutations is that of den Dunnen and Antonarakis [58].

-61delG (16%), and IVS9 -59delA (6%). These SNPs were present in a similar proportion of control DNA from healthy individuals, and are likely of no significance. Splicing abnormalities of FECH were also not identified, and we detected no coding mutations in ABCB7 or PUS1 in IASA/RARS patients. FECH mRNA expression in IASA/RARS patients was moderately increased compared to MDS patients without sideroblasts and to healthy controls (FECH expression compared to mean for normals: for IASA/RARS—2.10-fold increase; MDS without rings—1.87-fold increase). These results are almost identical to a pre-publication manuscript that appeared after our analysis was completed, reporting global gene expression microarray data from CD34+ cells in MDS patients [48]. In that report, FECH expression in RARS was 2.20-fold higher than healthy controls, and expression in refractory anemia (RA) was 1.35-fold elevated compared with controls; more generally, up-regulation of mitochondrial-related genes involved in heme biosynthesis was seen [48].

4. Discussion This analysis demonstrates that somatic missense and splicing mutations of ferrochelatase are not common in patients with IASA/RARS, and that expression of ferrochelatase in progenitor cells is increased compared to healthy controls, despite the moderately elevated FEP levels commonly observed in these disorders. Additionally, other candidate genes that have been identified as disrupted in inherited sideroblastic anemia syndromes (specifically, ABCB7 and PUS1) are also not commonly mutated in IASA/RARS. We did not analyze mitochondrial DNA in this study (mutated in Pearson marrow-pancreas syndrome), because acquired heteroplasmic mitochondrial DNA mutations had already been sought in patients with MDS, including RARS, by several groups, and such mutations were found to be uncommon [49,50]. Likewise, following the first report (in 1995) of late-onset presentation of sideroblastic anemia associated with germline ALAS2 mutations [11], this gene has been analyzed in IASA/RARS cases by several other groups of investigators, and mutations are uncommon (to our knowledge, only one somatic mutation has been identified [51]), so we did not repeat that analysis here. While it is possible that a small subset of IASA/RARS patients in our cohort might have had ALAS2 or mitochondrial DNA mutations, this finding would not change our conclusion that the molecular pathology of acquired sideroblastic anemia is likely to be distinct from that associated with the inherited syndromes. The elevated FEP levels in IASA suggested to us that either the activity of ferrochelatase might be defective in IASA erythroid precursors, or else that the redox state or suborganellar localization of elemental iron is aberrant in these

D.P. Steensma et al. / Leukemia Research 31 (2007) 623–628

conditions. Either of these possibilities could result in failure of ferrochelatase to incorporate ferrous iron into assembled protoporphyrin IX molecules, with resultant elevation in FEP. Although enzymatic activity assays that included heme synthase (ferrochelatase) had previously been reported as unrevealing [35], more recent descriptions of somatic deletions of FECH in patients with clonal myeloid disorders [25,29] suggested that this question should be revisited with molecular techniques. The present data would appear to exonerate ferrochelatase in most IASA/RARS cases, and attention should now focus on other factors, including other aspects of intracellular iron transport and metabolism. Elevated expression of ferrochelatase may represent a consequence of the iron accumulation or mitochondrial injury characteristic of IASA/RARS, rather than a cause, and appears to be part of a more general overexpression of heme-biosynthesis-related genes in these conditions [48]. Most of the iron present in the pathological mitochondria in ringed sideroblasts in both IASA/RARS and XLSA was recently found to be complexed with mitochondrial ferritin (MtF), which is encoded by the intronless FTMT gene at chromosome 5q21.3 and is localized to the mitochondrial matrix [52]. Remarkably, subsequent analyses showed that overexpression of MtF in sideroblastic anemias actually begins at an early stage of erythroid differentiation, well before the onset of mitochondrial iron accumulation [53]. This finding suggests that MtF accumulation may be an important contributor to the mitochondrial defect in RARS, rather than a later consequence of the accretion of potentially reactive iron species. Additionally, MtF expression is associated with increased expression of the transferrin receptor, possibly reflecting cellular sensing of a functional deficit of usable iron. In RARS, MtF and iron accumulation are associated with subsequent dissipation of the mitochondrial membrane potential and excessive apoptosis of developing erythroid cells [54], but not with a deficit in mitochondrial ATP production [53], thereby distinguishing ineffective erythropoiesis due to RARS from that associated with refractory anemia or other forms of bone marrow failure. Unlike other forms of MDS, global microarray expression analyses have not yet been reported in RARS [55]; such studies, if focused on early erythroid precursors and even more primitive hematopoietic cells, might illuminate other critical components of the pathobiology of acquired sideroblastic states. The combined phenotype of perinuclear mitochondrial iron accumulation, elevated FEP, overexpression of FECH and other genes, loss of the mitochondrial membrane potential, and excessive apoptosis may all be late markers of an early, as yet unknown pathogenic molecular event.

Conflict of interest The authors have no potential conflicts of interest to disclose.

627

Acknowledgements We thank Scott H. Kaufmann, PhD, MD for review of the manuscript and advice, and Joseph P. McConnell, PhD, for referral of patients with suspected EPP. This work was supported by award K12 CA90628 from the National Cancer Institute and a grant from the Robert A. Kyle Hematological Malignancies Fund, both to DPS, and an American Society of Hematology Medical Student award to KAH.

References [1] Alcindor T, Bridges KR. Sideroblastic anemias. Br J Hematol 2002;116(4):733–43. [2] Bjorkman SE. Chronic refractory anemia with sideroblastic bone marrow: a study of four cases. Blood 1956;11:250–9. [3] Bennett JM, et al. Proposals for the classification of the myelodysplastic syndromes. Br J Hematol 1982;51(2):189–99. [4] Vardiman JW, Harris NL, Brunning RD. The World Health Organization (WHO) classification of the myeloid neoplasms. Blood 2002;100(7):2292–302. [5] Yoshida Y, et al. Diagnostic and biological significance of sideroblastic erythropoiesis in the myelodysplastic syndromes. Int J Hematol 1998;67(2):137–44. [6] Cazzola M. Ring sideroblasts and myelodysplastic syndromes. J Clin Pathol 1983;36(12):1413–4. [7] Garand R, et al. Heterogeneity of acquired idiopathic sideroblastic anemia (AISA). Leuk Res 1992;16(5):463–8. [8] Gattermann N, Aul C, Schneider W. Two types of acquired idiopathic sideroblastic anemia (AISA). Br J Hematol 1990;74(1):45–52. [9] Bottomley SS. Congenital sideroblastic anemias. Curr Hematol Rep 2006;5(1):41–9. [10] Fleming MD. The genetics of inherited sideroblastic anemias. Semin Hematol 2002;39(4):270–81. [11] Cotter PD, et al. Late-onset X-linked sideroblastic anemia. Missense mutations in the erythroid delta-aminolevulinate synthase (ALAS2) gene in two pyridoxine-responsive patients initially diagnosed with acquired refractory anemia and ringed sideroblasts [see comment]. J Clin Invest 1995;96(4):2090–6. [12] Cazzola M, et al. Familial-skewed X-chromosome inactivation as a predisposing factor for late-onset X-linked sideroblastic anemia in carrier females. Blood 2000;96(13):4363–5. [13] Harigae H, et al. A novel mutation of the erythroid-specific gamma-aminolevulinate synthase gene in a patient with noninherited pyridoxine-responsive sideroblastic anemia. Am J Hematol 1999;62(2):112–4. [14] Cortesao E, et al. Onset of X-linked sideroblastic anemia in the fourth decade. Hematologica 2004;89(10):1261–3. [15] Bekri S, et al. A promoter mutation in the erythroid-specific 5aminolevulinate synthase (ALAS2) gene causes X-linked sideroblastic anemia. Blood 2003;102(2):698–704. [16] Furuyama K, et al. Late-onset X-linked sideroblastic anemia following hemodialysis. Blood 2003;101(11):4623–4. [17] Cotter PD, et al. Four new mutations in the erythroid-specific 5aminolevulinate synthase (ALAS2) gene causing X-linked sideroblastic anemia: increased pyridoxine responsiveness after removal of iron overload by phlebotomy and coinheritance of hereditary hemochromatosis. Blood 1999;93(5):1757–69. [18] Furuyama K, Sassa S. Multiple mechanisms for hereditary sideroblastic anemia. Cell Mol Biol (Noisy-le-grand) 2002;48(1):5–10. [19] Steensma DP, Gibbons RJ, Higgs DR. Acquired alpha-thalassemia in association with myelodysplastic syndrome and other hematologic malignancies. Blood 2005;105(2):443–52.

628

D.P. Steensma et al. / Leukemia Research 31 (2007) 623–628

[20] Nakahashi Y, et al. The molecular defect of ferrochelatase in a patient with erythropoietic protoporphyria. Proc Natl Acad Sci USA 1992;89(1):281–5. [21] Magnus IA, et al. Erythropoietic protoporphyria. A new porphyria syndrome with solar urticaria due to protoporphyrinaemia. Lancet 1961;2:448–51. [22] Todd DJ. Erythropoietic protoporphyria. Br J Dermatol 1994;131(6):751–66. [23] Bloomer J, et al. Molecular defects in ferrochelatase in patients with protoporphyria requiring liver transplantation. J Clin Invest 1998;102(1):107–14. [24] Rademakers LH, et al. Accumulation of iron in erythroblasts of patients with erythropoietic protoporphyria. Eur J Clin Invest 1993;23(2):130–8. [25] Aplin C, et al. Late-onset erythropoietic porphyria caused by a chromosome 18q deletion in erythroid cells. J Invest Dermatol 2001;117(6):1647–9. [26] Takeda Y, et al. Erythropoietic protoporphyria without cutaneous photosensitivity and with ringed sideroblasts in an atomic bomb survivor. Lancet 1996;347(8998):395–6. [27] Shirota T, et al. Myelodysplastic syndrome terminating in erythropoietic protoporphyria after 15 years of aplastic anemia. Int J Hematol 2000;72(1):44–7. [28] Lim HW, et al. Photosensitivity, abnormal porphyrin profile, and sideroblastic anemia. J Am Acad Dermatol 1992;27(2 Pt 2):287–92. [29] Goodwin RG, et al. Photosensitivity and acute liver injury in myeloproliferative disorder secondary to late-onset protoporphyria caused by deletion of a ferrochelatase gene in hematopoietic cells. Blood 2006;107(1):60–2. [30] Rothstein G, Lee R, Cartwright GE. Sideroblastic anemia with dermal photosensitivity and greatly increased erythrocyte protoporphyrin. New Engl J Med 1969;280(11):587–90. [31] Takeda Y, et al. Acquired hypochromic and microcytic sideroblastic anemia responsive to pyridoxine with low value of free erythrocyte protoporphyrin: a possible subgroup of idiopathic acquired sideroblastic anemia (IASA). Br J Hematol 1995;90(1):207–9. [32] Lee GR, Cartwright GE, Wintrobe MM. The response of free erythrocyte protoporphyrin to pyridoxine therapy in a patient with sideroachrestic (sideroblastic) anemia. Blood 1966;27(4):557–67. [33] Bottomley SS. Porphyrin and iron metabolism in sideroblastic anemia. Semin Hematol 1977;14(2):169–85. [34] Romslo I, et al. Sideroblastic anemia with markedly increased free erythrocyte protoporphyrin without dermal photosensitivity. Blood 1982;59(3):628–33. [35] Pasanen AV, et al. Haem biosynthesis in refractory sideroblastic anemia associated with the preleukaemic syndrome. Scand J Hematol 1981;27(1):35–44. [36] Jaffe ES, et al. WHO classification: tumours of haematopoietic and lymphoid tissues. Lyon: International Agency For Research on Cancer (IARC) Press; 2001. [37] Rufenacht UB, et al. Rapid molecular diagnosis of erythropoietic protoporphyria among Swiss patients. Clin Chem Lab Med 1998;36(10):763–5. [38] Di Pierro E, et al. A point mutation affecting an SP1 binding site in the promoter of the ferrochelatase gene impairs gene transcription and causes erythropoietic protoporphyria. Exp Hematol 2005;33(5):584–91.

[39] Gouya L, et al. Mutations in the ferrochelatase gene of four Spanish patients with erythropoietic protoporphyria. J Invest Dermatol 1998;111(3):406–9. [40] Chen FP, et al. Ferrochelatase gene mutations in erythropoietic protoporphyria: focus on liver disease. Cell Mol Biol (Noisy-le-grand) 2002;48(1):83–9. [41] Steensma DP, et al. Acquired somatic ATRX mutations in myelodysplastic syndrome associated with alpha thalassemia (ATMDS) convey a more severe hematologic phenotype than germline ATRX mutations. Blood 2004;103(6):2019–26. [42] Steensma DP, Tefferi A. The myelodysplastic syndrome(s): a perspective and review highlighting current controversies. Leuk Res 2003;27(2):95–120. [43] Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 2001;29(9):e45. [44] Ford RE, Ellefson RD. Quantitation of erythrocyte protoporphyrins and zinc protoporphyrin by high-pressure liquid chromatography. Clin Chem 1984;30:972. [45] Gouya L, et al. The penetrance of dominant erythropoietic protoporphyria is modulated by expression of wildtype FECH. Nat Genet 2002;30(1):27–8. [46] Risheg H, Chen FP, Bloomer JR. Genotypic determinants of phenotype in North American patients with erythropoietic protoporphyria. Mol Genet Metab 2003;80(1–2):196–206. [47] Whatley SD, et al. Autosomal recessive erythropoietic protoporphyria in the United Kingdom: prevalence and relationship to liver disease. J Med Genet 2004;41(8):e105. [48] Pellagatti A, et al. Gene expression profiles of CD34+ cells in myelodysplastic syndromes: involvement of interferon stimulated genes and correlation to FAB subtype and karyotype. Blood 2006. [49] Shin MG, et al. Mitochondrial DNA mutations in patients with myelodysplastic syndromes. Blood 2003;101(8):3118–25. [50] Gattermann N, Aul C, Schneider W. Is acquired idiopathic sideroblastic anemia (AISA) a disorder of mitochondrial DNA? Leukemia 1993;7(12):2069–76. [51] May A, Al-Sabah AI, Lawless SL. Abstract 3542: acquired sideroblastic anemia unresponsive to pyridoxine caused by a somatic mutation in ALA synthase 2. Blood 2005;106(11). [52] Cazzola M, et al. Mitochondrial ferritin expression in erythroid cells from patients with sideroblastic anemia. Blood 2003;101(5):1996–2000. [53] Tehranchi R, et al. Aberrant mitochondrial iron distribution and maturation arrest characterize early erythroid precursors in low-risk myelodysplastic syndromes. Blood 2005;106(1):247–53. [54] Matthes TW, et al. Increased apoptosis in acquired sideroblastic anemia. Br J Hematol 2000;111(3):843–52. [55] Pellagatti A, et al. Gene expression profiling in the myelodysplastic syndromes using cDNA microarray technology. Br J Hematol 2004;125(5):576–83. [56] Wang X, et al. Erythropoietic protoporphyria: four novel frameshift mutations in the ferrochelatase gene. J Invest Dermatol 1997;109(5):688–91. [57] Todd DJ, et al. Identification of a single base pair deletion (40 del G) in exon 1 of the ferrochelatase gene in patients with erythropoietic protoporphyria. Hum Mol Genet 1993;2(9):1495–6. [58] den Dunnen JT, Antonarakis SE. Nomenclature for the description of human sequence variations. Hum Genet 2001;109(1):121–4.