Correspondence 815 3 Bruckert P, Kappler R, Scherthan H, Link H, Hagmann F-G, Zankl H. 6 Sambani C, Trafalis DTP, Vessalas G, Politis G, Peristeris P, Double minutes and c-MYC amplification in acute myelogenous Nakopoulou L et al. Trisomy 6 and double minute in a leukemia: are they prognostic factors? Genet case of chronic myelomonocytic leukemia. Cancer Genet Cytogenet Cytogenet 2000; 120: 73–79. 1998; 106: 180–181. 4 Sen S, Sen P, Mulac-Jericevic B, Zhou H, Pirrotta V, Stass SA. 7 Crossen PE, Savage LM, Heaton DC, Morrison MJ. Characterization of Microdissected double-minute DNA detects variable patterns of the C-MYC amplicon in a case of acute myeloid leukemia with double chromosomal localizations and multiple abundantly expressed tran- minute chromosomes. Cancer Genet Cytogenet 1999; 112: 144–148. scripts in normal and leukemic cells. Genomics 1994; 19: 542–551. 8 Asker C, Mareni C, Coviello D, Ingvarsson S, Sessarego M, Origone P 5 Tanke HJ, Wiegant J, van Gijlswijk RPM, Bezrookove V, Pattenier H, et al. Amplification of c-myc and pvt-1 homologous sequences in acute Heetebrij RJ et al. New strategy for multi-colour fluorescence in situ nonlymphatic leukemia. Leuk Res 1988; 12: 523–527. hybridisation: COBRA: COmbined Binary RAtio labelling. Eur J Hum Genet 1999; 7: 2–11.

Reduced expression of H19 in bone marrow cells from chronic myeloproliferative disorders

Leukemia (2002) 17, 815–816. doi:10.1038/sj.leu.2402830 expressed three-fold less H19 (P ¼ 0.001) compared to healthy controls. CMML exhibited a 20-fold lower H19 level than control TO THE EDITOR cases (Po0.0001). When compared with each other, CMPD and From the paternally imprinted and maternally expressed H19 , CMML cases did not differ significantly with regard to H19 an untranslated RNA is generated that exerts growth regulatory expression, (Figure 1). functions during embryogenesis and acts as a potential tumor The function and role of H19 in normal and malignant suppressor in some types of cancer. It is a developmentally hematopoiesis largely remains unknown. A recent study demon- regulated gene and, thus, abundantly expressed in several fetal strated a tight regulation of H19 expression in lineage commitment. tissues, and repressed in most organs after birth.1 It has been The most intriguing data, although obtained in a small-scale study, elucidated that the untranslated RNA retains a high potency of were evidence of reduced H19 expression in precursor cells of antiproliferative effects and acts directly as a tumor suppressor.2 So far, no H19 protein could be identified.2 Little is known about the role of H19 in normal and neoplastic hematopoiesis. In precursor cells of patients with polycythemia vera, a reduced expression of H19 expression in normal hematopoiesis and CMPD H19 could be demonstrated.3 To assess the level of H19 in total bone marrow cells, we quantitatively focused on its expression in reactive lesions and chronic myeloproliferative disorders (CMPD). A total of 59 formalin-fixed, paraffin-embedded (FFPE) bone marrow trephines with histopathologically proven and clinically untreated CMPD (17 polycythemia vera (PV), 11 essential thrombocythemia 1 (ET), 11 chronic myeloid leukemia (CML), 10 idiopathic myelofi- brosis (IMF) in the cellular phase, 10 IMF with demonstrable fiber increase), 12 cases with chronic myelomonocytic leukemia (CMML), and 33 control cases (bone marrow biopsies with no H19 RNA level evidence of a neoplastic disease) were retrieved from the Bone (+/- standard error) Marrow Registry, Institute of Pathology, Medizinische Hochschule Hannover. Total RNA was extracted using an optimized protocol for the extraction of RNA from FFPE archival tissues.5 Total RNA (1 mg), pretreated with RNase-free DNase (RQ1, Promega, USA), was 0 transcribed to the complementary DNA (cDNA) according to the IMF IMF PV ET CML CMML control manufacture’s protocol using SuperScript II RnaseÀ Reverse cellular fibrotic Transcriptase (Invitrogen, Karlsruhe, Germany). Figure 1 A quantitative real-time RT-PCR assay was performed As a first step to assess a potential role of H19 in normal using an ABI PRISM 7700 Sequence Detection System (Applied hematopoiesis and myeloproliferative diseases, we established the Biosystems). Primer and probes for the quantitative analysis of H19 quantitative RNA expression in normal bone marrow cells. In both RNA and the housekeeping gene b-glucuronidase (b-GUS) were H19 normal bone marrow cells and those from reactive states, a forward primer 50-TGGTTGGAGTTGTGGAGACG; H19 reverse pri- relatively constant level could be found without interindividual mer 50-CACTTGCCAAGGTGGCTCA; and H19 probe 50– differences. Compared to the basal expression level in controls, CCTTGAGTCTCAGTACGAGTGTGCGTGAGT. b-GUS forward 0 bone marrow cells from CMPD expressed a significantly lower H19 primer 5 –CTCATTTGGAATTTTGCCGATT, b-GUS reverse primer 0 0 level. PV, ET, and CML cases showed an up to six fold reduced H19 5 –AGTGAAGATCCCCTTTTTA, and b-GUS probe 5 –TGAACAGT- CACCGACGAGAGTGCTGG. The DDCT–method for quantification of expression (Po0.0001), while IMF in the cellular and fibrotic phases 4 H19 RNA expression was used as previously described. In CMPD and CMML cases, significantly reduced H19 RNA levels were found compared to normal bone marrow cells. Both prefibrotic and fibrotic IMF exhibited a three-fold lower level, PV, ET, and CML exhibited a Correspondence: Dr H Kreipe, Institute of Pathology, Medizinische six-fold lower level, and CMML cases exhibited a 20-fold lower level Hochschule Hannover, Carl-Neuberg-Strasse 1, 30625 Hannover, compared to normal bone marrow cells. To illustrate the different Germany; Fax: 0049-511-532-5799 expression levels, the basal expression level of the control group was Received 17 September 2002; accepted 31 October 2002 set to 1.

Leukemia Correspondence 816 patients with PV.3 Expression of several associated with References positive regulation of the cell cycle and proliferation, for example, insulin-like growth factor 2 (IGF-2), seem to be regulated by H19 1 Looijenga LHJ, Verkerk AJMH, de Groot N, Hochberg AA, Oosterhuis RNA itself as a so-called .6 In a previous study, the JW. H19 in normal development and neoplasia. Mol Reprod Dev 1997; relaxation of IGF-2 imprinting and its biallelic expression in normal 46: 419–439. hematopoiesis could be demonstrated.7 This process seems to be 2 Hao Y, Crenshaw T, Moulton T, Newcomb E, Tycko B. Tumour- suppressor activity of H19 RNA. Nature 1993; 365: 764–767. specific for IGF-2, because the imprinting status of H19 in this study 3Nu´neˆz C, Bashein AM, Brunet CL, Hoyland JA, Freemont AJ, Bickle AM was found to be monoallelic and therefore unaffected. The et al. Expression of the imprinted tumour-suppressor gene H19 is tightly consequence of disrupted imprinting of IGF-2 in normal hematopoi- regulated during normal haematopoiesis and is reduced in haemato- esis could be an upregulation of its growth-promot- poietic precursors of patients with the myeloproliferative disease ing effects.7 These effects could be exaggerated in neoplastic polycythaemia vera. J Pathol 2000; 190: 61–68. hematopoietic proliferations. It has been proposed that H19 4 Livak KJ, Schmittgen TD. Analysis of relative data using real-time quantitative PCR and the 2ÀDDC - method. participates in the repression of IGF-2, at least in part, through 6 Methods 2001; 25: 402–408. effects on IGF-2 . Reduced expression of H19 in 5 Bock O, Kreipe H, Lehmann U. One-step extraction of RNA from CMPD therefore could enhance the growth-promoting effects of archival biopsies. Analyt Biochem 2001; 295: 116–117. IGF-2. In most adult tissues, H19 and IGF-2 are reciprocally 6 Wilkin F, Paquette J, Ledru E, Mamelin C, Pollak M, Deal CL. H19 imprinted. Loss of imprinting of IGF-2 has been described in many sense and antisense transgenes modify insulin-like growth-factor-II solid tumors as well as in CML in blast crisis and acute myeloid mRNA levels. Eur J Biochem 2000; 267: 4020–4027. 8,9 7 Morison IM, Eccles MR, Reeve AE. Imprinting of insulin-like growth leukemia. A loss of imprinting (LOI) of IGF-2 is associated with factor 2 is modulated during hematopoiesis. Blood 2000; 96: 3023– abnormal methylation and reduced expression of H19 in Wilms 3028. tumor.10 Both LOI and abnormal methylation could lead to an 8 Rhandawa GS, Cui H, Barletta JA, Strichman-Almashanu LZ, Talpaz M, altered expression of H19 and IGF-2. Our data clearly demonstrate Kantarjian HM et al. Loss of imprinting in disease progression in a significantly reduced expression of H19 as the hallmark in bone chronic myelogenous leukemia. Blood 1998; 91: 3144–3147. 9 Wu H, Weksberg R, Minden MD, Squire JA. Loss of imprinting marrow cells of CMPD compared to non-neoplastic states. Besides of human insulin-like growth factor II gene, IGF2, in acute allowing a better discrimination between CMPD and reactive myeloid leukemia. Biochem Biophys Res Commun 1997; 231: 466– hematopoietic proliferation, reduced H19 expression may play an 472. important role in the pathogenesis and growth regulatory dysfunc- 10 Steenman MJ, Rainier S, Dobry CJ, Grundy P, Horon IL, Feinberg AP. tion in these stem cell disorders. Loss of imprinting of IGF2 is linked to reduced expression and abnormal methylation of H19 in Wilms’ tumour. Nature Genet 1994; Oliver Bock Institute of Pathology, Medizinische Hochschule 7: 433–439. Jerome Schlue´ Hannover, Hannover, Germany Hans Kreipe

Age-related changes of P-glycoprotein-mediated rhodamine 123 efflux in normal human bone marrow hematopoietic stem cells

Leukemia (2003) 17, 816–818. doi:10.1038/sj.leu.2402853 evaluate age-related changes of P-gp function in normal human bone marrow stem cells. TO THE EDITOR Bone marrow samples from 46 individuals were collected (36 donors for transplant, four subjects submitted to thoracotomy for The MDR1 gene encodes the P-glycoprotein (P-gp), an energy- myocardial revascularization who were taking neither verapamil dependent transmembrane efflux pump for a variety of unrelated, nor digoxin, four with iron-deficiency anemia, one recovered from hydrophobic amphipathic drugs.1 It reduces cytoplasmic drug levels megaloblastic anemia, and one with immune thrombocytopenic and its cytotoxic effects, thus leading to multidrug resistance (MDR). purpura). These subjects were divided into four age groups: 8–19 P-gp is expressed by different normal cells, including the years (n ¼ 12), 20–35 years (n ¼ 17), 35–50 years (n ¼ 7), and 50–82 hematopoietic tissue. In hematopoietic stem cells, P-gp is physio- years (n ¼ 10). Subjects came from the same geographic region logically related to cell protection against toxic substances and (Southeastern Brazil) and samples were obtained after informed metabolites.1,2 Moreover, it has the ability to cause the MDR consent. For individuals submitted to myocardial revascularization, phenotype in leukemic cells.3 P-gp expression is a poor prognostic bone marrow fragments were taken immediately after thoracotomy factor in acute leukemia, particularly acute myeloid leukemia and rinsed with phosphate-buffered saline. All other samples were (AML), and is mainly associated with a more immature phenotype obtained by posterior iliac crest aspiration in EDTA-containing tubes and the expression of the CD34 antigen.4 The prevalence of P-gp and mononuclear cells separated by density gradient centrifugation. expression in AML progressively increases with age,5 being very CD34+ cells were purified from bone marrow mononuclear cells by high among elderly patients, but the reason for this finding is not indirect magnetic labeling (Miltenyi Biotec, Auburn, CA, USA), as completely understood.6 The aim of the present study was to previously described.7 P-gp activity was determined by efflux of Rh123 (Sigma), a fluorescent dye that is a substrate for P-gp, as previously described,7 and analyzed with a FACScan flow cytometer (Becton Dickinson, San Jose, CA, USA). The Rh123 Correspondence: RP Falca˜o, Department of Clinical Medicine, efflux index was calculated based on the ratio of mean fluorescence University of Sa˜o Paulo at Ribeira˜o Preto Medical School, Av. Bandeirantes, 3900 - Ribeira˜o Preto, 14049-900 SP, Brazil; Fax: +55 intensity for Rh123 in Rh123+verapamil/Rh123. The Spearman 16 633 1144 nonparametric correlation test followed by linear regression analysis Received 1 March 2002; accepted 15 November 2002 was employed to assess P-gp function changes as a function of age.

Leukemia