Wha-Soon Chung

Mobilization Kinetics of CD34+ Cells in Association with Modulation of CD44 and CD31 Expression during Continuous Intravenous Administration of G‐CSF in Normal Donors

The aim of the present study is to evaluate the kinetics of CD34+ cells and investigate the potential modulation of CD44 and CD31 expression on CD34+ cells during continuous i.v. administration of G‐CSF, thus to elucidate the possible mechanism of peripheral blood progenitor cell (PBPC) mobilization. Fifteen healthy donors were enrolled in this study. G‐CSF (10 μg/kg/day) was administered for four consecutive days through continuous 24‐h i.v. infusion. For measurement of complete blood counts, CD34+ cell levels and their expression of CD44 and CD31, PB sampling was performed immediately before the administration of G‐CSF (steady‐state) and after 4, 8, 24, 48, 72, 96, and 120 h of G‐CSF administration. The percentage and absolute number of CD34+ cells significantly increased at day 3 (0.55 ± 0.09%, 51.12 ± 24.83 × 103/ml) and day 4 (0.47 ± 0.09%, 46.66 ± 24.93 × 103/ml), compared to the steady‐state level (0.06 ± 0.09%, 2.03 ± 5.69 × 103/ml). At day 3 to day 5 following the onset of G‐CSF administration, a strong decrease of CD44 and CD31 expression was observed on mobilized CD34+ cells compared to controls: the relative fluorescence intensity of CD44 and CD31 was, respectively, 50%‐70% and 40%‐90% lower than that of controls. We conclude that continuous i.v. administration of G‐CSF apparently results in more rapid mobilization of CD34+ cells, and downregulation of CD44 and CD31 on CD34+ cells is likely to be involved in the mobilization of PBPC after treatment with G‐CSF.

Pseudoeosinophilia associated with malaria infection determined in the Sysmex XE-2100 hematology analyzer

Hemozoin is known to be an end product of hemoglobin digestion by the malaria parasite. Hemozoin is a birefringent crystal, and thus hemozoin-containing white blood cells (WBCs) may show the atypical light scattering pattern. The purpose of this study was to investigate pseudoeosinophilia associated with malaria infection using a Sysmex XE-2100 hematology analyzer (Sysmex Corporation, Kobe, Japan). The study group included 16 patients with malaria infection. Of these, 38% showed erroneously high eosinophil counts and atypical eosinophil distributions in the WBCs scattergram, which was due to the presence of hemozoin-containing neutrophils. In two patients, their erroneously high eosinophil counts declined as the parasitemia decreased with treatment. In conclusion, hematologists should consider the possibility of pseudoeosinophilia as a result of hemozoin-containing WBCs and confirm the WBC differential count by microscopy in cases of malaria infection.

Distinct patterns of apoptosis in association with modulation of CD44 induced by thrombopoietin and granulocyte‐colony stimulating factor during ex vivo expansion of human cord blood CD34+ cells

The insufficient number of haemopoietic stem cells (HSCs) in cord blood (CB) is the major potential limitation to widespread use of CB for marrow replacement. Cytokine‐mediated ex vivo expansion has been proposed as a means of increasing the number of CB HSCs for transplantation. However, the biology of CB HSCs during cytokine‐mediated ex vivo expansion, such as apoptosis or expression of adhesion molecules, has not yet been elucidated. We have investigated the patterns of apoptosis and CD44 expression on human CB CD34+ cells during ex vivo expansion. CD34+ cells isolated from human CB were cultured in a stroma‐free liquid culture system with thrombopoietin (TPO), flt3‐ligand (FL), stem cell factor (SCF), and/or granulocyte‐colony stimulating factor (G‐CSF). During the culture, for up to 5 weeks, apoptosis was measured by staining with 7‐amino‐actinomycin D (7‐AAD) along with concurrent immunophenotyping of CD34 and CD44 with three‐colour flow cytometry. In the cultures with TPO, an apoptotic fraction with down‐regulated CD44 appeared from the fourth day up to the second week. G‐CSF also induced apoptosis but in a different manner; the apoptotic fraction without down‐regulation of CD44 appeared unremittingly for up to 5 weeks. FL did not induce apoptosis or down‐regulation of CD44. These findings show that apoptosis is indeed involved in the regulation of CB CD34+ cells in ex vivo expansion and the patterns of apoptosis are dependent on the type of cytokines used. The distinct patterns of apoptosis suggest different mechanisms of TPO and G‐CSF in inducing apoptosis, which still remains to be elucidated.

CD34, RAB20, PU.1 and GFI1 mRNA expression in myelodysplastic syndrome.

Myelodysplastic syndrome (MDS) with hypocellular bone marrow (BM) is often difficult to distinguish from aplastic anemia (AA). Furthermore, the diagnosis of MDS with low blast counts and normal karyotype may be problematic. These issues highlight the need for a reliable marker for the diagnosis of MDS. This study was conducted to determine if changes of mRNA expression in any of the four selected genes would be useful markers for differentiation of hypoplastic MDS from AA, and MDS from benign disease, as well as to investigate whether mRNA expressions differ between MDS risk subgroups. Thirty‐five patients diagnosed with MDS, 27 patients with AA and 17 patients with benign diseases were included. The CD34, RAB20, PU.1 and GFI1 mRNA levels were measured by real‐time RT‐PCR. The CD34 mRNA expressions in hypoplastic MDS were higher than those found in AA. PU.1 and GFI1 mRNA expressions were significantly lower in MDS with low blast counts and normal karyotype than those of benign disease. High‐risk MDS showed higher CD34 expressions than those of low‐risk MDS. This study suggests that measurement of CD34 and GFI1 mRNA expressions could be useful as a diagnostic and prognostic marker for MDS.

Myeloid differentiation of human cord blood CD34+ cells during ex vivo expansion using thrombopoietin, flt3‐ligand and/or granulocyte‐colony stimulating factor

We investigated the phenotypic changes of human umbilical cord blood (CB) CD34+ cells during ex vivo expansion using thrombopoietin (TPO), flt3‐ligand (FL), and/or granulocyte‐colony stimulating factor (G‐CSF). During ex vivo expansion of CD34+ cells isolated from human CB for up to 5 weeks, surface expression of molecules on the cultured cells including CD64 (FcγRI), CD32 (FcγRII), CD16 (FcγRIII), CD11b (MAC‐1) and CD18 (β2‐integrin) was analysed by flow cytometry along with simultaneous measurement of apoptosis by 7‐aminoactinomycin D staining method. CD64, CD32 and/or CD18 expressing cells appeared in the cultures both with and without the addition of G‐CSF until the tenth day. However, without G‐CSF, CD16+ fractions did not appear and CD11b+ fractions were not maintained. With G‐CSF, the CD16+ or CD11b+ fractions appeared only from the second week. These results suggest that G‐CSF is necessary for the late stage of myeloid maturation during which CD16 and CD11b are expressed.

Clinical Characteristics of Hemophagocytic Lymphohistiocytosis Related to Kawasaki Disease

It is difficult to predict the prognosis or clinical course of secondary hemophagocytic lymphohistiocytosis (HLH) due to the various underlying causes. The authors analyzed the clinical and laboratory findings and outcomes in patients with HLH who had initially been diagnosed with Kawasaki disease (KD), and evaluated the clinical significance of each factor. Among the 21 patients with HLH, 5 had initially been diagnosed with KD and 16 had other etiologies. A comparative analysis was performed for fever duration, presence of cytopenia, serum ferritin, aspartate aminotransferase (AST), alanine aminotransferase (ALT), triglyceride, fibrinogen, hyponatremia, reactivation, and survival rate in those HLH patients associated with KD (group I) and other causes (group II). In patients in group I, a higher level of reactivation (20%), a lower survival rate (P = .001), higher AST (P = .031) and ferritin (P = .005), and frequent hyponatremia (P = .000) were found compared to patients in group II. Interestingly, patients in group I was older than the average of age of most KD patients. A high index of suspicion on the progression from KD to HLH would be mandatory when the KD patients show elevated AST and ferritin and the presence of hyponatremia, and especially so if the patient is of older age.

Acute promyelocytic leukemia with i(17)(q10) on G-banding and PML/RARA rearrangement by RT-PCR without evidence of PML/RARA rearrangement on FISH.

The cytogenetic hallmark of acute promyelocytic leukemia (APL) is t(15;17)(q22;q21) involving the PML gene on chromosome 15q22 and the RARA gene on chromosome 17q21. Fluorescence in situ hybridization (FISH) can detect almost all PML/RARA rearrangements including complex and cryptic rearrangements [Kohno et al., 2001; Zaccaria et al., 2002; Brockman et al., 2003; Fujita et al., 2003]. Here, we describe a case of APL with the karyotype 46,XX,i(17)(q10)[12]/46,XX[8] without evidence of PML/ RARA rearrangement by FISH, but PML/RARA fusion gene was molecularly confirmed by RT-PCR. To our knowledge, this is the first report of APL with i(17)(q10), lacking evidence of PML/RARA rearrangement on FISH. A 44-year-old female was transferred to our hospital because of presence of some immature cells in peripheral blood. The laboratory findings were as follows: leukocyte count was 1.5 · 10/l, haemoglobin 12.6 g/dl and platelet count 49 · 10/l. On the peripheral blood smear, 5% immature cells were found. The bone marrow aspirate was replaced by hypergranular promyelocytes (82%) with multiple Auer rods. Cytochemistry studies demonstrated that myeloperoxidase, sudan black B and specific esterase were all positive. The bone marrow biopsy showed 80% cellularity. Flow cytometry studies showed that the immature cells were positive for CD13 and CD33 and negative for HLA-DR and CD34. The patient was diagnosed as having APL. For cytogenetic analysis, unstimulated short-term cultures were set up using the bone marrow aspirate. The cytogenetic analysis of bone marrow showed 46,XX,i(17)(q10)[12]/46,XX[8] in the 20 metaphase cells examined (Figure 1). FISH study was performed using PML/RARA dual color, dual fusion probe (32-191013,Vysis, Des Plaines, IL) according to the manufacturer’s instructions. The FISH signals were viewed using a ·100 oil immersion objective along with a dual-pass filter or single-pass filter and analysed independently in a blinded fashion by two persons. Each person analysed 200 nuclei from different areas of the same slide, for a total of 400 nuclei. When a t(15;17) is present, the nuclei typically show two yellow fusion signals: one for PML/RARA fusion on derivative chromosome 15 and one for RARAPML fusion on derivative chromosome 17, along with one red signal and one green signal, reflecting the PML and RARA signals on normal chromosomes 15 and 17, respectively. In the present case, FISH analysis showed two red and three green signals in 55% of interphase cells. Two red signals (PML gene) indicated both normal chromosome 15 and three green signals (RARA gene) were present: one for normal chromosome 17 and two for i(17)(q10) (Figure 2). No yellow fusion signals indicating PML/RARA rearrangement were detected. In addition, no other RARA FISH signals could be detected that might indicate a complex translocation or RARA translocation not involving PML. To identify PML/RARA rearrangement, RT-PCR was performed using bone marrow aspirate. The molecular study showed L-form PML/RARA chimeric transcript. She was treated with combination chemotherapy (idarubicin and Ara-C) and all trans retinoic acid. At the end of induction therapy, her bone marrow aspirate revealed complete remission status, but cytogenetic study showed 10% of metaphase cells bearing i(17)(q10) in 20 metaphase cells examined. The interphase FISH analysis showed three RARA signals in 20% of interphase cells, indicating persistent i(17)(q10). The PML/RARA rearrangement by RT-PCR was still positive. After consolidation chemotherapy, a complete haematological remission was achieved and no evidence of PML/RARA rearrangement was found on cytogenetic study, FISH or RT-PCR. FISH has been a useful tool for detection of PML/ RARA rearrangements in APL including cases without apparent abnormalities of chromosome 15 or chromosome 17 by conventional cytogenetics [Kohno et al., 2001; Zaccaria et al., 2002; Brockman et al., 2003; Fujita et al., 2003; Grimwade et al.,1997]. One report demonstrated that the clinical sensitivity of a commercially available PML/RARA probe using two LETTER TO THE EDITOR INTERNATIONAL JOURNAL OF LABORATORY HEMATOLOGY

In vitro differentiation of natural killer T cells from human cord blood CD34+ cells

Summary. Natural killer T (NKT) cells are involved in innate immune defence and also in the regulation of adaptive immune responses. However, the development of NKT cells in vitro has not been fully characterized and culture conditions have not been fully optimized. In the present study, we found that an NKT cell fraction developed during the in vitro culture of cord blood (CB) CD34+ cells, and this was subsequently characterized both phenotypically and morphologically. CD34+ cells purified from 10 human CB were cultured in the presence of several cytokines and analysed by flow cytometry, light microscopy and electron microscopy. The NKT cell fraction, defined phenotypically (CD3+CD16+CD56+CD94+) as expressing the invariant T‐cell receptor Vα24 and Vβ11, appeared in the CD56hi fractions. Intracytoplasmic staining demonstrated that interferon‐γ and interleukin 4 (IL‐4) were detected in the CD56hi fractions. IL‐15 was essential and, in combination with either flt3‐ligand (FL) or stem cell factor (SCF), was sufficient to induce the development of NKT cells. The phenotype of the NKT cell fraction was CD45RO+CD45RA– and CD4+CD8α+. Morphologically, they were very large, with either round or oval nuclei, moderately condensed chromatins, voluminous weakly basophilic cytoplasm and various cytoplasmic granules such as dense core granules, multivesicular bodies, and intermediate form granules. When CD34+ cells purified from bone marrow (BM) were compared with those from CB, the latter were consistently more efficient at generating CD56hi NKT cell fractions. In conclusion, IL‐15 in combination with FL and/or SCF can induce the differentiation of NKT cells from human CB CD34+ cells.

Ultrastructural and phenotypic analysis of in vitro erythropoiesis from human cord blood CD34 + cells

Erythropoietin (EPO) induces erythropoiesis in vitro as well as in vivo, and the process of erythroid differentiation has been explored phenotypically and morphologically. However, morphological analysis of in vitro erythropoiesis of human hematopoietic progenitor cells at the ultrastructural level has not been reported before. In the present study, we have traced the ultrastructural changes of erythroid differentiation during ex vivo expansion of human cord blood (CB) CD34+ cells in the presence of EPO by electron microscopy (EM), along with concurrent phenotypic analysis. CD34+ cells purified from ten CBs by immunomagnetic selection were cultured in serum-free essential media in the presence of a combination of the several cytokines including EPO, thrombopoietin, flt3-ligand (FL), stem cell factor (SCF), granulocyte colony-stimulating factor, interleukin (IL)-3 and/or IL-11. Phenotypic analysis was performed by flow cytometric analysis for erythroid markers, including glycophorin C (GPC), Kell-related, glycophorin A (GPA), band 3, Lub, and RhD. Ultrastructural analysis was performed by electron-microscopic examination of the cultured cells stained with uranyl acetate and lead citrate. Phenotypic analysis revealed that in the absence of EPO, genuine erythroid fraction expressing the typical pattern of erythroid markers did not appear. The order of the above markers expressed in the cultured cells in the presence of EPO was GPC, Kell-related, GPA, band 3, Lub, and RhD, irrespective of the type of cytokine added. Of the cytokines used in combination with EPO, FL + IL-3 was the most efficient in inducing erythroid differentiation, which was followed by SCF + IL-3. EM examination demonstrated complete process of erythroid development from pronormoblasts to reticulocytes with nuclei having been extruded and mature erythrocytes. These results suggest that morphologically intact erythrocytes could be produced by ex vivo expansion of CB CD34+ cells using EPO.

Megakaryothrombopoiesis during ex vivo expansion of human cord blood CD34+ cells using thrombopoietin.

Thrombopoietin (TPO) is one of the most promising stimulants for ex vivo expansion of haematopoietic stem cells. Previously, we have found that TPO induces a characteristic pattern of apoptosis during ex vivo expansion of human cord blood (CB) CD34+ cells and that the TPO‐induced apoptotic cells belong to megakaryocyte (MK) lineage. In this study, we have examined the maturation of MK and platelet production in association with the TPO‐induced apoptosis. CD34+ cells, purified from human CB, were expanded in serum‐free conditions stimulated with TPO. Apoptosis was confirmed by terminal deoxynucleotidyl transferase‐mediated deoxyuridine triphosphate‐biotin nick end labelling (TUNEL) assay and electron microscopy (EM). Simultaneous measurement of DNA content and immunophenotyping revealed that the cells with higher DNA content (>8u2003N) constituted less than 5% of the CD41+ fractions until day 14, implying premature apoptosis of MKs before full polyploidization. Nevertheless, EM observation showed not only platelet territories but also newly produced platelets in which granules and microfilaments could be identified. Furthermore, flow cytometry demonstrated that the platelet fraction expressed P‐selectin and an activation motif on GPIIb/IIIa recognized by monoclonal antibody PAC‐1 upon stimulation with adenosine diphosphate (ADP). In addition, periodic acid‐Schiff (PAS)‐positive materials and nonspecific esterase activities could be demonstrated. Therefore, it is suggested that platelet production and the accompanying processes, rather than apoptosis only, be hastened during the ex vivo expansion of CB CD34+ cells when using TPO.