Increased apoptosis of circulating T cells in myelodysplastic syndromes

Available online at www.sciencedirect.com

Leukemia Research 31 (2007) 1641–1648

Increased apoptosis of circulating T cells in myelodysplastic syndromes Yumiko Shioi a , Hideto Tamura a , Norio Yokose a , Chikako Satoh a,b , Kazuo Dan a , Kiyoyuki Ogata a,∗ a

Division of Hematology, Department of Medicine, Nippon Medical School, Tokyo, Japan b Department of Bioregulation, Nippon Medical School, Kanagawa, Japan

Received 19 October 2006; received in revised form 13 March 2007; accepted 21 March 2007 Available online 7 May 2007

Abstract The mechanism of T cell lymphopenia in myelodysplastic syndromes (MDS) is unknown. We investigated apoptosis in freshly isolated and cultured lymphocytes; the latter were used to detect cells not yet apoptotic but destined for apoptosis. Apoptosis increased in both fresh and cultured T cells in MDS compared with those from healthy controls. Furthermore, in lymphopenic MDS patients the lymphocyte count correlated negatively with the degree of T cell apoptosis. MDS T cells showed increased Fas expression. However, in MDS but not in controls, the degree of T cell apoptosis was independent of the Fas expression level, and exogenous anti-Fas antibodies did not modulate T cell apoptosis. Mechanisms other than the Fas–Fas ligand pathway may induce T cell apoptosis in MDS. © 2007 Elsevier Ltd. All rights reserved. Keywords: Apoptosis; Myelodysplastic syndromes; T cell; Fas–Fas ligand pathway

1. Introduction Myelodysplastic syndromes (MDS) are malignant disorders of hematopoietic cells with poor prognosis that typically occur in the elderly and often transform into acute myeloid leukemia (AML) [1]. In each MDS patient, clonal hematopoietic progenitors show varying degrees of differentiation to myeloid cells. However, mainly due to apoptosis of partially or fully differentiated hematopoietic cells and insufficient differentiation capacity of the progenitors, anemia, neutropenia, and/or thrombocytopenia occur in MDS [2,3]. Although lymphocytes are not involved in a malignant clone in most MDS cases, lymphopenia is a common finding in MDS [4,5]. The immunological competence of hosts is considered to be important to protect against MDS clones, which is supported ∗ Corresponding author. Division of Hematology, Department of Medicine, Nippon Medical School, 1-1-5 Sendagi, Bunkyo-ku, Tokyo 1138603, Japan. Tel.: +81 3 3822 2131; fax: +81 3 5685 1793. E-mail address: [email protected] (K. Ogata).

0145-2126/$ – see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.leukres.2007.03.026

by the following observations. Immunosuppressive therapies increase mature myeloid cells in some MDS cases, however, the increased myeloid cells are thought to be clonal in origin [6]. MDS clones develop in a substantial proportion of patients with aplastic anemia who were immunosuppressed after treatment with anti-thymocyte globulin [7]. Lymphopenia in MDS is mainly due to a decrease in the number of T cells [4,5]. Few studies examined the mechanisms explaining T lymphopenia in MDS. Hamada et al. examined morphologically identifiable apoptosis of total lymphocytes in the peripheral blood (PB) [8]. They showed that, although the difference was statistically insignificant, a higher percentage of MDS lymphocytes exhibited apoptosis compared with normal lymphocytes. Amin et al. examined bone marrow (BM) samples and reported that apoptosis was increased in B cells but not in T cells in MDS [9]. In the present study, using PB samples, we examined T cell apoptosis in freshly isolated cells and cells cultured for 1–4 days; the latter system was used to detect cells preprogrammed in vivo to undergo apoptosis. We showed that T cell apoptosis

1642

Y. Shioi et al. / Leukemia Research 31 (2007) 1641–1648

was increased in both freshly isolated and cultured cells from MDS patients compared with cells from healthy controls. Furthermore, in lymphopenic MDS patients the lymphocyte count correlated negatively with the degree of T cell apoptosis. The increase in T cell apoptosis was more marked in MDS patients who had a high score on the International Prognostic Score System (IPSS) compared with other patients. We also investigated the mechanism of T cell apoptosis in MDS.

2. Materials and methods 2.1. Patients and controls Twenty-three patients with MDS, diagnosed according to the French–American–British (FAB) criteria [10] and treated at our institution, were enrolled in this study. Eight had refractory anemia (RA), 1 refractory anemia with ringed sideroblasts (RARS), 7 refractory anemia with excess blasts (RAEB), and 7 RAEB in transformation (RAEB-t). Their median age was 64.5 (range 29–86) years. Patients with secondary MDS and those who had had infections within 1 month before this study were excluded. Cytogenetic analyses were performed using standard G-banding with trypsinGiemsa staining. Karyotypes were interpreted using the International System for Cytogenetic Nomenclature criteria [11]. The IPSS of each MDS patient was determined according to the report by Greenberg et al. [12]. The control group comprised 25 healthy volunteers. We did not age-match the controls and MDS patients in this study because we confirmed that the degree of T cell apoptosis was not associated with age in preliminary experiments (upper part of Supplemental Table 1). This was also confirmed in MDS patients (see Section 3). The procedures followed were in accordance with the ethical standards of the Institutional Committee on Human Experimentation and with the Helsinki Declaration of 1975, as revised in 1983. 2.2. Cell separation Heparinized PB was obtained from the patients and controls after informed consent had been obtained. PB mononuclear cells (PBMCs) were prepared using FicollHypaque density-gradient centrifugation (Sigma, St. Louis, MO, USA). A portion of the PBMCs was immediately analyzed for apoptosis with flow cytometry (FCM). The remaining cells were cultured in RPMI 1640 medium containing 10% fetal bovine serum (FCS), 1 mM l-glutamine, 100 ␮g/ml penicillin, and 100 ␮g/ml streptomycin (designated complete medium in this paper) at a concentration of 5 × 105 cells/ml. On days 1 and 4 of culture, the cells were analyzed for apoptosis using FCM to detect cells preprogrammed in vivo to undergo apoptosis (spontaneous apoptosis), as described previously [13]. In some experiments, CD3+ lymphocytes were isolated by magnetic cell sorting as described previously [14].

2.3. Flow cytometry Cells were treated with human immunoglobulin to block nonspecific binding before antibody staining. To detect lymphocyte apoptosis, the cells were stained with one of the PE-conjugated mouse monoclonal antibodies (mAbs) to human lymphocyte antigens, fluorescein isothiocyanate (FITC)-conjugated annexin V and propidium iodide (PI) (Trevigen, Gaithersburg, MD, USA), according to the manufacturer’s instructions. To examine Fas (CD95) expression, the cells were stained with one of the FITC-conjugated mouse mAbs to human lymphocyte antigens and PE-conjugated mouse mAb to human CD95 (BD PharMingen, San Diego, CA, USA). To examine Fas ligand (FasL) expression, the cells were successively reacted with purified mouse anti-human FasL antibody (BD PharMingen), second antibody (FITCconjugated anti-mouse IgG), and PE-conjugated mAbs to one of the human lymphocyte and monocyte antigens. The mAbs to human lymphocyte and monocyte antigens included antibodies to human CD3, CD4, CD8, CD14, CD19, CD25, and CD56 (BD PharMingen). FCM was performed with the standard method, as described in our previous report [15]. Single-labeled cells were used to compensate for the fluorescence emission overlap of each fluorochrome into inappropriate channels. Isotype-matched negative controls were used in all assays. Analysis was performed on a FACScan (Becton Dickinson, Mountain View, CA, USA). 2.4. Induction of apoptosis via the Fas–Fas ligand pathway and by activation via CD3 receptor signaling To induce apoptosis through the Fas–FasL pathway, an agonistic anti-Fas mAb, CH11 (MBL, Nagoya, Japan), was added to the cultures of PBMCs or Jurkat cells (Riken Cell Bank, Tsukuba, Japan) at a final concentration of 400 ng/ml. Jurkat cells have been confirmed to be sensitive to Fas-mediated apoptosis [16]. To block apoptosis through the Fas–FasL pathway, an antagonistic anti-Fas mAb, ZB4 (MBL, Nagoya, Japan), was added to the cultures at a final concentration of 2 ␮g/ml. Concentrations of these antibodies were determined to be optimal in preliminary experiments. To examine activation-induced T cell apoptosis, culture dishes were precoated with anti-CD3 mAb (CRIS7) (NeoMarkers, Fremont, CA, USA) according to the method previously described [17]. Then PBMCs were cultured in the dishes for 1 or 2 days. 2.5. Analysis of caspase-3 and -8 activation PBMCs, which had been activated via CD3-mediated signaling for 1 day as described above, were subjected to caspase-3 analysis using a flow cytometric kit (BD PharMingen). In brief, the cells were stained first with FITCconjugated anti-CD3 antibody, washed, and then stained for the active form of caspase-3 according to the manufacturer’s instructions. For caspase-9 analysis, purified CD3+ T cells

Y. Shioi et al. / Leukemia Research 31 (2007) 1641–1648

were activated via CD3-mediated signaling for 1 day and then subjected to the analysis, which was performed using a caspase-9 colorimetric activity assay kit (Chemicon, Temecula, CA, USA) according to the manufacturer’s instructions. We used cell samples activated via CD3-mediated signaling because these assays, especially the caspase-9 assay, are not sufficiently sensitive to detect active caspases in unactivated T cells. 2.6. Statistical analysis Two groups of data with categorical variables were compared using the Chi-square test. Two groups of data with continuous variables were compared using Student’s t-test, the Mann–Whitney U test (unpaired data), and Wilcoxon signed-rank test (paired data). The latter two tests were used if the number of data in each group was less than 20. The correlation between two sets of variables was determined with the linear correlation test. A P-value of less than 0.05 was considered to represent a statistically significant difference.

3. Results 3.1. Increased apoptosis of peripheral CD3+ T cells in MDS We investigated T cell apoptosis in freshly isolated PBMCs and PBMCs that had been cultured in complete medium without any inducers of apoptosis (Fig. 1A). The percentages of annexin V+ cells (designated “apoptotic cells”) in total CD3+ T cells were significantly increased in freshly isolated PBMCs and PBMCs cultured for 1 and 4 days from the 23 MDS patients compared with those from the 25 controls (Fig. 1B). The percentages of annexin V+ PI− cells (designated “early apoptotic cells” according to the previous report [18]) in CD3+ T cells were also increased in freshly isolated and cultured PBMCs from MDS patients compared with those from controls (Fig. 1C). Both CD4+ and CD8+ T cells exhibited nearly the same degree of apoptosis in both the freshly isolated and cultured PBMCs (data not shown). Under the same experimental conditions, apoptosis was similarly increased in CD19+ B cells, but not in CD56+ cells, from MDS patients compared with those from healthy controls (data not shown). As observed in the healthy control group, the degree of T cell apoptosis was not associated with age in MDS patients (lower part of Supplemental Table 1). Seventeen of the above 23 MDS patients had lymphopenia (lymphocytes < 1500/␮L) (Supplemental Table 2). Among these 17 patients, the lymphocyte count in PB correlated negatively with the percentage of annexin V+ CD3+ cells in freshly isolated cells (without reaching statistical significance, P = 0.10, r = 0.41) and with the percentage of annexin V+ PI+ CD3+ cells (cells in late apoptosis) on day 4 of culture (P = 0.008, r = 0.65) (Fig. 2).

1643

3.2. Examination of the Fas–FasL pathway To investigate whether T cell apoptosis in MDS is mediated through the Fas–FasL pathway, the expression of Fas on T cells was determined in freshly isolated PBMCs from 6 controls and 11 MDS patients. The expression level of Fas on CD3+ T cells was significantly increased in MDS patients compared with that in healthy controls (mean ± S.D. of CD3+ Fas+ cells in total CD3+ cells: 33.4 ± 14.5% and 79.2 ± 19.0% in controls and MDS patients, respectively, P = 0.0018). The degree of Fas expression was similar between CD4+ cells and CD8+ cells (79.2 ± 19.0% and 78.2 ± 21.6% of these cell populations, respectively, in MDS patients). Nevertheless, a similar extent of apoptosis of CD3+ T cells was observed in both Fas+ and Fas− T cells from MDS patients (Fig. 3A). In contrast, CD3+ T cell apoptosis was observed much more commonly in Fas+ T cells compared with Fas− T cells from healthy controls (Fig. 3A). Next, we investigated the effects of agonistic and antagonistic anti-Fas mAbs on T cell apoptosis in MDS. We confirmed that an agonistic CH11 mAb induced apoptosis of Jurkat cells, which express high levels of Fas, and this induction of apoptosis was inhibited by an antagonistic anti-Fas mAb (ZB4) (Fig. 3B), as reported previously [19]. Meanwhile, the addition of CH11 mAb induced little apoptosis of T cells from either normal controls or MDS patients (Fig. 3B). Furthermore, the addition of ZB4 mAb tended to inhibit T cell apoptosis slightly in PBMCs from healthy controls but not in those from MDS patients (Fig. 3B). The above data on PBMCs from healthy controls were essentially the same as reported previously [13]. These data cannot formally rule out the contribution of the Fas–FasL pathway to the increased T cell apoptosis in MDS, however, another pathway probably plays a role in the T cell apoptosis in MDS. Examination of FasL expression on lymphocyte subsets and monocytes in fresh PB samples showed that, among the 12 MDS cases examined, FasL expression was detectable on CD4+ T cells from one case (RA) and CD14+ monocytes from another case (RAEB-t). No CD8+ or CD56+ cells from any MDS patient and none of the above cell subsets from six controls had detectable FasL. 3.3. Cell activation and apoptosis We then investigated the contribution of cell activation to T cell apoptosis. First, we examined the association between CD25 expression (a marker of cell activation in vivo) and the extent of T cell apoptosis. In healthy controls, the percentages of apoptotic cells were higher in CD25+ CD3+ T cells compared with CD25− CD3+ T cells in freshly isolated PBMCs, while no such difference was observed in the same cell fraction from MDS patients (Fig. 4A). When PBMCs were cultured in anti-CD3 mAb-coated or control IgG-coated dishes, T cell apoptosis was stimulated by the activation of anti-CD3 antibody in cells from both healthy controls and MDS patients. The T cell apoptosis

1644

Y. Shioi et al. / Leukemia Research 31 (2007) 1641–1648

Fig. 1. Increased T cell apoptosis in PBMCs from MDS patients. (A) An example of analyses of T cell apoptosis. Positivity for annexin V and PI were analyzed in CD3+ cells before (day 0) and on days 1 and 4 of culture. (B and C) Percentages of annexin V+ cells (apoptotic cells, B) and of annexin V+ PI− cells (early apoptotic cells, C) in CD3+ T cells before and after culture. Data are expressed as mean ± S.D. for 25 healthy controls (white bars) and 23 MDS patients (black bars). Data from two MDS patients on day 4 of culture were not included due to insufficient cell number for analysis.

induced was more marked in PBMCs from MDS patients compared with those from healthy controls on day 1 of culture and reached a very similar high level in PBMCs from both MDS patients and healthy controls on day 2 of culture (Fig. 4B).

3.4. Caspase-3 and -8 activation in T cells We then examined the activation of the main molecules involved in apoptosis, capase-3 and -8, in T cells that had been stimulated by the activation of anti-CD3 antibody

Y. Shioi et al. / Leukemia Research 31 (2007) 1641–1648

1645

Fig. 2. Relationship between circulating lymphocyte count and degree of apoptosis. The Y-axis shows the percentage of annexin V+ CD3+ cells in freshly isolated cells (left) and the percentage of annexin V+ PI+ CD3+ cells on day 4 of culture (right).

for 1 day. Consistent with the increased T cell apoptosis in MDS, both molecules showed greater activation in MDS T cells compared with T cells from healthy controls (Fig. 5).

3.5. Association between T cell apoptosis and FAB and IPSS in MDS When examining the data for each FAB subtype, the percentages of apoptotic CD3+ T cells tended to be higher in patients with advanced disease subtypes (Fig. 6A). A statistically significant difference was observed when data for the freshly isolated PBMCs and PBMCs cultured for 1 day were compared between RAEB-t and RA (fresh PBMCs: RAEB-t 27.8 ± 16.5%, RA 10.54 ± 9.4%, P = 0.0279) (PBMCs cultured for 1 day: RAEB-t 39.7 ± 20.3%, RA 20.4 ± 11.3%, P = 0.0372). When examining the data by IPSS category (i.e., Low, intermediate-1 [INT-1], INT-2, and High), the percentages of apoptotic and early apoptotic CD3+ T cells were higher in freshly isolated PBMCs from patients with INT2 and High compared with those from patients with INT-1 and Low (P = 0.0046, Fig. 6B). Similarly, the percentages of apoptotic CD3+ T cells in PBMCs cultured for 1 and 4 days from patients with INT-2 and High tended to be higher than in those from patients with INT-1 and Low (P = 0.0648 and 0.0448, respectively). 4. Discussion

Fig. 3. Examination of the Fas–FasL pathway. (A) The percentage of annexin V+ cells was determined for Fas+ CD3+ and Fas− CD3+ cells in freshly isolated PBMC. Data are expressed as mean ± S.D. for five healthy controls (white bars) and four MDS patients (black bars). NS, not significant. (B) Jurkat cells (shaded bars, n = 3) and PBMCs from nine controls (white bars) and ten MDS patients (black bars) were cultured with or without agonistic (CH11) and antagonistic (ZB4) anti-Fas mAbs. The plus and minus signs at the bottom indicate the presence and absence of antibodies, respectively. Percentages of annexin V+ CD3+ cells were analyzed on days 0 and 1 of culture. Data are expressed as mean ± S.D.

A variety of molecules, cells, and mechanisms are involved in the human immune system and they are tightly controlled in time and space [20]. Therefore, full elucidation of the immunological defects in MDS is challenging. In this study, we focused on the apoptosis of T cells, because lymphopenia in MDS is mainly due to T cell lymphopenia [4,5], and T lymphocytes are major players in tumor immunity. To the best of our knowledge, only one study examined T cell apoptosis in MDS and reported that it was not increased in the BM [9]. The important molecules involved in apoptosis, e.g., Fas, FasL, tumor necrosis factor (TNF)-␣, and TNF-related apoptosis-inducing ligand (TRAIL) have been identified. In MDS, the expression of Fas and FasL was reported to be increased in BM cells [21–23], while both soluble Fas and FasL levels in sera were reported to be normal in MDS [24,25]. In this study, we found that T cell apoptosis in MDS was increased in both freshly isolated PBMCs and PBMCs cul-

1646

Y. Shioi et al. / Leukemia Research 31 (2007) 1641–1648

Fig. 4. Relationship between cell activation and apoptosis. (A) The percentage of annexin V+ cells was determined for CD25+ CD3+ and CD25− CD3+ cells in freshly isolated PBMCs. Data are expressed as mean ± S.D. of five healthy controls (white bars) and four MDS patients (black bars). NS, not significant. (B) PBMCs from six controls (white bars) and ten MDS patients (black bars) were cultured in dishes coated with control IgG or anti-CD3 antibody. Percentages of annexin V+ CD3+ cells were analyzed on days 1 and 2 of culture. Data are expressed as mean ± S.D.

tured without exogenous inducers of apoptosis. The latter, designated spontaneous apoptosis, was shown to be increased in patients with head or neck cancer and melanoma [13,26]. Consistent with the increased T cell apoptosis, activation of caspase-3 and -8 was increased in MDS T cells. Furthermore, there was a negative correlation between the circulating lymphocyte count and the degree of T cell apoptosis in lymphopenic MDS patients. Therefore, it is likely that the increased T cell apoptosis contributes, at least in part, to the development of lymphopenia in MDS. In contrast, it was

reported previously that BM T cells were normal in both number [27] and in the degree of apoptosis [9]. To confirm this inconsistency in T cell apoptosis between PB and the BM, it will be necessary to examine T cell apoptosis simultaneously in both using the same analytical method. We also found that Fas expression on T cells from MDS patients was significantly increased. However, the extent of T cell apoptosis was not associated with the Fas expression level on T cells, and the addition of agonistic and antagonistic anti-Fas antibodies to the cell cultures did not influence T cell apoptosis in MDS. Mechanisms other than the Fas–FasL pathway may thus contribute to T cell apoptosis in MDS. The biological significance of the increased Fas expression on MDS T cells is unclear. Recent data have indicated that Fas and its downstream signalling components, in particular caspase-8, have diverse biological functions and can provide a costimulatory signal to T cells [28,29]. Therefore, increased Fas expression in MDS T cells may reflect the immune response of T cells to MDS clonal cells. In this study, detectable FasL expression on lymphocytes and monocytes in MDS was rare. However, when considering that FasL is a critical molecule, which should be tightly controlled in time and space [30], our data might not reflect the actual FasL expression status in vivo. We also investigated the contribution of cell activation to T cell apoptosis. The percentages of apoptotic cells were higher in CD25+ T cells (cells activated in vivo) compared with CD25− T cells in PBMCs from controls but not in those from MDS patients. Furthermore, when T cells were activated in vitro, a very high level of T cell apoptosis was induced in PBMCs from both healthy controls and MDS patients. We speculate that although cell activation is a potent in vitro inducer of T cell apoptosis in both healthy individuals and MDS patients, another inducer(s) of apoptosis may also function in MDS patients in vivo. TNF-␣-induced apoptosis, which has been considered to contribute to apoptosis of myeloid cells in MDS [22,31,32], is a candidate for such mechanisms and should be investigated in further studies. Finally, among FAB subtypes and risk categories defined by the IPSS, a widely accepted system to predict survival and leukemic transformation, T cell apoptosis was more marked in RAEB-t and higher-risk IPSS categories. Regarding apoptosis of CD34+ cells (clonal blasts) in MDS, it was shown that apoptosis of these cells was prominent in low-grade (earlystage) MDS but less marked in high-grade MDS [33]. We speculate that the changes in apoptotic status, such as the decrease in clonal blasts and increase in T cells, may contribute to disease progression in MDS. We also speculate that in early-stage MDS, T cells may be reacting to MDS clonal cells, which have some degree of differentiation potential, and thus be contributing to cytopenia. Does any causal relationship exist between clonal blasts and increased T cell apoptosis in MDS, especially in advanced stages of MDS? We reported that the level of circulating soluble interleukin-2 receptor (sIL-2R), which is a soluble form of the IL-2R ␣-chain (CD25) and has the potential to neutralize IL-2, is elevated in

Y. Shioi et al. / Leukemia Research 31 (2007) 1641–1648

1647

Fig. 5. Analysis of activation of caspase-3 and -8 in T cells. (A) An example of caspase-3 analysis. (B) The left panel shows percentages of CD3+ T cells expressing the active form of caspase-3 in all CD3+ T cells (white bar, six healthy controls; black bars, eight MDS patients). The right panel shows the caspase-8 activity of CD3+ T cells expressed as amounts of p-nitroaniline produced by 2 × 106 CD3+ T cells in the assays (white bar, five healthy controls; black bars, four MDS patients). Data are expressed as mean ± S.D.

MDS, especially in advanced stages [34,35]. Because the circulating sIL-2R level showed a positive correlation with the marrow blast mass and showed a negative correlation with the circulating T cell number in MDS [34], CD25 expressed on MDS blasts may be a source of circulating sIL-2R and the

increased sIL-2R molecules may suppress T cell proliferation or induce T cell apoptosis by depriving them of IL-2 [36]. In addition, we and others documented previously that de novo AML blasts express functional costimulatory molecules, B7H2 and B7-2, that suppress T lymphocyte function and are associated with disease progression in AML [37,38]. Furthermore, a signal mediated by another costimulatory molecule, B7-H1, was reported to induce T cell apoptosis [39]. We speculate that, similar to de novo AML, MDS blasts may express costimulatory molecules that induce T cell suppression. Further studies are needed to investigate the causal relationship between blasts and increased T cell apoptosis in MDS. In summary, this is the first report to show increased T cell apoptosis in MDS. That apoptosis was more marked in high-risk MDS patients and warrants further studies to clarify the mechanism and pathophysiological role of this phenomenon.

Acknowledgment Contributions: Y.S. performed experiments and analyzed data; H.T. designed the research, analyzed data and wrote the manuscript; N.Y. designed the research and analyzed data; C.S. performed experiments; K.D. performed morphologic analysis; K.O. designed the research, analyzed data and wrote the manuscript.

Appendix A. Supplementary data Fig. 6. T cell apoptosis as a function of FAB subtype and IPSS category. Box-and-whisker plots of data from 23 MDS patients for FAB subtypes (A) and for IPSS categories (B). The horizontal bars in each plot are the 10th, 25th, 50th, 75th, and 90th percentiles of data.

Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.leukres. 2007.03.026.

1648

Y. Shioi et al. / Leukemia Research 31 (2007) 1641–1648

References [1] Yoshida Y, Stephenson J, Mufti GJ. Myelodysplastic syndromes: from morphology to molecular biology. Part I. Classification, natural history and cell biology of myelodysplasia. Int J Hematol 1993;57:87–97. [2] Bogdanovic AD, Trpinac DP, Jankovic GM, Bumbasirevic VZ, Obradovic M, Colovic MD. Incidence and role of apoptosis in myelodysplastic syndrome: morphological and ultrastructural assessment. Leukemia 1997;11:656–9. [3] Raza A, Gezer S, Mundle S, Gao XZ, Alvi S, Borok R, et al. Apoptosis in bone marrow biopsy samples involving stromal and hematopoietic cells in 50 patients with myelodysplastic syndromes. Blood 1995;86:268–76. [4] Bynoe AG, Scott CS, Ford P, Roberts BE. Decreased T helper cells in the myelodysplastic syndromes. Br J Haematol 1983;54:97–102. [5] Hamblin T. Immunologic abnormalities in myelodysplastic syndromes. Hematol Oncol Clin North Am 1992;6:571–86. [6] Sloand EM, Mainwaring L, Fuhrer M, Ramkissoon S, Risitano AM, Keyvanafar K, et al. Preferential suppression of trisomy 8 compared with normal hematopoietic cell growth by autologous lymphocytes in patients with trisomy 8 myelodysplastic syndrome. Blood 2005;106:841–51. [7] Socie G, Henry-Amar M, Bacigalupo A, Hows J, Tichelli A, Ljungman P, et al. Malignant tumors occurring after treatment of aplastic anemia. European Bone Marrow Transplantation-Severe Aplastic Anaemia Working Party. N Engl J Med 1993;329:1152–7. [8] Hamada K, Takahashi I, Matsuoka M, Saika T, Mizobuchi N, Yorimitsu S, et al. Apoptosis of peripheral leukocytes in patients with myelodysplastic syndromes. Rinsho Ketsueki (in Japanese, with English abstract) 1998;39:1079–84. [9] Amin HM, Jilani I, Estey EH, Keating MJ, Dey AL, Manshouri T, et al. Increased apoptosis in bone marrow B lymphocytes but not T lymphocytes in myelodysplastic syndrome. Blood 2003;102:1866–8. [10] Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, et al. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol 1982;51:189–99. [11] Mitelman F. ISCN 1995. An International System for Human Cytogenetic Nomenclature. Basel, Switzerland: Karger; 1995. [12] Greenberg P, Cox C, LeBeau MM, Fenaux P, Morel P, Sanz G, et al. International scoring system for evaluating prognosis in myelodysplastic syndromes [erratum appears in Blood. 1998;91:1100]. Blood 1997;89:2079–88. [13] Saito T, Kuss I, Dworacki G, Gooding W, Johnson JT, Whiteside TL. Spontaneous ex vivo apoptosis of peripheral blood mononuclear cells in patients with head and neck cancer. Clin Cancer Res 1999;5:1263–73. [14] Ogata K, Satoh C, Tachibana M, Hyodo H, Tamura H, Dan K, et al. Identification and hematopoietic potential of CD45- clonal cells with very immature phenotype (CD45-CD34-CD38-Lin-) in patients with myelodysplastic syndromes. Stem Cells 2005;23:619–30. [15] Ogata K, Nakamura K, Yokose N, Tamura H, Tachibana M, Taniguchi O, et al. Clinical significance of phenotypic features of blasts in patients with myelodysplastic syndrome. Blood 2002;100:3887–96. [16] Takahashi S, Sato N, Takayama S, Ichimiya S, Satoh M, Hyakumachi N, et al. Establishment of apoptosis-inducing monoclonal antibody 2D1 and 2D1-resistant variants of human T cell lines. Eur J Immunol 1993;23:1935–41. [17] Tamura H, Dong H, Zhu G, Sica GL, Flies DB, Tamada K, et al. B7-H1 costimulation preferentially enhances CD28-independent T-helper cell function. Blood 2001;97:1809–16. [18] Bacso Z, Everson RB, Eliason JF. The DNA of annexin V-binding apoptotic cells is highly fragmented. Cancer Res 2000;60:4623–8. [19] Kim SG, Jong HS, Kim TY, Lee JW, Kim NK, Hong SH, et al. Transforming growth factor-beta 1 induces apoptosis through Fas ligand-independent activation of the Fas death pathway in human gastric SNU-620 carcinoma cells. Mol Biol Cell 2004;15:420–34.

[20] Bachmann MF, Kopf M. Balancing protective immunity and immunopathology. Curr Opin Immunol 2002;14:413–9. [21] Gersuk GM, Lee JW, Beckham CA, Anderson J, Deeg HJ. Fas (CD95) receptor and Fas-ligand expression in bone marrow cells from patients with myelodysplastic syndrome. Blood 1996;88:1122–3. [22] Gersuk GM, Beckham C, Loken MR, Kiener P, Anderson JE, Farrand A, et al. A role for tumour necrosis factor-alpha, Fas and Fas-Ligand in marrow failure associated with myelodysplastic syndrome. Br J Haematol 1998;103:176–88. [23] Bouscary D, De Vos J, Guesnu M, Jondeau K, Viguier F, Melle J, et al. Fas/Apo-1 (CD95) expression and apoptosis in patients with myelodysplastic syndromes. Leukemia 1997;11:839–45. [24] Shinohara K, Takahahi T, Nawata R, Oeda E. Decreased expression of the Fas ligand on peripheral blood mononuclear cells and undetectable levels of soluble Fas ligand in the serum of patients with aplastic anemia and myelodysplastic syndrome. Am J Hematol 1999;62:124–5. [25] Munker R, Midis G, Owen-Schaub L, Andreff M. Soluble FAS (CD95) is not elevated in the serum of patients with myeloid leukemias, myeloproliferative and myelodysplastic syndromes. Leukemia 1996;10:1531–3. [26] Saito T, Dworacki G, Gooding W, Lotze MT, Whiteside TL. Spontaneous apoptosis of CD8+ T lymphocytes in peripheral blood of patients with advanced melanoma. Clin Cancer Res 2000;6:1351–64. [27] Hilbe W, Eisterer W, Schmid C, Starz I, Silly H, Duba C, et al. Bone marrow lymphocyte subsets in myelodysplastic syndromes. J Clin Pathol 1994;47:505–7. [28] Chun HJ, Zheng L, Ahmad M, Wang J, Speirs CK, Siegel RM, et al. Pleiotropic defects in lymphocyte activation caused by caspase-8 mutations lead to human immunodeficiency. Nature 2002;419:395–9. [29] Newton K, Strasser A. Caspases signal not only apoptosis but also antigen-induced activation in cells of the immune system. Genes Dev 2003;17:819–25. [30] Linkermann A, Qian J, Janssen O. Slowly getting a clue on CD95 ligand biology. Biochem Pharmacol 2003;66:1417–26. [31] Sawanobori M, Yamaguchi S, Hasegawa M, Inoue M, Suzuki K, Kamiyama R, et al. Expression of TNF receptors and related signaling molecules in the bone marrow from patients with myelodysplastic syndromes. Leuk Res 2003;27:583–91. [32] Verhoef GE, De Schouwer P, Ceuppens JL, Van Damme J, Goossens W, Boogaerts MA. Measurement of serum cytokine levels in patients with myelodysplastic syndromes. Leukemia 1992;6:1268–72. [33] Parker JE, Mufti GJ, Rasool F, Mijovic A, Devereux S, Pagliuca A. The role of apoptosis, proliferation, and the Bcl-2-related proteins in the myelodysplastic syndromes and acute myeloid leukemia secondary to MDS. Blood 2000;96:3932–8. [34] Yokose N, Ogata K, Ito T, An E, Tamura H, Dan K, et al. Elevated plasma soluble interleukin 2 receptor level correlates with defective natural killer and CD8+ T-cells in myelodysplastic syndromes. Leuk Res 1994;18:777–82. [35] Ogata K, Yokose N, An E, Kamikubo K, Tamura H, Dan K, et al. Plasma soluble interleukin-2 receptor level in patients with primary myelodysplastic syndromes: a relationship with disease subtype and clinical outcome. Br J Haematol 1996;93:45–52. [36] Yokose N, Ogata K. Plasma soluble interleukin-2 receptors in patients with myelodysplastic syndromes. Leuk Lymphoma 1997;28:171–6. [37] Maeda A, Yamamoto K, Yamashita K, Asagoe K, Nohgawa M, Kita K, et al. The expression of co-stimulatory molecules and their relationship to the prognosis of human acute myeloid leukaemia: poor prognosis of B7-2-positive leukaemia. Br J Haematol 1998;102:1257–62. [38] Tamura H, Dan K, Tamada K, Nakamura K, Shioi Y, Hyodo H, et al. Expression of functional B7-H2 and B7.2 costimulatory molecules and their prognostic implications in de novo acute myeloid leukaemia. Clin Cancer Res 2005;11:5708–17. [39] Tamura H, Ogata K, Dong H, Chen L. Immunology of B7-H1 and its roles in human diseases. Int J Hematol 2003;78:321–8.