• MDS;
  • TNF;
  • Fas/Fas-L;
  • haemopoiesis


  1. Top of page
  2. Abstract
  6. Acknowledgements
  7. References

Apoptosis of haemopoietic cells in the marrow of patients with myelodysplastic syndrome (MDS) has been suggested as a mechanism for peripheral cytopenias. We determined the expression of Fas (CD95), Fas-Ligand (Fas-L) and TNF-α factors known to be involved in apoptosis, in the marrow of 44 patients with MDS and characterized their functional relevance in in vitro assays of haemopoiesis. Multidimensional flow cytometry revealed phenotypically aberrant blasts as defined by orthogonal light scatter and CD45 expression in the marrow of 24/44 patients. Among those blasts Fas expression was increased on CD34-positive cells and on cells co-expressing HLA-DR. In addition, Fas-L was expressed on some CD34+ cells of MDS patients but was never detected on CD34+ cells in normal marrow. Fas and Fas-L mRNAs as well as mRNA for TNF-α, known to increase Fas expression in normal marrow, were up-regulated in patients with MDS. TNF-α protein and sTNF-R1 levels in marrow plasma were higher in MDS patients than in controls (P < 0.002 and <0.003, respectively). However, results were dependent upon disease category: TNF-α levels were significantly higher in patients with refractory anaemia (RA) than in patients with RA with excess blasts (RAEB) or RAEB in transformation (RAEB-T) (P = 0.043). Conversely, the proportion of Fas-L-positive cells was lowest in patients with RA (P = 0.037). In marrow cultures, Fas-Ig, rhuTNFR:Fc or anti-TNF-α antibody, by blocking Fas or TNF mediated signals, respectively, significantly increased the numbers of haemopoietic colonies compared to untreated cells (P < 0.001, P < 0.003, P < 0.001, respectively). These results show significant dysregulation in the expression of TNF-α, Fas and Fas-L in the marrow from MDS patients. Altered expression of these molecules appears to be of functional relevance in the dysregulation of haemopoiesis in MDS and may be amenable to therapeutic interventions.

Myelodysplastic syndrome (MDS) comprises a group of disorders characterized by malignant transformation of early haemopoietic precursors and clonal proliferation and differentiation ( Heyman, 1991; Mayer & Canellos, 1996). The marrow is frequently normo- or hypercellular, whereas the peripheral blood shows cytopenia involving one or several lineages. These parameters may vary between the different subcategories of MDS which include refractory anaemia (RA), RA with ring sideroblasts (RARS), RA with excess blasts (RAEB) and RAEB in transformation (RAEB-T) ( Greenberg et al, 1997 ). Although MDS has been recognized as a complication of previous radiochemotherapy ( Heyman, 1991; Mayer & Canellos, 1996), most cases occur ‘spontaneously’ (primary MDS), often in older patients ( Mayer & Canellos, 1996). Clinical presentation and course may vary considerably between patients.

The pathogenesis of MDS is poorly understood ( Noel & Solberg, 1992; Heyman, 1991; Mayer & Canellos, 1996). It is likely that more than one pathway is involved. Recent studies have shown an increased rate of apoptosis (programmed cell death) in MDS marrow, either due to a lack of positive signals (growth factors) or an up-regulation of negative signals ( Smith & Smith, 1991; Ganser et al, 1991 ) such as tumour necrosis factor (TNF) alpha or Fas and its ligand ( Smith et al, 1994 ; Nagata & Golstein, 1995; Raza et al, 1995 ). Fas is a 45 kD type I membrane protein belonging to the TNF-α/nerve growth factor receptor family, and is expressed in various tissues including thymus, liver, heart and kidney ( Watanabe-Fukunaga et al, 1992 ; Smith et al, 1994 ; Nagata & Golstein, 1995). Fas mediates apoptosis following interaction with agonistic anti-Fas monoclonal antibody (mAb) (e.g. mAb CH-11) or its natural ligand, Fas-L ( Vignaux et al, 1995 ; Suda et al, 1995 , 1993; Watanabe-Fukunaga et al, 1992 ). Fas-L is a 40 kD type II transmembrane protein functionally related to TNF-α ( Suda & Nagata, 1994). Fas-L is prominently expressed in activated T cells ( Suda et al, 1995 , 1993), and more recently has been shown to be expressed on myeloid cells and tumour cells ( O'Connell et al, 1996 ; Badley et al, 1996; Hahne et al, 1996 ; Kiener et al, 1997 ; Liles et al, 1996 ).

Although interactions of Fas with Fas-L have been shown to play a role in the regulation of T- and B-cell development ( Rathmell et al, 1995 ; Dhein et al, 1995 ; Krammer et al, 1994 ; Alderson et al, 1995 ), a role in haemopoiesis is less well defined. Haemopoietic precursors in many patients with MDS express Fas at abnormally high levels ( Gersuk et al, 1996 ; Maciejewski et al, 1995a ; Hatake et al, 1994 ). There is also an increase in Fas-L expression on MDS marrow cells ( Gersuk et al, 1996 ). Since Fas-mediated signals play a central role in apoptosis ( Nagata & Golstein, 1995) and Fas signals are triggered by Fas-Ligand, it is conceivable that Fas-mediated apoptosis is involved in the haemopoietic failure seen in patients with MDS. Intramedullary cell death might explain the apparent discrepancy between cellular marrow and blood cytopenias ( Yoshida, 1993; Raza et al, 1995 ). Programmed cell death has also been described in the microenvironment of MDS marrow ( Raza et al, 1995 ), possibly a link to the defective interaction of haemopoietic cells with the microenvironment ( Mufti, 1992). It has been suggested that apoptosis was related to abnormal expression of cytokines such as TNF-α ( Bogdanovic et al, 1996 ). TNF-α up-regulates Fas on normal CD34+ haemopoietic precursors ( Maciejewski et al, 1995b ). Ongoing apoptosis is likely to exert stress on the marrow, and may favour the emergence and escape of ‘aberrant’ cells, especially if their responses to growth (or death) signals differ from those of ‘normal’ cells. In the present study we examined the marrow of patients with MDS for the expression of TNF-α, Fas and Fas-L and determined to what extent the blockade of signals mediated by TNF-α or Fas-L-affected haemopoiesis in in vitro culture systems.


  1. Top of page
  2. Abstract
  6. Acknowledgements
  7. References


Patients with MDS were referred to the Fred Hutchinson Cancer Research Center (FHCRC) for a consultation regarding treatment or for marrow or peripheral blood stem cell transplantation. Patient and disease characteristics are summarized in Table I.

Marrow cells and plasma

Bone marrow samples were obtained from 44 patients with MDS and from 12 normal volunteer donors who had given informed consent according to the procedures approved by the Institutional Review Board of the FHCRC. Samples were processed as described ( Srour et al, 1993 ; Greinix et al, 1992 ). For in vitro haemopoiesis, mononuclear cells from fresh marrow (MMNCs) were isolated by Ficoll-Hypaque density-gradient centrifugation and resuspended in the appropriate medium. For molecular studies additional cells were processed as described below. A small volume (1–1.5 ml) of second marrow aspirate was immediately placed on ice, centrifuged at 4°C and the plasma removed and cryopreserved at −70°C for later use (see below).

Antibodies and reagents

The following antibodies were purchased from commercial sources (unless otherwise noted, all antibodies were of murine origin): fluorescein isothiocyanate (FITC), phycoerythrin (PE) or peridinin chlorophyll protein (PerCP)-conjugated anti-human CD34 (HPLC-2, IgG1; Becton Dickinson, San Jose, Calif.); FITC-conjugated anti-human CD3, CD14, CD19, CD45 and HLA-DR (Becton Dickinson); PE-conjugated non-agonistic anti-human Fas (DX2, IgG1; PharMingen, San Diego, Calif.); agonistic anti-human Fas (IgM; Upstate Biotechnology, Lake Placid, N.Y.); anti-murine F(ab′)2 IgG1-FITC or PE control antibody (Becton Dickinson); affinity-purified rabbit anti-human Fas-L polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, Calif.); hamster anti-human Fas-L (clone 4H9) and biotinylated hamster anti-human Fas-L IgG (clone 4A5) (Immunotech, Westbrook, Maine): anti-rabbit IgG-PE control (Caltag, San Francisco, Calif.); anti-human transforming growth factor-β (TGFβ)1,2,3 (IgG1; Genzyme, Cambridge, Mass.; this mAb recognizes all three isoforms); anti-human TNF-α (IgG1; Biosource International, Camarillo, Calif.; for functional assays); anti-human TNF-α (Boehringer Mannheim Biochemicals, Indianapolis, Ind.; for ELISA); anti-human sTNF-R1 (R & D Systems, Minneapolis, Min.); anti-human TNF-α:HRP conjugate (Boehringer Mannheim Biochemicals, Indianapolis, Ind.); rabbit anti-human sTNF-R1 (Monosan, Netherlands); donkey anti-rabbit IgG:HRP conjugate (Jackson ImmunoResearch, West Grove, Pa.); chicken anti-human TGFβ1 (R & D Systems Inc., Minneapolis, Min.); and goat anti-chicken IgG:HRP conjugate (Kirkegaard & Perry Labs, Gaithersburg, Md.).

The following reagents were purchased from commercial sources: streptavidin-HRP (Vector Laboratories Inc., Burlingame, Calif.); TNB-ELISA substrate (Life Technologies, Gaithersburg, Md.); TNF-α (Boehringer Mannheim Biochemicals, Indianapolis, Ind.); rhuTNFr (R & D Systems, Minneapolis, Min.); and rhuTGFβ-1 (Sigma, St Louis, Mo.). rhuTNFR:Fc, a soluble fusion protein consisting of the extracellular domain of the TNF receptor and the Fc portion of human IgG1 was constructed, expressed and purified as previously described ( Mohler et al, 1997 ). Fas-Ig, a soluble fusion protein consisting of the extracellular domain of Fas and the Fc portion of human IgG1, was constructed, expressed and purified as previously described ( Kiener et al, 1997 ; Liles et al, 1996 ).

Phenotyping of marrow cells

Phenotypic analysis of marrow cells was carried out by flow cytometry. Briefly, for Fas and Fas-L detection 106 cells were incubated with PE-conjugated anti-human Fas mAb (DX2) or with affinity-purified anti-rabbit Fas-L polyclonal antibody and PE-conjugated (secondary) goat anti-rabbit IgG. For three-colour immunofluorescence, FITC-conjugated mAb to CD45 was used in combination with PerCP-conjugated anti-CD34 mAb and with either PE-conjugated anti-human Fas mAb, or with affinity-purified anti-rabbit Fas-L polyclonal antibody and PE-conjugated (secondary) goat anti-rabbit IgG. Cells labelled with irrelevant isotype-matched mAb or incubated with the secondary detection antibody alone served as negative controls. Cells were analysed by cytofluorometric analysis using a FACScan (Becton Dickinson). Multidimensional data analysis was used to separate several cell populations that differ in lineage, maturational stage and activation. Four populations or clusters of cells were identified using this depiction of data (lymphocytes, monocytes, myeloid cells and blast cells). The identity of cells in each cluster was based on the expression of cell surface antigens ( Terstappen et al, 1992 ). The differences from normal myeloblasts were similar to those described for aberrant blasts detected of MDS blasts in acute myeloid leukaemia ( Sievers et al, 1996 ). These differences could be categorized into four groups: lineage infidelity (lymphoid antigens expressed on myeloid cells), antigenic asynchrony (immature antigens co-expressed with antigens found on mature cells), antigenic absence (lack of an antigen normally expressed at a specific stage), antigenic intensity abnormalities (over- or under-expression of specific antigens). Agreement between two analysts was required before blast cells were defined as abnormal. The difference from normal antigen expression was at least five-fold.

In vitro haemopoiesis

Marrow cultures were carried out as described using the Dexter method (Sutherland et al, 1990; Reems & Torok-Storb, 1995). Briefly, stromal layers were established in T25 tissue culture flasks (Corning, N.Y.) using Iscove's modified Dulbecco's medium (GIBCO, Grand Island, N.Y.) supplemented with 12.5% each of heat-inactivated horse serum and fetal calf serum (FCS), 0.4 mg/ml of L-glutamine, 1 mmol/l sodium pyruvate, 10−6 mol/l hydrocortisone, 10−4 mol/l 2-mercaptoethanol, 100 U/ml penicillin, and 100 μg/ml of streptomycin ( Dexter et al, 1977 ). Flasks were demidepleted of cells at weekly intervals. At 2–3 weeks the stromal cells were irradiated at 1800 cGy, removed from the flask by a brief exposure to 0.05% trypsin, and 1.25 × 105 viable cells/ml were plated as supportive layer in flat-bottommed 48-well plates (Costar, Cambridge, Mass.). After 1–5 d, MMNCs (0.25–1.0 × 106 cells/well) were plated onto the stromal layers in replicates of three. Fas-Ig (50 μg/ml), anti-Fas mAb (100 ng/ml), anti-TNF-α mAb (10 μg/ml), rhuTNFR:Fc (5 μg/ml) or anti TGFβ1,2,3 mAb (10 μg/ml) were added to the culture wells as indicated. Cultures were maintained at 33°C in a humidified air/5% CO2 atmosphere. After 1 week the medium containing non-adherent cells was collected, and the plates were washed twice with Ca2+ and Mg2+ free Hanks balanced salt solution and 100 μl of 0.05% trypsin was added to each well. The adherent cells were removed and combined with the non-adherent cells and washed once. The cell mixture was then resuspended in 3.5 ml of 2 × Dulbecco's medium containing 40% FCS, mixed with an equal volume of 0.06% Agar (Difco, Detroit, Mich.) and aliquoted into 35 × 10 mm culture dishes (2 ml/dish). 10 μl of a cytokine mixture (including IL-1, IL-3, IL-6, G-CSF, GM-CSF and SCF at concentrations of 10 μg/ml each) was added to each dish. Untreated cells served as experimental controls. After 2 weeks the plates were scored for colony numbers. Groups of at least 40 cells were counted as colonies.

Isolation of RNA and reverse transcription

Total RNA was prepared from cell pellets using the RNeasyTM Total RNA kit (Qiagen Inc., Chatsworth, Calif.). Briefly, cells were lysed in buffer RLT (contains 4 M guanidinium isothiocyanate, 0.5% sarcosyl and 0.1 M 2β-mercaptoethanol) and homogenized using a QIAshredder spin column. After centrifugation, an equal volume of 70% ethanol was added to the lysate and the mixture was applied onto a RNeasyTM spin column. Total RNA, which binds to the silica gel membrane under high salt buffer conditions, was washed with buffers RW1 and RPE (supplied with the kit), and then eluted in DEPC-treated water. Finally, any remaining DNA was removed by digestion with RNase-free DNase. The concentration and purity of the RNA was determined by measuring the absorbance at A260/A280, and by gel electropheresis (the RNA, 1–2 μl, was denatured in a solution containing 20% formaldehyde [37% vol/vol], 50% formamide, 10% 10 × MOPS and 5 μg/ml ethidium bromide, and then analysed using a 20% formaldehyde–1.5% agarose gel (Amresco, Solon, Ohio)).

To synthesize cDNA, 3 μg of the total RNA (corresponding to 1–2 × 106 cells) was combined with 1 μl of Oligo (dT)12-18 (500 μg/ml; Gibco/BRL, Gaithersburg, Md,) and heated for 5 min to 70°C; the resulting heteroduplex was resuspended to 20 μl in reverse transcriptase buffer, and incubated for 30 min at 37°C with 200 U Moloney murine leukaemia virus (M-MLV) reverse transcriptase (Gibco/BRL). After adding a second 100 U aliquot of M-MLV, the reaction was continued for an additional 15 min prior to heat-inactivating the enzyme for 2 min at 80°C. cDNA was stored at −70°C. Samples were then thawed in batches for PCR amplification.

PCR amplification

The cDNA was amplified by PCR in a reaction mixture consisting of 50 m M KCl, 20 m M Tris-HCl (pH 8.4), 0.1% Triton X-100, 1 m M DTT, 2.0 m M MgCl2, all four dNTPs (each at 0.2 m M), 10 pmol of each oligonucleotide primers (see below) and 2.5 units of Taq DNA Polymerase (Boehringer Mannheim). Amplification was performed in 0.5 ml Gene Amp tubes in a final volume of 50 μl. The PCR mixes were overlaid with mineral oil and amplified for 30–40 cycles (see below) of denaturation (at 94°C for 1 min), annealing (at 55–65°C (see below) for 1 min, and extension (at 72°C for 1 min). PCR products were size-fractionated and analysed by 1.5% agarose gel electropheresis and normalized according to the amount of β-actin in the same cDNA sample. PCR product sizes were determined from comparison to the READY-LOADTM 100 bp DNA standard ladder (Gibco/BRL).

Primer sequences, expected product sizes, number of cycles and annealing conditions were as follows: β-actin (no. 1) 5′-GTGGGGCGCCCCAGGCACCA, β-actin (no.1) 3′-CTCCTTAATGTCACGCACGATTTC (548 bp fragment; 35 cycles; 55–65°C annealing); β-actin (no. 2) 5′-TCCTGTGGCATCCACGAAACT, β-actin (no. 2) 3′-GAAGCATTTGCGGTGGACGAT (304 bp fragment; 35 cycles; 55–65°C annealing); Fas 5′-ATGCTGGGCATCTGGACCCTCCTA, Fas 3′-TCTGCACTTGGTATTCTGGGTCCG (384 bp fragment; 35 cycles; 60°C annealing); Fas-L (no. 1) 5′-CTGGGGATGTTTCAGCTCTTC, Fas-L (no. 1) 3′-CTTCACTCCAGAAAGCAGGAC (231 bp fragment; 40 cycles; 60°C annealing); Fas-L (no. 2) 5′-CAAGTCCAACTCAAGGTCCATGCC, Fas-L (no. 2) 3′-CAGAGAGAGCTCAGATACGTTGAC (345 bp fragment; 35 cycles; 60°C annealing); FAP-1 5′-GAATACGAGTGTCAGACATGG, FAP-1 3′-AGGTCTGCAGAGAAGCAAGAATAC (607 bp fragment; 38 cycles; 60°C annealing); TNF-α 5′ -AAGCCTGTAGCCCATGTTGT, TNF-α 3′ -CAGATAGATGGGCTCATACC (330 bp fragment; 35 cycles; 55°C annealing); Note: In the multiplex PCRs, the β-actin primers were used at 1 pmol each per reaction.

Southern blotting for Fas and Fas-L

Electropheretically separated PCR products were transferred and fixed onto Hybond-N (Amersham, Aylesbury, U.K.) for examination of Fas and Fas-L specific sequences. The membrane was hybridized with either digoxigenin (DIG)-labelled human Fas probe (obtained from S. Nagata, Osaka Bioscience Institute, Osaka, Japan) or with DIG-labelled Fas-L probe (obtained from E. Mita, Osaka University School of Medicine, Osaka, Japan), and washed. Detection of DIG-labelled nucleotides was accomplished with a chemiluminescent reaction using CDP-starTM (Boehringer, Mannheim, Germany).

Enzyme-linked immunosorbent assays (ELISA)

The ELISA for Fas-L was performed essentially as described ( Tanaka et al, 1996 ). Briefly Immulon 2 plates (Dynatech, Chantilly, Va.) were coated with 100 μl/well of the 4H9 mAb (1 μg/ml) in 100 m M pH 9.5 Na bicarbonate overnight at 4°C. The plates were blocked for 2 h with 5% BSA. 100 μl of standard Fas-L or bone marrow plasma samples were added to the wells and incubated for 4 h at room temperature (RT). The plates were washed 10 × with PBS, 0.05% Tween 20 (PBS-T; SIGMA, St Louis, Mo.) and then 100 μl of the biotinylated 4A5 anti-Fas-L mAb (2 μg/ml) was added to each well. The plates were incubated for 2 h at RT, washed 10 × with PBS-T, and incubated with streptavidin-HRP for 1 h at RT. The plates were washed 15 × with PBS-T and then developed with TNB-ELISA substrate (Life Technologies). The assay was sensitive down to a concentration of 10 pg/ml Fas-L.

The concentrations of TNF-α, soluble TNF receptor (sTNF-R1; p55/60) and TGFβ1 in human bone marrow plasma were analysed by ELISA, developed by Allen Farrand, Shared Resources Cytokine Lab, FHCRC. Briefly, 96-well polystyrene plates (Costar, Cambridge, Mass.) were coated overnight at 4°C with either 2 μg/ml of anti-human TNF-α (diluted in 50 m M Na carbonate, pH 9.5), 0.5 μg/ml of anti-human sTNF-R1 (diluted in carbonate buffer), or with 1 μg/ml of anti-human TGFβ1,2,3 (diluted in PBS). After coating, the plates were washed three times with PBS-T. In the TNF-α ELISA, nonspecific activity was blocked by incubating the coated plates with 1% BSA/50 m M Na carbonate, pH 9.5, for 20 min at RT; for sTNF-R1, by incubating the plates with 5% Blotto/PBS (Non-Fat Dry Milk in PBS, Carnation) for 30 min at RT; and for TGFβ1, by incubating the plates with 1% BSA/TBS (tris buffered saline, Sigma) for 1 h at RT. The plates were then washed three times with PBS-T prior to the addition of samples.


Plasma samples were diluted 1:10 in assay buffer (1% BSA/5 m M EDTA/TBS–0.05% Tween 20) and incubated at 4°C. The following day the plates were washed three times with PBS-T. Captured TNF-α was detected by adding 0.05 units/ml of anti-human TNF-α:HRP conjugate (diluted in assay buffer) for 4 h at RT, plates were washed with three times PBS-T, and substrate 3,3′,5,5′, tetramethylbenzidinedihydrochloride (TMB with H2O2; Kirkegaard & Perry Labs) was added. The reaction was stopped with 1 M H2SO4. Optical density was determined at 450 nm using a microplate reader (Vmax, Molecular Devices, Sunnyvale, Calif.).


Plasma samples were diluted 1:10 in high salt assay buffer (1% mouse serum/5 m M EDTA/PBS-T + 0.5 M NaCl) and incubated at 4°C. The next day the plates were washed three times with PBS-T. Captured sTNF-R1 was developed by adding 0.1 μg/ml rabbit anti-human sTNF-R1 (in 1% mouse serum/PBS-T) for 2 h at RT. After washing three times with PBS-T, donkey anti-rabbit IgG:HRP conjugate (diluted 1:5000 in PBS-T) was added to each of the wells and incubated for 30 min at RT. The plates were again washed three times with PBS-T, and substrate was added as described above.


Plasma samples were acidified, neutralized, diluted 1:20 in assay buffer and incubated for 1 h at RT. Acidification was accomplished by adding an equal volume of 2.5 M acetic acid, 10 M urea to the samples and incubating at RT for 10 min; an equal volume of 2.7 NaOH, 1 M Hepes was then added to neutralize the sample. Next, the plates were washed three times with PBS-T. Captured TGFβ1 was developed by adding 1 μg/ml of chicken anti-human TGFβ1 (in 1% BSA/TBS-T) for 1 h at RT. After washing three times with PBS-T, goat anti-chicken IgG:HRP conjugate (diluted to 50 ng/ml in BSA/TBS-T) was added to each of the wells and incubated for 1 h at RT. The plates were again washed three times with PBS-T, and substrate was added as described above.

The concentrations of TNF-α, sTNF-R1 and TGFβ1 in the test samples were calculated from standardized curves prepared using either rhuTNF-α, rhuTNFr or rhuTGFβ1 (standards were diluted in the same low/high salt assay buffers as described above) by Four-Parameter Analysis (SoftMax Pro, Molecular Devices, Sunnyvale, Calif.). Inter-assay and intra-assay confidence variables were determined to be <10% with an assay sensitivity of <1 pg/ml for TNF-α and <2 pg/ml for both sTNF-R1 and TGFβ1. All samples, standards and controls were assayed in duplicates.

Statistical analysis

A Wilcoxon rank-sum test was used to compare measurements of protein concentrations between MDS and normal marrow, and colony numbers between untreated and treated MDS marrow. The Kruskall-Wallis test was used to test for correlations between patient or disease characteristics and in vitro test results. Significant differences between data from specific groups were defined as those with a P value leqslant R: less-than-or-eq, slant0.05.


  1. Top of page
  2. Abstract
  6. Acknowledgements
  7. References

Spectrum of MDS patients

Patient and disease characteristics are summarized in Table I. The relatively low median age of patients (48 years) is due to the referral pattern of patients, focusing on candidates for marrow transplantation. Among 44 patients studied, 35 were thought to have primary and nine secondary (therapy-related) MDS. 18 patients were categorized as RA, one as RARS, seven as RAEB, and 18 as RAEB-T. 30 patients had cytogenetic abnormalities and 14 did not.

Multidimensional flow cytometric analysis and sorting of subpopulations of marrow cells from MDS patients

Aberrant blasts

Preliminary studies had shown that marrow from many patients with MDS contained a population of blast cells with ‘aberrant’ characteristics as defined by CD45 expression and light-scatter properties ( Stelzer et al, 1993 ) (see Fig 1). By definition, blasts are infrequent in patients with RA, but are prominent in patients with RAEB-T. Although the expression of surface antigens (e.g. CD13, CD34, CD38, HLA-DR, etc.) is heterogenous on normal blasts, it is rather homogenous on aberrant blasts ( Sievers et al, 1996 ; Loken et al, 1992 ). Using multidimensional flow cytometry, we performed an analysis of cell surface expression of additional antigens (including CD3, CD13, CD14, CD19, Fas and Fas-L) in 24 patients with aberrant blasts to further define the phenotype of aberrant blasts. As expected, the proportion of blasts in the marrow (as defined by orthogonal light scatter and CD45 labelling) increased from 3.8 ± 1.6 (24.9 ± 12.2% CD34+) in patients with RA to 16.7 ± 10.3 (55.9 ± 23.8% CD34+) in patients with RAEB and 28.6 ± 23.0 (64.8 ± 25.8% CD34+) in patients with RAEB-T. In parallel the proportion of these CD34+ cells in the marrow increased from 1.4 ± 1.1% to 10.3 ± 6.9% and 21.6 ± 25.5% in patients with RA, RAEB and RAEB-T, respectively. Furthermore, in the majority of patients studied (15/24) the MDS blast population also showed homogenous staining for HLA-DR, CD33 and CD13 in contrast to the heterogenous pattern observed in normal marrow ( Terstappen et al, 1992 ). In addition, in 10/24 patients studied blast cells also expressed CD56. Thus, there was a progressive increase in relative and absolute numbers of blasts with aberrant phenotypic characteristics with progressive disease stages as determined by FAB classification.


Figure 1. % blasts (lymphoblasts and myeloblasts) for the normal marrow, and 25% lymphocytes, 7.0% monocytes, 23.8% myeloid forms and 38.0% blasts for the MDS marrow. The MDS patient depicted here was classified as RAEB-T. Shown are 5000 events/panel.

Download figure to PowerPoint

Fas and Fas-L expression

In addition to conventional surface markers, marrow blasts and lymphocytes were analysed for Fas and Fas-L expression, since aberrant expression of these markers has been recognized in patients with marrow failure ( Yoshida, 1993; Maciejewski et al, 1995a). Results on 24  patients and 10 controls are summarized in Table II. In the blast cell gate (see Fig 1), Fas expression was increased on CD34+ cells (MDS: 87 ± 11%; normal: 25 ± 11%) (P < 0.001) and on cells co-expressing HLA-DR+ (MDS: 75 ± 14%; normal: 47 ± 21%) (P < 0.05). Fas-L was not detectable on CD34+ precursors from normal controls. Fas-L was expressed, however, on CD34+ cells from MDS marrow, although the proportion of positive cells varied from patient to patient. In the lymphocyte gate of MDS marrow, both Fas (75 ± 17%; normal: 57 ± 13%) (P < 0.05) and Fas-L (17 ± 9%; normal: 2 ± 1%) (P < 0.01) were increased on CD3+ cells, and Fas-L on CD19+ cells (14 ± 2%; normal: 5 ± 2%) (P < 0.01). Fas-L expression was confirmed by experiments using Western blotting (not shown). One example of aberrant expression of Fas and Fas-L on CD34+ cells from a patient with MDS by flowcytometry is shown in Fig 2 A–D. The aberrant blast cells over-expressed CD34 and co-expressed Fas antigen and Fas-L.


Figure 2. In these figures a combination of forward and orthogonal light scattering (panel A), CD45 fluorescence and right-angle (side) light scatter (panel B), CD95 PE and CD34 PerCp staining (panel C) and Fas-L PE and CD34 PerCp staining (panel D) are shown simultaneously. The MDS patient was classified as RAEB. Analysis shows a leucocyte differential of 30% lymphocytes, 4.3% monocytes, 48.5% myeloid forms and 11.5% blasts (panel B). Myeloblasts show homogenous abnormal surface antigen expression of HLA-DR, CD38, CD13, CD33, but not CD11b, CD14, CD15, CD16 or any lymphoid antigens (data not shown). The blasts are intermediate in size by light scatter with decreased CD45 expression. Of these cells, 94.1% are CD34+, 83.0% are Fas+ (panel C) and 45.4% are Fas-L+ (panel D).

Download figure to PowerPoint

Molecular analysis

Fas and Fas-L

Total marrow mononuclear cell extracts were prepared from normal donors and from patients with MDS. Estimation of mRNA expression was based on band intensities of RT-PCR products electropheresed on agarose gels, or on analysis of autoradiographs following Southern blotting. In total cell preparations, Fas-L mRNA was up-regulated in all patients studied as shown by Southern analysis (Fig 3A) and by ethidium bromide staining of RT-PCR products (Fig 3B). In addition, preliminary experiments using highly purified CD34+ cells (>95% purity; prepared using Ceprate® affinity columns) ( Berenson et al, 1986 ) showed that expression of Fas-L was also prominent in these CD34+ cells in contrast to extremely low or undetectable levels in highly purified normal CD34+ cells (data not shown). Finally, the expression of Fas mRNA was increased in MDS patients, although less consistently than Fas-L, as compared to normals (Fig 3C).


Figure 3. Reverse transcriptase-PCR analysis of human Fas and Fas-L mRNA expression in marrow cells from normal controls and several MDS patients. (A) Identity of bands representing Fas-L was verified by hybridization of the blot with DIG-labelled probes (NML — normal controls; (B) identity of bands representing Fas-L by ethidium bromide staining (NBM — normal bone marrow); and (C) identity of bands representing Fas by Southern blotting. β-actin no. 2 (548 bp) served as an internal control in the multiplex PCR (NML — normal controls).

Download figure to PowerPoint

Soluble Fas-L in marrow plasma was measured by ELISA. Mean levels in MDS patients tended to be higher (ranging from 8.8 to 218 pg/ml; median of 54.5 pg/ml) than levels in controls (12.4–49.4 pg/ml; median 18.4 pg/ml) (P = 0.094).

FAP-1 is a protein tyrosine phosphatase that interacts with the cytosolic domain of Fas. FAP-1 expression has been shown to be highest in cells that are relatively resistant to Fas-mediated cytotoxicity ( Sato et al, 1995 ). We determined the expression of FAP-1 mRNA in marrow cells from several MDS patients by RT-PCR. As shown in Fig 4, FAP-1 expression was lower or absent in most MDS patients (more so in RAEB-T than in RA), as compared to normal marrow, consistent with the notion that cells in MDS marrow are more susceptible to the effects of Fas/Fas-L which are up-regulated in MDS patients.


Figure 4.  bp) served as an internal control in the multiplex PCR.

Download figure to PowerPoint

Cytokines and receptors

Since cytokines such as TNF-α are known to serve as negative regulators of haemopoiesis and have been shown to affect Fas expression, we determined TNF-α and sTNF-R1 expression on MDS marrow cells. As shown in Fig 5, TNF-α message was up-regulated consistently in patients with MDS. Furthermore, TNF-α protein levels in marrow plasma were significantly higher in MDS than in normal marrow ( Table III; P < 0.002). There was a significant correlation between TNF-α protein levels in marrow plasma and Fas expression on marrow blasts (P = 0.014). There was also an increase in soluble TNF receptor (sTNF-R1) in MDS marrow plasma (P < 0.003). No differences between normal and MDS marrow were observed for TGFβ1, another bifunctional cytokine involved in the regulation of haemopoiesis.


Figure 5. 48 bp) served as an internal control in the multiplex PCR.

Download figure to PowerPoint

Functional analysis

In vitro haemopoiesis

Marrow cells from 21 patients with MDS and 12 normal controls were propagated in long-term marrow cultures (LTMC) in vitro, and colony formation was determined in semi-solid medium. On a per cell basis the numbers of CFU-GMs derived from MDS marrow were an average 60% lower than those derived from normal marrow (Fig 6; P < 0.03), consistent with results reported by others ( Nagler et al, 1995 ). Colony frequency was lowest in patients with RA and highest in patients with RAEB-T. On average, 3.5 times more cells from patients with RA than from RAEB-T patients were required to generate a given number of colonies.


Figure 6. for normal marrow).

Download figure to PowerPoint

Since marrow cells from patients with MDS expressed Fas, Fas-L and TNF-α at higher levels than normal marrow, we speculated that blocking these receptors or ligands might protect colony formation. As shown in Fig 6, the addition of Fas-Ig to LTMCs significantly increased colony formation from MDS marrow as compared to untreated cultures (mean increase 104% [range 10–400%]; P < 0.001). The addition of Fas-Ig to cultures of normal marrow resulted only in a 25% increase in colony number relative to untreated cultures (P = 0.76). Also in agreement with the findings of decreased FAP-1 mRNA expression and increased Fas expression on MDS marrow cells was the observation that the addition of agonistic anti-Fas mAb resulted in a 75% decrease in colonies in MDS marrow compared to a ≈50% reduction in normal marrow. As illustrated in Fig 7, the addition of anti-TNF-α mAb or soluble rhuTNFR:Fc to cultures also increased colony numbers (110%; P < 0.001 and 25%; P<0.003, respectively). The addition of anti-TGF-β did not have a positive effect, and, in fact, slightly decreased colony numbers. These data are compatible with the hypothesis that both TNFR/TNF-α and Fas/Fas-L interactions are involved in inhibition of haemopoiesis in MDS marrow. The data are also consistent with the notion that the TNF-α signal occurs upstream of Fas ( Tanaka et al, 1996 ) and suggest that Fas is expressed on cells required for colony formation.


Figure 7. . Effect of anti-TNF-α, rhuTNFR:Fc, TGF-β on long-term marrow cultures. Normal or MDS marrow cells were cultured either under control conditions (none) or in the presence of anti TNF-α mAb (10 μg/ml), rhuTNFR:Fc (5 μg/ml) or anti-TGF-β1,2,3 mAb (10 μg/ml). Individual results (mean of triplicate cultures) expressed as percent of untreated concurrent control cultures (100%) are presented; the mean of all values is indicated by a horizontal bar (P < 0.001 for anti-TNF-α; P < 0.003 for rhuTNFR:Fc; P = NS for TGF-β).

Download figure to PowerPoint

Correlation of in vitro studies with patient and disease characteristics

The median values for each of the in vitro test parameters studied (i.e. TNF-α, sTNF-R1 and Fas-L values, Fas and Fas-L expression, colony counts) were compared across each of the three patient categories ( Table I). Results are summarized in Table IV. There was a correlation between the amount of TNF-α protein in marrow plasma and the disease category: Patients with RA had higher TNF-α values than patients with RAEB or RAEB-T (P = 0.043). As stated before, there was a suggestion of a correlation between TNF-α levels and Fas expression on blast cells although this did not reach significance level (P = 0.12). In addition, there was a strong correlation between Fas-L expression on aberrant blasts and the disease category: patients with RA had the lowest percentages of Fas-L-positive blast cells, whereas patients in the RAEB and RAEB-T groups had the highest values (P = 0.037). Since Fas-L is expressed at higher levels in the more advanced disease states, this finding may be of consequence if abnormal Fas-L+ blasts are responsible for triggering death in other cell types, i.e. the presence of the ligand for Fas on their own surface would preferentially result in death in other cells expressing Fas (but not the ligand). Marrow of patients with secondary MDS contained more Fas-L+ blasts than marrow from patients with primary MDS (P = 0.019). Data from colony-forming assays showed an inverse correlation between the disease category and the number of cells required to form a given number of colonies (P < 0.03), presumably a reflection of the concentration of progenitor cells present in the marrow. Finally, no correlation was found between the presence of normal cytogenetics or cytogenetic abnormalities and the aberrant expression of Fas or Fas-L on blast cells or lymphocytes, elevated marrow plasma TNF-α levels or the numbers of cells required to form colonies in functional assays.


  1. Top of page
  2. Abstract
  6. Acknowledgements
  7. References

The pathophysiology of haemopoietic failure in patients with MDS is not well understood. It has been hypothesized that a lack of haemopoietic growth factors is part of the mechanism; however, the administration of various growth factors (positive regulators of haemopoiesis) to patients with MDS has met with only limited success, i.e. only few patients have responded with improvement of blood cell counts over extended periods of time ( Greenberg, 1992; Ganser et al, 1991 ). Alternatively, it has been proposed that abnormal expression of negative regulators of haemopoiesis such as TNF-α may be responsible ( Maciejewski et al, 1995a ; Bogdanovic et al, 1996 ). TNF-α up-regulates Fas expression on normal CD34+ haemopoietic precursor cells, and we have shown that both Fas and Fas-L were up-regulated on marrow cells from MDS patients ( Gersuk et al, 1996 ). Since Fas-mediated signals play a central role in apoptosis and Fas signals are triggered by Fas-L, these findings suggest the possibility that Fas-mediated apoptosis is involved in haemopoietic failure as observed in patients with MDS.

In the present study we investigated TNF-α and Fas-mediated signals and their effect on haemopoiesis in patients with MDS. As suggested by previous reports, results show up-regulation of TNF-α ( Yamaguchi et al, 1996 ; Shetty et al, 1996 ), Fas and Fas-L in the marrow of patients with MDS ( Gersuk et al, 1996 ; Maciejewski et al, 1995a ). In contrast to normal marrow, Fas-L was also expressed on a proportion of CD34+ precursor cells, as part of the aberrant phenotype of blasts in MDS. Although the significance of this finding is not entirely clear, it is of note that the addition of Fas-Ig to LTMCs resulted in an increase in colony formation, apparently by protecting precursors against Fas-mediated death signals. Consistent with such a mechanism was the observation that treatment of haemopoietic precursor cells with an agonistic anti-Fas mAb resulted in a marked decline of colonies. The addition of anti-TNF-α mAb or the soluble fusion protein rhuTNFR:Fc also was associated with an increase in colony formation. In agreement with a model we described previously ( Hong et al, 1995 ; Lee et al, 1997 ), these results suggest that Fas/Fas-L signals affect haemopoiesis downstream of TNF-α, although the possibility of a direct effect of TNF-α independent of Fas on haemopoiesis was not formally excluded.

The regulation of Fas-mediated signals is complex ( Nagata, 1997). Several regulatory elements (adaptor molecules) interacting with the cytoplasmatic domain of Fas have been recognized. These include FADD (Fas-associating protein with death domain), Caspase-8 (originally termed FLICE, a FADD-like interleukin-1β-converting enzyme [ICE] protease), and FAP-1 (a protein tyrosine phosphatase [PTP], originally cloned from human basophils and termed PTP-BAS, which associates with the cytoplasmatic domain of Fas) ( Moller et al, 1994 ; Maekawa et al, 1994 ; Muzio et al, 1996 ; Alnemri et al, 1996 ; Berenson et al, 1986 ). Results of preliminary investigations indicate that levels of FAP-1 are reduced in MDS marrow. Decreased expression of FAP-1 would be expected to facilitate Fas signals and apoptosis since FAP-1 functions as a negative regulator of Fas. It may be important, therefore, to determine whether the expression of FAP-1 (and other adaptor molecules) differs between cell populations in MDS marrow and possibly between patients with different stages of MDS.

Important in this context is the determination of which cells — residual normal cells or aberrant and possibly clonal malignant cells — are inhibited and which cells are protected by interventions that block TNF or Fas mediated signals, such as rhuTNFR:Fc or Fas-Ig. Data from in vitro experiments similar to those summarized in Fig 7 suggest that normal precursors (at least as defined by the absence of a clonal cytogenetic marker) are present in the marrow and that interventions aimed at preventing cell death protect the growth of normal cells (unpublished observations). Whether such an intervention might also suppress the ‘escape’ of aberrant cells from regulatory control remains to be determined. A recent report by Raza et al (1996a ) indicates that haemopoiesis in patients with MDS can be improved by treatment with modalities aimed at TNF-α blockade. The present data provide phenotypic, functional and molecular ex vivo data which are in support of those findings, and suggest that manipulations which interfere with the function of negative regulators may favour normal rather than aberrant haemopoiesis.

The present data also illustrate further the rather heterogenous nature of MDS, which, beyond cell morphology and numbers, is expressed in cell phenotype, cytokine profile and probably at the level of molecular regulatory pathways. For example, the highest levels of TNF-α were observed in patients with RA. TNF up-regulates Fas, and prominent Fas expression on blasts in RA would render those cells highly susceptible to Fas-mediated cell death which would presumably result in low numbers of proliferating cells. Such a scenario is in agreement with at least one study showing a correlation between TNF-α levels and the extent of apoptosis ( Raza et al, 1996b ). In the present study Fas-L was expressed most prominently on blasts from patients with RAEB, RAEB-T and patients with secondary MDS, i.e. groups of patients who frequently show rapid disease progression. Increased Fas-L expression in these disease categories might be related to additional ‘activation’ steps and further enhance cell death in any residual normal precursors which express Fas as a consequence of increased levels of TNF-α. It will be important, however, to determine why aberrant (clonal) cells which also express Fas enjoy relative protection. Homonymous interactions of receptor and ligand are very effective, and even concurrent expression of Fas and Fas-L on the same cell has been shown to lead to cell death possibly via membrane folding that then permits for receptor/ligand interaction ( Nagata & Golstein, 1995). Protection could be related to alterations in Fas adaptor molecules (see above) or to Fas itself, e.g. in the form of splicing variants that lack the cytoplasmatic death domain as described recently by Cascino et al, (1996 ) in lymphoma-derived clones. Alternatively, dysregulation of apoptosis-related oncogenes such as bcl-2 and c-myc may be involved ( Rajapaksa et al, 1996 ). In fact, these questions are relevant not only to MDS but also to the maintenance of haemopoiesis in patients with acute and chronic myeloid leukaemia (CML). Selleri et al (1997 ) found that treatment with agonistic anti-Fas mAb inhibited in vitro colony formation more in CML progenitors than in normal progenitors. Iijima et al (1997 ) observed various levels of Fas expression on marrow cells of patients with acute myeloid leukaemia but found no correlation between the level of Fas expression and apoptosis. Clearly, further studies are needed to develop a better understanding of the regulation of apoptosis in these haemopoietic disorders.

In summary, results of the present analysis are consistent with a central role of negative regulators of haemopoiesis in MDS. The data indicate correlations between disease category, cell phenotype and negative regulators. Although further studies are needed to characterize differences between normal and aberrant cells, the present observations provide a basis for new treatment options for patients with MDS and suggest that different disease categories may benefit from different interventions.


  1. Top of page
  2. Abstract
  6. Acknowledgements
  7. References

This work was supported in part by grants HL36444, CA18029 and CA09515 from the National Institutes of Health, DHHS, Bethesda, Maryland. H.J.D. is also supported by a grant from the National Marrow Donor Program/Baxter Health Care Division.

We thank Denise Wells, Cherie Green, Mitch Hall and Keely Ghirardelli for their help in preparing and analysing samples for flow cytometry; Marina Lesnikova, Eileen Bryant, Fred Appelbaum, and Masaki Yamaguchi, for helpful discussions and technical assistance; Ted Gooley, for his help with statistical analyses; and Bonnie Larson and Harriet Childs for manuscript preparation.


  1. Top of page
  2. Abstract
  6. Acknowledgements
  7. References
  • 1
    Alderson, M.R., Tough, T.W., Davis-Smith, T., Braddy, S., Falk, B., Schooley, K.A., Goodwin, R.G., Smith, C.A., Ramsdell, F., Lynch, D.H. (1995) Fas ligand mediates activation-induced cell death in human T lymphocytes. Journal of Experimental Medicine, 181, 71 77.
  • 2
    Alnemri, E.S., Livingston, D.J., Nicholson, D.W., Salvesen, G., Thornberry, N.A., Wong, W.W., Yuan, J. (1996) Human ICE/CED-3 protease nomenclature. (Letter). Cell, 87, 171.
  • 3
    Badley, A.D., McElhinny, J.A., Leibson, P.J., Lynch, D.H., Alderson, M.R., Paya, C.V. (1996) Upregulation of Fas ligand expression by human immunodeficiency virus in human macrophages mediates apoptosis of uninfected T lymphocytes. Journal of Virology, 70, 199 206.
  • 4
    Berenson, R.J., Bensinger, W.I., Kalamasz, D. (1986) Positive selection of viable cell populations using avidin-biotin immunoadsorption. Journal of Immunological Methods, 91, 11 19.
  • 5
    Bogdanovic, A.D., Jankovic, &G.M., Colovic, M.D., Trpinac, D.P., Bumbasirevic, V.Z. (1996) Apoptosis in bone marrow of myelodysplastic syndrome patients. (Letter). Blood, 87, 3064.
  • 6
    Cascino, I., Papoff, G., De Maria, R., Testi, R., Ruberti, G. (1996) Fas/Apo-1 (CD95) receptor lacking the intracytoplasmic signaling domain protects tumor cells from Fas-mediated apoptosis. Journal of Immunology, 156, 13 17.
  • 7
    Dexter, T.M., Allen, T.D., Lajtha, L.G. (1977) Conditions controlling the proliferation of haemopoietic stem cells in vitro. Journal of Cellular Physiology, 91, 335-344.
  • 8
    Dhein, J., Walczak, H., Baumler, C., Debatin, K.M., Krammer, P.H. (1995) Autocrine T-cell suicide mediated by APO-1/(Fas/CD95). Nature, 373, 438 441.
  • 9
    Ganser, A., Seipelt, G., Hoelzer, D. (1991) The role of GM-CSF, G-CSF, interleukin-3, and erythropoietin in myelodysplastic syndromes. American Journal of Clinical Oncology, 14, (Suppl. 1), S34 S39.
  • 10
    Gersuk, G.M., Lee, J.W., Beckham, C.A., Anderson, J., Deeg, H.J. (1996) Fas (CD95) receptor and fas-ligand expression in bone marrow cells from patients with myelodysplastic syndrome (Letter). Blood, 88, 1122 1123.
  • 11
    Greenberg, P., Cox, C., LeBeau, M.M., Fenaux, P., Morel, P., Sanz, G., Sanz, M., Vallespi, T., Hamblin, T., Oscier, D., Ohyashiki, K., Toyama, K., Aul, C., Mufti, G., Bennett, J. (1997) International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood, 89, 2079 2088.
  • 12
    Greenberg, P.L. (1992) Treatment of myelodysplastic syndromes with hemopoietic growth factors. Seminars in Oncology, 19, 106 114.
  • 13
    Greinix, H.T., Storb, R., Bartelmez, S.H. (1992) Specific growth inhibition of primitive hematopoietic progenitor cells mediated through monoclonal antibody binding to major histocompatibility class II molecules. Blood, 80, 1950 1956.
  • 14
    Hahne, M., Rimoldi, D., Schroter, M., Romero, P., Schreier, M., French, L.E., Schneider, P., Bornand, T., Fontana, A., Lienard, D., Cerottini, J., Tschopp, J. (1996) Melanoma cell expression of Fas(Apo-1/CD95) ligand: implications for tumor immune escape. Science, 274, 1363 1366.
  • 15
    Hatake, K., Tomizuka, H., Ikeda, M., Tsunoda, J., Hoshino, Y., Ohtsuki, T., Kasahara, T., Yonehara, S., Miura, Y. (1994) The presence of apoptosis in refractory anemia of myelodysplasia. Molecular Biology of Haematopoiesis(ed. by N. Abraham, R. K. Shadduck, A. S. Levine and F. Takaku), p. 127. Intercept Ltd, Andover.
  • 16
    Heyman, M.R. (1991) Recent advances in biology and treatment of myelodysplasia. Current Opinion in Oncology, 3, 44 53.
  • 17
    Hong, D.-S., Beckham, C., Huss, R., Lee, J.W., Hockenbery, D., Ledbetter, J.A., Deeg, H.J. (1995) Major histocompatibility complex class II-mediated inhibition of hematopoiesis in long-term marrow cultures involves apoptosis and is prevented by c-kit ligand. Blood, 86, 3341 3352.
  • 18
    Iijima, N., Miyamura, K., Itou, T., Tanimoto, M., Sobue, R., Saito, H. (1997) Functional expression of Fas (CD95) in acute myeloid leukemia cells in the context of CD34 and CD38 expression: possible correlation with sensitivity to chemotherapy. Blood, 90, 4901 4909.
  • 19
    Kiener, P.A., Davis, P.M., Starling, G.C., Mehlin, C., Klebanoff, S.J., Ledbetter, J.A., Liles, W.C. (1997) Differential induction of apoptosis by Fas–Fas ligand interactions in human monocytes and macrophages. Journal of Experimental Medicine, 185, 1511 1516.
  • 20
    Krammer, P.H., Dhein, J., Walczak, H., Behrmann, I., Mariani, S., Matiba, B., Fath, M., Daniel, P.T., Knipping, E., Westendorp, M.O., Stricker, K., Baumler, C., Hellbardt, S., Germer, M., Peter, M.E., Debatin, K.M. (1994) The role of APO-1-mediated apoptosis in the immune system. Immunological Reviews, 142, 175 191.
  • 21
    Lee, J.W., Gersuk, G.M., Kiener, P.A., Beckham, C., Ledbetter, J.A., Deeg, H.J. (1997) HLA-DR-triggered inhibition of hematopoiesis involves Fas/Fas-ligand interactions and is prevented by c-kit ligand. Journal of Immunology, 159, 3211 3219.
  • 22
    Liles, W.C., Kiener, P.A., Ledbetter, J.A., Aruffo, A., Klebanoff, S.J. (1996) Differential expression of Fas (CD95) and Fas ligand on normal human phagocytes: implications for the regulation of apoptosis in neutrophils. Journal of Experimental Medicine, 184, 429 440.
  • 23
    Loken, M.R., Terstappen, L.W.M.M., Civin, C.I., Fackler, M.J. (1992) Flow cytometric characterization of erythroid, lymphoid and monomyeloid lineages in normal human bone marrow. Flow Cytometry in Hematology(ed. by O. D. Laerum and R. Bjerksnes), p.31. Academic Press, New York.
  • 24
    Maciejewski, J., Selleri, C., Anderson, S., Young, N.S. (1995a)Fas antigen expression on CD34+ human marrow cells is induced by interferon gamma and tumor necrosis factor alpha and potentiates cytokine-mediated hematopoietic suppression in vitro. Blood, 85, 3183 3190.
  • 25
    Maciejewski, J.P., Selleri, C., Sato, T., Anderson, S., Young, N.S. (1995b)Increased expression of Fas antigen on bone marrow CD34+ cells of patients with aplastic anaemia. British Journal of Haematology, 91, 245 252.
  • 26
    Maekawa, K., Imagawa, N., Nagamatsu, M., Harada, S. (1994) Molecular cloning of a novel protein-tyrosine phosphatase containing a membrane-binding domain and GLGF repeats. FEBS Letters, 337, 200 206.
  • 27
    Mayer, R.J. & Canellos, G.P. (1996) Primary myelodysplastic syndrome and secondary preleukemic disorders. Leukemia(ed. by E. S. Henderson, T. A. Lister and M. F. Greaves), p.513. Saunders, Philadelphia.
  • 28
    Mohler, K.M., Torrance, D.S., Smith, C.A., Goodwin, R.G., Stremler, K.E., Fung, V.P., Madani, H., Widmer, M.B. (1997) Soluble tumor necrosis factor (TNF) receptors are effective therapeutic agents in lethal endotoxemia and function simultaneously as both TNF carriers and TNF antagonists. Journal of Immunology, 151, 1548 1561.
  • 29
    Moller, N.P., Moller, K.B., Lammers, R., Kharitonenkov, A., Sures, I., Ullrich, A. (1994) Src kinase associates with a member of a distinct subfamily of protein-tyrosine phosphatases containing an ezrin-like domain. Proceedings of the National Academy of Sciences of the United States of America, 91, 7477 7481.
  • 30
    Mufti, G.J. (1992) A guide to risk assessment in the primary myelodysplastic syndrome. Hematology/Oncology Clinics of North America, 6, 587 606.
  • 31
    Muzio, M., Chinnaiyan, A.M., Kischkel, F.C., O'rourke, K., Shevchenko, A., Ni, J., Scaffidi, C., Bretz, J.D., Zhang, M., Gentz, R., Mann, M., Krammer, P.H., Peter, M.E., Dixit, V.M. (1996) FLICE, a novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95 (Fas/APO-1) death-inducing signaling complex. Cell, 85, 817 827.
  • 32
    Nagata, S. (1997) Apoptosis by death factor. Cell, 88, 355 365.
  • 33
    Nagata, S. & Golstein, P. (1995) The Fas death factor. Science, 267, 1449 1456.
  • 34
    Nagler, A., MacKichan, M.L., Negrin, R.S., Donlon, T., Greenberg, P.L. (1995) Effects of granulocyte colony-stimulating factor therapy on in vitro hemopoiesis in myelodysplastic syndromes. Leukemia, 9, 30 39.
  • 35
    Noel, P. & Solberg, L.A. (1992) Myelodysplastic syndromes: pathogenesis, diagnosis and treatment. Critical Reviews in Oncology/Hematology, 12, 193 215.
  • 36
    O'connell, J., O'sullivan, G.C., Collins, J., Shanahan, F. (1996) The Fas counterattack: Fas-mediated T cell killing by colon cancer cells expressing Fas ligand. Journal of Experimental Medicine, 184, 1075 1082.
  • 37
    Rajapaksa, R., Ginzton, N., Rott, L.S., Greenberg, P.L. (1996) Altered oncoprotein expression and apoptosis in myelodysplastic syndrome marrow cells. Blood, 88, 4275 4287.
  • 38
    Rathmell, J.C., Cooke, M.P., Ho, W.Y., Grein, J., Townsend, S.E., Davis, M.M., Goodnow, C.C. (1995) CD95 (Fas)-dependent elimination of self-reactive B cells upon interaction with CD4+ T cells. Nature, 376, 181 184.
  • 39
    Raza, A., Gezer, S., Gregory, S.A., Venugopal, P., Kaizer, H., Dong, L.M., Span, L., Hines, C., Thomas, R., Alvi, S., Mundle, S.D., Shetty, V., Borok, R., Loew, J., Reza, S., Robin, E.L., Rifkin, S.D., Alston, D., Hernandez, B.M., Shah, R., Hsu, W.-T., Dar, S., Preisler, H.D. (1996a)Anti-cytokine therapy produces responses in patients with myelodysplastic syndromes (MDS). (Abstract). Blood, 88, 580a.
  • 40
    Raza, A., Gezer, S., Mundle, S., Gao, X.Z., Alvi, S., Borok, R., Rifkin, S., Iftikhar, A., Shetty, V., Parcharidou, A., Loew, J., Marcus, B., Khan, Z., Chaney, C., Showel, J., Gregory, S., Preisler, H. (1995) Apoptosis in bone marrow biopsy samples involving stromal and hematopoietic cells in 50 patients with myelodysplastic syndromes. Blood, 86, 268 276.
  • 41
    Raza, A., Mundle, S., Shetty, V., Alvi, S., Chopra, H., Span, L., Parcharidou, A., Dar, S., Venugopal, P., Borok, R., Gezer, S., Showel, J., Loew, J., Robin, E., Rifkin, S., Alston, D., Hernandez, B., Shah, R., Kaizer, H., Gregory, S., Preisler, H. (1996b)A paradigm shift in myelodysplastic syndromes. Leukemia, 10, 1648 1652.
  • 42
    Reems, J.A. & Torok-Storb, B. (1995) Cell cycle and functional differences between CD34+/CD38hi and CD34+/38lo human marrow cells after in vitro cytokine exposure. Blood, 85, 1480 1487.
  • 43
    Sato, T., Irie, S., Kitada, S., Reed, J.C. (1995) FAP-1: a protein tyrosine phosphatase that associates with Fas. Science, 268, 411 415.
  • 44
    Selleri, C., Sato, T., Del Vecchio, L., Luciano, L., Barrett, A.J., Rotoli, B., Young, N.S., Maciejewski, J.P. (1997) Involvement of Fas-mediated apoptosis in the inhibitory effects of interferon-α in chronic myelogenous leukemia. Blood, 89, 957 964.
  • 45
    Shetty, V., Mundle, S., Alvi, S., Showel, M., Broady-Robinson, L., Dar, S., Borok, R., Showel, J., Gregory, S., Rifkin, S., Gezer, S., Parcharidou, A., Venugopal, P., Shah, R., Hernandez, B., Klein, M., Alston, D., Robin, E., Dominquez, C., Raza, A. (1996) Measurement of apoptosis, proliferation and three cytokines in 46 patients with myelodysplastic syndromes. Leukemia Research, 20, 891 900.
  • 46
    Sievers, E.L., Lange, B.J., Buckley, J.D., Smith, F.O., Wells, D.A., Daigneault-Creech, C.A., Shults, K.E., Bernstein, I.D., Loken, M.R. (1996) Prediction of relapse of pediatric acute myeloid leukemia by use of multidimensional flow cytometry. Journal of the National Cancer Institute, 88, 1483 1488.
  • 47
    Smith, C.A., Farrah, T., Goodwin, R.G. (1994) The TNF receptor superfamily of cellular and viral proteins: activation, costimulation, and death. Cell, 76, 959 962.
  • 48
    Smith, M.A. & Smith, J.G. (1991) The occurrence subtype and significance of haemopoietic inhibitory T cells (HIT cells) in myelodysplasia: an in vitro study. Leukemia Research, 15, 597 601.
  • 49
    Srour, E.F., Brandt, J.E., Briddell, R.A., Grigsby, S., Leemhuis, T., Hoffman, R. (1993) Long-term generation and expansion of human primitive hematopoietic progenitor cells in vitro. Blood, 81, 661 669.
  • 50
    Stelzer, G.T., Shults, K.E., Loken, M.R. (1993) CD45 gating for routine flow cytometric analysis of human bone marrow specimens. Annals of the New York Academy of Sciences, 677, 265 280.
  • 51
    Suda, T. & Nagata, S. (1994) Purification and characterization of the Fas-ligand that induces apoptosis. Journal of Experimental Medicine, 179, 873 879.
  • 52
    Suda, T., Okazaki, T., Naito, Y., Yokota, T., Arai, N., Ozaki, S., Nakao, K., Nagata, S. (1995) Expression of the Fas ligand in cells of T cell lineage. Journal of Immunology, 154, 3806 3813.
  • 53
    Suda, T., Takahashi, T., Golstein, P., Nagata, S. (1993) Molecular cloning and expression of the Fas ligand, a novel member of the tumor necrosis factor family. Cell, 75, 1169 1178.
  • 54
    Sutherland, H.J., Lansdorp, P.M., Henkelman, D.H., Eaves, A.C., Eaves, C.J. (1990) Functional characterization of individual human hematopoietic stem cells cultured at limiting dilution on supportive marrow stromal layers. Proceedings of the National Academy of Sciences of the United States of America, 87, 3584 3588.
  • 55
    Tanaka, M., Suda, T., Haze, K., Nakamura, N., Sato, K., Kimura, F., Motoyoshi, K., Mizuki, M., Tagawa, S., Ohga, S., Hatake, K., Drummond, A.H., Nagata, S. (1996) Fas ligand in human serum. Nature Medicine, 2, 317 322.
  • 56
    Terstappen, L.W., Safford, M., Konemann, S., Loken, M.R., Zurlutter, K., Buchner, T., Hiddemann, W., Wormann, B. (1992) Flow cytometric characterization of acute myeloid leukemia. Part II. Phenotypic heterogeneity at diagnosis. Leukemia, 6, 70 80.
  • 57
    Vignaux, F., Vivier, E., Malissen, B., Depraetere, V., Nagata, S., Golstein, P. (1995) TCR/CD3 coupling to Fas-based cytotoxicity. Journal of Experimental Medicine, 181, 781 786.
  • 58
    Watanabe-Fukunaga, R., Brannan, C.I., Itoh, N., Yonehara, S., Copeland, N.G., Jenkins, N.A., Nagata, S. (1992) The cDNA structure, expression, and chromosomal assignment of the mouse Fas antigen. Journal of Immunology, 148, 1274 1279.
  • 59
    Yamaguchi, M., Gersuk, G., Nadler, S., Deeg, H.J. (1996) HLA-DR mediated apoptosis and inhibition of hematopoiesis involves upregulation of tumor necrosis factor alpha. (Abstract). Experimental Hematology, 24, 1100.
  • 60
    Yoshida, Y. (1993) Hypothesis: apoptosis may be the mechanism responsible for the premature intramedullary cell death in the myelodysplastic syndrome. Leukemia, 7, 144 146.