www.elsevier.com/locate/yclim

Clinical Immunology 1

Defective tumor necrosis factor alpha-induced maturation of

monocyte-derived dendritic cells in patients with

myelodysplastic syndromes

Ilina Micheva*, Eleni Thanopoulou, Sotiria Michalopoulou, Marina Karakantza,

Alexandra Kouraklis-Symeonidis, Athanasia Mouzaki, Nicholas Zoumbos

Hematology Division, Department of Internal Medicine, Medical School and University Hospital, University of Patras, Patras, Greece

Received 29 June 2004; accepted 10 August 2004

Available online 18 September 2004

Abstract

Myelodysplastic syndromes (MDS) are clonal stem cell disorders, characterized by ineffective and dysplastic hematopoiesis. MDS

patients have a defective immune response manifested by increased susceptibility to bacterial infections, autoimmune phenomena, and high

incidence of lymphoid malignancies. Presently, we investigated the phenotype and function of monocyte-derived dendritic cells (MoDC) in

23 MDS patients and 15 controls at different stages of differentiation using the maturation stimuli tumor necrosis factor-alpha (TNF-a) and

LPS. Monocytes from MDS patients showed low potential to differentiate into dendritic cells (DC), as determined by low cell yield and

CD1a expression. MDS-MoDCs exhibited low expression of mannose receptor and reduced endocytic capacity. MDS-MoDCs showed a

diminished response to TNF-a with low CD83, CD80, and CD54 expression and allostimulatory capacity. In patients with 5q syndrome,

monocytes and MoDCs were positive for the 5q deletion, suggesting their origin from the malignant clone. Our data indicate that MoDCs in

MDS display quantitative and functional abnormalities that may contribute to the defective immune response of these patients.

D 2004 Elsevier Inc. All rights reserved.

Keywords: Myelodysplastic syndromes (MDS); Dendritic cells (DC); Maturation; Tumor necrosis factor-alpha (TNF-a)

Introduction

Myelodysplastic syndromes (MDS) are clonal disorders

of pluripotent hematopoietic stem cells occurring predom-

inantly in the elderly and characterized by ineffective hema-

topoiesis leading to peripheral blood cytopenias. There exist

clinical and laboratory data suggesting various defects of the

immune response in MDS such as increased susceptibility to

bacterial infections that are mainly due to the reduced

number and function of neutrophils [1], autoimmune

phenomena [2], and high incidence of lymphoid malignan-

cies [1–3]. In addition, progression of MDS to acute

myeloid leukemia and the increased risk of developing

1521-6616/$ - see front matter D 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.clim.2004.08.007

* Corresponding author. Hematology Division, Department of Internal

Medicine, Medical School and University Hospital, University of Patras,

Patras GR-26500, Greece. Fax: +30 2610 993950.

E-mail address: idm@med.upatras.gr (I. Micheva).

malignant tumors in MDS patients have been correlated to

impaired immune surveillance [2]. The reasons for the

immune disturbances observed in MDS patients remain

unclear; they may be due to primary defects of lymphoid

cells, of antigen (Ag) presenting cells (APC), or both.

Dendritic cells (DC) are the most potent APC, intimately

involved in the initiation and regulation of antigen-specific

immune responses [4]. DCs reside in a resting or immature

state in nonlymphoid tissues, where they efficiently capture

and process antigens. Upon activation, they initiate a dif-

ferentiation process that results in decreased Ag-processing

capacity, enhanced expression of MHC and costimulatory

molecules, and migration into secondary lymphoid organs,

where they trigger naive T-cells. This in vivo maturation

process is efficiently regulated and controlled by a complex

array of signals present in the DC microenvironment.

Recently, it has been shown that immature blood-derived

DCs in MDS originate from the malignant clone and exhibit

13 (2004) 310–317

I. Micheva et al. / Clinical Immunology 113 (2004) 310–317 311

phenotypic and functional deficiencies, and it was suggested

that DC dysfunction may account for the defective immune

response noted in MDS patients [5].

Circulating monocytes are considered a potential DC

precursor population, because they can differentiate into

DCs in vivo, providing a boost to the APC population in

sites of significant infection or inflammation [6]. In vitro,

DCs can be generated from peripheral blood monocytes

cultured with GM-CSF and IL-4 [7,8]. The monocyte-

derived DCs (MoDC) can be induced to mature by

immediate exposure to danger mediators such as bacterial

lipopolysaccharides (LPS) or proinflammatory cytokines,

mainly tumor necrosis factor-alpha (TNF-a) [9].

Presently, we studied the potential of peripheral blood

monocytes derived from the peripheral blood of patients

with MDS to differentiate into DCs. We further investigated

the phenotype and function of MoDCs at different stages of

maturation using the maturation stimuli TNF-a and LPS.

Materials and methods

Patients

Twenty-three patients with MDS and fifteen healthy

adults as controls were studied. Patients were diagnosed at

Patras University Hospital. Samples of heparinized blood

(20 ml) were drawn at the time of diagnosis and before the

administration of any treatment and from controls only

once. Informed consent was obtained from both patients and

controls. Patient data are presented in Table 1.

Table 1

Clinical characteristics of the MDS patients

MDS case FAB

category

Age

(years)

Sex

1 RA 67 M

2 RA 80 F

3 RA 60 F

4 RA 65 M

5 RA 72 M

6 RA 68 M

7 RA 62 M

8 RA 71 M

9 RARS 63 M

10 RARS 94 M

11 RARS 76 M

12 RARS 67 M

13 RARS 79 M

14 RAEB 69 F

15 RAEB 82 M

16 RAEB 58 M

17 RAEB 85 F

18 RAEB 69 F

19 RAEB 52 M

20 RAEB 69 F

21 CMML 80 M

22 CMML 61 M

23 CMML 69 M

Cells and culture conditions

Peripheral blood mononuclear cells (PBMC) were iso-

lated from venous blood by density gradient centrifugation

over Ficoll-Hypaque (Biochrom AG, Germany). CD3+ T-

cells were separated from PBMC using CD3 antibody-

coated immunomagnetic beads (Miltenyi Biotec, Glodbach,

Germany). CD3 T-cells were frozen in fetal bovine serum

(FBS) (Biochrom) with 10% dimethyl sulphoxide (DMSO)

(Sigma Chemical Co., St. Louis, MO) until further use.

PBMC were cultured at a concentration of 106 cells/ml in

culture medium (CM) composed of RPMI-1640, 10% FBS,

200 mmol/l l-glutamine, 50 Ag/ml streptomycin, and 50 U/

ml penicillin (all from Biochrom) for 2 h at 378C in a

humidified atmosphere of 5% CO2 in 25 cm2 tissue culture

flasks. The nonadherent cells were removed, the flasks were

washed, and the adherent cells were cultured in CM,

containing 100 ng/ml GM-CSF and 10 ng/ml IL-4

(Biosource, Nivelles, Belgium), for 5 days to obtain

immature DCs. MoDC were further cultured for 2 additional

days in plain CM or in the presence of 10 ng/ml TNF-a

(Biosource) or 0.1 Ag/ml lipopolysaccharide (LPS) (Escher-

ichia coli, serotype 055:B6) (Sigma). On day 5, cells were

counted using a hemocytometer. The percentage cell yield

was calculated as cell number obtained in culture over the

total cell number initially put into culture �100.

Flow cytometric analysis and sorting

Cells from the culture were washed, resuspended in FBS,

left for 1 h at 48C and then incubated with monoclonal

WBC

(� 109/l)

Mo

(%)/PBMC

Cytogenetics

9.2 4 46,XY,del(20q)

3.8 6 46,XX

6.3 6 46,XX

6.7 0.8 46,XY

2.8 0.7 47,XY,+8

5.6 4 46,XY

3.4 3.6 46,XY

3.78 7 46,XY

4.26 3.3 46,XY,del(5q)

8.1 3 46,XY

6.26 3 46,XY

4.8 3.6 46,XY

8.3 8 46,XY

3 4 46,XX

2.4 13 46,XY

1.7 0.5 46,XY

3.87 3 47,XX,+8,del5(q)

2.7 3.7 47,XX,+8

2.9 4 46,XY

5.2 10 46,XX,del(5q)

8.7 17 46,XY

12.5 28 46,XY

24.6 24 46,XY

Fig. 1. Mean DC yield after 5 days of culture in the presence of GM-CSF/

IL-4 for controls and MDS patients with RA, RARS, RAEB, and CMML.

I. Micheva et al. / Clinical Immunology 113 (2004) 310–317312

antibodies (mAbs) for 30 min at 48C. The following mAbs

were used: CD1a-PE, CD80-FITC, CD83-FITC, CD54-PE,

mannose receptor (MR)-PE (PharMingen, BD, Erembode-

gem, Belgium), HLA-DR-PE, CD11c-PE (BD), CD3-FITC

and PE, CD14-FITC (Immunotech, Villepinte, France),

TNFR1 (CD120a)-PE (Caltag Laboratories, Burlingame,

CA), and appropriate isotype-matched negative controls.

Nonviable cells were excluded from the analysis by gating

based on propidium iodide (PI) (PharMingen, BD). Samples

were analyzed on EPICS-XL (Beckman Coulter) flow

cytometer. Phosphate-buffered saline (PBS), supplemented

with 2% FBS was used as a staining buffer (SB) in the flow

cytometric assay. Before fluorescence in situ hybridization

(FISH), CD1a+/CD14� MoDCs were isolated with fluo-

rescence-activated cell sorting (FACS) (FACSVantage, BD)

and then used for cytospin preparation.

Mannose receptor-mediated endocytosis

DCs were incubated in CM with dextran-FITC (MW

40000, Sigma) at a concentration of 1 mg/ml for 30 min at

378C and also at 48C to determine the background uptake.

DCs were washed four times with ice-cold SB in a refrigera-

ted centrifuge and analyzed immediately by flow cytometry.

Dextran uptake was evaluated by the mean fluorescence

intensity (MFI) of the FITC channel and calculated in

arbitrary units (AU) as (MFI378C � MFI48C)/MFI48C.

Mixed lymphocyte reaction (MLR)

MoDCs were treated with mitomycin C (Sigma) (50 mg/

ml) as previously described [10]. DCs were then cocultured

with allogeneic normal CD3+ T-cells in varying ratios (1:5,

1:10, 1:20, and 1:100). Cultures were maintained in a

humidified atmosphere at 378C and 5% CO2 for 5 days.

Bromodeoxyuridin (BrdU), 10 AM, was added for the last

16 h. T-cell proliferation was assessed by the BrdU

incorporation using three-color staining flow cytometry

assay with anti-BrdU-FITC, CD3-PE monoclonal antibod-

ies, and 7-amino-actinomycin-D (7AAD) as instructed by

the manufacturers (PharMingen, BD).

FISH analysis

FISH was performed on CD1a+ MoDCs in two patients

with 5q abnormality. Cells were fixed in methanol–acetic

acid (3:1) and stored frozen at �208C until further use. FISH

analysis was performed using the LS1 EGR1/D5S721:

D5S23 dual-color DNA probe (Vysis Inc., Downers Grove,

USA), which hybridizes to regions 5q31 (SpectrumOrange

LS1 EGR1) and 5p15.2 (SpectrumGreen LS1 D5S72, D5

S23). Cells were denatured, dehydrated in a graded (70%,

85%, 100%) series of ethanol, air-dried, and hybridized

with the probe for a period of 16–18 h. Slides were washed

with x3 in 50% formamide/2� SSC, x1 in 2� SSC, and x1

in 2� SSC/0.1% NP-40. Each wash was at 468C for 10 min.

Slides were then counterstained with 125ng/ml DAPI

(4V,6-diamidine-2V-phenylindole; Boehringer Mannheim,

Germany) and viewed on a Zeiss Axioskop fluorescent

microscope. Image enhancement was performed using the

Isis 3 software (Metasystems GmbH, Germany).

Statistical analysis

Kolmogorov-Smirnov test was initially applied to deter-

mine whether the distribution of the values could be

considered normal. The independent data sets were com-

pared using independent two-tailed t tests. The paired

observations were analyzed using paired t tests. P b 0.05

was considered significant. The Pearson correlation test was

used to assess linear relationships between samples.

Results

Impaired differentiation of MDS blood monocytes into DC

At the end of the culture period (5 days) in the presence

of GM-CSF and IL-4, 8.2% F 3.8% of the PBMC from the

controls and 4.9% F 2.9% of the PBMC from the MDS

patients with RA, RARS, and RAEB differentiated into

MoDCs (P = 0.01). In the three cases with CMML, the cell

yield was higher, compared to the controls and the other

MDS types (Fig. 1). In all MDS cases, there was no

correlation between the percentage of monocytes in the

PBMC fraction at the beginning of the culture (range 0.5–

13%) and the DC yield (r = 0.388), whereas a positive

correlation was calculated for normal donors (r = 0.56).

Phenotypic characteristics of MoDC at different stages of

maturation

Immunophenotypic studies were performed on the

obtained immature MoDCs and on MoDCs stimulated with

TNF-a or LPS. Large cells with increased forward and side

scatter comprised the MoDC population. The percentage of

the viable cells, represented by the PI-negative DC

population, was higher than 92%; their myeloid origin

was confirmed by the expression of myeloid DC marker

CD11c. In MDS patients, the expression of CD1a and CD80

was

significan

tlylower

com

expressio

nof

otherDC-as

(CD83,HLA-D

R,andCD54

patien

tsandcontro

ls(Fig.2,

Theim

matu

reMDS-M

oD

nifican

tlylower

levels

than

M

2.24F

0.5

vs.MFI5.22F

2

Fig.2.P

CSF/IL

-4

gensbef

henotypeofDCsgenerated

andstim

ulated

with

TNF-a

ore

andafter

stimulatio

nin

M

pared

tothe

contro

ls.The

sociated

surfa

ce

molecules

)was

similar

betw

eenMDS

Table

2).

Csexpressed

TNFR1at

sig-

oDCsfro

mthecontro

ls(M

FI

.47,P=0.02)(Fig.3).

from

PBMCsin

thepresen

ceofGM-

orLPS.Expressio

nofDCsurface

anti-

DSpatien

tsandcontro

ls(*Pb0.05).

Table 2

Surface antigen expression (mean MFI F SD) on DCs from MDS patients (MDS) and from healthy subjects (C)

Immature DCa TNF-ab LPSc C—P values MDS—P values

C MDS P C MDS P C MDS P I/T I/L T/L I/T I/L T/L

CDla 11.1 F 6.1 6.8 F 4.7 0.034 4.3 F 2.1 8.2 F 5 0.07 3.7 F 1.8 3 F 2.6 0.552 0.02 0.07 0.577 0.976 0.011 0.01

CD83 0.4 F 0.3 0.47 F 0.3 0.793 2.2 F 1.1 0.9 F 0.6 b0.00l 3.5 F 1.9 2.3 F 1.6 0.155 0.001 0.003 0.087 0.013 0.004 0.002

CD80 2.1 F 0.4 0.7 F 0.5 b0.00l 3.5 F 1.5 1.2 F 1.2 b0.00l 6.2 F 3.9 6.2 F 3.1 0.992 0.033 0.025 0.085 0.099 0.002 0.001

CD54 54.1 F 15.7 62.2 F 27.9 0.328 103.1 F 21 66.4 F 25.1 0.005 88.3 F 9 110.7 F 29.9 0.065 0.001 b0.001 0.091 0.218 0.012 0.006

HLA-DR 54.9 F l8.8 51.1 F 20.4 0.619 117.2 F 38.9 91.8 F 31.6 0.105 180.6 F 32.6 166.7 F 34.5 0.427 0.001 b0.00l 0.004 b0.00l b0.00l b0.00l

MR 24.9 F 7.2 17.3 F 12.8 0.001 11 F 2.1 12.5 F 8.6 0.639 6.6 F 1.5 11.2 F 2.5 0.001 b0.00l 0.006 0.002 0.08 0.08 0.814

Differences between healthy volunteers and patients are analyzed using two-tailed t test; differences between immature and mature DCs and between maturation stimuli are calculated using t test for paired

samples; P values are shown for each analysis.a Immature DC (I).b DC stimulated with TNF-a (T).c DC stimulated with LPS (L).

I.Mich

evaet

al./Clin

icalIm

munology113(2004)310–317

313

Fig. 3. Comparison of the expression of TNFR1 (CD120a) on MoDCs

obtained in GM-CSF/IL-4: (a) results of the flow cytometry analysis of a

representative control and MDS patient; (b) comparison of the mean values

of expression.

Fig. 4. Dextran-FITC endocytosis (a) and MR expression (b) of immature,

TNF-a-, or LPS-matured MoDCs from patients with MDS and controls (*P

b 0.05).

Fig. 5. Allogeneic stimulatory activity of MoDCs. Mitomycin C-treated

DCs were cultured with allogeneic normal CD3 cells in different ratio (a).

T-cell proliferation was assessed by the BrdU incorporation. The mean

percentage of BrdU-positive proliferating lymphocytes for ratio 1:10

stimulator to responder cells before and after stimulation with TNF-a or

LPS in MDS patients and controls is represented on the graph (b) (*P b

0.05).

I. Micheva et al. / Clinical Immunology 113 (2004) 310–317314

In controls, culture of immature MoDCs with LPS and

TNF-a resulted in a significant up-regulation of DC surface

marker CD83, costimulatory molecule CD80, adhesion

molecule CD54, and HLA-DR. In MDS patients LPS was

an efficient stimulator of immature MoDCs, inducing

significant up-regulation of the intensity of expression of

CD83, CD80, HLA-DR, and CD54 and down-regulation of

the expression of CD1a. In contrast, TNF-a failed to induce

maturation of MDS-MoDCs, because alterations of the

expression of CD83, CD80, and CD54 markers were

negligible and remained significantly lower than in controls

(Fig. 2, Table 2).

Low endocytic capacity of MDS-MoDC with low expression

of mannose receptor

The capacity of MoDCs to uptake soluble antigens was

assessed by their ability to endocytose dextran-FITC. The

endocytic capacity of immature MDS-MoDCs was signifi-

cantly reduced in comparison to control-MoDC (2.0 F 1.1

AU vs. 3.42 F 0.6 AU, P = 0.03). When control immature

MoDCs were induced to maturation with either TNF-a or

LPS, a significant down-regulation of the dextran-FITC

endocytosis was observed. MDS-MoDCs, stimulated by

both maturation factors, exhibited no further decrease of

their endocytic activity (Fig. 4a).

The low dextran endocytosis of immature MDS-MoDCs

was accompanied by significantly lower, compared to

controls, expression of MR (P = 0.02) that was slightly

down-regulated upon maturation with both LPS and TNF-a

(Fig. 4b).

Allogeneic stimulatory capacity of MoDC at different stages

of maturation

We investigated the ability of immature as well as LPS-

or TNF-a-stimulated MoDCs to induce proliferation of T-

cells in an allogeneic MLR. The optimal DC–T-cell ratio

was 1:10 (Fig. 5a). Upon TNF-a stimulation, MDS-MoDCs

Fig. 6. FISH analysis of sorted CD1a+ MoDCs from a patient with 5q

syndrome (a) and a healthy subject (b). Normal nuclei appear with four

distinct signals, two green (5p) and two orange (5q), whereas the del

(5)(q13q33) aberration is indicated by nuclei containing two green and one

orange signal. (For interpretation of the references to color in this figure

legend, the reader is referred to the Web version of this article.)

I. Micheva et al. / Clinical Immunology 113 (2004) 310–317 315

failed to up-regulate the proliferative response in T-cells.

Immature MoDCs and MoDCs stimulated with LPS from

both MDS patients and controls did not differ in their ability

to induce proliferation of allogeneic T-cells (Fig. 5b).

Derivation of DC from the dysplastic clone in MDS

To determine the clonal origin of MoDCs in MDS, FISH

analysis was performed on sorted CD1a+ MoDCs from two

patients with 5q syndrome. In both cases, almost all CD1a+

DCs (94% and 95%, respectively) appeared to be derived

from the 5q deletion clone. Cell cytospins from controls

consistently had less than 6% of nuclei with 2 G + 1 O

signal pattern (Fig. 6).

Discussion

MDS are characterized by ineffective hematopoiesis with

impaired proliferation and maturation of one or more cell

lineages in the bone marrow, resulting in PB cytopenia.

Various immune disturbances in MDS such as increased

susceptibility to bacterial infections, autoimmune phenom-

ena, and high incidence of lymphoid malignancies [1–3]

reveal an underlying defect of the immune response in MDS

patients; the reasons for which still remain unclear.

A previous report [5] has shown that, in eight patients

with MDS, the immature MoDCs are phenotypically

defective with decreased expression of CD1a, CD54,

CD80, and class II molecules and exhibit low endocytic

as well as allostimulatory activity. In cases with trisomy 8

and monosomy 7 MDS, MoDCs were positive for the

cytogenetic aberration.

In the present study, we generated DCs from peripheral

blood monocytes in 23 patients with MDS and investigated

their phenotype and function in different stages of matura-

tion. The induction of DC with GM-CSF and IL-4 was

possible in all patients. However, we found that, in RA,

RARS, and RAEB, the yield of MDS-MoDCs was

significantly lower than in controls, whereas no difference

was found in CMML, in agreement to a previous study [11].

The low DC yield together with the fact that MDS

monocytes were positive for the 5q deletion suggests that

monocytes from patients with MDS originate from the

malignant clone and have low potential to differentiate into

DCs.

The immature MDS-MoDCs, generated in the presence

of GM-CSF and IL-4, had a reduced expression of CD1a

and CD80 surface antigens, but no difference was found

concerning the other DC-associated markers. Respectively,

the immature MoDCs in MDS did not differ from controls

in their ability to induce proliferation of allogeneic T-cells.

CD1a is a surface receptor involved in the antigen

uptake, and CD80 is a costimulatory molecule implicated in

the induction of antitumor immunity [12]. The reduced

expression of CD80 may be associated with the increased

incidence of other malignancies, and particularly lymphoid

neoplasms, observed in MDS patients and the escape of the

malignant clone from immune control with progression

towards acute leukemia.

The ability of DCs to take up and process antigens is

highly dependent on the stage of DC differentiation. The

high endocytic capacity is a specific property of cultured

immature DCs and is down-regulated upon maturation [7].

We studied the capacity of MDS-MoDCs to take up soluble

antigens via MR, a major antigen uptake receptor on

immature MoDCs. It has been previously reported that

mannosylated antigen is efficiently internalized by MR on

MoDCs, processed, and presented to T lymphocytes [13] We

found a reduction of the endocytic ability of immature DCs

with an insufficient down-regulation of endocytosis after

TNF-a and LPS induced maturation. We also observed a

significant reduction of the MR expression on DCs derived

from MDS patients, suggesting that a quantitative defect of

MR probably accounts for the impaired endocytosis.

The low endocytic ability may contribute to an inefficient

response to danger signals, contributing thus to the

infectious complications in MDS. Several studies have

shown that excessive apoptosis is a characteristic feature of

MDS [14,15]. DCs are involved in the apoptotic cell

I. Micheva et al. / Clinical Immunology 113 (2004) 310–317316

clearance, although not as efficiently as macrophages.

Physiologically, the apoptotic cell engulfment from DCs

prevents the onset of autoimmune reactions [16]. If

excessive apoptosis coincides with poor clearance, the

persistence of apoptotic bodies may promote an inflamma-

tory response or present a source of antigens [17]. In this

regard, the decreased endocytic capacity of immature DCs

may be implicated in the pathogenesis of autoimmune

disorders commonly seen in MDS.

Another interesting observation in the present study was

that MDS-MoDCs respond differently to the stimulating

agents LPS and TNF-a. LPS, but not TNF-a, induced

complete phenotypic and functional maturation of the MDS-

MoDCs. The bacterial cell wall constituent of LPS functions

as a physiological stimulus for DC maturation. In vitro LPS

stimulates MoDCs to produce IL-6, IL-8, IL-12, and TNF-

a, to up-regulate costimulatory molecules, and to accumu-

late rapidly stable peptide-loaded MHC class II molecules

[18]. TNF-a influences the DC growth and differentiation

characterized mainly by dendritic morphology, expression

of high levels of MHC, adhesion and costimulatory

molecules (HLA-DR, CD40, CD80, CD86), and low or

absent phagocytic but strong antigen-presenting capacity

[19]. Furthermore, it has also been reported that the

mediator of TNF-a induced signaling during DC maturation

is TNFR1 [20].

In the present study TNF-a failed to induce maturation of

MDS-DCs as shown by low expression of surface markers

(CD80, CD83, and CD54) that are presumably important for

DC–T-cell interaction. Besides the phenotypical defect,

TNF-a-stimulated DCs were poor inducers of T-cell

proliferation in MLR.

The maturation failure to TNF-a could be assigned to the

reduced expression of TNFR1 reported in our study. This

finding is of interest, since there is evidence for a role of

TNF-a in MDS pathogenesis. Several studies indicate that

TNF-a is elevated in the bone marrow [21,22] and

peripheral blood [23–25] of MDS patients. It is conceivable

that the chronic exposure to TNF-a could result in down-

regulation of TNFR1.

Alternatively, the differential effect of TNF-a and LPS

on the maturation of DCs could be related to defects of the

intracellular pathways triggered by TNF-a. The activation

of NF-nB family members is a critical control pathway for

differentiation of monocytes to DCs and for maturation of

DCs from antigen-processing to antigen-presenting cells

[26]. NF-nB transcription factor regulates a large number of

genes involved in immune responses, such as the proin-

flammatory cytokines (IL-1, IL-6, TNF-a) and cell surface

molecules including CD80 [27]. In this context, it can be

hypothesized that the difference in MDS-MoDCs matura-

tion by TNF-a and LPS reflects an underlying defect in the

capacity of TNF-a to activate NF-nB. It has been reported

that the capacity of DCs to activate T-cells following

CD40L treatment was enhanced compared with TNF-a

treatment, and this effect was NF-nB-dependent [28].

In conclusion, MDS-MoDCs belong to the abnormal

MDS clone and exhibit quantitative and functional abnor-

malities. Although functionally active when stimulated with

LPS, the MDS-MoDCs were numerically deficient and

showed low endocytic ability and impaired maturation after

TNF-a stimulation. This constellation of DC defects may

contribute to the defective immune response against

pathogens, escape and expansion of the malignant clone,

as well as autoimmune phenomena, observed in MDS

patients.

References

[1] P. Ruutu, Granulocyte function in myelodysplastic syndromes, Scand.

J. Haematol. 45 (1986) 66–70.

[2] T.J. Hamblin, Immunological abnormalities in myelodysplastic

syndromes, Semin. Hematol. 33 (1996) 150–162.

[3] J.A. Copplestone, G.J. Mufti, T.J. Hamblin, D.G. Oscier, Immuno-

logical abnormalities in myelodysplastic syndromes. II. Coexistent

lymphoid or plasma cell neoplasms: a report of 20 cases unrelated to

chemotherapy, Br. J. Haematol. 63 (1986) 149–159.

[4] D.N.J. Hart, Dendritic cells: unique leukocyte populations which

control the primary immune response, Blood 90 (1997) 3245–3287.

[5] G.M. Rigolin, J. Howard, A. Buggins, G. Sneddon, G. Castoldi,

W.J.R. Hirst, G.J. Mufti, Phenotypic and functional characteristics of

monocyte-derived dendritic cells from patients with myelodysplastic

syndromes, Br. J. Haematol. 107 (1999) 844–850.

[6] G.J. Randolph, K. Inaba, D.F. Robbiani, R.M. Steiman, W.A. Muller,

Differentiation of phagocytic monocytes into lymph node dendritic

cells in vivo, Immunity 11 (1999) 753–761.

[7] F. Sallusto, A. Lanzavecchia, Efficient presentation of soluble antigen

by cultured human dendritic cells is maintained by granulocyte/

macrophage colony-stimulating factor plus interleukin 4 and down-

regulated by tumor necrosis factor alpha, J. Exp. Med. 179 (1994)

1109–1118.

[8] L.J. Zhou, T.F Tedder, CD14 blood monocytes can differentiate

into functionally mature CD83 dendritic cells, Proc. Nat. Acad. Sci.

U. S. A. 93 (1996) 2588–2592.

[9] J.W. Young, Dendritic cells: expansion and differentiation with

hematopoietic growth factors, Curr. Opin. Hematol. 6 (1999)

135–144.

[10] L.L. Cavanagh, R.J. Saal, K.L. Grimmet, R. Thomas, Proliferation in

monocytes-derived dendritic cell cultures is caused by progenitor cells

capable of myeloid differentiation, Blood 92 (1998) 1598–1607.

[11] S. Vuckovic, D.B. Fearnley, S. Gunningham, R.L. Spearing, W.N.

Patton, D.N. Hart, Dendritic cells in chronic myelomonocytic

leukaemia, Br. J. Haematol. 105 (1999) 974–985.

[12] E.C. Guinan, J.G. Gribben, V.A. Boussiotis, G.J. Freeman, L.M.

Nadler, Pivotal role of the B7:CD28 pathway in transplantation

tolerance and tumor immunity, Blood 84 (1994) 3261–3282.

[13] A.J. Engering, M. Cella, D. Fluitsma, M. Brockhaus, E.C. Hoefsmit,

A. Lanzavecchia, J. Pieters, The mannose receptor functions as a high

capacity and broad specificity antigen receptor in human dendritic

cells, Eur. J. Immunol. 27 (1997) 2417–2425.

[14] P. Tsoplou, A. Kouraklis-Symeonidis, E. Thanopoulou, P. Zikos,

V. Orphanos, N.C. Zoumbos, Apoptosis in patients with myelodys-

plastic syndromes: differential involvement of marrow cells in

dgoodT versus dpoorT prognosis patients and correlation with

apoptosis-related genes, Leukemia 13 (1999) 1554–1563.

[15] J.A. Parker, G.J. Mufti, F. Rasool, A. Mijovic, S. Devereux, A.

Pagliuca, The role of apoptosis, proliferation, and the Bcl-2-related

proteins in the myelodysplastic syndromes and acute myeloid

leukemia secondary to MDS, Blood 96 (2000) 3932–3938.

I. Micheva et al. / Clinical Immunology 113 (2004) 310–317 317

[16] E.V. Sun, Y.F. Shi, Apoptosis: the quiet death silences the immune

system, Pharmacol. Ther. 92 (2001) 135–145.

[17] A. Rosen, L. Casiola-Rosen, Clearing the way to mechanisms of

autoimmunity, Nat. Med. 7 (2001) 664–665.

[18] M. Cella, A. Engering, V. Pinet, J. Pieters, A. Lanzavecchia,

Inflammatory stimuli induce accumulation of MHC class II com-

plexes on dendritic cells, Nature 388 (1997) 782–787.

[19] B. Chen, Y. Shi, J.D. Smith, D. Choi, J.D. Geiger, J.J. Mule, The role

of tumor necrosis factor-a in modulating the quantity of peripheral

blood-derived, cytokine driven human dendritic cells and its role in

enhancing the quality of dendritic cell function in presenting soluble

antigens to CD4+ T cells in vitro, Blood 91 (1998) 4652–4661.

[20] F. Lardon, H.W. Snoeck, Z.N. Berneman, V.F.I. Van Tendeloo, G. Nijs,

M. Lenjou, E. Henckaerts, C.J. Boeckxtaens, P. Vandenabeele, L.L.

Kestens, D.R. Van Bockstaele, G.L. Vanham, Generation of dendritic

cells from bone marrow progenitors using GM-CSF, TNF-a, and

additional cytokines: antagonistic effects of IL-4 and IFN-g and

selective involvement of TNF-a receptor-1, Immunology 91 (1997)

553–559.

[21] M. Kitagawa, I. Saito, T. Kuwata, S. Yoshida, S. Yamaguchi, M.

Takahashi, T. Tanizawa, R. Kamiyama, K. Hirokawa, Overexertion of

tumor necrosis factor (TNF)-alpha and interferon (IFN)-gamma by

bone marrow cells from patients with myelodysplastic syndromes,

Leukemia 11 (1997) 2049–2054.

[22] V. Shetty, S. Mundle, S. Alvi, M. Showel, L. Broady-Robinson, S. Dar,

R. Borok, J. Showel, S. Gregory, S. Rifkin, S. Gezer, A. Parcharidou, P.

Venugopal, R. Shah, B. Hernandez, M. Klein, D. Alston, E. Robin, C.

Dominquez, A. Raza, Measurement of apoptosis, proliferation and

three cytokines in 46 patients with myelodysplastic syndromes, Leuk.

Res. 20 (1996) 891–900.

[23] N. Zoumbos, A. Symeonidis, A. Kourakli, P. Katevas, P. Matsouka,

M. Perraki, V. Georgoulias, Increased levels of soluble interleukin-2

receptor and tumor necrosis factor in serum of patients with

myelodysplastic syndromes, Blood 77 (1991) 413–414.

[24] M. Koike, T. Ishiama, S. Tomoyasu, N. Tsuruoka, Spontaneous

cytokine overproduction by peripheral blood mononuclear cells from

patients with myelodysplastic syndromes and aplastic anemia, Leuk.

Res. 19 (1995) 639–644.

[25] L. Molnar, T. Berki, A. Hussain, P. Nemeth, H. Losonczy, Detection

of TNF alpha expression in the bone marrow and determination of

TNF alpha production of peripheral blood mononuclear cells in

myelodysplastic syndrome, Pathol. Oncol. Res. 6 (2000) 18–23.

[26] I. Baltathakis, O. Alcantara, D.H. Boldt, Expression of different NF-

nB pathway in dendritic cells (DCs) or macrophages assessed by gene

expression profiling, J. Cell. Biochem. 83 (2001) 281–290.

[27] A.S. Balwin, The NF-nB and InB proteins: new discoveries and

insights, Annu. Rev. Immunol. 14 (1995) 649–681.

[28] B.J. O’Sullivan, T. Ranjeny, CD40 ligation conditions dendritic cell

antigen-presenting function through sustained activation of NF-nB,J. Immunol. 168 (2002) 5491–5498.