Increased expression of cd40 on bone marrow cd34+ hematopoietic progenitor cells in patients with systemic lupus erythematosus: contribution to fas-mediated apoptosis

Vol. 60, No. 2, February 2009, pp 543–552 Increased Expression of CD40 on Bone Marrow CD34ϩ Hematopoietic Progenitor Cells in Patients With Katerina Pyrovolaki, Irene Mavroudi, Prodromos Sidiropoulos, Aristides G. Eliopoulos, Dimitrios T. Boumpas, and Helen A. Papadaki Objective. Patients with systemic lupus erythem-
portion of apoptotic cells and decreased the proportion
atosus (SLE) display increased apoptosis of bone mar-
of colony-forming cells as compared with untreated
row (BM) CD34؉ hematopoietic progenitor cells. This
cultures. The CD40L-mediated effects were amplified
study was undertaken to evaluate the expression of
following treatment with recombinant Fas ligand, sug-
CD40 and CD40L in the BM of SLE patients, and to
gesting that the effects of these ligands are synergistic.
explore the possible involvement of these molecules in
CD40L levels were significantly increased in long-term
apoptosis of CD34؉ cells.
BM culture supernatants and adherent layers of BM
Methods. The proportion and survival character-
cells from SLE patients as compared with controls.
istics of CD40؉ cells within the BM CD34؉ fraction
Conclusion. These data reveal a novel role for the
from SLE patients and healthy controls were evaluated
CD40/CD40L dyad in SLE by demonstrating that up-
by flow cytometry. The production of CD40L by BM
regulation and induction of CD40 on BM CD34؉ cells
stromal cells was assessed using long-term BM cultures,
from patients with SLE contribute to the amplification
and the effect of CD40L on the survival characteristics
of Fas-mediated apoptosis of progenitor cells.
and clonogenic potential of CD34؉ cells was evaluated
ex vivo by flow cytometry and clonogenic assays.
The emerging role of hematopoietic and mesen- Results. SLE patients displayed an increased
chymal stem cells in the pathophysiology and treatment proportion of CD40؉ cells within the CD34؉ fraction
of autoimmune diseases has stimulated interest in the as compared with controls. The CD34؉CD40؉ sub-
biologic properties of bone marrow (BM)–derived stem population contained an increased proportion of apo-
cells in these patients (1–4). In this context, we have ptotic cells compared with the CD34؉CD40؊ fraction
previously shown that patients with systemic lupus ery- in patients and controls, suggesting that CD40 is in-
thematosus (SLE) display low numbers of BM CD34ϩ volved in the apoptosis of CD34؉ cells. Stimulation of
hematopoietic progenitor cells, due to induction of patients’ CD34؉ cells with CD40L increased the pro-
apoptosis by autoreactive T lymphocytes (5). Thesefindings have been further substantiated by studies Dr. Eliopoulos’ work was supported in part by a grant from demonstrating low numbers of peripheral blood (PB) the Association of Cancer Research, UK.
CD34ϩ cells in association with increased apoptosis (6), Katerina Pyrovolaki, MD, Irene Mavroudi, BSc, Prodromos Sidiropoulos, MD, Aristides G. Eliopoulos, PhD, Dimitrios T.
as well as abnormal immunophenotypic characteristics Boumpas, MD, FACP, Helen A. Papadaki, MD, PhD: University of of BM CD34ϩ cells (7) in SLE patients. In addition, Crete School of Medicine, Heraklion, Crete, Greece.
clinical data have shown a low PB CD34ϩ progenitor Address correspondence and reprint requests to Helen A.
Papadaki, MD, PhD, Professor of Hematology, University Hospital of cell yield in SLE patients undergoing autologous stem Heraklion, PO Box 1352, Heraklion, Crete, Greece. E-mail: epapadak cell transplantation, in comparison with that in patients Submitted for publication May 2, 2008; accepted in revised Regarding the underlying pathophysiologic mecha- nisms, it has been shown that the Fas and Fas ligand in this study. All patients satisfied the American College of (FasL) systems, belonging to the tumor necrosis factor Rheumatology revised criteria for the diagnosis of SLE andhad evidence of active disease according to the established (TNF) receptor (TNFR) and TNF families, respectively, criteria (27). The SLE Disease Activity Index (SLEDAI) score is implicated, at least in part, in the apoptotic process of (scale 0–100) (28) was Ն4 in all patients. The characteristics of hematopoietic progenitor cells in SLE (5,7). Specifically, the patients are summarized in Table 1. Patients had discon- it has been suggested that increased local production of tinued any medication for at least 24 hours prior to BM interferon-␥ and FasL by autoreactive T lymphocytes aspiration. As controls, 20 hematologically healthy subjects,age- and sex-matched with the patients, were studied. Institu- results in up-regulation of Fas and apoptosis of CD34ϩ tional ethics committee approval was granted prior to the study. Informed consent, in accordance with the Declaration of CD40, a TNFR family member, and its ligand, Helsinki, was obtained from all patients.
CD40L (CD154) (10), have also been implicated in the BM sample preparation. BM aspirates (10 ml) ob-
pathophysiology of SLE (11,12). It has been shown that tained from the posterior iliac crest of patients and healthycontrols were immediately diluted 1:1 in Iscove’s modified overexpression of CD40L by activated T cells triggers a Dulbecco’s medium (IMDM; Gibco Invitrogen, Paisley, Scot- cascade of events in CD40ϩ target cells, resulting in land) supplemented with 100 IU/ml penicillin–streptomycin humoral immune dysregulation and autoantibody pro- (Gibco Invitrogen) and 10 IU/ml preservative-free heparin duction in SLE (12–15). Interestingly, a biologically (Sigma-Aldrich, St. Louis, MO). Diluted BM samples werecentrifuged on Histopaque 1077 (density 1.077 gm/cm3; Sigma- active, membrane-cleaved, soluble form of CD40L Aldrich) at 400g for 30 minutes at room temperature to obtain (sCD40L) has been identified and found to be increased BM mononuclear cells (BMMCs). Cell numbers and viability in patients’ sera, which contributes to the pathophysiol- were assessed after staining with trypan blue.
ogy of the disease by further augmenting the CD40- Immunophenotyping and 7-aminoactinomycin D (7-
AAD) staining. An indirect immunofluorescence technique
was used to evaluate the expression of CD40 and Fas antigens
The possible distribution of CD40 and its func- within the BM CD34ϩ cell fraction. Briefly, 1 ϫ 106 BMMCs tion on hematopoietic progenitor cells in SLE are, were stained with phycoerythrin (PE)–conjugated mouse however, entirely unknown. It is of interest that although anti-human CD34 monoclonal antibody (mAb) (QBEND-10; CD40–CD40L interactions have been associated with Immunotech, Marseilles, France) and fluorescein isothiocya- the rapid expansion of normal B lymphocytes, several nate (FITC)–conjugated mouse anti-human CD40 (5C3; BDBiosciences-PharMingen, San Jose, CA) or Fas (CD95) studies have shown that the activation of CD40 in (LOB 3/17; Serotec, Kidlington, UK) mAb. PE- and FITC- lymphoma, melanoma, carcinoma cells, and hepatocytes conjugated mouse IgG isotype-matched controls were used as results in suppression of cell proliferation and/or induc- negative controls. Following 30 minutes of incubation on ice, tion of apoptosis (19–23). One of the mechanisms cells were washed twice in phosphate buffered saline proposed to explain the proapoptotic function of CD40 (PBS)/1% fetal calf serum (FCS) (Gibco Invitrogen)/0.05%azide and were further stained with 7-AAD (Calbiochem- in malignant cells and hepatocytes is the CD40-inducible Novabiochem, La Jolla, CA) for the evaluation of the propor- up-regulation of Fas (23–26). However, whether such a tion of apoptotic cells, as previously described (29). Briefly, cooperative interaction between the CD40 and Fas 100 ␮l 7-AAD solution (200 ␮g/ml) was added to the cells, pathways is also implicated in the apoptosis of hemato- followed by suspension in 1 ml PBS and incubation on ice for poietic stem cells in SLE is unknown. Furthermore, 20 minutes in the dark. Following centrifugation, the superna-tant was removed, and cells were fixed in 500 ␮l 2% parafor- although several studies have demonstrated increased maldehyde solution (Sigma-Aldrich). Unstained fixed cells amounts of circulating sCD40L in SLE, the cytokine levels in the SLE BM microenvironment and the possi- Cell samples were analyzed on an Epics Elite model ble pathophysiologic significance of these cytokines in flow cytometer (Coulter, Miami, FL) within 30 minutes of SLE have so far not been investigated. Therefore, the fixation. Data were acquired and processed on 500,000 eventsto evaluate the proportion of CD34ϩ cells within the BMMCs, present study was performed to evaluate the expression the percentage of CD40ϩ and Fasϩ cells in the CD34ϩ cell of CD40/CD40L in BM hematopoietic progenitor cells fraction, and the proportion of 7-AAD–negative (live), and the BM microenvironment in patients with SLE, and 7-AAD–dim (apoptotic), and 7-AAD–bright (dead) cells to explore the possible involvement of this receptor/ within the CD40- or Fas-expressing CD34ϩ cells.
ligand dyad in the apoptotic depletion of CD34ϩ cells.
Effect of CD40L on the survival characteristics and
clonogenic potential of CD34؉ cells. Flow cytometric analysis.
CD34ϩ cells were isolated from the BMMCs of SLE patients
PATIENTS AND METHODS
and healthy controls by magnetic-activated cell sorting(MACS) analysis (MACS isolation kit; Miltenyi Biotec, Ber- Patients. Twenty patients with SLE (19 women and 1
gisch Gladbach, Germany) according to the manufacturer’s man; median age 43 years [range 19–73 years]) were enrolled protocol. The purity of the CD34ϩ cells was more than 96% in EXPRESSION OF CD40 AND Fas-MEDIATED APOPTOSIS OF BM CD34ϩ CELLS IN SLE Clinical and laboratory characteristics of the patients with systemic lupus erythematosus* * SLEDAI ϭ Systemic Lupus Erythematosus Disease Activity Index (scale 0–100); ANA ϭ antinuclear antibody; anti-dsDNA ϭ antibodies againstdouble-stranded DNA; Hgb ϭ hemoglobin; CNS ϭ central nervous system; PDN ϭ prednisolone; HCQ ϭ hydroxychloroquine; CYC ϭcyclophosphamide; MMF ϭ mycophenolate mofetil; ND ϭ not determined.
all experiments. We then cultured CD34ϩ cells in 96 round- results were expressed as the total number of colony-forming bottomed well plates at a density of 1 ϫ 104 CD34ϩ cells/well in the presence of 1 ␮g/ml recombinant human CD40L Long-term BM cultures (LTBMCs). For preparation
(rHuCD40L) (BMS308/2; Bender MedSystems, Vienna, Aus- of LTBMCs, 107 BMMCs were grown according to a standard tria) and/or 1 ␮g/ml rHuFasL (BMS309/2; Bender MedSys- technique (30), in 10 ml IMDM supplemented with 10% FCS, tems) in 200 ␮l IMDM/1% FCS. Following 48 hours of 10% horse serum (Gibco Invitrogen), 100 IU/ml penicillin– incubation, the cells were stained with 7-AAD in the same streptomycin, 2 mM L-glutamine, and 10Ϫ6M hydrocortisone manner as described above, for the evaluation of the propor- sodium succinate (Sigma-Aldrich), and incubated at 33°C in a tion of live, apoptotic, and dead CD34ϩ cells. In a number of fully humidified atmosphere of 5% CO . At weekly intervals, experiments, Fas antigen expression was also evaluated on cultures were examined for stromal layer formation, using an CD34ϩ cells that had been treated in the same manner as inverted microscope, and were fed by removing half of the described above, with 1 ␮g/ml rHuCD40L.
medium and replacing it with an equal volume of fresh IMDM Clonogenic assay. We cultured 3 ϫ 103 CD34ϩ cells supplemented as described above. At weeks 3–4, when a from SLE patients and normal controls in 1 ml IMDM confluent stromal layer was formed, cell-free supernatants supplemented with 30% FCS, 1% bovine serum albumin were stored at Ϫ72°C for the determination of sCD40L, using (BSA) (Gibco Invitrogen), 10Ϫ4M mercaptoethanol (Sigma- a commercially available (Bender MedSystems) enzyme-linked Aldrich), 0.075% sodium bicarbonate (Gibco Invitrogen), immunoabsorbent assay (ELISA). The sensitivity of the assay 2 mM L-glutamine (Sigma-Aldrich), and 0.9% methylcellulose is 0.6 ng/ml. The culture samples were tested by ELISA after (StemCell Technologies, Vancouver, British Columbia, Can- being concentrated using a Speedvac method (31).
ada), in the presence of 5 ng granulocyte–macrophage colony- For the assessment of CD40L expression in LTBMCs stimulating factor (R&D Systems, Minneapolis, MN), 50 ng of stroma, total messenger RNA (mRNA) was extracted from interleukin-3 (R&D Systems), and 2 IU erythropoietin the adherent cells of confluent LTBMCs, using the RNeasy (Janssen-Cilag, Bucks, UK) in the presence or absence of 1 ␮g mini kit (Qiagen, Hilden, Germany) according to the manu- rHuCD40L and/or 1 ␮g rHuFasL. Cultures were set up in facturer’s instructions. Contaminating DNA was removed by duplicate in 35-mm petri dishes and incubated at 37°C in a fully digestion with RNAse-free DNAse. The SuperScript Pream- humidified atmosphere of 5% CO . In a separate set of plification System (Gibco Invitrogen) was used for first-strand experiments, 105 BMMCs were cultured as described above, in complementary DNA (cDNA) synthesis from 1 ␮g total RNA, the presence or absence of 1 ␮g/ml CD40L neutralizing followed by reverse transcription–polymerase chain reaction antibody (MK13A4; Bender MedSystems) or with the same (PCR) with specific primers. PCR products were normalized quantity of PBS as a control. In all cases, on day 14 of the according to the amount of ␤ -microglobulin (␤ m) in the cultures, colonies were scored as previously described, and samples. We performed one tube reaction, and the primer Flow cytometric analysis of BMMCs in SLE patients and healthy controls* CD34ϩFasϩ cell fraction, 7-AADdim cells CD34ϩFasϪ cell fraction, 7-AADdim cells CD34ϩCD40ϩ cell fraction, 7-AADdim cells CD34ϩCD40Ϫ cell fraction, 7-AADdim cells * Except where indicated otherwise, values are the mean Ϯ SD. BMMC ϭ bone marrow mononuclear cell; SLE ϭ systemic lupus erythematosus;7-AADdim ϭ 7-aminoactinomycin D–dim (apoptotic).
concentration of CD40L was 7 times higher than the concen- Table 2. In accordance with previously reported data (5), the proportion of CD34ϩ cells within the BMMC frac- The forward and reverse primers were 5Ј-AGAATC- CTCAAATTGCGGC-3Ј and 5Ј-TGTGGGTATTTGCAGCT- tion was statistically significantly lower in SLE patients CTG-3Ј, respectively, for CD40L (PCR product size 286 bp), (mean Ϯ SD 1.14 Ϯ 0.35%) compared with healthy and 5Ј-TCCAACATCAACATCTTGGT-3Ј and 5Ј-TCCCCC- controls (2.06 Ϯ 0.62%; P Ͻ 0.0001), and the proportion AAATTCTAAGCAGA-3Ј, respectively, for ␤ m (PCR prod- of Fasϩ cells within the CD34ϩ cell compartment was uct size 123 bp). Conditions for 37 cycles of PCR amplificationfollowing initial denaturation were 94°C for 30 seconds, 62°C significantly higher in patients (20.69 Ϯ 19.53%) com- for 30 seconds, and 72°C for 45 seconds. PCR products were pared with controls (7.16 Ϯ 4.46%; P ϭ 0.0138). The electrophoresed on a 1.5% agarose gel and visualized under percentage of CD40-expressing CD34ϩ cells was signif- ultraviolet light by ethidium bromide staining. The positive icantly increased in SLE patients (10.79 Ϯ 6.90%) control was cDNA from the Ms5-CD40L–transfected cell line(provided by one of us [AGE]). Results were analyzed using compared with healthy controls (5.41 Ϯ 2.89%; P ϭ the ImageJ analysis program (National Institutes of Health, 0.0061). Moreover, a highly significant correlation was found between the proportion of CD40-expressing Statistical analysis. Data were analyzed using the
CD34ϩ cells and the proportion of Fas-expressing GraphPad Prism statistical program (GraphPad Software, SanDiego, CA) by means of the nonparametric Mann-Whitney CD34ϩ cells (r ϭ 0.5358, P ϭ 0.0004), suggesting a and Spearman’s tests. Student’s t-test for paired samples was parallel mode of expression of the 2 molecules in the used to compare differences in the proportion of Fasϩ cells study cell population (Figure 1). Notably, no statistically and the number of CFCs between cultures treated and those significant difference was found between patients and not treated with rHuCD40L or CD40L neutralizing antibody.
Group data are expressed as the mean Ϯ SD.
controls in the proportion of CD40ϩ cells detected inthe non-CD34ϩ BMMC fraction (7.03 Ϯ 2.65% and6.13 Ϯ 2.43%, respectively; P ϭ 0.818).
Survival characteristics of the CD34؉CD40؉
Expression of CD40 and Fas antigens on CD34؉
cells. Data on the survival characteristics of the patients’
cells. Results from flow cytometric analysis of the
CD34ϩ cells are presented in Table 2. In accordance CD34ϩ cells from patients with SLE are presented in with the results from previously reported studies (5), the EXPRESSION OF CD40 AND Fas-MEDIATED APOPTOSIS OF BM CD34ϩ CELLS IN SLE prominent role of the Fas antigen in the apoptosis ofBM CD34ϩ cells (5,32,33). Interestingly, an increasedrate of apoptosis was also observed in the CD34ϩFasϪcell fraction in the patients compared with the controls(P ϭ 0.0066), suggesting that additional mechanismsother than up-regulation of Fas antigen expression areoperating in the apoptotic depletion of CD34ϩ cells inSLE.
To investigate the possible involvement of CD40 in the apoptotic process of CD34ϩ cells, we studied thesurvival characteristics of the CD40ϩ and CD40Ϫ cellsubpopulations. In the gate of CD34ϩ cells, the propor-tion of apoptotic cells was significantly higher among theCD40ϩ cells than among the CD40Ϫ cells, in both thepatients (37.43 Ϯ 18.24% versus 11.85 Ϯ 9.72%; P Ͻ0.0001) and the controls (23.07 Ϯ 10.03% versus 5.41 Ϯ6.41%; P Ͻ 0.0001) (Table 2). In the gate of CD34Ϫcells, however, no statistically significant difference inthe percentage of apoptotic cells was documented be-tween the CD40ϩ and CD40Ϫ cell subpopulations, ineither the patients (2.40 Ϯ 2.42% versus 3.30 Ϯ 3.65%; P ϭ 0.5792) or the controls (2.47 Ϯ 1.80% versus 1.65 Ϯ1.64%; P ϭ 0.638). Furthermore, among all of thesubjects studied, a highly significant correlation wasnoted between the proportion of CD40ϩ cells and thepercentage of apoptotic cells within the CD34ϩ sub-population (r ϭ 0.4624, P ϭ 0.0027) (Figure 1), but not Figure 1. Correlations between the proportion of CD40ϩ cells and
within the non-CD34ϩ subpopulation (r ϭ 0.1887, P ϭ the percentages of Fasϩ cells (top) and apoptotic cells (bottom) within
0.818). These results indicate that overexpression of the the CD34ϩ cell fraction of bone marrow from patients with systemic CD40 antigen may have a role in the apoptotic depletion lupus erythematosus (n ϭ 20; solid circles) and healthy subjects (n ϭ20; open circles), as determined by linear regression analysis. Regres- of BM CD34ϩ cells in patients with SLE.
sion lines and the 95% confidence intervals are shown as solid lines All of the patients studied had active disease.
However, CD40 expression on CD34ϩ cells did notcorrelate with parameters of disease activity, such as theSLEDAI score, albumin levels, antinuclear antibody proportion of apoptotic cells within the CD34ϩ com- titers, and C3 values. Nevertheless, there was a correla- partment was significantly increased in SLE patients tion with hemoglobin levels (r ϭ Ϫ0.6689, P ϭ 0.0013), (26.03 Ϯ 26.40%) compared with healthy controls further indicating a possible apoptosis-inducing role of (4.47 Ϯ 2.21%; P Ͻ 0.0001). The percentage of apopto- CD40 on hematopoiesis in patients with SLE.
tic cells was higher in the CD34ϩFasϩ cell compart- Effect of CD40L on CD34؉ hematopoietic pro-
ment compared with the CD34ϩFasϪ cell compart- genitor cell death. To investigate the hypothesis that the
ment, in both the SLE patients (43.50 Ϯ 24.56% versus CD40 antigen may directly and/or indirectly have a role 22.89 Ϯ 28.38%; P ϭ 0.0098) and the healthy controls in the apoptotic process of CD34ϩ cells, we evaluated (25.60 Ϯ 14.15% versus 3.24 Ϯ 2.25%; P Ͻ 0.0001), the proportion of apoptotic cells following incubation of whereas no significant difference was documented in the purified CD34ϩ cells from SLE patients (n ϭ 3) and proportion of apoptotic cells detected in the Fasϩ and healthy controls (n ϭ 3) with rHuCD40L and/or rHu- FasϪ subpopulations of the non-CD34ϩ cell compart- FasL. The results are presented in Figure 2A. In SLE ment, in either the patients (2.87 Ϯ 2.60% and 3.74 Ϯ patients, the addition of rHuFasL significantly increased 3.97%, respectively; P ϭ 0.6848) or the controls (1.89 Ϯ the proportion of apoptotic cells (33.93 Ϯ 3.52%) com- 2.07% and 1.62 Ϯ 1.35%, respectively; P ϭ 0.7452).
pared with untreated (baseline) cultures (24.27 Ϯ These findings further highlight the previously reported 3.78%; P ϭ 0.0317). The addition of rHuCD40L also apoptotic cells compared with that in untreated cultures(53.26 Ϯ 11.23%; P ϭ 0.0013), compared with that incultures treated with rHuFasL alone (P ϭ 0.0466), orcompared with that in cultures treated with rHuCD40Lalone (P ϭ 0.0024). In healthy controls, the addition ofrHuCD40L and/or rHuFasL did not result in statisticallysignificant increases in the proportion of apoptotic cellsas compared with that in untreated cultures, probablybecause of the low levels of expression of surface CD40and Fas on normal CD34ϩ cells.
These data suggest that CD40L displays an indi- rect, rather than a direct, apoptosis-inducing effect onCD40-expressing CD34ϩ cells, by facilitating the FasL-mediated apoptotic process. To substantiate this hy-pothesis, we evaluated Fas antigen expression on immu-nomagnetically sorted CD34ϩ cells from SLE patients(n ϭ 12) following a 48-hour incubation with rHuCD40L.
We found that the proportion of Fasϩ cells significantlyincreased in cultures treated with rHuCD40L (48.60 Ϯ12.63%) as compared with that in untreated cultures(23.71 Ϯ 7.47%; P Ͻ 0.0001). The up-regulation of Fasexpression by CD40L may therefore explain, at least inpart, the amplifying effects of CD40L on Fas-mediatedCD34ϩ cell death.
Effect of CD40L on the clonogenic potential of
CD34؉ cells. To further characterize the effect of
CD40L on the function of CD34ϩ cells, we investigated
the clonogenic potential of CD34ϩ cells from SLE
patients (n ϭ 11) and healthy controls (n ϭ 11) in the
presence of rHuCD40L and/or rHuFasL. Results are
depicted in Figure 2B.
In cultures of patients’ cells treated with rHu- FasL, a significant decrease in CFC number (36.54 Ϯ Figure 2. Effect of CD40L on the survival characteristics and clono-
25.46 CFCs per 3 ϫ 103 CD34ϩ cells) was observed in genic potential of CD34ϩ cells. A, Proportion of apoptotic CD34ϩ
cells in the bone marrow of patients with systemic lupus erythematosus
comparison with that in untreated cultures (62.09 Ϯ (SLE) (n ϭ 3) and healthy subjects (n ϭ 3) following 48-hour 22.09 CFCs per 3 ϫ 103 CD34ϩ cells; P ϭ 0.0418). In incubation of the cells with recombinant human Fas ligand (rhFasL) (1 the presence of rHuCD40L, CFC numbers were simi- ␮g/ml) and/or recombinant human CD40L (1 ␮g/ml). Apoptosis was larly decreased (45.73 Ϯ 28.16 CFCs per 3 ϫ 103 CD34ϩ assessed by flow cytometry and 7-aminoactinomycin D staining. B,
cells) as compared with baseline; however, the differ- Numbers of colony-forming cells (CFCs) per 3 ϫ 103 CD34ϩ cells,determined using a clonogenic assay, in patients with SLE (n ence was not statistically significant. A profound de- healthy subjects (n ϭ 11) following 14-day culture of the cells with crease in CFC number was observed in the presence of recombinant human FasL (1 ␮g/ml) and/or recombinant human CD40L both rHuCD40L and rHuFasL (18.45 Ϯ 12.10 CFCs per (1 ␮g/ml). In both series of experiments, treated cultures were compared 3 ϫ 103 CD34ϩ cells) as compared with that in un- with untreated cultures (baseline) using the nonparametric Mann- treated cultures (P ϭ 0.0001) or compared with that in Whitney test. Bars show the mean and SD.
cultures treated with rHuCD40L alone (P ϭ 0.0256) orrHuFasL alone (P ϭ 0. 0459). Clonogenic assays using increased the proportion of apoptotic cells (30.20 Ϯ CD34ϩ cells from healthy subjects demonstrated that 1.47%), but not to a statistically significant extent as rHuCD40L and/or rHuFasL reduced the number of compared with untreated cultures (P ϭ 0.0645). The CFCs as compared with that in untreated cultures.
addition of both rHuCD40L and rHuFasL resulted in a However, these differences were not statistically signif- statistically significant increase in the proportion of icant. Collectively, these data suggest that CD40L exerts, EXPRESSION OF CD40 AND Fas-MEDIATED APOPTOSIS OF BM CD34ϩ CELLS IN SLE ng/ml) as compared with that in healthy controls (0.27 Ϯ0.18 ng/ml; P ϭ 0.0061).
Increase in clonogenic potential of SLE BM
progenitor cells by neutralization of CD40L. To further
substantiate the involvement of the CD40 pathway in
BM progenitor cell survival in SLE patients, we evalu-
ated the clonogenic potential of patient and normal
BMMCs in the presence or absence of a CD40L neu-
tralizing antibody. On the basis of our observations
indicating that CD40 was overexpressed on the patients’
CD34ϩ cells in the BMMC fraction (Table 2) and that
CD40L was overproduced by cells in the BM micro-
Figure 3. Concentration of CD40L in long-term bone marrow cul-
environment (Figure 3), it was anticipated that neutral- tures (LTBMCs). A, Cells from SLE patients (n ϭ 20) and healthy
ization of CD40L would increase the CFC number selec- controls (n ϭ 20) were harvested at confluence, and levels of CD40L,upon concentration in LTBMC supernatants, were measured by tively in SLE patient samples, but not in the controls.
enzyme-linked immunosorbent assay. The mean concentration and 95% confidence intervals are indicated by horizontal lines and shaded
areas, respectively. B, CD40L mRNA expression was determined in
LTBMC adherent layers of cells from all SLE patients and healthy
controls. Bars show the mean and SD cumulative data. Groups were
compared using the nonparametric Mann-Whitney test. See Figure 2
for other definitions.
mainly, a synergistic effect on the FasL-induced impair-ment of the clonogenic potential of CD40-expressingCD34ϩ cells, rather than a direct negative effect.
Increased CD40L expression in SLE LTBMCs.
To probe the pathophysiologic significance of the in-creased expression of CD40 on CD34ϩ cells frompatients with SLE, we evaluated CD40L levels andCD40L mRNA expression in LTBMCs, an experimentaldesign that represents an in vitro model mimicking theBM microenvironment (30). The levels of sCD40L werefirst determined in LTBMC supernatants. Results arepresented in Figure 3A.
SLE patients displayed increased sCD40L levels (21.97 Ϯ 21.07 ng/ml) as compared with healthy controls(9.42 Ϯ 3.63 ng/ml; P ϭ 0.0315), suggesting that localproduction of sCD40L is increased in the BM of SLEpatients. Individual sCD40L levels inversely correlated Figure 4. Effects of neutralization of CD40L on the clonogenic po-
with the proportion of CD34ϩ cells in the subjects tential of bone marrow progenitor cells from patients with SLE. Bone studied (r ϭ Ϫ0.6139, P Ͻ 0.0001), further indicating the marrow mononuclear cells (BMMCs) (105) from SLE patients (n ϭ 6) negative effect of sCD40L on the reserves of BM and healthy controls (n ϭ 6) were cultured using a standard clonogenicassay, in 1 ml Iscove’s modified Dulbecco’s medium supple- mented with 30% fetal calf serum, 1% bovine serum albumin, 10Ϫ4M To further substantiate the increased local pro- mercaptoethanol, 0.075% sodium bicarbonate, 2 mM L-glutamine, 0.9% duction of CD40L in SLE BM, we evaluated the cyto- methylcellulose, 5 ng granulocyte–macrophage colony-stimulating kine mRNA expression in LTBMC adherent cell ex- factor, 50 ng interleukin-3, and 2 IU erythropoietin in the presence tracts from patients and healthy controls. Cumulative or absence of 1 ␮g CD40L neutralizing antibody. Bars show the meanand SD number of CFCs obtained in SLE patients and controls in data are shown in Figure 3B. Consistent with the ELISA antibody-treated and untreated (baseline) cultures. Groups were com- data shown in Figure 4A, CD40L mRNA expression was pared using Student’s t-test for paired samples. See Figure 2 for other significantly increased in SLE patients (0.75 Ϯ 0.62 obtained from SLE patients (n ϭ 6) resulted in a displayed an increased proportion of apoptotic, Fas- statistically significant increase in the clonogenic poten- expressing cells within the CD34ϩ cell compartment as tial of BMMCs (93 Ϯ 47 CFCs per 105 BMMCs) as compared with the the levels in healthy subjects.
compared with that in cultures that did not receive To explore the possible association of CD40 anti-CD40L antibody (70 Ϯ 48 CFCs per 105 BMMCs; up-regulation with the apoptotic depletion of the hema- P ϭ 0.005) or that in cultures with the same quantity of topoietic progenitor cells in SLE patients, we per- PBS (69 Ϯ 46 CFCs per 105 BMMCs; P ϭ 0.003). In formed a subset analysis of the survival characteristics of contrast, neutralization of CD40L had no impact on the CD34ϩ cells on the basis of CD40 surface expression.
clonogenic potential of BMMCs obtained from control We found that the proportion of apoptotic CD34ϩ cells subjects (n ϭ 6) (159 Ϯ 16 CFCs per 105 BMMCs versus was significantly increased in the CD40ϩ cells as com- 162 Ϯ 17 CFC per 105 BMMCs in untreated cultures; pared with the CD40Ϫ cell fraction, in both patients and P ϭ 0.286) (Figure 4). Thus, our findings suggest that healthy controls, suggesting a possible involvement of CD40L is functionally expressed in BMMCs.
CD40 in the apoptotic process of BM hematopoieticprogenitor cells. In favor of this hypothesis was thestrong correlation between the proportion of CD40ϩ DISCUSSION
cells and the percentage of apoptotic CD34ϩ cells in the There is increasing evidence that immune dys- study population. In contrast, CD40 antigen expression regulation in patients with SLE may disturb the mor- did not affect the survival characteristics of cells in the phology and function of the BM microenvironment and CD34Ϫ BMMC fraction, which comprises mainly the may severely affect BM hematopoiesis (9,34). Auto- reactive lymphocytes and proinflammatory mediators CD40 ligation appears to have opposing effects have been primarily implicated in BM damage, in that on cell viability, depending on the cell target (11).
these elements suppress the hematopoiesis-supporting Interestingly, CD40-dependent apoptotic processes have capacity of BM stromal cells and accelerate the apo- been reported to be mediated via up-regulation of other ptosis of hematopoietic stem/progenitor cells (5). A TNFR family molecules, including Fas (22,24). More- varying degree of “unexplained” cytopenias, extreme over, a cooperative interaction between CD40 and Fas in cases of aplastic anemia, and insufficient CD34ϩ cell the promotion of the apoptotic process has been dem- yield upon mobilization are abnormalities attributed to onstrated in human hepatocytes (23), B lymphoma cells the immune-mediated damage of BM in SLE (35,36). A (38), and carcinoma cells (24). Our data demonstrating better understanding of the mechanisms that induce the that incubation of patients’ CD34ϩ cells with CD40L stem/progenitor cell damage in SLE would have major resulted in up-regulation of surface Fas indicate, for the clinical implications in SLE, in view of the current first time, that a cooperative mechanism between the interest in exploring the use of autologous stem cells in CD40 and Fas pathways also exists in BM CD34ϩ cells.
the treatment of severely affected patients (37), since In accordance with previously reported data (5), elucidating the effects of these cells may explain poten- SLE patients were found to display an increased propor- tial problems in the mobilization, harvesting, and mar- tion of Fasϩ cells in the CD34ϩ BMMC fraction, and incubation of purified CD34ϩ cells with FasL resulted in In the present study, we have investigated the a significant increase in the proportion of apoptotic cells, involvement of CD40–CD40L interactions in the apo- suggesting that functional Fas expression takes place in ptotic process of the BM hematopoietic stem/progenitor the BM. The strong correlation between the proportion cells in patients with SLE. Although these 2 TNFR/TNF of CD40- and Fas-expressing CD34ϩ cells in SLE family members are aberrantly expressed in patients’ PB patients indicates a possible cooperative mode of action lymphocytes and serum, they have never been studied in of these molecules in accelerating progenitor cell death.
the BM. We found that CD40 was minimally expressed Evidence to support this hypothesis comes from our on normal hematopoietic progenitor cells; however, the observation that ex vivo treatment of patients’ CD34ϩ proportion of CD40-expressing CD34ϩ cells was signif- cells with the combination of CD40L and FasL resulted icantly increased in SLE patients. In accordance with in amplification of apoptosis as compared with that in our previously reported findings on the reserves and cultures exposed to each ligand alone.
survival characteristics of BM CD34ϩ cells in SLE (5), It is of interest that treatment of patients’ CD34ϩ we found that SLE patients had a significantly lower cells with CD40L alone resulted in a low level of apo- percentage of CD34ϩ cells in the BMMC fraction and ptosis induction, suggesting that CD40 activation per se EXPRESSION OF CD40 AND Fas-MEDIATED APOPTOSIS OF BM CD34ϩ CELLS IN SLE may affect CD34ϩ cell survival. However, a number of reduce autoantibody production but also to ameliorate observations suggest that CD40 activation confers pre- cytopenias associated with BM dysfunction. The intra- dominantly indirect negative effects on the survival of cellular pathways and signaling cascades involved in this BM progenitor cells through the amplification of Fas- intriguing crosstalk between CD40 and Fas, the possible induced apoptosis. Thus, exposure of patients’ CD34ϩ implication of other TNF/TNFR family members, as cells to rHuCD40L resulted in up-regulation of Fas well as the effect of the treatment on CD40 expression expression. Moreover, we have shown that in vitro are interesting areas for further investigation.
exposure to a combination of CD40L and FasL induceda profound reduction in the number of CFCs in SLEpatients as compared with treatment with each cytokine ACKNOWLEDGMENTS
alone, and that the relative reduction in CFC numbers We thank the staff of the Departments of Rheumatol- was higher following FasL treatment as compared with ogy (especially Dr. Heraklis Kritikos) and Hematology at the University Hospital of Heraklion for providing the BM sam- It has been suggested that elevated levels of ples from patients and healthy controls.
sCD40L in SLE patient sera, in association with in-creased and prolonged expression of the membrane AUTHOR CONTRIBUTIONS
form of the molecule on patients’ T cells, may contribute Dr. Papadaki had full access to all of the data in the study and to the pathophysiology of the disease by amplifying the takes responsibility for the integrity of the data and the accuracy of the aberrant activation signals through the target cell CD40 data analysis.
Study design. Eliopoulos, Papadaki.
(16). In our study, we evaluated sCD40L in LTBMC Acquisition of data. Pyrovolaki, Mavroudi, Sidiropoulos, Boumpas,
supernatants of patients’ cells, reflecting the cytokine levels within the BM microenvironment. We found that Analysis and interpretation of data. Eliopoulos, Papadaki.
culture supernatant levels of sCD40L were significantly Manuscript preparation. Eliopoulos, Boumpas, Papadaki.
Statistical analysis. Papadaki.
increased in patients compared with controls, and thatthese levels correlated inversely with the proportion ofCD34ϩ cells. This finding coupled with the observed REFERENCES
CFC recovery following CD40L neutralization in pa- 1. Ikehara S. Bone marrow transplantation: a new strategy for tient, but not control, BMMC cultures further highlight intractable diseases. Drugs Today (Barc) 2002;38:103–11.
the negative effect that the CD40 pathway may exert on 2. Burt RK, Traynor AE, Craig R, Marmont AM. The promise of hematopoietic stem cell transplantation for autoimmune diseases.
hematopoiesis in patients with SLE. Consistent with the Bone Marrow Transplant 2003;31:521–4.
elevated sCD40L levels in the supernatants was the 3. Dazzi F, van Laar JM, Cope A, Tyndall A. Cell therapy for increased cytokine mRNA expression in LTBMC adher- autoimmune diseases. Arthritis Res Ther 2007;9:206.
4. Corcione A, Benvenuto F, Ferretti E, Giunti D, Cappiello V, ent layers derived from SLE patients compared with that Cazzanti F, et al. Human mesenchymal stem cells modulate B-cell in normal LTBMCs. The cellular origin of CD40L in the BM microenvironment in SLE is unknown. CD40L might 5. Papadaki HA, Boumpas DT, Gibson FM, Jayne DR, Axford JS, be produced by a variety of cells that constitute the Gordon-Smith EC, et al. Increased apoptosis of bone marrowCD34(ϩ) cells and impaired function of bone marrow stromal LTBMC adherent layers, such as monocyte/macrophages, cells in patients with systemic lupus erythematosus. Br J Haematol endothelial cells, and fibroblasts (11), but also might be produced by lymphocytes that persist in LTBMC condi- 6. Westerweel PE, Luijten RK, Hoefer IE, Koomans HA, Derksen RH, Verhaar MC. Haematopoietic and endothelial progenitor cells are deficient in quiescent systemic lupus erythematosus. Ann In summary, this study is the first to evaluate the expression of CD40/CD40L in the BM progenitor cell 7. Sun LY, Zhou KX, Feng XB, Zhang HY, Ding XQ, Jin O, et al.
Abnormal surface markers expression on bone marrow CD34(ϩ) microenvironment in patients with SLE and healthy cells and correlation with disease activity in patients with systemic subjects. Our data reveal a novel role for CD40 in BM lupus erythematosus. Clin Rheumatol 2007;26:2073–9.
homeostasis in SLE, by demonstrating that up- 8. Statkute L, Verda L, Oyama Y, Traynor A, Villa M, Shook T, et regulation of the CD40 pathway amplifies Fas-mediated al. Mobilization, harvesting and selection of peripheral blood stemcells in patients with autoimmune diseases undergoing autologous CD34ϩ hematopoietic progenitor cell death and con- hematopoietic stem cell transplantation. Bone Marrow Transplant tributes to defective BM progenitor cell reserve and function, both of which are characteristic of SLE. These 9. Papadaki HA. Autoreactive T-lymphocytes are implicated in the pathogenesis of bone marrow failure in patients with systemic results further substantiate the rationale of CD40/ lupus erythematosus. Leuk Lymphoma 2003;44:1301–7.
CD40L blockade as treatment of SLE, not only to 10. Fiumara P, Younes A. CD40 ligand (CD154) and tumour necrosis factor-related apoptosis inducing ligand (Apo-2L) in haematologi- 26. Lee JK, Seki N, Sayers TJ, Subleski J, Gruys EM, Murphy WJ, et cal malignancies. Br J Haematol 2001;113:265–74.
al. Constitutive expression of functional CD40 on mouse renal 11. Schonbeck U, Libby P. The CD40/CD154 receptor/ligand dyad.
cancer cells: induction of Fas and Fas-mediated killing by CD40L.
12. Koshy M, Berger D, Crow MK. Increased expression of CD40 27. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield ligand on systemic lupus erythematosus lymphocytes. J Clin Invest NF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25:1271–7.
13. Desai-Mehta A, Lu L, Ramsey-Goldman R, Datta SK. Hyper- 28. Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang DH, expression of CD40 ligand by B and T cells in human lupus and and the Committee on Prognosis Studies in SLE. Derivation of the its role in pathogenic autoantibody production. J Clin Invest SLEDAI: a disease activity index for lupus patients. Arthritis 14. Yellin MJ, Thienel U. T cells in the pathogenesis of systemic lupus erythematosus: potential roles of CD154-CD40 interactions and 29. Philpott NJ, Turner AJ, Scopes J, Westby M, Marsh JC, Gordon- costimulatory molecules. Curr Rheumatol Rep 2000;2:24–31.
Smith EC, et al. The use of 7-amino actinomycin D in identifying 15. Yazdany J, Davis J. The role of CD40 ligand in systemic lupus apoptosis: simplicity of use and broad spectrum of application erythematosus. Lupus 2004;13:377–80.
compared with other techniques. Blood 1996;87:2244–51.
16. Vakkalanka RK, Woo C, Kirou KA, Koshy M, Berger D, Crow 30. Coutinho LH, De Wynter EA, Will A, Testa NG. Clonal and MK. Elevated levels and functional capacity of soluble CD40 long-term cultures using human bone marrow. In: Testa NG, ligand in systemic lupus erythematosus sera. Arthritis Rheum Milineux G, editors. Haemopoiesis: a practical approach. Oxford: Oxford University Press; 1993. p. 75–106.
17. Kato K, Santana-Sahagun E, Rassenti LZ, Weisman MH, Tamura 31. Toure R, Kheadr E, Lacroix C, Moroni O, Fliss I. Production of N, Kobayashi S, et al. The soluble CD40 ligand sCD154 in systemic antibacterial substances by bifidobacterial isolates from infant lupus erythematosus. J Clin Invest 1999;104:947–55.
stool active against Listeria monocytogenes. J Appl Microbiol 18. Goules A, Tzioufas AG, Manousakis MN, Kirou KA, Crow MK, Routsias JG. Elevated levels of soluble CD40 ligand (sCD40L) in 32. Nagafuji K, Shibuya T, Harada M, Mizuno S, Takenaka K, serum of patients with systemic autoimmune diseases. J Auto- Miyamoto T, et al. Functional expression of Fas antigen (CD95) on hematopoietic progenitor cells. Blood 1995;8:883–9.
19. Funakoshi S, Longo DL, Beckwith M, Conley DK, Tsarfaty G, 33. Papadaki HA, Kritikos HD, Gemetzi C, Koutala H, Marsh JC, Tsarfaty I, et al. Inhibition of human B-cell lymphoma growth by Boumpas DT, et al. Bone marrow progenitor cell reserve and CD40 stimulation. Blood 1994;83:2787–94.
function and stromal cell function are defective in rheumatoid 20. Baker MP, Eliopoulos AG, Young LS, Armitage RJ, Gregory CD, arthritis: evidence for a tumor necrosis factor ␣-mediated effect.
Gordon J. Prolonged phenotypic, functional, and molecular change in group I Burkitt lymphoma cells on short-term exposure 34. Voulgarelis M, Giannouli S, Tasidou A, Anagnostou D, Ziakas to CD40 ligand. Blood 1998;92:2830–43.
PD, Tzioufas AG. Bone marrow histological findings in systemic 21. Von Leoprechting A, van der Bruggen P, Pahl HL, Aruffo A, lupus erythematosus with hematologic abnormalities: a clinico- Simon JC. Stimulation of CD40 on immunogenic human malig- pathological study. Am J Hematol 2006;81:590–7.
nant melanomas augments their cytotoxic T lymphocyte-mediated 35. Tagoe C, Shah A, Yee H, Belmont HM. Aplastic anemia in lysis and induces apoptosis. Cancer Res 1999;59:1287–94.
22. Georgopoulos NT, Steele LP, Thomson MJ, Selby PJ, Southgate J, systemic lupus erythematosus: a distinct presentation of acquired Trejdosiewicz LK. A novel mechanism of CD40-induced apoptosis aplastic anemia? J Clin Rheumatol 2001;7:377–83.
of carcinoma cells involving TRAF3 and JNK/AP-1 activation.
36. Burt RK, Fassas A, Snowden J, van Laar JM, Kozak T, Wulffraat Cell Death Differ 2006;13:1789–801.
NM, et al. Collection of hematopoietic stem cells from patients 23. Afford SC, Randhawa S, Eliopoulos AG, Hubscher SG, Young with autoimmune diseases. Bone Marrow Transplant 2001;28: LS, Adams DH. CD40 activation induces apoptosis in cultured human hepatocytes via induction of cell surface fas ligand expres- 37. Tyndall A, Daikeler T. Autologous hematopoietic stem cell trans- sion and amplifies fas-mediated hepatocyte death during allograft plantation for autoimmune diseases. Acta Haematol 2005;114: rejection. J Exp Med 1999;189:441–6.
24. Eliopoulos AG, Davies C, Knox PG, Gallagher NJ, Afford SC, 38. Xerri L, Bouabdallah R, Devilard E, Hassoun J, Stoppa AM, Birg Adams DH, et al. CD40 induces apoptosis in carcinoma cells F. Sensitivity to Fas-mediated apoptosis is null or weak in B-cell through activation of cytotoxic ligands of the tumor necrosis factor non-Hodgkin’s lymphomas and is moderately increased by CD40 superfamily. Mol Cell Biol 2000;20:5503–15.
ligation. Br J Cancer 1998;78:225–32.
25. Zhou F, Ajuebor MN, Beck PL, Le T, Hogaboam CM, Swain MG.
39. Papadaki HA, Marsh JC, Eliopoulos GD. Bone marrow stem cells CD154-CD40 interactions drive hepatocyte apoptosis in murine and stromal cells in autoimmune cytopenias. Leuk Lymphoma fulminant hepatitis. Hepatology 2005;42:372–80.

Source: http://mcb.med.uoc.gr/arxeia2/publications_2009-1_cd40_bone_marrow.pdf