SEARCH

SEARCH BY CITATION

Keywords:

  • chronic neutropenia;
  • T-lymphocytes;
  • myelosuppression;
  • interferon-γ;
  • Fas-ligand

Summary

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

To characterize the cellular components responsible for the impaired granulopoiesis in chronic idiopathic neutropenia (CIN), we investigated the origin of the proapoptotic cytokine producing cells in the bone marrow (BM) microenvironment of CIN patients. We found that the interferon gamma (IFNγ) and/or Fas-ligand expressing cells in patient BM mononuclear cells and long-term BM culture stroma cells were the CD3+ T-lymphocytes but not the CD14+ monocytes/macrophages. The percentage of activated T-lymphocytes was increased in patients’ BM as indicated by the proportions of human leucocyte antigen (HLA)-DR+, CD25+, CD38+, CD69+ and Fas+ cells within the CD3+ fraction. Intracellular IFNγ expression was higher in the BM than peripheral blood of the patients and was associated with increased BM T-lymphocyte numbers. In crossover experiments, patient CD3+ T-lymphocytes conferred autologous and allogeneic haemopoietic progenitor cell colony inhibition. Patients’ T-cell receptor repertoire and polymerase chain reaction analysis did not reveal any clonal T-lymphocyte expansion, suggesting the absence of a direct, antigen-driven recognition of CD34+ myeloid progenitor cells by patient T-lymphocytes. We conclude that CIN patients have increased number of activated T-lymphocytes in the BM, probably in the setting of a localized polyclonal immune reaction and that these cells confer an inhibitory effect on myelopoiesis through myelosuppressive cytokines including Fas-ligand and IFNγ.

Chronic idiopathic neutropenia (CIN) is defined as the prolonged and unexplained reduction in the number of circulating neutrophils below the lower limit of the normal distribution for a given ethnic population (Kyle & Linman, 1968; Dale et al, 1979; Haddy et al, 1999; Papadaki et al, 2001a). The condition may present either as isolated neutropenia or as neutropenia combined with thrombocytopenia and/or anaemia (Kyle, 1980; Papadaki et al, 2001b). It is typically characterized by its acquired and static character, female predominance (Kyle, 1980; Papadaki et al, 1999a) and the usually benign and uncomplicated course (Kyle, 1980). A human leucocyte antigen (HLA) class II (DRB1*1302) genetic predisposition defining a relative risk of 8·36 for the development of the disorder has also been reported (Papadaki et al, 2001c).

The cause of the disorder and the underlying mechanisms leading to neutropenia in the affected subjects are largely unknown. It has been reported that the in vitro growth potential of the granulocytic progenitors is impaired in CIN patients (Greenberg et al, 1980; Eliopoulos et al, 1990; Papadaki et al, 1999b) and that the disorder is usually characterized by a degree of bone marrow (BM) hypoplasia affecting mainly the postmitotic maturating pool of the granulocytic series, resulting in a mild to moderate degree of maturation arrest of the myeloid development (Young, 2000a; Papadaki et al, 2000). Recently, we have studied the reserves and survival characteristics of myeloid cells at sequential stages of granulocytic differentiation from the early progenitors to the mature neutrophils and we have found that the impaired granulopoiesis in these patients is associated with low frequency of the granulocyte progenitor CD34+/CD33+ cells and granulocyte-colony forming units (CFU-G) becasue of accelerated tumour necrosis factor alpha (TNFα)-induced Fas-mediated apoptosis within this particular cellular compartment (Papadaki et al, 2003). It is, therefore, conceivable that this hypoplastic type of CIN displays overlapping pathophysiological features with certain BM failure syndromes in which the apoptotic depletion of haemopoietic stem/progenitor cells has emerged as a central pathogenetic mechanism (Selleri et al, 1995; Young, 2000a; Papadaki & Eliopoulos, 2003).

Consistent with the heightened immune response that operate in certain BM failure syndromes (Young, 2000b; Barrett et al, 2000), CIN patients appear also to display immune dysregulation in the BM, as indicated by the aberrant local pro-inflammatory cytokine production, such as TNFα, interferon gamma (IFNγ) and Fas-ligand (FasL), that might explain, at least in part, the elevated Fas levels and the apoptotic depletion of patient progenitor cells (Papadaki et al, 2003). In the current study, using our recently published data as an entry point, we further probed the pathogenetic mechanism of CIN by investigating the origin of cells producing the pro-apoptotic cytokines in patients’ BM microenvironment. We also explored the possible existence of lymphocyte subpopulations with an activated phenotype pattern and myelosuppressive properties in patients’ BM and peripheral blood (PB), and examined whether specific, antigen-driven T-cell response or non-specific polyclonal T-cell activation is involved in the pathogenesis of the disease by characterizing the T-cell repertoire in CIN BM.

Patients

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

We studied 75 adults with CIN, 12 males and 65 females aged 16–76 years (median age 54 years), satisfying the previously reported diagnostic criteria for the disease (Dale, 1991; Palmblad et al, 2001; Papadaki et al, 2001a). In particular, the patients had neutrophil counts <1·8 × 109/l (mean 1·415 ± 0·31 × 109/l) for a period of 12–222 months (median duration 86 months), had no clinical, serological or ultrasonic evidence of any underlying disease known to be associated with neutropenia, no history of exposure to irradiation or use of chemical compounds or intake of drugs to which neutropenia might be ascribed, normal BM karyotype and negative serum leucoagglutination and immunofluorescence tests for anti-neutrophil antibodies. Cyclic and familial neutropenias were excluded by performing serial neutrophil enumerations in the patients and their family members. All patients studied displayed hypoplastic and left-shifted myeloid series in May–Grunwald–Giemsa stained marrow smears and trephine biopsy specimens. Detailed patient characteristics are presented in Table I. As controls, 47 healthy volunteers, age- and sex-matched with the patients, were studied. Informed consent according to the Helsinki Protocol was obtained from all subjects studied. The study was approved by the Ethics Committee of the Hospital.

Table I.  Clinical and laboratory data of the patients studied.
UPNAge (years)SexDuration (months)Hb (g/dl)WBC (×109/l)Neutro (×109/l)Lympho (×109/l)Mono (×109/l)Plts (×109/l)
  1. UPN, unique patient number; Hb, haemoglobin; WBC, white blood cells; neutro, neutrophils; lympho, lymphocytes; mono, monocytes; Plts, platelets.

160M5314·52·851·750·700·40185
229F20012·74·301·700·900·60200
353F15413·33·301·621·300·31192
453F8613·22·500·601·550·30200
566F5513·73·801·651·730·36211
633F8413·84·001·601·900·30188
730M3616·74·201·702·000·40300
856F5910·33·701·402·000·30292
955F6013·93·501·701·300·30189
1070F20013·53·701·601·700·40247
1154F6313·34·401·502·000·40219
1271F3613·34·101·502·300·30147
1365F11513·94·051·751·800·30253
1464F3011·53·651·751·600·30177
1532F3912·13·601·201·800·50259
1642F8012·83·951·751·810·21234
1753M11516·03·801·701·400·06190
1836F16113·03·101·760·930·29209
1950F10911·73·101·401·400·30121
2032F8412·83·301·501·400·30257
2160F9513·03·001·401·200·30210
2239F3613·03·101·401·300·21300
2348F16811·53·201·301·800·20276
2450F4812·63·701·702·200·30213
2545F14413·93·001·761·000·16160
2655F22212·44·301·702·200·30270
2770F2813·93·701·711·500·38181
2856F14212·82·801·601·000·20200
2947M16515·13·101·701·100·30270
3069F11813·23·401·701·300·30216
3170F13212·74·001·702·000·30182
3235F10111·73·601·701·500·30213
3360F6212·43·851·751·800·30126
3460F16613·84·501·702·500·30220
3570F13212·74·001·702·000·30182
3660F14313·14·501·702·200·50273
3760F7814·63·801·701·600·40235
3863F13712·94·101·402·000·50228
3955F8912·33·601·701·400·30121
4063M10813·43·001·601·000·60205
4165F14413·14·501·702·200·50273
4245F6012·94·101·781·700·50188
4335F6511·83·501·791·400·20214
4447F9713·33·301·601·300·30202
4526F16813·03·201·001·900·20300
4649F12011·92·601·101·100·40159
4763F19211·52·401·500·700·20166
4865F14413·14·501·702·200·50273
4966F10812·83·401·601·500·20181
5030M4815·23·501·601·400·40243
5148M13215·93·401·101·700·20196
5240F9111·53·301·701·300·20172
5319M2414·24·301·791·600·40259
5465F12012·13·601·201·800·50250
5525M6713·33·301·601·300·30202
5663F8414·14·701·702·400·55244
5752F15614·83·201·701·100·30260
5856F12012·73·401·501·500·30255
5922F4813·52·100·401·400·30194
6060F11814·02·601·200·800·50146
6151M1216·02·801·201·100·40140
6256F2414·13·601·701·640·16226
6330F2412·71·700·600·900·20200
6447F3612·93·001·101·400·40242
6556F12012·73·401·401·600·30240
6665F18014·64·801·702·500·40218
6753M6014·24·201·352·200·45249
6855F6313·33·401·501·500·40270
6965F6014·23·001·701·000·30172
7058F6013·03·301·601·200·40200
7134M6012·55·501·502·500·60183
7216F2412·42·401·001·200·10205
7332F1212·74·501·702·200·40319
7476F2412·94·301·602·200·40266
7533F2412·23·600·702·100·25163

Bone marrow samples

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

The BM cells obtained from posterior iliac crest aspirates, were immediately diluted 1:1 in Iscove's modified Dulbecco's medium (IMDM; Gibco, Invitrogen Corporation, Paisley, UK), supplemented with 100 IU/ml penicillin–streptomycin (PS; Gibco) and 10 IU/ml preservative-free heparin (Sigma, St Louis, MO, USA). Diluted BM samples were centrifuged on Lymphoprep (Nycomed Pharma AS, Oslo, Norway) at 400 g for 30 min at room temperature to obtain the bone marrow mononuclear cells (BMMCs).

Purification of BM cell subpopulations

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

The CD34+, CD3+ and CD14+ BM cell subpopulations representing the haemopoietic progenitors, T-lymphocytes and monocytes, respectively, were isolated from the BMMC fraction by magnetic-activated cell sorting (MACS) (Mitenyi Biotec GmbH, Bergisch Gladbach, Germany) according to the manufacturer's protocol. In all experiments, the purity of each subpopulation was >96%, as estimated by flow cytometry.

Long-term BM cultures

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

Standard long-term BM cultures (LTBMCs) were grown from 107 BMMCs according to the standard technique (Coutinho et al, 1993) in 10 ml IMDM supplemented with 10% fetal bovine serum (FBS; Gibco), 10% horse serum (Gibco), 100 IU/ml PS, 2 mmol l-glutamine (Sigma) and 10−6 mol hydrocortisone sodium succinate (Sigma), and incubated at a 33°C-5%CO2 fully humidified atmosphere. At weekly intervals, the cultures were fed by demi-depopulation and assessed morphologically for the adherent layer formation. At week 4, confluent stromal layers from patients and control subjects were trypsinized and the CD3+ and CD14+ lymphoid and monocytic elements, respectively, were immunomagnetically sorted as described above.

Reverse transcription polymerase chain reaction and Southern blot analysis

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

Total messenger RNA (mRNA), extracted (RNeasy mini kit; Qiagen GmbH, Hilden, Germany) from total BMMCs or adherent cells of confluent LTBMCs, was reverse transcribed (SUPERSCRIPT Preamplification system; Gibco) and amplified by reverse transcription polymerase chain reaction (RT-PCR) for the detection of IFNγ and FasL. In the IFNγ and/or FasL-expressing samples, mRNA was also extracted from purified CD3+ and CD14+ cells derived from the BMMC- and LTBMC-adherent cell fraction. RT-PCR specific conditions and primers, as well as the hybridization probes for IFNγ and FasL, have been described previously (Papadaki et al, 2003). PCR products were normalized according to the amount of β2-microglobulin in the samples. The conditions for 25 cycles of PCR were 94°C for 30 s, 55°C for 30 s and 72°C for 30 s. The positive control for IFNγ and FasL was cDNA from PBMCs or Jurkat cells, respectively, following induction with 100 ng/ml phorbol myristate acetate (PMA, Sigma) plus 1 μmol/l ionomycin (Sigma) for 4 h (Oyaizu et al, 1995).

Flow cytometry and BM immunohistochemistry

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

Two-colour flow cytometry was used for the analysis of BM and PB lymphocyte subpopulations. Briefly, 100 μl aliquots of diluted BM or ethylenediaminetetraacetic acid (EDTA)-anticoagulated PB samples were stained for 30 min on ice with a combination of phycoerythrin (PE)- or fluorescein isothiocyanate (FITC)-conjugated mouse antihuman monoclonal antibodies (mAbs; Beckman Coulter, Marseille, France). In particular, anti-CD3 (UCHT1) or anti-CD8 (B9.11) or anti-CD4 (13B8.2) mAb was combined with each of the following mAbs representing T-cell activation markers: anti-HLA-DR (B8.12.2), anti-CD25 (interleukin-2 receptor; B1.49.9), anti-CD95 (Fas; UB2), anti-CD38 (T16) and anti-CD69 (TP1.55.3). Similarly, anti-CD19 (J4.119) mAb was combined with anti-CD23 (9P25) or anti-CD69 mAb, representing markers of B-cell activation. BM samples were also stained with each of the following mAbs: anti-CD2-PE (39C1.5), anti-CD16-PE (3G8), anti-CD56-PE (N901) and anti-CD57-PE (NC1). PE- or FITC-conjugated mouse IgG of the appropriate isotype served as negative controls. Following two washes with PBS-1% FBS-0·05% sodium azide, contaminating red cells were lysed with 0·12% formic acid and samples were fixed in 0·2% paraformaldehyde using the Q-prep reagent system (Beckman Coulter). Analysis on 10 000 events was performed in an Epics Elite model flow cytometer (Coulter, Miami, FL, USA) in the gate of cells with low forward and low side-scatter properties where lymphocytes are included. Results were expressed as proportions of cells expressing each mAb. Furthermore, by dividing the proportions of double positive cells using the above described mAb combinations by the percentages of total CD3+, CD4+, CD8+ or CD19+ cells, we estimated the proportions of activated cells within each lymphocyte subpopulation (Maciejewski et al, 1994).

For intracellular IFNγ staining of BM or PB CD3+ cells, 106 BMMCs or PBMCs isolated with Lymphoprep, were diluted in 1000 μl IMDM supplemented with 50 ng/ml PMA, 4 μg/ml ionomycin and 40 μg/ml Brefeldin A (Calbiochem, La Jolla, CA, USA) and incubated at 37°C-5% CO2 for 4 h. The cells were then washed with PBS-1% FBS-0·05% sodium azide and incubated in the dark for 30 min on ice with FITC-conjugated anti-CD3 mAb. After washing with the above buffer, cells were fixed and permeabilized with the IntraPrep intracellular staining kit (Beckman Coulter) and incubated with PE-conjugated mouse antihuman IFNγ mAb (clone 45·15) in the dark for 30 min on ice. Following one wash, cells were resuspended in PBS-2% paraformaldeyde before flow cytometric analysis. Cells incubated with isotypic control antibodies were used as negative controls in each step. Results of intracellular cytokine staining were expressed as the percentage of cells expressing IFNγ within the BM or PB CD3+ cell population.

To substantiate the BM flow cytometric findings, buffered formalin-fixed and EDTA-decalcified patient BM trephine biopsies were immunohistochemically stained with mAbs (DakoCytomation, Glostrup, Denmark) against CD3 (PC3/188A), CD4 (MT310), CD8 (C8/144B), CD5 (CD5/54/F6), CD20 (L26), CD79a (JCB117), and CD45RO (OPD4) by means of the avidin–biotin complex (ABComplexes) technique (Hsu & Raine, 1981).

Assessment of clonality of BM CD3+ cells

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

Clonality assessment of T cells in whole BM samples was performed by quantitative analysis of different variable regions of the TcR β chain (Vβ repertoire) of BM CD3+ cells by means of flow cytometry using the IOTest Beta Mark kit (Beckman Coulter) according to the manufacturer's instructions. The kit contains mixtures of conjugated TcR Vβ antibodies corresponding to 24 different specificities (70% coverage of normal human TcR Vβ repertoire).

T cell clonality assessment was also performed by DNA-based (QIAamp DNA Blood Mini kit; Qiagen) PCR and heteroduplex analysis of TcRβ, TcRγ and TcRδ gene rearrangements following recent suggestions from the BIOMED-2 collaborative effort enabling the detection of virtually all clonal T-cell populations (van Dongen et al, 2003). Specifically, for TcRβ studies, multiple Vb and Jb primers, theoretically covering all functional Vb and Jb gene segments, were used; this approach also permitted the detection of incomplete Db–Jb rearrangements. In the case of TcRγ rearrangements, BIOMED primers enabled the detection of all rearranging Vγ segments except ψVγB (very rarely rearranged) as well as four of five Jγ segments (Jγ1·2 was excluded so as to avoid false identification as clonal products of Vγ9–Jγ1·2 canonical rearrangements, which are detected in approximately 1% of blood T-lymphocytes and increase in frequency with age). For TcRδ rearrangements, the BIOMED protocol enabled the detection of approximately 95% of known rearrangements (Vδ (D)Jδ, Vδ–Dδ, Dδ–Dδ and Dδ–Jδ).

Co-culture of BM CD34+ and CD3+ cells

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

A standard clonogenic progenitor cell assay was used for the co-culture of 1·7 × 104 patient or normal purified CD34+ BM cells in 1 ml of methylcellulose culture medium (Stem Cell Technologies, Vancouver, BC, Canada) with 1·7 × 105 autologous purified CD3+ cells in the presence of 5 ng/ml granulocyte-macrophage colony stimulating factor (R&D Systems, Minneapolis, MN, USA), 50 ng/ml interleukin-3 (R&D Systems) and 2 IU/ml erythropoietin (Janssen-Cilag, Athens, Greece) (Papadaki et al, 2003). Crossover experiments, in which patient or normal CD34+ cells were co-cultured with normal or patient CD3+ cells, respectively, were also performed. In another set of experiments, the CD34+ cells were preincubated with the autologous or allogeneic CD3+ cells at 37°C for 4 hours before culture to assess the effect of the cell-to-cell contact in colony inhibition. Cultures containing only patient or normal CD34+ cells were used as controls in each experiment. On day 14, the number of CFU-G, CFU-macrophage (CFU-M) and CFU-granulocyte/macrophage (CFU-GM) were scored according to established criteria (Coutinho et al, 1993) and the sum of the above colonies was defined as total CFUs. The percentage of colony inhibition was calculated by dividing the difference in colony numbers between the control culture and test culture by the number of the respective colonies in the control culture.

Statistical analysis

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

Data were analysed using the GraphPAd Prism statistical PC program (GraphPad Software, San Diego, CA, USA) by means of the Student's t-test for paired and unpaired samples. Chi-square test was used to define differences in IFNγ and FasL expression in LTBMC stromal cell and BMMC extracts, between patients and control subjects. Grouped data were expressed as mean ± 1 SD. P ≤ 5% was considered to be statistically significant.

Identification of IFNγ and/or FasL producing cells in CIN BM

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

Based on previous data that suggested an impaired haemopoietic-supporting capacity of CIN patient LTBMC stromal layers, probably because of abnormal IFNγ and FasL expression (Papadaki et al, 2003), we sought to identify the origin of the apoptotic cytokine-producing cells in the adherent cell population. Of the 30 CIN stromal cell extracts studied, IFNγ and/or FasL expression was identified in 28 samples (93·3%), while expression of both cytokines was detected in 24 cases (80·0%). In keeping with previous reports, IFNγ and/or FasL expression was not found in any of the normal controls studied (Papadaki et al, 2003). In all CIN patients, the IFNγ and/or FasL-producing cells were the CD3+ but not the CD14+ immunomagnetically sorted cells, suggesting the presence of activated T-lymphocytes in patients’ stroma (Fig 1).

image

Figure 1. Detection of IFNγ and FasL mRNA in purified CD3+ but not CD14+ cells from LTBMCs and BMMCs in CIN. Total mRNA extracted from immunomagnetically sorted CD3+ and CD14+ lymphoid and monocytic elements of trypsinized LTBMC adherent cell layers and BMMCs was subjected to RT-PCR analysis for IFNγ and FasL detection using specific primers. PCR products were electrophoresed on a 1·5% agarose gel and visualized under ultraviolet light by ethidium bromide staining. Results of a representative CIN patient are shown. As positive controls for IFNγ and FasL expression, cDNA was obtained from PBMCs and Jurkat cells, respectively, stimulated with PMA (100 ng/ml) plus ionomycin (1 μmol/l) for 4 h. β2-microglobulin (β2m) was used as control for cDNA amplification. M.W.M., molecular weigh marker.

Download figure to PowerPoint

As lymphocytes constitute the minor cell population in LTBMC stromal layers, we also examined the IFNγ and FasL mRNA expression in RT-PCR products derived from total BMMCs, as their lymphocyte content better reflects the BM lymphocyte function and status of activation. In the IFNγ and FasL-expressing BMMC samples, RT-PCR analysis was also performed in the CD3+ and CD14+ purified cell populations. IFNγ and/or FasL mRNA expression was detected in BMMC extracts of 24 of 28 patients studied (85·7%), while expression of both cytokines was found in 20 cases (71·4%). IFNγ and/or FasL were not detected in the BMMC extracts of any of the control subjects studied. In accordance with the stromal cell data, the IFNγ and/or FasL-producing cells were CD3+ but not CD14+ cells of the BMMC population (Fig 1), a finding corroborating further the presence of activated T-lymphocytes in the BM microenvironment of CIN patients.

Flow cytometric analysis of BM and PB lymphocyte subsets

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

To explore whether the expression of IFNγ and FasL in CIN BM is associated with quantitative changes and/or an activated immunophenotypic profile of lymphocytes, we next performed a flow cytometric study of BM cells in CIN patients (n = 33) and healthy controls (n = 15). The percentage of CD3+ cells in patients’ BM was significantly increased, compared to normal controls (69·0 ± 7·7% vs. 63·3 ± 7·1%, P = 0·0191). This increase was paralleled with a reduction in the proportion of CD19+ cells in the patients, compared with controls (14·8 ± 6·2% vs. 19·1 ± 7·1%, P = 0·0345), suggesting higher T-lymphocyte numbers in CIN BM. Consistent with this finding was the statistically significant increase in the percentage of BM CD2+ cells in the patients compared with controls (74·6 ± 5·7% vs. 66·0 ± 6·8%, P < 0·001). No significant difference was documented between patients and control subjects in the CD4/CD8 cell ratio (1·45 ± 0·66 vs. 1·44 ± 0·51), suggesting a parallel rise in both the CD4+ and CD8+ lymphocyte subpopulations in patients’ BM. Similarly, no statistically significant difference was found between patients and healthy controls in the percentage of CD16+ cells (11·2 ± 4·8% vs. 12·9 ± 5·0%), CD56+ cells (10·87 ± 6·7% vs. 9·9 ± 8·8%) and CD57+ cells (12·65 ± 5·9% vs. 12·5 ± 4·8%). Furthermore, the immunohistochemical study of patient BM biopsies demonstrated the presence of 15–18% lymphocytes among total nucleated cells, mostly of T-cell origin (CD3+, CD5+, CD45RO+, CD79a, CD20), in a predominantly interstitial and, to a lesser extent, nodular pattern (Fig 2).

image

Figure 2. T-lymphocyte predominance in BM trephine biopsy immunohistochemically stained in a CIN patient. Buffered formalin-fixed and EDTA-decalcified BM trephine biopsy specimens were studied immunohistochemically by means of the avidin–biotin complex (ABComplex) technique routinely applied in our laboratory to assess patient BM lymphocyte subpopulations. (A) Increased proportion of CD3+ cells; (B) rare CD20+ cells (original magnification ×,200, Olympus light microscope, Olympus Optical Company, Hamburg, Germany).

Download figure to PowerPoint

The flow cytometric analysis of BM T- and B-lymphocyte subsets bearing activation markers is presented in Table II. The percentage of activated T-lymphocytes was significantly higher in the patients compared with the controls, as indicated by the proportions of HLA-DR+ (P = 0·0246), CD25+ (P = 0·0465), CD38+ (P = 0·0002), CD69+ (P = 0·0121) and Fas+ (P < 0·0001) cells within the CD3+ cell population (Fig 3). To further characterize the activated cellular components, a subset analysis was performed within the CD4+ and CD8+ subpopulations. The proportions of cells bearing the above activation markers were significantly higher in the patients compared with the controls in both the CD4+ (P = 0·0214, P = 0·0342, P = 0·0025, 0·0392 and 0·0005, respectively) and the CD8+ (P = 0·0402, 0·0443, 0·0002, 0·0001 and 0·0017, respectively) cell fractions, suggesting the presence of activated helper and cytotoxic T cells in patients’ BM. In contrast, no statistically significant difference was found in the percentage of activated B cells between patients and control subjects as indicated by the proportions of CD23+ (32·6 ± 14·8% and 32·1 ± 13·7%, respectively) and CD69+ (28·2 ± 15·2% and 29·0 ± 4·5%, respectively) cells detected within the CD19+ cell population (P = 0·9211 and 0·8498 respectively).

Table II.  Flow cytometric analysis of bone marrow lymphocyte subsets.
 CIN patients (n = 33)Healthy controls (n = 15)P-value*
  1. All values are expressed as mean ± 1 SD.

  2. *Comparison of values between patients and healthy controls was performed with the Student's t-test. P ≤ 0·05 were considered as statistically significant.

CD3+ cell fraction
 % HLA-DR+ cells6·69 ± 3·834·07 ± 3·100·0246
  Median (range)6·60 (1·12–14·49)3·90 (0·16–9·21)
 % CD25+ cells7·57 ± 3·745·54 ± 1·140·0465
  Median (range)6·50 (1·60–20·50)5·00 (3·50–7·50)
 % CD38+ cells23·86 ± 13·929·24 ± 3·020·0002
  Median (range)25·40 (4·30–45·60)10·00 (2·90–15·00)
 % CD69+ cells21·60 ± 8·0516·06 ± 2·010·0121
  Median (range)18·30 (12·60–37·40)15·00 (13·00–20·00)
 % Fas+ cells27·35 ± 13·1311·55 ± 6·54<0·0001
  Median (range)34·50 (3·49–42·20)8·20 (3·70–27·80)
CD4+ cell fraction
 % HLA-DR+ cells6·46 ± 4·483·52 ± 2·400·0214
  Median (range)5·00 (2·19–15·91)1·70 (0·94–7·40)
 % CD25+ cells10·96 ± 5·449·32 ± 1·650·0342
  Median (range)9·30 (3·00–26·40)9·10 (7·77–12·12)
 % CD38+ cells23·77 ± 17·249·12 ± 5·790·0002
  Median (range)18·90 (3·80–57·30)7·70 (6·06–23·26)
 % CD69+ cells13·83 ± 6·2710·15 ± 3·500·0392
  Median (range)13·40 (6·40–26·70)10·40 (6·40–15·10)
 % Fas+ cells27·63 ± 12·5114·59 ± 7·030·0005
  Median (range)35·10 (2·70–46·00)12·10 (6·30–31·30)
CD8+ cell fraction
 % HLA-DR+ cells11·09 ± 7·696·43 ± 5·460·0402
  Median (range)8·80 (1·76–30·43)4·20 (0·40–15·15)
 % CD25+ cells4·09 ± 3·032·35 ± 1·650·0443
  Median (range)3·20 (1·40–16·90)2·00 (0·90–5·60)
 % CD38+ cells23·89 ± 10·1113·37 ± 0·620·0002
  Median (range)22·20 (6·35–47·00)13·16 (12·12–13·90)
 % CD69+ cells33·71 ± 10·1422·59 ± 2·280·0001
  Median (range)32·70 (18·50–54·80)24·30 (19·90–24·50)
 % Fas+ cells22·47 ± 13·2410·79 ± 4·150·0017
  Median (range)22·60 (2·66–44·40)8·50 (3·50–20·30)
image

Figure 3. Flow cytometric analysis of BM T lymphocytes for the surface expression of activation markers. Total BM cells from CIN patients and normal controls were double stained with PE- or FITC-conjugated mouse antihuman anti-CD3 mAb and anti-Fas, anti-CD38, anti-HLA-DR, anti-CD25 or anti-CD69 mAb representing T-cell activation markers. Representative dot plots demonstrating the increased proportions of activated CD3+ lymphocyte subsets in a CIN patient compared with a healthy control are shown. Quadrants were set on the basis of BM cells stained with PE- and FITC-conjugated mouse IgG of appropriate isotype and analysis was performed in the gate of cells with low forward and low side-scatter properties where lymphocytes are included.

Download figure to PowerPoint

To evaluate whether the presence of activated T-lymphocytes in CIN BM is associated with the existence of circulating activated T-lymphocytes, a separate flow cytometric study was performed in patients’ PB. The percentage of CD3+ cells did not differ significantly between patients and controls (74·8 ± 7·9% vs. 75·4 ± 5·9% respectively). However, the proportion of activated T cells was significantly higher in the patients than in the controls, as indicated by the percentage of HLA-DR+ (P = 0·0012), CD25+ (P = 0·0013), CD38+ (P = 0·0017), CD69+ (P = 0·0021) and Fas+ (P = 0·0044) cells detected within the CD3+ cell fraction, suggesting the presence of activated T-lymphocytes in patients’ PB. Consistent with the BM data, the proportion of cells bearing the above activation markers was significantly higher in the patients, compared with controls, in both the CD4+ (P = 0·0121, 0·0002, 0·0363, 0·0003 and 0·0300, respectively) and the CD8+ (P < 0·0001, 0·0067, 0·0017, 0·0404 and 0·0028, respectively) cell populations (Table III). No significant differences were found in the proportion of PB CD19+ cells between patients and control subjects and no significant differences were demonstrated between patients and controls in the proportion of cells carrying the activation markers CD23+ and CD69+ in the CD19+ cell population (Table III).

Table III.  Flow cytometric analysis of peripheral blood lymphocyte subsets.
 CIN patientsHealthy controlsP value*
  1. All values are expressed as mean ± 1SD.

  2. *Comparison of values between patients and healthy controls was performed with the Student's t-test. P ≤ 0·05 were considered as statistically significant.

CD3+ cell fraction
 % HLA-DR+ cells4·79 ± 3·352·56 ± 1·940·0012
  Median (range)4·00 (0·10–13·20)1·70 (0·10–7·30)
  n4931
 % CD25+ cells11·44 ± 6·156·62 ± 2·730·0013
  Median (range)10·65 (0·20–24·60)6·55 (2·70–11·60)
  n5020
 % CD38+ cells13·40 ± 11·676·34 ± 4·510·0017
  Median (range)10·05 (0·50–48·20)6·30 (1·45–14·70)
  n6031
 % CD69+ cells21·67 ± 3·4918·27 ± 4·280·0021
  Median (range)22·10 (16·50–28·70)17·20 (12·10–26·90)
  n3322
 % Fas+ cells30·90 ± 11·8925·13 ± 8·440·0449
  Median (range)33·60 (7·80–54·90)25·85 (12·40–45·70)
  n4722
CD4+ cell fraction
 % HLA-DR+ cells4·86 ± 2·563·27 ± 1·950·0121
  Median (range)4·60 (1·00–10·40)2·90 (0·60–6·10)
  n6720
 % CD25+ cells15·48 ± 6·989·13 ± 4·200·0002
  Median (range)15·00 (0·30–28·30)8·85 (2·80–17·60)
  n6520
 % CD38+ cells20·18 ± 14·5213·50 ± 5·410·0363
  Median (range)15·80 (3·00–64·60)13·60 (3·80–23·10)
  n5220
 % CD69+ cells28·23 ± 11·3017·84 ± 6·340·0003
  Median (range)25·80 (13·00–50·96)17·10 (8·70–30·20)
  n3322
 % Fas+ cells34·17 ± 12·4427·62 ± 8·930·0300
  Median (range)33·45 (8·60–62·20)28·05 (15·00–46·40)
  n4822
CD8+ cell fraction
 % HLA-DR+ cells7·32 ± 5·642·74 ± 2·61<0·0001
  Median (range)5·80 (0·80–22·20)1·80 (0·30–12·50)
  n6931
 % CD25+ cells5·47 ± 4·502·54 ± 2·490·0067
  Median (range)4·30 (0·10–28·60)1·60 (0·10–7·70)
  n6720
 % CD38+ cells20·89 ± 12·1211·63 ± 5·600·0017
  Median (range)19·10 (2·00–49·50)10·45 (3·60–27·30)
  n4820
 % CD69+ cells33·75 ± 10·7927·67 ± 9·890·0404
  Median (range)30·00 (20·00–50·96)26·30 (19·80–59·50)
  n3222
 % Fas+ cells25·61 ± 11·2617·11 ± 7·290·0028
  Median (range)23·85 (4·50–52·80)16·70 (8·90–35·30)
  n4822

Intracellular IFNγ staining

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

Based on the IFNγ mRNA expression in patients’ BM microenvironment, we used flow cytometry to evaluate the intracellular cytokine expression in BM and PB lymphocytes as an additional marker of T-cell activation. Less than 1% of BM or PB CD3+ cells expressed IFNγ in both patients (n = 15) and healthy controls (n = 15). However, following a 4-h incubation of cells with PMA and ionomycin, CIN patients showed a reduced proportion of IFNγ expressing PB CD3+ cells compared with controls (12·6 ± 5·8% vs. 21·1 ± 11·2%, P = 0·015). In contrast, patient BM CD3+ cells contained a significantly higher proportion of IFNγ expressing cells than the controls upon stimulation (19·0 ± 4·8% vs. 14·5 ± 6·1%, P = 0·0312) (Fig 4). In addition, the proportion of IFNγ expressing CD3+ cells was significantly higher in BM, compared with PB, within the group of patients (P = 0·003) but not in the group of controls (P = 0·055). These data indicate recruitment of activated T- lymphocytes in patients’ BM.

image

Figure 4. Intracellular IFNγ staining of BM lymphocytes upon stimulation. BMMCs from CIN patients and healthy controls were stimulated with PMA and ionomycin in the presence of brefeldin A and stained with anti-CD3 mAb for surface expression and with anti-IFNγ mAb for intracellular expression. (A) The scattergram of forward scatter (FS) versus side scatter (SSC) that enabled the gating on BM lymphocytes (low FS and low SSC properties) (R1). (B) A three-dimension contour of the expression of surface CD3 and intracellular IFNγ in stimulated patient BMMCs. (D) The intracellular IFNγ expression in the gate of CD3+ BMMCs [R2 in (C)] after stimulation in a healthy control (pale grey) and in a patient with CIN (dark grey). The negative control is also depicted (black histogram).

Download figure to PowerPoint

Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

To investigate whether the presence of activated T-lymphocytes in patients’ BM microenvironment displays pathogenetic significance or is simply part of a secondary immune reaction, we studied the effect of patient BM CD3+ cells on autologous or allogeneic normal granulocytic progenitor cell growth. The mean colony data from three sets of such experiments are presented in Table IV and the mean percentage of in vitro inhibition of total CFU formation is depicted in Fig 5.

Table IV.  Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth.
 PreincubationCFU-GCFU-M + CFU-GMTotal CFU
  1. A total of 1·7 × 104 CIN patient or normal purified CD34+ BM cells were cultured in 1 ml methylcellulose culture medium supplemented with growth factors in the presence of 1·7 × 105 purified normal or patient CD3+ cells, with or without preincubation of CD34+ with CD3+ cells at 37°C for 4 h before culture. Cultures containing only patient or normal CD34+ cells were used as controls in each experiment. Data are expressed as mean colony number per 1 ml of culture medium ± SD in three sets of experiments.

  2. Statistically significant differences from the respective baseline values are indicated by asterisks. *P < 0·01; **P < 0·001; ***P < 0·05.

CIN CD34+ cells 55·00 ± 4·7335·00 ± 5·0090·33 ± 1·53
 Plus CIN T-lymphocytes3·67 ± 1·53*29·00 ± 2·6532·66 ± 2·52**
 Plus CIN T-lymphocytes+3·33 ± 1·15*27·67 ± 0·5831·00 ± 1·00**
 Plus normal T-lymphocytes48·67 ± 6·6628·67 ± 4·7378·33 ± 10·12
 Plus normal T-lymphocytes+48·33 ± 8·0227·67 ± 7·0976·00 ± 12·12
Normal CD34+ cells 76·67 ± 2·8940·67 ± 1·15117·33 ± 2·52
 Plus CIN T-lymphocytes3·67 ± 1·53**33·67 ± 1·53*37·33 ± 1·53**
 Plus CIN T-lymphocytes+3·33 ± 1·15*27·67 ± 0·58***48·67 ± 8·39*
 Plus normal T-lymphocytes79·33 ± 2·8943·00 ± 4·36122·33 ± 6·51
 Plus normal T-lymphocytes+80·67 ± 4·5145·00 ± 4·00125·33 ± 8·50
image

Figure 5. Inhibition of colony formation by purified CD34+ cells in the presence of purified CIN CD3+ lymphocytes. 1·7 × 104 CIN or normal BM CD34+ cells were co-cultured 1·7 × 105 autologous CD3+ cells, with or without a 4-h preincubation, in a standard clonogenic progenitor cell assay. Additionally, CIN or normal CD34+ cells were co-cultured with normal or CIN CD3+ cells with or without (w/o) a 4-h preincubation. Cultures containing only patient or normal CD34+ cells were used as controls in each experiment. On day 14, CFUs were scored as described and the percentage of inhibition was calculated by dividing the difference in CFU numbers between the control culture and test culture by the number of CFUs in the control culture. The mean percentage of colony inhibition by patient or normal CD34+ cells was significantly higher in the presence of patient CD3+ lymphocytes than in the presence of normal lymphocytes, with and without preincubation. No statistically significant difference was found between experiments performed with and without 4-h preincubation of CD34+ and CD3+ cells.

Download figure to PowerPoint

The number of total CFUs obtained by patient CD34+ cells co-cultured with autologous CD3+ lymphocytes with or without a 4-h preincubation, was significantly reduced compared with CFUs obtained in cultures containing only patient CD34+ cells (both P < 0·001). The mean percentage of colony inhibition was 63·1 ± 1·5% and 66·4 ± 0·2% respectively. Notably, the decrease of colony formation in the presence of patient lymphocytes, with or without preincubation, was because of the lower number of CFU-G (both P < 0·01) with a mean percentage of CFU-G inhibition 94·1 ± 1·7% and 93·4 ± 2·4% respectively. In contrast, no statistically significant difference was obtained in the total CFU-M plus CFU-GM colony formation in the presence of patient lymphocytes irrespectively of cell preincubation. In the presence of allogeneic normal CD3+ lymphocytes with or without preincubation, a decrease in total CFU formation by patient CD34+ cells was also noted but the difference from the respective baseline cultures was not statistically significant. In these experiments, the mean percentage of CFU inhibition was 13·4 ± 9·7% and 16·0 ± 12·0% respectively.

When normal CD34+ cells were co-cultured with patient CD3+ lymphocytes with and without a 4-h preincubation, the number of total CFUs obtained was significantly lower compared with baseline number (P < 0·0001 and P < 0·001 respectively). This decrease was because of the lower number of both the CFU-G (P < 0·01 and P < 0·001 respectively) and the total CFU-M plus CFU-GM (P < 0·05 and P < 0·01 respectively) colonies. The mean percentage of CFU inhibition was elevated to 68·2 ± 0·6% and 58·4 ± 7·9%, respectively, with a CFU-G inhibition of 85·9 ± 6·8% and 95·3 ± 1·9%, respectively, and total CFU-M plus CFU-GM inhibition of 6·6 ± 7·2% and 17·2 ± 2·5% respectively. In the presence of autologous normal CD3+ cells with or without a 4-h preincubation, an increase in CFU formation was obtained although the difference from the baseline was not statistically significant and the mean percentage of increase was 4·3 ± 5·2% and 6·8 ± 6·8% respectively. All these ex vivo findings indicate that activated T-lymphocytes in patients’ BM microenvironment may also exert in vivo myelosuppression, therefore contributing to the development of impaired granulopoiesis in CIN.

Assessment of clonality of BM T cells

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

Having demonstrated the existence of activated T-lymphocytes with myelosuppressive properties in CIN BM, we next tested the hypothesis of clonal expansion of BM CD3+ T cells in the patients. A clonal proliferation might reflect a specific, antigen-driven recognition of BM myeloid progenitor cells by patient T-lymphocytes. However, TcR Vβ repertoire flow cytometric analysis of BM CD3+ cells did not reveal a monoclonal or oligoclonal expansion in any of the 25 CIN patients studied, or any of the 38 BM CIN samples analysed following BIOMED-2 protocols was found to display a monoclonal TcRβ, TcRγ or TcRδ gene rearrangement in DNA-PCR and heteroduplex analysis, suggesting that an antigen-driven T cell attack on the myeloid progenitor cells cannot be postulated in CIN patients.

Discussion

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References

The growth and development of haemopoietic cells in the BM occur in intimate association with cells and biosynthetic components of the BM microenvironment. In addition to native cells of mesenchymal and haemopoietic origin, the BM microenvironment contains periphery-derived immune cells, such as lymphocytes and monocyte/macrophages, which either settle in the BM or continuously traffic between BM and PB. It is, therefore, conceivable that inflammatory cells from the periphery may affect the growth and survival characteristics of BM haemopoietic stem cells and their progeny. An autoimmune process targeting the BM haemopoietic progenitor cells has been implicated in the BM damage associated with certain BM failure syndromes (Young, 2000b; Barrett et al, 2000; Zoumbos et al, 1985; Kook et al, 2001), while an immune-mediated suppression of haemopoiesis as a bystander phenomenon has been implicated in marrow failure associated with rheumatoid arthritis and systemic lupus erythematosus (Papadaki et al, 2001d, 2002).

Recent evidence suggests that the BM microenvironment may exert an inhibitory effect on myelopoiesis in certain cases of CIN and may have a role in the pathogenesis of the disease, at least in part, because of overproduction of pro-inflammatory cytokines that induce Fas-mediated apoptotic depletion of the granulocytic progenitor cells (Papadaki et al, 2003). In the current study we have explored further the basis of the BM microenvironment defect in CIN patients by assessing its functional properties as an integral unit of various cell types but also, by identifying individual abnormal cellular components. As LTBMCs represent the most physiological in vitro model to study haemopoiesis, we have used this culture system to establish stromal layers consisting of macrophages, lymphocytes and cells of mesenchymal origin to mimic the in vivo BM microenvironment.

We found that 93·3% of CIN patients abnormally expressed IFNγ and/or FasL mRNA in their LTBMC adherent cell extracts. Although IFNγ and FasL are mainly produced by activated T-lymphocytes, circumstantial evidence has shown that monocyte/macrophages may also express these cytokines upon activation (Rothfuchs et al, 2001; Kitagawa et al, 1998). Therefore, to identify the cellular elements responsible for IFNγ and/or FasL overproduction in BM microenvironment in our patients, we separately assessed lymphocytes and monocytes purified from patient LTBMC stromas and BMMC fractions. IFNγ and/or FasL mRNA were detectable in cell extracts derived from the lymphocyte but not the CD14+ nonocyte/macrophage component, a finding suggesting the presence of activated T-lymphocytes in patients’ BM. The increased percentage of cells bearing activation markers in flow cytometric analysis of BM and PB T-lymphocytes corroborates this hypothesis. Interestingly, although an activation immunophenotypic pattern was observed in both PB and BM T-lymphocytes of the patients, the intracellular IFNγ expression upon stimulation was markedly increased in the BM but not in the PB T-lymphocytes of the patients compared with controls. Furthermore, IFNγ expression was significantly increased in BM compared with PB T- lymphocytes in the patients but not in the controls, suggesting a more prominent activation status of BM T cells compared with PB in CIN. Elevated proportion of T-lymphocytes was also detected immunophenotypically and immunohistochemically in patient BM specimens. Taken together, the above findings suggest that activated T-lymphocytes preferentially accumulate in BM rather than in PB of CIN patients.

The increased proportion of activated T-lymphocytes in patients’ BM raises the issue of their possible involvement in the pathogenesis of the abnormal granulopoiesis in CIN and, more precisely, in the Fas-mediated apoptotic death of CD34+/CD33+ granulocytic progenitor cells previously described in CIN patients (Papadaki et al, 2003). A T-lymphocyte-mediated progenitor cell inhibition manifested in vitro by the co-culture suppression of colony formation by patient lymphocytes as well as by the activated state of PB and BM cytotoxic lymphocytes, has been implicated in the immune suppression of haemopoiesis associated with aplastic anaemia (Maciejewski et al, 1995, Young, 2000b) and myelodysplastic syndromes (Dunbar & Saunthararajah, 2000). The favourable haematological responses of a proportion of such patients to immunosuppressive treatment strongly support a pathogenetic role of T-lymphocytes in these conditions (Barrett et al, 2000; Young, 2000b; Selleri et al, 2002). In CIN patients, however, the usually benign and uncomplicated course of the disease does not ethically permit immunosuppressive therapy and, therefore, clinical evidence substantiating the T-lymphocyte-mediated myelosuppression in vivo cannot be obtained.

To investigate the functional role of activated T-lymphocytes in the pathogenesis of CIN, we performed co-culture experiments of patient T-lymphocytes with autologous and allogeneic normal CD34+ cells. A significant inhibition of both autologous and allogeneic CFU growth by patient CD3+ T-lymphocytes was observed, suggesting a myelosuppressive role of T-lymphocytes in CIN patients. The fact that no statistically significant difference was found in colony inhibition between cultures with and without preincubation of CD34+ with CD3+ cells indicates the prominent role of a soluble inhibitory activity rather than a cell-to-cell contact inhibitory effect. The precise mechanism, however, for the selective CD34+/CD33+ cell inhibition in CIN patients remains unclear as no convincing evidence for involvement of a specific, antigen-driven T-cell response against the granulocytic progenitor cells was demonstrated in our patients. In fact, if a granulocytic progenitor cell-associated antigen was responsible for inciting T-lymphocytes to attack BM CD34+/CD33+ progenitors, then clonal or oligoclonal expansion of T-lymphocytes should be observed in CIN BM. However, such a clonal expansion was not identified in either the flow cytometric analysis of the TcR Vβ repertoire or the study of the TcRβ, TcRγ and TcRδ gene rearrangements with PCR. All these findings provide strong support for a polyclonal T-lymphocyte activation in CIN patients, resulting in myelosuppression by means of production of myelosuppressive cytokines rather than by a cell-to-cell interaction with the granulocyte progenitor cells.

In conclusion, patients with CIN have an increased proportion of activated T-lymphocytes in both BM and PB. The expansion of T cells in patients’ BM is associated with overexpression of IFNγ and FasL, two molecules with potent inhibitory effects on myelopoiesis. BM-derived T-lymphocytes exert significant suppression on myeloid progenitor cell growth in vitro. We suggest that a localized polyclonal cellular immune reaction, resulting in overproduction of inhibitory cytokines and proapoptotic molecules with subsequent Fas-mediated apoptotic death of CD34+/CD33+ granulocytic progenitor cells, represents an important pathogenetic mechanism for the disease.

References

  1. Top of page
  2. Summary
  3. Patients, materials and methods
  4. Patients
  5. Bone marrow samples
  6. Purification of BM cell subpopulations
  7. Long-term BM cultures
  8. Reverse transcription polymerase chain reaction and Southern blot analysis
  9. Flow cytometry and BM immunohistochemistry
  10. Assessment of clonality of BM CD3+ cells
  11. Co-culture of BM CD34+ and CD3+ cells
  12. Statistical analysis
  13. Results
  14. Identification of IFNγ and/or FasL producing cells in CIN BM
  15. Flow cytometric analysis of BM and PB lymphocyte subsets
  16. Intracellular IFNγ staining
  17. Effect of BM CD3+ cells from CIN patients on CD34+ progenitor cell growth
  18. Assessment of clonality of BM T cells
  19. Discussion
  20. Acknowledgements
  21. References
  • Barrett, J., Saunthararajah, Y. & Molldrem, J. (2000) Myelodysplastic syndrome and aplastic anemia: distinct entities or diseases linked by a common pathophysiology? Seminars in Hematology, 37, 1529.
  • Coutinho, L.H., Gilleece, M.H., De Wynter, E.A., Will, A. & Testa, N.G. (1993) Clonal and long-term cultures using human bone marrow. In: Haemopoiesis. A Practical Approach (ed. by N.G.Testa & G.Molineux), pp. 75106. Oxford University Press, Oxford.
  • Dale, D.C. (1991). Neutrophil disorders: Benign, quantitative abnormalities of neutrophils. In: Hematology (ed. by W.J.Williams, E.Beutler, A.J.Erslev & M.A.Lichtman), pp. 807816. McGraw-Hill Publishing Company, New York, MY.
  • Dale, D.C., Guerry, D., Wewerka, J.R., Bull, J.M. & Chusid, J. (1979) Chronic neutropenia. Medicine (Baltimore), 58, 128144.
  • Dunbar, C.E. & Saunthararajah, Y. (2000) Myelodysplastic syndromes. In: Bone Marrow Failure Syndromes (ed. by N.S.Young), pp. 6998. W.B. Saunders Company, Philadelphia, PA.
  • Eliopoulos, G.D., Meletis, J., Fessas, P. & Anagnou, N.P. (1990) Defective CSA-dependent granulopoiesis in patients with chronic idiopathic and drug-induced neutropenia. Haematologia, 23, 101109.
  • Greenberg, P.L., Mara, B., Steed, S. & Boxer, L. (1980) The chronic idiopathic neutropenia syndrome: correlation of clinical features with in vitro parameters of granulocytopoiesis. Blood, 55, 915921.
  • Haddy, T.B., Rana, E.R. & Castro, O. (1999) Benign ethnic neutropenia: what is a normal absolute neutrophil count? Journal of Laboratory and Clinical Medicine, 133, 1522.
  • Hsu, S.M. & Raine, L. (1981) Protein A, avidin and biotin in immunohistochemistry. Journal of Histochemistry and Cytochemistry, 29, 13491353.
  • Kitagawa, M., Yamaguchi, S., Takahashi, M., Tanizawa, T., Hirokawa, K. & Kamiyama, R. (1998) Localization of Fas and Fas ligand in bone marrow cells demonstrating myelodysplasia. Leukemia, 12, 486492.
  • Kook, H., Zeng, W., Guibin, C., Kirby, M., Young, N.S. & Maciejewski, J.P. (2001) Increased cytotoxic T cells with effector phenotype in aplastic anemia and myelodysplasia. Experimental Hematology, 29, 12701277.
  • Kyle, R.A. (1980) Natural history of chronic idiopathic neutropenia. The New England Journal of Medicine, 302, 908909.
  • Kyle, R.A. & Linman, J.W. (1968) Chronic idiopathic neutropenia. A newly recognized entity? The New England Journal of Medicine, 279, 10151019.
  • Maciejewski, J.P., Hibbs, J.R., Anderson, S., Katevas, P. & Young, N.S. (1994) Bone marrow and peripheral blood lymphocyte phenotype in patients with bone marrow failure. Experimental Hematology, 22, 11021110.
  • Maciejewski, J.P., Selleri, S., Sato, T., Cho, H.J., Keefer, L.K., Nathan, C.F. & Young, N.S. (1995) Nitric oxide suppression of human hematopoiesis in vitro. Contribution to inhibitory action of interferon-gamma and tumor necrosis factor-alpha. Journal of Clinical Investigation, 96, 10851092.
  • Oyaizu, N., Than, S., McCloskey, T.W. & Pahwa, S. (1995) Requirement of p56lck in T-cell receptor/CD3-mediated apoptosis and Fas-ligand induction in Jurkat cells. Biochemical and Biophysical Research Communications, 213, 9941001.
  • Palmblad, J., Papadaki, H.A. & Eliopoulos, G.D. (2001) Acute and chronic neutropenias. What is new? Journal of Internal Medicine, 250, 476491.
  • Papadaki, H.A. & Eliopoulos, G.D. (2003) The role of apoptosis in the pathophysiology of chronic neutropenias associated with bone marrow failure. Cell Cycle, 2, 447451.
  • Papadaki, H.A., Xylouri, I., Coulocheri, S., Kalmanti, M., Kafatos, A. & Eliopoulos, G.D. (1999a) Prevalence of chronic idiopathic neutropenia of adults among an apparently healthy population living on the island of Crete. Annals of Hematology, 78, 293297.
  • Papadaki, H.A., Giouremou, K. & Eliopoulos, G.D. (1999b) Low frequency of myeloid progenitor cells in chronic idiopathic neutropenia of adults may be related to increased production of TGF-beta1 by bone marrow stromal cells. European Journal of Haematology, 63, 154162.
  • Papadaki, H.A., Chatzivassili, A., Stefanaki, K., Koumaki, V., Kanavaros, P. & Eliopoulos, G.D. (2000) Morphologically defined myeloid cell compartments, lymphocyte subpopulations, and histological findings of bone marrow in patients with nonimmune chronic idiopathic neutropenia of adults. Annals of Hematology, 79, 563570.
  • Papadaki, H.A., Palmblad, J. & Eliopoulos, G.D. (2001a) Non-immune chronic idiopathic neutropenia of adult: an overview. European Journal of Haematology, 67, 3544.
  • Papadaki, H.A., Eliopoulos, D.G., Valatas, V., Eliopoulos, G.D. (2001b) Anemia of chronic disease is the most frequent type of anemia seen in patients with chronic idiopathic neutropenia of adults. Annals of Hematology, 80, 195200.
  • Papadaki, H.A., Eliopoulos, G.D., Coulocheri, S.A., Spyropoulou, M. & Stavropoulos-Giokas, C. (2001c) Increased frequency of HLA-DRB1*1302 haplotype in patients with nonimmune chronic idiopathic neutropenia of adults. Blood, 97, 580581.
  • Papadaki, H.A., Boumpas, D.T., Gibson, F.M., Jayne, D.R., Axford, J.S., Gordon-Smith, E.C., Marsh, J.C.W. & Eliopoulos, G.D. (2001d) Increased apoptosis of bone marrow CD34+ cells and impaired function of bone marrow stromal cells in patients with systemic lupus erythematosus. British Journal of Haematology, 115, 167174.
  • Papadaki, H.A., Kritikos, H.D., Gemetzi, C., Koutala, H., Marsh, J.C.W., Boumpas, D.T. & Eliopoulos, G.D. (2002) Bone marrow progenitor cell reserve and function and stromal cell function are defective in rheumatoid arthritis: evidence for a tumor necrosis factor alpha-mediated effect. Blood, 99, 16101619.
  • Papadaki, H.A., Eliopoulos, A.G., Kosteas, T., Gemetzi, C., Damianaki, A., Koutala, H., Bux, J. & Eliopoulos, G.D. (2003) Impaired granulocytopoiesis in patients with chronic idiopathic neutropenia is associated with increased apoptosis of bone marrow myeloid progenitor cells. Blood, 101, 25912600.
  • Rothfuchs, A.G., Gigliotti, D., Palmblad, K., Andersson, U., Wigzell, H. & Rottenberg, M.E. (2001) IFN-alpha, beta-dependent, IFN-gamma secretion by bone marrow-derived macrophages controls an intracellular bacterial infection. Journal of Immunology, 167, 64536461.
  • Selleri, C., Sato, T., Anderson, S., Young, N.S. & Maciejewski, J.P. (1995) Interferon-gamma and tumor necrosis factor-alpha suppress both early and late stages of hematopoiesis and induce programmed cell death. Journal of Cell Physiology, 165, 538546.
  • Selleri, C., Maciejewski, J.P., Catalano, L., Ricci, P., Andretta, C., Luciano, L. & Rotoli, B. (2002) Effects of cyclosporine on hematopoietic and immune functions in patients with hypoplastic myelodysplasia: in vitro and in vivo studies. Cancer, 95, 19111922.
  • van Dongen, J.J., Langerak, A.W., Bruggemann, M., Evans, P.A., Hummel, M., Lavender, F.L., Delabesse, E., Davi, F., Schuuring, E., Garcia-Sanz, R., van Krieken, J.H., Droese, J., Gonzalez, D., Bastard, C., White, H.E., Spaargaren, M., Gonzalez, M., Parreia, A., Smith, J.L., Morgan, G.J., Kneba, M. & MacIntyre, E.A. (2003) Design and standarization of PCR primers and protocols for detection of clonal immunoglobulin and T-cell receptor gene recombinations in suspect lymphoproliferations: report of the BIOMED-2 Concerted Action BMH4-CT98-3936. Leukemia, 17, 22572317.
  • Young, N.S. (2000a) Acquired aplastic anemia. In: Bone Marrow Failure Syndromes (ed. by N.S.Young), W.B. Saunders Company, Philadelphia, PA.
  • Young, N.S. (2000b) Hematopoietic cell destruction by immune mechanisms in acquired aplastic anemia. Seminars in Hematology, 37, 314.
  • Zoumbos, N.C., Gascon, P., Djeu, J.Y., Trost, S.R. & Young, N.S. (1985) Circulating activated suppressor T lymphocytes in aplastic anemia. The New England Journal of Medicine, 312, 257265.