• bone marrow;
  • T lymphocytes;
  • myelodysplastic syndrome;
  • immunosuppresion


  1. Top of page
  2. Abstract
  6. References

We have demonstrated that 44% of myelodysplastic syndrome (MDS) patients with cytopenia have a haematological response to antithymocyte globulin (ATG). Three ATG responders and two non-responders with refractory anaemia were further studied for lymphocyte-mediated inhibition of bone marrow using a standard CFU-GM assay. In responders, peripheral blood lymphocytes (PBL) added at a 5:1 ratio suppressed CFU-GM by 54 ± 9% (P = 0.04) and was reversed by ATG treatment. Pre-treatment marrow depleted of CD3 lymphocytes, increased CFU-GM by 32% (P = 0.02) in an ATG responder, but not in a non-responder. CD3 lymphocytes from 6-month post-treatment marrow did not inhibit pre-treatment CFU-GM, indicating ATG had affected the T cells. Pre-treatment marrow depleted of CD8 lymphocytes, increased CFU-GM by 60% (P = 0.01) and 49% (P = 0.03) in two ATG responders, but not in a non-responder. Inhibition required cell–cell interaction through MHC I. TCR Vβ families, analysed by SSCP, changed from clonal to polyclonal in one ATG responder after 6 months, but clones persisted in a non-responder. These results indicate patients with refractory anaemia who respond to ATG have CD8 T-cell clones that mediate MHC-I-restricted suppression of CFU-GM which are replaced by polyclonal T cells that do not suppress CFU-GM after ATG treatment.

Myelodysplastic syndrome (MDS) is characterized by a varying degree of pancytopenia, the aetiology of which is unclear ( Weimar et al, 1994 ; Sawada, 1996; Duhrsen & Hossfeld, 1996). Since some pancytopenic patients with hypoplastic MDS have shown increased bone marrow function with immunosuppressive therapy ( Tichelli et al, 1988 ; Young et al, 1988 ), we hypothesized that pancytopenia in MDS may be in part lymphocyte-mediated. Immunosuppression with antithymocyte globulin (ATG) might therefore improve bone marrow function by reducing lymphocyte-mediated marrow suppression. Although other investigators have demonstrated T-cell-mediated marrow inhibition in some patients with MDS ( Sugawara et al, 1992 ; Smith & Smith, 1991; Bagby et al, 1980 ), the mechanism of inhibition and the effect of immunosuppressive treatment on abnormal clonal T-cell populations is unknown.

We recently showed that 11/25 cytopenic patients with MDS and cellular bone marrows responded to treatment with antithymocyte globulin with red cell transfusion independence and increased platelet and neutrophil counts ( Molldrem et al, 1997 ). To further investigate the role of immunosuppression in MDS, we studied three ATG responders and two non-responders to characterize changes in the T-cell repertoire and to detect lymphocyte-mediated suppression of bone marrow CFU-GM before and after ATG treatment. We show here that some MDS patients had an abnormal T-cell receptor (TCR) Vβ repertoire and had T-cell-mediated suppression of CFU-GM growth. Such patients responded to ATG treatment with a reduction in CD8+ lymphocyte-mediated suppression of CFU-GM growth and a return towards a normal TCR Vβ repertoire.


  1. Top of page
  2. Abstract
  6. References


25 patients with MDS and cytopenia were treated in a phase II trial evaluating the effect of ATG on recovery from cytopenia ( Molldrem et al, 1997 ). All patients required regular red cell transfusions prior to ATG. The protocol was approved by the Institutional Review Board of the National Heart, Lung and Blood Institute. After informed consent was obtained, patients were given ATG (horse ATG from Pharmacia & Upjohn, Bridgewater, New Jersey), 40 mg/kg/d for 4 d with prednisone 1 mg/kg/d for 10 d which was tapered over 7 d. Bone marrow and blood samples for study were obtained immediately before and 6 months after ATG treatment. Of 14 patients with refractory anaemia (FAB subtype RA), three ATG responders and two non-responders with a comparable degree of cytopenia were selected for study based on the availability of sufficient material for the colony inhibition assays described. Patient details and their outcome after ATG are shown in Table I.


Peripheral blood mononuclear cells (PBMC) and bone marrow mononuclear cells (BMC) were obtained by centrifugation over Ficoll-Hypaque density gradient (Organon Teknika Company, Durham, N.C.) and were either used fresh for RNA extraction or frozen in RPMI-1640 complete medium (CM; 25 mmol/l HEPES buffer, 2 mmol/l L-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin; GIBCO-BRL, Gaithersberg, Md.) supplemented with 20% heat-inactivated pooled human AB serum (HS; Pel-Freez, Brown Deer, Wis.) and 10% dimethyl sulphoxide (DMSO) and stored in liquid nitrogen for use in CFU-GM assays.

Lymphocyte depletion and selection

BMC were depleted of CD3+ cells by immunoabsorption with anti-CD3 coated beads) Dynal, Oslo, Norway), and CD8+ depletion by immunoabsorption with anti-CD8 coated beads (Dynal, Oslo, Norway) according to the manufacturer's instructions. Previously cryopreserved BMC were thawed and washed three times in CM supplemented with 10% HS and 5 × 106 cells were combined with 25 μl of beads (1 × 107 beads/ml) in a test tube for 10 min at 4°C. Cells were then separated from the BMC sample by magnetic separation. Non-absorbed marrow cells were used for plating in methylcellulose for CFU-GM assays. The immunoadherent CD3+ or CD8+ lym-phocytes were collected separately and added back to BMC previously depleted of lymphocytes in some experiments.

Colony inhibition assays

Previously cryopreserved marrow-derived cells were thawed and washed three times with CM supplemented with 10% HS and cell viability, measured by trypan blue (ABI, Columbia, Md.) exclusion, was consistently >90%. Marrow cells were then suspended in 0.5 ml of CM and combined with 2.5 ml of methylcellulose containing 50 ng/ml recombinant human stem cell factor (rhSCF), 10 ng/ml rhIL-3, 10 ng/ml rhGM-colony stimulating factor (rhGM-CSF) and 30% fetal bovine serum (FBS), but without erythropoietin (Methocult GF H4534; Stem Cell Technologies, Vancouver, British Columbia, Canada) to achieve a final cell concentration of 1 ×105 cells/ml. Cells were then plated in triplicate in 24-mm wells at 1 ml per well and incubated at 37°C before counting CFU-GM on day 14.

Peripheral blood lymphocytes (PBL) were thawed and washed three times with CM supplemented with 10% HS and added to BMC at a ratio of 5:1 and then incubated at 37°C for 4 h in 0.5 ml of CM supplemented with 10% HS. These co-cultures were then added to 2.5 ml of methylcellulose for CFU-GM assays.

In some experiments previously immunoabsorbed CD3+ or CD8+ lymphocytes were combined with CD3+ or CD8+ lymphocyte-depleted marrow cells taken from the patient prior to ATG treatment at a ratio of 5:1 and co-incubated for 4 h at 37°C in 0.5 ml of CM + 10% HS. At the end of 4 h, cells were combined with 2.5 ml of Methocult GF H4534 and plated in triplicate at 1 × 105 target cells/ml as described above.

In experiments where supernatant was added to lymphocyte-depleted marrow target cells, 0.5 ml of cell supernatant was collected after a 4 h co-incubation of CD8+ lymphocytes with BMC at a ratio of 5:1. The supernatant was then added to separate CD8-depleted marrow cells which were combined with 2.5 ml of Methocult GF H4534 and plated in triplicate at 1 × 105 target cells/ml as described above.

Mouse monoclonal antibody W6/32 (ATCC, Rockville, Md.) against HLA-A, B and C was used to block MHC-I recognition of target cells by lymphocytes. 20 μl of antibody was added to CD8+ lymphocyte-depleted BMC and incubated for 30 min at 37°C prior to the addition of CD8+ lymphocytes at an E:T ratio of 5:1. Cells were suspended in 0.5 ml of CM + 10% HS and combined with 2.5 ml of Methocult GF H4534 and plated in triplicate at 1 × 105 target cells/ml as previously described.

CFU-GM were counted on day 14 and the mean colony counts of three wells was calculated with standard deviations. Statistical significance was determined using the paired two-tailed Student's t-test.

RNA extraction and cDNA synthesis

PBMC or BMC from cryopreserved samples or freshly isolated blood samples were washed once and transferred into 1.8 ml eppendorf tubes. Total RNA extraction was carried out using the RNA STAT-60 kit (TEL-TEST B Inc., Friendswood, Texas) according to the manufacturer's recommended procedure. Primary strand cDNA synthesis was carried out using 0.6 μg of total RNA with the cDNA synthesis kit commercially obtained from Perkin Elmer (Roche Molecular Systems Inc., Branchburg, N.J.). To avoid variations in cDNA synthesis between samples, aliquots of a stock cDNA mixture were used. One vial of cDNA from the stock preparation was used for each reaction to which 0.6 μg of RNA was added. The final volume of each reaction was 20 μl. After cDNA synthesis (60 min at 42°C), 80 μl of DEPC treated H2O was added and cDNA was either stored at −80°C or used for polymerase chain reaction (PCR).

Quantitation of Vβ families using PCR

For PCR, 2.5 μl of cDNA was combined with 25 p M of each primer set (a Vβ sense primer and a Cβ antisense primer as previously published ( Jiang et al, 1997 )) and the reaction was performed using dNTP at 200 μM, and PCR buffer with final concentrations of 1.5 m M MgCl2, 75 m M KCl, 10 m M Tris-HCl, and Taq polymerase buffered to pH 9.2. The final volume was 25 μl and PCR was performed for the number of cycles indicated below using the following conditions: 95°C for 1 min, 55°C for 1 min, 72°C for 1 min with final extension at 72°C for 8 min. A negative control without the addition of cDNA and a positive control using glyceraldehyde phosphate dehydrogenase (GADPH)-specific primers were also used for each PCR assay. For semi-quantitative PCR, 2.22 kBq radioactive dCTP (Amersham International, Amersham, U.K.) was added to each reaction and 26 cycles of PCR was performed. In order to reduce variation between PCR assays, a stock solution containing dNTP, PCR buffer, primers, and Taq polymerase was prepared. When needed for PCR, only H2O, 32P and cDNA from the individual patient samples was added to the stock solution.

Fifteen microlitres of the PCR product were loaded with DNA loading dye (bromphenol blue and xyline) onto an 8% acrylamide gel (13 ml of 30% acrylamide:protogel (National Diagnostics, Atlanta, Ga.), 32 ml H2O 5 ml 10× TBE, 0.5 ml 10% ammonium persulphate, 50 μl TEMED (Bio-Rad Lab, Hercules, Calif.), for a 50 ml gel). The electrophoresis was performed at 250 mV for 3 h. The gel was then dried, exposed to a phosphor imager, and scanned (Molecular Dynamics, Sunnyvale, Calif.) for quantitation of the signals. Percentage expression of Vβ subfamilies was then calculated by dividing each Vβ signal by the sum of all Vβ signals.

Single-stranded DNA conformational polymorphism (SSCP) analysis

PCR, performed for all of the Vβ families using PBMC samples obtained from patients before and 6 months after ATG treatment, was carried out for 32 cycles under identical conditions as outlined above. 4 μl of PCR product was then mixed with 16 μl of stop solution (95% formamide, 0.3% Bromphenol Blue, 0.3% xyline, 10 m M EDTA). The resultant suspension was heated to 95°C for 10 min and immediately transferred to ice cold water before loading onto a 10% polyacrylamide gel in TBE. The gel was run under a constant temperature of 26°C using a circulating water pump system at 310 mV for 1.5 h, and then stained in TBE+ silver-green with gentle agitation for 30 min before photographing.

Flow cytometry and antibodies

PBMC and BM cells were stained using fluorochrome-labelled monoclonal antibodies for CD3, CD4, CD8, CD16, CD56 and CD45. Total nucleated cells and lymphocyte subsets were determined using a CD45 and side-scatter gate. Ten thousand events were acquired with the Becton Dickinson FACScan for each cell marker, and data were analysed using Lysis II (Becton Dickinson, San Jose, Calif.) software.


  1. Top of page
  2. Abstract
  6. References

Effect of ATG treatment

All five patients had similar degrees of cytopenia before ATG treatment and retained the morphologic characteristics of RA without excess blasts during the study period ( Table I). None of the patients had cytogenetic abnormalities. Patients UPN 12, 21 and 16 responded within 10 weeks to ATG treatment, with independence from red cell and platelet transfusions and a sustained increase in the neutrophil count to >×109/l. All three responses were maintained at the time of the 6-month post-treatment studies. UPN 6 and 10 showed no response to ATG; they continued to require transfusions and had no increase in neutrophil or platelet counts ( Molldrem et al, 1997 ).

CFU-GM are suppressed by pre-treatment but not post-treatment PBL in ATG responders

Table II shows the number of day 14 CFU-GM from each patient before and 6 months after ATG therapy. The number of colonies increased by 52 ± 24% after 6 months in the responders, whereas the colonies remained unchanged in the non-responders (P < 0.05). PBL collected prior to ATG treatment were then added to BM cells (also collected prior to ATG) at a 5:1 ratio for 4 h prior to plating in methylcellulose for counting CFU-GM, as described. The percent decrease in the number of CFU-GM when PBL were co-incubated with BM is also shown in Table II. In the responders, PBL added to BM decreased the number of CFU-GM by 54 ± 9% (P = 0.03). This effect was reversed by 6 months after ATG. In contrast, no significant change in the number of CFU-GM was noted in the non-responders, either before or after ATG.

T-cell depletion increases CFU-GM in ATG-responders

The effect of T lymphocytes on CFU-GM was examined by plating BMC or CD3+ lymphocyte-depleted BMC in methylcellulose from UPN 12 (responder) and UPN 6 (non-responder). Fig 1 shows that CFU-GM increased by 32% (P = 0.02) when BMC from UPN 12 were depleted of CD3+ lymphocyte prior to ATG treatment. By 6 months, after the patient had recovered from cytopenia, CD3+ lymphocyte depletion did not increase CFU-GM. In contrast, UPN 6 showed no effect of CD3+ lymphocyte depletion on CFU-GM either before or after ATG treatment.


Figure 1. 4 of culture. Three replicate wells were used to determine CFU-GM and data are displayed as mean colony counts ±standard deviation.

Download figure to PowerPoint

Reversal of the CD3+ lymphocyte depletion-induced increase in CFU-GM seen in UPN 12 could have resulted from a change in either the BMC or the lymphocytes after ATG treatment. To investigate this, CD3+ lymphocytes positively selected from UPN 12 BMC 6 months after treatment were combined at a ratio of 5:1 with previously CD3+ lymphocyte-depleted BMC from before ATG. Fig 1C shows that there was no effect on CFU-GM when the post-ATG CD3+ lymphocytes were added back to the pre-ATG BMC, demonstrating that the lymphocytes, and not the BMC, had been altered by ATG treatment.

To examine which T lymphocyte subset was affected by ATG treatment, BMC from before and after ATG treatment were depleted of CD8+ lymphocytes in two responders (UPN 21 and 16) and one non-responder (UPN 10) and then plated in methylcellulose. Day 14 CFU-GM were compared to BMC not depleted of CD8+ lymphocytes. Fig 2 shows that in both responders before ATG, CFU-GM increased by 60% and 49% in UPN 21 and 16, respectively (P = 0.01 and P = 0.03, respectively), when BMC were depleted of CD8+ lymphocytes. After ATG treatment, however, CD8+ lymphocyte depletion had no effect on CFU-GM. In contrast, CD8+ lymphocyte depletion in the non-responder UPN 10 had no effect on CFU-GM either before or after ATG treatment.


Figure 2. 1 and 16, respectively (P = 0.01 and P = 0.03, respectively). There was no change in CFU-GM in the non-responder (UPN 10) when CD8+ lymphocytes were depleted. (B) 6 months after ATG treatment, CD8+ lymphocyte depletion had no effect on CFU-GM in either the ATG-responders or the non-responder. Three replicate wells were used to determine CFU-GM and data are displayed as mean colony counts ±standard deviation.

Download figure to PowerPoint

Inhibition of CFU-GM by CD8+ T cells is MHC-I restricted

Because CD8+ lymphocyte depletion in the responders increased CFU-GM prior to ATG, the effect of adding back CD8+ lymphocytes to BMC was investigated. BMC from two responders (UPN 21 and 16) and one non-responder (UPN 10) were depleted of CD8+ cells and incubated overnight. The following day CD8+ lymphocytes were co-incubated with the previously-depleted BMC at a ratio of 5:1 at 37°C for 4 h. Fig 3A shows that, prior to ATG treatment, adding back a five-fold excess of CD8+ lymphocytes caused 57% and 36% inhibition of CFU-GM in the two responders (P = 0.01 and P = 0.04, respectively). When mouse monoclonal antibody to HLA-A, B and C was incubated for 30 min with BMC prior to adding back the CD8+ lymphocytes, CFU-GM numbers were restored by 44% and 35% in UPN 21 and 16, respectively. There was no change in CFU-GM in the non-responder, UPN 10, after CD8+ lymphocytes were added to BMC. The CD8+ lymphocyte mediated inhibition of CFU-GM was reversed following ATG treatment in the two responders, but no inhibition was demonstrated in the non-responder, as shown in Fig 3B.


Figure 3. 5% in ATG-responders UPN 21 and UPN 16, respectively, whereas no effect was seen in the non-responder. (B) 6 months after ATG treatment, CD8+ lymphocytes added back to previously CD8+ depleted marrow had no effect on CFU-GM in either ATG-responders or the non-responder. Three replicate wells were used to determine CFU-GM and data are displayed as mean colony counts ±standard deviation.

Download figure to PowerPoint

Pre-treatment cell supernatant from a responder (UPN 21), taken after 4 h of co-culture of CD8+ lymphocytes with previously CD8+ lymphocyte depleted BMC, was added back to BMC from UPN 21 to evaluate the possible effect of soluble mediators of CFU-GM inhibition. Fig 4 shows there was no difference in day 14 CFU-GM when supernatant was added to pre-treatment or post-treatment marrow, compared to marrow alone, suggesting that cell contact was necessary for CFU-GM inhibition by CD8+ lymphocytes.


Figure 4.  h co-incubation of CD8+ lymphocytes with bone marrow at a ratio of 5:1 has no effect on CFU-GM in an ATG-responder (UPN 21). Supernatants from before or after ATG treatment, added back to fresh bone marrow depleted of CD8+ lymphocytes, did not inhibit CFU-GM before or after ATG treatment, indication that cell contact between CD8+ lymphocytes and bone marrow target cells is required for colony inhibition. Three replicate wells were used to determine CFU-GM and data are displayed as mean colony counts ±standard deviation.

Download figure to PowerPoint

Total lymphocytes and CD8+ T-cell numbers are unchanged by 6 months after ATG

The total number of peripheral blood and marrow lymphocytes was examined by flow cytometry in all five patients before and 6 months after ATG. Dual-positive lymphocyte subsets of CD3+CD8, CD3+CD4 and CD16+CD56 cells were also determined. There was no significant change in either the total number of lymphocytes or the CD4+, CD8+ or NK lymphocyte subset distribution after ATG treatment in any of the patients (data not shown). Therefore the effect of ATG on CFU-GM was not the result of any substantial change in the total number of blood or marrow lymphocytes in the responders compared to the non-responders. In addition, bone marrow cellularity and the number of CD34 cells did not change in any of the five patients during the study period (data not shown).

TCR Vβ repertoire is normalized after ATG treatment only in responsive patients

Since CD8+ lymphocytes reversibly inhibited CFU-GM in ATG responders, and because there was no change in the total lymphocyte number in marrow or peripheral blood, a more sensitive technique was used to investigate abnormally expanded T-cell clones in two ATG responders (UPN 12 and 21) and one non-responder (UPN 6). Each of the 25 TCR Vβ subfamilies was amplified by PCR from both pre-treatment and 6 months post-treatment peripheral blood samples. Percentage expression of each Vβ subfamily, calculated by dividing each Vβ signal by the sum of all Vβ signals, revealed the normal pattern of uneven Vβ family distribution in both patients. Each of the subfamilies was investigated by SSCP analysis in one ATG-responder (UPN 21) and one non-responder (UPN6). As shown in Fig 5, in two of the responders Vβ 6, 8, 13 and 14 were over-represented before ATG treatment and fell by >5% after treatment. In contrast, no Vβ family fluctuations of >5% were seen in the non-responder. After ATG treatment, the prominent clones were lost in the responder (UPN 21), and there was an expansion in the number of additional clones (Fig 6). In contrast, the single bands in the non-responder (UPN 6) did not change after ATG treatment. A T-cell clone-specific for the PPD antigen derived from a separate individual is shown for comparison as the control clone. Taken together, this demonstrates that in the ATG-responsive patients dominant T-cell clones were present prior to ATG treatment, and that these were replaced by a polyclonal T-cell population.


Figure 5. Fig 5. Changes in the expression of T-cell receptor (TCR) Vβ subfamilies were greater in two ATG-responders (UPN 12 and 21) 6 months after ATG treatment, than in a non-responder (UPN 6), when compared to pre-treatment bone marrow lymphocytes. mRNA from bone marrow mononuclear cells was reverse transcribed and amplified by PCR using primers specific for each Vβ subfamily. Percentage expression of each Vβ subfamily was calculated by dividing each Vβ signal by the sum of all Vβ signals, and results are expressed as the change in the percentage of expression of each Vβ subfamily before and 6 months after ATG treatment.

Download figure to PowerPoint


Figure 6. months after (post) ATG treatment. Single bands correspond to prominent T-cell clones, and a smear corresponds to an oligo- or polyclonal T-cell repertoire. A T-cell clone specific for the PPD antigen, established by limiting dilution from a separate individual, is shown for comparison as control clone.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  6. References

In a series of patients with MDS and RA, treatment with ATG resulted in a clinical improvement of the cytopenia in 44% of patients ( Molldrem et al, 1997 ). This suggested to us that in responding patients ATG might have abrogated a lymphocyte-mediated marrow suppression intrinsic to the disease. The five patients we studied (three with a haematological response and two without) were selected based on availability of sufficient samples of peripheral blood and bone marrow in addition to a comparable degree of cytopenia. They showed no change in the marrow cellularity or blast cell content, no disease progression, and no change in the percentage of CD3+ or CD8+ T cells in the blood or BM over the study period. In the responders we found evidence of T-cell-mediated inhibition of marrow progenitors using a CFU-GM colony inhibition assay: When PBL were added to patient BM samples before treatment 36–69% inhibition of CFU-GM occurred but there was no significant inhibition in the non-responders. Depletion of CD3+ or CD8+ T cells from the marrow produced a significant increase in CFU-GM growth in the responders. This effect was lost in post-ATG treatment marrow. Blocking antibody experiments showed that CFU-GM inhibition was MHC class I restricted. The effect did not appear to be cytokine-mediated since supernatant from CD8+ lymphocyte and BM co-cultures did not inhibit patient's CFU-GM. However, we did not investigate specific cytokine secretion during co-culture. The changes in colony growth after ATG treatment appeared to be due to an effect on lymphocytes and not on CFU-GM, since addition of post-treatment T lymphocytes to pre-treatment BM cells did not inhibit CFU-GM. Furthermore, despite a haematological response, CFU-GM numbers did not change significantly in responders. In contrast in the two non-responders we found no evidence of lymphocyte-mediated colony inhibition in either pre- or post-treatment marrow.

We next investigated the T-cell Vβ repertoire to look for abnormalities suggesting clonal T-cell expansion in MDS and to identify changes in the repertoire following ATG treatment. We compared the Vβ repertoire in two responders with a series of nine normal individuals. The pattern of Vβ representation differed significantly from the normal (Fig 5), and evidence of oligoclonality by SSCP was seen in Vβ11, 14, 24, in one responder (Fig 6). After ATG treatment the oligoclonal bands were replaced by a diffuse polyclonal pattern. No comparable changes were seen in the non-responder.

These results therefore suggest that in a subset of patients with refractory anaemia, a haematological response to ATG was associated with the loss of an oligoclonally expanded T-cell population responsible for suppressing marrow function. The mechanism of cytopenia in patients with MDS is not understood. Several studies in MDS show that a high percentage of marrow cells undergo apoptosis ( Raza et al, 1995 , 1996; Lepelley et al, 1996 ; Rajapaksa et al, 1996 ). This has been ascribed to abnormalities of growth and differentiation intrinsic to the myelodysplastic progenitor cell. A T-lymphocyte-mediated attack on myelodysplastic progenitor cells could also induce apoptosis through a classical fas/fas ligand mechanism ( Selleri et al, 1997 ). This would require recognition by the T cells of target antigens unique to MDS cells.

It is not clear how ATG eliminates such autoreactive T cells. ATG is a polyclonal antibody binding to numerous lymphocyte surface molecules causing lymphopenia from direct and indirect lymphocytotoxicity and stimulating cytokine release ( Smith et al, 1985 ; Plantanias et al, 1987 ; Nimer et al, 1991 ; Tong et al, 1991 ). ATG also triggers fas ligand expression on most lymphocytes and increases susceptibility to apoptosis of only preactivated lymphocytes ( Genestier et al, 1998 ). Therefore, in the MDS patients, ATG may have preferentially eliminated clonally expanded activated T cells, thus allowing the normal repertoire to re-expand.

In conclusion, these finding raise important questions about the role of autoreactive lymphocytes in the aetiology of cytopenia in MDS. Since lymphocyte-mediated inhibition of CFU-GM does not occur in every cytopenia patient with MDS, the T-cell response appears to be secondary to the primary marrow disorder.


  1. Top of page
  2. Abstract
  6. References
  • 1
    Bagby, G.C., Gabourel, J.D., Linman, J.W. (1980) Gluticocorticoid therapy in the preleukemic syndrome (hemopoietic dysplasia). Annals of Internal Medicine, 92, 55 58.
  • 2
    Duhrsen, U. & Hossfeld, D.K. (1996) Stromal abnormalities in neoplastic bone marrow diseases. Annals of Hematology, 73, 53 70.
  • 3
    Genestier, L., Fournel, S., Flacher, M., Assossou, O., Revillard, J., Bonnefoy-Berard, N. (1998) Induction of fas (Apo-1, CD95)-mediated apoptosis of activated lymphocytes by polyclonal antithymocyte globulins. Blood, 91, 2360 2368.
  • 4
    Jiang, Y.Z., Mavroudis, D.A., Dermime, S., Molldrem, J., Hensel, N.F., Barrett, A.J. (1997) Preferential usage of T cell receptor (TCR) V beta by allogeneic T cells recognizing myeloid leukemia cells: implications for separating graft-versus-leukemia effect from graft-versus-host disease. Bone Marrow Transplantation, 19, 899 903.
  • 5
    Lepelley, P., Campergue, L., Grardel, N., Preudhomme, C., Cosson, A., Fenaux, P. (1996) Is apoptosis a massive process in myelodysplastic syndromes? British Journal of Haematology, 95, 368 371.
  • 6
    Molldrem, J.J., Caples, M., Mavroudis, D., Plante, M., Young, N.S., Barrett, A.J. (1997) Antithymocyte globulin for patients with myelodysplastic syndrome. British Journal of Haematology, 99, 699 705.
  • 7
    Nimer, S.D., Golde, D.W., Kwan, K., Lee, K., Clark, S., Camplin, R. (1991) In vitro production of granulocyte-macrophage colony-stimulating factor in aplastic anemia: possible mechanisms of action of antithymocyte globulin. Blood, 78, 163 168.
  • 8
    Plantanias, L., Gascon, P., Bielory, L., Griffith, P., Nienhuis, A.W., Young, N.S. (1987) Lymphocyte subsets and lymphokines following anti-thymocyte globulin therapy in patients with aplastic anaemia. British Journal of Haematology, 66, 433 443.
  • 9
    Rajapaksa, R., Ginzton, N., Rott, L.S., Greenberg, P.L. (1996) Altered oncoprotein expression and apoptosis in myelodysplastic syndrome marrow cells. Blood, 88, 4275 4287.
  • 10
    Raza, A., Gezer, S., Mundle, S., Gao, X.Z., Alvi, S., Borok, R., Rifkin, S., Iftikhar, A., Shetty, V., Parcharidou, A. (1995) Apoptosis in bone marrow biopsy samples involving stromal and hematopoietic cells in 50 patients with myelodysplastic syndromes. Blood, 86, 268 276.
  • 11
    Raza, A., Mundle, S., Shetty, V., Alvi, S., Chopra, H., Span, L., Parcharidou, A., Dar, S., Venugopal, P., Borok, R., Gezer, S., Showel, J., Loew, J., Robin, E., Rifkin, S., Alston, D., Hernandez, B., Shah, R., Kaizer, H., Gregory, S. (1996) Novel insights into the biology of myelodysplastic syndromes: excessive apoptosis and the role of cytokines. International Journal of Hematology, 63, 265 278.
  • 12
    Sawada, K. (1996) Proliferation and differentiation of myelodysplastic CD34+ cells. Leukemia and Lymphoma, 22, 43 51.
  • 13
    Selleri, C., Sato, T., Del Vecchio, L., Luciano, L., Barrett, A.J., Rotoli, B., Young, N.S. (1997) Involvement of Fas-mediated apoptosis in the inhibitory effects of interferon-alpha in chronic myelogenous leukemia. Blood, 89, 957 964.
  • 14
    Smith, A.G., O'Reilly, R.J., Hansen, J.A., Martin, P.J. (1985) Specific antibody-blocking activities in antilymphocyte globulin as correlates of efficacy for the treatment of aplastic anemia. Blood, 66, 721 723.
  • 15
    Smith, M.A. & Smith, J.G. (1991) The occurrence subtype and significance of haemopoietic inhibitory T cells (HIT cells) in myelodysplasia: an in vitro study. Leukemia Research, 15, 597 601.
  • 16
    Sugawara, T., Endo, K., Shishido, T., Sato, A., Kameoka, J., Fukuhara, O., Yoshinaga, K., Miura, A. (1992) T cell-mediated inhibition of erythropoiesis in myelodysplastic syndromes. American Journal of Hematology, 41, 304 305.
  • 17
    Tichelli, A., Gratwohl, A., Wuersch, A., Nissen, C., Speck, B. (1988) Antilymphocyte globulin for myelodysplastic syndrome? British Journal of Haematology, 68, 139 140.
  • 18
    Tong, J., Bacigalupo, A., Piaggio, G., Figari, O., Sogno, G., Marmont, A. (1991) In vitro response of T cells from aplastic anemia patients to antilymphocyte globulin and phytohemagglutinin: colony-stimulating activity and lymphokine production. Experimental Hematology, 19, 312 316.
  • 19
    Weimar, I.S., Bourhis, J.H., De Gast, G.C., Gerritsen, W.R. (1994) Clonality in myelodysplastic syndromes. Leukemia and Lymphoma, 13, 215 221.
  • 20
    Young, N., Griffith, P., Brittain, E., Elfenbein, G., Gardner, F., Huang, A., Harmon, D., Hewlett, J., Fay, J., Mangan, K. (1988) A multicenter trial of antithymocyte globulin in aplastic anemia and related diseases. Blood, 72, 1861 1869.