Biology of Blood and Marrow Transplantation
Volume 10, Issue 4 , Pages 259-268, April 2004

Antiviral immunity and T-regulatory cell function are retained after selective alloreactive T-cell depletion in both the HLA-identical and HLA-mismatched settings

  • Jeffrey K. Davies

      Affiliations

    • Department of Haematology, Royal Free & University College Medical School, London, United Kingdom UK
    • Corresponding Author InformationCorrespondence and reprint requests: J.K. Davies, MRCP, MRCPath, Department of Haematology, Royal Free Campus, Royal Free & University College Medical School, London NW3 2PF, UK
    • ,
  • Mickey B.C. Koh

      Affiliations

    • Department of Haematology, Royal Free & University College Medical School, London, United Kingdom UK
    • ,
  • Mark W. Lowdell

      Affiliations

    • Department of Haematology, Royal Free & University College Medical School, London, United Kingdom UK

Received 1 October 2003; accepted 1 December 2003.

Article Outline

Abstract 

Nonselective T-cell depletion reduces the incidence of severe graft-versus-host disease after allogeneic hematopoietic stem cell transplantation, but the cost is delayed and disordered antigen-specific immune reconstitution and increased infection. We use a method of selective depletion of alloreactive T cells expressing the activation marker CD69 after coculture with stimulator cells in a modified or standard mixed lymphocyte reaction. The technique has been shown to reduce alloreactivity while retaining third-party responses in vitro and, in a mismatched murine model, led to donor T-cell engraftment with a virtual absence of graft-versus-host disease and increased survival. We show in a human HLA-mismatched and unrelated HLA-identical setting that this technique retains >80% of specific cellular antiviral activity by cytomegalovirus-tetramer analysis and cytomegalovirus/Epstein-Barr virus peptide-stimulated interferon-γ ELISpot assay. Furthermore, CD4+ CD25+ T-regulatory cells are not removed by this method of selective allodepletion and retain their function in suppressing allogeneic proliferative responses. Preservation of antiviral cytotoxic T lymphocytes in selectively allodepleted stem cell grafts would lead to improved antiviral immunity after transplantation. The retention of immunosuppressive CD4+ CD25+ T-regulatory cells could lead to more ordered immune reconstitution and further suppress alloreactive responses after transplantation.

Keywords:  T-cell depletion, Graft-versus-host disease, Allogeneic hematopoietic stem cell transplantation, T-regulatory cells

 

Back to Article Outline

Introduction 

Graft-versus-host disease (GVHD) is a cause of major morbidity and mortality in allogeneic hematopoietic stem cell (AHSC) transplantation because of the recognition by immunocompetent T cells within the AHSC graft of nonself HLA or minor histocompatability antigens (mHags) present on recipient cells. It has long been established that nonselective T-cell depletion of the AHSC graft reduces the incidence of severe GVHD in both the HLA-matched sibling and HLA-matched unrelated donor transplant setting and even with haploidentical transplants 1, 2, 3.

Although this maneuver leads to a reduction in GVHD mortality after transplantation, the removal of pathogen-specific T cells leads to a delay in quantitative and qualitative immune reconstitution [4]. This translates into delayed recovery of antigen-specific immune responses in the recipient and an increase in the incidence and severity of a broad spectrum of infections after transplantation.

Several methods of selective depletion of alloreactive cells have been shown to be effective at reducing the allogeneic response while maintaining third-party reactivity. All rely on the stimulation of live responder cells (from the donor cell pool) by irradiated stimulator (recipient) cells in a unidirectional mixed lymphocyte reaction (MLR) or modified MLR (mMLR). Alloreactive responder cells may then be identified (and subsequently removed) by their expression of activation antigens—such as CD69 5, 6, CD25 7, 8, 9, 10, 11, or both [12]—or destroyed by FAS-FAS ligand-mediated apoptosis [13] or by photodynamic purging [14]. In vitro studies and murine models have been published for all these techniques demonstrating effective reduction of alloreactivity/GVHD.

Cytomegalovirus (CMV) infection after AHSC transplantation is a significant cause of morbidity and mortality and has a considerable economic effect [15]. A significant number of adults who receive AHSC transplants have had serologically detectable past exposure to CMV. In the series of Meyer et al. [16] from 1986, as many as 70% of CMV-seropositive transplant recipients developed clinically significant CMV infection, and CMV disease was the most common cause of death in such patients. Prophylactic treatment with ganciclovir or acyclovir has only partially reduced post-AHSCT CMV infection and has the theoretical disadvantage of selecting drug-resistant viral strains [17].

The strategy of monitoring patients for viral shedding by the detection of viral genome by (real-time) polymerase chain reaction reduces CMV infection and death in the first 100 days after AHSCT but may have the effect of postponing infection to a later stage after transplantation [18]. Such preemptive treatment is associated with adverse effects, especially (reversible) neutropenia that often requires cessation of treatment. The crucial importance of T cell-mediated immunity in controlling CMV is underlined by the increased rate of CMV infection seen in nonselective T cell-depleted AHSCTs [19].

The cytotoxic T lymphocyte (CTL) response is of critical importance for control of CMV infection, and it has been estimated that 1% to 5% of CTLs in healthy CMV-seropositive individuals have specificity for CMV [20]. In controlling CMV disease, CTL responses are believed to be the main effector arms of the adaptive immune system, although they are critically dependent on CD4+ T-cell help [21]. In the absence of CD4+ help, there may be persistence of virus despite the presence of CD8+ CMV-specific CTLs as the latter become anergic [22].

Nonselective T-cell depletion to reduce GVHD, especially in the unrelated donor setting or in nonidentical transplants, removes both CD8+ CMV-specific CTLs and CD4+ CMV-specific T-helper cells and slows the recovery of CMV-specific immunity. A major advantage of selective alloreactive cell depletion is that it leaves the donor CD8+ and CD4+ T-cell pool intact, and it has been shown to retain functional anti-CMV activity in the HLA-mismatched setting [11]. Epstein-Barr virus (EBV)-driven posttransplantation lymphoproliferative disorder (PTLD) is another cause of significant morbidity and mortality after AHSCT. EBV-specific CTL responses are of paramount importance in PTLD, and infusion of very small numbers of donor-derived EBV-specific CTLs are very effective at suppressing PTLD. Selective alloreactive cell depletion may leave donor EBV-specific CTLs intact in the donor cell pool and confer more active EBV immunity after AHSC transplantation.

CD4+ CD25+ T-regulatory cells, which form fewer than 10% of resting human CD4+ lymphocytes, are now recognized as playing an important role in the suppression of immune responses [23]. These cells are required for the ex vivo induction of tolerance to alloantigen via co-stimulatory blockade and for the inhibition of allogeneic skin graft rejection. There is also evidence that these cells can suppress the alloreactive response and limit the development of GVHD. Depletion of CD4+ CD25+ cells from the donor T-cell inoculum or in vivo CD25 depletion of the recipient before transplantation resulted in increased GVHD mediated by CD4+ or whole T cells in mice. The infusion of freshly purified donor CD4+ CD25+ cells modestly inhibited GVHD when administered in equal numbers with whole CD4+ cells but at higher levels significantly inhibited rapidly lethal GVHD [24].

The kinetics of upregulation of CD69 expression on alloantigen-activated lymphocytes have been determined, and in a unidirectional MLR, the maximal expression of CD69 occurs after 72 to 96 hours—somewhat earlier than the maximal expression of CD25 [25]. CD69 is thus a suitable antigen to use to selectively deplete alloantigen-activated lymphocytes while maintaining maximal cell viability. CD69 has been used to identify and immunomagnetically deplete alloreactive cells after mixed lymphocyte culture with irradiated recipient cells. An mMLR [26] with cytokine pretreatment of stimulator cells has been used to increase their alloimmunogenicity and thus maximize the activation of alloreactive cells in HLA-matched donor/recipient pairs in vitro. After depletion of alloreactive cells, the nonalloreactive cell fractions showed significantly reduced proliferative responses against the original matched stimulators and had preservation of responses to third-party stimulator cells [5].

This technique has also been tested in a completely mismatched nonobese diabetic/severe combined immunodeficient murine model of aggressive GVHD. Murine recipients of infusions of nonmanipulated major histocompatability complex class I and class II mismatched donor T cells experienced a rapid onset of acute (and generally fatal) GVHD. This model is akin to aggressive transfusion-associated GVHD. Recipients of lymphocyte infusions depleted of CD69+ alloreactive donor cells ex vivo and monitored for 10 weeks after infusion demonstrated significantly improved survival (71.4% compared with 12.5% in the nonmanipulated group) and the absence of clinical or histologic evidence of GVHD despite the presence of circulating donor lymphocytes demonstrated by flow cytometry [6].

We have used the CD69-based selective allodepletion technique to remove alloreactive cells from both HLA-identical unrelated donor/recipient pairs and totally HLA-mismatched donor/recipient pairs in which the donor was known to be CMV immunoglobulin (Ig)G positive. CMV-reactive T cells were enumerated before and after selective allodepletion by CMV-peptide specific interferon (IFN)-γ ELISpot assay and by CMV-peptide HLA A0201 tetramer analysis. EBV-reactive T cells by IFN-γ ELISpot were also enumerated before and after selective alloreactive cell depletion in the HLA-mismatched setting. CD4+ CD25+ T-regulatory cell depletion should not occur with our CD69-based strategy, and to confirm this, we enumerated and tested the functionality of CD4+ CD25+ T-regulatory cells after CD69-based selective allodepletion.

Back to Article Outline

Materials and methods 

HLA-A-, -B-, and -DR-identical unrelated donor/recipient pairs and complete HLA-A-, -B-, and -DR-mismatched donor/recipient pairs were chosen with donors seropositive for CMV IgG or EBV nuclear antigen IgG (Table 1). Written consent was obtained for the use of all blood samples. Peripheral blood mononuclear cells (PBMCs) were obtained by density gradient centrifugation (Lymphoprep; Nycomed-Amersham, Buckinghamshire, UK), and cells were used fresh or after storage in liquid nitrogen.

Table 1. HLA Typing of CMV IgG-Positive Responders in Sample Pairs Used for Selective Depletion of Alloreactive Cells and Peptides Used in CMV Peptide-Stimulated IFN-γ ELISpot Assays before and after Selective Depletion of Alloreactive Cells
PairStimulator (Recipient)Responder (Donor) HLA TypingCMV Peptide Used in IFN-γ ELISpot
1HLA mismatchedA3B0702TPRVTGGGAM
A26B0801
2HLA mismatchedA0201B0702NLVPMVATV
A2402B1401
3HLA mismatchedA0101B0702TPRVTGGGAM
A0101B37
4HLA mismatchedA0201B14NLVPMVATV
A25B58
5HLA mismatchedA0201B44NLVPMVATV
A74B38
6HLA mismatchedA0201N/ANLVPMVATV
7HLA mismatchedA0201N/ANLVPMVATV
8Matched at HLA-A, -B, and -DRA0101B0801DANDIYRIF
A0101B0801
9Matched at HLA-A, -B, and -DRA0201B3501NLVPMVATV
A0301B3701
10Matched at HLA-A, -B, and -DRA0101B0702TPRVTGGGAM
A0101B37
11Matched at HLA-A, -B, and -DRA0201B0702NLVPMVATV
A2402B1401
12Matched at HLA-A, -B, and -DRA0201B4402NLVPMVATV
A3101B3503

N/A indicates not assessed.

Alloreactivity was assessed by a primary unidirectional MLR in an unmodified form in the HLA-mismatched pairs. Irradiated (30-Gy) stimulators were cocultured in RPMI 1640 supplemented with 10% human AB serum and 1% penicillin/streptomycin (complete media; all Gibco, Paisley, UK) with live responder cells in 96-well plates (Nunc, Rochester, NY) with a stimulator-responder ratio of 1:1 at 37°C, 5% CO2, and 60% humidity. After 108 hours, 1 μCi of tritiated thymidine (Amersham, Buckinghamshire, UK) was added per 105 responder cells. The plates were harvested at 120 hours on a harvester, and proliferative alloreactive responses were measured as tritiated thymidine incorporation in responder cells by using a beta-scintillation counter (Wallac, Turku, Finland).

In the HLA-matched unrelated donor/recipient samples, the primary MLR was modified by pretreatment of the stimulator cells with OKT3 or cytokines. OKT3 was obtained from a fermentation system by using a murine hybridoma-secreting human mitogenic OKT3 (American Type Culture Collection cell line CRL-8001). The concentration of OKT3 in hybridoma supernatant was quantified by murine IgG enzyme-linked immunosorbent assay, and 500 ng of OKT3 per well was immobilized on a 24-well plate by overnight incubation at 4°C. Wells were then washed twice, and 106 stimulator PBMCs were added per well in complete media. The plates were incubated for 48 hours as described previously, and the cells washed twice before use as stimulators in the MLR. No additional cytokines were added to the MLR in this strategy.

In the HLA-matched unrelated donor/recipient samples, the primary MLR was alternately modified by pretreatment of stimulator cells with recombinant human IFN-γ 1000 IU/106 cells and recombinant human tumor necrosis factor-α 1000 IU/106 cells for 24 hours (R&D Systems, Abingdon, UK). Stimulator cells were then diluted in complete medium and washed twice before use as stimulators in the MLR. Recombinant human interleukin-4 (R&D Systems) was added at a dose of 1000 IU/106 responder cells to the MLR with this strategy. All experiments were performed in triplicate.

Selective allodepletion was achieved by performing a bulk MLR in the HLA-mismatched setting or a bulk OKT3 and cytokine mMLR in the HLA-matched setting by following the protocols outlined previously. However, after 72 hours of coculture, the cells were sorted to deplete the responder cell fraction of CD69+ alloreactive cells. Depletion was achieved by using CD69/fluorescein isothiocyanate (FITC) antibody (Becton Dickinson, Oxford, UK) and anti-FITC microbeads with single use depletion columns and a magnetically activated cell sorter (Miltenyi Biotech, Cologne, Germany). The depleted cell fraction was used as the responder cell fraction in a secondary MLR to confirm reduction in first-party alloreactivity and preservation of third-party alloreactivity. The depletion efficiency (%) of CD69+ cells was defined as

CMV tetramer analysis was performed on unmanipulated PBMCs and on selectively allodepleted PBMCs on HLA-A0201-positive donors by using a phycoerythrin (PE)-conjugated HLA-A0201-restricted human CMV AE42 peptide tetrameric construct (a gift from the Anthony Nolan Research Institute, London, UK). A total of 1 μg of tetramer was incubated with 106 cells for 30 minutes and then counterstained with CD3-allophycocyanin and CD8-FITC (Becton Dickinson), and 104 events were acquired on a FACSCalibur flow cytometer (Becton Dickinson). Data were subsequently analyzed with CellQuest Pro software (BD Biosciences).

A selection of HLA class I-restricted immunogenic human CMV virus peptides were used as stimulatory peptides in an IFN-γ ELISpot assay (MabTech, Nacka, Sweden) by using polyvinylidene difluoride-lined 96-well plates (Millipore, Watford, UK). Peptides used were the A0201-restricted pp65 tegmental-derived nonamer AE42 (NLVPMVATV), the B0702-restricted decamer TPRVTGGGAM, and the B0801-restricted nonamer DANDIYRIF (Proimmune, Oxford UK) 27, 28, 29. Peptides were stored at a high concentration in dimethyl sulfoxide at −70°C and diluted to 10 μg/mL in complete media before use.

The HLA specificity of the AE42 peptide was confirmed before use in a T2-binding assay and was also shown to prevent the binding of PE-conjugated HLA-A0201-restricted human CMV AE42 peptide tetramer to the CD8+ CTLs of a CMV IgG-positive HLA-A0201 subject (data not shown). Stimulatory peptides were also tested in the IFN-γ ELISpot assay on healthy CMV IgG-positive individuals of suitable HLA class I type. Briefly, polyvinylidene difluoride-lined 96-well plated were washed with 70% ethanol, and a capture antibody was added at 1.5 μg per well and incubated overnight at 4°C. After washing and blocking with RPMI 1640 containing 10% fetal calf serum (FCS) (150 μL per well) for 1 hour, cells were washed and live PBMCs were added (105 to 2 × 105 per well).

A total of 1 μg of stimulatory peptide was used per 105 live responder cells incubated for 20 hours as described previously. Cells were then removed and wells washed before the addition of 100 μL of diluted biotinylated antibody in filtered phosphate-buffered saline (PBS) with 0.5% FCS to each well and incubated for 90 minutes at room temperature. Wells were then washed 6 times with sterile, filtered PBS. Then 100 μL of streptavidin/alkaline phosphatase antibody (diluted 1:1000 in filtered PBS/0.5% FCS) was added to each well and incubated for 1 hour at room temperature. Wells were washed 6 times with sterile filtered PBS, and 100 μL of color change substrate (Bio-Rad, Hemel Hempsted, UK) was added per well. Plates were agitated gently at room temperature for 10 to 20 minutes until spots appeared. Color development was stopped by washing in tap water 200 μL per well 3 times. Plates were left to air-dry at room temperature overnight, and wells were photographed with an Olympus (Tokyo, Japan) digital camera and dissection microscope.

Negative controls were cells plus no peptide and cells plus dummy (HLA-mismatched) peptide. Positive controls were cells stimulated with phytohemagglutinin (10 ng/106cells; Sigma-Aldrich, Dorset, UK). All experiments were performed in triplicate. ELISpot-responding cells were expressed as

The frequency of CD3+ CD8+ CMV-responding cells was expressed as (% CD3+CD8+ PBMCs) × number of responding cells. The frequency of CMV-reactive CD3+ CD8+ CTLs was determined in unmanipulated donor samples and in selectively allodepleted donor samples in both the HLA-matched and HLA-mismatched donor/recipient pairs.

A variety of EBV-derived peptides stimulate EBV-CTL responses. We chose to use the SVRDRLARL peptide derived from the latent EBV nuclear antigen 3A, which demonstrates immunodominance in HLA-A0201 individuals [30]. EBV-CTL activity was measured by peptide-stimulated IFN-γ ELISpot by using the reagents and methodology for CMV peptide ELISpot outlined previously.

The frequency of CD4+ CD25+ T-regulatory cells was measured before and after selective allodepletion by using CD25-PE and CD4-peridinin chlorophyll-a protein antibodies (Becton Dickinson), and a FACSCalibur flow cytometer was used for analysis with CellQuest Pro software. After selective allodepletion, the CD69-depleted fraction was sorted by flow cytometry to select the CD4+ CD25+ T-regulatory cell fraction. These sorted T-regulatory cells were added at a ratio of 1:1 to an allogeneic mismatched MLR with first-party stimulators and responders autologous to the T-regulatory cells to assess the immunosuppressive capacity of the T-regulatory cells after selective allodepletion. The T-regulatory negative fraction from the sort was also added to a first-party allogeneic MLR with responders autologous to the T-regulatory negative fraction as a negative control.

The unpaired t test with the Welch correction in the presence of unequal variance and the paired t test were used as tests of statistical significance, and a P value of <.05 was chosen to reject the null hypothesis.

Back to Article Outline

Results 

Allostimulation was effective and allodepletion was efficient 

In the HLA-mismatched pairs, CD69 was upregulated on 5.9% ± 5% (mean ± SD) of responder CD3+ cells, 4.9% ± 2.3% of CD3+ CD4+ cells, and 4.8% ± 4.2% of CD3+ CD8+ cells after allostimulation in the standard MLR at 72 hours. The depletion efficiency of these cells was 92% ± 7%, 91% ± 8%, and 85% ± 10%, respectively.

In the HLA-matched pairs, CD69 expression at 72 hours was not increased above autologous control in the standard MLR (data not shown; P > .05). In contrast, allostimulation in the OKT3 mMLR led to CD69 expression on 20% ± 11% of CD3+ responder cells, 24% ± 15% of CD3+ CD4+ cells, and 15% ± 7% of CD3+ CD8+ cells at 72 hours. The time scale of maximal responder CD69 expression was similar to that seen in HLA-mismatched pairs, although upregulation of CD69 occurred somewhat earlier (data not shown). The depletion efficiency of these cells was 92% ± 6%, 91% ± 8%, and 83% ± 14%, respectively. In the same donor/recipient pairs, allostimulation in the cytokine mMLR led to CD69 expression on 7.2% ± 9% of CD3+ responder cells, 8.3% ± 9% of CD3+ CD4+ cells, and 5% ± 6% of CD3+ CD8+ cells at 72 hours. The time scale of maximal responder CD69 expression was similar to that seen in HLA-mismatched pairs (data not shown). The depletion efficiency of these cells was 86% ± 16%, 86% ± 15%, and 88% ± 12%, respectively. These results are summarized in Table 2.

Table 2. CD69 Expression on Alloreactive Responder Cells and Depletion Efficiency of CD69+ Cells in HLA-Mismatched Standard MLR and HLA-Matched OKT3 and Cytokine-Stimulated MLRs: Mean (SD)
VariableHLA-Mismatched Pairs: Standard MLRHLA-Matched Pairs
CD3+CD3+ CD4+CD3+ CD8+OKT3 MLRCytokine MLR
CD3+CD3+ CD4+CD3+ CD8+CD3+CD3+ CD4+CD3+ CD8+
% Responder CD69+∗5.9 (5)4.9 (2)4.8 (4)20 (11)24 (15)15 (7)7.2 (9)8.3 (10)5.0 (6)
CD69+ depletion efficiency92 (7)91 (8)85 (10)92 (6)91 (8)83 (14)86 (16)86 (15)88 (12)

∗ Above autologous control.

CD69 responses seen after allostimulation in HLA-matched pairs with both the OKT3 mMLR and the cytokine mMLR were not significantly different from those seen after stimulation with HLA-mismatched third-party controls (data not shown). The yield in all depletion procedures was >85% of viable nonalloreactive CD3+ cells.

Allodepletion selectively reduced first-party proliferation in MLRs 

In HLA-mismatched donor/recipient pairs, the first-party alloreactivity in the secondary MLR assay was reduced to 3.98% ± 4.5% of that seen in the primary MLR with retention of third-party proliferative responses (73.2% ± 24.3% of the predepletion value). In HLA-matched donor/recipient pairs, the first-party alloreactivity was reduced to 0.8% ± 0.9% of the predepletion value after selective allodepletion with OKT3-pretreated stimulator cells, with retention of third-party proliferative responses (58% of predepletion values). In HLA-matched donor/recipient pairs, the first-party alloreactivity was reduced to 2.6% ± 4.4% of the predepletion value after selective allodepletion with cytokine-pretreated stimulator cells, with retention of third-party proliferative responses (133% of predepletion values).

CMV tetramer-positive cells were retained after selective alloreactive cell depletion 

HLA-A0201 restriction of the tetramer was demonstrated by tetramer binding to CD3+ CD8+ cells in an HLA-A0201 CMV IgG-positive subject but not in an HLA-B0701 CMV IgG-positive subject. The AE42 peptide specificity was demonstrated by blocking of tetramer binding to CD3+ CD8+ cells in an HLA-A0201 CMV IgG-positive subject by treatment of PBMCs with exogenous AE42 peptide in excess.

In HLA-mismatched donor/recipient pairs after selective allodepletion, 93% ± 9% of CMV-AE42 peptide tetramer-positive CTLs were retained. In HLA-matched donor/recipient pairs after OKT3-stimulated MLR selective allodepletion, 88.5% ± 36% of CMV-AE42 peptide tetramer-positive CTLs were retained, and after cytokine-stimulated MLR-selective allodepletion, 87.0% ± 18.4% of CMV-AE42 peptide tetramer-positive CTLs were retained. An example of AE42 peptide tetramer-positive CTL preservation after HLA-matched selective allodepletion is shown in Figure 1.

  • View full-size image.
  • Figure 1. 

    CMV CTLs are retained after CD69+-based selective allodepletion. a, HLA-A2 CMV AE42 (NLV) peptide tetramer-binding CD3+ cells in an unmanipulated donor cell pool. b, HLA-A2 CMV AE42 (NLV) peptide tetramer-binding CD3+ cells in an HLA-matched selectively allodepleted fraction with OKT3-stimulated MLR. c, HLA-A2 CMV AE42 (NLV) peptide tetramer-binding CD3+ cells in an HLA-matched selectively allodepleted fraction with cytokine-stimulated MLR. (This figure show dot plots from a representative experiment.)

There was no significant difference in paired 2-tailed t test results between the frequency of CMV-AE42 peptide tetramer-binding CTLs before and after selective allodepletion in the HLA-mismatched donor/recipient pairs or in the HLA-matched pairs selectively allodepleted with either the OKT3-stimulated MLR or the cytokine mMLR.

Functional CMV-specific CTLs were retained after selective alloreactive cell depletion 

The CMV peptides and ELISpot assays were optimized and validated in healthy CMV IgG-positive individuals of known HLA class I types. The AE42 peptide was further validated in a T2-binding assay in which the addition of HLA-A0201-restricted peptide to transporter-associated protein-deficient HLA-A0201 T2 cells in culture resulted in stabilization and an increase of expression of HLA-A0201 on the surface of T2 cells of 100% (whereas the addition of an HLA B0702-restricted CMV-peptide did not alter the expression of HLA-0201). The AE42 peptide also increased CMV-specific CTL-mediated killing in a cytotoxicity assay with T2 target cells (data not shown).

In HLA-mismatched donor/recipient pairs after selective allodepletion, 89.2% ± 26.9% of the functional CMV-reactive CTLs by IFN-γ ELISpot were retained. In HLA-matched donor/recipient pairs after OKT3-stimulated MLR-selective allodepletion, 72.3% ± 39.5% of the functional CMV-reactive CTLs by IFN-γ ELISpot were retained. In HLA-matched donor/recipient pairs after cytokine-stimulated MLR-selective allodepletion, 80.7% ± 17.9% of the functional CMV-reactive CTLs by IFN-γ ELISpot were retained.

There was no significant difference in a paired 2-tailed t test between the frequency of functional CMV-reactive CTLs by IFN-γ ELISpot before and after selective allodepletion in the HLA-mismatched donor/recipient pairs or in the HLA-matched pairs selectively allodepleted with either the OKT3-stimulated MLR or the cytokine mMLR. Preservation of both the AE42 peptide tetramer-positive CTLs and CMV-reactive CTLs by IFN-γ ELISpot in both the HLA-mismatched and -matched setting is shown in Figure 2.

  • View full-size image.
  • Figure 2. 

    Functional CMV CTLs are retained after CD69+-based selective allodepletion. Retention of CMV CTLs by HLA-A2 CMV AE42 (NLV) peptide tetramer and retention of functional CMV CTLs by CMV peptide-stimulated IFN-γ ELISpot after selective allodepletion expressed as a percentage of the of value seen in unmanipulated cells.

Functional EBV-specific CTLs were retained after selective alloreactive cell depletion 

Functional EBV-specific CTLs were measured by IFN-γ ELISpot in the HLA-mismatched setting. After selective allodepletion, there was no significant difference in the number of EBV-reactive CTLs compared with the predepletion samples (data not shown).

CD4+CD25+ T-regulatory cells were retained after selective allodepletion 

The level of CD4+ CD25+ T-regulatory cells present in the responder cell population after allostimulation at 72 hours was not significantly altered from the baseline value in unmanipulated cells (CD4+ CD25+ after allostimulation, 14.6% ± 7.9% versus the baseline value, 17.9% ± 1.0%; P = .6) or from the level seen after autologous cell culture (CD4+ CD25+ after allostimulation, 14.6% ± 7.9% versus autologous cell culture, 15.8% ± 9.7%; P = .9).

The level of CD4+ CD25+ T-regulatory cells present in the responder cell population after selective allodepletion at 72 hours was not significantly altered from the predepletion value (CD4+ CD25+ after allostimulation, 14.6% ± 7.9% versus after depletion, 11.4% ± 2.2%; P = .54). In HLA-matched donor/recipient pairs, no difference in the levels of CD4+ CD25+ T-regulatory cells was found either after allostimulation or after selective allodepletion in either the OKT3-stimulated MLR or the cytokine mMLR depletion strategies (data not shown). In both the HLA-mismatched and -matched samples, CD69+ selective allodepletion removed a small fraction of CD25+ cells that also expressed CD25; most of the CD4+ CD25+ cells remained CD69− and were retained in the selectively allodepleted fraction (Figure 3).

  • View full-size image.
  • Figure 3. 

    CD4+ CD25+ T-regulatory cells are not removed by CD69+ selective allodepletion. Allostimulation in the OKT3-stimulated MLR in HLA-matched samples leads to an appearance of both CD69+ CD25− and CD69+ CD25+ alloreactive CD3+ CD4+ T cells, which are removed by selective allodepletion based on CD69 expression. T-regulatory cells (CD4+ CD25+ CD69−) are not removed by this process. A, Allostimulated responder cells; B, untreated responder cells; C, selectively allodepleted responder cells.

CD4+ CD25+ T-regulatory cells present in selectively allodepleted cell fractions retain immunosuppressive function 

CD4+ CD25+ T-regulatory cells sorted after selective allodepletion retained their immunosuppressive function by suppressing allogeneic MLR responses to first-party stimulators. In the mismatched setting, the mean first-party allogeneic MLR response was significantly reduced to 18.9% ± 9.4% of the control value by the addition of sorted selectively allodepleted CD4+ CD25+ T-regulatory cells autologous to and at equal numbers to the MLR responders (P = .005). Addition of autologous non-T-regulatory cells from selectively allodepleted cell fractions in similar quantities did not significantly alter the first-party allogeneic MLR response in this setting (Figure 4).

  • View full-size image.
  • Figure 4. 

    CD4+ CD25+ T-regulatory cells present in selectively allodepleted cells retain immunosuppressive function. Proliferative responses to first-party stimulators in HLA-mismatched allogeneic MLRs is significantly reduced after the addition of CD4+ CD25+ T-regulatory cells (autologous to the responders) from selectively allodepleted cell fractions, whereas proliferation is not significantly affected after the addition of the CD4+ CD25+ T-regulatory cell negative fraction from selectively allodepleted cell fractions.

Back to Article Outline

Discussion 

All modalities of selective alloreactive cell depletion rely on the principle of stimulation of alloreactive cells in a live responder (donor) cell population by irradiated stimulator (recipient) cells in a unidirectional mixed lymphocyte culture. In the HLA-A-, -B-, and -DR-identical setting, we used a technique of pretreatment of stimulator cells with mitogenic OKT3 to increase the capacity of the stimulator cell pool (mixed PBMCs) to present alloantigens. Human T-cell clones have been shown to be capable of presenting antigens [31]. Although the stimulation of T cells with mitogenic OKT3 has been reported to lead to anergy in responder cells when such T cells alone are used as nonprofessional antigen-presenting cells in the MLR [32], here we show that the HLA-matched unrelated donor/recipient pairs tested demonstrated a significant upregulation of CD69 on responder cells and a significant proliferative response in the OKT3-stimulated MLR with mixed PBMCs as stimulators (akin to that seen with the established cytokine mMLR). The stimulation and expansion of T-cell stimulator cells with mitogenic OKT3 has been described recently as part of the protocol for clinical scale application of a selective allodepletion technique [33]. Both this work and other work within our group has shown that OKT3-treated/expanded T cells significantly upregulate molecules involved in alloantigen presentation, such as CD80 and CD86 [34]. Although it could be argued that important alloantigens presented only by monocytes would be lost by the exclusive use of T cells as APCs, our 48-hour pretreatment of PBMCs with OKT3 retains a proportion (5%–10%) of monocytes within the APC pool. Moreover, the use of predominantly T cells as APCs might lead to retention of identified myeloid or B-cell-restricted tumor-associated alloantigens 35, 36, 37 with preservation of a graft-versus-leukemia effect.

In both the HLA-mismatched and -matched unrelated donor/recipient pairs, immunomagnetic depletion of CD69+ responder cells was efficient and led to a reduction in proliferation to the original stimulators to <5% of predepletion values and preservation of proliferative responses to third-party controls. Yields of viable nonalloreactive T cells were commonly 85% to 90%, meaning that the nonalloreactive donor T-cell pool remains quantitatively intact with this strategy. Depletion efficiency of alloreactive T cells was greater than 90%, and residual CD69+ cells left in responder cell pools only weakly expressed the antigen. If infused into the recipient, such cells might still be capable of mounting an alloreactive response and causing GVHD, although if still bound to CD69 antibody they might be easily destroyed by the recipient’s reticuloendothelial system. In the published nonobese diabetic/severe combined immunodeficient mouse model of CD69-selective allodepletion, the depletion efficiency of 93% was sufficient to completely abrogate GVHD in sublethally irradiated major histocompatibility complex nonidentical mice [6]. However, the ability of small numbers of alloreactive cells infused to recipients in the context of the cytokine release that accompanies pretransplantation conditioning remains a potential concern.

Responses to highly immunogenic HLA-disparate antigens might mask responses against minor histocompatability antigens, which might stimulate responders more weakly or more slowly in the MLR. The cytokine mMLR has been shown to be predictive of clinical GVHD in HLA-identical AHSC transplant recipients after 120 hours of coculture, indicating that alloresponders can proliferate to mHags in this time scale [26]. Both the cytokine and the OKT3 pretreatment strategy in the HLA-identical pairs we used led to a maximal CD69 responder expression by 72 hours: thus, alloresponses to mHags do not seem to be delayed with respect to HLA-disparate antigens.

Greater than 80% of HLA-A0201 CMV AE42 peptide tetramer-reactive CTLs were retained in the allodepleted fraction in both the HLA-mismatched and HLA-matched unrelated donor pairs. This is in keeping with a recent study of CD25-based selective allodepletion in HLA-mismatched pairs [11].

However, tetramer analysis gives no indication of functionality; tetramer-binding cells have been reported without cytolytic function, especially in the absence of CD4 cell help [22]. Baseline levels of CMV tetramer-binding CTLs were 3- to 5-fold higher than baseline CMV-reactive CTLs by IFN-γ ELISpot assay, which is in keeping with published data [38].

Thus, the IFN-γ ELISpot assay, a functional test of CMV reactivity, gives us important additional information. We have shown that the great majority of CMV-reactive CTLs by IFN-γ ELISpot assay are retained after selective allodepletion in the HLA-matched setting with HLA-A0201, -B0702, and -B0801 responders, adding to published data using mismatched pairs with HLA-A0201 responders [11].

The dominant CMV peptide response in every individual can vary, even between individuals of the same HLA class I type [39], and thus additional CMV responses may have been retained that cannot be identified by a single HLA-restricted peptide approach.

In the HLA-mismatched setting, we have shown that functional EBV immunity is preserved after selective depletion of alloreactive cells. EBV and other viruses that are controlled primarily by a T cell-mediated immune response are important pathogens after AHSCT; these include respiratory syncytial virus and adenovirus 40, 41, 42. Life-threatening complications arising from infection from all of these viruses after AHSCT have been treated successfully with donor lymphocyte infusions, with presumed transfer of antiviral CTLs 43, 44, 45. Specific cellular antiviral immunity to such viruses might also be preserved by selective alloreactive T-cell depletion of the donor T-cell pool.

Our data also demonstrate the quantitative preservation of CD4+ CD25+ T-regulatory cells after selective allodepletion in both HLA-mismatched donor/recipient and matched unrelated donor/recipient pairs. Moreover, the functionality of such CD4+ CD25+ T-regulatory cells was preserved after selective allodepletion, as demonstrated by retention of their immunosuppressive capability when added to autologous live responder cells in an allogeneic MLR.

In common with CD69, expression of CD25 (the low-affinity interleukin-2 receptor) is upregulated on activated T cells. Alloreactive cells that upregulate CD69 after allostimulation come from both the CD8 and CD4 subsets: some alloreactive cells express only CD69, some express only CD25, and some express both. It is possible that the CD25+ cell population in the allodepleted cell fractions contains some alloreactive cells that have upregulated CD25 after presentation of alloantigens in the MLR. However, reduction of first-party proliferative responses after selective allodepletion to <5% of their predepletion values indicates that any CD69− CD25+ cells that have upregulated CD25 in response to alloantigens and remain after depletion are unable to make a significant proliferative response to first-party stimulators and are thus unlikely to contribute significantly to in vivo allogeneic responses. Purified CD4+ CD25+ cells from allodepleted cell fractions maintained the phenotype in vitro of T-regulatory cells [23] by reducing proliferative responses to first-party stimulators, and it thus seems unlikely that any significant population of alloreactive cells had appeared within the CD4+ CD25+ cell fraction or, if they had, that their responses were overwhelmed by the suppressive effects of the true T-regulatory cells. Other markers of T-regulatory cells are being identified (intracellular CD152 and CD150) [46] and might be useful in further distinction between T-regulatory cells and alloreactive cells in the donor cell pool.

The role of CD4+ CD25+ T-regulatory cells in suppressing alloreactivity is increasingly being recognized, and the loss of such cells from the donor pool might result in increased GVHD. Loss of T-regulatory function is seen in individuals with autoimmune and atopic conditions [47]. Nonselective T-cell depletion leads to disordered immune reconstitution and the loss of hierarchical control, which may be manifested as autoimmune disorders such as autoimmune hemolytic anemia [48].

Selective allodepletion based on the upregulation of CD25 will remove or destroy not only alloreactive cells that have upregulated CD25 on allostimulation, but also any cells that express CD25 constitutively, including CD4+ CD25+ T-regulatory cells: such techniques will therefore deplete the donor T-cell pool of the T-regulatory compartment with the potential of increased alloreactivity and disordered immune reconstitution.

Back to Article Outline

Acknowledgements 

We thank the Anthony Nolan Research Institute for provision of expert HLA typing and the Department of Virology, Royal Free Hospital, for viral serology. The Medical Research Council (J.K.D.) and the Leukaemia Research Fund (M.B.C.K.) supported this work.

Back to Article Outline

References 

  1. Atkinson K, Farrelly H, Cooley M, O’Flaherty E, Downs K, Biggs J. Human marrow T cell dose correlates with severity of subsequent acute graft-versus-host disease. Bone Marrow Transplant. 1987;2:51–57
  2. Ash RC, Casper JT, Chitambar CR, et al.  Successful allogeneic transplantation of T-cell-depleted bone marrow from closely HLA-matched unrelated donors. N Engl J Med. 1990;322:485–494
  3. Aversa F, Tabilio A, Velardi A, et al.  Treatment of high-risk acute leukemia with T-cell-depleted stem cells from related donors with one fully mismatched HLA haplotype. N Engl J Med. 1998;339:1186–1193
  4. Keever CA, Small TN, Flomenberg N, et al.  Immune reconstitution following bone marrow transplantation (comparison of recipients of T-cell depleted marrow with recipients of conventional marrow grafts). Blood. 1989;73:1340–1350
  5. Koh MB, Prentice HG, Lowdell MW. Selective removal of alloreactive cells from haematopoietic stem cell grafts (graft engineering for GVHD prophylaxis). Bone Marrow Transplant. 1999;23:1071–1079
  6. Koh MB, Prentice HG, Corbo M, Morgan M, Cotter FE, Lowdell MW. Alloantigen-specific T-cell depletion in a major histocompatability complex fully mismatched murine model provides effective graft-versus-host disease prophylaxis in the presence of lymphoid engraftment. Br J Haematol. 2002;118:108–116
  7. Cavazzana-Calvo M, Fromont C, Le Deist F, et al.  Specific elimination of alloreactive T cells by an anti-interleukin-2 receptor B chain-specific immunotoxin. Transplantation. 1990;50:1–7
  8. Cavazzana-Calvo M, Stephan JL, Sarnacki S, et al.  Attenuation of graft-versus-host disease and graft rejection by ex vivo immunotoxin elimination of alloreactive T cells in an H-2 haplotype disparate mouse combination. Blood. 1994;83:288–298
  9. Mavroudis DA, Jiang YZ, Hensel N, et al.  Specific depletion of alloreactivity against haplotype mismatched related individuals by a recombinant immunotoxin (a new approach to graft-versus-host disease prophylaxis in haploidentical bone marrow transplantation). Bone Marrow Transplant. 1996;17:793–799
  10. Mavroudis DA, Dermime S, Molldrem J, et al.  Specific depletion of alloreactive T cells in HLA-identical siblings (a method for separating graft-versus-host and graft-versus-leukaemia reactions). Br J Haematol. 1998;101:565–570
  11. Amrolia PJ, Muccioli-Casadei G, Yvon E, et al.  Selective depletion of donor allo-reactive T-cells without loss of anti-viral or anti-leukemic responses. Blood. 2003;102:2292–2299
  12. Fehse B, Frerk O, Goldmann M, Bulduk M, Zander AR. Efficient depletion of alloreactive donor T lymphocytes based on expression of two activation-induced antigens (CD25 and CD69). Br J Haematol. 2000;109:644–651
  13. Hartwig UF, Robbers M, Wickenhauser C, Huber C. Murine acute graft-versus-host disease can be prevented by depletion of alloreactive T lymphocytes using activation-induced cell death. Blood. 2002;99:3041–3049
  14. Chen BJ, Cui X, Liu C, Chao NJ. Prevention of graft-versus-host disease while preserving graft-versus-leukemia effect after selective depletion of host-reactive T cells by photodynamic cell purging process. Blood. 2002;99:3083–3088
  15. Squifflet JP, Legendre C. The economic value of valacyclovir prophylaxis in transplantation. J Infect Dis. 2002;186(suppl 1):S116–S122
  16. Meyers JD, Flournoy N, Thomas ED. Risk factors for cytomegalovirus infection after human marrow transplantation. J Infect Dis. 1986;153:478–488
  17. Prentice HG, Gluckman E, Powles RL, et al.  Impact of long-term acyclovir on cytomegalovirus infection and survival after allogeneic bone marrow transplantation. European Acyclovir for CMV Prophylaxis Study Group. Lancet. 1994;343:749–753
  18. Ljungman P. Immune reconstitution and viral infections after stem cell transplantation. Bone Marrow Transplant. 1998;21(suppl 2):S72–S74
  19. Bacigalupo A. Second EBMT Workshop on reduced intensity allogeneic hemopoietic stem cell transplants (RI-HSCT). Bone Marrow Transplant. 2002;29:191–195
  20. Gillespie GM, Wills MR, Appay V, et al.  Functional heterogeneity and high frequencies of cytomegalovirus-specific CD8(+) T lymphocytes in healthy seropositive donors. J Virol. 2000;74:8140–8150
  21. Gamadia LE, Remmerswaal EB, Weel JF, Bemelman F, van Lier RA, Ten Berge IJ. Primary immune responses to human CMV (a critical role for IFN-gamma-producing CD4+ T cells in protection against CMV disease). Blood. 2003;101:2686–2692
  22. Zajac AJ, Blattman JN, Murali-Krishna K, et al.  Viral immune evasion due to persistence of activated T cells without effector function. J Exp Med. 1998;188:2205–2213
  23. Taams LS, Vukmanovic-Stejic M, Smith J, et al.  Antigen-specific T cell suppression by human CD4+CD25+ regulatory T cells. Eur J Immunol. 2002;32:1621–1630
  24. Taylor PA, Lees CJ, Blazar BR. The infusion of ex vivo activated and expanded CD4(+)CD25(+) immune regulatory cells inhibits graft-versus-host disease lethality. Blood. 2002;99:3493–3499
  25. Craston R, Koh M, McDermott A, Ray N, Prentice HG, Lowdell MW. Temporal dynamics of CD69 expression on lymphoid cells. J Immunol Methods. 1997;209:37–45
  26. Bishara A, Brautbar C, Nagler A, et al.  Prediction by a modified mixed leukocyte reaction assay of graft-versus-host disease and graft rejection after allogeneic bone marrow transplantation. Transplantation. 1994;57:1474–1479
  27. Currier JR, Kuta EG, Turk E, et al.  A panel of MHC class I restricted viral peptides for use as a quality control for vaccine trial ELISPOT assays. J Immunol Methods. 2002;260:157–172
  28. Longmate J, York J, La Rosa C, et al.  Population coverage by HLA class-I restricted cytotoxic T-lymphocyte epitopes. Immunogenetics. 2001;52:165–173
  29. Wills MR, Carmichael AJ, Mynard K, et al.  The human cytotoxic T-lymphocyte (CTL) response to cytomegalovirus is dominated by structural protein pp65 (frequency, specificity, and T-cell receptor usage of pp65-specific CTL). J Virol. 1996;70:7569–7579
  30. Tan LC, Gudgeon N, Annels NE, et al.  A re-evaluation of the frequency of CD8+ T cells specific for EBV in healthy virus carriers. J Immunol. 1999;162:1827–1835
  31. Hewitt CR, Feldmann M. Human T cell clones present antigen. J Immunol. 1989;143:762–769
  32. Sidhu S, Deacock S, Bal V, Batchelor JR, Lombardi G, Lechler RI. Human T cells cannot act as autonomous antigen-presenting cells, but induce tolerance in antigen-specific and alloreactive responder cells. J Exp Med. 1992;176:875–880
  33. Solomon SR, Tran T, Carter CS, et al.  Optimized clinical-scale culture conditions for ex vivo selective depletion of host-reactive donor lymphocytes (a strategy for GvHD prophylaxis in allogeneic PBSC transplantation). Cytotherapy. 2002;4:395–406
  34. Davies JK, Koh MB, Prentice HG, Lowdell MW. Potentiating the MLR prior to selective depletion of alloreactive cells. Biol Blood Marrow Transplant. 2002;9:67; (abstr.)
  35. Jiang YZ, Mavroudis DA, Dermime S, Molldrem J, Hensel NF, Barrett AJ. 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 Transplant. 1997;19:899–903
  36. Dolstra H, Fredrix H, Preijers F, et al.  Recognition of a B cell leukemia-associated minor histocompatibility antigen by CTL. J Immunol. 1997;158:560–565
  37. Dolstra H, Fredrix H, Maas F, et al.  A human minor histocompatibility antigen specific for B cell acute lymphoblastic leukemia. J Exp Med. 1999;189:301–308
  38. Lechner F, Cuero AL, Kantzanou M, Klenerman P. Studies of human antiviral CD8+ lymphocytes using class I peptide tetramers. Rev Med Virol. 2001;11:11–22
  39. Kern F, Surel IP, Faulhaber N, et al.  Target structures of the CD8(+)-T-cell response to human cytomegalovirus (the 72-kilodalton major immediate-early protein revisited). J Virol. 1999;73:8179–8184
  40. Khanna R, Burrows SR. Role of cytotoxic T lymphocytes in Epstein-Barr virus-associated diseases. Ann Rev Microbiol. 2000;54:19–48
  41. Alwan WH, Record FM, Openshaw PJ. CD4+ T cells clear virus but augment disease in mice infected with respiratory syncytial virus. Comparison with the effects of CD8+ T cells. Clin Exp Immunol. 1992;88:527–536
  42. Walls T, Shankar AG, Shingadia D. Adenovirus (an increasingly important pathogen in paediatric bone marrow transplant patients). Lancet Infect Dis. 2003;3:79–86
  43. Papadopoulos EB, Ladanyi M, Emanuel D, et al.  Infusions of donor leukocytes to treat Epstein-Barr virus-associated lymphoproliferative disorders after allogeneic bone marrow transplantation. N Engl J Med. 1994;330:1185–1191
  44. Kishi Y, Kami M, Oki Y, et al.  Donor lymphocyte infusion for treatment of life-threatening respiratory syncytial virus infection following bone marrow transplantation. Bone Marrow Transplant. 2000;26:573–576
  45. Hromas R, Cornetta K, Srour E, Blanke C, Broun ER. Donor leukocyte infusion as therapy of life-threatening adenoviral infections after T-cell-depleted bone marrow transplantation. Blood. 1994;84:1689–1690
  46. Wood KJ, Sakaguchi S. Regulatory T cells in transplantation tolerance. Nat Rev Immunol. 2003;3:199–210
  47. Papiernik M. Natural CD4+ CD25+ regulatory T cells. Their role in the control of superantigen responses. Immunol Rev. 2001;182:180–189
  48. Drobyski WR, Potluri J, Sauer D, Gottschall JL. Autoimmune hemolytic anemia following T cell-depleted allogeneic bone marrow transplantation. Bone Marrow Transplant. 1996;17:1093–1099

PII: S1083-8791(03)00522-6

doi:10.1016/j.bbmt.2003.12.001

Biology of Blood and Marrow Transplantation
Volume 10, Issue 4 , Pages 259-268, April 2004

Article Tools

* Image Usage & Resolution