|
|||||||||
|
|
|
|
|
|
|
|
|
|
|
Prepublished online as a Blood First Edition Paper on January 9, 2003; DOI 10.1182/blood-2002-09-2859.
IMMUNOBIOLOGY
From the Department of Microbiology and Immunology,
University of Miami School of Medicine, Miami, FL; Burnham Institute,
La Jolla, CA; Basic Research Program, SAIC-Frederick,
National Cancer Institute at Frederick, MD.
Engraftment failure following allogeneic bone marrow (BM)
transplantation is of clinical concern particularly involving
T-cell-depleted inoculum and transplantations for aplastic anemia.
Immune resistance by lymphoid and natural killer (NK)
populations with "barrier" function is well established. Major
histocompatibility complex (MHC)-identical marrow allografts
were examined to investigate effector pathways in non-NK-mediated
resistance. Barrier function was examined in cytotoxic normal and
deficient B6 (H-2b) recipients primed to donor minor
histocompatibility antigen (MiHA) prior to BM transplantation.
Host resistance was sensitively evaluated by colony-forming
unit (CFU) assays to directly assess for donor progenitor
cell (PC) and peripheral chimerism. B6 host CD8+ T
cells but not CD4+ or NK1.1+ cells effected
rejection of primitive (CFU-HPP [high-proliferative potential]) and lineage-committed (CFU-IL3/GM [interleukin
3/granulocyte macrophage]) allogeneic donor progenitors. To
address complementation by the cytotoxic pathways existing in singly
deficient (perforin or FasL) recipients, cytotoxically double (perforin
plus FasL) deficient (cdd) recipients were used. Resistance in B6-cdd
recipients was comparable to that of wild-type B6 recipients
and was also dependent on CD8+ T cells. A "triple"
cytotoxic deficient model, involving transplantation of
TNFR1 In the setting of bone marrow (BM) transplantation
between genetically disparate individuals, allogeneic immune responses pose complex obstacles to successful outcome. For example,
host-versus-graft (HVG) responses can effect resistance against the
allograft. Such immune resistance must be overcome for the successful
engraftment of donor progenitor cells (PCs).1-3 Host
resistance, or "barrier" activity, has necessitated use of
conditioning regimens to diminish recipient immune function and, more
recently, stimulated attempts to infuse high doses of marrow/PCs to
overcome host resistance.4-10 Higher incidences of graft
failure both (1) following BM transplantation with T-cell-depleted
(TCD) marrow and (2) in multiply transfused (eg, aplastic anemic)
individuals continue to engender clinical concern regarding marrow
(progenitor cell) graft resistance.1,2,11
Studies by Martin and colleagues12,13 demonstrated
the importance of cytotoxic function in donor T cells to overcome
resistance and establish successful long-term engraftment. Findings by
Murphy et al14 and others15-18 have
demonstrated that acute rejection in unsensitized mice can involve
natural killer (NK), CD8+ T cells and other populations.
Although significant advances have been made in understanding the
nature of host resistance against bone marrow allografts, particularly
in mouse models of transplantation, a crucial question concerns
specifically which molecules in T/NK cell populations mediate the
barrier response.
This laboratory has been investigating the involvement of cytotoxic
effector function in marrow graft rejection.19-21
Recipients singularly lacking the ability to mediate perforin or
FasL-dependent killing maintained the capacity to effect some level of
resistance against major histocompatibility complex (MHC)-mismatched
allografts and strong resistance against MHC-matched
allografts.20,21 Studies investigating granzyme
B-defective recipients also failed to detect diminished barrier
responses following MHC-mismatched allogeneic stem cell
transplantations.22 However, NK-dependent resistance can
exhibit deficits in some strains with perforin deficiency.23,24 Thus, the contributions of various host
effector pathways remain unclear and, importantly, may differ,
depending on the effector cell populations involved.
The present studies investigated host resistance in recipients
sensitized to donor antigens to specifically investigate effector pathways in non-NK (ie, T-cell-mediated) marrow graft resistance. To
exclude any potential involvement of the intact cytotoxic pathway functioning in single cytotoxic deficient mice, resistance was examined
in perforin and FasL cytotoxically double deficient (cdd) recipients.21,22 A triple cytotoxic deficient model was
used to examine if the absence of TNFR1 Mice
The gld point mutation of B6-cdd mice was identified by polymerase
chain reaction (PCR) with primers designed with the terminal nucleotide pairing to either wild-type FasL or to the gld
mutation.27 To prevent possible mispairing to the
templates, a mismatch was introduced at the second nucleotide from the
3'-terminal of the FasL-wt or FasL-gld primers (C6A) to destabilize the
3'-end binding. Each of these primers together with an antisense primer
(FasL-as) was used in PCR to identify the gld mutation of B6-cdd mice.
A 563-bp fragment was amplified from samples with wild-type FasL gene
or gld mutation when FasL-wt or FasL-gld primers were used, respectively. All random mice examined in the colony contained homozygous gld mutation.
The gld mutation was further confirmed by real-time PCR and melting
point analysis using LightCycler (Roche Diagnostics, Indianapolis, IN).
DNA samples were amplified with fluorescent-labeled primers (5'-TGA GGA
TCT GGT GCT AAT GG-3' and 5'-AAT ATT CCT GGT GCC CAT GA-3'). The
melting temperatures (Tm) of the PCR products were tested by binding to
a 6-carboxy-fluorescein-labeled anchor probe and a detection probe.
Fragments containing wild-type FasL gene sequence exhibited a Tm of
52°C versus Tm of fragments containing gld mutation of 58°C. DNA
samples from B6-cdd mice in the colony tested at random all contained
homozygous gld mutation.
B6-cdd mice develop spontaneous "cdd syndrome," resulting in early
death by 15 weeks.25 Therefore, all cytotoxic double deficient mice were given transplants between 49 and 56 days (7-8 weeks) of age. B6-FasLgld/gld mice do not
exhibit such early death; therefore, B6-cdd phenotype is also monitored
by life span of genotyped mice not used as bone marrow transplant (BMT)
donors. Colony life span is 97.1 ± 9.8 days, consistent with
published results.25
TNFR1- and TNFR2-deficient mice on C3H and BALB/c backgrounds were
generated from breeding pairs
B6.129-Tnfrsflatm1Mak and
B6.129S2-Tnfrsflbtm1Mwm (Jackson Laboratories).
R1+/ Priming against minor histocompatibility antigen (MiHA) or MHC
alloantigens
Bone marrow transplantation Femurs and tibias were harvested from appropriate female donors as previously described.21 T cells were removed by incubation with anti-Thy1.2 monoclonal antibody (mAb; HO13.4 ascites diluted 1:200) followed by 10% vol/vol Low-Tox M rabbit complement (Accurate Chemical & Scientific, Westbury, NY) at 37°C for 45 minutes. T cells present were reduced from approximately 2.0% to below the level (flow cytometry) of detection (< 0.2%). Recipient mice were exposed to 9.0 Gy lethal total-body irradiation (TBI) from an open beam 60Co gamma (50 cGy/minute). Twenty-four hours later, 2.0 × 106 TCD bone marrow cells (BMCs) were infused intravenously, and mice were maintained on acidified/antibiotic water (pH < 2.2, 100 mg/L neomycin sulfate, 10 mg/L polymyxin B sulfate).Adoptive transfer of allograft resistance to unprimed recipients B6 mice were primed with BALB.B cells twice prior to BM transplantation. Spleen and LN cells from these primed mice were then fractionated to deplete lymphocyte subsets as previously described.21 NK cells were removed by in vivo depletion of B6 donors using antibody against NK1.1 (PK136 ascites, 100 µL) injected intraperitoneally 48 hours prior to removing spleen and LNs. Aliquots of cells from these donors were treated with anti-NK1.1 mAb plus C' in vitro prior to transfer. NK cell-depleted populations contained less than 1.0% (fluorescein isothiocyanate [FITC]-conjugated antihamster immunoglobulin) staining cells. To delete T-cell subsets, spleen and LN cells from primed or naive animals were treated with anti-CD8 or anti-CD4 mAbs and C' (1-2 times). Alternatively, lymphocytes from primed B6 mice were used to produce CD8-enriched and CD8-depleted fractions by immunomagnetically labeling with anti-CD8 MACS beads followed by column separation (Miltenyi Biotec, Bergisch Gladbach, Germany). Cells were than analyzed to assess enrichment/depletion ( 87%/ 2.1%). Desired numbers of
these cells or unfractionated populations from primed or unprimed
(control) B6 mice were infused intravenously into unprimed B6 mice 48 hours before BM transplantation. One day later, these mice were
administered 9.0 Gy TBI and received 2 × 106 BMT-TCD
BALB.B marrow 24 hours later.
In vitro CFU-GM and CFU-IL3 assays Recipients were killed 5 days after BM transplantation. Pooled spleen cells (1 × 105) were cultured in 1 mL of a mixture containing -modification of Eagle medium with
nucleosides (-MEM), 0.86% methyl cellulose (Methocult; StemCell
Technologies, Vancouver, BC), 30% fetal calf serum (FCS), 250 µM
2-mercaptoethanol, 2 mM L-glutamine, 100 U/mL penicillin,
100 µg/mL streptomycin, and 50 U/mL recombinant murine granulocyte-macrophage colony stimulating factor (GM-CSF) or
interleukin 3 (IL-3) (R&D Systems, Minneapolis, MN).28
Triplicate cultures were maintained at 37°C, 5% CO2 for
4 days. Cell aggregates containing 25 cells or more were scored as
individual colonies (colony-forming unit [CFU]) on day 4. Results are
presented as average CFU-GM or -IL3 ± SD/group as previously
described.21
For CFU-HPP (high-proliferative potential) assay, BM cells were
harvested from femurs and tibias of recipient animals 12 days after BM
transplantation. Cells were pooled and washed, and
1 × 105 cells were seeded in 1.5 mL 0.3% agarose (low
melting point agarose; GIBCO BRL) containing semisolid culture
media in Iscoves modified Dulbecco medium (IMDM) supplemented
with 20% fetal calf serum, 100 ng/mL stem cell factor (PEPROTECH,
Rocky Hill, NJ), 50 U/mL IL-3, and 2700 U/mL IL-1 Detection of genetic markers for primitive progenitors HPP colonies were isolated at day 6 of culture. Cells from the pooled colonies were lysed, and total RNA was prepared using RNeasy kit (Qiagen). Reverse transcription (RT)-PCR was performed using the total HPP RNA to detect expression of progenitor markers. PCR primers used were (1) stem cell antigen-1 (Sca-1), 5'CAATGTAGCAGTTCCCAATG3' and 5' CAGGGGCTATAAAGGCAAAA3'; (2) IL-3R,
5' GAACAGATTCCACCATGGCCTCCTTG 3' and 5'GTCTCCACTACGGACACTTCTGTC3'; and
(3) glyceraldehyde phosphate dehydrogenase (GAPDH), 5'
ATGACCACAGTCCATGCCAT 3' and 5' GCCTGCTTCACCACCTTCTT 3'.
Chimerism To assess peripheral blood and other tissues for chimerism following BM transplantation of MHC (H-2b)-matched C3H.SW donor inoculum into B6 recipients, Ly9.1 mAb staining was performed. Ly9.1 is expressed on most hematopoietic cells of C3H (Ly9.1+) but not BL/6 (Ly9.1 ) background
mice. Results, therefore, represent a conservative estimate of donor
cell chimerism in the compartments examined.
BM transplantation in presence of anti-TRAIL antibody Monoclonal anti-TRAIL antibody N2B229 or an isotype-matched control rat immunoglobulin G2a (IgG2a) antibody was injected intraperitoneally on days 0 and 1 (200 µg/injection) after BM transplantation.Statistics For each transplantation experiment, BM cells (BMCs) were harvested from a minimum of 2 donor mice, pooled, and transplanted into a minimum of 2 recipient mice for every experimental group. CFU values are presented as an average of triplicate (duplicate for HPP-CFU cultures) wells with SD. Statistical significance for the difference between each experimental group was examined by 2-tailed t test and reported as P values where appropriate as previously reported.20 In all experiments presented, the difference between the CFU values obtained from primed mice compared with the numbers obtained from naive mice of the same strain were highly significant (P < .05) and, therefore, were not included in the tables or figures. Each experiment was performed independently a minimum of 2 times; most were performed more than 3 times.
High-proliferative potential colony forming cells (HPP-CFCs) are targets of MiHA-disparate host resistance Previously, we examined host resistance against hematopoietic PCs in 2 MHC (H-2b)-matched MiHA-mismatched transplant models: C3H.SW or BALB.B marrow B6 recipients.21 These studies
demonstrated that resistance was dependent on priming (Figure
1) against the donor MiHA prior to
transplantation and a host CD3+ population.21
However, CFU-IL3 and CFU-GM are lineage-committed progenitors and do
not represent pluripotent hematopoietic stem cell (HSC) populations
responsible for establishing longer term chimerism and renewing
multilineage cell populations.
HPP-CFCs are pluripotent cells that share many features with
HSCs.30 To determine whether these pluripotent progenitor
populations were also subject to rejection, host resistance was
examined in the MiHA-disparate BALB.B
Marrow allograft resistance is mediated by host
CD8+NK1.1 cells remaining in
the primed spleen plus LN cell population that was transferred into
naive B6 recipients. Following BM transplantation there was no
diminution in the resistance transferred by this CD4-depleted versus
C'-only treated population (Figure 3D). CD4/-deficient
mice failed to exhibit resistance against BALB.B BM transplantation
following the standard donor MiHA-priming protocol (Figure
4). However, resistance in these
recipients could be reconstituted following transfer of a syngeneic
CD4+ population prior to donor MiHA priming (Figure 4).
Together, these data demonstrate that
CD8+CD4NK1.1 populations from
primed mice function as barrier cells in MHC-matched marrow allograft
resistance and require a host CD4+ T-cell response for
generation of this effector population.
Perforin/FasL B6-cdd mice resist MiHA-mismatched marrow allografts Granzyme B-defective recipients effectively rejected C3H.SW progenitors following priming against the donor MiHA prior to transplantation (data not shown). This observation is consistent with and extends the previous finding that perforin-deficient recipients are also capable of efficient resistance against MiHA-mismatched allografts.21 A mixed-marrow inoculum containing allogeneic (BALB.B) and syngeneic, congenic (B6-Ly5.1) marrow transplanted into granzyme B/ recipients, paralleled
previous findings using cytotoxically healthy recipients21
and indicated that in the absence of granule-dependent killing (see
"Marrow allograft resistance in cytotoxic double deficient mice is
CD8+ T-cell dependent"), the efficiency as well as
specificity of the resistance was not altered (data not shown).
Because either the granule or Fas-mediated pathways can induce rapid
and potent apoptotic signaling, a cytotoxic deficiency in one effector
pathway could be compensated for by the concurrently functioning
alternative pathway. If true, simultaneous impairment of both would
result in the loss of barrier function. To address this possibility,
B6-cdd (perforin and FasL = cdd) mice were used as recipients. All
B6-cdd mice used in these studies were screened for homozygous perforin
deficiency by PCR prior to transplantation. Because
FasLgld/gldB6Pfp+/
Marrow allograft resistance in cytotoxic double-deficient mice is CD8+ T-cell dependent To determine the cell population required for resistance in B6-cdd recipients, adoptive transfer experiments were performed using cells from B6-cdd donors primed against BALB.B MiHA (Figure 5). Unfractionated spleen cells were prepared from B6-wt and B6-cdd mice primed against BALB.B antigens. Aliquots containing 1 × 107 CD8+ cells were transferred into syngeneic unprimed B6 recipients prior to irradiation and BM transplantation using the model (Figure 1) described earlier. Recipients receiving either B6-wt or B6-cdd populations demonstrated effective allograft resistance compared with control (ie, untransferred) naive mice (P < .05). The same total number of B6-cddBALB.B cells transferred following depletion of CD8+ T cells failed to demonstrate resistance after BALB.B marrow transplantation (Figure 5). In contrast, transplantation of CD8+ T-cell-enriched fractions (containing 7.1 × 106 CD8+ T cells, ie 71% versus the number contained in unfractionated populations) again demonstrated efficient resistance. These findings demonstrated that CD8+ T cells generated after MiHA priming in B6-cdd mice can efficiently transfer resistance against donor PCs.
Resistance against TNFR1 / or
TNFR2/ donors was transplanted into healthy B6
recipients previously primed to donor C3H.SW MiHA (Figure
6). B6-wt recipient mice primed by
administration of healthy C3H.SW cells exhibited efficient resistance
to marrow allografts containing PCs from healthy C3H.SW donors or
donors lacking TNFR1 (Figure 6A). These findings were verified when
B6-wt recipients were primed prior to BM transplantation using cells
from R1/ (6B) or R2/ (6C) mice and
found to strongly resist transplantations containing progenitors from
these TNFR1/ and R2/ marrow donors,
respectively.
Next, the ability of B6-cdd recipients to resist marrow allografts
containing PCs lacking the ability to signal via TNFR1 or R2 was
examined. TNFR1
To corroborate and extend these observations, resistance was examined
in a second T-cell-dependent model between MHC-mismatched donors and
recipients (Figure 9). Previous studies
have demonstrated that, although NK cells play a crucial role in the
resistance against MHC-mismatched marrow grafts in naive recipients,
the resistance is virtually completely T-cell dependent if the
recipient is sensitized against donor antigens prior to BM
transplantation.15 B6-wt and B6-cdd recipients
(H-2b) were, therefore, primed to complete MHC class I and
II mismatched donor MHC H-2d alloantigens prior to
transplantation. As previously detected in MiHA-mismatched transplants,
cytotoxically normal B6-wt mice effectively and equivalently rejected
TNFR1
Resistance occurs in the simultaneous absence of TRAIL, perforin, FasL, and TNFR1 signaling The triple cytotoxic defective resistance model using TNFR1/ marrow donors was extended to examine for TRAIL
involvement. Studies using the anti-TRAIL mAb N2B2 demonstrated its in
vivo ability to interfere with tumor clearance.31,32 TRAIL
transfectants, but not control transfectant cells, were shown to
effectively lyse L929 target cells as previously reported, and the mAb
N2B2 effectively inhibited this TRAIL-mediated cytotoxic
killing31 (M.M. and R.B.L., unpublished observations,
April 2002). We found that this anti-TRAIL mAb (N2B2) markedly
augmented metastatic carcinoma lesions over a 2-week time period in
vivo.33 B6-cdd recipients primed to C3H.SW MiHA were given
transplants of TNFR1/ marrow. This same anti-TRAIL mAb
administered in the tumor experiments was used to examine rejection in
the triple deficient model. Groups received 2 injections of anti-TRAIL
antibody or the isotype control immunoglobulin on days 0 and +1. Day 5 CFU assay indicated that efficient resistance occurred in the
recipients receiving anti-TRAIL antibody (Figure
10A). A second experiment was performed
using the MHC-mismatched model. B6-cdd recipients injected with
anti-TRAIL antibody efficiently rejected 1 × 107
allogeneic H-2d-R1/ marrow (Figure 10B). We
have also found (T.J.S. et al, unpublished observations, October
2002) that treatment of BMCs with recombinant TRAIL (1000 ng/mL) overnight in vitro prior to transfer to irradiated syngeneic
recipients does not affect immune reconstitution, strongly suggesting
that marrow precursors are not particularly sensitive to TRAIL-mediated
apoptosis. In total, these results demonstrated that in the
simultaneous absence of 4 cytotoxic effector pathways used by cytologic
T lymphocytes (CTLs), strong host resistance could be
detected.
Significant advances have been made in understanding the nature of barrier responses against bone marrow allografts, particularly in mouse transplant models. The acute rejection of allogeneic marrow can occur within several days of transplantation in irradiated recipients.34 Elegant studies by Murphy et al14 and others15-18 have demonstrated that acute rejection in unsensitized mice can involve NK, CD8+ T cells and other populations. Previous findings by several laboratories have showed that CD4+ as well as CD8+ T cells can participate in rejection of MHC and MiHA marrow allografts.35,36 CTLs have been identified in patients rejecting marrow grafts,37 and, thus, it has generally been considered that cytotoxic function by host T cells is a crucial effector pathway in marrow allograft resistance. Experimental studies have isolated and cloned antidonor alloreactive T cells from primates and mice rejecting hematopoietic grafts38,39 and used them to transfer resistance to nonresistant recipients.39 Studies by Kernan et al37 reported that the rejection of HLA nonidentical marrow grafts was associated with the presence of host antidonor-specific CTLs. Nonetheless, demonstration that transferred CTLs can mediate resistance does not conclusively constitute mea culpa of cytotoxic function in the transplantation setting. A crucial question concerns which molecules in CTL (and non-CTL) populations can mediate resistance function(s). To carefully analyze the potential effector pathways used by T
cells during the rejection of progenitor cell allografts, models of
resistance involving recipients sensitized to donor antigens have been
analyzed.21 The use of MHC-matched, MiHA-mismatched strains together with a priming event results in a T-cell-dependent barrier, found here to be transferrable by
CD8+NK1.1 Studies from several laboratories have failed to detect
diminished host resistance in mice with severe cytotoxic deficiencies. In BMT models involving varying donor/recipient genetic disparities, perforin- or FasL-defective (MHC class I/II + MiHA; MiHA only), granzyme B KO (P6F1), and granzyme A promoter diphtheria toxin transgenic animals (MHC class I/II + MiHA; and P6F1) have failed to show defects in marrow graft resistance,
respectively.19-22,40 Studies using | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
1) and BM transplantation (day 0). Recipients
received varying doses of donor or syngeneic BM-TCD and
resistance was analyzed by CFU activity in recipient spleens (day 5)
and donor (ie, C3H.SW) chimerism by staining with Ly9.1 mAb.
) or an isotype-matched control antibody (
) was injected
(200 µg) intraperitoneally on day 0 just prior to BM transplantation
and again 24 hours later. Recipients were examined by CFU-IL3 assay.
(A) Naive B6-cdd, 3682 ± 294; primed B6-cdd injected with
isotype control, 80 ± 25; primed B6-cdd injected with N2B2,
7 ± 12; naive B6-gld, perf+/
2µ KO class
I-deficient transplanted marrow hav