Biology of Blood and Marrow Transplantation
Volume 11, Issue 12 , Pages 957-971, December 2005

Cytolytic Pathways Used by Effector Cells Derived from Recipient Naive and Memory T Cells and Natural Killer Cells in Resistance to Allogeneic Hematopoietic Cell Transplantation

Department of Microbiology and Immunology, University of Miami Miller School of Medicine, Miami, Florida

Received 25 April 2005; accepted 11 July 2005.

Article Outline

Key Words:  Cytotoxic pathways , Resistance to hematopoietic transplants , Effector NK , T naive and T memory cells , Conditions of transplant

 

Back to Article Outline

Introduction 

During the past decade, 2 major cytolytic pathways used by natural killer (NK) and T cells have been thoroughly described. Both the perforin/granzyme and death receptor (ie, fas/fas ligand [fasL]) pathways ultimately result in apoptosis of the targeted population [1]. Studies have been preformed using mutant and knockout strains to investigate NK and T-cell impairment in viral and tumor immune responses, as well as in transplantation models, to investigate the contribution of these pathways in resistance to allogeneic cell engraftment [2, 3, 4, 5, 6, 7, 8, 9]. Several early studies yielded interesting results with regard to cytolytic deficiencies and so-called allograft resistance. For example, major histocompatibility complex (MHC)–heterozygous recipients were observed to reject donor parental marrow despite granzyme B deficiency [4]. Additionally, lethally conditioned MHC-disparate recipients deficient in perforin or expressing defective fasL were also able to reject allogeneic hematopoietic grafts after a standard dose of bone marrow (BM) was transplanted [2].

By using a different approach in which diphtheria toxin was transgenically expressed under a granzyme B promoter to broadly eliminate cytolytic barrier cell populations, results were obtained that were similar to those described previously: no detection was reported of diminution in NK resistance to hematopoietic stem cell (HSC) grafts [10]. In contrast, adoptively transferred peripheral blood mononuclear cells from syngeneic donors treated with isoleucyl methyl ester (which lyses cytotoxic granule-containing cells [11]) failed to reject most dog leukocyte antigen–mismatched BM grafts [12]. Such studies were followed by investigations that examined how the simultaneous deficit of cytolytic molecules affected resistance mediated by T and NK cells [6, 7, 8, 9, 13]. Although initially perhaps surprising to many investigators, the results of these studies have consistently demonstrated that resistance in most models remains relatively intact despite the absence of perforin or fasL and, in some transplant conditions, efficient when multiple cytolytic pathways are lacking [2, 4, 5, 6, 9, 10] (see below). Such findings—that the host can apparently summon a variety of potent effector pathways to thwart successful engraftment of progenitor cell allografts and hoped-for tolerance after transplantation—are not only biologically intriguing, but also important from the clinical perspective. This review discusses findings obtained from several transplantation models investigating how resistance to stem/progenitor cells is mediated, particularly with respect to the involvement of cytolytic pathways used by NK cells and effector cells derived from naive and memory T-cell populations in recipients conditioned for allogeneic HSC grafts. On the basis of published findings to date, we hypothesized that transplantation conditions determine the relative importance of host cytotoxic pathways in resisting allogeneic hematopoietic cell engraftment.

Back to Article Outline

The “barrier” 

A major obstacle toward more universally successful allogeneic hematopoietic stem/progenitor cell transplantation (HSPCT) is circumvention of the immune-mediated resistance present in the recipient [14]. This so-called barrier is composed of multiple cell types, including NK and T-cell populations (Table 1) [15, 16, 17, 18, 19, 20]. Moreover, depending on the recipient’s state of antigen exposure, effector cells may be generated from a host T-cell compartment composed of both naive T-cell and memory T-cell populations. One report noted anti-HLA class I antibody in a patient who rejected an allogeneic BM graft, and a recent study observed that preformed antibody against MHC-disparate grafts can contribute to rejection, presumably as a result of anti-MHC antibody binding to cell-surface donor antigens [21] (P. Taylor and B. Blazar, University of Minnesota, unpublished data). Although antibodies against minor histocompatibility antigen (MiHA) peptides have been identified in HSPCT recipients [22, 23], there have been no reports of antibody-mediated resistance against MHC-matched hematopoietic grafts. Notably, increased levels of bcl-2 in CD8+ memory T cells selectively favor their survival versus naive T cells after total body irradiation [24]. Overall, despite considerable insult to the host immune system resulting from chemotherapy, radiation treatment, or both (ie, antitumor therapy/conditioning before HSPCT), all of these cell populations can persist and mediate immune responses, thus preventing successful long-term hematopoietic engraftment by transplanted donor stem/progenitor cells [25, 26]. The continued development and implementation of reduced-intensity conditioning protocols to diminish regimen-related toxicity and complications of immune deficiency (ie, infection) underscores the need to inhibit/regulate host resistance [27, 28]. Moreover, recently it has been found that clinically, the loss of donor chimerism in these nonmyeloablative settings can be associated with higher rates of relapse [27].

Table 1. Cell Populations That Can Effect Resistance in Experimental Models of Hematopoietic/Stem and Progenitor Cell Transplantation
Variable NK⁎Naive T Cell†Memory T CellReferences
Naive host (unprimed to donor alloantigen)
MHC mismatched
Hostα Donor MHC recognition
A → A/BNoneYes‡NoNo18, 33
A → BFullYesYesNo17, 19, 20
A → BClass I onlyYesYesNo34
A → BClass II onlyNoYesNo34, 35
MHC matched
A.2 → A.1NoneNoYesNo105§
Sensitized host (primed to donor alloantigen)
MHC mismatched
A → BFullYesYesYes9, 26, 72∥
MHC matched
A.2 → A.1NoneNoYesYes6, 9, 13, 73

⁎ This population is responsible for early (days 1-5) resistance.

† This population contributes after 5 to 6 days after transplantation.

‡ Italic signifies the effector population predominantly responsible for resistance under the conditions of transplantation listed.

§ Z.Z., A.S., and R.B.L., unpublished data.

∥ Anti-donor MHC antibody may contribute to resistance in MHC class I/II–mismatched recipients sensitized to donor antigens (P. Taylor and B. Blazar, unpublished data).

Experimental models of allograft resistance, principally in the mouse, have led to a greater understanding of the cellular players that mediate resistance in different transplantation settings (Table 1). However, the cells within the heterogeneous hematopoietic stem/progenitor cell (HSPC) population that are targeted for rejection and, importantly, the ultimate fate for such cells continue to be unclear. Regarding the former, resistance activity against short-term repopulating and committed progenitor cells may or may not be identical to activity against long-term HSCs, in part as a consequence of potentially different repopulation locations and kinetics of these progenitors [29, 30, 31, 32]. Concerning the fate of targeted cells, donor HSPC target populations recognized by host NK and/or T-cell effector cells could undergo apoptosis. Alternatively, the resistance phenotype would also be manifest if HSPCs were functionally inhibited but persistent in the recipient, unable to proliferate or undergo differentiation (Figure 1). Moreover, the molecular mechanisms by which the host immune system effects rejection of donor progenitors remains to be fully elucidated. It is interesting to note that the transplantation setting itself must be considered because resistance in naive versus antigen-sensitized recipients or after nonablative versus ablative conditioning could understandably lead to different effector pathways (Table 1; Figure 1). Recent findings from our own laboratory and others support the notion that effector mechanisms used by NK and primed T-cell populations may be nonoverlapping or possibly distinct (see below) [6, 8, 9, 13]. Therefore, identifying the precise molecular pathways responsible for progenitor allograft resistance remains a primary challenge for immunologists as a prelude to developing approaches that selectively diminish a patient’s antigraft response while preserving his or her overall immune function, including antitumor responses early after transplantation.

  • View full-size image.
  • Figure 1. 

    Summary of parameters that affect the outcome of engraftment after allogeneic hematopoietic cell transplantation. Factors such as the nature of pretransplantation conditioning, prior sensitization to donor alloantigen, and the genetic disparity between the donor and the recipient can affect the cellular constituency of the host immune barrier to alloengraftment. This barrier could be mediated by the induction of apoptosis or other effector pathways. Ab indicates antibody; PC, progenitor cell.

Back to Article Outline

Effector cells and transplantation conditions 

Several cellular constituents that contribute to the allograft resistance barrier have been well characterized in experimental models in the mouse (Table 1). The acute rejection of allogeneic marrow can occur within several days of transplantation in irradiated recipients [33]. Several elegant studies [15, 16, 17, 18, 19, 20] demonstrated that acute rejection in unsensitized mice can involve NK cells, NKT, CD8+ T cells, and other populations. Findings by Vallera and colleagues [34] showed that CD4+ and CD8+ T cells can reject T cell–depleted marrow. These studies demonstrated that T-cell subsets can act independently of each other in this process because marrow from C57BL/6 congenic mice differing at a single MHC class I (bm1) or class II (bm12) locus was capable of being rejected by host CD4+ or CD8+ cells, respectively. Sprent and colleagues [35] further demonstrated that transfer of CD4+ T cells into class II–disparate recipients resulted in marrow destruction, and the authors proposed that this hematopoietic attack was dependent on class II recognition of progenitor cells because an MHC class II−/− inoculum was not eliminated. We have recently observed that rejection of T cell–depleted BM transplanted from MHC-matched/MiHA-mismatched donors occurs in nonablatively conditioned CD8−/− recipients; this supports the notion that the host CD4+ subset can contribute to the rejection of marrow cells (Z.Z. and R.B.L., unpublished data). We have also examined resistance in MiHA-disparate recipients primed to donor alloantigens [6, 9]. Anti-CD8 antibody treatment of such hosts abolishes resistance, thus indicating that a population of CD8+CD4− T-memory (CD44+Ly6Chi; A.S. and R.B.L., unpublished data) cells was responsible for resistance in these recipients [13] (Table 1).

An important characteristic of the resistance process against progenitor cell allografts is that there is a kinetic difference between the actions by host NK and T-cell populations. Three approaches for assessing the presence of donor transplanted cells are generally used in resistance studies (Table 2). These methods have different strengths and limitations but together have enabled insights into the kinetics and other aspects of the rejection process. In murine models using iodeoxyuridine uptake, NK cells were shown to effect rejection early (within 48 hours) after transplantation [33, 36]. Recent studies in our laboratory directly assessing transplanted donor progenitor populations as an indicator of resistance have identified functional transplanted progenitor cells in host compartments as early as 2 hours after transplantation [37]. Donor progenitor cells were virtually undetectable at 24 to 48 hours after transplantation in lethally conditioned MHC-mismatched allogeneic recipients unsensitized to donor antigens, whereas significant donor colony-forming unit/interleukin (IL)–3 and colony-forming unit/high proliferative potential populations were clearly present at these times in NK-depleted (anti-NK1.1 monoclonal antibody [mAb]) mice (A.S. and R.B.L, unpublished data). In contrast to NK effects, the kinetics of T cell–mediated resistance are clearly distinct. In recipients not sensitized to donor antigens, resistance cannot be demonstrated for approximately 5 to 7 days after transplantation in MHC-mismatched models after transplantation of overriding BM cell doses (too many cells for the NK compartment to eliminate) into MHC-mismatched recipients [19, 38]. Our recent observations regarding resistance in unsensitized MHC-matched/MiHA-mismatched recipients—in which NK cells are not contributory—reflect similar kinetics, because diminishing levels of peripheral chimerism could not be detected before 5 to 7 days after transplantation (Z.Z. and R.B.L., unpublished data). In contrast, resistance in MHC-matched mice sensitized to donor allogeneic antigens can be detected within 2 days [13].

Table 2. Methods for Analyzing the Presence of Transplanted Cell Populations after Progenitor Cell Allografts
MethodAdvantagesLimitations
Spleen cell proliferation assayEarly assessmentCannot discriminate between different cell lineages and does not enable long-term assessment
Chimerism in central/peripheral compartmentCan determine multiple hematopoietic lineages in multiple compartments; can assess long termCannot enable early assessment
Progenitor cell assayVery early assessment; can determine specific lineagesCannot assess mature cell populations

The identification of NK and T cells in the resistance process led to questions regarding how these populations mediate barrier activity. Are cytolytic pathways induced by these cells required for effective resistance (Figure 2)? As noted previously, transgenic mice defective in NK- and T cell–mediated cytotoxic responses continued to resist allogeneic HSCs [10]. Our own studies have demonstrated strong resistance to BM cell allografts in mice deficient in perforin- and fasL-dependent killing if recipients possess T cells previously primed to donor alloantigens [6, 7], whereas others found resistance in unsensitized recipients intact in the singular absence of perforin, fasL, granzyme B, and other cytotoxic pathways (see below) [2, 4, 6, 7, 8, 9, 39]. These findings raise the question: how critical is cytotoxicity effected by NK and T-cell populations in resistance to allogeneic HSPC?

  • View full-size image.
  • Figure 2. 

    Lytic pathways that could contribute to resistance against allogeneic hematopoietic stem/progenitor cells. The primary cell-mediated lytic pathways (ie, perforin/granzyme and fasL-fas), as well as recently discovered death ligand/receptor pairs that could be expressed by effector and progenitor cells.

Back to Article Outline

NK cell–mediated resistance to allogeneic hematopoietic progenitor cell grafts 

Contrary to the classic rules of T cell–mediated rejection, Cudkowicz and Bennett [40] were the first to demonstrate that F(1) mice were capable of rejecting marrow derived from a parental donor strain despite the absence of an MHC alloantigen disparity. The cells primarily responsible for this so-called hybrid, as well as allogeneic MHC resistance, were later identified as NK cells [15, 18, 41] and thoroughly reviewed [42]. NK cell alloreactivity occurs in part as the result of the absence (“missing self”) of signaling by transmembrane NK-inhibitory receptors, which require engagement of self-MHC, thereby permitting NK-activating receptors to signal induction of effector pathways [43, 44]. Subsequent to these seminal studies, transplantations involving MHC-mismatched recipients never exposed to donor alloantigen demonstrated that the initial participants in this acute resistance process are also NK cells [19].

Despite the importance of NK cells in marrow allograft resistance, the target population(s) of the NK cell within the HSPC inoculum is currently unknown. Some experiments suggest that the HSC itself may not be susceptible to lysis by the NK cell. For example, purified HSCs cocultured before transplantation with recipient NK cells from MHC-disparate animals showed no differences in their capacity to rescue lethally irradiated syngeneic recipients [31]. It is possible that the signals necessary to make the HSCs susceptible to NK-mediated attack are absent in vitro; alternatively, we hypothesize that more committed progenitor cell populations may be the targets of NK-mediated lysis in vivo [32]. HSCs could be capable of avoiding recognition by NK cell populations, for example, by expressing a paucity of activation-inducing ligands [45]. It is interesting to note that stem cell populations in general may not be readily lysed by NK cells. Despite their inability to induce inhibitory Ly49-mediated signaling because of low or no MHC class I expression [46, 47], studies in our laboratory have recently shown that neural stem cells were also not lysed by syngeneic or allogeneic NK cells, regardless of the absence or presence (after cytokine induction) of self-MHC class I molecules [48]. Perhaps the antigens required for NK recognition by activating receptors are not present on these progenitor populations. Recent studies reported that other stem cell populations, ie, embryonic stem cells, also do not seem susceptible to NK effector cell–mediated lysis [49]. It is interesting to consider that stem cells may inherently posses the ability to evade or diminish host immune responses. Several studies have reported the ability of stem cell populations, including mesenchymal stem cells and HSCs, to inhibit or downregulate immune responses in vitro and in vivo [50, 51]. Notably, mesenchymal stem cell populations have recently been reported to posses the capacity to inhibit both naive and memory antigen–specific T cells [51].

Although NK cell involvement in acute graft rejection is well established in mice, evidence for NK-mediated rejection in clinical transplant recipients is difficult to establish. Clinical findings have been reported that indicate that certain HLA-C determinants correlate with an increased risk of graft rejection. HLA-C is most often recognized by killer inhibitory receptors (KIRs) on human NK cells [52]; therefore, “missing” HLA-C determinants on donor cells would likely result in a strong host anti-donor NK reactivity. Currently, there is no adequate method to conclusively discriminate NK from T-cell involvement in clinical graft failure. In the context of progenitor cell transplantations, alloreactive human NK populations have been recently proposed to mediate antilymphohematopoietic tumor effects after HSPCT [53]. This interpretation was based on clinical findings in which greater graft-versus-leukemia activity after allogeneic bone marrow transplantation (BMT) reportedly was associated with donor/recipient pairs in instances in which the recipient lacked ligands for KIR on donor cells. In contrast, a recent study by another group failed to observe such a correlation between graft-versus-leukemia and KIR mismatch in a small series of nonmyeloablative clinical transplantations [54]. Regardless, although it seems that NK populations could also be important in mediating allogeneic graft versus host under appropriate genetic conditions, clinical findings to support this pathway remain elusive.

Back to Article Outline

The involvement of cytotoxic pathways in NK-mediated resistance to progenitor cell grafts 

Several studies in murine models have investigated potential mediators of NK-dependent marrow allograft rejection [4, 7, 8, 10]. Such responses could be mediated via cytotoxic pathways, the elaboration of cytokines that affect hematopoiesis, or some presently unknown mechanism. To effect target cell lysis, NK cells have been shown to depend on either the perforin/granzyme exocytosis pathway or the tumor necrosis factor (TNF) family of death-inducing ligands [55, 56, 57, 58, 59, 60]. The most well characterized death ligand is fasL (CD95L), which is expressed by both NK and activated T cells [1, 56, 58]. Upon ligation with their concomitant ligand, TNF receptor (TNFR) family death receptors can effect several cellular programs, including apoptosis—a process dependent on recruitment and activation of procaspase 8 (see review [61]). Alternatively, or in addition to cytolytic molecules, activated NK cells can elaborate multiple cytokines, including interferon-γ, transforming growth factor-β, IL-12, and TNF-α. With respect to progenitor cell engraftment, each of the cytokines noted has been shown to affect progenitor cell function and hematopoiesis [62, 63, 64, 65, 66, 67, 68]. Because cytokines such as transforming growth factor-β can mediate both stimulatory and inhibitory activity on different progenitor populations and because individual cytokine effects can be altered in the presence of other cytokines, it is difficult to understand how an individual cytokine could be globally responsible for allograft rejection.

Several studies have examined the involvement of the 2 major cytotoxic effector pathways used by NK cells, ie, perforin/granzyme and fasL, in HSPC rejection [2, 4, 5, 7, 8]. Initial studies from this laboratory using B6 (H2b) perforin−/− mice that underwent transplantation with MHC-disparate marrow from BALB/c (H2d) donors showed, to our surprise, that rapid, ie, NK-dependent, resistance occurred in these recipients [2]. These findings clearly indicated that under the conditions of those transplantations (lethal total body irradiation and complete MHC-mismatched BM), perforin-deficient NK cells seemed to be fully capable of resisting a standard dose (2 × 106) of transplanted allogeneic BM. Granzyme B, although required for rapid perforin-dependent lysis of target cells in vitro, also was not required for HSPC rejection in vivo because granzyme B−/− recipients were capable of rejecting BM grafts as efficiently as wild-type mice in a hybrid resistance model [4]. Additional studies in perforin-deficient mice found that although perforin was not required for NK-mediated rejection, the NK cell compartment could be overridden by transplantation of fewer HSPCs in perforin-deficient mice compared with perforin-competent recipients [8]. Such observations support a role for perforin/granzymes for optimal NK-dependent resistance. Genetic differences between mouse strains and environmental conditions in animal housing facilities were also found to influence the perforin dependence of NK-mediated HSPC rejection. B6.129 perforin−/− mice exhibited a deficiency in NK-mediated rejection only after transfer from a specific pathogen–free environment to a conventional housing facility [5]. In these experiments, rejection in the absence of perforin also proceeded even in the presence of blocking mAb to interferon-γ and IL-12: this suggests that these 2 molecules may also not be required in this rejection process. In total, the findings to date indicate that whereas perforin is an important molecule that NK cells use in resistance, distinct cytotoxic or noncytotoxic pathways can clearly effect rejection if the perforin/granzyme pathway is absent. A question remaining is the relative contribution of such pathways when perforin is present in the resisting host NK populations. Regardless, it is important to consider that intervention strategies targeting the inhibition of perforin-mediated pathways in host NK cells may not prevent resistance, depending on the conditions of the transplantation.

In addition to the perforin/granzyme effector pathway, the involvement of death receptor pathways has also been studied in murine models of NK-mediated resistance. Baker et al. [2] found that transplantation of 2 × 106 BM cells into lethally irradiated recipients expressing nonfunctional fasL were capable of rejecting MHC-mismatched BM allografts, thereby demonstrating that this ligand was not required for this NK-mediated resistance. To prevent possible compensation by a perforin pathway in this model (and by fasL in the perforin-deficient recipients noted previously), lpr (fas-deficient) donor marrow was infused into perforin−/− mice, thus creating a double-deficient transplant. Transplantation of MHC-disparate marrow from C3H.HeJ lpr (H2k) donors was also resisted in perforin−/− recipients [2]. Thus, even though resistance by H2b recipients against H2k donors is genetically “weak” [33], NK-mediated rejection of the “standard” (2 × 106) marrow dose seemed to occur in the absence of the 2 major cytotoxic pathways, with the caveat that the lpr mutation is “leaky”; therefore, some fas expression could not be ruled out in marrow cells [69]. Nonetheless, these findings led us to hypothesize that alternative pathways, distinct from the perforin/granzyme and fasL/fas pathways, are likely to exist in recipients after transplantation that can effect resistance to progenitor allografts.

An eloquent approach to examine the involvement of death receptors other than fas used a strategy in which HSPCs were transduced with a retroviral vector that expressed the caspase 8 inhibitor cFLIP [7]. Transplantation of these class I–deficient progenitor populations into perforin−/− recipients resulted in diminished resistance in this NK-dependent model. One interpretation of this finding was that some caspase 8–dependent extrinsic apoptotic signal may mediate an additional pathway by which NK cells can effect allograft resistance [7]. It is interesting to note that blockade of the caspase 8–dependent pathways in this study did not confer complete protection from NK-mediated rejection: this suggests that a third mechanism distinct from perforin or TNF family death receptors may be contributory. Furthermore, studies in which B6.129 or B6 perforin−/− mice underwent transplantation with B6-lpr transporter-associated protein (TAP)-deficient marrow (thus resulting in NK-dependent resistance) found that resistance was weaker in the concomitant absence of perforin- and fasL-mediated lytic pathways, particularly in the B6.129 recipients [5]. These findings again demonstrate a role for perforin- and fasL-dependent effector pathways in NK-mediated resistance. It is interesting to note that neutralizing anti-TNF mAb administered to B6.129 or B6 perforin−/− recipients did not further inhibit this NK-dependent resistance [5, 8]. In summary, the findings published to date have observed that in the absence of individual cytolytic pathways, some or almost no dimunition of resistance against BM allografts is observed in NK-dependent models [2, 4, 5, 7, 8, 9, 10, 39]. However, concomitant absence of the major cytolytic pathways significantly reduces NK-dependent resistance.

Back to Article Outline

T cell–mediated resistance to allogeneic hematopoietic progenitor cell grafts 

The NK compartment represents the initial immunologic barrier that a donor HSPC must circumvent after transplantation of MHC-mismatched murine BM if engraftment is to take place. However, presumably in part because of the numbers of NK cells and limited expansion of the compartment, this barrier can be overridden by increasing the number of transplanted cells [19]. Subsequently, surviving host anti-donor reactive T cells provide a second and stronger immune barrier to successful engraftment that, however, can also be overridden with large (ie, mega) doses of progenitor cell–containing inoculum [70, 71]. As noted previously, T-cell involvement, including both CD4+ and CD8+ subsets, is well established in murine and canine models of allograft resistance (Z.Z. and R.B.L., unpublished data) [9, 19, 34, 35, 72, 73]. Although there is a paucity of information regarding NK-mediated resistance in humans, several studies have implicated host T cells in clinical cases of graft failure [27, 74, 75, 76, 77]. In both HLA-matched and -mismatched allogeneic recipients, host CD8+ and CD4+ T cells capable of anti-donor cytolytic activity have been isolated from the peripheral blood of patients undergoing acute allograft rejection [74, 76, 77, 78]. Additionally, early recovery of the host CD8+ compartment positively correlates with an increased risk of graft rejection, and it has recently been suggested that residual host cytotoxic T lymphocytes (CTLs) and IL-2–producing CD4+ T-helper cells are responsible for the unstable mixed chimerism observed after nonmyeloablative conditioning and transplantation [27].

Not surprisingly, individuals unintentionally sensitized (eg, aplastic anemia patients) or experimental animal recipients intentionally sensitized to donor antigens possess even stronger barriers to successful allogeneic engraftment compared with recipients who are naive with respect to donor antigens [6, 9, 79, 80, 81, 82, 83, 84, 85, 86, 87]. As noted previously, in murine models involving MHC-matched/MiHA-mismatched (ie, T cell–mediated) transplantations, recipients possessing primed host T-cell populations (ie, memory T cells) have been shown to resist hematopoietic allogeneic grafts with enhanced kinetics compared with unsensitized recipients [6, 9, 13, 38]. In contrast to the NK barrier, it is extremely difficult to override resistance in mice possessing anti-donor memory T-cell populations. Studies in our laboratory have found that megadosing the HSPC inoculum, ie, transplanting up to 3 × 107 donor MiHA-mismatched marrow cells, continues to be efficiently rejected by 5 days after transplantation in recipients previously sensitized against donor antigens [9].

The graft rejection that occurs in patients with severe aplastic anemia who have received HSPCT from HLA-identical sibling donors is most likely the result of prior transfusion-induced sensitization to donor MiHA [85, 88, 89]. Such responses likely involve host anti-donor T-cell populations, including T memory cell effector pathways. With respect to such “sensitized” recipients (Figure 1), recent findings regarding the unexpected presence of antigen-specific T cells in mothers many years after childbirth may be important to consider regarding resistance to engraftment after clinical HSPCT in such individuals (Figure 3). It is well established that pregnancy can lead to allogeneic immunization and development of antibodies against paternal HLA and erythrocytes, as well as other antigens [90, 91]. It is interesting to note that direct evidence has recently been reported involving women who have never received transplants or transfusions indicating the presence of antigen-specific T cells directed against paternal transplantation antigens [92, 93]. After parturition, exposure to paternally derived alloantigens can apparently sensitize maternal T cells to MiHAs [93, 94, 95, 96]. A recent study identified circulating T cells in 5 (all female) out of 17 donors with T-cell receptors specific for H-Y–encoded antigens [92]. Another investigation examined peripheral blood from healthy multiparous females and, by using HLA-A2 tetramer reagents, identified HA-1–, HA-2–, and H-Y–specific T cells in 4 of 7 donors up to 22 years after their last delivery [97]. Although presently unknown, perhaps the risk factor of graft-versus-host disease associated with female to male and female to female transplantations involves responses by transferred alloreactive donor memory T cells after allogeneic clinical BMT [98].

  • View full-size image.
  • Figure 3. 

    Potential barrier composition in women after allogeneic HSPCT. Depending on the genetic disparity between donor and recipient and the potential presence of anti-donor reactive T memory cells, the cellular constituency of the immune-mediated barrier to alloengraftment may be composed of multiple cell populations in the recipient. For example, sensitization to paternally derived alloantigens during pregnancy is one setting in which anti-donor alloantigen-reactive memory T cells could be generated.

In contrast to experimental animals typically housed in relatively pathogen-free environments, individuals are continually exposed to multiple environmental and pathogenic antigens throughout their lifetimes. Because many such antigen encounters will result in memory cell generation, a reasonable hypothesis regarding the resultant memory T-cell populations is that some cells may possess an alloreactive T-cell receptor [99]. This phenomenon, termed heterologous immunity, was reportedly responsible for heightened rejection of cardiac allografts observed in recipients >1 month after lymphocytic choriomeningitis virus infection [99]. The generation of T cells with cross-reactivity against HLA antigens has been demonstrated for HLA-B8–restricted Epstein-Barr virus–specific T-cell clones and is consistent with observations in mice reporting the generation of alloreactive T cells after exposure to murine cytomegalovirus and other viruses [99, 100, 101, 102]. This group of findings suggests that memory T cells specific for alloantigen may clinically be considerably more frequent than previously appreciated. If this is true, it is conceivable and even likely that a significant number of clinical transplantations may involve a host memory T-cell component that might influence the strength of resistance after an allogeneic HSPCT. Because memory cells are more difficult to tolerize than naive T-cell populations, regulation of such cells to circumvent a host’s barrier toward the establishment of HSPC engraftment would be a challenging task during the posttransplantation period [103].

Back to Article Outline

The involvement of cytotoxic pathways in T cell–mediated resistance to progenitor cell grafts 

Similar to NK cell barrier activity, the precise molecular pathway(s) by which T-cell populations mediate resistance to allogeneic HSPC is also not clear. The findings to date concerning the use of cytotoxic pathways by host T cells are particularly intriguing. Our own experimental studies have investigated whether CTLs are a prerequisite for effective resistance against allogeneic MiHA-mismatched progenitor cells [6, 9, 13]. As noted previously, studies have isolated and cloned anti-donor alloreactive CTLs from primates as well as mice rejecting hematopoietic grafts [26, 104]. Subsequently, several murine CTL clones were found to be capable of transferring resistance to nonresistant recipients [104], and several clinical studies have identified the presence of host anti-donor CD8+ clones [76, 78] in patients who experienced graft failure. Some of these clones were able to lyse donor-derived targets in vitro. However, the presence of anti-donor CTLs in recipients after transplantation or the fact that transferred CTLs can mediate resistance does not constitute mea culpa in the transplantation setting or conclusively demonstrate that CTLs are required for the resistance process. In fact, accumulating evidence from studies in murine models continues to demonstrate that efficient T (as well as NK) cell–mediated resistance against allogeneic progenitor cell grafts can occur in the absence of the major components of cellular cytotoxic pathways [2, 3, 4, 5, 6, 7, 8, 9, 10].

Several experimental models have been designed to specifically study resistance principally—if not exclusively—mediated by host T cells (Table 1). One approach examines the resistance that occurs after transplantations between MHC-identical/MiHA-disparate donor/recipient pairings [6, 9, 105]. Because NK cells contribute a significant barrier against MHC-disparate transplantations, MHC-matched/MiHA-mismatched transplantation models effectively enable isolated study of T cells in the rejection process. Other approaches specifically deplete the host NK compartment by administering antibodies directed against the NK1.1 or DX5 molecules before transplantation, whereas other models take advantage of the fact that only the T-cell compartment can be specifically and markedly expanded in response to alloantigen exposure (ie, priming) [19]. Resistance in such sensitized MiHA-disparate recipients therefore should reflect the nature of a primed, ie, memory, response in that large numbers of donor cells should be rejected with rapid kinetics [9, 13, 38].

Experimental findings from our laboratory transplanting BM from MHC-matched donors into recipient mice sensitized to donor MiHA and singly or doubly deficient in the major mediators of cytotoxicity have demonstrated that T cell–dependent HSPCT rejection is virtually unimpaired and proceeds in a remarkably rapid and efficient manner [6, 9]. The host barrier in MiHA-disparate recipients sensitized to donor alloantigens requires a radioresistant CD8+ T cell–dependent pathway [9, 13]. Mice either individually or simultaneously lacking perforin or fasL did not demonstrate any defects in the ability to reject MHC-matched/MiHA-disparate or MHC-disparate grafts (Table 3) [2] as determined by both kinetics of rejection (both B6 wild type and B6 cytotoxic double-defective [cdd] mice reject within 48 hours of BMT), even after transplantation of high numbers of BM cells [6]. In total, these studies eliminated the possibility of compensation between granule exocytosis–dependent and fas-dependent cytolytic contributions effecting rejection and indicate that (1) an efficient resistance pathway is mediated in the absence of concomitant perforin and fasL and (2) if differing effector pathways are used in cytotoxically normal and deficient mice, the barrier response seems to arise in the same time period after transplantation.

Table 3. Cytotoxic Effector Pathways Examined in Resistance against Allogeneic HSPCT
Resistance ModelCytotoxic DeficitBarrier Activity in HostCells InvolvedReferences
Host unsensitized to donor antigens
Hematopoietic histoincompatibility (parent into F1 transplant)Granzyme BIntactNK4
MHC class I⁎PerforinIntact (clean environment) 5
PerforinPartial loss (conventional) 5
Perforin + CD95Partial loss 8
Death receptorsPartial loss 7
MHC class I/IIPerforinIntactNK + T2, 39†
CD95LIntact 2, 39†
MHC matchedPerforinIntactTTable 4
CD95LIntact Table 4
Perforin and fasLDiminished Table 4
Host sensitized to donor antigens
MHC class I/IIPerforin + CD95L + TNFR1 + TRAILIntactT9
MHC matchedPerforinIntactT6
CD95LIntact
Granzyme BIntact 9
Perforin + CD95L + TNFR1 + TRAILIntact 9
Perforin + CD95L + TRAIL + TL1a + TWEAKIntactT13

⁎ Resistance in these studies was mediated by host NK cells because of the absence of MHC class I on syngeneic donor (ie, TAP−/− or β2u−/− origin) marrow.

† Studies using purified allogeneic HSCs.

The participation of other death ligand receptor signaling pathways by host T cells has also been examined [9, 13]. B6 cdd mice have been challenged with BM from TNFR1- or TNFR2-deficient donors. TNF can initiate an apoptotic program either through the death domain containing TNFR1 or through TNFR2 [9, 106] to effectively create cytotoxic triple-defective transplants. It is interesting to note that no impairment in resistance has been observed in these transplants. Examination of TNF involvement in NK-mediated marrow graft resistance by using anti-TNF mAb infusion also failed to identify a contribution by this lytic pathway in these models [5].

Recently, several additional TNF family apoptosis-inducing receptors/ligands have been identified, including DR3/TL1a, TWEAK/Fn14, TRAIL/DR4, LTβ/LIGHT, and DR6, that also could be proposed to play a role in the resistance process [61, 107, 108, 109, 110, 111, 112]. To study the involvement of several of these molecules in resistance dependent on T cells, B6 cdd mice sensitized against donor alloantigens underwent transplantation with the administration of high levels of blocking mAbs to disrupt the signaling dependent on 3 ligand-receptor pairs (Table 3). These experiments consistently found that B6 cdd T cells again mediated efficient resistance with the disruption of TRAIL-, TWEAK-, DR3-, fasL-, and perforin-mediated cytotoxicity, even when all 5 pathways were concomitantly disrupted [13]. Taken together, the data from this group of studies support the notion that in antigen-sensitized individuals, in whom resistance is principally dependent on responses by effector cells derived from the host memory T-cell pool, known cytotoxic pathways do not seem to be required to mediate rejection against HSPC populations. However, one cannot formally exclude the contribution of a presently uncharacterized death receptor pathway capable of mediating strong resistance. Moreover, although apoptosis can proceed through the extrinsic pathways discussed previously that involve signaling through death receptors, programmed cell death can also occur via an intrinsic or mitochondrial-dependent pathway initiated by diverse events such as growth factor deprivation or DNA damage (Figure 1). Therefore, if the target HSPC population does undergo apoptosis during resistance, it may not be due entirely or at all to an extrinsic, death receptor–dependent pathway. At present, the involvement of the intrinsic apoptotic pathway in HSPC rejection has not been examined and should therefore not be excluded.

Recent studies in our laboratory have been examining T cell–mediated resistance after MiHA-mismatched BMT in nonmyeloablatively conditioned cytotoxically impaired recipients who have not been sensitized to donor antigens (Z.Z. and R.B.L., unpublished data). Resistance was evident in recipients lacking perforin (MiHA-mismatched donors, Table 4, line 2; MHC-mismatched donors, Table 4, line 5) and recipients lacking functional fasL (Table 4, line 3) that was indistinguishable from that in cytotoxically unimpaired (B6 wild-type) recipients (Table 4). These findings could indicate that either pathway is sufficient to effect resistance in such recipients. A recent study also found that transplantation of purified allogeneic stem cells into recipients with defects in individual cytotoxic pathways did not result in altered resistance [39]. Recent findings in our laboratory indicate that combined deficiency in both perforin and fasL diminishes resistance in recipients of MHC- or MiHA-mismatched transplants (Table 4, lines 4 and 6). Such results suggest that, similar to the findings in NK-dependent models, these observations support a requirement for cytotoxic function for optimal resistance by effector cells arising from naive host T cells in unsensitized recipients.

Table 4. Resistance after Mini-BMT in Unsensitized Recipients Occurs in the Individual but Not Combined Absence of Host Perforin- and fasL-Dependent Cytotoxicity
Recipient (5.5 Gy)Marrow Donor⁎Lytic Pathway(s) Not Used by the HostDonor Chimerism (>1%) 3 wk after BMT†
1.B6-wtC3H.SWNone0/11
2.B6-perf−/−C3H.SWPerforin0/9
3.B6-gld+/+C3H.SWfasL0/14
4.B6-cddC3H.SWPerforin+fasL11/14
5.B6-perf−/−(B6 × C3H)F1Perforin0/3
6.B6-perf−/−(B6 × C3H)F1-lpr‡Perforin+fasL3/3

Wt indicates wild type; perf, perforin; cdd, cytotoxic double deficient; gld, generalized lymphoproliferative disorder.

⁎ A total of 4 × 106 MHC-matched, MiHA-mismatched C3H.SW and 1 × 107 MHC-mismatched (B6 × C3H)F1 BM cells (T-cell depleted) were transplanted.

† Donor chimerism was assessed at weekly intervals by analysis of peripheral blood staining for Ly9.1+ (C3H.SW) or H2Kk (C3H) expression.

‡ The lpr mutation results in very low levels of fas messenger RNA and almost undetectable levels of cell-surface fas expression [118].

Back to Article Outline

Summary and future directions 

Host immunologic resistance against allogeneic HSPCT is of clinical concern, particularly in cases involving an HLA disparity and T cell–depleted progenitor cell inoculum [113]. The increasing trend to incorporate less ablative conditioning regimens will continue to demand the understanding and regulation of recipient anti-MHC and MiHA-specific allogeneic immune responses after transplantation [27]. During the past 3 decades, experimental studies have identified important cellular effector populations of resistance operative in a variety of allogeneic transplantation settings, including donor/recipient pairs who are MHC mismatched and MHC matched and in the context of prior sensitization to donor alloantigens (Figure 1; Table 1). T and NK cells are clearly important host elements after MHC-disparate transplantations, whereas T cells are important in situations in which there is a MiHA disparity only or in recipients in whom prior sensitization to donor antigens has occurred. Because these same cell populations are critical for immune defense against infectious agents and malignancies after transplantation, it is extremely important to identify the effector mechanisms in HSPC resistance to enable selective inhibition of rejection while minimally affecting their host defense capability.

In HLA-disparate HSPCTs and transplantations in which there has been no prior sensitization to donor alloantigens, experimental studies suggest that NK cells can contribute to acute resistance. The role of cytotoxicity in NK-mediated rejection and the lytic pathways used by these cells is complex. Although the perforin/granzyme exocytosis pathway is required for in vitro NK cytotoxicity, it was not found to be required for all NK-mediated rejection events in experiments that used MHC-disparate as well as TAP−/− HSPCT grafts [2, 4, 5]. However, in some experimental settings, perforin-deficient recipients exhibit a diminished ability to effect NK-mediated resistance [5, 7, 8]. Some of the nonperforin cytolytic effector activity by NK cells seems to result from the use of caspase 8–mediated extrinsic apoptotic pathways, although the specific ligand/receptor interactions remain to be elucidated [56, 59]. Intriguingly, with the disruption of both perforin- and caspase 8–dependent pathways, NK cells continue to retain some capacity to reject allogeneic HSPCT [7]. Elaboration of this effector pathway may lead to insights on how to limit NK-mediated rejection of HSPC while maintaining selective NK cell functions against viral pathogens or malignancies in recipients.

In addition to the barrier effected by NK cells, T cells also contribute resistance after MHC-mismatched HSPCT. In recipients of transplants that have been HLA matched and in individuals who have received prior exposure by natural or iatrogenic means to donor alloantigen, recipient T cells are the major immune barrier to successful donor engraftment. Although the presence of anti-donor CTL activity correlates with resistance in patients, the requirement for this functional population within both CD8 and CD4 subsets for rejection has not been formally demonstrated. In several experimental models, the transfer of in vitro–derived CTL populations into nonresistant recipients was shown to confer resistance [104] (M.K. and R.B.L., unpublished data). It is interesting to note that the singular absence of perforin or fasL does not abrogate the ability to reject MHC-matched MiHA-mismatched marrow (Table 4) or MHC-mismatched stem cells in nonablatively and ablatively conditioned recipients, respectively [39]. This does not eliminate a role for cytotoxicity in these responses, but it is consistent with the possibility that more than a single effector pathway may be invoked by the host to resist engraftment. Models specifically examining resistance between MHC-matched MiHA-mismatched donor/recipient combinations in which the recipient has been sensitized to donor alloantigens have repeatedly found that the combined absence of perforin and fasL, together with TNFR1 or 2 absence—or together with the inhibition of additional death ligand/death receptor (TRAIL-DR5, TL1a-DR3, and TWEAK-Fn14) signaling pathways—does not abrogate and seems not to dampen effective graft resistance [6, 9, 13].

Thus, despite extensive investigation of cytotoxicity in experimental immune-mediated graft resistance, the studies to date cannot rule out the presence and contribution of noncytotoxic mechanisms. As noted previously, allogeneic HSCs can be rejected competently in unsensitized mice that express a diphtheria toxin transgene regulated by the granzyme A promoter [10]. Theoretically, this model results in elimination of CTLs and NK cells. One interpretation of this finding is that non-CTL populations of effector T cells may participate in resistance; this is consistent with the notion that cytotoxic mechanisms are not required in the barrier against allogeneic progenitor cell engraftment. However, our recent studies in nonmyeloablated recipients of MiHA-mismatched BM demonstrate that perforin- or fasL-mediated function is, in fact, required for the allograft resistance observed. In contrast, several studies involving recipients sensitized against donor alloantigens support the possibility that T cell–mediated resistance against HSPCT grafts may be effected through noncytotoxic pathways [6, 9, 13]. Whether such pathways are compensatory in the absence of cytotoxic mediators or are of primary importance is presently unclear. Regardless, these observations are important to appreciate in the design of strategies to support progenitor cell engraftment and inhibit immune-mediated resistance. If, in fact, under any transplantation conditions, cytotoxic pathways are not used by the host, what is the overall fate of the HSPC in such a rejection process? Do these progenitor populations undergo apoptosis or alternatively persist in the host, perhaps functionally inhibited in their capacity to proliferate or differentiate (Figure 1)? For example, CD70 (found on activated T cells) signaling via CD27 can inhibit myeloid differentiation and long-term repopulation in vivo [114, 115]. Additionally, cytokine and growth factor signaling can regulate progenitor cell proliferation and lineage commitment [116, 117].

To account for the experimental findings reported to date regarding the cell populations that mediate resistance and the role of cytotoxicity by these cells in the barrier to hematopoietic engraftment, a model must include the results pertaining to resistance by NK and T cells in both sensitized and unsensitized progenitor cell transplant recipients. The hypothesis we have proposed is that the effector pathways used by the various host populations mediating resistance differ and are, therefore, at most only partially overlapping (Figure 4). We interpret the findings to indicate that the importance of cytotoxicity in resistance pathways against progenitor cell allografts differs according to the source of the host effector populations responsible for the barrier response, as well as the sensitization (ie, against donor antigens) status of the recipient. We speculate that host effector cells derived from NK or naive T cells, ie, effector cells, which arise in recipients who are not sensitized to donor graft antigens, require some cytotoxic-dependent pathway (eg, perforin/granzymes) to mediate strong and highly efficient resistance (Figure 4, left). In contrast, effector cells derived from host memory T-cell populations, ie, effector cells generated in recipients previously primed to donor antigens, do not require known cytotoxic molecules to effect highly efficient resistance (Figure 4, right). It is interesting to speculate that such an immunologic pathway would not likely be limited to resistance against allogeneic hematopoietic cells and thus may have significantly broader importance. One consequence of this hypothesis is that the successful targeting (antibody, small interfering RNA, antisense, and so on) of cytotoxic pathways to inhibit their function would have markedly different effects on subsequent engraftment according to the conditions of the transplantation (Figure 1). It is interesting to note that the findings reported to date identifying individual cytotoxic pathways that are not required—together with observations that in some transplantations, many of these pathways can be simultaneously inhibited without dampening allograft resistance—lend credence to the notion that identifying the appropriate effector pathways of resistance could lead to strategies specifically blocking these while leaving many significant host immune functions intact early after transplantation. Emerging insights into the cellular and molecular mechanisms underlying rejection of allogeneic HSPCs will facilitate progress toward the development of more effective nonmyeloablative conditioning protocols.

  • View full-size image.
  • Figure 4. 

    A proposed model for the involvement of cytotoxic effector pathways in immune-mediated resistance to allogeneic hematopoietic stem/progenitor cell engraftment. The model presents the hypothesis that depending on the source of the effector cells that are generated in recipients after transplantation, the contribution of cytotoxic pathways will differ. For example, if effector cells are derived from naive host CD8+ T cells (or NK cells), cytolytic activity via perforin- or fasL-dependent pathways is critical for generation of a strong barrier to engraftment. In contrast, effector cells derived from host CD8+ memory T cells mediate potent barrier activity independently of these 2 pathways.

Back to Article Outline

References 

  1. Lowin B , Hahne M , Mattmann C , Tschopp J . Cytolytic T-cell cytotoxicity is mediated through perforin and Fas lytic pathways . Nature . 1994;370:650–652
  2. Baker MB , Podack ER , Levy RB . Perforin- and Fas-mediated cytotoxic pathways are not required for allogeneic resistance to bone marrow grafts in mice . Biol Blood Marrow Transplant . 1995;1:69–73
  3. Shizuru JA , Jerabek L , Edwards CT , Weissman IL . Transplantation of purified hematopoietic stem cells (requirements for overcoming the barriers of allogeneic engraftment) . Biol Blood Marrow Transplant . 1996;2:3–14
  4. Graubert TA , Russell JH , Ley TJ . The role of granzyme B in murine models of acute graft-versus-host disease and graft rejection . Blood . 1996;87:1232–1237
  5. Bennett M , Taylor PA , Austin M , et al.   Cytokine and cytotoxic pathways of NK cell rejection of class I-deficient bone marrow grafts (influence of mouse colony environment) . Int Immunol . 1998;10:785–790
  6. Jones M , Komatsu M , Levy RB . Cytotoxically impaired transplant recipients can efficiently resist major histocompatibility complex-matched bone marrow allografts . Biol Blood Marrow Transplant . 2000;6:456–464
  7. Taylor MA , Chaudhary PM , Klem J , Kumar V , Schatzle JD , Bennett M . Inhibition of the death receptor pathway by cFLIP confers partial engraftment of MHC class I-deficient stem cells and reduces tumor clearance in perforin-deficient mice . J Immunol . 2001;167:4230–4237
  8. Taylor MA , Ward B , Schatzle JD , Bennett M . Perforin- and Fas-dependent mechanisms of natural killer cell mediated rejection of incompatible bone marrow cell grafts . Eur J Immunol . 2002;32:793–799
  9. Komatsu M , Mammolenti M , Jones M , Jurecic R , Sayers TJ , Levy RB . Antigen-primed CD8+ T cells can mediate resistance, preventing allogeneic marrow engraftment in the simultaneous absence of perforin-, CD95L-, TNFR1-, and TRAIL-dependent killing . Blood . 2003;101:3991–3999
  10. Aguila HL , Hershberger RJ , Weissman IL . Transgenic mice carrying the diphtheria toxin A chain gene under the control of the granzyme A promoter (expected depletion of cytotoxic cells and unexpected depletion of CD8 T cells) . Proc Natl Acad Sci U S A . 1995;92:10192–10196
  11. Thiele DL , Lipsky PE . The immunosuppressive activity of L-leucyl-L-leucine methyl ester (selective ablation of cytotoxic lymphocytes and monocytes) . J Immunol . 1986;136:1038–1048
  12. Loughran TP , Raff RF , Graham TC , et al.   Transfusion of autologous cytotoxic cells leads to failure of unrelated, DLA-nonidentical marrow grafts . Exp Hematol . 1990;18:1126–1131
  13. Zimmerman ZF , Shatry AM , Deyev V , et al.   Effector cells derived from host CD8 memory T cells mediate rapid resistance against minor histocompatibility antigen-mismatched allogeneic marrow grafts without participation of perforin, Fas ligand, and the simultaneous inhibition of 3 tumor necrosis factor family effector pathways . Biol Blood Marrow Transplant . 2005;11:576–586
  14. Storb R , Weiden PL , Graham TC , Thomas ED . Failure of engraftment and graft-versus-host disease after canine marrow transplantation. Two phenomena linked to but not exclusively determined by known antigens of the major histocompatibility complex . Transplant Proc . 1978;10:113–118
  15. Murphy WJ , Kumar V , Bennett M . Rejection of bone marrow allografts by mice with severe combined immune deficiency (SCID). Evidence that natural killer cells can mediate the specificity of marrow graft rejection . J Exp Med . 1987;165:1212–1217
  16. Yankelevich B , Knobloch C , Nowicki M , Dennert G . A novel cell type responsible for marrow graft rejection in mice. T cells with NK phenotype cause acute rejection of marrow grafts . J Immunol . 1989;142:3423–3430
  17. Raff RF , Deeg HJ , Loughran TP , et al.   Characterization of host cells involved in resistance to marrow grafts in dogs transplanted from unrelated DLA-nonidentical donors . Blood . 1986;68:861–868
  18. Lee LA , Sergio JJ , Sykes M . Natural killer cells weakly resist engraftment of allogeneic, long-term, multilineage-repopulating hematopoietic stem cells . Transplantation . 1996;61:125–132
  19. Murphy WJ , Kumar V , Bennett M . Acute rejection of murine bone marrow allografts by natural killer cells and T cells (differences in kinetics and target antigens recognized) . J Exp Med . 1987;166:1499–1550
  20. Takeda K , Moore MW , Dennert G . Acute rejection of marrow grafts in mice. Dependence on and independence of functional TcR in the rejection process . J Immunol . 1994;152:4407–4416
  21. Barge AJ , Johnson G , Witherspoon R , Torok-Storb B . Antibody-mediated marrow failure after allogeneic bone marrow transplantation . Blood . 1989;74:1477–1480
  22. Miklos DB , Kim HT , Zorn E , et al.   Antibody response to DBY minor histocompatibility antigen is induced after allogeneic stem cell transplantation and in healthy female donors . Blood . 2004;103:353–359
  23. Zorn E , Miklos DB , Floyd BH , et al.   Minor histocompatibility antigen DBY elicits a coordinated B and T cell response after allogeneic stem cell transplantation . J Exp Med . 2004;199:1133–1142
  24. Grayson JM , Zajac AJ , Altman JD , Ahmed R . Cutting edge (increased expression of Bcl-2 in antigen-specific memory CD8+ T cells) . J Immunol . 2000;164:3950–3954
  25. Storb R . Critical issues in bone marrow transplantation . Transplant Proc . 1987;19:2774–2781
  26. Reisner Y , Benbassat I , Douer D , Kaploon A , Schwartz E , Ramot B . Demonstration of clonable alloreactive host T cells in a primate model for bone marrow transplantation . Proc Natl Acad Sci U S A . 1986;83:4012–4015
  27. Kraus AB , Shaffer J , Toh HC , et al.   Early host CD8 T-cell recovery and sensitized anti-donor interleukin-2-producing and cytotoxic T-cell responses associated with marrow graft rejection following nonmyeloablative allogeneic bone marrow transplantation . Exp Hematol . 2003;31:609–621
  28. Lowry PA , Tabbara IA . Peripheral hematopoietic stem cell transplantation (current concepts) . Exp Hematol . 1992;20:937–942
  29. Jordan CT , Lemischka IR . Clonal and systemic analysis of long-term hematopoiesis in the mouse . Genes Dev . 1990;4:220–232
  30. Jones RJ , Wagner JE , Celano P , Zicha MS , Sharkis SJ . Separation of pluripotent haematopoietic stem cells from spleen colony-forming cells . Nature . 1990;347:188–189
  31. Aguila HL , Weissman IL . Hematopoietic stem cells are not direct cytotoxic targets of natural killer cells . Blood . 1996;87:1225–1231
  32. Koh CY , Welniak LA , Murphy WJ . Lack of correlation between an assay used to determine early marrow allograft rejection and long-term chimerism after murine allogeneic bone marrow transplantation (effects of marrow dose) . Biol Blood Marrow Transplant . 2005;11:252–259
  33. Davenport C , Kumar V , Bennett M . Rapid rejection of H2k and H2k/b bone marrow cell grafts by CD8+ T cells and NK cells in irradiated mice . J Immunol . 1995;155:3742–3749
  34. Vallera DA , Taylor PA , Sprent J , Blazar BR . The role of host T-cell subsets in bone-marrow rejection directed to isolated major histocompatibility complex class-I versus class-II differences of Bm1 and Bm12 mutant mice . Transplantation . 1994;57:249–256
  35. Sprent J , Surh CD , Agus D , Hurd M , Sutton S , Heath WR . Profound atrophy of the bone marrow reflecting major histocompatibility complex class II-restricted destruction of stem cells by CD4+ cells . J Exp Med . 1994;180:307–317
  36. Sentman CL , Streilein R , Kumar V , et al.   Kinetics and location of bone marrow cell graft rejection by irradiated mice . Nat Immun Cell Growth Regul . 1988;7:95–105
  37. Shatry AM , Jones M , Levy RB . The effect of the spleen on compartmental levels and distribution of donor progenitor cells after syngeneic and allogeneic bone marrow transplants . Stem Cells Dev . 2004;13:51–62
  38. Dennert G , Anderson CG , Warner J . T-killer cells play a role in allogeneic bone-marrow graft-rejection but not in hybrid resistance . J Immunol . 1985;135:3729–3734
  39. Scheffold C , Scheffold YC , Cao TM , Gworek J , Shizuru JA . Cytokines and cytotoxic pathways in engraftment resistance to purified allogeneic hematopoietic stem cells . Biol Blood Marrow Transplant . 2005;11:1–12
  40. Cudkowicz G , Bennett M . Peculiar immunobiology of bone marrow allografts. I. Graft rejection by irradiated responder mice . J Exp Med . 1971;134:83–102
  41. Kiessling R , Hochman PS , Haller O , Shearer GM , Wigzell H , Cudkowicz G . Evidence for a similar or common mechanism for natural killer cell activity and resistance to hemopoietic grafts . Eur J Immunol . 1977;7:655–663
  42. Barao I , Murphy WJ . The immunobiology of natural killer cells and bone marrow allograft rejection . Biol Blood Marrow Transplant . 2003;9:727–741
  43. Liao NS , Bix M , Zijlstra M , Jaenisch R , Raulet D . MHC class I deficiency (susceptibility to natural killer (NK) cells and impaired NK activity) . Science . 1991;253:199–202
  44. Ljunggren HG , Karre K . In search of the ‘missing self’ (MHC molecules and NK cell recognition) . Immunol Today . 1990;11:237–244
  45. Anderson SK , Ortaldo JR , McVicar DW . The ever-expanding Ly49 gene family (repertoire and signaling) . Immunol Rev . 2001;181:79–89
  46. McLaren FH , Svendsen CN , Van der Meide P , Joly E . Analysis of neural stem cells by flow cytometry (cellular differentiation modifies patterns of MHC expression) . J Neuroimmunol . 2001;112:35–46
  47. Modo M , Mellodew K , Rezaie P . In vitro expression of major histocompatibility class I and class II antigens by conditionally immortalized murine neural stem cells . Neurosci Lett . 2003;337:85–88
  48. Mammolenti M , Gajavelli S , Tsoulfas P , Levy R . Absence of major histocompatibility complex class I on neural stem cells does not permit natural killer cell killing and prevents recognition by alloreactive cytotoxic T lymphocytes in vitro . Stem Cells . 2004;22:1101–1110
  49. Drukker M , Katz G , Urbach A , et al.   Characterization of the expression of MHC proteins in human embryonic stem cells . Proc Natl Acad Sci U S A . 2002;99:9864–9869
  50. Bartholomew A , Sturgeon C , Siatskas M , et al.   Mesenchymal stem cells suppress lymphocyte proliferation in vitro and prolong skin graft survival in vivo . Exp Hematol . 2002;30:42–48
  51. Krampera M , Glennie S , Dyson J , et al.   Bone marrow mesenchymal stem cells inhibit the response of naive and memory antigen-specific T cells to their cognate peptide . Blood . 2003;101:3723–3729
  52. Scott I , O’Shea J , Bunce M , et al.   Molecular typing shows a high level of HLA class I incompatibility in serologically well matched donor/patient pairs (implications for unrelated bone marrow donor selection) . Blood . 1998;92:4864–4871
  53. Ruggeri L , Capanni M , Urbani E , et al.   Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants . Science . 2002;295:2097–2100
  54. Koenecke C , Shaffer J , Alexander SI , et al.   NK cell recovery, chimerism, function and recognition in recipients of haploidentical hematopoietic cell transplantation following nonmyeloablative conditioning using a humanized anti-CD2 mAb, Medi-507 . Exp Hematol . 2003;31:911–923
  55. Young JD , Hengartner H , Podack ER , Cohn ZA . Purification and characterization of a cytolytic pore-forming protein from granules of cloned lymphocytes with natural killer activity . Cell . 1986;44:849–859
  56. Arase H , Arase N , Saito T . Fas-mediated cytotoxicity by freshly isolated natural-killer-cells . J Exp Med . 1995;181:1235–1238
  57. Kagi D , Ledermann B , Burki K , et al.   Cytotoxicity mediated by T cells and natural killer cells is greatly impaired in perforin-deficient mice . Nature . 1994;369:31–37
  58. Oshimi Y , Oda S , Honda Y , Nagata S , Miyazaki S . Involvement of Fas ligand and Fas-mediated pathway in the cytotoxicity of human natural killer cells . J Immunol . 1996;157:2909–2915
  59. Kayagaki N , Yamaguchi N , Nakayama M , et al.   Expression and function of TNF-related apoptosis-inducing ligand on murine activated NK cells . J Immunol . 1999;163:1906–1913
  60. Degliantoni G , Murphy M , Kobayashi M , Francis MK , Perussia B , Trinchieri G . Natural killer (NK) cell-derived hematopoietic colony-inhibiting activity and NK cytotoxic factor. Relationship with tumor necrosis factor and synergism with immune interferon . J Exp Med . 1985;162:1512–1530
  61. Thorburn A . Death receptor-induced cell killing . Cell Signal . 2004;16:139–144
  62. Maciejewski J , Selleri C , Anderson S , Young NS . Fas Antigen expression on CD34(+) human marrow-cells is induced by interferon-gamma and tumor-necrosis-factor-alpha and potentiates cytokine-mediated hematopoietic suppression in-vitro . Blood . 1995;85:3183–3190
  63. Sato T , Selleri C , Young NS , Maciejewski JP . Hematopoietic inhibition by interferon-gamma is partially mediated through interferon regulatory factor-I . Blood . 1995;86:3373–3380
  64. Eng VM , Car BD , Schnyder B , et al.   The stimulatory effects of interleukin (IL)-12 on hematopoiesis are antagonized by IL-12-induced interferon-gamma in-vivo . J Exp Med . 1995;181:1893–1898
  65. Snoeck HW , Weekx S , Moulijn A , et al.   Tumor necrosis factor alpha is a potent synergistic factor for the proliferation of primitive human hematopoietic progenitor cells and induces resistance to transforming growth factor beta but not to interferon gamma . J Exp Med . 1996;183:705–710
  66. Kawano Y , Takaue Y , Hirao A , et al.   Synergy among erythropoietin, interleukin-3, stem-cell factor (C-Kit ligand) and interferon-gamma on early human hematopoiesis . Stem Cells . 1994;12:514–520
  67. Allavena P , Scala G , Djeu JY , et al.   Production of multiple cytokines by clones of human large granular lymphocytes . Cancer Immunol Immunother . 1985;19:121–126
  68. Murphy WJ , Keller JR , Harrison CL , Young HA , Longo DL . Interleukin-2-activated natural-killer-cells can support hematopoiesis in vitro and promote marrow engraftment in vivo . Blood . 1992;80:670–677
  69. Cui H , Ju ST , Sherr DH . Functional expression of Fas (CD95) protein in autoimmune lpr mice . Cell Immunol . 1996;174:35–41
  70. Lapidot T , Faktorowich Y , Lubin I , Reisner Y . Enhancement of T-cell-depleted bone marrow allografts in the absence of graft-versus-host disease is mediated by CD8+ CD4− and not by CD8− CD4+ thymocytes . Blood . 1992;80:2406–2411
  71. Bachar-Lustig E , Rachamim N , Li HW , Lan F , Reisner Y . Megadose of T cell-depleted bone marrow overcomes MHC barriers in sublethally irradiated mice . Nat Med . 1995;1:1268–1273
  72. Storb R , Deeg HJ , Applebaum FR , et al.   Failure of anti-Ia monoclonal antibody to abrogate transfusion-induced sensitization and prevent marrow graft rejection in DLA-identical canine littermates . Transplantation . 1988;45:505–506
  73. Storb R , Deeg HJ . Failure of allogeneic canine marrow grafts after total-body irradiation. Allogeneic “resistance” versus transfusion-induced sensitization . Transplantation . 1986;42:571–580
  74. Kernan NA , Flomenberg N , Dupont B , O’Reilly RJ . Graft rejection in recipients of T-cell-depleted HLA-nonidentical marrow transplants for leukemia. Identification of host-derived antidonor allocytotoxic T lymphocytes . Transplantation . 1987;43:842–847
  75. Fleischhauer K , Kernan NA , O’Reilly RJ , Dupont B , Yang SY . Bone marrow-allograft rejection by T lymphocytes recognizing a single amino acid difference in HLA-B44 . N Engl J Med . 1990;323:1818–1822
  76. Marijt WAF , Kernan NA , Diazbarrientos T , et al.   Multiple minor histocompatibility antigen-specific cytotoxic T-lymphocyte clones can be generated during graft-rejection after HLA-identical bone-marrow transplantation . Bone Marrow Transplant . 1995;16:125–132
  77. Fleischhauer K , Zino E , Mazzi B , et al.   Peripheral blood stem cell allograft rejection mediated by CD4(+) T lymphocytes recognizing a single mismatch at HLA-DP beta 1*0901 . Blood . 2001;98:1122–1126
  78. Marijt WA , Veenhof WF , Goulmy E , Willemze R , van Rood JJ , Falkenburg JH . Minor histocompatibility antigens HA-1-, -2-, and -4-, and HY-specific cytotoxic T-cell clones inhibit human hematopoietic progenitor cell growth by a mechanism that is dependent on direct cell-cell contact . Blood . 1993;82:3778–3785
  79. Storb R , Epstein RB , Rudolph RH , Thomas ED . The effect of prior transfusion on marrow grafts between histocompatible canine siblings . J Immunol . 1970;105:627–633
  80. Storb R , Rudolph RH , Graham TC , Thomas ED . The influence of transfusions from unrelated donors upon marrow grafts between histocompatible canine siblings . J Immunol . 1971;107:409–413
  81. Storb R , Buckner CD , Fefer A , et al.   Marrow transplantation in aplastic anemia . Transplant Proc . 1974;6:355–358
  82. Deeg HJ , Shulman HM , Schmidt E , Yee GC , Thomas ED , Storb R . Marrow graft rejection and veno-occlusive disease of the liver in patients with aplastic anemia conditioned with cyclophosphamide and cyclosporine . Transplantation . 1986;42:497–501
  83. Storb R , Prentice RL , Thomas ED . Marrow transplantation for treatment of aplastic anemia. An analysis of factors associated with graft rejection . N Engl J Med . 1977;296:61–66
  84. Storb R , Thomas ED , Buckner CD , et al.   Marrow transplantation in thirty “untransfused” patients with severe aplastic anemia . Ann Intern Med . 1980;92:30–36
  85. Storb R , Prentice RL , Thomas ED , et al.   Factors associated with graft rejection after HLA-identical marrow transplantation for aplastic anaemia . Br J Haematol . 1983;55:573–585
  86. Parkman R , Rappeport J , Camitta B , Levey RH , Nathan DG . Successful use of multiagent immunosuppression in the bone marrow transplantation of sensitized patients . Blood . 1978;52:1163–1169
  87. Champlin RE , Horowitz MM , van Bekkum DW , et al.   Graft failure following bone marrow transplantation for severe aplastic anemia (risk factors and treatment results) . Blood . 1989;73:606–613
  88. Loughran TP , Storb R . Treatment of aplastic anemia . Hematol Oncol Clin North Am . 1990;4:559–575
  89. Ramsay NK , Kim TH , McGlave P , et al.   Total lymphoid irradiation and cyclophosphamide conditioning prior to bone marrow transplantation for patients with severe aplastic anemia . Blood . 1983;62:622–626
  90. Levine P , Stetson RE . Landmark article July 8, 1939. An unusual case of intra-group agglutination. By Philip Levine and Rufus E Stetson . JAMA . 1984;251:1316–1317
  91. Van Rood JJ , Eernisse JG , Van Leeuwen A . Leucocyte antibodies in sera from pregnant women . Nature . 1958;181:1735–1736
  92. James E , Chai JG , Dewchand H , Macchiarulo E , Dazzi F , Simpson E . Multiparity induces priming to male-specific minor histocompatibility antigen, HY, in mice and humans . Blood . 2003;102:388–393
  93. Tekolf WA , Shaw S . In vitro generation of cytotoxic cells specific for human minor histocompatibility antigens by lymphocytes from a normal donor potentially primed during pregnancy . J Exp Med . 1983;157:2172–2177
  94. Singal DP , Wadia YJ , Naipaul N . In vitro cell-mediated cytotoxicity to the male specific (H-Y) antigen in man . Hum Immunol . 1981;2:45–53
  95. Mutis T , Gillespie G , Schrama E , Falkenburg JH , Moss P , Goulmy E . Tetrameric HLA class I-minor histocompatibility antigen peptide complexes demonstrate minor histocompatibility antigen-specific cytotoxic T lymphocytes in patients with graft-versus-host disease . Nat Med . 1999;5:839–842
  96. Komlos L , Vardimon D , Notmann J , et al.   Role of children’s sex in mixed mother-child lymphocyte culture reactivity . Am J Reprod Immunol . 1990;22:4–8
  97. Verdijk RM , Kloosterman A , Pool J , et al.   Pregnancy induces minor histocompatibility antigen-specific cytotoxic T cells (implications for stem cell transplantation and immunotherapy) . Blood . 2004;103:1961–1964
  98. Gratwohl A , Baldomero H , John L , et al.   Peripheral blood pressure cells transplants across a major histocompatibility barrier in rabbits (positive effects of a higher number of precursor cells?) . Acta Haematol . 1996;95:176–180
  99. Adams AB , Williams MA , Jones TR , et al.   Heterologous immunity provides a potent barrier to transplantation tolerance . J Clin Invest . 2003;111:1887–1895
  100. Yang H , Welsh RM . Induction of allospecific and virus-specific memory cytotoxic T cells during acute arenavirus infections . Med Microbiol Immunol (Berl) . 1986;175:137–139
  101. Yang H , Welsh RM . Induction of alloreactive cytotoxic T cells by acute virus infection of mice . J Immunol . 1986;136:1186–1193
  102. Finberg R , Cantor H , Benacerraf B , Burakoff S . The origins of alloreactivity (differentiation of prekiller cells to viral infection results in alloreactive cytolytic T lymphocytes) . J Immunol . 1980;124:1858–1860
  103. London CA , Lodge MP , Abbas AK . Functional responses and costimulator dependence of memory CD4+ T cells . J Immunol . 2000;164:265–272
  104. Nakamura H , Gress RE . Graft-rejection by cytolytic T-cells—specificity of the effector mechanism in the rejection of allogeneic marrow . Transplantation . 1990;49:453–458
  105. Luznik L , Engstrom LW , Iannone R , Fuchs EJ . Posttransplantation cyclophosphamide facilitates engraftment of major histocompatibility complex-identical allogeneic marrow in mice conditioned with low-dose total body irradiation . Biol Blood Marrow Transplant . 2002;8:131–138
  106. Baker SJ , Reddy EP . Transducers of life and death (TNF receptor superfamily and associated proteins) . Oncogene . 1996;12:1–9
  107. Chinnaiyan AM , O’Rourke K , Yu GL , et al.   Signal transduction by DR3, a death domain-containing receptor related to TNFR-1 and CD95 . Science . 1996;274:990–992
  108. Migone TS , Zhang J , Luo X , et al.   TL1A is a TNF-like ligand for DR3 and TR6/DcR3 and functions as a T cell costimulator . Immunity . 2002;16:479–492
  109. Kaplan MJ , Lewis EE , Shelden EA , et al.   The apoptotic ligands TRAIL, TWEAK, and Fas ligand mediate monocyte death induced by autologous lupus T cells . J Immunol . 2002;169:6020–6029
  110. Nakayama M , Harada N , Okumura K , Yagita H . Characterization of murine TWEAK and its receptor (Fn14) by monoclonal antibodies . Biochem Biophys Res Commun . 2003;306:819–825
  111. Nakayama M , Ishidoh K , Kojima Y , et al.   Fibroblast growth factor-inducible 14 mediates multiple pathways of TWEAK-induced cell death . J Immunol . 2003;170:341–348
  112. Kaplan MJ , Ray D , Yung RL , Richardson BC . Trail and tweak mediate antigen specific and autoreactive macrophage killing by D10 cells . [abstract] FASEB J . 2000;14:A554
  113. Madrigal JA , Scott I , Arguello R , Szydlo R , Little AM , Goldman JM . Factors influencing the outcome of bone marrow transplants using unrelated donors . Immunol Rev . 1997;157:153–166
  114. Nolte MA , Arens R , van Os R , et al.   Immune activation modulates hematopoiesis through interactions between CD27 and CD70 . Nat Immunol . 2005;6:412–418
  115. Hintzen RQ , Lens SM , Beckmann MP , Goodwin RG , Lynch D , van Lier RA . Characterization of the human CD27 ligand, a novel member of the TNF gene family . J Immunol . 1994;152:1762–1773
  116. Bourette RP , De Sepulveda P , Arnaud S , Dubreuil P , Rottapel R , Mouchiroud G . Suppressor of cytokine signaling 1 interacts with the macrophage colony-stimulating factor receptor and negatively regulates its proliferation signal . J Biol Chem . 2001;276:22133–22139
  117. Du Y , Tsai S , Keller JR , Williams SC . Identification of an interleukin-3-regulated aldoketo reductase gene in myeloid cells which may function in autocrine regulation of myelopoiesis . J Biol Chem . 2000;275:6724–6732
  118. Watanabe-Fukunaga R , Brannan CI , Copeland NG , Jenkins NA , Nagata S . Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis . Nature . 1992;356:314–317

PII: S1083-8791(05)00457-X

doi:10.1016/j.bbmt.2005.07.006

Biology of Blood and Marrow Transplantation
Volume 11, Issue 12 , Pages 957-971, December 2005

Article Tools

* Image Usage & Resolution