|Home | About | Journals | Submit | Contact Us | Français|
Hematopoietic stem cell transplantation (HSCT) has been utilized for treatment of many hematologic malignancies, genetic and metabolic disorders, and hemoglobinopathies such as sickle cell disease and thalassemia. It also induces donor-specific tolerance to organ and tissue transplants. The widespread success of HSCT is hampered by the toxicities of immunosuppression and development of graft-versus-host disease (GVHD). The mechanism of induction of transplantation tolerance (reciprocal donor/host) is still an elusive challenge in allogeneic HSCT. An understanding of the mechanisms for induction of tolerance and the critical cells involved in this process has resulted in novel cell-based therapies poised to be translated to clinical application. The focus of this review is those cells of interest.
Bone marrow-derived plasmacytoid dendritic cells induce naïve T cells to differentiate to become antigen-specific regulatory T cells (Treg), creating a milieu for the induction of transplantation tolerance. Recently, CD8+/TCR− facilitating cells (FC), a novel cell population in mouse bone marrow, have been shown to potently enhance engraftment of allogeneic HSC without causing GVHD. The predominant subpopulation of FC resembles plasmacytoid precursor dendritic cells. FC induce antigen-specific Treg in vivo. Notably, FC address one major concern that has prevented the implementation of Treg cell therapy in the clinic: to expand Treg and have them remain tolerogenic in vivo. FC are novel in that they induce an antigen-specific regulatory milieu in vivo. The discovery of FC has opened new alternatives to expanded criteria in bone marrow transplantation that were previously restricted to human leukocyte antigen-matched recipients. The focus of this review is to cover what is currently known about the mechanism of FC action in inducing tolerance and preventing GVHD and hostversus-graft reactivity.
Hematopoietic stem cell transplantation (HSCT) has been utilized in the treatment of hematologic malignancies, hemoglobinopathies, genetic and metabolic disorders. It has also been shown to induce remission in patients with rheumatologic and autoimmune disorders and to induce tolerance to allotransplants.1 However, the complication of graft-versus-host disease (GVHD) and limited availability of HLA-matched donors has limited the widespread clinical application of HSCT. An approach to allow mismatched transplants and avoid GVHD would be a major advance. This review focuses on the discovery of tolerogenic facilitating cells (FC), their similarities with plasmacytoid dendritic cells (pDC), and their translation from bench to bedside and in parallel back to the bench.
Plasmacytoid dendritic cells have been recently identified as a special type of antigen presenting cell (APC).2 They exhibit poor phagocytic properties and are able to both activate and/or inhibit T cells depending upon the environment. They respond to virus and DNA/RNA via toll-like receptors, with production of type 1 interferon. In transplantation, pDC acquire alloantigen in the allograft and then circulate to the peripheral lymph nodes. In the lymph nodes, pDC present alloantigen and induce the generation of CD4+/CD25+/FoxP3+ regulatory T cells (Treg) to control the immune response. As such, they are cells of interest for tolerance induction. One major limitation to the use of pDC in clinical trials has been to expand the cells to obtain sufficient numbers for transplantation yet maintain their tolerogenic features.3
Facilitating cells constitute less than 0.04% of whole bone marrow. They are phenotypically characterized as a CD8α+ population devoid of any conventional αβ- and γδ-TCR surface expression (Fig. 1). Previous studies have shown that the CD8+/TCR− FC are a heterogeneous population of cells.4,5 The major subpopulation shares phenotypic and functional characteristics with plasmacytoid precursor dendritic cells (p-preDC), the immature form of pDC. Both FC and p-preDC respond to CpG-oligodeoxynucleotides (CpG-ODN) and toll-like receptors (TLR). Removal of p-preDC FC completely abrogates facilitation. However, p-preDC FC alone do not facilitate as effectively as FC total. Additional FC subpopulations include NK FC and CD19 FC (Fig. 1). These various FC components may act together to induce and modulate efficient engraftment and maintain the tolerogenic features. The FC effect requires direct cell:cell contact with HSC.6 To further understand the mechanism of FC function, a clear understanding of the interrelated nature of p-preDC and Treg is necessary.
Dendritic cells (DC), which are also derived from hematopoietic bone marrow progenitor cells, can be divided into two main groups: myeloid dendritic cells (mDC) and pDC. Murine pDC display a B220/CD45RA+/CD11clow phenotype and also express Ly6C, Ly49Q, SiglecH and Bst2.7 Human pDC express BDCA-2, BDCA-4, ILT7 and CD123.7 pDC are characterized by their ability to secrete type-I IFN in response to viral infections. pDC express TLR7 and TLR 9 and low levels of major histocompatibility complex (MHC) class II and co-stimulatory molecules. They respond to CpG-ODN stimulation to produce IFNα, IL-12 and acquire the ability to present antigen. pDC play a role in activating both the innate and adaptive immune systems.
Under normal conditions, pDC leave the bone marrow and migrate to T cell-rich areas of secondary lymphoid tissues, as well as through the marginal zones of the spleen.8–10 pDC freshly isolated either from peripheral blood (PB) or lymphoid tissues express MHC class I and II antigens on their surface and low to undetectable levels of the co-stimulatory molecules CD80, CD86 and CD40.11–13 They exist in low number in tissues but accumulate in lymphoid and non-lymphoid tissues under pathological conditions.14 Induction of tolerance can be avoided by decreasing the numbers of pDC or preventing their migration to lymph nodes and reciprocally tolerance can be re-established by adoptive transfer of pDC.
After alloantigen presentation, pDC are capable of long-term induction of Treg in both mature and immature forms. In HSC transplantation, pDC from the donor or the recipient, and even third party, are able to induce tolerance. They migrate to the local lymph node after being stimulated by the graft, acquire alloantigens and convert naïve T cells to Treg. pDC and their precursors mediate tolerance to allografts by inducing Treg development in vivo. IL-10 and transforming growth factor (TGF)β are believed to be the main cytokines involved.15,16 Previous studies indicated that induction of Treg depends upon direct interaction between CD4+ T cells and pDC in the lymph nodes of allograft recipients. pDC require the expression of CD62L in order to migrate to the lymph nodes.17 Other studies found that increased levels of IL-10 and decreased IFNγ plus upregulation of indoleamine-2,3-dioxygenase (IDO) induce the conversion of naïve CD4+ T cells to Treg.18
pDC are hypothesized to promote the establishment of an immunomodulatory environment with cytokine production, ligand and cell marker interaction and/or IDO enzyme regulation. pDC deplete the tryptophan-favoring downstream metabolites through the expression of IDO. T cells act ineffectively under decreased tryptophan. The first observation by Munn et al. that pregnant female mice aborted semi-allogeneic but not syngeneic fetuses after contact with an inhibitor of IDO, 1-methyl-DL-tryptophan (1-MT), showed evidence for a role for tryptophan in tolerance. IDO is expressed in macrophages and trophoblasts. Munn et al. later also showed that B220+/CD8α+ DC induce IDO by ligation of CD80/86 by the cytotoxic T lymphocyte antigen-4 (CTLA4), known to be expressed in Treg. This suppression was inhibited with 1-MT and later demonstrated in an IDO−/− deficient mouse model.20 The CTLA4 expressed on Treg act to induce IDO in the APC that then can restrict an environment with depleted tryptophan. Tryptophan metabolites also act on naïve T cells to induce them to produce FoxP3 (Fig. 2).21
IDO function can be demonstrated in a model of immunoregulation where an immature DC can turn to a mature stage with IDO competency leading to a tolerogenic over immunogenic milieu depending on the cytokines present and environmental stimulation.22,23 IFNγ is a potent inducer of IDO, in the tolerogenic stage IFNγ may have an anti-inflammatory role.24 DC populations express enzymes that catalyze essential amino acids which may represent the requirement for a special environment in and around the tolerated organ.25 Until now, experimental studies demonstrated that IDO plays an important role in tolerance of liver grafts and IDO gene transfer into pancreatic cells prolonged graft survival,26 and gene transfection with IDO prevented allogeneic lung transplant rejection.27
Treg tolerance, originally termed T cell suppression or infectious tolerance, arises from the active suppression of the T cell effector functions necessary for allograft rejection and autoimmunity by a unique “regulatory” T-cell population.28 Treg can be divided into two subsets: natural and adaptive. Naturally-occurring CD4+/CD25+ Treg account for 5% to 10% of all CD4+ T cells. They are not antigen-specific. Natural Treg constitutively express CD25, the alpha chain of the interleukin (IL)-2 receptor.29 The selection of CD4+/CD25+ Treg by antigens expressed in the thymus is suggested by transgenic studies.30 These cells can proliferate in the periphery after exposure to such antigens.31 The fact that natural T reg are not antigen-specific makes them less attractive as a cell-based therapy to induce tolerance.
Adaptive Treg are antigen-specific. They are induced in the periphery from naïve CD4+ T cells and produce specific suppressive cytokines such as IL-10 and TGFβ. The mechanism of action of CD4+/CD25high Treg in suppression is still unclear. Studies however showed that it is probably a cell-contact dependent mechanism with suppressive cytokines like IL-10 and TGFβ involved in vitro and in vivo.32
Donor-derived CD4+/CD25+ Treg prevent lethal GVHD after allogeneic bone marrow transplantation across MHC class I and II barriers in mice.33,34 Donor Treg enhanced engraftment of T-cell depleted bone marrow cells in an MHC-mismatched B6→BALB/c animal model.35 A recent study showed that in vitro culture of donor-derived Treg with irradiated allogeneic recipient splenocytes and IL-2, resulted in expansion of donor-derived Treg (recipient-specific Treg) that can effectively suppress GVHD without compromising immunity after HSC transplantation.36
Studies have indicated that recipient-derived Treg play an important role in induction of transplantation tolerance early after transplantation. Bayer's reported that in a syngeneic mouse model, five weeks after transplantation of T-cell depleted bone marrow cells from B6 (CD45.2) into conditioned syngeneic B6 (CD45.1) recipient mice, the majority of Treg (80%) were recipient (CD45.1)-derived. A similar outcome was observed in an allogeneic model (B6→BALB/c).37 In our own studies using a competitive HSC repopulation assay, purified donor B6 HSC + recipient nonobese diabetic (NOD) mice HSC plus CD8+/TCR− B6 FC were transplanted into NOD mice conditioned with 950 cGy of total body irradiation (TBI). Donor- and recipient-derived CD4+/CD25+/FoxP3+ Treg were detectable in thymus, spleen, and bone marrow at starting two weeks after transplantation. The majority of Treg were recipient-derived (88%–92%). Secondary transfer studies confirmed that these Treg were antigen-specific in that they allowed engraftment of donor-derived HSC but not MHC-disparate third-party HSC (Fig. 3).38 Although Treg were found as early as two weeks post-primary transplantation, they did not exhibit regulation until ≥5 weeks. At that time FoxP3 expression was significantly increased.38 These observations support the hypothesis that both donor Treg and host Treg contribute to enhance HSC engraftment and prevent GVHD. In contrast to Treg from wild type B6, chimeric Treg are significantly more effective in suppressing the proliferation of effector T cells in vitro. The discovery of FC may address one major obstacle that has prevented the use of DC-based therapies: to induce tolerance to maintain the tolerogenic properties in vivo and avoid conversion to immune activation.
The function of p-preDC FC in facilitation and prevention of GVHD has been most extensively studied in a mouse model.5,38,39 CD8+/TCR− FC have been shown to enhance engraftment of, HSC in syngeneic and allogeneic recipients without causing GVHD.5,40–42 FC are a heterogeneous cell population. The largest FC subpopulation is B220+/CD11c+/CD11b− p-preDC. Other subpopulations include CD19+ FC, NK1.1+DX5+ FC and CD3ε+ FC. P-preDC FC display characteristic plasmacytoid morphology and produce interferon-α, interferon-γ, tumor necrosis factor (TNF)α, and other cytokines in response to CpG-ODN stimulation.5 P-preDC FC have capacity to differentiate into mature dendritic cells by upregulating MHC class II, CD86 and CD80 activation markers, as do p-preDC.5,6,43
Studies showed that p-preDC FC were functional to facilitate HSC engraftment.5 B10.BR HSC alone or in combination with p-preDC FC, FC without p-preDC FC or FC total were transplanted into ablated B10 mice. Engraftment was significantly enhanced in recipients of HSC + p-preDC FC compared with the recipients of HSC alone. Facilitation was significantly impaired in recipients of HSC + FC without p-preDC FC. However, durable engraftment was significantly prolonged in recipients of HSC + FC total compared to the recipients of HSC + p-preDC FC (Fig. 4).
FC promote engraftment of purified HSC in allogeneic recipient mice without clinical or histological evidence of GVHD.40,44 In a murine semi-allogeneic parent→F1 combination model of lethal GVHD, conditioned recipients were injected with donor HSC alone or in combination with splenic αβ-TCR+ T cells, bone marrow-derived CD8+/TCR+ T cells or CD8+/TCR− FC. Lethal GVHD occurred in recipients of HSC + splenic T cells or bone marrow-derived T cells. Recipients of HSC + FC did not show significant clinical or histologic evidence of GVHD. HSC + FC recipients exhibited increased expression of transforming growth factor (TGFβ) and the induction of the regulatory T-cell genes including CTLA4, GITR and FoxP3.44
To test whether FC-mediated facilitation of allogeneic HSC engraftment and GVHD-avoidance occurred induction of Treg in vivo, the production of Treg were evaluated after HSC + FC transplantation in B6→NOD mouse model.38 In two weeks, the highest numbers of Treg were present in spleen and thymus, with absolute numbers increasing in the peripheral blood, spleen, and bone marrow over time. These phenotypic Treg acquired FoxP3 expression by week three and were functional at ≥5 weeks. Removal of p-preDC FC from FC resulted in impaired generation of Treg, suggesting that p-preDC FC play a critical role in induction of regulatory T cells in vivo.
Another mechanism by which FC act on the more primitive subpopulations of HSC is to promote survival and function of HSC via production of physiologically relevant levels of TNFα to the HSC as part of the hematopoietic microenvironment, representing a trophic type of “nurse cell.” FC from TNFα deficient mice (TNFα−/−) exhibit impaired facilitation in vivo and loss of function in vitro to enhance HSC clonogenicity.6 Co-culture of FC with HSC induces production of low levels of TNFα by FC. There is a critical requirement for TNFα in the FC effects on HSC which require direct cell: cell contact between FC and HSC. FC did not produce TNFα if separated from HSC in transwell assays. The effect of FC is not due to an increase in proliferation or differentiation of HSC, because the HSC co-cultured with FC retained the ability to generate more multipotent colonies in vitro, as well as to promote durable engraftment of HSC in vivo.
In summary, pDC, FC and p-preDC FC have immunogenic and tolerogenic properties and effectively induce Treg through cytokine release, interaction with ligands and cellular markers and/or IDO enzyme upregulation. FC may address one major concern that has slowed the translation of tolerogenic Treg and DC cell-based therapies to the clinic: how to maintain the tolerogenic features in vivo after expansion and/or infusion. The full understanding of the mechanism of induction and maintenance of tolerance has and will provide novel cell-based tolerogenic therapeutic approaches that are currently in early stages of translation into the clinic for treatment of many chronic conditions, including the induction of tolerance to renal allografts.
The authors thank Haval Shirwan for review of the manuscript and helpful comments; Carolyn DeLautre for manuscript preparation; the staff of the animal facility for outstanding animal care.
This work was supported in part by NIH R01 DK069766 and NIH 5RO1 HL063442 and JDRF 1-2006-1466, The Department of the Army, Office of Army Research. [Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Office of Army Research. This publication was made possible by Award No. W81XWH-07-1-0185, No. W81XWH-09-2-0124, and No. WH1XWH-10-1-0688 from the Office of Army Research], the Commonwealth of Kentucky Research Challenge Trust Fund, and the W.M. Keck Foundation.
S. Ildstad is the founding scientist and director of Regenerex, a biotech start-up company; it has not been capitalized.