OBJECTIVE—Plasmacytoid precursor dendritic cell facilitating cells (p-preDC FCs) play a critical role in facilitation of syngeneic and allogeneic hematopoietic stem cell (HSC) engraftment. Here, we evaluated the phenotype and function of CD8+/TCR FCs from NOD mice.

RESEARCH DESIGN AND METHODS—The phenotype of CD8+/TCR FCs was analyzed by flow cytometry using sorted FCs from NOD, NOR, or B6 mice. The function of NOD FCs was evaluated by colony-forming cell (CFC) assay in vitro and syngeneic or allogeneic HSC transplantation in vivo.

RESULTS—We report for the first time that NOD FCs are functionally impaired. They fail to facilitate engraftment of syngeneic and allogeneic HSCs in vivo and do not enhance HSC clonogenicity in vitro. NOD FCs contain subpopulations similar to those previously described in B6 FCs, including p-preDC, CD19+, NK1.1+DX5+, and myeloid cells. However, the CD19+ and NK1.1+DX5+ subpopulations are significantly decreased in number in NOD FCs compared with disease-resistant controls. Removal of the CD19+ or NK1.1+DX5+ subpopulations from FCs did not significantly affect facilitation. Notably, Flt3 ligand (FL) treatment of NOD donors expanded FC total in peripheral blood and restored facilitating function in vivo.

CONCLUSIONS—These data demonstrate that NOD FCs exhibit significantly impaired function that is reversible, since FL restored production of functional FCs in NOD mice and suggest that FL plays an important role in the regulation and development of FC function. FCs may therefore be linked to diabetes pathogenesis and prevention.

We were the first to identify a CD8+/TCR facilitating cell (FC) population in mouse bone marrow that facilitates hematopoietic stem cell (HSC) engraftment across major histocompatibility complex (MHC) barriers without causing graft-versus-host disease (GVHD) (13). FCs are a heterogeneous cell population, with the predominant subpopulation resembling B220+/CD11c+/CD11b plasmacytoid precursor dendritic cells (p-preDC) (4). p-preDC FCs display characteristic plasmacytoid morphology and produce interferon (IFN)-α, tumor necrosis factor (TNF)-α, and other cytokines in response to CpG-oligodeoxynucleotides (CpG-ODN) (4). p-preDC FCs also have the capacity to differentiate into mature dendritic cells (DCs) by upregulating MHC class II, CD86, and CD80 activation markers (4). Removal of p-preDC FCs completely abrogates facilitation, suggesting that p-preDC FCs play a critical role in facilitation (4).

NOD mice develop spontaneous autoimmune diabetes due to defects in both peripheral and central tolerance mechanisms (5). Several regulatory defects have been described in NOD mice, including islet-reactive T-cells that escape deletion, impaired generation of regulatory T-cells (Treg), inhibitory cytokines, aberrant professional antigen-presenting cell (APC) function, and low levels of NK cell activity (6,7). B-cells also contribute to the development of diabetes in NOD mice in their role as professional APCs (8). The role of p-preDC in pathogenesis of autoimmune disease has been addressed (9). Several groups reported abnormalities in DC phenotype and function in human type 1 diabetes and NOD mice (10,11). An understanding of which specific DC subsets play a critical role in maintenance of self-tolerance and prevention of diabetes may allow novel cell-based therapies to be used in the clinic for disease prevention.

In the present study, we evaluated the phenotype and function of NOD FCs. We report for the first time that NOD FCs are functionally impaired in vivo and in vitro. As in disease-resistant controls, the B220+/CD11c+/CD11bp-preDC FC subpopulation represents the major subpopulation of CD8+/TCR FCs in NOD bone marrow. The CD19+ or NK1.1+DX5+ FC subpopulations were significantly decreased in NOD FCs compared with those from B6 or MHC-congenic diabetes-resistant NOR mice (12). NOR FCs significantly enhanced engraftment of NOR HSCs. In striking contrast, NOD FCs were completely impaired in function and did not facilitate HSC engraftment. Similarly, NOD FCs were impaired in function in vitro. NOR p-preDC FCs were more efficient at granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin (IL)-6, macrophage inflammatory protein (MIP)-1α/CCL3, Rantes/CCL5, and TNF-α production in response to CpG compared with NOD p-preDC FCs. Removal of the CD19+ or NK1.1+DX5+ FC subpopulations did not significantly impair facilitation. Notably, FL treatment of NOD mice expanded FCs in peripheral blood (PB), and these FL-PB-FCs significantly enhanced engraftment of HSCs. The fact that FL treatment restored the function of NOD FCs suggests that FL may represent a key cytokine for the development and function of FCs. FCs may therefore be a critical link in diabetes pathogenesis and prevention and may provide a novel cell-based approach to restore self-tolerance and regulation in treatment of type 1 diabetes.

Mice.

Four- to six-week-old NOD mice (male and female; Taconic Laboratories, Germantown, NY), female nonobese resistant (NOR) mice, male C57BL/6 mice, and female C57BL/10SnJ mice (Jackson Laboratory, Bar Harbor, ME) were used. Animals were housed in the barrier facility at the Institute for Cellular Therapeutics (Louisville, KY) and were cared for according to National Institutes of Health animal care guidelines.

Antibodies.

All monoclonal antibodies (mAbs) used in this study were purchased from BD Biosciences (San Diego, CA). c-Kit+Sca-1+Lin (HSC) sorting experiments used the following mAbs: stem cell antigen-1 (Sca-1) phycoerythrin (PE), c-Kit allophycocyanin (APC), and the lineage panel consisting of the following: CD8α fluorescein isothiocyanate (FITC), Mac-1 FITC, B220 FITC, Gr-1 FITC, γδ-TCR FITC, and β-TCR FITC. CD8+/TCR FC sorting experiments used β-TCR FITC, γδ-TCR FITC, and CD8α PE. CD8+/TCR/CD19 sorting experiments used CD8α PE, β-TCR FITC, γδ-TCR FITC, and CD19 APC. CD8+/TCR/NK1.1DX5 cells were sorted by using CD8α APC, β-TCR FITC, γδ-TCR FITC, NK1.1 PE, and DX5 PE.

Sorting of HSCs and FCs.

HSCs and FCs were isolated from bone marrow by multiparameter, live sterile cell sorting (FACSVantage SE; Becton Dickinson, Mountainview, CA), as previously described (4). Briefly, bone marrow was isolated and collected in a single cell suspension at a concentration of 100 × 106 cells/ml in sterile cell sort media, containing sterile 1× Hank's balanced salt solution without phenol red, 2% heat-inactivated fetal bovine serum, 10 mmol · l−1 · ml−1 HEPES buffer, and 30 μl/ml gentamicin (Gibco, Grand Island, NY). Directly labeled mAbs were added at saturating concentrations, and the cells were incubated for 30 min and washed with cell sort media.

Phenotypic analysis of sorted CD8+/TCR FCs.

Sorted FCs (purity was ≥95%) were incubated with Fc receptor block (anti-CD16/CD32) before staining with lineage-specific markers: anti-CD11 FITC, CD11b APC, CD14 FITC, NK1.1 FITC, DX5 FITC, B220 PerCP, Gr-1 APC, and CD19 APC, as previously described (4). Sorted FCs were analyzed for p-preDC FCs as B220+/CD11c+/CD11b using Cell Quest Software (Becton Dickinson).

DC maturation and cytokine production.

FCs were cultured alone or with 1 μmol/l TLR-9 ligand CpG-ODN 1668 (TCCATGACGTTCCGATGCT) (Gibco BRL Custom Primers) (13) for 18 h. Supernatants were assayed for cytokines by Linco Diagnostic (St Charles, MO) using LINCOplex Multiplex immunoassay, and cells were stained with anti-CD80, anti-CD86, or anti-MHC class II I-Ad (39-10-8) FITC mAbs, with appropriately matched isotype controls.

HSC and/or FC transplantation.

In the syngeneic model, recipients were conditioned with 950 cGy total body irradiation (TBI) from a Cesium source (Nordion, ON, Canada) and transplanted with 500 HSCs ± 30,000 FC populations by tail vein injection ≥6 h after irradiation (14). In the allogeneic model, recipients conditioned with 950 cGy TBI were transplanted with 5,000 HSCs ± 30,000 FL-PB FCs (4).

Colony-forming cell assay.

HSCs were cultured at a 1:2 ratio with or without FCs in methylcellulose containing mouse growth factors (MethoCult GF M3434; StemCell Technologies, Vancouver, BC, Canada) in duplicate at 37°C in 5% CO2 and a humidified atmosphere (13). After 14 days, colonies containing more than 50 cells were scored.

Chimerism testing.

Engraftment of donor cells was evaluated by PBL typing using three-color flow cytometry, as previously described (15).

FC morphology.

Wright-Giemsa staining was performed on cytospins of 100,000 FCs after being fixed in methanol. Slides were examined for dendritic morphology under optical microscopy.

Statistical analysis.

Experimental data were evaluated for significant differences using Student's t test; P < 0.05 was considered significant. Graft survival was calculated according to the Kaplan-Meier method (4).

NOD FCs exhibit specific and significant differences in subpopulations compared with normal controls.

In normal mice, the CD8+/TCR FC (FC total) population is heterogeneous, with the dominant subpopulation phenotypically resembling p-preDC (B220+/CD11c+/CD11b) (4). Smaller percentages of B-cell (CD19+), NK cell (NK1.1+DX5+), granulocyte (Gr-1+), and monocyte (CD14+) subpopulations are also present in FC total from normal mice (4). We found that NOD and NOR FCs are comprised of similar distinct heterogeneous subpopulations (Fig. 1A and C) and show a heterogeneous morphology with Wright-Giemsa staining under light microscopy (Fig. 1B and D). p-preDC FCs represent the major CD8+/TCR FC subpopulation in all strains (female and male NOD mice, female NOR mice, and male B6 mice) examined (Fig. 1E). The B220+/CD11c+ FC population in female and male NOD mice is significantly increased compared with control NOR or B6 mice (Fig. 1H; P < 0.05). The B220/CD11c+/CD11b+ subset is significantly decreased compared with NOR mice (Fig. 1F; P < 0.007). As previously shown, the dominant cell population in CD19+ FC is pre–B-cells (B220+/CD11c/intracytoplasmic IgM+) (4). Of the female NOD FCs, 14% were CD19+, which is significantly decreased compared with NOR and B6 mice (Fig. 1I, P < 0.05). Approximately 0.27% of NOD FCs are CD19+/CD11c+/B220+ cells (Fig. 1G), which is not significantly different than the control strain. DCs with a similar phenotype from normal LN and spleen have been shown to function as p-preDC (16). Of the female NOD FC total, B220+/NK1.1+DX5+ and B220+/Gr-1+ populations were significantly decreased compared with B6 FCs (Fig. 1K and J). The B220+/CD14+ population was not significantly different in all strains examined (Fig. 1L).

FCs produce cytokines and upregulate activation markers after stimulation.

We evaluated whether NOD FCs resemble NOR FCs in response to CpG-ODN stimulation. CD86 was upregulated on NOR FCs, whereas CD80 and class II expression was similar in the absence of CPG stimulation (Fig. 2A and B). However, whereas CD86 was upregulated on NOD FCs, CD80 expression was markedly decreased with stimulation (Fig. 2A and B). After CpG-ODN stimulation, the majority of NOR FCs were in a more activated state compared with NOD FCs, as evidenced by their dendritic morphology (Fig. 2C, right panel). In contrast, NOD FCs did not exhibit a mature morphology after CpG treatment (Fig. 2C, left panel).

We also examined chemokine and cytokine production by NOD and NOR p-preDC FCs after CpG-ODN stimulation. In the presence of CpG-ODN, p-preDC FCs produced more MIP-1α/CCL3, RANTES/CCL5, interferon-γ–induced protein (IP)-10, IL-6, and TNF-α compared with the level of those in absence of stimulation (Fig. 2D). Notably, p-preDC FCs from NOR mice produced higher amounts of IL-6 (5×), RANTES/CCL5 (3.5×), MIP-1α/CCL3 (2.1×), and TNF-α (1.9×) compared with NOD p-preDC FCs (Fig. 2D). In addition, we found that NOR p-preDC FCs produce GM-CSF more efficiently in response to CpG-ODN stimulation, whereas NOD p-preDC FCs do not (Fig. 2D). Taken together, these data demonstrate that NOD p-preDC FCs are impaired in their ability to produce chemokines and cytokines after CpG-ODN stimulation.

NOD CD8+/TCR FC function is significantly impaired in vivo.

We next examined the ability of NOD FCs to facilitate HSC engraftment using a syngeneic model (13,14). NOD recipients were ablatively conditioned with 950 cGy TBI and reconstituted with 500 HSCs ± 30,000 FCs sorted from NOD donors. Only 4 of 13 (31%) recipients of HSCs plus FCs and 4 of 17 (24%) recipients of HSCs engrafted survived up to 130 days (Fig. 3B). In striking contrast with normal controls (4), NOD FCs did not improve HSC engraftment in NOD recipients, as evidenced by the similar engraftment of HSCs with FCs compared with the HSCs alone (P = 0.579).

We then examined the function of NOR FCs. NOR mice are MHC-congenic to NOD mice, but do not develop diabetes (Fig. 3A) (5). Five (31%) of 16 recipients of HSCs alone engrafted and survived up to 130 days. In contrast, 70% (7 of 10) of recipients of HSCs plus FCs engrafted long term with survival over 130 days (Fig. 3C). Therefore, NOR FCs significantly enhance engraftment of HSCs in limiting numbers of HSCs (P = 0.029).

To assess whether NOR FCs facilitate engraftment of NOD HSCs, 500 NOD HSCs plus 30,000 NOR FCs (n = 15) were transplanted into NOD recipients conditioned with 950 cGy. All recipients of HSCs alone expired before 130 days after transplantation (Fig. 3D). In striking contrast, the majority of (11 of 21) animals transplanted with NOD HSCs and NOR FCs survived over 130 days, demonstrating that NOR FCs also facilitate engraftment of NOD HSCs (Fig. 3D). As expected, NOD FCs did not enhance engraftment of NOR HSCs (n = 16; Fig. 3E).

NOD CD8+/TCR FCs failed to promote generation of colonies from HSCs.

To evaluate the function of NOD FCs in vitro, we tested them using the CFC assay, which enumerates the number of mono-lineage and multi-lineage colonies generated by HSCs (13). NOR HSCs cocultured with NOR FCs for 18 h, then cultured in methylcellulose for 14 days, generated significantly more colonies compared with NOR HSCs alone (n = 3; P = 0.011; Fig. 4B). In contrast, NOD FCs failed to enhance colony formation when cultured with NOD HSCs (n = 3; P = 0.422; Fig. 4C). Figure 4A shows representative appearance of CFC-GM and CFC–granulocyte-erythroid-macrophage-megakaryocyte (GEMM) for NOD HSC. FCs alone did not generate colonies (Fig. 4B and C).

Removal of CD19+ or NK1.1+DX5+ cells from FCs does not significantly impair facilitation.

To define the function of CD19+ or NK1.1+DX5+ FC subpopulations, HSCs, CD8+/TCR, or CD8+/TCR/CD19 cells were sorted from NOR mice and tested in the syngeneic assay for in vivo facilitation (Fig. 5A). Four of nine (44%) recipients of HSCs plus CD8+/TCR/CD19 FCs exhibited long-term engraftment and survived at least 110 days (Fig. 5B). Five of eight (63%) animals given HSC + CD8+/TCR FCs survived up to 110 days (Fig. 5B). There was no significant difference in survival between the HSC plus FC total group compared with the HSC plus FC, from which CD19+ FCs had been depleted (P = 0.49). Four of 17 (23%) of recipients transplanted with HSC alone survived up to 110 days (Fig. 5B). These data suggest that the CD19+ subpopulation may not play an important role in facilitation and therefore that the low numbers of these cells was not the cause of ineffective facilitation by NOD FCs.

We next evaluated the contribution of the NK1.1+DX5+ FC subpopulation to total FC function. Donor NK cells have the potential to promote HSC engraftment and suppress GVHD in allogeneic transplantation (17). Our previous data showed that ∼4–6% of FCs are NK1.1+DX5+ cells (4). In NOD mice, 1–1.5% of FCs express NK1.1+DX5+. To test the contribution of the NK1.1+DX5+ FC subpopulation to FC function, HSCs, CD8+/TCR FCs, and CD8+/TCR/NK1.1DX5 cells were sorted from the marrow of B6 donors; 58% (7/12) of recipients of HSC plus CD8+/TCR FCs survived up to 110 days, and 42% (5/12) of HSC plus CD8+/TCR/NK1.1DX5 recipients survived over 110 days (Fig. 5C). Survival of both groups was significantly enhanced compared with the group that received HSCs alone (P = 0.009).

FL-mobilized NOD FCs facilitate HSC engraftment in allogeneic recipients.

We previously reported that FL treatment of NOD mice restored production of defective mature myeloid DCs and plasmacytoid DCs in spleen and pancreatic lymph nodes and significantly increased Treg in pancreatic lymph nodes (18). This was associated with a significant delay in diabetes progression. To test whether FL treatment can restore the function of NOD FCs, we evaluated the phenotype and function of FL-PB FCs. NOD mice were treated with FL for 10 days. FCs were sorted from peripheral blood, and sorted FCs were stained with B220, CD11c, CD19, NK1.1DX5, and CD11b mAbs. There was a significant increase in B220/CD11c+/CD11b+ DC and NK1.1+DX5+ subpopulations in FL-PB FCs (Fig. 6A). The percentage of CD19+ FCs and p-preDC FCs remained at the same levels as untreated NOD bone marrow FCs (Fig. 6A).

We next evaluated whether FL treatment can restore the facilitating function of NOD FCs. FL mobilized NOD PB FCs were in a more activated state than untreated NOD bone marrow FCs, as evidenced by their dendritic morphology (Figs. 6B and 1B). To test function of FL-PB FCs, HSCs were sorted from bone marrow of untreated NOD mice and FCs from the PB of FL-treated NOD mice. Conditioned B10 recipients received 5,000 HSCs plus 30,000 FL-PB FCs. Control mice were transplanted with 5,000 HSCs ± 30,000 FCs from bone marrow of untreated NOD mice. FL-PB FCs significantly enhanced engraftment of HSCs, as evidenced by 63% of recipients (n = 8) that received HSC plus FL-PB FCs surviving 120 days (Fig. 6C). A total of 13% and 20% of recipients of HSCs alone (n = 9) or HSCs plus FCs (n = 8) from untreated NOD mice survived over 120 days, respectively (Fig. 6C).

To confirm that recipients of HSCs plus FL-PB FCs exhibited durable engraftment and multi-lineage reconstitution, animals were followed for >4 months. Three-color flow cytometric analysis was performed. Recipients of HSCs alone showed the presence of cells of donor origin including DCs (CD11c), macrophage (Mac-1) and granulocytes (Gr-1), and NK cells (NK1.1DX5) and the presence of low levels of T-cells (CD8, CD4, αβ-TCR + γδ-TCR) and B-cells (B220) (Fig. 6D). In contrast, recipients of HSCs plus FL-PB FCs showed donor chimerism for multi-lineages, including T-cells, B-cells, NK cells, macrophages, and granulocytes (Fig. 6E).

CD8+/TCR FCs play an important role in facilitating engraftment of syngeneic and allogeneic HSCs (1,14,19). CD8+/TCR FCs (FC total) are heterogeneous, including the following phenotypic subpopulations: p-preDC, B-cells, NK cells, and monocytes (4). The p-preDC FC subpopulation represents the majority of FCs and plays a critical role in this complex network (4). Removal of the p-preDC component from the FCs results in complete loss of facilitation, confirming that p-preDCs are the primary component in facilitation (4). However, p-preDC FCs facilitate HSC engraftment significantly less efficiently than total FCs (4), suggesting that other FC subpopulations expressing the B-cell and NK cell phenotypes may play a collaborative role in facilitation. FCs induce generation of Treg in vitro and in vivo (20,21). More recently, Treg have been shown to enhance engraftment of HSCs in syngeneic recipients (22). The fact that the FC maintains its tolerogenic function in vivo addresses one of the major concerns regarding DC-based therapies: how to avoid immune activation and maintain tolerogenicity after infusion in vivo (23). As such, FCs may offer a novel cell-based therapeutic approach to induce tolerance in the clinic for treatment of autoimmune disorders.

In the present study, we evaluated the function of FCs in the context of a mouse model for type 1 diabetes, a systemic autoimmune disease (5). We found that NOD FCs, as a heterogeneous population, share phenotypic characteristics similar to those previously described for wild-type FCs (4). However, the percentages of both CD19+ and NK1.1+DX5+ cells in NOD FCs were significantly decreased compared with NOR or B6 FCs. NOD FCs were functionally impaired and failed to facilitate engraftment of HSCs in both syngeneic and allogeneic recipients, as well as in an in vitro assay for FCs. Notably, FL treatment expanded PB FCs in NOD mice and restored the ability of NOD FCs from the PB to facilitate engraftment of allogeneic HSCs. We propose that the defective function of NOD FCs may be due to an abnormal activation status of the p-preDC FC subpopulation or the presence of impaired function of a collaborative subpopulation in FCs such as B-cells or NK cells. This hypothesis offers an attractive explanation for the mechanism by which FCs enhance HSC engraftment in vivo and induce tolerance.

The majority of FCs are B220+/CD11c+/CD11b and share characteristic features of p-preDC, including morphology and phenotype, secretion of similar cytokines and chemokines, and maturation after stimulation with CpG-ODN (4). However, their distinct differences, including the fact that FCs produce IL-10 whereas p-preDC do not, may offer an understanding as to how p-preDC FCs remain tolerogenic in vivo. p-preDCs play an important role in regulating innate and adaptive immune responses. They selectively express toll-like receptors (TLR)-7 and TLR-9 and are specialized in rapid secretion of type 1 IFN after viral stimulation (24). p-preDC can promote the function of NK cells, B-cells, T-cells, and myeloid DCs through type 1 IFN during an antiviral immune response and differentiate into unique types of mature DCs, which directly regulate the function of T-cells and thus link innate and adaptive immune responses. We were the first to show that p-preDC have the potential to facilitate engraftment of HSCs (4). p-preDCs also induce tolerance to heart allografts (25). Several studies have demonstrated an association between the pathogenesis of autoimmune diseases (systemic lupus erythematosus, Sjogren's Syndrome, and dermatomyositis) and defective function of IFN-α producing p-preDC (2628). NOD mice exhibit an abnormal DC phenotype and function (10,11). We show here for the first time that FCs from NOD mice exhibit a functional defect in facilitating HSC engraftment in vivo and impaired function in vitro as well. However, the fact that FL treatment of NOD donors results in production of functional FCs implies that the defect is probably not cell intrinsic, but rather due to a lacking signal or activated state.

FL plays a critical role in the development of p-preDC in humans and mice (29,30). The ability of FL to promote p-preDC development in vivo was confirmed by experiments showing that administration of FL into human volunteers led to an increase in the number of PB p-preDCs in humans and that FL transgenic mice have increased numbers of p-preDCs, where FL-deficient mice have fewer p-preDCs (31). Our own work showed that treatment of prediabetic NOD mice with FL significantly decreased insulitis and progression to diabetes and was associated with a significant increase in myeloid DCs, plasmacytoid DCs, and Treg (18). When DCs from NOD mice bone marrow are treated with nuclear factor κB–specific ODN in vitro, administration of DCs into NOD mice can effectively prevent the onset of diabetes (32). FL is also a key cytokine for FC generation and expansion, as evidenced by FL–bone marrow culture and the mobilization of FC in peripheral blood (4). FL-mobilized PB FCs promote the establishment of donor chimerism and tolerance induction. In the present study, we showed that FL treatment can restore the function of NOD FCs, demonstrating that FL can promote that development and function of FCs in NOD mice.

The importance of myeloid DCs on their ability to trigger B-cell growth and differentiation has been addressed (33). A recent study showed that the p-preDCs regulate B-cell function by producing IFN-α and IL-6, which thereby induces B-cell differentiation (34). Depletion of p-preDC from human blood mononuclear cells abrogates the secretion of immunoglobulins in response to influenza virus and affects the differentiation of activated B-cells into plasma cells through the secretion of IFN-αβ and IL-6 (34). Several experimental models in animals have shown that B-cells are involved in inducing T-cell tolerance in vivo (35,36). The role of B-cells in autoimmune diseases may occur through several mechanistic pathways that include self-reactive antibodies, secretion of inflammatory cytokines, participation in antigen presentation, and augmentation of T-cell activation (37). In NOD mice, the B-cell subpopulation (CD19+) within the total FC population is present at a much lower frequency compared with NOR and B6 controls. The function of the CD19+ FC subpopulation remains elusive. Removal of this subpopulation from normal donors did not impair facilitation. It is formally possible that the CD19+ FC subpopulation does not contribute to FC function or that there is redundancy in the system that is contributed from another FC collaborative subpopulation.

NOD FCs also contain significantly lower numbers of NK1.1+DX5+ cells compared with B6 or NOR mice. It was unclear whether the failure of FC function was due to decreased numbers of the NK FC subpopulation. NK cells have been shown to play a major role in regulating early engraftment of allogeneic bone marrow cells, both by activation as well as inhibition of immune responses (38). Adoptive transfer of activated NK cells early after transplantation inhibits GVHD and promotes graft-versus-tumor (GVT) in the mouse model (39). Recently, various studies have investigated the interaction between NK cells and plasmacytoid DCs, suggesting that the cross-talk between NK cells and DCs leads to NK cell activation and DC maturation (40). NOD mice exhibit an abnormally low level of NK cell activity (7,41) and a defect in NK/T-cells (42). To evaluate whether NK FCs were involved in facilitation of HSC engraftment, we performed allogeneic HSC transplantation (B6 → C3H) using FCs depleted of the NK FC subpopulation. There was no difference in engraftment in mice that receive HSC plus FC total vs. FCs depleted of NK FCs, suggesting that NK FCs did not contribute to facilitation.

The interaction of DCs and Treg in the regulation of innate and adaptive immune responses has been reported (43). The consequences of DC interactions with Treg depend on the phenotype and maturation status of DCs. Mature DCs have the unique ability to promote natural Treg expansion, which limits immune responses to self-antigen (43,44), whereas immature DCs induce the generation of adaptive Treg, which control the immune response to non–self-antigen (45). Mice lacking CD28 or its ligands have decreased numbers of CD4+/CD25+ cells (46). These data suggest that both CD86 and CD80 are capable of delivering the signals that promote Treg generation. It has been shown that on the NOD mouse background, genetic ablation of CD86 results in only a subtle decrease in Treg numbers (20–30% reduction). The remaining Treg were largely dependent on CD80, as the additional injection of anti-CD80 blocking mAbs substantially decreased the number of CD4+/CD25+ cells (∼70% reduction). These data support a more important role for CD80 than CD86 in generation of Treg (47). Notably, we found that NOD FCs exhibit significantly impaired upregulation of CD86 after stimulation with CpG. Similarly, and in contrast with FCs from diabetes-resistant donors, they failed to produce GM-CSF and produced significantly lower levels of IL-6 after CpG stimulation. Several groups have reported that NOD mice exhibited reduced Treg frequency (48,49), and their impaired suppressive function has been linked to diabetes pathogenesis (50). The fact that wild-type FCs can induce the generation of Treg, but only in the presence of CpG-ODN (21), and that they are impaired in function in diabetes-prone NOD mice suggests that FCs may also play a distinct role in diabetes pathogenesis.

In conclusion, our data reveal a novel defect in NOD FC function that is restored by treatment with FL. Our data suggest the critical role of FL in developing and maintaining the function of FCs. These findings may have clinical implications for the treatment of type 1 diabetes and possibly other autoimmune disease states.

FIG. 1.

NOD FC total is a heterogeneous population. A and C: CD8+/TCR FCs were sorted from NOD bone marrow, blocked using the anti-Fc receptor Abs and stained with anti-B220, anti-CD19, anti-CD11b, anti-NK1.1, anti-DX5, anti-Gr-1, and anti-CD14 mAbs. Flow cytometric profiles are representative of at least three experiments in NOD or NOR mice. B and D: Morphology of sorted NOD or NOR CD8+/TCR FCs were examined after Wright-Giemsa staining with optical microscopy. E–L: Comparison of phenotype of FCs from NOD, NOR, and B6 bone marrow. Represented are the means ± SD of three independent experiments. *P < 0.05; **P < 0.007.

FIG. 1.

NOD FC total is a heterogeneous population. A and C: CD8+/TCR FCs were sorted from NOD bone marrow, blocked using the anti-Fc receptor Abs and stained with anti-B220, anti-CD19, anti-CD11b, anti-NK1.1, anti-DX5, anti-Gr-1, and anti-CD14 mAbs. Flow cytometric profiles are representative of at least three experiments in NOD or NOR mice. B and D: Morphology of sorted NOD or NOR CD8+/TCR FCs were examined after Wright-Giemsa staining with optical microscopy. E–L: Comparison of phenotype of FCs from NOD, NOR, and B6 bone marrow. Represented are the means ± SD of three independent experiments. *P < 0.05; **P < 0.007.

FIG. 2.

Expression of activation markers on FC. A: Sorted CD8+/TCR FCs (100,000) from NOD or NOR bone marrow were cultured with medium or CpG-ODN for 18 h and stained with anti-MHC class II I-Ad, CD86, or CD80 FITC–labeled mAbs or isotype controls. The data shown are representative of three experiments. B: Level of expression of activation markers on FCs with or without CpG-ODN stimulation. The results are the percent of CD8+/TCR FC total from three separate experiments. C: The morphology of FCs was examined using Wright-Giemsa staining under optical microscopy after CpG-ODN stimulation. D: p-preDC FCs were cultured with CpG-ODN or medium only. Culture cell-free supernatants were collected after 18 h and MIP-1α/CCL3, RANTES/CCL5, IP-10, IL-6, GM-CSF, and TNF-α were measured by LINCO plex multiplex immunoassay. Data showed an average of two separate experiments.

FIG. 2.

Expression of activation markers on FC. A: Sorted CD8+/TCR FCs (100,000) from NOD or NOR bone marrow were cultured with medium or CpG-ODN for 18 h and stained with anti-MHC class II I-Ad, CD86, or CD80 FITC–labeled mAbs or isotype controls. The data shown are representative of three experiments. B: Level of expression of activation markers on FCs with or without CpG-ODN stimulation. The results are the percent of CD8+/TCR FC total from three separate experiments. C: The morphology of FCs was examined using Wright-Giemsa staining under optical microscopy after CpG-ODN stimulation. D: p-preDC FCs were cultured with CpG-ODN or medium only. Culture cell-free supernatants were collected after 18 h and MIP-1α/CCL3, RANTES/CCL5, IP-10, IL-6, GM-CSF, and TNF-α were measured by LINCO plex multiplex immunoassay. Data showed an average of two separate experiments.

FIG. 3.

NOD FCs are functionally impaired in vivo. HSCs (500) and FCs (30,000) were sorted from donor marrow and transplanted into syngeneic recipients. HSCs and FCs were mixed before injection. A: Haplotype pedigree of NOD and NOR mice. B: Survival of NOD recipients of NOD HSCs with or without NOD FCs. C: Survival of NOR recipients of HSCs and FCs or HSCs alone from NOR donors. D: Survival of NOD recipients of HSCs from NOD mice with or without additional FCs from NOR mice. E: Survival of NOR recipients of HSCs and FCs or HSCs alone, HSCs from NOR mice, and FCs from NOD mice. Results are from three to five separate transplant experiments for each experimental group.

FIG. 3.

NOD FCs are functionally impaired in vivo. HSCs (500) and FCs (30,000) were sorted from donor marrow and transplanted into syngeneic recipients. HSCs and FCs were mixed before injection. A: Haplotype pedigree of NOD and NOR mice. B: Survival of NOD recipients of NOD HSCs with or without NOD FCs. C: Survival of NOR recipients of HSCs and FCs or HSCs alone from NOR donors. D: Survival of NOD recipients of HSCs from NOD mice with or without additional FCs from NOR mice. E: Survival of NOR recipients of HSCs and FCs or HSCs alone, HSCs from NOR mice, and FCs from NOD mice. Results are from three to five separate transplant experiments for each experimental group.

FIG. 4.

NOD FCs fail to promote HSC colony formation in vitro. CFC assays were performed on sorted HSCs plus FCs from NOD or NOR mice. A: Representative appearance of colonies at 14 days. B: HSCs (2,000) and FCs (4,000) were sorted from bone marrow of NOR mice; results are expressed as CFC frequency per 1,000 HSCs from three different experiments. C: HSCs and FCs from NOD mice; data represent three different experiments. Each rectangle (▪) represents one individual sample. The dotted lines link samples from the same experiments. Averaged data from three experiments are presented as means ± SE. (Please see http://dx.doi.org/10.2337/db08-0356 for a high-quality digital representation of this figure.)

FIG. 4.

NOD FCs fail to promote HSC colony formation in vitro. CFC assays were performed on sorted HSCs plus FCs from NOD or NOR mice. A: Representative appearance of colonies at 14 days. B: HSCs (2,000) and FCs (4,000) were sorted from bone marrow of NOR mice; results are expressed as CFC frequency per 1,000 HSCs from three different experiments. C: HSCs and FCs from NOD mice; data represent three different experiments. Each rectangle (▪) represents one individual sample. The dotted lines link samples from the same experiments. Averaged data from three experiments are presented as means ± SE. (Please see http://dx.doi.org/10.2337/db08-0356 for a high-quality digital representation of this figure.)

FIG. 5.

Removal of CD19+ or NK1.1+DX5+ FC subpopulations did not significantly affect facilitation. A: BM cells were stained with anti-CD8 PE or APC, anti–TCR-β FITC, anti–TCR-γδ FITC, and anti-CD19 APC, or anti-NK1.1DX5 PE, and gated R2 for CD8+/TCR FCs (middle panel), and CD8+/TCRCD19 or CD8+/TCRNK1.1DX5 subpopulation was sorted (left panel). B: Survival of syngeneic recipients transplanted with HSCs and CD8+/TCR FCs or CD8+/TCRCD19 NOR recipients were conditioned with 950 cGy TBI and transplanted with 500 HSCs alone or in combination with 30,000 CD8+/TCR FCs (HSC + FC group), CD8+/TCRCD19 (HSC + CD8+/TCRCD19 group). C: Survival of recipients transplanted with HSC plus CD8+/TCR FCs or CD8+/TCRNK1.1DX5 in an allogeneic model (B6 → C3H). Results are from three to four separate transplant experiments.

FIG. 5.

Removal of CD19+ or NK1.1+DX5+ FC subpopulations did not significantly affect facilitation. A: BM cells were stained with anti-CD8 PE or APC, anti–TCR-β FITC, anti–TCR-γδ FITC, and anti-CD19 APC, or anti-NK1.1DX5 PE, and gated R2 for CD8+/TCR FCs (middle panel), and CD8+/TCRCD19 or CD8+/TCRNK1.1DX5 subpopulation was sorted (left panel). B: Survival of syngeneic recipients transplanted with HSCs and CD8+/TCR FCs or CD8+/TCRCD19 NOR recipients were conditioned with 950 cGy TBI and transplanted with 500 HSCs alone or in combination with 30,000 CD8+/TCR FCs (HSC + FC group), CD8+/TCRCD19 (HSC + CD8+/TCRCD19 group). C: Survival of recipients transplanted with HSC plus CD8+/TCR FCs or CD8+/TCRNK1.1DX5 in an allogeneic model (B6 → C3H). Results are from three to four separate transplant experiments.

FIG. 6.

FL-mobilized PB FCs facilitate HSC engraftment in allogeneic recipients. Recombinant human FL (expressed from CHO cells) (provided by Amgen, Thousand Oaks, CA) was diluted in 0.1% mouse serum albumin (MSA; Sigma, St. Louis, MO) at a concentration of 100 μg/ml. Donor NOD female mice were injected with 10 μg FL once daily subcutaneously from day 0 to day 9. Control mice received saline injections. A: Flow cytometric analysis of subpopulations in sorted FL-mobilized PB FCs. B: Sorted FL-PB FCs were examined after Wright-Giemsa staining by optical microscopy. C: Survival of recipients of HSC plus FL-mobilized PB FCs in allogeneic model (NOD → B10). B10 recipients were conditioned with 950 cGy TBI and transplanted with 10,000 HSCs from untreated NOD donors either alone or mixed with 30,000 purified FCs from untreated NOD bone marrow, or from FL-mobilized PB FCs from NOD mice. D and E: Multilineage typing of representative B10 recipients of NOD HSCs alone or recipients of NOD HSCs plus FL-PB FCs. Multilineage data are from PB 3 months after transplantation and analyzed based on the lymphoid and myeloid gate. Data shown are from one representative recipient. A total of three to five recipients were analyzed per group.

FIG. 6.

FL-mobilized PB FCs facilitate HSC engraftment in allogeneic recipients. Recombinant human FL (expressed from CHO cells) (provided by Amgen, Thousand Oaks, CA) was diluted in 0.1% mouse serum albumin (MSA; Sigma, St. Louis, MO) at a concentration of 100 μg/ml. Donor NOD female mice were injected with 10 μg FL once daily subcutaneously from day 0 to day 9. Control mice received saline injections. A: Flow cytometric analysis of subpopulations in sorted FL-mobilized PB FCs. B: Sorted FL-PB FCs were examined after Wright-Giemsa staining by optical microscopy. C: Survival of recipients of HSC plus FL-mobilized PB FCs in allogeneic model (NOD → B10). B10 recipients were conditioned with 950 cGy TBI and transplanted with 10,000 HSCs from untreated NOD donors either alone or mixed with 30,000 purified FCs from untreated NOD bone marrow, or from FL-mobilized PB FCs from NOD mice. D and E: Multilineage typing of representative B10 recipients of NOD HSCs alone or recipients of NOD HSCs plus FL-PB FCs. Multilineage data are from PB 3 months after transplantation and analyzed based on the lymphoid and myeloid gate. Data shown are from one representative recipient. A total of three to five recipients were analyzed per group.

Published ahead of print at http://diabetes.diabetesjournals.org on 20 June 2008.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ for details.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

This work was supported in part by National Institutes of Health grants R01 DK069766 and 5R01 HL063442; Juvenile Diabetes Research Foundation grants 1-2005-1037 and 1-2006-1466; the Department of the Navy, Office of Naval Research; the Department of the Army, Office of Army Research; the Commonwealth of Kentucky Research Challenge Trust Fund; the W.M. Keck Foundation; and the Jewish Hospital Foundation.

The authors thank Dr. Thomas C. Mitchell for review of the manuscript and helpful comments; Michael Tanner, Lala-Rukh Hussain, and Barry Udis for technical assistance; Carolyn DeLautre for manuscript preparation; and the staff of the University of Louisville animal facility for outstanding animal care.

1.
Kaufman CL, Colson YL, Wren SM, et al.: Phenotypic characterization of a novel bone-marrow derived cell that facilitates engraftment of allogeneic bone marrow stem cells.
Blood
84
:
2436
–2446,
1994
2.
Gandy KL, Domen J, Aguila HL, et al.: CD8+TCR+ and CD8+TCR cells in whole bone marrow facilitate the engraftment of hematopoietic stem cells across allogeneic barriers.
Immunity
11
:
579
–590,
1999
3.
Schuchert MJ, Wright RD, Colson YL: Characterization of a newly discovered T-cell receptor beta-chain heterodimer expressed on a CD8+ bone marrow subpopulation that promotes allogeneic stem cell engraftment.
Nat Med
6
:
904
–909,
2000
4.
Fugier-Vivier I, Rezzoug F, Huang Y, et al.: Plasmacytoid precursor dendritic cells facilitate allogeneic hematopoietic stem cell engraftment.
J Exp Med
201
:
373
–383,
2005
5.
Anderson MS, Bluestone JA: The NOD mouse: a model of immune dysregulation.
Annu Rev Immunol
23
:
447
–485,
2005
6.
Serreze DV, Gaskins HR, Leiter EH: Defects in the differentiation and function of antigen presenting cells in NOD/Lt mice.
J Immunol
150
:
2534
–2543,
1993
7.
Kataoka S, Satoh J, Fujiya H, et al.: Immunologic aspects of the nonobese diabetic (NOD) mouse: abnormalities of cellular immunity.
Diabetes
32
:
247
–253,
1983
8.
Serreze DV, Fleming SA, Chapman HD, et al.: B lymphocytes are critical antigen-presenting cells for the initiation of T cell-mediated autoimmune diabetes in nonobese diabetic mice.
J Immunol
161
:
3912
–3918,
1998
9.
Cao W, Liu YJ: Innate immune functions of plasmacytoid dendritic cells.
Curr Opin Immunol
19
:
24
–30,
2007
10.
Steptoe RJ, Ritchie JM, Harrison LC: Increased generation of dendritic cells from myeloid progenitors in autoimmune-prone nonobese diabetic mice.
J Immunol
168
:
5032
–5041,
2002
11.
Poligone B, Weaver DJ Jr, Sen P, et al.: Elevated NF-kappaB activation in nonobese diabetic mouse dendritic cells results in enhanced APC function.
J Immunol
168
:
188
–196,
2002
12.
Prochazka M, Serreze DV, Frankel WN, et al.: NOR/Lt mice: MHC-matched diabetes-resistant control strain for NOD mice.
Diabetes
41
:
98
–106,
1992
13.
Rezzoug F, Huang Y, Tanner MK, et al.: TNFα is critical to facilitation of hematopoietic stem cell engraftment and function.
J Immunol
180
:
49
–57,
2008
14.
Grimes HL, Schanie CL, Huang Y, et al.: Graft facilitating cells are derived from hematopoietic stem cells and functionally require CD3, but are distinct from T lymphocytes.
Exp Hematol
32
:
946
–954,
2004
15.
Huang Y, Rezzoug F, Chilton PM, et al.: Matching at the MHC class I K locus is essential for long-term engraftment of purified hematopoietic stem cells: a role for host NK cells in regulating HSC engraftment.
Blood
104
:
873
–880,
2004
16.
Munn DH, Sharma MD, Hou D, et al.: Expression of indoleamine 2,3-dioxygenase by plasmacytoid dendritic cells in tumor-draining lymph nodes.
J Clin Invest
114
:
280
–290,
2007
17.
Ruggeri L, Capanni M, Urbani E, et al.: Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants.
Science
295
:
2097
–2100,
2002
18.
Chilton PM, Rezzoug F, Fugier-Vivier I, et al.: Flt3-ligand treatment prevents diabetes in NOD mice.
Diabetes
53
:
1995
–2002,
2004
19.
Bridenbaugh S, Kenins L, Bouliong-Pillai E, et al.: Clinical stem-cell sources contain CD8+CD3+ T-cell receptor-negative cells that facilitate bone marrow repopulation with hematopoietic stem cells.
Blood
111
:
1735
–1738,
2008
20.
Colson YL, Christopher K, Glickman J, et al.: Absence of clinical GVHD and the in vivo induction of regulatory T cells following facilitating cell transplantation.
Blood
104
:
3829
–3835,
2004
21.
Taylor KN, Shinde-Patil VR, Cohick E, et al.: Induction of FoxP3+CD4+25+ regulatory T cells following hemopoietic stem cell transplantation: role of bone marrow-derived facilitating cells.
J Immunol
179
:
2153
–2162,
2007
22.
Shatry AM, Levy RB: Pre-transplant infusion of donor CD4+ CD25+ T cells suppresses host antidonor MiHA-specific CD8 T cells and facilitates stable mixed chimerism following MHC-matched allogeneic marrow transplant (Abstract).
American Society of Hematology
118
:
a955
,
2007
23.
Morelli AE, Thomson AW: Tolerogenic dendritic cells and the quest for transplant tolerance.
Nat Rev Immunol
7
:
610
–621,
2007
24.
Kawai T, Akira S: Innate immune recognition of viral infection.
Nat Immunol
7
:
131
–137,
2006
25.
Abe M, Wang Z, De CA, et al.: Plasmacytoid dendritic cell precursors induce allogeneic T-cell hyporesponsiveness and prolong heart graft survival.
Am J Transplant
5
:
1808
–1819,
2005
26.
Pascual V, Banchereau J, Palucka AK: The central role of dendritic cells and interferon-alpha in SLE.
Curr Opin Rheumatol
15
:
548
–556,
2003
27.
Gottenberg JE, Cagnard N, Lucchesi C, et al.: Activation of IFN pathways and plasmacytoid dendritic cell recruitment in target organs of primary Sjogren's syndrome.
Proc Natl Acad Sci U S A
103
:
2770
–2775,
2006
28.
Greenberg SA, Pinkus JL, Pinkus GS, et al.: Interferon-alpha/beta-mediated innate immune mechanisms in dermatomyositis.
Ann Neurol
57
:
664
–678,
2005
29.
Gilliet M, Boonstra A, Paturel C, et al.: The development of murine plasmacytoid dendritic cell precursors is differentially regulated by FLT3-ligand and granulocyte/macrophage colony-stimulating factor.
J Exp Med
195
:
953
–958,
2002
30.
Chen W, Antonenko S, Sederstrom JM, et al.: Thrombopoietin cooperates with FLT3-ligand in the generation of plasmacytoid dendritic cell precursors from human hematopoietic progenitors.
Blood
103
:
2547
–2553,
2004
31.
Manfra DJ, Chen SC, Jensen KK, et al.: Conditional expression of murine Flt3 ligand leads to expansion of multiple dendritic cell subsets in peripheral blood and tissues of transgenic mice.
J Immunol
170
:
2843
–2852,
2003
32.
Ma L, Qian S, Liang X, et al.: Prevention of diabetes in NOD mice by administration of dendritic cells deficient in nuclear transcription factor-kappaB activity.
Diabetes
52
:
1976
–1985,
2003
33.
Jego G, Pascual V, Palucka AK, et al.: Dendritic cells control B cell growth and differentiation.
Curr Dir Autoimmun
8
:
124
–139,
2005
34.
Jego G, Palucka AK, Blanck JP, et al.: Plasmacytoid dendritic cells induce plasma cell differentiation through type I interferon and interleukin 6.
Immunity
19
:
225
–234,
2003
35.
Eynon EE, Parker DC: Small B cells as antigen-presenting cells in the induction of tolerance to soluble protein antigens.
J Exp Med
175
:
131
–138,
1992
36.
Yuschenkoff VN, Sethna MP, Freeman GJ, et al.: Coexpression of B7–1 and antigen blocks tolerance induction to antigen presented by resting B cells.
J Immunol
157
:
1987
–1995,
1996
37.
Martin F, Chan AC: B cell immunobiology in disease: evolving concepts from the clinic.
Annu Rev Immunol
24
:
467
–496,
2006
38.
Raulet DH, Vance RE, McMahon CW: Regulation of the natural killer cell receptor repertoire.
Annu Rev Immunol
19
:
291
–330,
2001
39.
Asai O, Longo DL, Tian ZG, et al.: Suppression of graft-versus-host disease and amplification of graft-versus-tumor effects by activated natural killer cells after allogeneic bone marrow transplantation.
J Clin Invest
101
:
1835
–1842,
1998
40.
Colonna M, Trinchieri G, Liu YJ: Plasmacytoid dendritic cells in immunity.
Nat Immunol
5
:
1219
–1226,
2004
41.
Ogasawara K, Hamerman JA, Hsin H, et al.: Impairment of NK cell function by NKG2D modulation in NOD mice.
Immunity
18
:
41
–51,
2003
42.
Wang B, Geng YB, Wang CR: CD1-restricted NK T cells protect nonobese diabetic mice from developing diabetes.
J Exp Med
194
:
313
–320,
2001
43.
Tang Q, Bluestone JA: Plasmacytoid DCs and T(reg) cells: casual acquaintance or monogamous relationship?
Nat Immunol
7
:
551
–553,
2006
44.
Yamazaki S, Iyoda T, Tarbell K, et al.: Direct expansion of functional CD25+ CD4+ regulatory T cells by antigen-processing dendritic cells.
J Exp Med
198
:
235
–247,
2003
45.
Kared H, Masson A, Adle-Biassette H, et al.: Treatment with granulocyte colony-stimulating factor prevents diabetes in NOD mice by recruiting plasmacytoid dendritic cells and functional CD4+CD25+ regulatory T-cells.
Diabetes
54
:
78
–84,
2005
46.
Salomon B, Lenschow DJ, Rhee L, et al.: B7/CD28 costimulation is essential for the homeostasis of the CD4+CD25+ immunoregulatory T cells that control autoimmune diabetes.
Immunity
12
:
431
–440,
2000
47.
Bour-Jordan H, Salomon BL, Thompson HL, et al.: Costimulation controls diabetes by altering the balance of pathogenic and regulatory T cells.
J Clin Invest
114
:
979
–987,
2004
48.
Pop SM, Wong CP, Culton DA, et al.: Single cell analysis shows decreasing FoxP3 and TGFbeta1 coexpressing CD4+CD25+ regulatory T cells during autoimmune diabetes.
J Exp Med
201
:
1333
–1346,
2005
49.
Kukreja A, Cost G, Marker J, et al.: Multiple immuno-regulatory defects in type-1 diabetes.
J Clin Invest
109
:
131
–140,
2002
50.
Lindley S, Dayan CM, Bishop A, et al.: Defective suppressor function in CD4(+)CD25(+) T-cells from patients with type 1 diabetes.
Diabetes
54
:
92
–99,
2005

Supplementary data