Technical Advance: In vitro production of distinct dendritic-like antigen-presenting cells from self-renewing hematopoietic stem cells

Transcription

1 f TECHNICL DVNCE Technical dvance: In vitro production of distinct dendritic-like antigen-presenting cells from self-renewing hematopoietic stem cells Rebecca. Hinton and Helen C. O Neill 1 Immunology and Stem Cell Group, Research School of iology, ustralian National University, Canberra CT, ustralia RECEIVED JUNE 20, 2011; REVISED OCTOER 5, 2011; CCEPTED OCTOER 20, DOI: /jlb STRCT novel CD11c lo CD11b hi MHC-II CD8 dendritic-like cell (L-DC) develops in cocultures of bone marrow over splenic stroma. L-DCs are distinct from other DC subsets and have potential importance in spleen for immunity to blood-borne antigens. s production is maintained in cultures for 12 months, L-DC development evidently depends on self-renewing progenitors. To improve this culture system, highly purified HSCs were sorted from bone marrow and used to establish cocultures. Nonadherent cells produced were analyzed for surface marker expression and capacity to activate/inhibit T cells. Cocultures produced a pure population of L-DCs for up to 12 months, which were strong activators of CD8 T cells. The in vitro production of a pure population of L-DCs from HSCs in numbers amenable to in vitro assays of function and development therefore represents an important advance. J. Leukoc. iol. 91: ; Introduction Very few cultures for in vitro hematopoiesis have been described that exactly replicate in vivo differentiation. In terms of DC differentiation from HSPCs, described culture systems are generally supplemented with inflammatory cytokines and do not reflect steady-state hematopoiesis. For example, DCs developing in vitro under the influence of GM-CSF and TNF- reflect inflammatory DCs and are distinct from cdc and pdc subsets produced in vitro under support from Flt3L [1]. It is now known that cdc and pdc develop from a multipotential macrophage/dc progenitor [2] and a more committed, common dendritic progenitor, both of which express Flt3, the receptor for Flt3L [3]. bbreviations: 6 C57L/6J, cdc conventional DC, F /F Flt3 /Flt3 subsets, respectively, Flt3L fms-like tyrosine kinase-3 ligand, FSC forward scatter, HSC hematopoietic stem cell, HSPC hematopoietic stem/progenitor cell, KLS ckit Lin Sca1, L-DC long-term culture DC, Lin lineage, OV-FITC FITC-conjugated ovalbumin, pdc plasmacytoid DC, PI propidium iodide, sdmem supplemented DMEM medium, SSC side scatter, STX3 murine splenic stromal cell line This lab reported previously that cocultures of HSPCs from bone marrow and spleen over the splenic STX3 stromal line support the differentiation and long-term production of distinct, myeloid-like DCs [4]. s the cells produced display similar morphology, phenotype, and functional properties to cells generated in spleen long-term cultures [5, 6], they have been termed L-DCs [4]. Cultures support continuous hematopoiesis with production of L-DCs for 12 months and provide the appropriate niche for hematopoiesis. L-DCs are unique as PCs, in that they activate CD8 T cells but not CD4 T cells [6 8]. Their role in cross-presentation of blood-borne antigen without activation of CD4 T cells is distinct amongst DCs. We recently published evidence for an in vivo equivalent PC subset to L-DC [8]. This is present only in spleen and is rare and difficult to isolate. Cells are highly endocytic for antigen and more accessible than other known spleen DC subsets for uptake of blood-borne antigen [8]. Murine HSCs were purified as Flt3 and Flt3 subsets of KLS cells in bone marrow and cocultured over STX3 to assess capacity for L-DC production. These subsets, differing by expression of Flt3, reflect HSCs, which are short-term (F KLS; 6 weeks) or long-term (F KLS; 20 weeks), reconstituting in mice [9]. F KLS cocultures generated a pure population of L-DCs over a long period in numbers amenable to further study of this rare cell type. MTERILS ND METHODS nimals Specific pathogen-free 6 mice, aged 4 6 weeks, were obtained from the John Curtin School of Medical Research (Canberra, CT, ustralia). Cell culture Cells were cultured in DMEM, supplemented with 10% heat-inactivated FCS (JRH iosciences, Lenexa, KS, US), 10 mm HEPES, 2 mm L-glutamine, 100 U/ml penicillin, 100 g/ml streptomycin, 4 g/l glucose, 6 mg/l folic acid, 36 mg/l L-asparagine, 116 mg/l L-arganine HCl, and M 2-ME (sdmem). The spleen stromal cell line STX3 [4] was passaged every 3 4 days by scraping a loose 25% of the adherent monolayer 1. Correspondence: Research School of iology, ustralian National University, Canberra CT 0200, ustralia. helen.oneill@anu.edu.au /12/ Society for Leukocyte iology Volume 91, February 2012 Journal of Leukocyte iology 341

2 and transferring cells into a new flask. Cells were incubated in 5% CO 2 in air with 97% humidity at 37 C. Isolation and preparation of HSCs one marrow was isolated from the femurs and tibias of 6 mice and cell suspensions prepared by RC lysis. Lin bone marrow was prepared using MCS magnetic beads, separators, and columns (Miltenyi iotec, Gladbach, Germany), as described previously [7]. Cells were exposed to a biotin-labeled, Lin-depletion cocktail containing antibodies specific for the Lin markers 7-4, CD5, CD11b, CD45R, Ly6G/C, and Ter119 (Miltenyi iotec). Cell suspensions were incubated with antibiotin MCS microbeads and then applied to a MS or LS column, the column placed in the magnetic field of a SuperMCS II separator, and unlabeled cells allowed to filter into a new tube, followed by three washes with MCS labeling buffer (PS/0.5% S/2 mm EDT). HSCs were sorted from Lin bone marrow by staining cells with additional Lin-depletion cocktail, along with antibodies specific for the DC markers CD11c (HL3; iolegend, San Gabriel, C, US) and CD11b (M1/70; iolegend), and the HSC markers ckit (28; eioscience, San Diego, C, US), Flt3 (2F10; eioscience), and Sca1 (E ; ecton Dickinson, San Jose, C, US), to sort subsets of F KLS and F KLS cells. Preparation of cocultures The capacity of STX3 to support hematopoiesis from HSCs was assessed by overlay of cell suspensions above stromal cell monolayers, followed by coculture for 4 8 weeks. STX3 cells were grown to 90% confluency ( cells) in a 25-cm 2 tissue-culture flask before addition of overlay cells in 5 ml sdmem. Production of cells was analyzed by antibody staining of nonadherent cells collected at medium change. Cells were photographed by brightfield and phase microscopy using a DM IRE2 inverted research microscope (Leica, North Ryde, NWS, ustralia), equipped with a DFC digital camera (Leica). Images were processed using Leica IM4.0 software. ntibody staining of cells ntibody staining was used to delineate cell subsets. Staining involved cells/well of a flexible 96-well polystyrene microtiter plate (Corning, Corning, NY, US). Cells were incubated with anti-cd16/32 ("Fc lock"; eioscience), diluted in FCS buffer (DMEM/1% FCS/0.1% NaN 3 ) at 4 C for 15 min to block FcRs. further 25- l primary antibody cocktail was then added and cells incubated for 25 min. Cells were washed twice with 150 l FCS buffer. If secondary conjugates were used, 25 l asa cocktail was added to each well, and cells incubated at 4 C for 25 min. Cells were washed twice and resuspended in 50 l FCS buffer. For sorting, cells were stained in the absence of 0.1% NaN 3. Prior to multicolor flow cytometric analysis, 5 l 100 g/ml PI (Sigma-ldrich, St. Louis, MO, US) was added for discrimination of dead cells, which were analyzed using an LSRII flow cytometer (ecton Dickinson) and sorted using a FCSria II flow cytometer (ecton Dickinson). D FCSDiva software (ecton Dickinson) was used to set voltage parameters and perform compensation. Data were processed using FlowJo software (Tree Star, shland, OR, US) for postacquisition gating and manual compensation. nalysis involved FSC (log analysis), fluorochrome binding (log analysis), and SSC (linear analysis). For analysis and sorting of dendritic and myeloid subsets, cells were stained with 220 (R3-62; eioscience), CD11b (M1/70; iolegend), CD11c (N418; eioscience), CD8 (53-6.7; eioscience), CD24 (M1/69; eioscience), Gr1 (R6-8C5; ecton Dickinson), and MHC-II (F ; eioscience). Isotype control antibodies were used to set gates for sorting, and sorted subsets were reanalyzed flow cytometrically to check purity. Endocytosis Cells were assessed for capacity to endocytose antigen after in vitro exposure to OV-FITC (Molecular Probes, Eugene, OR, US). Cells were placed on ice for 10 min before addition of 100 g/ml OV-FITC in a total volume of 100 l sdmem. Cells were incubated at 37 C for 45 min before endocytosis was halted by addition of 100 l chilled PS/0.1% NaN 3. Cells were washed three times before analysis by flow cytometry. Cell proliferation Cells were labeled with CFSE (Molecular Probes) for analysis of cell division. CFSE was added to cells at 10 g/ml and samples vortexed immediately and incubated at room temperature for 5 min. Cells were then washed twice by addition of 1 ml ice-cold medium. Following culture for various times, cell division was assessed flow cytometrically in terms of reduction in CFSE as cells divide. Data are expressed as a percentage of cells that have divided one or more times. T cell activation MLR was used to investigate the ability of L-DC to activate allogeneic T cells or syngeneic T cells as controls. L-DCs were collected as nonadherent cells from cocultures, established from 6 Lin bone marrow or HSCs. Control CD11c DCs were isolated from 6 and C/H spleens by antibody-mediated magnetic bead selection. T cells, as responders, were prepared from 6 and C/H spleens by depletion of myeloid and cells using antibody-coated magnetic beads and then CFSE-labeled. T cells were cultured at cells/well in a 96-well plate with ratios of 1:10, 1:100, and 1:1000 PCs:T cells. T cells alone provided a negative control. MLRs in a volume of 200 l medium were incubated at 37 C for 4 days ahead of flow cytometric analysis of proliferation in terms of reduction in level of CFSE staining. In separate experiments, cells were stained with antibodies to detect proliferated CD4 and CD8 T cells. To detect whether L-DC asserted any inhibitory effect on T cell activation, graded numbers of L-DCs were added to MLRs, established with freshly isolated CD11c DCs (1:10) and T cells ( ). Potential inhibitors were added to give dilutions of 1:10, 1:100, and 1:1000 L-DCs:T cells. Statistical analysis Data have been presented as mean se, where sample size n. The Wilcoxon rank sum test (P 0.05) was therefore used to confirm statistical significance. Where a normal distribution could be assumed, 95% confidence intervals have been calculated. RESULTS ND DISCUSSION STX3 stroma supports HSC To quantify the proportion of cells amongst sorted HSCs, which formed colonies over STX3 stroma, sorted cells were plated using the flow cytometer single cell deposition facility at one cell/well in microtiter plates containing confluent STX3 stroma (Fig. 1). Plates were monitored for development of colonies. fter 25 days of coculture, colony-forming cells represented 14% of sorted F KLS cells and 10% of sorted F KLS cells (Fig. 1). ased on 95% confidence limits, these values were not significantly different. This high frequency of colony-forming cells confirms that colonies are arising from self-renewing progenitors within the sorted HSC subsets, rather than from low-frequency, replicating, non-hsc contaminants as a result of sorting error, estimated at 1% of sorted cells. STX3 stroma supports long-term in vitro hematopoiesis Sorted HSC subsets or Lin bone marrow cells as controls were cultured in replicate cocultures over STX3. Cocultures were analyzed over 4 weeks for cell production in terms of morphology and staining for CD11b, CD11c, CD8, and MHC-II. 342 Journal of Leukocyte iology Volume 91, February

3 TECHNICL DVNCE Hinton and O Neill Hematopoiesis of dendritic-like cells in vitro Propor on of CFC HSC subset Colony frequency* 95% confidence interval 10/ / *HSC were plated at 1 cell/well above STX3 stroma and colonies counted at 25 days Figure 1. Growth of HSC colonies over STX3 stroma. () one marrow (M) of 10, 4- to 6-week-old 6 mice was Lin-depleted using MCS bead technology. Cells were stained with the Lin antibody cocktail, CD11b, CD11c, c-kit, Sca-1, and Flt3, along with PI for detection of live PI cells. F KLS cells were sorted as a PI Lin CD11b CD11c ckit Sca1 Flt3 subset, and F KLS cells were sorted as a PI Lin CD11b CD11c ckit Sca1 Flt3 subset. Numbers in gates represent percent positive cells, as set using isotype control antibodies. () Colony production from F KLS and F KLS cocultures over STX3 after 25 days was estimated in terms of proportion of colony-forming cells (CFC) amongst HSC subsets. fter 7 days, cells produced within F KLS and F KLS cocultures demonstrated distinct DC morphology (Fig. 2). Using flow cytometry, large (FSC hi ) cells were gated as CD11b CD11c DCs. Large cells generated in Lin bone marrow cocultures were evenly distributed as MHC-II hi, MHC-II lo, and MHC-II subsets, whereas cells produced in F KLS cocultures comprised mainly MHC-II lo and MHC-II cells, and cells produced in F KLS cocultures were exclusively MHC-II. ll MHC- II DCs were found to be FSC hi SSC hi, whereas MHC-II lo and MHC-II hi cells were found to have lower FSC and SSC profiles. lthough Lin bone marrow and F KLS cocultures produced a higher overall yield of DCs, 20 30% of these cells were MHC-II, whereas the F KLS population comprised 95% MHC-II DCs (Fig. 2C). This difference was significant (P 0.05; Wilcoxon rank sum test). No CD8 cdcs were produced in any of the three coculture types (data not shown). fter 14 days, cells with DC morphology were again observed within all cocultures (Fig. 3), although cells produced did not express MHC-II. Distinct CD11b lo CD11c and CD11b hi CD11c populations could be gated (Fig. 3). F KLS cocultures produced a much higher and significant percentage (37%; P 0.05; Wilcoxon rank sum test) of CD11b lo CD11c cells than did Lin bone marrow (7%) or F KLS (5%) cocultures (Fig. 3C). fter 4 weeks, cells produced in all cocultures were uniformly FSC hi CD11b CD11c lo CD8 MHC-II cells, and high endocytic capacity was shown by uptake of OV-FITC, a phenotype corresponding to L-DC (Fig. 3D). They also lacked expression of the pdc marker 220 and the myelomonocytic marker Gr1. - Lin - M C % amongst cells Lin M F KLS MHC-II - MHC-II lo MHC-II hi Figure 2. Cell production in STX3 cocultures. F KLS and F KLS cells were seeded into four replicate flasks at 10 4 cells/5 ml. Lin bone marrow cells as a positive control were seeded at 10 5 cells/5 ml. () Cell production in cocultures established with HSCs was photographed at 7 days using bright-field microscopy (scale bars, 100 m). () Nonadherent cells were collected at 7 days, and antibody staining for CD11b, CD11c, CD8, and MHC-II and flow cytometry used to delineate hematopoietic cell subsets produced in cocultures. Cells were stained with PI for gating PI live cells. Large cells were gated on SSC versus FSC. DCs were identified as CD11b CD11c cells, and distinct subsets were quantified as percent MHC-II, MHC-II lo, and MHC-II hi subsets amongst all CD11c CD11b DCs. Numbers in gates reflect percent positive cells as set by isotype control antibodies. Representative staining results for one of four replicate cocultures are shown, and mean production of MHC-II, MHC-II lo, and MHC-II hi subsets of CD11b CD11c cells is presented in C as mean se (n 4). Only one control Lin bone marrow coculture was established. Volume 91, February 2012 Journal of Leukocyte iology 343

4 Lin - M C % amongst cells D Lin - M Figure 3. Production of L-DC in cocultures established with HSCs. Cocultures were established by overlay of HSC subsets F KLS and F KLS over STX3, with Lin bone marrow as a control. () Cocultures were photographed at 14 days using bright-field microscopy (scale bars, 100 m). () ntibody staining and flow cytometry were used to delineate DC subsets within cocultures, as described in Fig. 2. (C) t 14 days, DCs were identified as CD11b lo CD11c and CD11b hi CD11c cells. The proportion of each subset amongst total cells produced was compared for cocultures established with Lin bone marrow, and the F KLS and F KLS subsets of HSC. Data represent mean se (n 4). (D) Nonadherent cells produced in cocultures of F KLS bone marrow HSCs over STX3 were collected at 27 days and stained with fluorochrome-conjugated antibodies specific for CD11b, CD11c, 220, CD8, Gr-1 and MHC-II. Cells were also incubated at 37 C for 40 min with OV-FITC to assess capacity for endocytosis. The finding that L-DC can be generated from F KLS cells is consistent with previous data showing that STX3 stromal cells do not produce Flt3L [10]. The presence of the L-DC progenitor amongst the F KLS also distinguishes L-DCs from cdcs and pdcs [11]. L- DCs develop via a Flt3L-independent pathway and do not arise from described common dendritic progenitors, which are distinct Flt3 Lin ckit lo Sca1 progenitors of cdcs and pdcs [12]. L-DCs activate CD8 but not CD4 T cells fter 24 days, nonadherent cells were collected from cocultures and tested for capacity to stimulate T cell proliferation in a MLR. L-DCs raised in F KLS or F KLS cocultures did not initiate an allogeneic CD4 T cell response, in contrast to control CD11c DCs isolated from spleen, which initiated an allogeneic but not a syngeneic response (Fig. 4). However, although DCs derived from F KLS cocultures did not stimulate CD8 T cell proliferation, DCs taken from F KLS cocultures demonstrated weak capacity for CD8 T cell stimulation: 16% T cell proliferation compared with 45% for control spleen DCs (Fig. 4). This response was specific for allogeneic (C/H) not syngeneic (6) T cells. DCs derived from cocultures established with F KLS or F KLS HSC subsets therefore differ in functional capacity, with DCs developing from the more primitive F KLS subset, showing distinct ability to activate CD8 T cells. These results are consistent with previous findings that DCs produced in long-term spleen cultures are highly endocytic with capacity to stimulate CD8 T cells [6, 8], and that L-DCs generated in cocultures of whole bone marrow over STX3 display a similar function [4]. In this respect, L- Spl DC (6) Spl DC (6) Figure 4. T cell activation capacity of L-DCs. Cocultures of F KLS and F KLS bone marrow HSCs sorted from 6 mice were established over STX3 stroma. Nonadherent cells were collected from cocultures after 24 days and tested capacity to activate T cells in a 4-day MLR. T cells were isolated from spleens of C/H and 6 mice, depleted of and myeloid cells using magnetic bead technology, and then labeled with CFSE. CD11c control DCs were freshly isolated from 6 spleen (Spl DC) using MCS magnetic bead technology. T cells were cultured together with the coculture-derived L-DCs or splenic DCs as controls in graded dilutions of 10:1, 100:1, and 1000:1 T cell:dc. CD4 and CD8 T cell subsets were identified flow cytometrically along with their CFSE profile. Data are presented as percent divided T cells for C/H (C; allogeneic) and 6 (syngeneic) CD4 T cells (), and CD8 T cells (). Controls included T cells only. Results are representative of duplicate experiments. 344 Journal of Leukocyte iology Volume 91, February

5 TECHNICL DVNCE Hinton and O Neill Hematopoiesis of dendritic-like cells in vitro DCs differ from the cdc and pdc subsets, which activate CD4 and CD8 T cells. L-DCs do not inhibit T cell activation Nonadherent cells collected after 24 days were also tested as inhibitors of MLRs. T cells were isolated from spleens of allogeneic and syngeneic mice, labeled with CFSE, and combined with freshly isolated CD11c spleen DCs at a 100:1 ratio. Cells produced in cocultures were then added in graded dilutions of 10:1, 100:1, and 1000:1 T cell:inhibitor. fter 4 days, cells were stained with antibodies and flow cytometry used to identify CD4 and CD8 T cells and their CFSE profile. When potential inhibitors were allogeneic with T cells, little change in T cell proliferation was observed for CD8 or CD4 T cells (Fig. 5), although DCs derived from F KLS cocultures did induce weak proliferation of CD4 T cells (Fig. 5). When inhibitors were instead syngeneic with T cells, enhanced proliferation of CD4 and CD8 T cells was observed (Fig. 5). DCs derived from F KLS cocultures stimulated T cell activation, reflective of uptake of apoptotic, allogeneic spleen DCs by L- DCs with presentation for MHC-restricted activation of T cells. F KLS cocultures, which produced a distinct subset of CD11b lo CD11c hi MHC-II DCs, along with L-DCs, did produce cells that inhibited T cell proliferation at the highest concentration (10:1; Fig. 5) and could reflect regulatory DCs reported previously in cocultures of HSCs over splenic endothelial cells [13]. Indeed, regulatory DCs have also been reported to differentiate from myeloid cells and could be transiently produced in F KLS cocultures [14]. L-DCs can be maintained in STX3 cocultures in excess of 12 months Cocultures established with Lin bone marrow or with F KLS or F KLS HSCs were maintained for over 12 months. During this time, the stromal layer rolls up and is re-established several times. Previous studies confirmed that overlaid cells are not the source of new stromal growth [4]. Cocultures examined after 6 months produced cells with clear DC morphology (Fig. 6). When nonadherent cells were stained with antibody, the majority of cells within F KLS or F KLS cocultures was CD11b CD11c lo CD8 MHC-II cells, highly endocytic for OV- FITC and reflective of L-DCs (Fig. 6). Cells produced failed to show expression of MHC-II, a marker of mature cdc, or CD24, a marker of cdc precursors [15, 16]. Cell production in Lin bone marrow and F KLS cocultures was similar at 12 months (Fig. 6C), whereas production within F KLS flasks had slowed significantly. Over this period, F KLS cocultures consistently yielded constant numbers of L-DCs at biweekly medium change with the number of nonadherent cells collected at five- and ten-fold above the input number of HSCs. Cocultures therefore provide cells in numbers sufficient to investigate the function of L-DCs in in vivo and in vitro assays. lthough F KLS and F KLS cocultures produced CD11b hi CD11c lo MHC-II cells, which morphologically and phenotypically resemble L-DCs, F KLS cells produced these exclusively, whereas F KLS cells generated an additional CD11b lo CD11c hi MHC-II subset, reflective of myeloid DCs. s the latter subset was not seen after 28 days (all data not shown), it appears to represent a transient population, arising from preformed myeloid precursors contained within the sorted F KLS population. The L-DC progenitor is most highly enriched amongst the F KLS subset of bone marrow. ased on the longevity of L-DC production, an L-DC progenitor is also present, along with myeloid DC precursors within the F KLS subset of HSCs. The F KLS progenitor appears to be a more primitive progenitor, which may differentiate to become a Flt3 progenitor when cultured over STX3. nother explanation is that the L-DC pro- CD4 + T cells ded cells % divid % divide ed cells CD8 + T cells T cell : Inhibitor (6) (6) (6) (6) Figure 5. L-DCs are not inhibitors of T cell activation. Nonadherent cells were collected from cocultures of F KLS and F KLS bone marrow HSCs from 6 mice after 24 days and tested for their capacity to inhibit the activation of T cells by splenic DCs. The latter were freshly isolated from 6 and C/H spleens using MCS. T cells were isolated from the spleens of C/H and 6 mice, depleted of and myeloid cells using magnetic bead technology, labeled with CFSE, and cocultured with spleen DCs at a ratio of 100:1, along with graded numbers of potential inhibitors in T cell:inhibitor ratios of 10:1, 100:1, and 1000:1. fter 4 days, cells were stained with antibody and analyzed flow cytometrically to detect the CFSE profile of CD4 and CD8 T cells. Data are presented as percent divided T cells based on CFSE staining for CD4 T cells () and CD8 T cells () in each of C/H anti-6 and 6 anti-c/h reactions. The dashed lines show response in the absence of added "inhibitor". Results are representative of duplicate experiments. T cell : Inhibitor Volume 91, February 2012 Journal of Leukocyte iology 345

6 SSC C Lin - Figure 6. Long-term production of L-DCs from HSCs in vitro. Cocultures M were established by overlay of F KLS and F KLS bone marrow HSCs or Lin bone marrow (as a control) over STX3 stroma. Data shown are reflective of four replicate cocultures. () Cocultures were photographed after 6 months using bright-field microscopy (scale bars, 100 m). () Nonadherent cells from cocultures established for 6 months were stained with fluorochrome-conjugated antibodies specific for CD11b, CD11c, and MHC-II to detect cell types produced as described in Fig. 2. Cells were also incubated at 37 C for 40 min with OV-FITC to assess capacity for endocytosis. (C) Cells produced in Lin bone marrow and F KLS cocultures maintained for 12 months were stained with fluorochrome-conjugated antibodies specific for CD11b, CD11c, MHC-II, and CD24. Prior to flow cytometry, cells were stained with PI for gating live PI cells. For analysis of cell production, large cells were gated based on SSC versus FSC. Numbers in gates represent percent positive cells, as set by isotype control antibodies. genitor in this subset is a Flt3 progenitor but that Flt3 is not a distinguishing marker of progenitors. In summary, this paper reports a reliable coculture system involving a splenic stromal cell line and purified HSCs, which produce L-DCs. It represents a controlled, ready source of distinct PCs for further analysis of their immune function, as well as a stromal niche for hematopoiesis which is readily amenable to studies on signaling for hematopoiesis and L-DC production. UTHORSHIP R..H. and H.C.O. conceived of and designed the experiments, interpreted the data, and wrote the paper. R..H. performed the experiments. CKNOWLEDGMENTS This work was supported by project grants to H.C.O.: #P061 from the ustralian Stem Cell Centre and # from the National Health and Medical Research Council of ustralia. R..H. was supported by an NU Graduate School Fellowship. The authors recognize the statistical advise of Terry O Neill (Statistics Department, NU). REFERENCES 1. Xu, Y., Zhan, Y., Lew,. M., Naik, S. H., Kershaw, M. H. (2007) Differential development of murine dendritic cells by GM-CSF versus Flt3 ligand has implications for inflammation and trafficking. J. Immunol. 179, Fogg, D. K., Sibon, C., Miled, C., Jung, S., ucouturier, P., Littman, D. R., Cumano,., Geissmann, F. (2006) clonogenic bone marrow progenitor specific for macrophages and dendritic cells. Science 311, Onai, N., Obata-Onai,., Schmid, M.., Ohteki, T., Jarrossay, D., Manz, M. G. (2007) Identification of clonogenic common Flt3 M-CSFR plasmacytoid and conventional dendritic cell progenitors in mouse bone marrow. Nat. Immunol. 8, Periasamy, P., Tan, J. K., Griffiths, K. L., O'Neill, H. C. (2009) Splenic stromal niches support hematopoiesis of dendritic-like cells from precursors in bone marrow and spleen. Exp. Hematol. 37, O'Neill, H. C., Wilson, H. L., Quah,., bbey, J. L., Despars, G., Ni, K. (2004) Dendritic cell development in long-term spleen stromal cultures. Stem Cells 22, Quah,., Ni, K., O'Neill, H. C. (2004) In vitro hematopoiesis produces a distinct class of immature dendritic cells from spleen progenitors with limited T cell stimulation capacity. Int. Immunol. 16, Tan, J. K., Periasamy, P., O'Neill, H. C. (2010) Delineation of precursors in murine spleen that develop in contact with splenic endothelium to give novel dendritic-like cells. lood 115, Tan, J. K. H., Quah,. J. C., Griffiths, K. L., Periasamy, P., Hey, Y-Y., O'Neill, H. C. (2011) Identification of a novel antigen cross-presenting cell type in spleen. J. Cell. Mol. Med. 15, Lai,. Y., Lin, S. M., Kondo, M. (2005) Heterogeneity of Flt3-expressing multipotent progenitors in mouse bone marrow. J. Immunol. 175, Despars, G., O'Neill, H. C. (2004) role for niches in the development of a multiplicity of dendritic cell subsets. Exp. Hematol. 32, D'mico,., Wu, L. (2003) The early progenitors of mouse dendritic cells and plasmacytoid predendritic cells are within the bone marrow hemopoietic precursors expressing Flt3. J. Exp. Med. 198, Liu, K., Victora, G. D., Schwickert, T.., Guermonprez, P., Meredith, M. M., Yao, K., Chu, F. F., Randolph, G. J., Rudensky,. Y., Nussenzweig, M. (2009) In vivo analysis of dendritic cell development and homeostasis. Science 324, Zhang, M., Tang, H., Guo, Z., n, H., Zhu, X., Song, W., Guo, J., Huang, X., Chen, T., Wang, J., Cao, X. (2004) Splenic stroma drives mature dendritic cells to differentiate into regulatory dendritic cells. Nat. Immunol. 5, Tang, H., Guo, Z., Zhang, M., Wang, J., Chen, G., Cao, X. (2006) Endothelial stroma programs hematopoietic stem cells to differentiate into regulatory dendritic cells through IL- 10. lood 108, Hunte,. E., Capone, M., Zlotnik,., Rennick, D., Moore, T.. (1998) cquisition of CD24 expression by Lin CD43 220(low)ckit(hi) cells coincides with commitment to the cell lineage. Eur. J. Immunol. 28, Naik, S. H., Metcalf, D., van Nieuwenhuijze,., Wicks, I., Wu, L., O'Keeffe, M., Shortman, K. (2006) Intrasplenic steady-state dendritic cell precursors that are distinct from monocytes. Nat. Immunol. 7, KEY WORDS: hematopoiesis progenitors multiparameter FCS 346 Journal of Leukocyte iology Volume 91, February