Flexible fate commitment of E2-2high common DC progenitors ...

© The Japanese Society for Immunology. 2017. All rights reserved. For permissions, please e-mail: [email protected]

Flexible fate commitment of E2-2high common DC progenitors implies tuning in tissue microenvironments

Nobuyuki Onai1,2, Jumpei Asano1, Rumiko Kurosaki1, Shoko Kuroda1 and Toshiaki Ohteki1 1Department of Biodefense Research, Medical Research Institute, Tokyo Medical and Dental University (TMDU), 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8510, Japan2Department of Immunology, Kanazawa Medical University, 1-1 Daigaku, Uchinada, Kahoku, Ishikawa 920-0293, Japan

Correspondence to: T. Ohteki; E-mail: [email protected]

Received 12 June 2017, editorial decision 19 October 2017; accepted 26 October 2017

Abstract

The basic helix-loop-helix transcription factor E2-2 is essential for the development of plasmacytoid dendritic cells (pDCs) but not conventional DCs (cDCs). Here, we generated E2-2 reporter mice and demonstrated that an E2-2high fraction among common DC progenitors, which are a major source of pDCs and cDCs in the steady state, strictly gave rise to pDCs in the presence of Flt3 (Fms-like tyrosine kinase receptor-3) ligand ex vivo or in the secondary lymphoid organs when transferred in vivo. However, in the small intestine, some of these E2-2high progenitors differentiated into cDCs that produced retinoic acid. This transdifferentiation was driven by signaling via the common β receptor, a receptor for the cytokines IL-3, IL-5 and GM-CSF, which are abundant in the gut. In the presence of GM-CSF and Flt3 ligand, E2-2high-progenitor-derived cDCs consistently induced Foxp3+ Treg cells ex vivo. Our findings reveal the commitment and flexibility of E2-2high progenitor differentiation and imply that pertinent tuning machinery is present in the gut microenvironment.

Keywords: βc cytokines, common dendritic cell progenitors (CDP), E2-2, plasmacytoid dendritic cell

Introduction

Plasmacytoid dendritic cells (pDCs) were originally identified as plasmacytoid T cells or plasmacytoid monocytes in the T-cell area of the lymph node (1, 2) and were subsequently defined as cells that naturally produce type I interferon in the human tonsils and peripheral blood (3). These cells were eventually termed pDCs because they acquire DC morph-ology and stimulate T cells when activated (4, 5). The cap-acity of pDCs to produce large amounts of type I interferon is due to their selective expression of the nucleic acid sen-sors TLR7 and TLR9, and pDCs have been implicated in the pathogenesis of autoimmune diseases such as lupus and psoriasis (6).

Conventional DCs (cDCs) and pDCs in the lymphoid organs share the same developmental pathway in terms of their cytokine signaling and progenitors (7). Steady-state DC development is largely dependent on Fms-like tyrosine kin-ase receptor-3 (Flt3) and its ligand (Flt3L), as shown by their gene-deficient and transgenic mice (8–12). We and another group previously identified DC-restricted progenitors, that is, common DC progenitors (CDPs) (13–15). CDPs are highly proliferative and give rise strictly to pDCs and cDCs, and

not to other cell lineages, whether in vitro or in vivo. CDPs consist of CD115+ and CD115− subfractions (13, 15). The CD115+ subset produces many more cDCs than pDCs, while the CD115− subset is a major source of pDCs in the steady state. CDPs differentiate into cDC1 and cDC2 cells via pre-cDC1s and pre-cDC2s, respectively (16, 17), and into CCR9+ pDCs via CCR9− pDCs (18–20). Single-cell analysis revealed that both CD115+ and CD115− CDPs include progenitors that are bi-potent for pDCs and cDCs and other progenitors that are uni-potent for either pDCs or cDCs (13, 15), implying that progenitors that are already committed toward either pDCs or cDCs at the CDP stage are present in the bone marrow (BM). In this context, transcription factors (TFs) have been identified that are critical for developing and maintaining DC subsets (21, 22). One of the E-protein TFs, E2-2 (also known as Tcf4), is essential for the development and maintenance of pDCs in both mice and humans (23–25). Mice carrying an E2-2 gene deficiency in hematopoietic cells have impairments in pDC development and in CpG-induced type I interferon production in vivo (23). A conditional depletion of the E2-2 gene induces the conversion from pDCs into CD8α+ cDC-like

International Immunology, Vol. 29, No. 10, pp. 443–455doi:10.1093/intimm/dxx058Advance Access publication 2 November 2017

Dow

nloaded from https://academ

ic.oup.com/intim

m/article/29/10/443/4587487 by guest on 14 January 2022

mailto:[email protected]?subject=

cells (25). In Pitt-Hopkins syndrome, a disease caused by a heterozygous E2-2 deficiency, pDCs express cell surface markers abnormally and fail to produce type I interferon in the peripheral blood (23).

To identify pDC-restricted progenitors, we generated E2-2-Kusabira-Orange (E2-2-KuOr) reporter mice using BAC transgenic (Tg) technology. We identified E2-2high progeni-tors among CDPs as pDC-restricted progenitors in the BM of E2-2-KuOr reporter mice. These E2-2high progenitors gave rise only to pDCs, and not cDCs, in the secondary lymphoid organs, whereas in the small intestine, E2-2high progenitors not only gave rise to pDCs but also differentiated into cDCs to some extent, in a manner dependent on common β chain (βc)-related cytokines that are enriched in the gut. These find-ings imply that a pertinent tuning mechanism is present in the gut.

Methods

MiceC57BL/6J (B6, Clea), B6.SJL-ptprca mice congenic at the CD45 locus (B6.SJL mice), E2-2+/− (26), E2-2-KuOr BAC Tg, CD11c-DTR (27), Csf2rb−/− (28) and OT-II (Taconic) mice were maintained in our specific pathogen-free (SPF) facility. All animal experiments were approved by the Institutional Animal Care Committees of Tokyo Medical and Dental University.

BM chimeric miceControl E2-2+/+ and E2-2−/− fetal liver (FL) cells were prepared from 14.5 days postcoitum embryos, genotyped by PCR and pooled. The E2-2+/+ and E2-2−/− FL cells (1 × 106) were injected into lethally X-irradiated (10 Gy, Faxitron) B6.SJL mice (CD45.1+CD45.2−), and cell progenies in the BM and spleen were analyzed 8 weeks later. DC-depleted mice were gener-ated by isolating BM cells (1 × 106) from B6.SJL × CD11c-DTR-F1 mice (CD45.1+CD45.2+), injecting the cells into lethally irradiated B6.SJL mice, and transplanting the sorted progenitors into diphtheria toxin (DT)-treated mice (8 ng g−1 body weight) 8 weeks later. The DT-treated mice were eutha-nized 8 days after transplantation, and the cell progenies in the spleen and gut were analyzed. Vehicle control (PBS) or 8 µg of βc cytokines (IL-3, IL-5 and GM-CSF) (Biolegend) were subcutaneously injected into E2-2high CDP transplanted mice on day 1, 3, 5 and 7 after transplantation. The spleen prog-enies were analyzed on day 8.

Generation of E2-2-KuOr BAC Tg miceThe BAC clone (RP23-1630D18) covering most of the E2-2 locus was modified using the RED/ET recombination tech-nique (Gene Bridges, Germany). The KuOr-SV40 polyA fragment was ligated into an FRT PGK-gb2-Neo expres-sion cassette consisting of a prokaryotic promoter and a neomycin-resistance gene flanked by an FRT sequence (Gene Bridges). Homology arms for the second exon of E2-2 were ligated into both ends of the reporter cassette by PCR amplification. Escherichia coli carrying the BAC was transformed with the RED/ET expression plasmid pSC101-BAD-gbaA. Recombinants were identified by PCR to detect

the kanamycin-resistance gene. The Kan/Neo cassette was excised by introducing the FLPe expression plasmid (Gene Bridged) and inducing FLPe expression. Recombinant BAC DNAs were purified using a BAC100 column (Nippon Genetics, Tokyo), linearized and injected into fertilized eggs. The E2-2-KuOr BAC Tg mice were genotyped by PCR using the primer pair 5′-ACTCCAGAGACGCTGGTCTT-3′ and 5′-GCTGCCGTCCATGTAGTAGCGCATCTTC-3′. All pro-cedures for generating E2-2-KuOr mice were performed by Laboratory Animal Resource Center (LARC), Faculty of Medicine, University of Tsukuba.

Cell sorting and flow cytometric analysisBM lineage negative (Lin−) cells were immunomagnetically pre-enriched using PE-Cy5-conjugated antibodies against lineage antigens, including CD3ε (145-2C11), CD4 (GK1.5), CD8α (53-6.7), B220 (RA3-6B2), CD19 (MB19-1), CD11c (N418), MHC class II (I-A and I-E; M-15/114.15.2), CD11b (M1/70), Gr-1 (RB6-8C5), TER119 (TER119), NK1.1 (PK136), anti-Cy5 MicroBeads and autoMACSpro (Miltenyi Biotec). To isolate E2-2+ and E2-2− CDPs, BM Lin− cells were then stained with FITC-anti-CD115 (AFS-98), APC-anti-CD135 (A2F10.1), PE-Cy7-anti-CD117 (ACK2) and Brilliant Violet 421-anti-CD127 (A7P34) (all from BioLegend). To assess the number of hematopoietic stem cells (HSCs) and hematopoietic stem and progenitor cells (HSPCs), BM Lin− cells were stained with antibodies against CD117 (APC), Sca-1 (PE-Cy7), CD34 (FITC), CD135 (Brilliant Violet 421), CD115 (APC), Ly6c (HK1.4; Brilliant Violet 510), CD11c (APC-Cy7), I-A/I-E (PE-Cy7), CD172a (P84; Biotin) and streptavidin (APC-Cy7) (all from BioLegend). The cells were sorted and analyzed on a FACSAria III or a FACSCanto II (BD Biosciences) with FlowJo software (TreeStar). Cells were cultured as described previ-ously (13).

Quantitative RT–PCRTotal RNA was extracted using the RNeasy Mini Kit (Qiagen), and cDNA was synthesized using random hexamers and SuperScript III reverse transcriptase. For real-time PCR, cDNA products equivalent to the RNAs from 500 cells were ampli-fied using LightCycler®480 SYBR Green I Master (Roche Diagnostics). The data were normalized to the amount of gapdh RNA in each sample. The primers used for real-time PCR were as follows: E2-2, sense, 5′-TGAGATCAAATCCGACGA-3′ and antisense, 5′-CGTTATTGCTAGATCTTGACCT-3′; Gapdh sense, 5′-TCCACCACCCTGTTGCTGTA-3′ and antisense, 5′-ACCACAGTCCATGCCATCAC-3′. Primers were synthe-sized by Eurofins Genomics.

Aldehyde dehydrogenase activity assayAldehyde dehydrogenase (ALDH) activity was determined using the ALDEFLUOR staining kit (StemCell Technology) according to the manufacturer’s instructions. Lamina pro-pria cells were isolated from the small intestine of progen-itor-transplanted DC-depleted chimeric mice. The cells were stained with antibodies against CD45.1 (A20; PE-Cy7), CD45.2 (104; Pacific Blue), CD103 (2E7; APC), CD11b (APC-Cy7), CD11c (Biotin) and streptavidin-BV510. The cells

444 Flexible fate in E2-2high CDPs

Dow


ic.oup.com/intim


were then resuspended in ALDEFLUOR assay buffer contain-ing ALDEFLUOR substrate (150 nM) with or without the ALDH inhibitor DEAB (100 µM; Sigma) and incubated for 30 min at 37°C.

Treg cell differentiation assaySorted E2-2high and E2-2low CDPs were cultured in the pres-ence of hFlt3L (30 ng ml−1) with or without mGM-CSF (50 ng ml−1) for 7 days, after which pDCs or CD11c+MHC class II+ cDCs derived from E2-2high and E2-2low CDPs were sorted out and pulsed with 1 µM ovalbumin peptide for 1 h. The CD4+CD62L+CD44− naive T cells were isolated from OT-II mice. The DC (1 × 103 cells) and naive CD4+ T cells (1 × 104 cells) were co-cultured in the presence of 3 ng ml−1 hTGFβ1 (Peprotech) and 100 U ml−1 IL-2 (Biolegend) for 5 days. Foxp3+ Treg cells were detected by staining with monoclonal antibodies against CD4 (FITC), Foxp3 (Fjk-16; Biotin, eBio-science) and streptavidin-APC (Biolegend) according to the manufacturers’ instructions.

Statistical analysisThe statistical significance of the obtained values was ana-lyzed by Student’s two-tailed t-test. A P value <0.05 was con-sidered significant.

Results

E2-2 is essential for CDP differentiation into pDCsChimeric mice reconstituted with E2-2−/− FL cells completely lack pDC differentiation both in vivo and ex vivo and accu-mulate aberrant CD135+CD11c+Ly6c+PDCA-1− cells in the BM (23). However, it is not known whether the number of CDPs and other progenitors is altered in the chimeric mice. We generated the same chimeric mice by transplanting E2-2−/− and E2-2+/+ FL cells into lethally irradiated mice (hereafter E2-2−/− and E2-2+/+ chimeras) (Fig. 1). The E2-2−/− and E2-2+/+ chimeras were comparable in both the percentage and the total numbers of CD115− and CD115+ CDPs (Fig. 1A, B and E). To examine the developmental potential of CDPs in the E2-2−/− and E2-2+/+ chimeras, we sorted CDPs from the BM of these mice and cultured the cells in the presence of Flt3L for 8 days. As expected, the CDPs from E2-2+/+ chimeras most efficiently differentiated into pDCs and cDCs, respect-ively, and the cDCs derived from both CDP subsets appropri-ately contained both CD24high and CD24low subsets (Fig. 1C and F). In contrast, both CDPs from E2-2−/− chimeras failed to produce pDCs, and solely gave rise to cDCs composed of CD24high and CD24low subpopulations (Fig. 1D and F). In addition, both the percentage and the total numbers of HSCs and other progenitor cells, including long-term and short-term HSCs, multipotent progenitors (MPPs), common myeloid progenitors (CMPs), granulocyte-macrophage progenitors (GMPs), megakaryocyte-erythroid progenitors (MEPs), com-mon lymphoid progenitors (CLPs), macrophage and DC pro-genitors (MDPs), common monocyte progenitors (cMoPs), and pre-cDCs, were unaltered in E2-2−/− chimeras, compared with E2-2+/+ chimeras (Supplementary Figure S1, available at International Immunology Online). We also confirmed that

CD135+CD11c+Ly6c+PDCA-1− cells accumulated in the BM of the E2-2−/− chimeras (data not shown) as reported previ-ously (23). These results suggested that E2-2 has an essen-tial role in CDP differentiation into pDCs.

Visualization and identification of pDC-committed progenitorsTo visualize E2-2 expression and examine the pDC differ-entiation potential of E2-2-expressing cells, we generated BAC Tg mice carrying the KuOr fluorescence reporter gene under control of the E2-2-promoter region. The KuOr cDNA was introduced into the second exon of the E2-2 gene in the BAC clone using the Red/ET recombination technique (Supplementary Figure S2A, ‘Methods’ section, available at International Immunology Online). Using the E2-2-KuOr BAC Tg mice (hereafter E2-2-KuOr mice), we first counted DC-related progenitors/precursors in the BM (Supplementary Figure 2B, available at International Immunology Online) and DC subsets and monocytes in the spleen (Supplementary Figure 2C, available at International Immunology Online) of the E2-2-KuOr mice. There were no differences between WT and E2-2-KuOr mice in the numbers of macrophage and DC progenitors (MDPs), CDPs and pre-cDCs and pre-pDCs in the BM, or in the number of DCs and monocytes in the spleen (Supplementary Figure 2B and C, available at International Immunology Online).

We next evaluated the E2-2-KuOr expression levels in Lin–

CD117lowCD135− cells, MDPs and CDPs by multicolor flow cytometry. E2-2-KuOr florescence signals were high in CDPs, intermediate in MDPs and low in Lin–CD117lowCD135－cells (Fig. 2A), consistent with each subset’s potential to develop into pDCs (15). Based on the E2-2-KuOr expression level, we subdivided CDPs into three fractions, namely, E2-2-KuOrhigh (R1), E2-2-KuOrint (R2) and E2-2-KuOrlow (R3) (Fig. 2B), and used qPCR to confirm that the KuOr mRNA expression lev-els were correlated with E2-2 transcription (Fig. 2C). These results enabled us to identify pDC progenitors within the CDPs. We then sorted the R1–R3 fractions and cultured them in the presence of Flt3L for 8 days (Fig. 2D–F). Notably, the R1 fraction exclusively gave rise to pDCs expressing E2-2-KuOr, Siglec-H and CD45RA, but not CD11b. The R2 frac-tion gave rise to both pDCs and cDCs, while most of the R3 progenies were cDCs. These results clearly indicated that the E2-2 expression level in CDPs determines their pDC develop-mental potential, and that E2-2high CDPs are pDC-committed progenitors.

Non-CDPs with pDC developmental potential in the BMWe used the E2-2-KuOr mice to search for other progenitors in the BM that could differentiate into pDCs. Since DC devel-opmental potential is confined to the Flt3+ fraction in the BM (11, 29, 30), E2-2high (R1) and E2-2low cells (R2) were sorted from Lin−Flt3+ cells and cultured in the presence of Flt3L (Fig. 3A and B). Importantly, Lin−Flt3+E2-2high cells prominently gave rise to pDCs, whereas Lin−Flt3+E2-2low cells mainly produced cDCs (Fig. 3B and E), suggesting that the E2-2 level in Lin−Flt3+ cells reflects their pDC differentiation potential regardless of the CD117 level. In this context, the Lin−Flt3+ cells can be subdivided into CD117+ and CD117low/int fractions; the former

Flexible fate in E2-2high CDPs 445

Dow


ic.oup.com/intim


Fig. 1. E2-2 is essential for the pDC differentiation potential but not for the generation of CDPs. (A, B, E) Flow cytometry profiles (A and B) and absolute numbers (E) of CD115− and CD115+ CDPs in the BM of B6.SJL mice (CD45.1) 8 weeks after the transplantation of 2 × 106 FL cells of E2-2+/+ (A) and E2-2−/− (B) mice (CD45.2). (C, D, F) The ex vivo DC differentiation potential of CD115− and CD115+ CDPs from mice after FL cell transplantation as in A and B. Cells (2 × 104) were cultured in the presence of human Flt3L (30 ng ml−1). (C, D) Flow cytometry profiles and (F) absolute numbers of pDC (CD45RA+CD11cint) and cDC subpopulations (CD24high and CD24low cDCs) on day 8 of culture. Data are representa-tive of three independent experiments. Error bars show the mean ± SEM. *P < 0.01.


Dow


ic.oup.com/intim


Fig. 2. Identification of pDC-committed progenitors in CDPs. (A) E2-2-KuOr expression in Lin−CD117−CD135− cells, CDPs and MDPs in the BM of E2-2-KuOr mice. (B) Flow cytometric sorting of CDP subpopulations. Boxed areas: R1 fraction, E2-2high CDPs; R2, E2-2int CDPs; R3, E2-2low CDPs. (C) KuOr and E2-2/Tcf4 mRNA levels in the R1–3 fractions were analyzed by qPCR. (D, E) Ex vivo DC differentiation from sorted R1, R2 and R3. Cells (3 × 104) from R1–3 were cultured in the presence of human Flt3L (30 ng ml−1) for 8 days. Flow cytometry profiles (D) and absolute numbers (E) of pDCs (PDCA-1+Siglec-H+CD11cint) and cDC subpopulations (CD24highcDCs and CD24lowcDCs). (F) CD45RA or CD11b expres-sion (shaded) and corresponding isotype controls (open) on pDCs derived from E2-2high CDPs. Data are representative of three independent experiments. Error bars show the mean ± SEM. *P < 0.001, **P < 0.005.


Dow


ic.oup.com/intim


Fig. 3. Restriction of pDC developmental potential to E2-2high cells in the BM. (A) E2-2-KuOr expression of Lin−CD135+ cells from the BM of E2-2-KuOr mice. (B, E) Ex vivo DC differentiation from sorted R1 and R2 fractions. Cells were cultured in the presence of human Flt3L (30 ng ml−1). (B) Flow cytometry profiles and (E) absolute numbers of pDC (PDCA-1+Siglec-H+CD11cint) and cDC subpopulations (CD24high cDCs and CD24low cDCs) on day 8 of culture. (C) E2-2-KuOr and CD110 (c-Mpl) expression by LMPPs (Lin−CD117+Sca-1+CD34+CD135+) (R3) and Lin−CD117+Sca-1+CD34−CD135− cells. (D) E2-2highCD110+LMPPs + FL with or without TPO (30 ng ml−1). Flow cytometry profiles (D) and absolute numbers (F) of pDC and cDC subpopulations on day 8 of culture. Data are representative of three independent experiments. Error bars show the mean ± SEM. *P < 0.01, **P < 0.05.


Dow


ic.oup.com/intim


includes lymphoid-primed multipotent progenitors (LMPPs) (31), CMPs (30) and MDPs (19), and the latter consist of CDPs and CLPs (15). Therefore, we further examined LMPPs, the lineage level at which progenitors with DC developmen-tal potential branch off (13, 14). We divided LMPPs, defined as Lin−CD117+Sca-1+CD34+CD135+ cells, by E2-2-KuOr and by the thrombopoietin (TPO) receptor, also known as CD110, which is encoded by c-Mpl (Fig. 3C). The CD110+E2-2high LMPPs were cultured in the presence of Flt3L alone or Flt3L with TPO, which enhances pDC differentiation from CDPs (15). Notably, the CD110+E2-2high LMPPs exclusively differentiated into pDCs, and the number of pDCs increased in the presence of Flt3L plus TPO (Fig. 3D and F). These results suggested that E2-2high non-CDPs in Lin−Flt3+ cells—in other words, E2-2high LMPPs—are pDC-committed progenitors.

Developmental potential of E2-2high CDPs in vivoTo evaluate the developmental potential of E2-2high CDPs in vivo, we generated DC-depleted chimeric mice, in which injected progenitor-derived DCs find appropriate spaces to inhabit. In brief, B6.SJL mice (CD45.1+CD45.2−) were lethally irradiated and reconstituted with the BM cells of B6.SJL × CD11c-DTR26 F1 mice (CD45.1+CD45.2+) (Fig. 4A). Eight weeks after the transplantation, E2-2high and E2-2low CDPs (2 × 104) from WT mice (CD45.1−CD45.2+) were injected into DT-treated chimeras (hereafter DC-depleted mice), and we analyzed the cells’ progenies in the spleen 8 days later (Fig. 4B and D). Consistent with the results of the ex vivo experi-ments, E2-2high CDPs exclusively gave rise to pDCs (Siglec-H+CD11cint), and E2-2low CDPs mainly gave rise to cDCs (Siglec-H−CD11c+), which consisted of cDC1s and cDC2s in the spleen (Fig. 4B and D). Notably, the E2-2high CDPs pro-duced 10 times more pDCs than did the E2-2low CDPs (Fig. 4D).

We also analyzed progenies of the injected cells in the small intestine. Surprisingly, the E2-2high CDPs, which only dif-ferentiated into pDCs in the spleen, differentiated partly into cDCs, such as CD103+CD11b− and CD103+CD11b+ cDCs, in the small intestine (Fig. 4C, upper panels and Fig. 4E). As expected, E2-2low CDPs differentiated into CD103+CD11b− and CD103+CD11b+ cDCs in the small intestine (Fig. 4C, lower panels and Fig. 4E) (32). The expression levels of CD80 and E2-2-KuOr in the cDC progenies derived from E2-2high and E2-2low CDPs were comparable while the level of CD86 was slightly higher in the cDC subset derived from E2-2low CDPs compared with those in the cDC subsets derived from E2-2high CDPs (Supplementary Figure 3A, B and D, avail-able at International Immunology Online). Hardly any prog-enies from the injected cells were detected in the colon or liver (Supplementary Figure 3E, available at International Immunology Online). Thus, the E2-2high CDPs gave rise only to pDCs in the secondary lymphoid organs, but flexibly con-verted in part into cDCs in the small intestine.

Common βc signals drive the E2-2high CDPs’ conversion into cDCs in the small intestineTo determine the mechanisms that induce the conversion of E2-2highCDPs in the small intestine, we examined cytokines that share the βc receptor—IL-3, IL-5 and GM-CSF (hereafter βc cytokines)—because GM-CSF inhibits pDC development and

enhances cDC differentiation driven by Flt3L ex vivo (33, 34). We first measured the expression of βc cytokine mRNAs in the small intestine and other tissues by qPCR. All of the βc cytokines were enriched in the small intestine compared with other tissues (Fig. 5A). On the basis of these results, we next addressed the effect of these βc cytokines on DC differentiation ex vivo. Whole BM cells were cultured in the presence of Flt3L with or without βc cytokines. All of the βc cytokines significantly inhibited pDC development, and IL-3 and GM-CSF, but not IL-5, enhanced cDC differentiation (Fig. 5B). Stem cell factor, a βc-independent cytokine, did not affect the Flt3L-dependent DC differentiation (Fig. 5B).

To demonstrate the importance of βc cytokine signaling in converting the CDP differentiation from being toward pDCs to cDCs in the small intestine, we injected neutralizing antibod-ies against βc cytokines (IL-3, IL-5 and GM-CSF) into E2-2high CDP-transplanted chimeric mice and examined the E2-2high CDP progenies in the spleen and small intestine. Antibody treatment did not affect the balance of E2-2high CDP-derived DC subsets in the spleen (Fig. 6A), suggesting that the βc cytokines are not involved in the development of DCs in the secondary lymphoid organs (35). In contrast, the same anti-body treatment significantly inhibited the conversion from pDC to cDC differentiation and enhanced pDC differentiation in the small intestine (Fig. 6B). To further confirm the import-ance of βc cytokines in inducing this conversion from pDC to cDC, we generated E2-2-KuOr Csf2rb−/− mice. E2-2high CDPs sorted from control E2-2-KuOr Csf2rb+/+ and E2-2-KuOr Csf2rb−/− mice were transplanted into DC-depleted mice as in Fig. 4. Consistent with the results of our blocking experiments with antibodies, the conversion of pDC to cDC differenti-ation was significantly inhibited in the progenies of Csf2rb−/− E2-2high CDPs (Fig. 6C). In the small intestine, the potential of Csf2rb−/−E2-2low CDPs to develop into CD103+CD11b− cDCs was not altered, while the number of CD103+CD11b+ cDCs from Csf2rb−/−E2-2low CDPs was significantly reduced com-pared with that of Csf2rb+/+ E2-2low CDPs (Fig. 6D). In add-ition, injecting βc cytokines into the E2-2high CDP-transplanted chimeric mice significantly inhibited the E2-2high CDPs’ differ-entiation into pDCs and reciprocally promoted their develop-ment toward cDCs in the spleen (Supplementary Figure 4, available at International Immunology Online). Consistent with these results, the over-expression of GM-CSF inhibits pDC development in the spleen (36). These results suggested that βc cytokines alter the fate of E2-2high CDPs and promote their differentiation into cDCs in the small intestine.

CDPs differentiate into pDCs via CCR9− pre-pDCs. The pDC roles include some lymphocyte-like functions, such as IgH D-J rearrangement (37), which is detected in CCR9+ pDCs but not in CCR9− pre-pDCs or CDPs (15, 20). To determine the cellular origin of the converted cDCs in the small intestine, we used PCR to detect IgH D-J rearrange-ment in E2-2high CDPs, CCR9− pre-pDCs and CCR9+ pDCs in the BM, and in E2-2high CDP-derived DC progenies in the small intestine of E2-2high CDP-transplanted chimeric mice. Notably, IgH rearrangement was first detectable in the CCR9+ pDCs and pDCs in all tissues examined, including the small intestine, but was not detected in the earlier pro-genitors (i.e. E2-2high CDPs and CCR9− pre-pDCs) in the BM,


Dow


ic.oup.com/intim


Fig. 4. In vivo developmental potential of E2-2high and E2-2low CDPs. (A) BM cells from B6.SJL × CD11c-DTR F1 mice (CD45.1+CD45.2+) were transplanted into lethally irradiated B6.SJL (CD45.1+CD45.2+) mice. Eight weeks later, DT was repeatedly injected into the chimeric mice to deplete DC subsets. On the next day, 5 × 104 sorted E2-2high CDPs or E2-2low CDPs (both were CD45.1−CD45.2+) were transplanted into the DC-depleted chimeric mice. The progenies of these cells were analyzed on day 8. (B) Flow cytometry profiles of the spleen 8 days after intra-BM transplantation of E2-2high CDPs and E2-2low CDPs (both were CD45.1−CD45.2+) into DC-depleted chimeric mice. Progenies were stained for CD45.1, CD45.2, CD11c, Siglec-H, CD8α and CD11b. (C) Flow cytometry profiles of the small intestine 8 days after intra-BM transplantation of 5 ×104 sorted E2-2high CDPs and E2-2low CDPs (both were CD45.1−CD45.2+) into DC-depleted chimeras (CD45.1+CD45.2+). Progenies were stained for CD45.1, CD45.2, CD11c, Siglec-H, CD11b and CD103. (D, E) Absolute cell counts of DC subsets in (D) splenic and (E) small intes-tine progenies from E2-2high CDPs and E2-2low CDPs, respectively. Data are representative of three independent experiments. Error bars show the mean ± SEM. *P < 0.001.


Dow


ic.oup.com/intim


or in cDCs in the spleen and small intestine. Importantly, this activity was detected in the pDCs but not cDCs derived from E2-2highCDPs in the small intestine (Fig. 6E), suggest-ing that the E2-2high CDP-derived cDCs in the small intestine were of CCR9− pre-pDC origin. To confirm the contribution of CCR9− pre-pDCs in vivo, we sorted CCR9− pre-pDCs and CCR9+ pDCs from the BM of WT mice and transplanted them separately into DC-depleted mice. Both CCR9− pre-pDCs and CCR9+ pDCs gave rise only to pDCs in the spleen (Fig. 6F). In contrast, CCR9− pre-pDCs gave rise not

only to pDCs, but also to a number of CD103+CD11b− and CD103+CD11b+ cDCs in the small intestine (Fig. 6G and H). The expression levels of CD80 and E2-2-KuOr in the cDC progenies derived from CCR9− pre-pDCs and E2-2low CDPs were comparable, while the level of CD86 in the CCR9− pre-pDC-derived cDCs was slightly lower than that in the E2-2low CDP-derived cDCs, and instead resembled that in the E2-2high CDP-derived cDCs (Supplementary Figure 3A–D, available at International Immunology Online). In this context, we could not detect any CCR9+ pDC-derived

Fig. 5. βc-related cytokines drive pDC progenitor differentiation into cDCs. (A) The mRNA expression of βc-related cytokines in various tissues. Total RNA was isolated from the small intestine (SI: 4 parts), large intestine (LI, 2 parts), BM, spleen (SP), lymph node (LN), liver (Liv), kidney (Kid) and brain (BR) and subjected to cDNA synthesis. The mRNA levels of βc-related cytokines (IL-3, IL-5 and GM-CSF) were evaluated by qPCR. Data were normalized to GAPDH expression. Data are representative of three independent experiments. (B) Whole BM cells were cul-tured in the presence of Flt3L with or without βc-related cytokines (IL-3, IL-5 and GM-CSF), and DC progenies were analyzed on day 8. Data represent the mean ± SD from three independent experiments. *P < 0.01, **P < 0.05.


Dow


ic.oup.com/intim


Fig. 6. Common βc signals drive the differentiation of E2-2high CDPs into cDCs in the small intestine. (A, B) Sorted E2-2high CDPs (5 × 104) were transplanted into the BM cavity of DC-depleted mice, and the mice were treated with control antibodies or neutralizing antibodies against βc-related cytokines (IL-3, IL-5 and GM-CSF). Bars show the percentage of donor cell-derived DC subsets in the spleen (A) and small intestine (B). (C, D) E2-2high (C) or E2-2low (D) CDPs from the E2-2-KuOr/Csf2rb+/+ or E2-2-KuOr/Csf2rb−/− mice were transplanted into the BM cavity of DC-depleted mice. Bars show the percentage of donor cell-derived DC subsets in the small intestine. (E) Genomic DNA was isolated from E2-2high CDPs, CCR9− pre-pDCs and CCR9+ pDCs in the BM; pDCs in the peripheral blood (PB); pDCs and cDCs derived from E2-2high CDPs in the small intestine (SI) of E2-2high CDP-transplanted chimeric mice; and DC subsets in the small intestine and spleen (SP) of WT mice. DH-JH rearrangement was evaluated by PCR. (F–H) Flow cytometry profiles of the (F) spleen and (G) small intestine 5 days after the transplantation of 5 × 105 sorted CCR9− pre-pDCs or CCR9+ pDCs into the BM cavity of DC-depleted mice. Bars show the percentage of donor cell-derived DC subsets in the small intestine (H). Data are representative of three independent experiments. Error bars show the mean ± SEM. *P < 0.01, **P < 0.05.


Dow


ic.oup.com/intim


Fig. 7. E2-2high CDP-derived cDCs in the small intestine express ALDH and induce Treg cells ex vivo. (A) Flow cytometry profile and (B) per-centage of ALDEFLUOR+ cells in the E2-2high CDP-derived CD103+CD11b− cDCs, CD103+CD11b+ cDCs and pDCs in the small intestine of mice prepared as in Fig. 6C and E. (C) Flow cytometry profile of ALDEFLUOR+ cells in E2-2high and E2-2low CDP-derived CD11c+MHC class II+ cDCs ex vivo. The cells were incubated with ALDEFLUOR in the absence or presence of the ALDFLUOR inhibitor DEAB. (D) Flow cytometry profile and (E) percentage of Foxp3+ Treg cells differentiated from naive OT-II cells that were co-cultured with cDCs or pDCs derived from E2-2high CDPs, and cDCs derived from E2-2low CDPs in the presence of OVA peptide (1 µM), human TGF-β1 (3 ng ml−1) and IL-2 (100 U ml−1) ex vivo. Foxp3+ Treg cells were analyzed on day 5. Data are representative of three independent experiments. Error bars show the mean ± SEM. *P < 0.01.


Dow


ic.oup.com/intim


cells in the small intestine (Fig. 6G and H). On the basis of these results, we concluded that the converted cDCs were derived from CCR9− pre-pDCs, a downstream progenitor of E2-2high CDPs.

E2-2high CDP-derived cDCs express ALDH and induce Treg cell differentiationIn gut-associated lymphoid tissue, including the small intes-tine, CD103+ DCs are critically involved in inducing Treg cells from naive T cells by producing retinoic acid (RA) and TGF-β (38, 39). ALDH converts retinal into RA, and GM-CSF induces ALDH expression in CD103+ DCs (40). Therefore, we examined the ALDH activity in E2-2high CDP-derived cDCs in the lamina propria of the small intestine. We detected ALDH activity in both CD103+CD11b− and CD103+CD11b+ cDCs, which were derived from E2-2high CDPs, and there were more ALDH+ cells among the CD103+CD11b− cDCs than the CD103+CD11b+ cDCs (Fig. 7A, upper panels and Fig. 7B). The ALDH activity was abolished by adding the ALDH inhibitor DEAB (Fig. 7A, lower panels). The pDCs derived from E2-2high CDPs did not have ALDH activity (Fig. 7A and B). In this context, the ALDH activity in the cDC subsets derived from CCR9− pre-pDCs was slightly lower than in those derived from E2-2low CDPs, and in-stead resembled the ALDH activity in the E2-2high CDP-derived cDCs (Fig. 7A and B; Supplementary Figure 5, available at International Immunology Online). Likewise, the cDCs derived from E2-2high CDPs in the presence of Flt3L and GM-CSF ex vivo showed ALDH activity (Fig. 7C). Finally, we co-cultured the cDCs with naive T cells to assess their capacity to induce Treg cells. The cDCs, but not pDCs derived from E2-2high CDPs in the presence of Flt3L and GM-CSF, efficiently induced Foxp3+ Treg cells, which was comparable to the Treg cell induction by cDCs derived from E2-2low CDPs ex vivo (Fig. 7D and E). Collectively, these results indicate that E2-2high CDP-derived cDCs acquire ALDH activity and induce Treg cells under the influence of micro-environments that are rich in βc-dependent cytokines, such as the intestine.

Discussion

We previously identified two distinct DC-restricted progenitors, CD115+ and CD115− CDPs (13, 15). In the present study, we used E2-2 reporter mice to further identify E2-2high progeni-tors within the CDPs. E2-2high CDPs strictly gave rise to pDCs ex vivo and in the secondary lymphoid organs in vivo, while E2-2low CDPs mainly gave rise to cDCs. Recently, cDC1- and cDC2-restricted progenitors were identified among pre-cDC progenitors (16, 17). Taking these findings together with our present study, we propose the following model: CDPs are sub-divided into E2-2high and E2-2low cells, of which the E2-2high cells differentiate into CCR9+ pDCs via CCR9− pre-pDCs while the E2-2low cells give rise to pre-cDCs. The pre-cDCs further differ-entiate into cDC1s and cDC2s via pre-cDC1s and pre-cDC2s, respectively.

E2-2 is essential for the development and maintenance of pDCs in both mice and humans (23). We confirmed the importance of E2-2 in pDC development in chimeric mice reconstituted with E2-2−/− FL cells. The development of HCS and HSPCs was not altered in the chimeric mice; in particular,

both CD115+ and CD115− CDPs were present in normal num-bers in the BM. However, CDPs derived from the E2-2−/− FL cells could not develop into pDCs and gave rise solely to cDC subsets, implying that the E2-2 expression in CDPs was im-portant for their pDC differentiation potential. By generating and analyzing E2-2-KuOr mice, we found that pDC develop-ment was regulated by a gradient of E2-2 expression in CDPs: E2-2high CDPs gave rise to only pDCs, E2-2int CDPs gave rise to both pDCs and cDCs, and E2-2low CDPs gave rise mainly to cDCs. These results are consistent with the dogma that TFs determine the cell lineage and that their expression level and activity are critical for cell fate determination (41). In this con-text, the commitment toward DC lineages occurs at the LMPP level (42). In this study, we found pDC progenitors among the LMPPs (i.e. E2-2high LMPPs) in the BM of E2-2-KuOr mice, which might identify the first point at which DC development branches from other lineages in hematopoiesis.

The wide and multi enhancer regions for E2-2/Tcf4 gene have recently been reported (43). In this context, our E2-2-KuOr BAC Tg does not contain the enhancer regions, suggesting that the brightness of KuOr expression is not optimized. Nevertheless, the fluorescence reporter KuOr mRNA expression levels in CDPs and other progenitors were clearly correlated with E2-2 transcription (Fig. 2C) and their pDC developmental potential (Fig. 2B–E, 3, and 4), likely due to the activation of E2-2 transcrip-tion by E2-2 itself (43). Indeed, multiple E2-2 binding sites were identified in the 5’ region of our E2-2-KuOr BAC Tg (data not shown). Interestingly, E2-2high CDPs gave rise not only to pDCs, but also to CD103+CD11b− and CD103+CD11b+ cDC sub-sets in the small intestine. This conversion was induced by βc cytokines, which are abundant in the small intestine. Supporting this notion, CCR9− pre-pDCs have developmental plasticity, and GM-CSF, produced by mouse epithelial cell line, inhibits E2-2 expression in CCR9− pre-pDCs and promotes their differenti-ation into cDCs ex vivo (20). A standing question in hemato-poiesis is whether a particular cytokine provides an instructive signal that affects progenitors’ developmental direction (44). For example, an ectopic GM-CSF signal on CLPs induces cell fate conversion into the myeloid lineage (45). Enforced Flt3 expres-sion on MEPs induces their differentiation into DCs and myeloid cells (11). Lymphocytic choriomeningitis virus (LCMV) infection induces the conversion of pDCs into cDCs in the BM (46). We found that pDC-committed E2-2high CDPs differentiated into cDC subsets in the small intestine in a βc-cytokine-dependent man-ner. Given that the converted CD103+ cDCs expressed ALDH, which is an essential enzyme for RA synthesis, and induced Treg cells ex vivo, our findings imply that this conversion was accom-plished by an environmentally pertinent tuning machinery and is relevant for inducing Treg cells and maintaining gut homeostasis.

Supplementary data

Supplementary data are available at International Immunology Online.

Funding

This work was supported by the Naito Foundation (N.O.), the Uehara Memorial Foundation (N.O.), the NOVARTIS founda-tion (N.O.), the Takeda Science Foundation (N.O. and T.O.),


Dow


ic.oup.com/intim


a Grant-in-Aid for Scientific Research (A) from the Ministry of Education, Science, Sports and Culture of Japan (T.O.) and the Joint Usage/Research Program of the Medical Research Institute, Tokyo Medical and Dental University (N.O.).

Acknowledgements

We thank H. Kamioka for secretarial support, Y. Zhuang (Duke University) for E2-2+/– mice, S. Jung (Weissmann Institute of Science) for CD11c-DTR mice and R. Nishinakamura (Kumamoto Universiity) for Csf2rb–/– mice. N.O. conceived the study, performed experiments, ana-lyzed data and wrote part of the manuscript. J.A., R.K. and S.K. assisted N.O. T.O. supervised the overall project and wrote the manuscript.

Conflicts of interest statement: the authors declared no conflicts of interest.

References

1 Lennert, K. and Remmele, W. 1958. Karyometric research on lymph node cells in man. I. Germinoblasts, lymphoblasts & lym-phocytes. Acta Haematol. 19:99.

2 Facchetti, F. C., de Wolf-Peeters, D. Y., Mason, K. et al. 1998. Plasmacytoid T cells. Immunohistochemical evidence for their monocyte/macrophage origin. Am. J. Pathol. 133:15.

3 Siegal, F. P., Kadowaki, N., Shodell, M. et al. 1999. The nature of the principal type 1 interferon-producing cells in human blood. Science 284:1835.

4 Colonna, M., Trinchieri, G. and Liu, Y. J. 2004. Plasmacytoid den-dritic cells in immunity. Nat. Immunol. 5:1219.

5 Liu, Y. J. 2005. IPC: professional type 1 interferon-producing cells and plasmacytoid dendritic cell precursors. Annu. Rev. Immunol. 23:275.

6 Gilliet, M., Cao, W. and Liu, Y. J. 2008. Plasmacytoid dendritic cells: sensing nucleic acids in viral infection and autoimmune dis-eases. Nat. Rev. Immunol. 8:594.

7 Shortman, K. and Naik, S. H. 2007. Steady-state and inflammatory dendritic-cell development. Nat. Rev. Immunol. 7:19.

8 Maraskovsky, E., Brasel, K., Teepe, M. et al. 1996. Dramatic increase in the numbers of functionally mature dendritic cells in Flt3 ligand-treated mice: multiple dendritic cell subpopulations identified. J. Exp. Med. 184:1953.

9 McKenna, H. J., Stocking, K. L., Miller, R. E. et al. 2000. Mice lack-ing flt3 ligand have deficient hematopoiesis affecting hematopoi-etic progenitor cells, dendritic cells, and natural killer cells. Blood 95:3489.

10 Miller, G., Pillarisetty, V. G., Shah, A. B., Lahrs, S. and DeMatteo, R. P. 2003. Murine Flt3 ligand expands distinct dendritic cells with both tolerogenic and immunogenic properties. J. Immunol. 170:3554.

11 Onai, N., Obata-Onai, A., Tussiwand, R., Lanzavecchia, A. and Manz, M. G. 2006. Activation of the Flt3 signal transduction cas-cade rescues and enhances type I interferon-producing and den-dritic cell development. J. Exp. Med. 203:227.

12 Waskow, C., Liu, K., Darrasse-Jèze, G. et al. 2008. The receptor tyrosine kinase Flt3 is required for dendritic cell development in peripheral lymphoid tissues. Nat. Immunol. 9:676.

13 Onai, N., Obata-Onai, A., Schmid, M. A., Ohteki, T., Jarrossay, D. and Manz, M. G. 2007. Identification of clonogenic common Flt3+M-CSFR+ plasmacytoid and conventional dendritic cell pro-genitors in mouse bone marrow. Nat. Immunol. 8:1207.

14 Naik, S. H., Sathe, P., Park, H. Y. et al. 2007. Development of plas-macytoid and conventional dendritic cell subtypes from single precursor cells derived in vitro and in vivo. Nat. Immunol. 8:1217.

15 Onai, N., Kurabayashi, K., Hosoi-Amaike, M. et al. 2013. A clo-nogenic progenitor with prominent plasmacytoid dendritic cell developmental potential. Immunity 38:943.

16 Schlitzer, A., Sivakamasundari, V., Chen, J. et al. 2015. Identification of cDC1- and cDC2-committed DC progenitors reveals early lineage priming at the common DC progenitor stage in the bone marrow. Nat. Immunol. 16:718.

17 Grajales-Reyes, G. E., Iwata, A., Albring, J. et al. 2015. Batf3 maintains autoactivation of Irf8 for commitment of a CD8a+ con-ventional DC clonogenic progenitor. Nat. Immunol. 16:708.

18 Naik, S. H., Metcalf, D., van Nieuwenhuijze, A. et al. 2006. Intrasplenic steady-state dendritic cell precursors that are distinct from monocytes. Nat. Immunol. 7:663.

19 Liu, K., Victora, G. D., Schwickert, T. A. et al. 2009. In vivo analysis of dendritic cell development and homeostasis. Science 324:392.

20 Schlitzer, A., Loschko, J., Mair, K. et al. 2011. Identification of CCR9− murine plasmacytoid DC precursors with plasticity to dif-ferentiate into conventional DCs. Blood 117:6562.

21 Merad, M. and Manz, M. G. 2009. Dendritic cell homeostasis. Blood 113:3418.

22 Watowich, S. S. and Liu, Y. J. 2010. Mechanisms regulating den-dritic cell specification and development. Immunol. Rev. 238:76.

23 Cisse, B., Caton, M. L., Lehner, M. et al. 2008. Transcription factor E2-2 is an essential and specific regulator of plasmacytoid den-dritic cell development. Cell 135:37.

24 Nagasawa, M., Schmidlin, H., Hazekamp, M. G., Schotte, R. and Blom, B. 2008. Development of human plasmacytoid dendritic cells depends on the combined action of the basic helix-loop-helix factor E2-2 and the Ets factor Spi-B. Eur. J. Immunol. 38:2389.

25 Ghosh, H. S., Cisse, B., Bunin, A., Lewis, K. L. and Reizis, B. 2010. Continuous expression of the transcription factor e2-2 maintains the cell fate of mature plasmacytoid dendritic cells. Immunity 33:905.

26 Zhuang, Y., Cheng, P. and Weintraub, H. 1996. B-lymphocyte development is regulated by the combined dosage of three basic helix-loop-helix genes, E2A, E2-2, and HEB. Mol. Cell. Biol. 16:2898.

27 Jung, S., Unutmaz, D., Wong, P. et al. 2002. In vivo deple-tion of CD11c+ dendritic cells abrogates priming of CD8+ T cells by exogenous cell-associated antigens. Immunity 17:211.

28 Nishinakamura, R., Nakayama, N., Hirabayashi, Y. et al. 1995. Mice deficient for the IL-3/GM-CSF/IL-5 beta c receptor exhibit lung pathology and impaired immune response, while beta IL3 receptor-deficient mice are normal. Immunity 2:211.

29 D’Amico, A. and 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. 198:293.

30 Karsunky, H., Merad, M., Cozzio, A., Weissman, I. L. and Manz, M. G. 2003. Flt3 ligand regulates dendritic cell development from Flt3+ lymphoid and myeloid-committed progenitors to Flt3+ den-dritic cells in vivo. J. Exp. Med. 198:305.

31 Adolfsson, J., Månsson, R., Buza-Vidas, N. et al. 2005. Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryocytic potential a revised road map for adult blood lin-eage commitment. Cell 121:295.

32 Bogunovic, M., Ginhoux, F., Helft, J. et al. 2009. Origin of the lam-ina propria dendritic cell network. Immunity 31:513.

33 Gilliet, M., Boonstra, A., Paturel, C. et al. 2002. 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.

34 Esashi, E., Wang, Y. H., Perng, O., Qin, X. F., Liu, Y. J. and Watowich, S. S. 2008. The signal transducer STAT5 inhibits plas-macytoid dendritic cell development by suppressing transcription factor IRF8. Immunity 28:509.

35 Vremec, D., Lieschke, G. J., Dunn, A. R., Robb, L., Metcalf, D. and Shortman, K. 1997. The influence of granulocyte/macrophage colony-stimulating factor on dendritic cell levels in mouse lymph-oid organs. Eur. J. Immunol. 27:40.

36 Zhan, Y., Vega-Ramos, J., Carrington, E. M., Villadangos, J. A., Lew, A. M. and Xu, Y. 2012. The inflammatory cytokine, GM-CSF, alters the developmental outcome of murine dendritic cells. Eur. J. Immunol. 42:2889.

37 Corcoran, L., Ferrero, I., Vremec, D. et al. 2003. The lymphoid past of mouse plasmacytoid cells and thymic dendritic cells. J. Immunol. 170:4926.

38 Coombes, J. L., Siddiqui, K. R., Arancibia-Cárcamo, C. V. et al. 2007. A functionally specialized population of mucosal CD103+ DCs induces Foxp3+ regulatory T cells via a TGF-beta and retinoic acid-dependent mechanism. J. Exp. Med. 204:1757.


Dow


ic.oup.com/intim


39 Sun, C. M., Hall, J. A., Blank, R. B. et al. 2007. Small intestine lamina propria dendritic cells promote de novo generation of Foxp3 Treg cells via retinoic acid. J. Exp. Med. 204:1775.

40 Yokota, A., Takeuchi, H., Maeda, N. et al. 2009. GM-CSF and IL-4 synergistically trigger dendritic cells to acquire retinoic acid-pro-ducing capacity. Int. Immunol. 21:361.

41 Orkin, S. H. 2000. Diversification of haematopoietic stem cells to specific lineages. Nat. Rev. Genet. 1:57.

42 Naik, S. H., Perié, L., Swart, E. et al. 2013. Diverse and heritable lin-eage imprinting of early haematopoietic progenitors. Nature 496:229.

43 Grajkowska, L. T., Ceribelli, M., Lau, C. M. et al. 2017. Isoform-specific expression and feedback regulation of E

protein TCF4 control dendritic cell lineage specification. Immunity 46:65.

44 Metcalf, D. 1999. Stem cells, pre-progenitor cells and lineage-committed cells: are our dogmas correct? Ann. NY Acad. Sci. 872:289; discussion 303.

45 Kondo, M., Scherer, D. C., Miyamoto, T. et al. 2000. Cell-fate con-version of lymphoid-committed progenitors by instructive actions of cytokines. Nature 407:383.

46 Zuniga, E. I., McGavern, D. B., Pruneda-Paz, J. L., Teng, C. and Oldstone, M. B. 2004. Bone marrow plasmacytoid dendritic cells can differentiate into myeloid dendritic cells upon virus infection. Nat. Immunol. 5:1227.


Dow


ic.oup.com/intim


Flexible fate commitment of E2-2high common DC progenitors ...

Documents

Transcript of Flexible fate commitment of E2-2high common DC progenitors ...