Cutting Edge: ACVRL1 Signaling Augments CD8 α+ Dendritic Cell Development Rohit Verma, Hemant Jaiswal, Kuldeep Singh Chauhan, Monika Kaushik and Prafullakumar Tailor This information is current as of September 29, 2021. J Immunol 2016; 197:1029-1034; Prepublished online 15 July 2016; doi: 10.4049/jimmunol.1501849 http://www.jimmunol.org/content/197/4/1029 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2016/07/14/jimmunol.150184 Material 9.DCSupplemental

References This article cites 31 articles, 12 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/197/4/1029.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 29, 2021

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2016 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Th eJournal of Cutting Edge Immunology

Cutting Edge: ACVRL1 Signaling Augments CD8a+ Dendritic Cell Development Rohit Verma, Hemant Jaiswal, Kuldeep Singh Chauhan, Monika Kaushik, and Prafullakumar Tailor Dendritic cells (DCs) are a collection of different sub- netic proteins [BMPs], growth-differentiation factor, NODAL, types, each of which is characterized by specific surface activins/inhibins, and anti-Mullerian€ hormone) involves com- markers, -expression patterns, and distinct func- binationsoftwotypesofreceptors:typeI(alsoknownas tions. Members of the IFN regulatory factor family play activin –like [ALKs]1–7]) and type II (TGF- critical roles in DC development and functions. Re- bRII, II, activin receptor IIB, bone morpho- cently, Irf8 was shown to activate TGF-b signaling, genetic receptor II, and anti-Mullerian€ hormone receptor) which led to exacerbated neuroinflammation in the (7, 8). Activated receptor complex leads to phosphorylation of Downloaded from experimental autoimmune encephalomyelitis mouse receptor-regulated Smad proteins (R-SMADs). Specificity in model. We analyzed the effect of Irf8 on TGF-b/bone the activation of R-SMADs is defined by binding of ligand to morphogenetic protein pathway–specific in DCs specific type I receptor during signaling (e.g., SMAD1, and identified Acvrl1, a type I TGF-b superfamily re- SMAD5, and SMAD8 are activated by BMPs through ALK1–3 and ALK6, and TGF-b activates SMAD2–3 through

ceptor, as a gene strongly induced by Irf8 expression. http://www.jimmunol.org/ ALK4–5 signaling) (8). Phosphorylated R-SMADs associate Among various DC subtypes, Acvrl1 is differentially with the partner SMAD4 and move to the nucleus where this expressed in CD8a+ DCs. ACVRL1 signaling aug- a+ complex (R-SMAD and SMAD4) can interact with other mented Irf8-directed classical CD8 DC develop- partners to regulate gene transcription. Further, TGF-b sig- ment. Irf8 expression is essential for plasmacytoid naling is also shown to cross-talk with other signaling pathways a+ DC and CD8 DC development, and this study dem- (7). Thus, through multilayered signal-transduction events and onstrates that ACVRL1 signaling plays a pivotal role interactions with other signaling pathways, the TGF-b super- whereby it suppresses plasmacytoid DC development family controls many cellular events, immune functions, and a+ b while enhancing that of CD8 DCs, thus contribut- tumor metastasis (7). TGF- promotes the development of by guest on September 29, 2021 ing to DC diversity development. The Journal of DCs, and TGF-b–null mice have a defect in the development Immunology, 2016, 197: 1029–1034. of epidermal Langerhans cells (LCs) (9, 10). A recent study (11) identified another TGF-b superfamily ligand, BMP7, as endritic cells (DCs) are represented by various sub- an instructive factor for human epidermal LCs. TGF-b1was classes, each of which has differential patterns of reported to upregulate Irf8 during DC development (12). Irf8 D , anatomical distribution, and ability was shown to enhance transcription of Itgb8 gene that converts for specific cytokine production and Ag presentation (1–3). latent TGF-b to the active form at the interface of APCs and 2 2 CD4+ DC, CD8a+ DC, CD4 CD8a double-negative DC, naive T cells (13). We analyzed the role of Irf8 in the regulation and plasmacytoid DC (pDC) are major DC subtypes in spleen; of TGF-b/BMPsignalinginDCsusingacommercially equivalent populations of these subtypes can be found in dif- available PCR-based array specific to mouse TGFb/BMP sig- ferent body sites, along with tissue-specific DCs (1–4). All DC naling pathway genes. We report that Irf8 expression induces subtypes developed from common DC progenitors in bone Acvrl1 (activin receptor–like 1; also known as ALK1) + marrow (5, 6). Members of the IFN regulatory factor (IRF), gene, which is expressed specifically in the CD8a DC subtype. HLH, BATF, ETS, and STAT families of transcription factors, Further, we demonstrate that ACVRL1 signaling enhances de- + along with other transcription factors, play a critical role in velopment of the CD8a DC subtype. specific DC subtype development (3). Recent studies high- lighted the significance of Notch, Wnt, and TGF-b signaling Materials and Methods pathways in the generation of DC diversity (3). Signaling by All animal work conformed to the guidelines of the institute animal ethics ligands of the TGF-b superfamily (TGF-b,bonemorphoge- committee at the National Institute of Immunology. BXH-2 mice were

Laboratory of Innate Immunity, National Institute of Immunology, New Delhi, Delhi The online version of this article contains supplemental material. 110067, India Abbreviations used in this article: ALK, activin receptor–like kinase; BMP, bone mor- Received for publication August 19, 2015. Accepted for publication June 16, 2016. phogenetic protein; cat. no., catalog number; ChIP, chromatin immunoprecipitation; DC, dendritic cell; FL-BMDC, FLT3 ligand–derived bone marrow DC; IRF, IFN- This work was supported by the National Institute of Immunology Core Fund and a regulatory factor; LC, Langerhans cell; pDC, plasmacytoid DC; Q-PCR, quantitative Ramalingaswami Fellowship award (to P.T.). PCR; R-SMAD, receptor-regulated Smad protein; TSS, transcription start site. Address correspondence and reprint requests to Dr. Prafullakumar Tailor, Laboratory of Innate Immunity, National Institute of Immunology, Aruna Asaf Ali Marg, New Delhi, Copyright Ó 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 Delhi 110067, India. E-mail address: [email protected]

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1501849 1030 CUTTING EDGE: ACVRL1 SIGNALING ENHANCES CD8a+ DC DEVELOPMENT procured from The Jackson Laboratory and crossed with C57BL/6 female mice and splenic DC populations were analyzed by flow cytometry, as described for three generations to eliminate vertical transmission of ecotropic murine earlier (14–16). Lineage-negative cells were purified from mouse bone mar- leukemia virus. Progeny were crossed to generate mice homozygous for row using a Lineage Cell Depletion Kit (Miltenyi Biotec). CD4+ DCs, 2 2 mutation (IRF8R294C) and littermate controls (IRF8WT). Mice were used after CD8a+ DCs, CD4 CD8a DCs, and B220+ pDCs, as defined earlier (15), genotyping by PCR sequencing, and experiments were performed at 8–12 wk were purified from splenic low-density cells by FACSAria III sorter using anti- of age. FLT3 ligand–derived bone marrow DCs (FL-BMDCs), DC9 cells, CD4–allophycocyanin-Cy7 and anti-B220–allophycocyanin (BD Biosciences), Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 1. Acvrl1 gene is differentially expressed by CD8a+ DCs. (A) TGF-b/BMP pathway–specific gene-expression profiling was performed on control and 2 2 Irf8-expressing DC9 (Irf8 / background) cells. Acvrl1 gene was highly induced in the Irf8-expressing population. An increase in the Acvrl1 transcript level was validated by Q-PCR analysis of Irf8-expressing DC9 cells. Data are an average of six experiments. (B) Analysis of subtype-specific gene expression from sorted mouse splenic DC subtype populations suggested that Acvrl1 is specifically expressed in CD8a+ DCs. Data are representative of two independent experiments. (C) TGF-b/BMP pathway–specific gene-expression profiling was also performed on BMDCs from BXH-2 (Irf8R294C DC) and littermate control (Irf8WT DC) mice. Acvrl1 and Id2 genes common to analysis in (A) and (C) are shown in bold text. Results were also validated by Q-PCR analysis on BMDCs from BXH-2 (Irf8R294C) and littermate control (Irf8WT) mice. Data are an average of four individual mice. (D) DC9 cells were transduced with retroviruses expressing transcription factors (as shown). Populations were selected for 48 h with puromycin antibiotic, and Acvrl1 transcript levels were measured by Q-PCR. Data are representative of three independent experiments. (E) Schematic diagram shows the organization of human and mouse exons in the Acvrl1 gene (shown up to ∼10 kb). Blue arrows show known TSSs and red arrow shows proposed TSS in mouse. Translation start sites are shown as red bars. Exon (Ex) and Intron-Exon boundary (In-Ex)-specific Q-PCR analysis of Acvrl1 gene transcription from Irf8R294C DC and control Irf8WT DC mice (F) and from control and Irf8-expressing DC9 cells (G) also correlates with the proposed TSS. Results are representative of two independent experiments. Ex12 and Ex13 were analyzed by Q-PCR as control only to check the continuity of transcript. (H)59 RACE analysis demonstrated the expected amplicon from Irf8WT DCs, whereas Irf8R294C DCs show a very faint band correlating with very low transcription. The expected PCR amplicon can be seen in DC9 cells upon Irf8 expression (lower panel). (I)ChIPassayfrom FL-BMDCs confirmed the recruitment of IRF8 to the proposed binding sites. Results are representative of three independent experiments. (J)IRF8bindingsites(SIand SII) can be seen as minor peaks in the Acvrl1 gene from recently reported IRF8 ChIP-sequencing studies (22, 23). Error bars in graphs represent 6 SD. The Journal of Immunology 1031 anti-CD8–FITC (BioLegend), and anti-CD11c–PE (eBioscience) Abs. For (Fig. 1A). To assess the significance of Acvrl1 in DC biology, detection of SMAD phosphorylations, FL-BMDCs (day 8) were serum we examined its expression in different DC subtype pop- starved for 2 h, and stimulations were performed with 10 ng/ml TGF-b1 and 10 ng/ml BMP9 (catalog number [cat. no.] 100–21 and cat. no. 120– ulations from spleen. Each DC subtype population showed 07, respectively; PeproTech) in 10% FBS-supplemented media for 1 h. expression of a subtype-specific transcript; this study sug- Cell lysates were prepared, and equal amounts of protein were separated on gested that Acvrl1 is specifically expressed in the CD8a+ DC 4–12% NuPAGE gel (Invitrogen). Immunoblotting was performed using 463/465 subtype (Fig. 1B). To confirm the selectivity of Acvrl1 gene anti–phospho-SMAD1/5 (Ser ) and anti-SMAD1 (), + anti-SMAD2/3 (BD Biosciences), and anti–phospho-SMAD2 (Ser465/467; expression in CD8a DCs, we conducted PCR array analysis Chemicon International) Abs and detected as described earlier (14). For for BMDCs from mice carrying a BXH-2 (Irf8R294C) muta- analysis of Id2 gene induction by BMP9, FL-BMDCs and NIH3T3 cells tion in the Irf8 gene. Irf8 is essential for the development of were serum starved by culture in 0.2% FBS–supplemented media for 24 h a+ (with 100 ng/ml Flt-3L for FL-BMDCs), and transcript levels were pDCs and CD8 DCs, and mice homozygous for the R294C measured by quantitative PCR (Q-PCR) in control (PBS) or ACVRL1-Fc Irf8 mutation showed a selective block in the develop- (ALK-1Fc, cat. no. 770-MA; R&D Systems) after 2 h of BMP9 (10 ng/ml) ment of CD8a+ DCs. The PCR array for BXH-2 mice stimulation and serum supplementation (17). confirmed that Acvrl1 is indeed a CD8a+ DC–specific marker Acvrl1 (amplified from FL-BMDCs) and Bmp9 (cat. no. MG50684-M; Sino Biological) open reading frames were cloned into pMSCV-puro retro- (Fig. 1C). We (14) recently demonstrated that Irf8 increases viral vector (Clontech) (14). Retroviral gene-transfer experiments and Q-PCR Batf3 and Id2 transcript expression, and these transcription assays were performed as described earlier (14). Details of primer sequences factors have a synergistic effect on Irf8-directed classical designed for PCR analysis are described in Supplemental Fig. 1A. a+ For TGF-b/BMP pathway–specific gene-expression profiling, DC9 cells CD8 DC development. Consistent with these findings, were transduced with control and Irf8-expressing pMig-IRES-hCD8t retro- coexpression of Batf3 and Id2 led to a synergistic increase in Downloaded from viruses, and populations were purified at day 4 by magnetic sorting (14, 18). Irf8-induced Acvrl1 expression (Fig. 1D). To understand the b Similarly, to analyze the effect of BXH-2 mutation in Irf8 on TGF- /BMP regulation of Acvrl1 transcription in murine DCs, we ana- pathway–specific gene expression, FL-BMDCs from mice carrying a BXH-2 mutation (Irf8R294C DC) and littermate control mice (Irf8WT DC) were lyzed human and mouse Acvrl1 gene using the Ensembl analyzed. RNA extraction and cDNA preparation were performed per the genome browser (Fig. 1E). Based on exon-specific Q-PCR manufacturers’ instructions (RNeasy Mini Kit, cat. no. 74104; QIAGEN and analyses, we predicted that murine DCs would have a tran- Superscript II Reverse Transcriptase, cat. no. 18064; Invitrogen, respectively) and cDNA samples were subjected to RT2 Profiler PCR Array mouse TGFb/ scription start site (TSS) upstream of exon 6, correlating with http://www.jimmunol.org/ BMP signaling pathway (cat. no. PAMM-035ZC; QIAGEN). Data from two the previously reported (20) human ACVRL1 transcript independent experiments were evaluated using a threshold value of five-fold change in RT2 profiler PCR array data analyzer template provided by the manufacturer (SABiosciences/QIAGEN). 59 RACE was performed per the manufacturer’s instructions (cat. no. 18374-058; Invitrogen) using primer GSP1 for cDNA synthesis and GSP2 paired with AAP for PCR amplification. PCR amplicons were cloned in a pGEM-T Easy vector (Promega) and confirmed by sequencing. Chromatin immunoprecipitation (ChIP) was performed to confirm IRF8 binding in 7 Acvrl1 gene, following a scaled-down version (for 10 cells) of Agilent’s ChIP- by guest on September 29, 2021 on-chip Mammalian protocol (version 9.2), using anti-IRF8 Ab and normal goat IgG as a control (both from Santa Cruz Biotechnology). IRF8 binding was confirmed by Q-PCR using primers flanking the proposed binding sites (SI and SII); Hprt gene was used as negative control. The significance of the increase in CD8a+ DC frequencies or the decrease in pDC frequencies upon BMP9 supplementation was calculated using a paired one-tailed Student t test. Results and Discussion Acvrl1 is differentially expressed in CD8a+ DCs TGF-b/BMP signaling has pleiotropic effects on immune cell development and functions. BMP7 signaling was reported to play a critical role in LC development (11). A recent study (13) demonstrated that IRF8 directly controls Itgb8 gene in APCs and triggers the development of regulatory T cells and Th17 cells by activating latent TGF-b. TGF-b treatment induces Id2 gene expression in DCs (19). Id2 expression is essential for classical CD8a+ DC development (19), and we (14) recently reported that Irf8 expression leads to an increase in Id2 transcript levels. These findings suggest that Irf8 may play an important role in controlling TGF-b signaling in DCs. Hence, to study the effect of Irf8 expression on TGF- FIGURE 2. ACVRL1 signaling enhances CD8a+ DC development. DC9 b/BMP signaling, we performed our study using a commer- cells were transduced with retroviruses expressing the indicated genes and cially available TGF-b/BMP pathway–focused PCR array– selected for 48 h with puromycin. DC subtype–specific transcript levels were 2 2 based approach (QIAGEN) in DC9 (Irf8 / ) cells (14). measured by Q-PCR after puromycin selection. Puromycin-selected pop- ulations were treated with CpG (1826, 1 mg/ml) for 24 h and analyzed for Retroviral expression of Irf8 in DC9 cells led to the devel- a A a+ CD8 surface expression by flow cytometry (14). Coexpression of Acvrl1 ( ) opment of pDCs and CD8 DC-like cells with an increase and Bmp9 (B) with Irf8 led to a synergistic effect on Irf8-directed CD8a+ DC in subtype-specific transcripts (14). With this PCR array, we development concomitant with an increase in subtype-specific transcripts. identified the Acvrl1 gene (Alk1,atypeIreceptoroftheTGF-b Data from flow cytometry and Q-PCR are representative of three indepen- superfamily), which is highly induced by Irf8 in DC9 cells dent experiments. Error bars in graphs represent 6 SD. 1032 CUTTING EDGE: ACVRL1 SIGNALING ENHANCES CD8a+ DC DEVELOPMENT

FIGURE 3. BMP9-ACVRL1 signaling differentially enhances CD8a+ DC development. (A) FL-BMDC cultures supplemented with BMP9 (10 ng/ml) showed a decrease in the pDC fraction (B220+CD11c+ and B220+SiglecH+CD11c+) and an increase in the CD8a+ DC equivalent population (CD24hiMacIintCD11c+ 2 2 and B220 CD103+Sirpa ). Data are representative of independent FL-BMDC cultures from nine mice. (B) Different populations, as analyzed in (A), are represented as a bar graph. Error bars represent 6 SD. (C) FL-BMDC cultures from lineage-negative bone marrow cells supplemented with BMP9 showed a decrease in the pDC fraction (B220+CD11c+ and B220+SiglecH+CD11c+) and an increase in the CD8a+ DC equivalent population (CD24hiMacIintCD11c+). Data are representative of two independent experiments.

(GenBank:NM 001077401.1) starting from exon 3 (Fig. 1F, (26, 27), also synergistically increased Irf8-directed CD8a+-like

1G). The proposed TSS upstream of exon 6 is also supported DC development (Fig. 2B). These results suggested that Downloaded from by intron-exon boundary-specific Q-PCR transcript analyses ACVRL1 signaling may play a critical role in the selective (Fig. 1F, 1G). Amplicons from 59 RACE analysis confirmed development of CD8a+-like DCs. ACVRL1 is essential for the proposed TSS in murine DCs (Fig. 1H), and it correlates the growth and survival of vertebrates, and its constitutive with a dominant TSS (at chr15, 100,965,054 bp) seen in CAGE-TSS data in the FANTOM5 promoterome collection

using the ZENBU genome browser. We identified IRF8- http://www.jimmunol.org/ binding sequences [site I: GTTTCACTTCC at 21260 and site II: GGTTTCCATGG at 2148 position from the major TSS; as per an earlier report (21) and using PROMO soft- ware, respectively] in mouse Acvrl1 gene and confirmed IRF8 binding in proximity to these sites by ChIP assay (Fig. 1I). IRF8 binding sites confirmed by ChIP analysis can also be seen as peaks upstream of the TSS in recently reported IRF8 ChIP-sequencing studies (22, 23). IRF8 is recruited near the by guest on September 29, 2021 major TSS of Acvrl1 gene selectively in CD24+ cDCs (CD8a+ DC equivalent population) but not in pDCs (Fig. 1J). This observation suggests that CD8a+ DC–specific factors help the differential recruitment of IRF8 near the major TSS of Acvrl1 gene and may explain the selective expression of Acvrl1 in CD8a+ DCs as opposed to pDCs. Future studies, with em- phasis on the identification of CD8a+ DC–specific factors that aid IRF8 recruitment, would help to better understand the regulation of Acvrl1 gene expression in DCs. Analysis of Acvrl1 gene expression using the ImmGen database (24) also confirmed our finding that Acvrl1 is differentially expressed in + splenic CD8a DCs (Supplemental Fig. 2A). High expression FIGURE 4. Mechanistic understanding of ACVRL1 signaling in DC di- of Acvrl1 in common DC progenitors and committed inter- versification. (A) FL-BMDC cultures from IRF8WT and IRF8R294C mice were + + + mediate stages suggests that it might play a role in guiding supplemented with BMP9 (10 ng/ml), and pDC (B220 SiglecH CD11c )and a+ hi int + CD8a+ DC subtype development from the precursor stage CD8 DC equivalent (CD24 MacI CD11c ) populations were analyzed. Data are the average of three independent experiments. (B)IRF8WT and (Supplemental Fig. 2B–D) (23–25). ImmGen database analysis IRF8R294C FL-BMDCs were serum starved for 2 h and stimulated with TGF-b of DC subtypes from different tissues showed selectively high or BMP9 for 1 h in serum-containing medium. DCs were harvested poststim- + levels of Acvrl1 in CD8a DCs (or its equivalent DC pop- ulation and analyzed for SMAD phosphorylation. TGF-b–specific SMAD-2 ulation), thus making it a good candidate for a CD8a+ DC– phosphorylation was comparable in both DC populations, whereas BMP9-spe- specific marker (Supplemental Fig. 2E). cific phosphorylation of SMAD1/5/8 was defective in IRF8R294C DCs. Data are representative of two independent experiments. (C)NIH3T3cellsandFL- BMDCs (day 8) were serum starved (0.2% FBS) for 24 h (with 100 ng/ml Flt3-L ACVRL1 signaling enhances murine CD8a+ DC development for FL-BMDCs), and transcript levels were measured after 2 h of BMP9 (10 ng/ml) To understand the significance of ACVRL1 signaling in Irf8- stimulation (17). BMP9-stimulated Id2 gene induction was blocked by the directed DC development, we coexpressed Irf8 and Acvrl1 in presence of ACVRL-Fc in the culture media. For experiments with NIH3T3 cells (left panel), the gray bar represents control, and the black bar represents BMP9- DC9 cells. ACVRL1 signaling augmented Irf8-directed D a+ stimulated samples. Data are the average of three experiments. ( ) FL-BMDCs CD8 -like DC development concomitant with the increase (day 8) were serum starved (0.2% FBS) for 24 h with 100 ng/ml Flt3-L, and in subtype-specific transcripts (Fig. 2A). Similarly, coex- Id2 transcript levels were measured after 2 h of FBS stimulation, at the con- pression of Irf8 and Bmp9 (growth-differentiation factor 2), centrations shown (17). Data are representative of two independent experiments. one of the high-affinity physiological ligands for ACVRL1 Error bars in graphs represent 6 SD. The Journal of Immunology 1033 knockout mouse model is embryonic lethal (28). Circulating in FL-BMDC cultures (Fig. 4D) (17). Together, we dem- BMP9 was detected at a concentration of 2–12 ng/ml in onstrate that, among different DC subtypes, the Acvrl1 gene is human serum, and it was demonstrated that ACVRL1- differentially expressed in CD8a+ DCs (or its equivalent DC stimulating activity in serum is attributable to BMP9 (29). population), and ACVRL1 signaling contributes to the gen- To further understand the significance of ACVRL1 signaling eration of DC diversity by specifically enhancing CD8a+ DC in DC development, we cultured FL-BMDCs in the presence development. of BMP9 (10 ng/ml). BMP9 supplementation of DC cultures led to a selective increase in CD8a+ DC development and suppression of pDCs (Fig. 3A, 3B). Observed changes in the Acknowledgments frequency of DC subtypes also correlated with absolute cell This work benefited from data assembled by the ImmGen Consortium. We thank Dr. K. Ozato (National Institute of Child Health and Human numbers (Fig. 3, Supplemental Fig. 1B, 1C). A recent report Development/National Institutes of Health) and Dr. R. Mukhopadhyaya (30) suggested that BMP10, another high-affinity ligand for for careful reading of the manuscript. ACVRL1, can be present in the active form in serum. Sup- plementation of BMP10 also led to a selective increase in CD8a+ DC development and suppression of pDCs, as ob- Disclosures served in BMP9-supplemented cultures (data not shown). The authors have no financial conflicts of interest. Experiments with retroviral expression of Irf8 (Fig. 2) did not show a reduction in pDC transcripts, probably because Irf8 Downloaded from was expressed with a viral promoter and, hence, was not under References 1. Haniffa, M., M. Collin, and F. Ginhoux. 2013. Ontogeny and functional spe- physiological control. FL-BMDC cultures from mouse bone cialization of dendritic cells in human and mouse. Adv. Immunol. 120: 1–49. marrow lineage-negative cells further confirmed that BMP9 2. Shortman, K., and Y.-J. Liu. 2002. Mouse and human dendritic cell subtypes. Nat. a+ Rev. Immunol. 2: 151–161. suppressed pDC development and enhanced CD8 DC 3. Merad, M., P. Sathe, J. Helft, J. Miller, and A. Mortha. 2013. The dendritic cell development from initial developmental stages (Fig. 3C). lineage: ontogeny and function of dendritic cells and their subsets in the steady state R294C and the inflamed setting. Annu. Rev. Immunol. 31: 563–604. http://www.jimmunol.org/ Irf8 mutation was unable to induce expression of Acvrl1 4. Tailor, P., T. Tamura, and K. Ozato. 2006. IRF family proteins and type I inter- in DCs (Fig. 1C, Supplemental Fig. 1D). These results sug- feron induction in dendritic cells. Cell Res. 16: 134–140. 5. Onai, N., A. Obata-Onai, M. A. Schmid, T. Ohteki, D. Jarrossay, and M. G. Manz. gest a mechanistic basis for the differential sensitivity of + + R294C 2007. Identification of clonogenic common Flt3 M-CSFR plasmacytoid and Irf8 FL-BMDC culture to suppression of pDC devel- conventional dendritic cell progenitors in mouse bone marrow. Nat. Immunol. 8: opment by BMP9 (Fig. 4A). BMP signaling results in the 1207–1216. 6. Naik, S. H., P. Sathe, H.-Y. Park, D. Metcalf, A. I. Proietto, A. Dakic, S. Carotta, phosphorylation of SMAD1/5/8, whereas TGF-b signaling M. O’Keeffe, M. Bahlo, A. Papenfuss, et al. 2007. Development of plasmacytoid leads to phosphorylation of SMAD2/3. BMP9 signaling in and conventional dendritic cell subtypes from single precursor cells derived in vitro and in vivo. Nat. Immunol. 8: 1217–1226. FL-BMDCs led to selective and efficient phosphorylation of 7. Guo, X., and X.-F. Wang. 2009. Signaling cross-talk between TGF-b/BMP and

SMAD1/5/8, indicating that ACVRL1 signaling is active other pathways. Cell Res. 19: 71–88. by guest on September 29, 2021 8. Shi, Y., and J. Massague´. 2003. Mechanisms of TGF-b signaling from cell mem- during DC development (Fig. 4B). Id2 plays a pivotal role in brane to the nucleus. Cell 113: 685–700. DC subtype differentiation; enforced expression of Id2 spe- 9. Strobl, H., E. Riedl, C. Scheinecker, C. Bello-Fernandez, W. F. Pickl, K. Rappersberger, O. Majdic, and W. Knapp. 1996. TGF-b 1 promotes in vitro cifically blocks pDC development, whereas it is essential for + + development of dendritic cells from CD34 hemopoietic progenitors. J. Immunol. CD8a DC development (3, 19, 31). DCs are represented 157: 1499–1507. by diverse subtypes and Acvrl1 is differentially expressed 10. Borkowski, T. A., J. J. Letterio, C. L. Mackall, A. Saitoh, X. J. Wang, D. R. Roop, + R. E. Gress, and M. C. Udey. 1997. A role for TGFbeta1 in Langerhans cell bi- by CD8a DC subtype; aside from DC diversity, using ology. Further characterization of the epidermal Langerhans cell defect in TGFbeta1 NIH3T3, a fibroblast cell line model, we show that BMP9 null mice. J. Clin. Invest. 100: 575–581. 11. Yasmin, N., T. Bauer, M. Modak, K. Wagner, C. Schuster, R. Ko¨ffel, M. Seyerl, stimulation leads to efficient induction of Id2 gene expression J. Sto¨ckl, A. Elbe-Burger,€ D. Graf, and H. Strobl. 2013. Identification of bone (Fig. 4C). BMP9 stimulation of FL-BMDCs also induced Id2 morphogenetic protein 7 (BMP7) as an instructive factor for human epidermal R294C Langerhans cell differentiation. J. Exp. Med. 210: 2597–2610. gene transcription at the early stage (2 h), although Irf8 12. Ju, X.-S., D. Ruau, P. Ja¨ntti, K. Sere´, C. Becker, E. Wiercinska, C. Bartz, DCs showed low levels of BMP9-stimulated Id2 gene in- B. Erdmann, S. Dooley, and M. Zenke. 2007. Transforming growth factor b1 up- regulates interferon regulatory factor 8 during dendritic cell development. Eur. J. duction, probably as the result of the defect in Acvrl1 receptor Immunol. 37: 1174–1183. expression (Supplemental Fig. 1E). BMP9 stimulation of FL- 13. Yoshida, Y., R. Yoshimi, H. Yoshii, D. Kim, A. Dey, H. Xiong, J. Munasinghe, BMDCs specifically induced the Id2 gene; this was efficiently I. Yazawa, M. J. O’Donovan, O. A. Maximova, et al. 2014. The transcription factor IRF8 activates integrin-mediated TGF-b signaling and promotes neuro- blocked by the presence of ACVRL1-Fc in culture, whereas inflammation. Immunity 40: 187–198. pDC and CD8a+ DC subtype–specific transcripts remained 14. Jaiswal, H., M. Kaushik, R. Sougrat, M. Gupta, A. Dey, R. Verma, K. Ozato, and P. Tailor. 2013. Batf3 and Id2 have a synergistic effect on Irf8-directed classical unaffected (Fig. 4C). These results imply that the observed CD8a+ dendritic cell development. J. Immunol. 191: 5993–6001. effect of BMP9 was specifically mediated by signaling through 15. Tailor, P., T. Tamura, H. C. Morse, III, and K. Ozato. 2008. The BXH2 mutation in IRF8 differentially impairs dendritic cell subset development in the mouse. Blood the ACVRL1 receptor and suggest an explanation for the 111: 1942–1945. + observed tilt in DC differentiation toward the CD8a DC 16. Gilliet, M., A. Boonstra, C. Paturel, S. Antonenko, X.-L. Xu, G. Trinchieri, subtype (Figs. 1C, 4, Supplemental Fig. 1D, 1E). BMP9 A. O’Garra, and Y.-J. Liu. 2002. The development of murine plasmacytoid den- dritic cell precursors is differentially regulated by FLT3-ligand and granulocyte/ transcripts remained undetected and BMP10 transcripts were macrophage colony-stimulating factor. J. Exp. Med. 195: 953–958. negligible in FL-BMDCs, suggesting that ACVRL1 signaling 17. Kurooka, H., T. Nakahiro, K. Mori, K. Sano, and Y. Yokota. 2012. BMP signaling a+ is responsible for serum-induced Id2 expression. Biochem. Biophys. Res. Commun. mediated–enhancement of CD8 DC development was not 420: 281–287. guided by an autocrine mechanism but may occur via a 18. Tamura, T., P. Thotakura, T. S. Tanaka, M. S. Ko, and K. Ozato. 2005. Identi- fication of target genes and a unique cis element regulated by IRF-8 in developing paracrine mode through the presence of ligand in serum. The macrophages. Blood 106: 1938–1947. observation that serum supplementation–induced Id2 tran- 19. Hacker, C., R. D. Kirsch, X.-S. Ju, T. Hieronymus, T. C. Gust, C. Kuhl, T. Jorgas, S. M. Kurz, S. Rose-John, Y. Yokota, and M. Zenke. 2003. Transcriptional pro- scription was blocked by ACVRL1-Fc in the culture medium filing identifies Id2 function in dendritic cell development. Nat. Immunol. 4: 380– confirmed the presence of ACVRL1 ligand in the serum used 386. 1034 CUTTING EDGE: ACVRL1 SIGNALING ENHANCES CD8a+ DC DEVELOPMENT

20. Garrido-Martin, E. M., F. J. Blanco, A. Fernandez-L, C. Langa, C. P. Vary, 26. Mitchell, D., E. G. Pobre, A. W. Mulivor, A. V. Grinberg, R. Castonguay, U. E. Lee, S. L. Friedman, L. M. Botella, and C. Bernabeu. 2010. Characterization T. E. Monnell, N. Solban, J. A. Ucran, R. S. Pearsall, K. W. Underwood, et al. of the human Activin-A receptor type II-like kinase 1 (ACVRL1) promoter and its 2010. ALK1-Fc inhibits multiple mediators of angiogenesis and suppresses tumor regulation by Sp1. BMC Mol. Biol. 11: 51. growth. Mol. Cancer Ther. 9: 379–388. 21. Shin, D.-M., C.-H. Lee, and H. C. Morse, III. 2011. IRF8 governs expression of 27. Scharpfenecker, M., M. van Dinther, Z. Liu, R. L. van Bezooijen, Q. Zhao, genes involved in innate and adaptive immunity in human and mouse germinal L. Pukac, C. W. G. M. Lo¨wik, and P. ten Dijke. 2007. BMP-9 signals via ALK1 center B cells. PLoS One 6: e27384. and inhibits bFGF-induced endothelial cell proliferation and VEGF-stimulated 22. Kc, W., A. T. Satpathy, A. S. Rapaport, C. G. Brisen˜o, X. Wu, J. C. Albring, angiogenesis. J. Cell Sci. 120: 964–972. E. V. Russler-Germain, N. M. Kretzer, V. Durai, S. P. Persaud, et al. 2014. L-Myc 28. Oh, S. P., T. Seki, K. A. Goss, T. Imamura, Y. Yi, P. K. Donahoe, L. Li, expression by dendritic cells is required for optimal T-cell priming. Nature 507: K. Miyazono, P. ten Dijke, S. Kim, and E. Li. 2000. Activin receptor-like kinase 1 243–247. modulates transforming growth factor-beta 1 signaling in the regulation of angio- 23. Grajales-Reyes, G. E., A. Iwata, J. Albring, X. Wu, R. Tussiwand, W. Kc, genesis. Proc. Natl. Acad. Sci. USA 97: 2626–2631. N. M. Kretzer, C. G. Brisen˜o, V. Durai, P. Bagadia, et al. 2015. Batf3 maintains 29. David, L., C. Mallet, M. Keramidas, N. Lamande´, J. M. Gasc, S. Dupuis-Girod, autoactivation of Irf8 for commitment of a CD8a(+) conventional DC clonogenic H. Plauchu, J. J. Feige, and S. Bailly. 2008. Bone morphogenetic protein-9 is a progenitor. Nat. Immunol. 16: 708–717. circulating vascular quiescence factor. Circ. Res. 102: 914–922. 24. Heng, T. S., M. W. Painter, K. Elpek, V. Lukacs-Kornek, N. Mauermann, 30. Jiang, H., R. M. Salmon, P. D. Upton, Z. Wei, A. Lawera, A. P. Davenport, S. J. Turley, D. Koller, F. S. Kim, A. J. Wagers, N. Asinovski, et al; Immunological N. W. Morrell, and W. Li. 2016. The prodomain-bound form of bone morpho- Genome Project Consortium. 2008. The Immunological Genome Project: networks genetic protein 10 is biologically active on endothelial cells. J. Biol. Chem. 291: of gene expression in immune cells. Nat. Immunol. 9: 1091–1094. 2954–2966. 25. Schlitzer, A., V. Sivakamasundari, J. Chen, H. R. Sumatoh, J. Schreuder, J. Lum, 31. Spits, H., F. Couwenberg, A. Q. Bakker, K. Weijer, and C. H. Uittenbogaart. 2000. B. Malleret, S. Zhang, A. Larbi, F. Zolezzi, et al. 2015. Identification of cDC1- and Id2 and Id3 inhibit development of CD34(+) stem cells into predendritic cell (pre- cDC2-committed DC progenitors reveals early lineage priming at the common DC DC)2 but not into pre-DC1. Evidence for a lymphoid origin of pre-DC2. J. Exp. progenitor stage in the bone marrow. Nat. Immunol. 16: 718–728. Med. 192: 1775–1784. Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021