Expression of the Atypical Chemokine Receptor ACKR4 Identifies a Novel

Expression of the Atypical Chemokine Receptor ACKR4 Identifies a Novel Population of Intestinal Submucosal Fibroblasts That Preferentially Expresses This information is current as Endothelial Cell Regulators of September 26, 2021. Carolyn A. Thomson, Serge A. van de Pavert, Michelle Stakenborg, Evelien Labeeuw, Gianluca Matteoli, Allan McI Mowat and Robert J. B. Nibbs J Immunol published online 14 May 2018 Downloaded from http://www.jimmunol.org/content/early/2018/05/11/jimmun ol.1700967 http://www.jimmunol.org/ Supplementary http://www.jimmunol.org/content/suppl/2018/05/11/jimmunol.170096 Material 7.DCSupplemental

Why The JI? Submit online.

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

by guest on September 26, 2021 • No Triage! Every submission reviewed by practicing scientists

• 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 © 2018 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published May 14, 2018, doi:10.4049/jimmunol.1700967 The Journal of Immunology

Expression of the Atypical Chemokine Receptor ACKR4 Identiﬁes a Novel Population of Intestinal Submucosal Fibroblasts That Preferentially Expresses Endothelial Cell Regulators

Carolyn A. Thomson,*,1 Serge A. van de Pavert,† Michelle Stakenborg,‡ Evelien Labeeuw,‡ Gianluca Matteoli,‡ Allan McI Mowat,*,2 and Robert J. B. Nibbs*,2

Atypical chemokine receptors (ACKRs) are expressed by discrete populations of stromal cells at speciﬁc anatomical locations where they control leukocyte migration by scavenging or transporting chemokines. ACKR4 is an atypical receptor for CCL19, CCL21, and

CCL25. In skin, ACKR4 plays indispensable roles in regulating CCR7-dependent APC migration, and there is a paucity of mi- Downloaded from gratory APCs in the skin-draining lymph nodes of Ackr4-deficient mice under steady-state and inflammatory conditions. This is caused by loss of ACKR4-mediated CCL19/21 scavenging by keratinocytes and lymphatic endothelial cells. In contrast, we show in this study that Ackr4 deficiency does not affect dendritic cell abundance in the small intestine and mesenteric lymph nodes, at steady state or after R848-induced mobilization. Moreover, Ackr4 expression is largely restricted to mesenchymal cells in the intestine, where it identifies a previously uncharacterized population of fibroblasts residing exclusively in the submucosa. Com- 2 +

pared with related Ackr4 mesenchymal cells, these Ackr4 fibroblasts have elevated expression of genes encoding endothelial cell http://www.jimmunol.org/ regulators and lie in close proximity to submucosal blood and lymphatic vessels. We also provide evidence that Ackr4+ fibroblasts form physical interactions with lymphatic endothelial cells, and engage in molecular interactions with these cells via the VEGFD/ VEGFR3 and CCL21/ACKR4 pathways. Thus, intestinal submucosal fibroblasts in mice are a distinct population of intestinal mesenchymal cells that can be identified by their expression of Ackr4 and have transcriptional and anatomical properties that strongly suggest roles in endothelial cell regulation. The Journal of Immunology, 2018, 201: 000–000.

he intestine is a complex and dynamic organ requiring leukocytes can move and migrate (1–3). These processes are es- continuous interactions between many different haemo- sential for tissue development and repair, maintenance of ho- T poietic and nonhaemopoietic (stromal) cell types. These meostasis, adaptation to environmental challenges, and response by guest on September 26, 2021 include resident and migratory leukocytes, epithelial cells, blood to disease. vessel endothelial cells (BECs), lymphatic endothelial cells Leukocyte migration in the intestine and elsewhere is dependent (LECs), neurons, and various populations of mesenchymal cells, on chemokines, which engage conventional G-protein–coupled such as pericytes, smooth muscle cells, fibroblasts, and myofi- chemokine receptors on leukocytes (4). There are also four broblasts. Mesenchymal cells play many important roles in the atypical chemokine receptors (ACKRs) that regulate chemokine- physiological and immunological functions of the intestine (1–3). driven migration by transcytosing, sequestering, or scavenging In addition to providing the three-dimensional matrix of the tissue, chemokines (5). ACKRs are typically expressed by stromal cells they supply growth factors critical for epithelial cell renewal, in highly defined microanatomical niches, where they have been regulate endothelial cell function, contribute to innate and adaptive shown to be capable of regulating leukocyte trafficking into, immune responses, and provide surfaces upon which interstitial within, and out of tissues (5). ACKR4, previously known as

*Institute of Infection, Immunity and Inﬂammation, College of Medical, Veterinary by Research Foundation Flanders Grant G0D8317N and Ph.D. fellowship ZKD1563 and Life Sciences, University of Glasgow, Glasgow G12 8TA, United Kingdom; (to M.S.). †Centre d’Immunologie de Marseille-Luminy, Aix Marseille Universite´, CNRS, The microarray data presented in this article have been submitted to the National INSERM, 13288 Marseille Cedex 9, France; and ‡Laboratory of Mucosal Immunol- Center for Biotechnology Information’s Gene Expression Omnibus database (https:// ogy, Department of Chronic Diseases, Metabolism and Ageing, Translational Re- www.ncbi.nlm.nih.gov/geo/) under accession number GSE113665. search Center for Gastrointestinal Disorders, Catholic University Leuven, BE-3000 Leuven, Belgium Address correspondence and reprint requests to Prof. Robert J.B. Nibbs, University of Glasgow, 120 University Place, Glasgow G12 8TA, U.K. E-mail address: robert. 1Current address: Department of Physiology and Pharmacology, Cumming School of [email protected] Medicine, University of Calgary, Calgary, AB, Canada. The online version of this article contains supplemental material. 2A.M.M. and R.J.B.N. contributed equally to this work. Abbreviations used in this article: ACKR, atypical chemokine receptor; AF, Alexa ORCIDs: 0000-0003-3906-7539 (C.A.T.); 0000-0002-7147-4380 (S.A.v.d.P.); 0000- Fluor; BEC, blood vessel endothelial cell; BMP, bone morphogenetic protein; CM, 0002-4086-0289 (M.S.); 0000-0002-2902-4976 (G.M.); 0000-0001-9389- circular muscle; DC, dendritic cell; DN, double negative; DSS, dextran sodium 3079 (A.M.M.); 0000-0002-8150-0044 (R.J.B.N.). sulfate; FDR, false discovery rate; FRC, ﬁbroblastic reticular cell; IESC, intestinal Received for publication July 6, 2017. Accepted for publication April 10, 2018. epithelial stem cell; iMC, intestinal mesenchymal cell; LC, Langerhans cell; LEC, lymphatic endothelial cell; LM, longitudinal muscle; LN, lymph node; LP, lamina This work was supported by a project grant (MR/L000598) from the United Kingdom propria; ME, muscularis externa; MHCII, MHC class II; MLN, mesenteric LN; Medical Research Council (to C.A.T., A.M.M., and R.J.B.N.). S.A.v.d.P. was sup- QPCR, quantitative real-time PCR; SCS, subcapsular sinus; SI, small intestine; ported by A*MIDEX Chaire d’Excellence Grant ANR-11-IDEX-0001-002, Fonda- SLN, skin-draining LN; aSMA, a smooth muscle actin; WT, wild-type. tion pour la Recherche Me´dicale Jeunes Equipes Grant AJE20150633331, and France Bio Imaging Grant ANR-10-INBS-04-01, and G.M., E.L., and M.S. were supported Copyright Ó 2018 by The American Association of Immunologists, Inc. 0022-1767/18/$35.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1700967 2 ACKR4 AND SUBMUCOSAL FIBROBLASTS IN THE INTESTINE

CCRL1 or CCX–CKR, binds with high affinity to CCL19 and purchased from Biolegend: CD103–PE (2E7), CD11c–PE/Cy7 (N418), CCL21, which direct leukocyte migration through CCR7, and to CD146–PerCP/Cy5.5 and CD146–APC (ME-9F1), B220–PerCP/Cy5.5 ε CCL25, the sole ligand for CCR9 (6–9). By internalizing and (RA3–6B2), CD3 –PerCP/Cy5.5 (145-2C11), CD31–BV605 and CD31– PE (390), CD45–PerCP/Cy5.5, CD55–PE (RIKO-3), CD66a–APC (Mab- degrading its ligands, ACKR4 can regulate their abundance and CC1), CD105–APC (MJ7/18), CD106–PE (429), CD9–Alexa Fluor (AF) shape chemokine gradients (10–12). In vivo, this has been most 647 (MZ3), CD90.2–PE (30-H12), EPCAM–PerCP/Cy5.5 and EPCAM– studied in the skin, where, under steady-state and inflammatory APC/Cy7 (G8.8), F4/80–BV605 and F4/80–APC (BM8), GP38–PE/ conditions, the CCR7-dependent trafficking of Langerhans cells Cy7 (8.1.1), ICAM1–PE (YN1/1.7.4), Ly6G–PE (1A8), MHC class II (MHCII)–BV510 (M5/114.15.2), NK1.1–PerCP/Cy5.5 (PK136), PDGFRa– (LCs) and dendritic cells (DCs) to draining lymph nodes (LNs) is PE (APA5), and SCA1–APC (D7). The following anti-mouse Abs controlled by ACKR4 expression on keratinocytes, a subset of were purchased from eBioscience (San Diego, CA): CD11b–AF700 dermal LECs, and by LECs lining the ceiling of the LN subcap- and CD11b–eFluor 450 (M1/70), CD45–AF700 and CD45–eFluor 450 sular sinus (SCS) (11–15). CCR7 is also essential for the migra- (30-F11), CD166–PE (ALC48), Ly6C–eFluor 450 (HK1.4), and Ter119– tion of DCs from the small intestine (SI) to the mesenteric LNs eFluor 450 (TER-119). BD Biosciences (Franklin Lakes, NJ) supplied CD29–PE (HM b1-1), CD34–AF647 (RAM34), and Siglec F–PE (MLNs) (9), but the function of ACKR4 in this context has not (E50-2440). Dead cells were excluded using 7AAD (BioLegend) or Fix- been explored. Ackr4 transcripts are detectable in the SI and colon able Viability Dye eFluor 780 (eBioscience). Stained cells were analyzed of mice (7), and GFP+ cells are present in the SI of Ackr4gfp/+ or sorted using an LSR II (BD Biosciences) or a FACSAria I, IIu, or III . reporter mice (13), although the identity of these cells is not clear. (BD Biosciences). The purity of all sorted cell populations was 95%. Analysis was performed using FlowJo version 10 (TreeStar, Ashland, OR). In this article, we report that Ackr4 deficiency does not alter migratory DC abundance in the SI or MLNs, either at steady state CCL19AF647 uptake assay or after R848-induced DC mobilization. Thus, ACKR4 in the SI, Prior to staining of surface markers, 1–2 3 106 cells were incubated in Downloaded from in contrast to the skin, serves no detectable indispensable role in 250 mg/ml AF647-labeled CCL19 (CCL19AF647; Almac, Craigavon, U.K.) regulating DC trafficking to draining LNs. We also find that Ackr4 in R10 medium containing 20 mM HEPES (Thermo Fisher) for 1 h at expression in the SI and colon, unlike the skin, is largely restricted 37˚C, as described previously (21–23). + to a subset of mesenchymal cells. These Ackr4 cells represent a Dextran sodium sulfate–induced colitis novel and discrete population of fibroblasts that reside exclusively Mice received 2% dextran sodium sulfate (DSS) (MP Biomedicals, Santa in the intestinal submucosa and that, to our knowledge, have not http://www.jimmunol.org/ + Ana, CA) in sterile distilled water ad libitum for 5 d. Control ani- been characterized before. We show that these Ackr4 fibroblasts mals received sterile water. Colitis scoring was performed as described preferentially express an array of genes encoding endothelial cell previously (24). regulators and that they lie in close proximity to blood and lym- R848-induced DC mobilization phatic vessels in submucosa. We also provide evidence that LECs 2 2 and Ackr4+ fibroblasts physically interact and that these two cell WT and Ackr4 / mice received an i.p. injection of 100 mg of R848 m types are equipped to engage in unique reciprocal molecular in- (InvivoGen, San Diego, CA) in 100 l of sterile PBS. Control animals were injected with 100 ml of sterile PBS alone. Mice were culled 12 or teractions through the VEGFD/VEGFR3 and CCL21/ACKR4 axes. 24 h following injection.

Immunofluorescent microscopy by guest on September 26, 2021 Materials and Methods Mice Tissues were fixed for 3 h in 1% methanol-free paraformaldehyde (Thermo Fisher) in PBS, washed for 1 h in PBS, and then frozen in Optimal Cutting Wild-type (WT), Ackr42/2 [Ackr4tm1.1Rjbn: www.informatics.jax.org/ Temperature compound (Cell Path, Newton, U.K.). Embedded tissue was allele/key/625599 (16)], Ackr4gfp/+, and Ackr4gfp/gfp (Ackr4tm1Ccbl: www. then cut into 7-mm sections, which were blocked using PBS containing 1% informatics.jax.org/allele/key/817740) (13) mice, all on a C57BL/6 back- BSA (Sigma-Aldrich) and 10% donkey serum (Sigma-Aldrich) before ground, were bred and maintained in specific pathogen-free conditions in being stained in the dark for .8 h with a primary Ab mixture. The the Central Research Facility at the University of Glasgow. Female mice following anti-mouse monoclonal Abs were used: CD45-eFluor 450 6–11 wk of age were used in all experiments, which were performed under (30-F11), unlabeled endomucin (rat-derived; V.7C7), and LYVE1-eFluor the auspices of U.K. Home Office Licenses. 660 (ALY7) from eBioscience; PDGFRa-biotinylated (APA5) (Bio- Legend); and CD31-purified (MEC 13.3) (BD Biosciences). The following Preparation of single-cell suspensions polyclonal anti-mouse Abs were also used: anti-CCL21 and anti-VEGFR3/ Flt4 (both generated in goat; R&D Systems, Minneapolis, MN) and anti–a Single-cell suspensions were generated from intestinal tissue by enzymatic a digestion as described previously (17–19). Briefly, excised SI and colons smooth muscle actin ( SMA) (rabbit-derived; Abcam, Cambridge, U.K.). were washed in 2 mM EDTA in HBSS for 3 3 20 min at 37˚C. Tissue was Unlabeled primary Abs were visualized using polyclonal, directly conju- then digested in R10 medium (10% FCS, 2 mM EDTA in RPMI 1640 gated Abs, including donkey anti-goat AF555, donkey anti-goat AF647, [Thermo Fisher, Waltham, MA]) containing 1 mg/ml collagenase VIII donkey anti-rat AF594, goat anti-rat AF555, and goat anti-rabbit AF647, (Sigma-Aldrich, St Louis, MO) for SI digests or 0.65 mg/ml collagenase V all from Thermo Fisher. Biotinylated primary Abs were visualized using (Sigma-Aldrich), 0.45 mg/ml collagenase D (Roche Diagnostics, Mann- DyLight 549 streptavidin (Vector Laboratories, Burlingame, CA). When heim, Germany), 1 mg/ml Dispase (Thermo Fisher), and 30 mg/ml DNase I biotinylated Abs were used, tissue sections were also blocked using an (Roche Diagnostics) for colon digests. Digests were then passed through a Avidin/Biotin Blocking kit (Vector Laboratories) in accordance with the Ackr4gfp/+ 40-mm cell strainer, and the cells were pelleted at 400 3 g for 5 min. manufacturer’s instructions. GFP expression in samples from mice was directly detected by fluorescence microscopy without using anti- In some experiments, cells from the intestinal muscularis externa (ME) containing the longitudinal muscle (LM) and circular muscle (CM) were GFP Abs. Images were acquired on a Leica SP5 confocal laser–scanning isolated as described previously (20). Briefly, ME from SI and colon was microscope or an EVOS FL Cell Imaging System (Thermo Fisher). dissected from the mucosa and enzymatically digested in MEMa medium RNA isolation (Lonza, Verviers, Belgium) containing 100 mg/ml penicillin, 100 mg/ml strep- tomycin, 50 mM 2-ME, 5% FCS, 0.5 mg/ml protease type I (Sigma-Aldrich), Purified cells were pelleted and lysed in 500 ml of RLT lysis buffer (Qiagen, 0.25 mg/ml collagenase type IV (Sigma-Aldrich), and 5 U/ml DNase I (Sigma- Hilden, Germany). Lysed cells were homogenized using QIAshredders Aldrich) for 15 min at 37˚C. Cell suspensions were filtered through a 40-mm (Qiagen). Under RNase-free conditions, RNA was isolated using an cell strainer, and the cells were pelleted at 1500 rpm for 5 min. RNeasy Micro kit (Qiagen). Genomic DNA contaminants were digested on column using the RNase-free DNase kit (Qiagen). Flow cytometry and cell sorting Transcriptomics Cells (1–6 3 106) were blocked using anti-CD16/CD32 Ab (BioLegend, San Diego, CA), and then labeled with fluorescently conjugated Abs in the Microarray assays were performed on GeneChip Mouse Transcriptome dark as described previously (17–19). The following anti-mouse Abs were Arrays 1.0 by Almac Diagnostics. Briefly, purified total RNAwas converted The Journal of Immunology 3 into sense-strand cDNA and amplified using an Ovation Pico WTA System LN stromal cells can be subdivided according to expression of V2 kit (NuGEN, San Carlos, CA). Amplified cDNA was then fragmented CD31 and GP38 into GP382CD31+ BECs, GP38+CD31+ LECs, and labeled using an Encore Biotin module (NuGEN). Fragmented cDNA GP38+CD312 fibroblastic reticular cells (FRCs), and a “double was then hybridized to GeneChip Mouse Transcriptome Arrays 1.0 (Affymetrix, Santa Clara, CA). Procedures were carried out as described negative” (DN) population containing pericytes and other rare by the manufacturers. Data sets were analyzed using GeneSpring GX stromal cell types (28). A substantial proportion (28–52%) of software. Data were normalized using a variant of the robust multichip LECs were GFP+ in the MLN of Ackr4gfp/+ mice, whereas all average method (RMA16). Normalized data were analyzed using unpaired other stromal cells lacked GFP (Fig. 1B, 1E). t tests to determine the significance of gene expression differences. The 2 resulting p values were adjusted for multiple comparisons using the Among CD45 cells in digests of SI and colon, epithelial cells + 2 Benjamini–Hochberg multiple testing correction at a false discovery rate were identified as EPCAM , whereas EPCAM cells were sub- 2 (FDR) of 0.1. Differentially expressed genes were assigned gene ontology divided into GP38 CD31+ BECs, GP38+CD31+ LECs, GP38+ terms and grouped into biological processes using the Database for An- CD312 cells (referred to hereafter as iMCs), and a GP382CD312 notation, Visualization and Integrated Discovery Bioinformatics Resources DN population (Fig. 1C). In contrast to the MLN, virtually all version 6.8 (https://david.ncifcrf.gov). Analysis was performed using gfp/+ 2 protocols developed by Huang et al. (25, 26). Significance of enrichment LECs in the SI and colon of Ackr4 mice were GFP (Fig. 1D, was determined using a modified Fisher’s exact test, and a Benjamini– 1E). GFP was also barely detectable in epithelial cells, BECs, and Hochberg multiple testing correction was used to correct for the rate of DN cells in the SI and colon of Ackr4gfp/+ mice (Fig. 1D). The vast type I errors. Enrichment of biological processes was considered signifi- + # majority of GFP cells in these tissues were in the iMC population cant if p 0.05. The microarray data have been deposited in NCBI’s Gene ∼ ∼ Expression Omnibus database (https://www.ncbi.nlm.nih.gov/geo/) and are (Fig. 1D), with, on average, 25 and 35% of iMCs expressing accessible through the Gene Expression Omnibus Series accession number GFP in the SI and colon, respectively (Fig. 1F).

GSE113665. To determine if Ackr4 expression in the intestine might be Downloaded from gfp/+ Quantitative real-time PCR modified by inflammation, we induced colitis in Ackr4 mice by administering 2% DSS in the drinking water. All DSS-treated Total RNA was reverse transcribed with random primers using Quan- mice had developed clinical colitis by day 5 of treatment titect Reverse Transcription kit (Qiagen). Quantitative real-time PCR (Supplemental Fig. 1A–C), and this led to a decrease in the pro- (QPCR) amplifications were performed in triplicate using PerfeCTa + ∼ SYBR Green FastMix (Quanta Biosystems, Gaithersburg, MD) as de- portion of colonic iMCs that were GFP from an average of 30 scribed previously (27). A 500-mM mix of forward and reverse primers to ∼20%. However, the overall pattern of GFP expression in http://www.jimmunol.org/ was used per reaction. Primers were designed using Primer3 Input Ackr4gfp/+ mice was unchanged, with the dominant GFP+ pop- software (version 0.4.0) and generated by Integrated DNA Technolo- ulations being iMCs in SI and colon and LECs in the MLN gies. Primer sequences were as follows (59 to 39): Adm:59-TCT TGG ACT TTG GGG TTT TG-39 and 59-ATT CTG TGG CGA TGC TCT G- (Supplemental Fig. 1D–G, data not shown). Thus, colonic in- 39; Ccl21:59-CCA ACT CAC AGG CAA AGA GG-39 and 59-GCC flammation is not associated with a change in the distribution of AGG TAA GAA AGG GAT GG-39; Ctsh:59-TCG CCA GTG AAA Ackr4 expression. AAT CAG G-39 and 59-CCC GTG GTA GAG AAA GTC CA-39; Fap: 59-CGC ACA GAC CAA GAA ATA CAA G-39 and 59-CAA ACT GAG iMCs and MLN LECs express functional ACKR4 protein GCA GAA TCA-39; Figf:59-AGC CAG GAG AAC CCT TGA TT-39 and 59-GGGCAACAGTGACAGCAAC-39; Grem1:59-CCT TTC Next, we sought to identify cells expressing ACKR4 protein. Using AGT CTT GCT CCT TCT G-39 and 59-CGT GTG ACC CTT TTC samples from Ackr4-deficient [Ackr4tm1.1Rjbn (16)] mice as con- by guest on September 26, 2021 TTG-39; Il6:59-TTC CAT CCA GTT GCC TTC TT-39 and 59-ATT trols, commercially available anti-ACKR4 Abs repeatedly failed 9 9 TCC ACG ATT TCC CAG AG-3 ; Serpine1:5-GTA AAC GAG AGC to provide convincing detection of ACKR4 in the intestine of WT GGC ACA G-39 and 59-CCG AAC CAC AAA GAG AAA GG-39;and Tek:59-TGA GAT GCC TGC GTT TAC TG-39 and 59-TTG TAG TTC mice by flow cytometry or immunofluorescence microscopy (data TGT CTG CCG TTG T-39. QPCR reactions were performed using a not shown). We therefore used fluorescent chemokine uptake as- Prism 7900HT Sequence Detection System (Life Technologies, Invi- says, a technique that we have used to successfully and sensitively trogen, CA) for 40 cycles. Target gene expression was normalized to detect ACKR expression in other contexts (12, 21–23). Single-cell expression of a reference gene, Tbp, which encodes the TATA-box suspensions of MLN, SI, and colon from WT and Ackr4-deficient binding protein: preliminary data indicated that this gene had very AF647 similar levels of expression among different intestinal mesenchymal mice were incubated ex vivo with CCL19 andanalyzedbyflow cell (iMC) populations. The sequences of the Tbp-specific primers cytometry (Fig. 2). Consistent with the analysis of GFP expression in were 59-TGC TGT TGG TGA TTG GT-39 and 59-AAC TGG CTT GTG Ackr4gfp/+ mice (Fig. 1), ACKR4-dependent CCL19AF647 uptake was TGG GAA AG-39. Fold change values were calculated using the 2DD restricted to LECs in MLN (Fig. 2A) and iMCs in SI and colon following equation: fold change = 2 CT, where CT is the cycle threshold. The mean DCT of the control samples was used as a (Fig. 2B–D). Thus, iMCs and MLN LECs express ACKR4 protein calibrator. capable of internalizing CCL19. Statistical analysis Ackr4 deficiency does not affect DC migration from the Data were analyzed using Prism 6 software (GraphPad, San Diego, CA) and intestine to the MLN are represented as mean 6 1 SD. Statistical tests used are indicated in the ACKR4 regulates CCR7-dependent trafficking of DCs and LCs figure legends. from the skin under steady-state and inflammatory conditions, and Ackr4-deficient skin-draining LNs (SLNs) contain fewer DCs/LCs Results than WT controls, whereas the skin contains more (11–13). DC Ackr4 is expressed by MLN LECs and a subset of iMCs in trafficking from the SI to MLN is also CCR7-dependent (29–31). steady state and during inflammation However, we found no significant differences in the numbers or To characterize Ackr4 expression in MLN and intestine, we relative proportions of total DCs, or their subsets, in the SI lamina used flow cytometry to examine GFP expression by CD45+ propria (LP) or MLN of WT and Ackr4-deficient mice (Fig. 3A, leukocytes and CD452 stromal cells in Ackr4gfp/+ reporter mice 3B and data not shown; DC gating strategy in Supplemental Fig. (Ackr4tm1Ccbl) (13), using WT mice as controls. As in periph- 2A, 2B). This was also the case after injection of R848, a synthetic eral lymphoid tissues, blood and skin (11–13), CD45+ cells in TLR7/8 agonist that stimulates rapid CCR7-dependent mobiliza- 2 the MLN, SI or colon were entirely GFP (Fig. 1A). However, tion of DC from the SI LP to the MLN (30, 32) (Fig. 3C, 2 GFP+ cells were present in the CD45 compartment of all Supplemental Fig. 2C). ACKR4 is therefore dispensable for tissues (Fig. 1A). steady-state and R848-induced DC migration from SI to MLN. 4 ACKR4 AND SUBMUCOSAL FIBROBLASTS IN THE INTESTINE Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

FIGURE 1. Ackr4 is expressed by LECs in the MLN and mesenchymal cells in the intestine. (A) Overlaid histogram ﬂow cytometry plots showing GFP expression by CD45+ (top panels) and CD452 cells (bottom panels) among single, live Ter1192 cells in cell suspensions of the MLN, SI, and colon of WT and Ackr4gfp/+ (Ackr4tm1Ccbl) mice. (B) Left panel, Flow cytometry contour plot of stromal cell populations among single, live CD452Ter1192 cells in the MLN, identifying GP38+CD312 FRCs, GP38+CD31+ LECs, GP382CD31+ BECs and GP382CD312 DN cells. Right panels, Representative overlaid histogram plots showing GFP expression by stromal cell populations in the MLN of WT and Ackr4gfp/+ mice. (C) Flow cytometric analysis of stromal cell populations among single, live CD452Ter1192 cells in the SI and colon, identifying EpCAM+ epithelial cells (histograms) and GP38+CD312 iMCs, LECs, BECs, and DN cells in the EpCAM2 population (contour plots). (D) Representative overlaid histograms showing GFP expression by stromal cell populations in the SI and colon of WT and Ackr4gfp/+ mice. In (A), (B), and (D), the numbers on the histogram plots indicate (Figure legend continues) The Journal of Immunology 5 Downloaded from

FIGURE 2. ACKR4 is expressed as a functional protein on a subset of iMCs and MLN LECs. Single-cell suspensions prepared from WT and Ackr4- 2/2 tm1.1Rjbn AF647

deficient (Ackr4 ) mice (Ackr4 ) were incubated in medium containing CCL19 for 1 h at 37˚C, then labeled with fluorescent Abs and http://www.jimmunol.org/ analyzed by flow cytometry. Stromal cell subsets were identified from among live, single CD452Ter1192 cells. (A–C) Representative overlaid histogram plots showing uptake of CCL19AF647 by LECs and GP38+CD312 cells (i.e., FRCs or iMCs) from (A) MLN, (B) SI, and (C) colon of WT and Ackr42/2 mice. The numbers on the plots indicate the percentage of CCL19AF647-positive cells in the WT samples. (D) Mean percentage of CCL19AF647-positive cells (61 SD) in the GP38+CD312 population in the MLN, SI, and colon of WT and Ackr42/2 mice (n = 3/4 per group). Data are representative of two individual experiments. ****p , 0.0001, unpaired Student t test, comparing data from the same tissue from WT versus Ackr42/2 mice.

Ackr4 expression deﬁnes a discrete subset of Thus, although iMCs comprise a complex mixture of different intestinal ﬁbroblasts cell types, Ackr4+ iMCs in the SI and colon appear to be a discrete

We next explored iMC heterogeneity and considered whether and relatively homogeneous subset of fibroblasts. by guest on September 26, 2021 Ackr4 expression identifies a specific subset of iMCs. This was Ackr4+ fibroblasts are restricted to the intestinal submucosa done by analyzing iMCs from Ackr4gfp/+ mice for expression of + 12 mesenchymal cell markers: CD9, CD29 (integrin b1), CD54 Next, we sought to determine if Ackr4 fibroblasts occupy a (ICAM1), CD55 (DAF), CD66a (CEACAM-1), CD90 (Thy1), specific anatomical niche. Analysis of transverse sections of SI gfp/+ CD105 (endoglin), CD106 (VCAM1), CD140a (PDGFRa), and colon from Ackr4 mice by fluorescence microscopy + CD146 (MCAM), CD166 (ALCAM), and SCA1. showed that GFP cells were absent from the LP but abundant in All these markers, except CD105, CD166, VCAM1, and the submucosa, particularly in regions directly underlying the LP CEACAM-1, were detected on some or all iMCs in both SI and (Fig. 5A, 5B). Unexpectedly, cells of the LM layer of the SI and + colon (Fig. 4, data not shown). iMCs were divided into three colon were also GFP (Fig. 5A, 5B). No green fluorescent cells populations based on expression of CD146 and Ackr4 (Fig. 4A, were present in the submucosa or muscle layers of control WT SI 4B). A small population (P1 in Fig. 4A, 4B) was uniformly and colon sections (Supplemental Fig. 3A). Immunohistological CD146+CD9+CD29+ and lacked Ackr4, ICAM1, PDGFRa, and analysis of both tissues revealed that all the GFP+ cells in the CD34 (Fig. 4C, 4D). In contrast, the larger CD1462Ackr42 and submucosa were PDGFRa+, and virtually all PDGFRa+ cells in CD1462Ackr4+ populations (P2 and P3 in Fig. 4A, 4B) were this region were GFP+ (Fig. 5C, data not shown). In contrast, uniformly SCA1+ICAM1+PDGFRa+, had lower expression of GFP+ cells in the aSMA+ LM layer of the SI and colon lacked CD9 and CD29, and were mostly (P2) or all (P3) CD34+ (Fig. 4C, PDGFRa, although PDGFRa+ cells lacking GFP and aSMA were 4D). Colonic iMCs typically expressed less CD55, but more CD9 readily detectable between, and less frequently within, the LM and and CD90, than SI iMCs (Fig. 4C, 4D). CD1462Ackr42 and CM (Fig. 5D, 5E). PDGFRa+ cells were also abundant in the LP, CD1462Ackr4+ iMCs from the same tissue differed in their ex- particularly in the region underneath the epithelium, but these 2 pression of CD34 and CD55 (SI and colon) and CD90 (colon cells were uniformly GFP (Fig. 5C, 5F–G). All GFP+ iMCs only), each of which was expressed more uniformly by Ackr4- detected in the flow cytometry experiments were PDGFRa+ (P3 in expressing cells (Fig. 4C, 4D). The surface phenotype of Ackr4+ Fig. 4), so they must all be submucosal fibroblasts: the GFP+ cells iMCs, particularly the presence of CD34 and PDGFRa, indicates in the LM layer are presumably not liberated alive when preparing that these cells are fibroblasts. single-cell suspensions. In contrast, the GFP2PDGFRa+ iMC

the percentage of GFP+ cells in Ackr4gfp/+ samples. Flow cytometry plots in (A)–(D) are representative of those from at least three individual experiments involving at least three mice of each genotype. (E and F) Percentage of cells expressing GFP in LEC (E) and GP38+CD312 cell (F) populations in the MLN (n = 6), SI (n = 23), and colon (n = 20) of Ackr4gfp/+ mice. Mean 6 1 SD is indicated. Data are pooled from two or more individual experiments. 6 ACKR4 AND SUBMUCOSAL FIBROBLASTS IN THE INTESTINE Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

FIGURE 3. Ackr4 deficiency has no detectable effect on steady-state or R848-induced migration of intestinal DCs to the MLN. At steady state, or after R848 treatment, DC subsets in the MLN and SI of WT and Ackr4-deficient (Ackr42/2) mice (Ackr4tm1.1Rjbn) were identified by flow cytometry and quantified (for gating, see Supplemental Fig. 2A, 2B). (A) Numbers of CD11c+MHCII+F4/802 resident DC and CD11c+MHCIIhiF4/802 migratory DC subsets in the MLN of steady-state WT and Ackr42/2 mice. (B) Numbers of CD11c+MHCII+F4/802 DC subtypes in the SI of steady-state WT and Ackr42/2 mice. (C) Numbers of cells of each DC subtype in the CD11c+MHCIIhiF4/802 migratory DC population in the MLN of WT and Ackr42/2 mice, 12 and 24 h after i.p. administration of 100 mg of R848 and in PBS-treated controls. In all graphs, data from individual mice are shown, along with the mean (+1 SD) for each group. No statistically significant differences are present between WT and Ackr42/2 data, and this was also seen in two or more repeat experiments.

population identified by flow cytometry (P2 in Fig. 4) presumably cytometry (20) (Supplemental Fig. 2D). Lymphocytes were contains the PDGFRa+ cells that we observed in the LP and be- barely detectable in the ME of either the colon or the SI (data tween and within the LM and CM layers. notshown),butDCs(CD45+CD642CD11c+MHCIIhi)were Given the location of Ackr4-expressing cells, we were in- present. However, neither the overall number of ME DCs nor terested in whether Ackr4-deficiency was associated with any the size of individual ME DC subsets was different between changes in the DC and lymphocyte content of the ME. This WT and Ackr4-deficient mice. structure was therefore dissected from the SI and colon of Ackr4 expression therefore shows exquisite anatomical restric- WT and Ackr4-deficient mice, digested, and analyzed by flow tion in the intestine and defines a discrete population of fibroblasts The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

FIGURE 4. Ackr4 expression and surface immunophenotyping identifies discrete subsets of iMCs. (A and B) Flow cytometric histogram plots showing CD146 and GFP expression by live, single CD452GP38+CD312 iMCs from the (A) SI and (B) colon of Ackr4gfp/+ (Ackr4tm1Ccbl) mice. P1 cells are defined as CD146+GFP2; P2 cells are identified as CD1462GFP2; and P3 cells are identified as CD1462GFP+. The percentage of all iMCs found in the P1, P2, and P3 populations is indicated. (C and D) Histogram plots showing expression of SCA1, ICAM1, PDGFRa, CD9, CD90, CD29, CD55, and CD34 by P1, P2 and P3 in the (C) SI and (D) colon. Results are representative of at least three individual experiments. 8 ACKR4 AND SUBMUCOSAL FIBROBLASTS IN THE INTESTINE Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

FIGURE 5. Ackr4+ fibroblasts populate the intestinal SM. (A and B) Representative fluorescent microscopic images showing the location of GFP expression (green) in transverse sections of (A) SI and (B) colon of Ackr4gfp/+ (Ackr4tm1Ccbl) mice costained with DAPI (blue). The image on the right of each panel is a magnified version of the region shown in the box on the image to its left. Original magnification 310. (C) Representative immunofluorescent microscopic image of section of colon from Ackr4gfp/+ mouse immunostained with Abs against PDGFRa (red) and LYVE1 (purple) and counterstained with DAPI (blue). Left, PDGFRa (red) only; Right, composite image of all colors. The SM and LP are indicated and separated by the dotted line. (D–G) Representative immunofluorescent microscopic images of sections of (D and F)SIor(E and G) colon of Ackr4gfp/+ mice immunostained with Abs against PDGFRa (red) and aSMA (purple) and counterstained with DAPI (blue). In (D) and (E), LM and CM layers are labeled and their boundaries marked with dotted lines; in (F) and (G), LP and Lu are marked. Scale bars are shown on the images. Images are representative of at least two individual experiments, each involving two or more mice. Lu, lumen; SM, submucosa. The Journal of Immunology 9 Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

FIGURE 6. Ackr4+ fibroblasts show elevated expression of multiple endothelial cell regulators. (A) FACS gating strategy used to isolate Ackr4+ and Ackr42 cells from among live, single CD452GP38+CD312 cells from the SI of Ackr4gfp/+ (Ackr4tm1Ccbl) mice. (B) Heatmap showing the relative signal intensities of the 165 transcripts that were differentially expressed between the Ackr4+ and Ackr42 cells (fold change $2; adjusted p value ,0.05; FDR 0.1). (C) Volcano plot showing differences between Ackr4+ and Ackr42 cells in their expression of all genes analyzed, against the statistical significance of those expression differences. Genes shown in red have a fold change $2 and an adjusted p value ,0.1: those on the right-hand side are more highly expressed in Ackr4+ cells; those on the left are more highly expressed in Ackr42 cells. Ackr4 and genes analyzed in (E) are highlighted. Significance was calculated using a modified t test and adjusted using a Benjamini–Hochberg multiple testing correction. (D) The biological (Figure legend continues) 10 ACKR4 AND SUBMUCOSAL FIBROBLASTS IN THE INTESTINE Downloaded from

FIGURE 7. Expression of selected genes encoding regulators of endothelial cells. RNA was isolated from populations of cells that had been purified http://www.jimmunol.org/ by FACS sorting from the SI of WT mice. cDNA was prepared and expression of eight genes encoding endothelial cell regulators was analyzed by QPCR in endothelial cells (CD452CD31+), leukocytes (CD45+), iMCs (CD452GP38+CD312), and epithelial cells (CD452EPCAM+). Gene names are indicated at the top of each graph. Mean expression in iMCs is set to 1. Data points from individual mice are shown, along with means (61 SD). Only significant differences between iMCs and other cell types are shown. *p , 0.05, **p , 0.01, ***p , 0.001, one-way ANOVA. resident in the submucosa, but Ackr4 deficiency does not lead to encode secreted or surface proteins (Fig. 6C, Supplemental any detectable change in DC abundance in the ME. Table I). Moreover, several other genes significantly elevated in + + Ackr4 fibroblasts encode secreted proteins that have also been

Ackr4 fibroblasts show elevated expression of genes encoding by guest on September 26, 2021 implicated in endothelial cell regulation, including Pi16 (Pepti- regulators of endothelial cells dase inhibitor 16) (33), Postn (periostin) (34–36), Ism1 (Isthmin-1) To identify genes specifically expressed by submucosal fibroblasts, (37), Slit2 (38, 39), Ntn4 (Netrin-4) (40–42), and Chrdl1 (Chordin- + 2 we FACS-purified Ackr4 fibroblasts and Ackr4 iMCs from the like 1) (43). gfp/+ SI of Ackr4 mice (Fig. 6A) and compared their transcriptomes To validate the transcriptomic data, we compared the expression using Mouse Transcriptome Arrays (Fig. 6B–D). In addition to of eight of these endothelial cell regulators (Adm, Ctsh, Fap, Figf, Ackr4, 169 entities encompassing ∼125 characterized genes were Grem1, Il6, Tek, and Serpine1) in GFP+ and GFP2 iMCs purified differentially expressed between the two groups by at least 2-fold from the SI of Ackr4gfp/+ mice. GFP+ cells more strongly (p , 0.05; FDR 0.1) (Fig. 6B, 6C, Supplemental Table I). Inter- expressed all eight genes (Fig. 6E). Moreover, six of the genes we + estingly, more than half of the genes elevated in Ackr4 fibroblasts analyzed (Adm, Fap, Figf, Grem1, Tek, and Serpine1) were more encode secreted or cell surface proteins (Supplemental Table I). highly expressed by SI iMCs than by endothelial cells, leukocytes, Many of these are components, or regulators, of the extracellular or epithelial cells from the SI (Fig. 7), which, together with the matrix. Cytokines and their regulators were also present. It ap- data in Fig. 6E, indicate that Ackr4+ submucosal fibroblasts are + pears likely, therefore, that Ackr4 fibroblasts shape the submu- the dominant source of these endothelial cell regulators in the cosal microenvironment to regulate the behavior of cells resident intestine. in, or migrating through, this region of the intestine. Collectively, these observations indicate that Ackr4+ fibroblasts Interestingly, gene ontology clustering of genes elevated in have the potential to create a microenvironment that controls the + Ackr4 fibroblasts identified “positive regulation of angiogenesis” biology of endothelial cells. as the most significantly enriched biological process (Fig. 6D). + The genes responsible for this association were Adm (Adreno- Ackr4 fibroblasts are intimately associated with blood and medullin), Col8a1 (collagen type VIII a1), Ctsh (cathepsin H), lymphatic vessels in the submucosa Fap (fibroblast activation protein-a), Figf (VEGFD), Grem1 In light of the transcriptomic and QPCR data, we performed (Gremlin-1), Il6 (IL-6), Serpine1 (plasminogen activator inhibitor-1), immunofluorescence microscopy on sections of intestine from and Tek (angiopoietin receptor Tie2), all but one of which Ackr4gfp/+ mice to examine the physical relationship between

processes most signiﬁcantly upregulated in Ackr4+ cells compared with Ackr42 cells, as determined using the Database for Annotation, Visualization and In- tegrated Discovery Bioinformatics Resources. (E) Relative expression of eight genes in Ackr4+ and Ackr42 cells, as determined by QPCR analysis of cDNA from cells FACS-sorted from among live single CD452GP38+CD312 cells from the SI. Gene names are indicated at the top of each graph. Data points from individual mice are shown, along with means (61SD).MeanexpressioninAckr42 cells is set to 1. *p , 0.05, **p , 0.01, ***p , 0.001, Student t test. The Journal of Immunology 11 Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

FIGURE 8. Intestinal Ackr4+ fibroblasts interact with the vasculature of the SM. (A and B) Representative images captured by fluorescence microscopy from sections of SI from Ackr4gfp/+ (Ackr4tm1Ccbl) mice immunostained with fluorescent Abs against (A) CD31 (red) and LYVE-1 (purple) or (B) LYVE1 only (purple) and costained with DAPI (blue). In (A), the images in the left panel show individual colors; a composite image is shown in the right panel. In (B), images in (Bi) and (Bii) show high-power views of the areas highlighted in the left panel, with arrows indicating sites of physical interaction between LYVE1+ cells and GFP+ cells. (C) Image of section of colon from Ackr4gfp/+ mouse immunostained with Abs against VEGFR3 (blue), CD45 (cyan), and blood vessel protein endomucin (EMCN; red). Right panels show high-power views of the areas (Ci) and (Cii) highlighted in left panel. Scale bars are shown. Images are representative of those taken from at least three independent experiments. SM, submucosa.

Ackr4+ fibroblasts and endothelial cells lining blood and lym- fibroblasts in the steady-state intestine (Figs. 6E, 7). Endothe- phatic vessels. This revealed that Ackr4+ fibroblasts reside in close lial cells (Fig. 9A), and specifically LYVE1+ LECs (Fig. 9B), proximity to both CD31+LYVE12 blood vessels and CD31int were also the principal source of the ACKR4 ligand CCL21 in LYVE1+ lymphatic vessels in the submucosa of the SI (Fig. 8A) the intestine. Although Ccl21 transcripts were undetectable in and colon (Supplemental Fig. 3B). Many Ackr4+ fibroblasts iMCs (Fig. 9A), under high magnification, small deposits of showed particularly close physical interactions with lymphatic CCL21 protein could be seen associated with GFP+ fibroblasts vessels, and were often seen projecting extensions onto the in the intestine of Ackr4gfp/+ mice (Fig. 9C). These were vir- LYVE1+ LECs lining these structures (Fig. 8B). Moreover, LECs tually absent from GFP+ submucosal fibroblasts in Ackr4- in the submucosa expressed VEGFR3 (Fig. 8C, Supplemental deficient Ackr4gfp/gfp mice (Fig. 9C), indicating that they are Fig. 3C), a receptor for VEGFD. VEGFD is encoded by the likely generated by ACKR4-mediated uptake of LEC-derived Figf gene, which shows strong preferential expression by Ackr4+ CCL21. 12 ACKR4 AND SUBMUCOSAL FIBROBLASTS IN THE INTESTINE

FIGURE 9. ACKR4-dependent association of LEC-derived CCL21 with Ackr4+ ﬁbroblasts in the SM. (A)cDNA was prepared from RNA extracted from endothelial cells (CD452CD31+), leukocytes (CD45+), iMCs (CD452GP38+ CD312), and epithelial cells (CD452 EPCAM+) that had been FACS-sorted from the SI. Expression of Ccl21 was Downloaded from analyzed by QPCR. Mean expression by iMCs is set to 1. Data points from individual mice are shown, along with means (61 SD). ***p , 0.001, one-way ANOVA. (B and C) Representative images of sections of SI from (B)and http://www.jimmunol.org/ (C) Ackr4gfp/+ or (C) Ackr4-deﬁcient Ackr4gfp/gfp (Ackr4tm1Ccbl)mouseimmu- nostained with Abs against CCL21 (red) and LYVE1 (purple) and costained with DAPI (blue). In (B), the images in the left panel show individual colors; a composite image is shown in the right panel. In (C), lower panels show high-power views of the areas highlighted in the panel above. Scale bars are shown. Images are by guest on September 26, 2021 representative of two or more individual experiments. SM, submucosa.

Thus, Ackr4+ fibroblasts are a rich source of endothelial cell and molecular interactions with LECs. Together, these data pro- regulators and reside in close proximity to blood and lymphatic vide novel insights into the complexity and heterogeneity of the vessels in the submucosa. They form intimate physical associa- iMC compartment and suggest a specific role for Ackr4+ fibro- tions with LECs, and are likely to regulate the function of these blasts in regulating blood and lymphatic vessels. cells by producing the VEGFR3 ligand VEGFD and by scav- Like keratinocytes, dermal LECs, and LECs on the SCS of the enging LEC-derived CCL21 using ACKR4. SLNs (12), subsets of iMCs and MLN LECs were capable of mediating ACKR4-dependent uptake of ACKR4 ligands. This was Discussion evident in CCL19AF647 uptake experiments performed ex vivo and To our knowledge, our study provides the first detailed analysis of by the comparison of anti-CCL21 immunostaining of GFP+ cells ACKR4 in the intestine and it reveals fundamental differences in in the intestines of Ackr4gfp/+ mice and Ackr4-deficient Ackr4gfp/gfp the expression and function of this ACKR in the skin and intestine. mice. However, the functional significance of ACKR4-mediated Furthermore, we have identified a previously uncharacterized chemokine scavenging in the intestine remains unclear. Ackr4 population of intestinal fibroblasts that expresses ACKR4 together deficiency disrupts steady-state and inflammation-driven APC with genes encoding regulators of endothelial cell function. These trafficking from skin to SLN (11–13), but we could find no evi- Ackr4+ fibroblasts associate closely with the submucosal vascu- dence for this in the intestinal immune system, despite APC mi- lature, and we have provided evidence that they engage in physical gration to draining LNs being profoundly dependent on CCR7 in The Journal of Immunology 13 both tissues (8, 9, 14, 15, 29–31). This likely reflects differences in and other GP38+ iMCs (data not shown). Particularly prominent the anatomical location of the ACKR4-expressing cells in the two among genes preferentially expressed by P1 cells were those tissues. In skin, ACKR4 is strongly expressed by keratinocytes, encoding markers of enteric glial cells, including Plp1, S100b, which may explain why the migration of epidermal LCs is par- Gfap, and Sox10 (52) (data not shown). ticularly sensitive to Ackr4 deficiency (11, 12). In contrast, the vast Our combined transcriptomic, QPCR, and microscopic analyses majority of migratory intestinal DCs are in the mucosal LP and strongly suggest that an important function of Ackr4+ fibroblasts is will use CCR7 to exit via LP-resident, CCL21-expressing lym- to regulate endothelial cells. The submucosa is rich in blood and phatic vessels that are some distance from the Ackr4+ cells. We lymphatic vessels, and capillaries in the LP originate from these considered the possibility that ACKR4 might regulate the migra- vessels. Thus, Ackr4+ fibroblasts are well positioned to regulate tory behavior of CCR7+ cells in the deeper layers of the intestine capillary growth and endothelial cell function in both arms of the in response to CCL21 derived from submucosal LECs. However, vasculature. Many of the genes preferentially expressed by Ackr4+ the rare DC populations present in the ME under steady-state fibroblasts encode proteins that regulate BECs. Interestingly, in conditions were not detectably affected by the absence of ACKR4. thymus, ACKR4 also appears to be expressed by stromal cells A potential activity of ACKR4 in the mucosa of the SI that would adjacent to blood vessels, specifically those at the cortico- not be relevant in the skin could be the regulation of CCL25 medullary junction where thymocyte progenitors enter and mature function. CCL25 contributes to the recruitment of activated CCR9+ thymocytes leave the thymus (48). However, the close physical T and B cells into the LP and epithelium of the SI (9). It is association between Ackr4+ fibroblasts and LECs in the submu- expressed by intestinal epithelial cells, most prominently those at cosa was particularly striking, and we found strong evidence of the base of the crypts (44), and it is conceivable that ACKR4- potential cross-talk between these two cell types through the Downloaded from mediated scavenging by neighboring submucosal fibroblasts could CCL21/ACKR4 and VEGFD/VEGFR3 pathways. Moreover, help maintain CCL25 gradients in this region. However, our un- several other genes preferentially expressed by Ackr4+ fibroblasts published data (C.A. Thomson, A.M. Mowat, and R.J.B. Nibbs) encode proteins that regulate LECs, including Tie2 (53, 54), have shown that plasmacytoid DCs, whose presence in the SI LP adrenomedullin (55), netrin-4 (41), and periostin (35). is dependent on CCR9 (45), are present in normal numbers in the The lymphatic vessel network of the intestine is unique. Not only

SI of Ackr4-deficient mice. Furthermore, we have found normal does it act as a crucial conduit for tissue fluid and migratory DCs, http://www.jimmunol.org/ numbers of T and B cells in the SI LP of steady-state Ackr4- but it also plays a crucial role in metabolism by mediating the deficient mice, and that activated intestinal T cells home normally transport of dietary fat out of the intestine in the form of chylo- to this site in these animals. They also have unaltered suscepti- microns. These lipoprotein particles enter the lymphatic capillaries bility to DSS colitis. Ackr4-deficient mice also have normal ar- in SI villi and travel in the lymph to the bloodstream. Moreover, chitecture and composition of intestinal secondary lymphoid under steady-state conditions lymphatic vessels in the intestine, organs such as the MLNs and Peyer’s patches. Although we unlike those in other tissues, are in a permanent state of regen- cannot formally exclude a role for ACKR4 in the development of eration and proliferation (56, 57). This is dependent on VEGFR2- smaller organized lymphoid structures, such as cryptopatches and and VEGFR3-induced expression of the Notch ligand delta–like 4 isolated lymphoid follicles (46), this seems unlikely in view of the (DLL4) (57). Smooth muscle cell–derived VEGFC is the key by guest on September 26, 2021 normal numbers of T and B cells present in the LP of Ackr4- VEGFR3 ligand in this context, but VEGFD also makes a con- deficient mice. Further studies are therefore required to determine tribution (56), and Gremlin1, which is a VEGFR2 agonist (58), the immunological functions of ACKR4 in the intestine. could play a role too. The strong and restricted expression of Adm GFP expression in Ackr4gfp/+ mice has been used to identify by Ackr4+ fibroblasts is also particularly interesting, given that discrete populations of stromal cells, including subpopulations of adrenomedullin is critical for the development and maintenance of thymic epithelial cells (47, 48) and LECs on the SCS ceiling of lymphatic vessels (55). Indeed, recent elegant work has shown LNs (11). A key finding of our study is that GFP expression in that inducing the conditional deletion of Calcrl (which encodes Ackr4gfp/+ mice also identifies a discrete population of fibroblasts the adrenomedullin receptor) exclusively in the LECs of adult in the mouse intestine. To our knowledge, these cells have not mice leads to intestinal lymphangiectasia, impaired lipid uptake, been described previously; indeed, prior to our study, the sub- defective LEC expression of DLL4 and junctional proteins, and mucosal stromal compartment was very poorly characterized. failure to resolve intestinal inflammation (59). Together, these Ackr4+ fibroblasts, which our surface immunophenotyping findings indicate that a number of the genes we have shown as suggests are a relatively homogeneous population of cells, are being highly and specifically expressed by submucosal Ackr4+ clearly anatomically distinct from mesenchymal cells located in fibroblasts encode mediators that control homeostasis of the local the LP, such as myofibroblasts, LP fibroblasts, crypt stromal cells lymphatic vasculature. (49), PDGFRa+ subepithelial cells (50), pericytes and smooth Their high expression of Grem1 raises the possibility that muscle cells (1). They are also distinct from the PDGFRa+ Ackr4+ fibroblasts may also be capable of regulating epithelial cell fibroblast-like cells that we observed in, and between, the LM and turnover in the intestine. In addition to acting as a VEGFR2 ag- CM layers and that have been reported previously by others (51). onist (58), Gremlin1 is an antagonist of bone morphogenetic In our FACS analysis, these various Ackr42 mesenchymal cells proteins (BMPs) 2, 4, and 7 (60), and this activity is thought to were either in the small GP382CD312 DN population or in the contribute to the maintenance of intestinal epithelial stem cells GP38+CD312 cell subsets that we designated P1 and P2. P2 (IESCs) (49, 61). Indeed, previous work has shown that CD34+ contained all PDGFRa+ iMCs that lacked Ackr4 expression. GP38+ iMCs are a major source of Grem1 transcripts in the mouse These cells appear to share many features with Ackr4+ iMCs, as intestine and can induce Gremlin1-dependent expansion of IESCs evident from their surface immunophenotype and the relatively in intestinal organoids in vitro (49). These CD34+GP38+ iMCs small number of differentially expressed genes identified in the were reported to be present in the LP (49), but because we have transcriptomic analysis. The surface immunophenotype of the P1 found that all Ackr4+ fibroblasts express CD34 and GP38, a large population suggests that these cells are less closely related to other fraction of CD34+GP38+ iMCs must reside in the submucosa. GP38+ iMCs. Indeed, transcriptomic analysis indicated that many Moreover, our data suggest that these submucosal cells express hundreds of genes are differentially expressed between P1 cells more Grem1. Indeed, when we purified Ackr4+ and Ackr42 cells 14 ACKR4 AND SUBMUCOSAL FIBROBLASTS IN THE INTESTINE directly from the CD34+GP38+ iMC population of the SI, tran- 11. Ulvmar, M. H., K. Werth, A. Braun, P. Kelay, E. Hub, K. Eller, L. Chan, . B. Lucas, I. Novitzky-Basso, K. Nakamura, et al. 2014. The atypical chemokine scripts encoding Gremlin1 were 4-fold more abundant in the receptor CCRL1 shapes functional CCL21 gradients in lymph nodes. Nat. + + Ackr4 cells (data not shown). The Ackr4 cells also showed higher Immunol. 15: 623–630. expression of genes encoding Wnt2b and R-spondin1 (data not 12. Bryce, S. A., R. A. M. Wilson, E. M. Tiplady, D. L. Asquith, S. K. Bromley, A. D. Luster, G. J. Graham, and R. J. B. Nibbs. 2016. ACKR4 on stromal cells shown), two other important IESC regulators reported to be spe- scavenges CCL19 to enable CCR7-dependent trafficking of APCs from inflamed cifically expressed by CD34+GP38+ iMCs (49, 62–64). In addition, skin to lymph nodes. J. Immunol. 196: 3341–3353. our transcriptomic data show that, compared to Ackr42 fibroblasts, 13. Heinzel, K., C. Benz, and C. C. Bleul. 2007. A silent chemokine receptor reg- + ulates steady-state leukocyte homing in vivo. Proc. Natl. Acad. Sci. USA 104: Ackr4 fibroblasts have elevated expression of Chrdl1, which en- 8421–8426. codes the BMP4 antagonist Chordin-like 1 protein (61), whereas 14. Ohl, L., M. Mohaupt, N. Czeloth, G. Hintzen, Z. Kiafard, J. Zwirner, T. Blankenstein, G. Henning, and R. Fo¨rster. 2004. CCR7 governs skin dendritic Bmp3, Bmp5 and Bmp7 were among the small number of genes whose cell migration under inflammatory and steady-state conditions. Immunity 21: + expression was significantly lower in Ackr4 cells (Supplemental 279–288. Table I). Collectively, these observations suggest that Ackr4+ fibro- 15. Braun, A., T. Worbs, G. L. Moschovakis, S. Halle, K. Hoffmann, J. Bo¨lter, A. Mu¨nk, and R. Fo¨rster. 2011. Afferent lymph-derived T cells and DCs use blasts are well equipped to regulate the IESC niche, and although, different chemokine receptor CCR7-dependent routes for entry into the lymph like GREM1+ mesenchymal cells in the human colon (61), they do node and intranodal migration. Nat. Immunol. 12: 879–887. not appear to lie immediately adjacent to the crypt bases where the 16. Comerford, I., R. J. Nibbs, W. Litchfield, M. Bunting, Y. Harata-Lee, S. Haylock-Jacobs, S. Forrow, H. Ko¨rner, and S. R. McColl. 2010. The atypical most primitive IESCs are found, it remains possible that by producing chemokine receptor CCX-CKR scavenges homeostatic chemokines in circula- Wnt2b, R-spondin1, and BMP antagonists, these cells can contribute tion and tissues and suppresses Th17 responses. Blood 116: 4130–4140. 17. Platt, A. M., C. C. Bain, Y. Bordon, D. P. Sester, and A. M. Mowat. 2010. An to IESC regulation and epithelial cell turnover. independent subset of TLR expressing CCR2-dependent macrophages promotes To our knowledge, our studies are the first to characterize colonic inflammation. J. Immunol. 184: 6843–6854. Downloaded from submucosal fibroblasts in mice. It will now be of interest to dissect 18. Scott, C. L., C. C. Bain, and A. M. Mowat. 2017. Isolation and identification of intestinal myeloid cells. Methods Mol. Biol. 1559: 223–239. their contribution to the maintenance of homeostasis in the SI and 19. Bain, C. C., and A. M. Mowat. 2012. CD200 receptor and macrophage function colon and to characterize their function in other contexts, such as in the intestine. Immunobiology 217: 643–651. during development, injury, inflammation, or tumorigenesis when 20.Farro,G.,M.Stakenborg,P.J.Gomez-Pinilla, E. Labeeuw, G. Goverse, M. Di Giovangiulio, N. Stakenborg, E. Meroni, F. D’Errico, Y. Elkrim, et al. 2017. angiogenesis and lymphangiogenesis are prominent and vitally CCR2-dependent monocyte-derived macrophages resolve inflammation and restore gut important processes. It will also be important to characterize, in the motility in postoperative ileus. Gut 66: 2098–2109. http://www.jimmunol.org/ 21. Ford, L. B., V. Cerovic, S. W. F. Milling, G. J. Graham, C. A. H. Hansell, and context of health and disease, the phenotype and function of R. J. B. Nibbs. 2014. Characterization of conventional and atypical receptors for submucosal mesenchymal cells in the human intestine. Ultimately, the chemokine CCL2 on mouse leukocytes. J. Immunol. 193: 400–411. a deeper understanding of the molecular and cellular properties of 22. Hansell, C. A. H., C. Schiering, R. Kinstrie, L. Ford, Y. Bordon, I. B. McInnes, C. S. Goodyear, and R. J. B. Nibbs. 2011. Universal expression and dual function submucosal fibroblasts and other mesenchymal cell populations of the atypical chemokine receptor D6 on innate-like B cells in mice. Blood 117: could reveal novel therapeutic approaches to treat diseases af- 5413–5424. fecting the intestine. 23. Ford, L. B., C. A. H. Hansell, and R. J. B. Nibbs. 2013. Using fluorescent chemokine uptake to detect chemokine receptors by fluorescent activated cell sorting. Methods Mol. Biol. 1013: 203–214. Acknowledgments 24. Bordon, Y., C. A. H. Hansell, D. P. Sester, M. Clarke, A. M. Mowat, and R. J. B. Nibbs.

2009. The atypical chemokine receptor D6 contributes to the development of experi- by guest on September 26, 2021 We thank staff at Glasgow University’s Central Research Facility for an- mental colitis. J. Immunol. 182: 5032–5040. imal husbandry. 25. Huang, W., B. T. Sherman, and R. A. Lempicki. 2009. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat. Protoc. 4: 44–57. Disclosures 26. Huang, W., B. T. Sherman, and R. A. Lempicki. 2009. Bioinformatics enrich- The authors have no financial conflicts of interest. ment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res. 37: 1–13. 27. Thomson, C. A., A. McColl, J. Cavanagh, and G. J. Graham. 2014. Peripheral inflammation is associated with remote global gene expression changes in the References brain. J. Neuroinflammation 11: 73. 1. Powell, D. W., I. V. Pinchuk, J. I. Saada, X. Chen, and R. C. Mifflin. 2011. 28. Chang, J. E., and S. J. Turley. 2015. Stromal infrastructure of the lymph node and Mesenchymal cells of the intestinal lamina propria. Annu. Rev. Physiol. 73: coordination of immunity. Trends Immunol. 36: 30–39. 213–237. 29. Worbs, T., U. Bode, S. Yan, M. W. Hoffmann, G. Hintzen, G. Bernhardt, 2. Owens, B. M. J., and A. Simmons. 2013. Intestinal stromal cells in mucosal R. Fo¨rster, and O. Pabst. 2006. Oral tolerance originates in the intestinal immune immunity and homeostasis. Mucosal Immunol. 6: 224–234. system and relies on antigen carriage by dendritic cells. J. Exp. Med. 203: 519–527. 3. Owens, B. M. J. 2015. Inflammation, innate immunity, and the intestinal stromal 30. Schulz, O., E. Jaensson, E. K. Persson, X. Liu, T. Worbs, W. W. Agace, and cell niche: opportunities and challenges. Front. Immunol. 6: 319. O. Pabst. 2009. Intestinal CD103+, but not CX3CR1+, antigen sampling cells 4. Bachelerie, F., A. Ben-Baruch, A. M. Burkhardt, C. Combadie`re, J. M. Farber, migrate in lymph and serve classical dendritic cell functions. J. Exp. Med. 206: G. J. Graham, R. Horuk, A. H. Sparre-Ulrich, M. Locati, A. D. Luster, et al. 3101–3114. 2013. International Union of Pharmacology. LXXXIX. Update on the extended 31. Johansson-Lindbom, B., M. Svensson, O. Pabst, C. Palmqvist, G. Ma´rquez, family of chemokine receptors and introducing a new nomenclature for atypical R. Fo¨rster, and W. W. Agace. 2005. Functional specialization of gut CD103+ chemokine receptors. [Published erratum appears in 2014 Pharmacol Rev. 66: dendritic cells in the regulation of tissue-selective T cell homing. J. Exp. Med. 467.] Pharmacol. Rev. 66: 1–79. 202: 1063–1073. 5. Nibbs, R. J. B., and G. J. Graham. 2013. Immune regulation by atypical che- 32. Yrlid, U., V. Cerovic, S. Milling, C. D. Jenkins, L. S. Klavinskis, and mokine receptors. Nat. Rev. Immunol. 13: 815–829. G. G. MacPherson. 2006. A distinct subset of intestinal dendritic cells responds 6. Gosling, J., D. J. Dairaghi, Y. Wang, M. Hanley, D. Talbot, Z. Miao, and selectively to oral TLR7/8 stimulation. Eur. J. Immunol. 36: 2639–2648. T. J. Schall. 2000. Cutting edge: identification of a novel chemokine receptor that 33. Hazell, G. G. J., A. M. G. Peachey, J. E. Teasdale, G. B. Sala-Newby, binds dendritic cell- and T cell-active chemokines including ELC, SLC, and G. D. Angelini, A. C. Newby, and S. J. White. 2016. PI16 is a shear stress and TECK. J. Immunol. 164: 2851–2856. inflammation-regulated inhibitor of MMP2. Sci. Rep. 6: 39553. 7. Townson, J. R., and R. J. B. Nibbs. 2002. Characterization of mouse CCX-CKR, 34. Shao, R., S. Bao, X. Bai, C. Blanchette, R. M. Anderson, T. Dang, a receptor for the lymphocyte-attracting chemokines TECK/mCCL25, SLC/ M. L. Gishizky, J. R. Marks, and X.-F. Wang. 2004. Acquired expression of mCCL21 and MIP-3beta/mCCL19: comparison to human CCX-CKR. Eur. periostin by human breast cancers promotes tumor angiogenesis through up- J. Immunol. 32: 1230–1241. regulation of vascular endothelial growth factor receptor 2 expression. Mol. 8. Girard, J.-P., C. Moussion, and R. Fo¨rster. 2012. HEVs, lymphatics and ho- Cell. Biol. 24: 3992–4003. meostatic immune cell trafficking in lymph nodes. Nat. Rev. Immunol. 12: 35. Kudo, Y., S. Iizuka, M. Yoshida, P. T. Nguyen, S. B. S. M. Siriwardena, 762–773. T. Tsunematsu, M. Ohbayashi, T. Ando, D. Hatakeyama, T. Shibata, et al. 2012. 9. Mowat, A. M., and W. W. Agace. 2014. Regional specialization within the in- Periostin directly and indirectly promotes tumor lymphangiogenesis of head and testinal immune system. Nat. Rev. Immunol. 14: 667–685. neck cancer. PLoS One 7: e44488. 10. Comerford, I., S. Milasta, V. Morrow, G. Milligan, and R. Nibbs. 2006. The 36. Chen, G., I. Nakamura, R. Dhanasekaran, E. Iguchi, E. J. Tolosa, P. A. Romecin, chemokine receptor CCX-CKR mediates effective scavenging of CCL19 in vitro. R. E. Vera, L. L. Almada, A. G. Miamen, R. Chaiteerakij, et al. 2017. Tran- Eur. J. Immunol. 36: 1904–1916. scriptional induction of periostin by a sulfatase 2-TGFb1-SMAD signaling axis The Journal of Immunology 15

mediates tumor angiogenesis in hepatocellular carcinoma. Cancer Res. 77: 632– 50. Kurahashi, M., Y. Nakano, L. E. Peri, J. B. Townsend, S. M. Ward, and 645. K. M. Sanders. 2013. A novel population of subepithelial platelet-derived growth 37. Venugopal, S., M. Chen, W. Liao, S. Y. Er, W.-S. F. Wong, and R. Ge. 2015. factor receptor a-positive cells in the mouse and human colon. Am. J. Physiol. Isthmin is a novel vascular permeability inducer that functions through cell- Gastrointest. Liver Physiol. 304: G823–G834. surface GRP78-mediated Src activation. Cardiovasc. Res. 107: 131–142. 51. Kurahashi, M., H. Zheng, L. Dwyer, S. M. Ward, S. D. Koh, and K. M. Sanders. 38. Jones, C. A., N. Nishiya, N. R. London, W. Zhu, L. K. Sorensen, A. C. Chan, 2011. A functional role for the ‘fibroblast-like cells’ in gastrointestinal smooth C. J. Lim, H. Chen, Q. Zhang, P. G. Schultz, et al. 2009. Slit2-Robo4 signalling muscles. J. Physiol. 589: 697–710. promotes vascular stability by blocking Arf6 activity. Nat. Cell Biol. 11: 1325– 52. Rao, M., B. D. Nelms, L. Dong, V. Salinas-Rios, M. Rutlin, M. D. Gershon, and 1331. G. Corfas. 2015. Enteric glia express proteolipid protein 1 and are a transcrip- 39. Rama, N., A. Dubrac, T. Mathivet, R.-A. Nı´ Cha´rthaigh, G. Genet, B. Cristofaro, tionally unique population of glia in the mammalian nervous system. Glia 63: L. Pibouin-Fragner, L. Ma, A. Eichmann, and A. Che´dotal. 2015. Slit2 signaling 2040–2057. through Robo1 and Robo2 is required for retinal neovascularization. Nat. Med. 53. Tammela, T., A. Saaristo, M. Lohela, T. Morisada, J. Tornberg, C. Norrmeń, 21: 483–491. Y. Oike, K. Pajusola, G. Thurston, T. Suda, et al. 2005. Angiopoietin-1 promotes 40. Lejmi, E., L. Leconte, S. Pe´dron-Mazoyer, S. Ropert, W. Raoul, S. Lavalette, lymphatic sprouting and hyperplasia. Blood 105: 4642–4648. I. Bouras, J.-G. Feron, M. Maitre-Boube, F. Assayag, et al. 2008. Netrin-4 in- 54. Morisada, T., Y. Oike, Y. Yamada, T. Urano, M. Akao, Y. Kubota, H. Maekawa, hibits angiogenesis via binding to neogenin and recruitment of Unc5B. Proc. Y. Kimura, M. Ohmura, T. Miyamoto, et al. 2005. Angiopoietin-1 promotes Natl. Acad. Sci. USA 105: 12491–12496. LYVE-1-positive lymphatic vessel formation. Blood 105: 4649–4656. 41. Larrieu-Lahargue, F., A. L. Welm, K. R. Thomas, and D. Y. Li. 2010. Netrin-4 55. Klein, K. R., and K. M. Caron. 2015. Adrenomedullin in lymphangiogenesis: induces lymphangiogenesis in vivo. Blood 115: 5418–5426. from development to disease. Cell. Mol. Life Sci. 72: 3115–3126. 42. Lejmi, E., I. Bouras, S. Camelo, M. Roumieux, N. Minet, C. Lere´-Deán, 56. Nurmi, H., P. Saharinen, G. Zarkada, W. Zheng, M. R. Robciuc, and K. Alitalo. T. Merkulova-Rainon, G. Autret, C. Vayssettes, O. Clement, et al. 2014. Netrin-4 2015. VEGF-C is required for intestinal lymphatic vessel maintenance and lipid promotes mural cell adhesion and recruitment to endothelial cells. Vasc. Cell absorption. EMBO Mol. Med. 7: 1418–1425. 6: 1. 57. Bernier-Latmani, J., C. Cisarovsky, C. S. Demir, M. Bruand, M. Jaquet, 43. Kane, R., C. Godson, and C. O’Brien. 2008. Chordin-like 1, a bone morpho- S. Davanture, S. Ragusa, S. Siegert, O. Dormond, R. Benedito, et al. 2015. DLL4 genetic protein-4 antagonist, is upregulated by hypoxia in human retinal peri- promotes continuous adult intestinal lacteal regeneration and dietary fat trans- cytes and plays a role in regulating angiogenesis. Mol. Vis. 14: 1138–1148. port. J. Clin. Invest. 125: 4572–4586. Downloaded from 44. Kunkel, E. J., J. J. Campbell, G. Haraldsen, J. Pan, J. Boisvert, A. I. Roberts, 58. Mitola, S., C. Ravelli, E. Moroni, V. Salvi, D. Leali, K. Ballmer-Hofer, E. C. Ebert, M. A. Vierra, S. B. Goodman, M. C. Genovese, et al. 2000. Lym- L. Zammataro, and M. Presta. 2010. Gremlin is a novel agonist of the major phocyte CC chemokine receptor 9 and epithelial thymus-expressed chemokine proangiogenic receptor VEGFR2. Blood 116: 3677–3680. (TECK) expression distinguish the small intestinal immune compartment: epi- 59. Davis, R. B., D. O. Kechele, E. S. Blakeney, J. B. Pawlak, and K. M. Caron. thelial expression of tissue-specific chemokines as an organizing principle in 2017. Lymphatic deletion of calcitonin receptor-like receptor exacerbates in- regional immunity. J. Exp. Med. 192: 761–768. testinal inflammation. JCI Insight 2: e92465. 45. Wendland, M., N. Czeloth, N. Mach, B. Malissen, E. Kremmer, O. Pabst, and 60. Brazil, D. P., R. H. Church, S. Surae, C. Godson, and F. Martin. 2015. BMP

R. Fo¨rster. 2007. CCR9 is a homing receptor for plasmacytoid dendritic cells to signalling: agony and antagony in the family. Trends Cell Biol. 25: http://www.jimmunol.org/ the small intestine. Proc. Natl. Acad. Sci. USA 104: 6347–6352. 249–264. 46. Pabst, O., H. Herbrand, M. Friedrichsen, S. Velaga, M. Dorsch, G. Berhardt, 61. Kosinski, C., V. S. W. Li, A. S. Y. Chan, J. Zhang, C. Ho, W. Y. Tsui, T. L. Chan, T. Worbs, A. J. Macpherson, and R. Fo¨rster. 2006. Adaptation of solitary intestinal R. C. Mifﬂin, D. W. Powell, S. T. Yuen, et al. 2007. Gene expression patterns of lymphoid tissue in response to microbiota and chemokine receptor CCR7 signal- human colon tops and basal crypts and BMP antagonists as intestinal stem cell ing. J. Immunol. 177: 6824–6832. niche factors. Proc. Natl. Acad. Sci. USA 104: 15418–15423. 47. Ribeiro, A. R., C. Meireles, P. M. Rodrigues, and N. L. Alves. 2014. Intermediate 62. de Lau, W., N. Barker, T. Y. Low, B.-K. Koo, V. S. W. Li, H. Teunissen, expression of CCRL1 reveals novel subpopulations of medullary thymic epithelial P. Kujala, A. Haegebarth, P. J. Peters, M. van de Wetering, et al. 2011. Lgr5 cells that emerge in the postnatal thymus. Eur. J. Immunol. 44: 2918–2924. homologues associate with Wnt receptors and mediate R-spondin signalling. 48. Lucas, B., A. J. White, M. H. Ulvmar, R. J. B. Nibbs, K. M. Sitnik, W. W. Agace, Nature 476: 293–297. W. E. Jenkinson, G. Anderson, and A. Rot. 2015. CCRL1/ACKR4 is expressed 63. Kim, K.-A., M. Kakitani, J. Zhao, T. Oshima, T. Tang, M. Binnerts, Y. Liu, in key thymic microenvironments but is dispensable for T lymphopoiesis at B. Boyle, E. Park, P. Emtage, et al. 2005. Mitogenic inﬂuence of human R-

steady state in adult mice. Eur. J. Immunol. 45: 574–583. spondin1 on the intestinal epithelium. Science 309: 1256–1259. by guest on September 26, 2021 49. Stzepourginski, I., G. Nigro, J.-M. Jacob, S. Dulauroy, P. J. Sansonetti, G. Eberl, 64. Valenta, T., B. Degirmenci, A. E. Moor, P. Herr, D. Zimmerli, M. B. Moor, and L. Peduto. 2017. CD34+ mesenchymal cells are a major component of the G. Hausmann, C. Cantu`, M. Aguet, and K. Basler. 2016. Wnt ligands secreted by intestinal stem cells niche at homeostasis and after injury. Proc. Natl. Acad. Sci. subepithelial mesenchymal cells are essential for the survival of intestinal stem USA 114: E506–E513. cells and gut homeostasis. Cell Rep. 15: 911–918. Supplementary Data.

Expression of the atypical chemokine receptor ACKR4 idenﬁes a novel populaon of intesnal submucosal ﬁbroblasts that preferenally expresses endothelial cell regulators

Carolyn A Thomson, Serge A van de Pavert, Michelle Stakenborg, Evelien Labeeuw, Gianluca Maeoli, Allan McI Mowat, Robert J B Nibbs

110 5 A Control Control 110 5 **** 105 ** ** DSS 4 Ackr4Controlgfp/+ Control DSS **** gfp/+ 105 ** 4 Ackr4DSS DSS 100 ** 3 100 3 * 95 2

DSS Score * 95 2 90 1 DSS Score

% original body weight 90 1 85 0 % original body weight 0 1 2 3 4 5 0 1 2 3 4 5 85 0 0 1 2 Days 3 4 5 0 1 2 3 Days4 5 Day Day B C Eosinophils PMN Monocytes * 10 20 6 4 Control ) ) ) DSS 3 3 9 3 15 3 4 8 10 2 7 2 5 1 Total cells (x10 cells Total Total cells (x10 cells Total Total cells (x10 cells Total

Colon length (cm) 6

5 0 0 0

DSS DSS DSS DSS Control Control Control Control

D Epithelial cells! DN! iMC! LEC! BEC! WT Control WT DSS Ackr4gfp/+ Control Ackr4gfp/+ DSS % of Max

GFP GFP GFP GFP GFP E F G GP38+CD31-! iMC! iMC! LEC!

40 *** 8 15000 40 Ackr4Controlgfp/+ Control gfp/+

) Ackr4DSS DSS 30 5 6 30

20 4 10000 20 cells (x10 cells MFI GFP + 10 2 10 GFP % Expressing GFP Expressing % % Expressing GFP Expressing % 0 0 5000 0

cLP cLP siLP MLN DSS DSS siLP MLN Control Control

Supplementary Figure 1. Ackr4 expression is restricted to colonic iMCs and MLN LECs in colic mice. Ackr4gfp/+ (Ackr4tm1Ccbl) and WT mice were given 2% DSS in the drinking water (DSS group) or maintained on sterile water (Control) for 5 days. (A) Change in original body weight (le) and clinical scores (right) in Ackr4gfp/+ mice (n=4-5 mice/group). *p<0.05, **p<0.01, ****p<0.0001, two-way ANOVA. (B) Colon lengths in Ackr4+/gfp mice at day 5. *p<0.05, unpaired Student’s t test. (C) Numbers of inflammatory leukocytes isolated from the colon of Ackr4gfp/ + mice at day 5. Eosinophils (SiglecF+), neutrophils (PMN; Ly6G+CD11bhi), and monocytes (SiglecF-Ly6G-CD11b +Ly6C+) were idenfied by flow cytometry aer gang for single, live, CD45+ leukocytes. (D) Representave overlaid flow cytometry histogram profiles showing GFP expression by colonic stromal cell subsets (GP38-CD31- ‘double negave’ cells (DN); intesnal mesenchymal cells (iMC); lymphac endothelial cells (LEC); blood vessel endothelial cells (BEC)). (E) Percentage of GP38+CD31- cells expressing GFP in colon, small intesne (SI) and MLN of Ackr4gfp/+ mice on day 5. ***p<0.001, two-way ANOVA. (F) Numbers of colonic iMCs expressing GFP, and the mean fluorescence intensity (MFI) of GFP expression, in Ackr4gfp/+ mice on day 5. (G) Percentage of LECs expressing GFP in the colon, SI and MLN of Ackr4gfp/+ mice on day 5. In A-C and E-G, datapoints from individual mice are shown, along with means ± 1SD. A Small intestine CD103 CD11c CD11c CD11c CD11c CD11c

MHC II Lin F4/80 CD11b B MLN CD11c CD11c CD11c CD11c CD11c CD103

MHC II Lin MHC II CD11b

C + - + + D CD103 CD11b CD103 CD11b Myeloid cells All DCs CD11cDC+MHCII subsets+CD64 - DCs 20 20 40 40 25 **** 10000 ) ) ) ) ) WT 5 WT 5 5 ) ****WT 5 5 SI 3 WT WTWT 15 15 30 3020 Ackr4-ACKR4/--/- ACKR4-/- Ackr4-/- Ackr4-Ackr4-/- /- 1000 10 10 20 2015

10 5 5 10 10 100 Nocellsof Nocellsof Total cells (x10 Total cells (x10 Nocellsof (x10 Total cell count Totalcell Total cells (x10 Total cells (x10 5

0 0 0 count (x10Totalcell 0 + + + - 0 MHCII cells 10 CD11c MHCII CD64 DCs PBS R848PBS R848 PBS R848PBS R848 8 DCs 10000 Mo/MF colon )

3 WT WTWT EosinophilsNeutrophils -/- -/- 6 Macrophages Ackr4 1000 Ackr4-Ackr4 /- - + - - CD103- CD11b- CD103 CD11b CD103 CD11b CD103+ CD11b-CD103+ CD11b+CD103- CD11b+

6 6 25 254 100 ) ) ) ) ) WTWT 5 5 WT 5 5 5 20 20 -/- Nocellsof -/- Nocellsof 2 Ackr4-ACKR4/- ACKR4 count Totalcell 10 4 4

15 15count (x10Totalcell 0 1 10 10 2 2 DCs 5 5 Mo/MF Total cells (x10 Total cells (x10 Total cells (x10 Total cells (x10 Nocellsof (x10 EosinophilsNeutrophils Macrophages 0 0 0 0 CD103- CD11b- CD103+ CD11b-CD103+ CD11b+CD103- CD11b+

PBS R848PBS R848 PBS R848PBS R848

Supplementary Figure 2. (A-B) Flow cytometric gang strategies used to idenfy DCs. (A) Small intesne. CD11c+MHCII+ cells were idenfied among single, live, CD45+ cells, before lineage (Lin)+ cells (CD3+ T cells and B220+ B cells/plasmacytoid DC) and F4/80+ macrophages were excluded. The remaining convenonal DCs were then divided into four populaons based on the expression of CD103 and CD11b. (B) MLN. Cells expressing lineage markers CD3 and B220 were excluded from single, live, CD45+CD11c+MHCII+ cells in the MLN. Migratory DCs were then idenfied as MHCIIhi, with remaining CD11c+MHCII+ cells classified as resident DCs. Migratory DCs were divided into four populaons based on expression of CD103 and CD11b. (C) Number of DCs in small intesne of WT and Ackr4-/- mice 24 hours aer R848 treatment. Numbers of cells within each of the four convenonal DC populaons in the small intesne of WT and Ackr4-deficient (Ackr4tm1.1Rjbn) (Ackr4-/-) mice 24 hours aer i.p. administraon of PBS or 100µg R848. Individual data points are shown, along with means + 1SD. Data are representave of three individual experiments. (D) DCs in the intesnal muscularis externa. Graphs show the total number of DCs (CD45+CD64-CD11c+MHCIIhi) (le hand graphs), and the number of cells in each DC subset (defined according to CD103 and CD11b expression) (right hand graphs), in single cell suspensions of the muscularis externa prepared from the small intesne (SI) and colon of WT and Ackr4-deficient (Ackr4tm1.1Rjbn) (Ackr4-/-) mice. Individual data points are shown, along with means +/- 1SD. In C and D, no stascally significant differences are present between WT and Ackr4-/- groups for any cell type. A DAPI DAPI GFP GFP

SM DAPI GFP

LP DAPI GFP LYVE1 CD31 LYVE1 CD31 100μm C VEGFR3 GFP CD45 EMCN SM LP

LM CM

VEGFR3 GFP CD45 EMCN Supplementary Figure 3. Ackr4-expressing iMCs are found adjacent to blood and lymphac vessels. (A) Representave fluorescent microscopy images of transverse secons of small intesne (le) and colon (right) from wild-type mice co-stained with DAPI (blue). Images were captured with sengs equivalent to those used on secons from Ackr4gfp/+ mice (Fig 5A-B) and show absence of GFP signal in WT intesne. (B-C) Fluorescent microscopy images of secons of intesne from Ackr4gfp/+ (Ackr4tm1Ccbl) mice immunostained with fluorescently labeled anbodies against the proteins indicated on each image. GFP idenfies Ackr4-expressing cells (B) Colon. Le panels show individual colors; right panel shows composite image. LP, lamina propria; SM, submucosa. (C) Small intesne. Right panel is a higher magnificaon of region boxed in the le panel. Arrows idenfy GFP+ fibroblasts adjacent to VEGFR3+ LECs. LP, lamina propria; SM, submucosa; CM, circular muscle; LM, longitudinal muscle. All images shown are representave of data from at least two experiments in which two or more mice were analyzed.

Transcripts significantly MORE abundant in GFP+ cells. Gene FC P-value Gene FC P-value Gene FC P-value Il6 8.79 8.72E-05 Postn 3.07 1.13E-06 Dusp1 2.26 3.51E-04 Cd81 8.66 8.28E-08 LOC100861693 3.03 3.78E-04 Sgms2 2.24 1.92E-07 Chodl 7.97 8.34E-07 Grem1 3.03 1.68E-06 Gpr133 2.24 2.31E-04 Gm11309 6.65 1.09E-04 Ism1 3.01 6.13E-04 NONCODE1 2.23 1.41E-04 1700019G06Rik 6.55 1.41E-04 NONCODE1 2.97 4.33E-06 Ctsh 2.22 3.61E-06 Adm 5.30 1.16E-04 Slit2 2.94 5.78E-06 Plxdc2 2.22 1.16E-04 Gm11314 5.20 4.78E-04 Ntn4 2.91 9.64E-04 Adamtsl1 2.20 2.99E-05 Gfpt2 5.15 1.70E-04 NONCODE1 2.90 4.09E-05 Slc16a2 2.17 3.80E-04 Cilp 5.11 1.28E-04 Aebp1 2.84 8.82E-06 Nbia 2.16 2.10E-04 Piezo2 5.10 3.72E-07 Adamtsl3 2.83 9.08E-07 NONCODE3 2.15 6.83E-04 Pcolce2 5.00 1.71E-07 NONCODE2 2.79 1.88E-05 Flrt3 2.15 5.76E-05 Prss23 4.89 4.88E-05 NONCODE1 2.79 5.36E-04 Fndc1 2.15 5.01E-04 Olfml2b 4.59 1.43E-05 Fam180a 2.70 5.65E-04 Foxp2 2.15 4.83E-06 Serpine1 4.58 2.12E-04 Fign 2.70 3.41E-04 NONCODE 1 2.15 1.35E-04 NONCODE1 4.54 2.10E-06 LOC100862089 2.69 3.29E-04 Prlr 2.14 1.04E-05 Npy1r 4.40 3.77E-04 Rbp1 2.68 7.12E-04 Gm11816 2.13 7.68E-04 Ebf1 4.25 4.02E-07 Fam71a 2.68 3.23E-04 Serpini1 2.12 5.97E-04 Figf 3.63 1.43E-04 Lgi2 2.67 9.71E-04 Enpp2 2.11 3.41E-04 Gdf6 3.58 2.85E-04 Osr1 2.65 3.45E-04 Pdgfrl 2.11 2.56E-05 Dhrs3 3.57 4.39E-06 Cpxm1 2.61 1.53E-06 Galnt16 2.11 1.02E-04 Ar 3.48 5.56E-07 Col14a1 2.50 5.50E-05 Id3 2.10 4.74E-04 Col8a1 3.45 5.78E-07 Tek 2.48 1.18E-05 Smpdl3a 2.09 8.48E-05 Ier2 3.43 1.68E-04 Hk2 2.47 6.81E-04 Chrdl1 2.09 4.41E-04 Pi16 3.42 4.65E-05 Epyc 2.47 2.55E-05 Gm15079 2.09 9.76E-06 NONCODE2 3.42 7.78E-05 Pla2g2e 2.45 1.13E-04 Pde12 2.08 4.87E-04 Mt2 3.41 2.85E-04 Klf4 2.45 1.39E-04 Cldn10 2.08 3.12E-04 Flrt2 3.36 8.36E-05 Gm17034 2.43 8.67E-04 Cgnl1 2.08 5.20E-05 NONCODE1 3.35 6.28E-04 Cd248 2.41 2.37E-04 NONCODE 1 2.08 3.98E-04 Rcan1 3.31 6.27E-05 Ackr4 2.40 1.19E-04 Adcyap1r1 2.08 7.03E-04 Gpc3 3.28 6.54E-07 Igp6 2.39 8.79E-05 Bag3 2.08 4.11E-04 Gm26802 3.28 3.43E-04 Cd68 2.38 7.25E-04 Abi3bp 2.08 7.81E-06 Gm12158 3.23 1.55E-05 Cpz 2.36 1.91E-05 Gm17035 2.08 4.88E-04 NONCODE1 3.22 6.24E-06 NONCODE1 2.36 7.61E-08 Syne1 2.06 1.87E-04 NONCODE1 3.19 5.63E-05 Y0G0104A12 2.33 3.16E-06 Ugdh 2.04 7.47E-05 Scara5 3.15 9.60E-04 Lsamp 2.31 1.77E-05 Gm26889 2.01 3.97E-05 Lbp 3.14 3.59E-05 NONCODE1 2.31 5.16E-04 Pvr 2.01 7.76E-04 NONCODE3 3.12 2.26E-04 NONCODE3 2.30 1.50E-04 NONCODE 3 2.00 4.42E-06 Fam129a 3.12 4.09E-07 Myc 2.30 1.90E-04 Gm17083 2.00 5.37E-04 Vldlr 3.11 1.28E-04 Cadm3 2.28 1.00E-04 Fap 3.08 2.05E-04 Scd1 2.27 8.50E-04 Transcripts significantly LESS abundant in GFP+ cells. Gene FC P-value Gene FC P-value Gene FC P-value Mmp10 -19.23 6.16E-08 Il34 -2.84 4.37E-04 Casq2 -2.39 3.87E-06 Pdlim3 -6.57 1.61E-04 Hpse2 -2.78 1.09E-04 NONCODE1 -2.37 9.17E-04 Acta2 -6.18 1.53E-05 Hmcn1 -2.78 2.25E-06 NONCODE1 -2.36 2.78E-06 NONCODE3 -4.98 3.42E-06 Gm16054 -2.76 1.17E-06 Chp2 -2.31 9.18E-04 NONCODE1 -4.36 1.04E-04 Myh11 -2.75 7.17E-05 Ptgs1 -2.29 2.40E-04 LOC100862378 -4.27 2.67E-04 Cxcl14 -2.73 1.14E-04 Nid2 -2.23 1.47E-05 Bmp5 -4.22 2.60E-05 NONCODE3 -2.71 4.20E-04 NONCODE2 -2.20 1.01E-04 Ednrb -4.12 1.27E-04 Lphn3 -2.70 9.23E-07 NONCODE2 -2.19 1.15E-05 Igp3 -3.90 3.14E-06 NONCODE2 -2.68 3.53E-05 Kitl -2.19 9.48E-04 Srpx2 -3.65 7.48E-06 Mir5124 -2.68 1.64E-04 NONCODE1 -2.16 2.11E-04 Abcc9 -3.63 1.19E-05 Sept4 -2.67 6.24E-06 Epas1 -2.15 7.35E-04 Tagln -3.23 4.28E-04 Col12a1 -2.67 2.14E-04 Hpse -2.08 9.86E-04 Bmp7 -3.15 2.05E-04 Bmp3 -2.64 3.11E-04 Pdlim1 -2.08 1.60E-04 Aldh1a3 -2.98 2.45E-05 Tnc -2.55 1.28E-04 Foxf2 -2.07 1.55E-05 NONCODE3 -2.93 2.97E-05 Rdh10 -2.55 9.90E-04 Lama1 -2.07 2.25E-05 Gpr81 -2.89 9.50E-05 NONCODE3 -2.46 1.47E-05 NONCODE1 -2.06 5.36E-05 Col6a5 -2.86 5.55E-06 NONCODE3 -2.42 3.62E-04 Pde1c -2.05 6.70E-04 NONCODE1: Non-coding transcript idenfied by NONCODE:SenseNoExonic; NONCODE2: Non-coding transcript idenfied by NONCODE:Ansense; NONCODE3: Non-coding transcript idenfied by NONCODE:Linc Supplementary Table 1: Genes showing stascally significant >2-fold differenal expression between the GFP+ and GFP- subsets of CD45-CD31-GP38+CD146- cells isolated by from the small intesne of Ackr4gfp/+ (Ackr4tm1Ccbl) mice. FC, fold change. Of the transcripts more abundant in GFP+ SI fibroblasts, those encoding secreted factors are labeled in red, while those encoding cell surface proteins are green.