Conventional Dendritic Cells Impair Recovery after Myocardial Infarction Jun Seong Lee, Se-Jin Jeong, Sinai Kim, Lorraine Chalifour, Tae Jin Yun, Mohammad Alam Miah, Bin Li, Abdelilah This information is current as Majdoubi, Antoine Sabourin, Tibor Keler, Jean V. Guimond, of September 28, 2021. Elie Haddad, Eui-Young Choi, Slava Epelman, Jae-Hoon Choi, Jacques Thibodeau, Goo Taeg Oh and Cheolho Cheong

J Immunol 2018; 201:1784-1798; Prepublished online 10 Downloaded from August 2018; doi: 10.4049/jimmunol.1800322 http://www.jimmunol.org/content/201/6/1784 http://www.jimmunol.org/ Supplementary http://www.jimmunol.org/content/suppl/2018/08/08/jimmunol.180032 Material 2.DCSupplemental References This article cites 76 articles, 28 of which you can access for free at: http://www.jimmunol.org/content/201/6/1784.full#ref-list-1

Why The JI? Submit online. by guest on September 28, 2021

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

• 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. The Journal of Immunology

Conventional Dendritic Cells Impair Recovery after Myocardial Infarction

Jun Seong Lee,*,†,‡ Se-Jin Jeong,x Sinai Kim,x Lorraine Chalifour,{ Tae Jin Yun,†,‖ Mohammad Alam Miah,*,† Bin Li,*,# Abdelilah Majdoubi,*,‡ Antoine Sabourin,*,‡ Tibor Keler,** Jean V. Guimond,†† Elie Haddad,*,‡‡ Eui-Young Choi,xx Slava Epelman,{{,‖‖,## Jae-Hoon Choi,*** Jacques Thibodeau,*,‡ Goo Taeg Oh,x and Cheolho Cheong*,†,‖,1

Ischemic myocardial injury results in sterile cardiac inflammation that leads to tissue repair, two processes controlled by mono- nuclear phagocytes. Despite global burden of cardiovascular diseases, we do not understand the functional contribution to path- ogenesis of specific cardiac mononuclear phagocyte lineages, in particular dendritic cells. To address this limitation, we used detailed + +

lineage tracing and genetic studies to identify bona fide murine and human CD103 conventional dendritic cell (cDC)1s, CD11b Downloaded from cDC2s, and plasmacytoid DCs (pDCs) in the heart of normal mice and immunocompromised NSG mice reconstituted with human CD34+ cells, respectively. After myocardial infarction (MI), the specific depletion of cDCs, but not pDCs, improved cardiac function and prevented adverse cardiac remodeling. Our results showed that fractional shortening measured after MI was not influenced by the absence of pDCs. Interestingly, however, depletion of cDCs significantly improved reduction in fractional shortening. Moreover, fibrosis and cell areas were reduced in infarcted zones. This correlated with reduced numbers of cardiac

macrophages, neutrophils, and T cells, indicating a blunted inflammatory response. Accordingly, mRNA levels of proinflamma- http://www.jimmunol.org/ tory cytokines IL-1b and IFN-g were reduced. Collectively, our results demonstrate the unequivocal pathological role of cDCs following MI. The Journal of Immunology, 2018, 201: 1784–1798.

yocardial infarction (MI) caused by a thrombotic oc- forces within the myocardium lead to adverse remodeling changes clusion of a coronary artery is the most frequent cause in the remote, uninfarcted myocardial segments (3–5). Ischemic M of cardiac dysfunction and heart failure (1). Ischemia heart disease is the leading cause of death worldwide. Myocardial results in oxygen and nutrient deprivation, leading to car- necrosis after MI triggers immuno-inflammatory reactions that are diomyocyte cell death in the nonperfused territory. This is ac- integral to the healing process but also contribute to left ventric- companied by an intense sterile inflammatory response and a ular (LV) dysfunction (6, 7). Such inflammation results from the by guest on September 28, 2021 gradual collagen-driven scar formation that is required to replace activation of tissue-resident immune cells, which produce pro- the necrotic myocardium (2). Ultimately, altered hemodynamic inflammatory cytokines and chemokines capable of attracting

*De´partement de Microbiologie, Infectiologie et Immunologie, Universite´ de Montre´al, Que´be´coise d’Hypertension Arte´rielle – Le Fonds de Recherche du Que´bec - Sante´ J1 Montreal, Quebec H3T 1J4, Canada; †Institut de Recherches Cliniques de Montre´al, Mon- scholarship. treal, Quebec H2W 1R7, Canada; ‡Laboratoire d’Immunologie Mole´culaire, Universite´ de x J.S.L., G.T.O., J.-H.C., and C.C. designed the experiments. J.S.L. executed the ma- Montre´al, Montreal, Quebec H3T 1J4, Canada; Immune and Vascular Cell Network Re- jority of experiments presented. L.C. performed myocardial infarctions. S.-J.J., S.K., search Center, National Creative Initiatives, Department of Life Sciences, Ewha Womans { and E.-Y.C. did Masson trichrome staining and wheat germ agglutinin staining as University, Seoul 120-750, South Korea; Lady Davis Institute, Division of Experimental ‖ well as echocardiography. T.J.Y. performed phagocytosis assay and marker analysis Medicine, McGill University, Montreal, Quebec H3T 1E2, Canada; Division of Experi- for plasmacytoid dendritic cells. M.A.M. and B.L. prepared mouse and humanized mental Medicine, Department of Medicine, McGill University, Montreal, Quebec H4A 3J1, mouse hearts for FACS. A.M. and A.S. helped analyze flow cytometry data for Canada; #De´partement de Biologie Mole´culaire, Universite´ de Montre´al, Montreal, Quebec identification of dendritic cell subsets. T.K. provided human Flt3L. E.H. and J.V.G. H3T 1J4, Canada; **Celldex Therapeutics, Hampton, NJ 08827; ††Centre de Sante´ et de provided humanized mice. J.S.L., S.E., J.-H.C., G.T.O., J.T., and C.C. interpreted the Services Sociaux Jeanne-Mance, Montreal, Quebec H2T 2R9, Canada; ‡‡Centre Hospitalier xx data. J.S.L., S.E., J.-H.C., G.T.O., and J.T. wrote the manuscript. Universitaire Sainte-Justine Research Center, Montreal, Quebec H3T 1C5, Canada; Gang- nam Severance Hospital, Yonsei University College of Medicine, Seoul 06273, South Address correspondence and reprint requests to Dr. Jacques Thibodeau or Dr. Goo Korea; {{Peter Munk Cardiac Center, Toronto, Ontario M5G 2N2, Canada; ‖‖Ted Rogers Taeg Oh, Laboratoire d’Immunologie Mole´culaire, De´partement de Microbiologie, Centre for Heart Research, Toronto, Ontario M5G 1X8, Canada; ##Toronto General Hospital Infectiologie et Immunologie, Universite´ de Montre´al, Faculte´ de Me´decine, C.P. Research Institute, University Health Network, Toronto, Ontario M5G 2C4, Canada; and 6128 Succursale Centre-Ville, 2900 Boulevard Edouard Montpetit, Montreal, QC ***Department of Life Science, College of Natural Sciences, Research Institute for Natural H3C 3J7, Canada (J.T.) or National Creative Research Center for Immune and Vas- Sciences, Hanyang University, Seoul 04763, South Korea cular Cell Network, Department of Life Sciences, Ewha Womans University, Seodaemoon-gu, Seoul 120-750, Korea (G.T.O.). E-mail addresses: Jacques. 1Deceased. [email protected] (J.T.) or [email protected] (G.T.O.) ORCIDs: 0000-0001-6675-3265 (J.S.L.); 0000-0002-6375-5334 (S.-J.J.); 0000-0002- The online version of this article contains supplemental material. 5985-7733 (S.K.); 0000-0002-2053-4793 (A.M.); 0000-0003-3732-0190 (E.-Y.C.); 0000-0002-0790-9634 (J.T.). Abbreviations used in this article: BM, bone marrow; cDC, conventional DC; DC, dendritic cell; DN, double-negative; DT, diphtheria toxin; DTR, DT receptor; Flt3L, Received for publication March 5, 2018. Accepted for publication July 6, 2018. Flt3 ligand; hu-mice, humanized mice; LN, lymph node; LV, left ventricular; MF, This work was supported by grants from the Canadian Foundation for Innovation macrophage; MHC II, MHC class II; MI, myocardinal infarction; moDC, monocyte (John R. Evans Leaders Fund to C.C.) and the Canadian Institutes of Health Research DC; pDC, plasmacytoid DC; poly I:C, polyinosinic-polycytidylic acid; pre-cDC, (CIHR) (MOP 125933 to C.C. and J.T., MOP 136802 to J.T., and MOP 133050 to precursor cDC; RA, right atrium; Seg, segment; Treg, regulatory T cell; TTC, tri- C.C.), by Canadian HIV Cure Enterprise Team Grant HIG-133050 (to C.C. and E.H.), phenyltetrazolium chloride; WT, wild type. and by a National Research Foundation of Korea grant funded by the Korean gov- ernment (2012R1A3A2026454 to G.T.O.). J.S.L. was supported by the Fonds de Copyright Ó 2018 by The American Association of Immunologists, Inc. 0022-1767/18/$35.00 Recherche du Que´bec – Nature et Technologies. J.T. holds the Saputo Research Chair. C.C. was a recipient of a CIHR New Investigator Award and held a Socie´te´ www.jimmunol.org/cgi/doi/10.4049/jimmunol.1800322 The Journal of Immunology 1785 leukocytes from the blood into the region of cardiac injury (8). mouse experiments were performed according to guidelines of the Cana- Various animal and human studies have provided evidence that the dian Council on Animal Care. expansion and/or recruitment of inflammatory leukocytes such as Heart single-cell isolation and flow cytometry mononuclear phagocytes, dendritic cells (DCs), monocytes, and macrophages (MFs) either enhance or suppress the ability of the Cardiac single cells were isolated according to a previously described method (25) with minor modifications. Mouse hearts were incubated for myocardium to recover after cardiac injury (9). One key limitation 40 min at 37˚C with gentle shaking in HBSS containing 675 U/ml has been the inability to distinguish the function of MFs and DCs, collagenase I, 18.75 U/ml collagenase XI, and 9 U/ml hyaluronidase which can express similar cell surface markers, such as MHC (Sigma-Aldrich). Cells were centrifuged on a discontinuous Percoll class II (MHC II), CD11c, and F4/80 (10). In this context, it is (Sigma-Aldrich) gradient (40 and 80%), and leukocytes were collected. Mouse spleen and lymph nodes (LNs) were digested in 400 U/ml of critical to precisely define the origin, developmental features, and collagenase D (Roche) in RPMI 1640 for 30 min at 37˚C. Foxp3+ T cells functions of DCs in the heart. were stained using a Foxp3 staining buffer set (eBioscience) according DCs are professional APCs that bridge innate and adaptive to a previously described method (31); all other intracellular stainings immunity (11). In mice, their ontogeny has been the subject of were performed using a Fixation/Permeabilization Solution Kit (BD many important discoveries in recent years (12). It is now estab- Biosciences). Stained cells were acquired or sorted using an LSRFor- tessa flow cytometer or FACSAria III (BD Biociences), respectively, lished that common DC progenitors are heterogeneous and give and analyzed with FlowJo (Treestar). Sorted cells were cytospined at rise to plasmacytoid DCs (pDCs) or precursor DCs, which all 500 rpm for 5 min using Cytospin4 (Thermo Fisher Scientific) for move from the bone marrow (BM) to the blood, the secondary phagocytosis analysis. lymphoid organs, and to nonhematopoietic tissues (13). These Immunohistochemistry and confocal microscopy cells express MHC II molecules and can act as Ag-presenting cells Downloaded from to power proatherogenic T cell immunity (14). However, through CD11c-YFP mice hearts were perfused with cold PBS and fixed in 4% formaldehyde for 1 h. After careful removal of the pericardial fat and blood the production of IDO, pDCs also promote tolerance by modu- vessels, hearts were processed using standard procedures, embedded in lating regulatory T cell (Treg)-suppressive functions during au- OCT, frozen, and sectioned (12-mm thick). Cryosections were singly toimmunity and atherosclerosis (15, 16). immunostained with the indicated Abs. Immunoreactive proteins were BM and blood precursor cDCs (pre-cDCs) are heterogeneous visualized using an ImmPRESS-AP Anti-Rabbit Ig Polymer Detection Kit and an ImmPRESS HRP Anti-Mouse Ig Polymer Detection Kit. All ex- and already committed to the conventional DC (cDC)1 and cDC2 periments were performed according to the manufacturer’s instructions. http://www.jimmunol.org/ lineages (17). It is not clear, however, if pre-cDCs can be found in Confocal images were acquired along the z axis using an LSM 710 laser other tissues, such as the heart. In lymphoid organs, the resident scanning confocal microscope (Carl Zeiss), and Z-stacked images were cDC1s and cDC2s express CD8a and CD4, respectively. In non- analyzed with Imaris software (Bitplane). lymphoid tissues such as the skin, cDC1s and cDC2s are rather Abs and reagents characterized by the expression of CD103 and CD11b, respec- tively (18). In addition to these phenotypic differences, cDCs are Abs to mouse Foxp3 for FACS analysis was from eBioscience. All other Abs were from BioLegend. Cells were incubated in culture supernatant from the functionally distinguishable. cDC1s are dependent on the tran- 2.4G2 hybridoma (Fc Receptor Block, HB-197; ATCC) prior to staining + scription factors IRF8, ID2, and BatF3. These CD103 cDCs are with the indicated Abs. Polyinosinic-polycytidylic acid (poly I:C; Sigma- potent cross-presenting cells involved in Th1 polarization in re- Aldrich) was given i.v. at a dose of 25 mg to induce DC maturation. by guest on September 28, 2021 sponse to pathogens. cDC2s, rather, rely on IRF4 and several Cell culture additional transcriptional factors, such as RelB, RBP-J, and IRF2 for their development and promote activation of Th2 and Th17 For phagocytosis assays, cardiac cells from Flt3L-injected C57BL/6 mice were incubated with Fluoresbrite YG Carboxylate Microspheres cells (19–23). As a whole, it is well established that DCs, although 1.0 mm (Polysciences) at 37˚C for 1 h. Then, cells were incubated in Fc poorly phagocytic, are highly efficient in the capture, processing, receptor block and stained with Abs at 4˚C for cell sorting. All mouse and presentation of Ags to T cells (24, 25). cells were incubated in RPMI 1640 containing 5% FBS, 13 antibiotic- In previous studies, the respective roles of cDCs and pDCs in a antimycotic, GlutaMAX, MEM nonessential amino acids, and 2-ME clinically relevant disease could not be established because of the (Life Technologies). lack of tools to separately track or deplete specific subsets. Although BM chimeras and depletion of DCs pDCs can be selectively depleted by diphtheria toxin (DT) in mice Although in the mouse hematopoietic cell compartment the expression of expressing the DT receptor (DTR) under the control of the human genes under the control of Zbtb46 and the human BDCA2 promoters is BDCA2 promoter, a similar approach to track cDCs has only been restricted to cDCs and pDCs, respectively, DT-driven depletion experi- developed recently (26). Indeed, Zbtb46 is not shared by other ments were performed in BM chimeras to rule out any interference from hematopoietic cell types, and its expression marks the cDC lineage the possible depletion of nonimmune cells (26, 28, 32). C57BL6/J male mice (8-wk-old) were lethally g-irradiated (2 3 5 Gy, 4 h apart) and (27, 28). In this study, we sought to characterize DC subsets in reconstituted for 5 wk with BM of BDCA2-DTR or Zbtb46-DTR mice. For healthy and infarcted myocardium of C57BL/6 mice and human- systemic DC depletion, mice were injected i.v. with DT (20 ng/g weight; in ized mice (hu-mice). Critically, using depletion strategies based on PBS; Sigma Chemical, St. Louis, MO). BDCA2 and Zbtb46, we addressed the functional impact of deleting MI surgery individual DC subsets on ischemic cardiac injury. Cardiac surgery was performed by the Surgery Core of the Lady Davis Institute (33). C57bl/6J reconstituted with Zbtb46-DTR or BDCA2-DTR mouse BM Materials and Methods for 5 wk were anesthetized with isoflurane and intubated. Analgesia was Mice provided by an injection of slow release buprenorphine. A permanent occlusion 2/2 in blood flow downstream of the left anterior descending coronary artery C57BL/6, Ldlr (29), CX3CR1-GFP, Zbtb46-GFP, Zbtb46-DTR, and ∼ NSG (NOD/SCID/gc-null) mice were purchased from The Jackson Lab- 2 mm distal to the left atrial appendage was created using a 7-0 silk suture 2/2 ligature. Using a similar surgery, it was estimated that this ligation places oratory (Bar Harbor, ME). Flt3 mice (30) (I. Lemischka, Mount Sinai ∼ School of Medicine) were a generous gift of M. Nussenzweig. BDCA2- 38% of the left ventricle at risk for infarction (34). DTR mice (26) were from M. Colonna (Washington University, St. Louis, Tissue harvesting, histology, and morphometric analyses MO). For hu-mice, NSG mice reconstituted with human CD34+ cells (hu-mice) were used. To expand the number of DCs, mice were injected Mice were deeply euthanized, and the hearts were perfused with saline i.p. with 2 mg of human Flt3 ligand (Flt3L) per day for nine consecutive followed by catheter (Anatech, Battle Creek, MI). To measure myocardial days (15). All mice were bred under specific pathogen-free conditions. All infarct size, hearts were sectioned transversely into 3-mm thick slices and 1786 cDCs IMPAIR RECOVERY AFTER MYOCARDIAL INFARCTION incubated in 2% triphenyltetrazolium chloride (TTC) at 37˚C for 30 min to further characterized using a combination of CD103 and CD11b identify the noninfarcted and infarcted areas. The infarcted area appeared out of the same exclusionary gate set. We observed that the ex- as a yellow-white area that was unstained by TTC. The true infarct size on pression of these markers was mutually exclusive, defining three the TTC-stained slices was measured directly and was calculated as the + 2 2 + ratio (%) of cumulative infarct area to the entire LV area using ImageJ subpopulations: CD103 CD11b cDC1s, CD103 CD11b software (National Institutes of Health). cDC2s, and double-negative (DN) cDCs. To further characterize the putative cDC1s and cDC2s, total cDCs were gated on CD11b+ Masson trichrome staining or CD103+ cells and analyzed for the expression of lineage and To measure myocardial fibrosis, Masson trichrome stain was used for each subset-specific markers (Fig. 1B–E). Their cDC origin was con- paraffin-embedded transaxial LV section. We stained two slides of each firmed, as both subsets were Zbtb46+ (Fig. 1B, 1C). In addition, heart (PBS, n = 10 and DT, n = 8) with the Masson trichrome staining and quantified fibrous infarct regions (blue color in red, green, and blue mode cardiac cDC1s express Clec9a, Flt3/CD135, DEC205, CD24, and and white color in black/white mode). The total levels of fibrosis in the CD283, whereas cDC2s specifically expressed low levels of areas of the fibrous infarct regions were measured with ImageJ software CD115/M-CSFR, F4/80, CX3CR1, and Ly6C (Fig. 1B–E). cDCs and expressed as percentages of the total LV surface. An independent did not express NK cells, B cells, T cells, granulocytes, or pDCs investigator who was blinded to the experimental groups quantified all markers (Supplemental Fig. 1A, 1B). Interestingly, we observed experiments. dichotomous expression of XCR1 and CD172a in mature cDC Wheat germ agglutinin staining (Fig. 1F) (17). We also identified similar dichotomous expression To quantify individual myocyte size, some sections were stained with in the DN cDC population, suggesting that these cells may rep- fluorescein FITC-conjugated wheat germ agglutinin (5 mg/ml; Invitrogen, resent a novel population of tissue pre-cDCs (Fig. 1F).

Grand Island, NY). Myocyte area was assessed with ImageJ software. The Downloaded from mean cardiomyocyte area was evaluated by measurement of 500 cells per Bona fide cDCs in the heart heart (6–9 hearts per genotype). cDCs display a variety of distinctive developmental, functional, and Echocardiography phenotypic characteristics (12). We performed in vivo and ex vivo Echocardiography was performed under anesthesia by inhalation of 1.5–2% experiments to monitor some of these parameters in the mouse heart, isoflurane (Forane isoflurane; Abbott Scandinavia AB, Sweden). Images including the impact of depleting Zbtb46-expressing cells on the were acquired using the Vevo 2100 system (VisualSonics) with a 30-MHz DC compartment. First, to definitely confirm the DC origin of the http://www.jimmunol.org/ linear array transducer. The left ventricle was investigated in both long- populations identified above, we examined whether Flt3L could axis and short-axis views during 3 wk after MI. We selected the two- affect cardiac cDC numbers. Flt3L is a DC growth factor, promoting dimensional image slice in which infarction-related regional wall motion abnormality was most significant, then used the M-mode scan, which is the survival and differentiation of progenitors (35). Injection of Flt3 to crossing site of the most significant regional wall motion abnormalities. To mice was shown to dramatically increase the numbers of mature reduce measurement variability by observer, after image acquisition, au- DC subpopulations (36). Mice treated with Flt3L showed an increase tomatic border detection software was used to measure LV systolic and in total DCs (CD11c+ MHC II+) in the heart (Fig. 1G, top) as well as diastolic dimensions, and then fractional area change was calculated using those values. All the measurement variables such as LV end dia- in the spleen (Supplemental Fig. 1C). In contrast, the proportion + + stolic and systolic dimensions, stroke volume, and cardiac output are in of CD11b CD64 MFs, in which development is dependent on the Supplemental Table I, and TTC-stained data is in the Supplemental CD115 rather than CD135 (12, 37), was decreased (Fig. 1G, bot- by guest on September 28, 2021 Fig. 4F. tom). Absolute numbers revealed that Flt3 increases predominantly + Quantitative real-time PCR analysis thenumberofCD103 and DN DCs subsets (Fig. 1H, 1I). Cardiac CD103+ DCs were particularly sensitive to the effect of Flt3L, Total heart RNA was extracted from a single-cell suspension using TRIzol showing a 5-fold increase (Fig. 1I). Interestingly, heart DN cDCs Lysis Reagent (5 PRIME GmbH Hamburg, Germany), and cDNA was cells showed a dramatic expansion (∼30-fold)inresponseto9d prepared using the RevertAid First Strand cDNA Synthesis Kit (Fermentas). 2 2 Quantitative real-time PCR was performed using StepOnePlus Real-Time of Flt3 treatment, both in the heart (CD103 CD11b )andspleen PCR systems (Applied Biosystems) instrument and SYBR Green PCR (CD8a2 CD11b2) (Fig. 1H, 1I, Supplemental Fig. 1D). The Master Mix (KAPA Biosystems) and was quantified on StepOne Software effect of Flt3L was mediated through the FLT3 receptor (CD135), v2.3 (Applied Biosystems) using DCT method and b-actin as a control. as the expansion of total cDC numbers was not observed in Statistics CD135-deficient (FLT3 KO) mice (Fig. 1J). All statistical significance between two groups was tested using a Mann– Next, to confirm the cDC nature of the cardiac cell subsets Whitney U test with two-tailed p values unless indicated otherwise. described above, we tested the impact of DT in lethally irradiated R-value was obtained from linear regression analysis. Data are presented as C57BL/6 mice reconstituted with Zbtb46-DTR BM. In these mean 6 SD. chimeric mice, DT selectively eliminated heart, spleen, and LN DCs (CD11c+ MHC II+) (Fig. 2A, Supplemental Fig. 1E). All Results cardiac DC subsets were depleted, although some DN cells Characterization of cardiac mouse DCs (CD1032CD11b2) remained after DT treatment (Fig. 2B). Although heart MFs at the steady state have been extensively We also examined the levels of IRF4 and IRF8, as strong ex- studied (9), the nature of the resident DC populations remains to pression of these transcription factors delineates the two main cDC be fully characterized. To better define the cardiac DC subsets in subsets (38). Although IRF4 was more strongly expressed in C57BL/6 mice, single-cell suspensions from pooled healthy hearts CD11b+ than in CD103+ cDCs, we found more IRF8 in CD103+ were isolated and analyzed by flow cytometry (Fig. 1). First, cDCs than in CD11b+ cDCs (Fig. 2C, bottom panels). The granulocytes (Ly6G+ CD11b+), T (CD3+), B (CD19+), and NK CD1032 CD11b2 DN population appeared to be mostly IRF4lo (CD49b+) cells were gated out from the CD45+ leukocyte pop- and bimodal for the expression of IRF8. Interestingly, when ulation, as these have been shown to be intravascular contaminants mice were given Flt3L, cDC1 numbers greatly increased, and not within cardiac tissue during steady state (Fig. 1A) (9). whereas cDC2s were only marginally affected (Fig. 2C, upper MFs were specifically identified as CD64+ CD11b+. To identify panels). However, the IRF8hi subset of DN DCs strongly pDCs, we analyzed the Ly6G2 CD642 CD32 CD192 CD49b2 responded to the growth factor, and their numbers greatly in- MHC IIlo CD11b2 CD11clo subset and gated on PDCA1+ Ly6c+ creased in the heart (Fig. 2C, 2D). Thus, when total DCs were cells, as described (13). Putative cDCs (CD11c+ MHC II+) were analyzed for IRF8 and IRF4 expression, the large increase in the The Journal of Immunology 1787 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 1. Identification of mouse cardiac DCs under steady state. (A) Cardiac single-cell suspensions were pooled from 3–5 hearts of WT C57BL/6 and analyzed by flow cytometry. SSClow and FSClow dead cells and doublets were excluded (data not shown). CD45+ cells were further characterized, and the various subpopulations of cardiac leukocytes are visualized altogether on the dot blot (right) according to the expression of CD11c and MHC II. (B–E)Cardiac CD103+ DCs (red line) and CD11b+ DCs (blue line) were further characterized phenotypically using a series of mAbs and their specific isotype controls (gray filled lines). All markers were surface stained except Clec9a, which was stained following permeabilization. Zbtb46 (Zbtb46+/+ [WT] and Zbtb46gfp/+)and CX3CR1 (CX3CR1+/+ [WT] and CX3CR1gfp/+) reporter mice were used for assessing the expression of these molecules. Cardiac MFswereincludedas positive controls for the assessment of Ly6C, CD64, and MertK expression. (F) The expression of CD172a and XCR1 was analyzed in specific DC subsets. Absolute numbers are shown. (G–J) WT and Flt3/Flk2/CD135-deficient mice were injected i.p. with PBS (2)orwith2mg of Flt3L (+) each day for nine consecutive days. Representative FACS plots from at least three experiments are shown. (G) Percentages of cardiac total DCs (top) and MFs (bottom). (H) Absolute numbers of total cardiac DCs (top) and DC subsets (bottom). (I) Relative numbers of cardiac DC subsets in Flt3-treated mice normalized to the control PBS group. Graphs indicate the mean 6 SD; n =10.(J) Percentages of cardiac total DCs from WT and Flt3 KO mice treated or not with Flt3L. 1788 cDCs IMPAIR RECOVERY AFTER MYOCARDIAL INFARCTION Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 2. Bona fide cDCs in the mouse heart. Lethally irradiated C57BL6/J mice were reconstituted with BM from WT or Zbtb46-DTR mice. (A) Representative FACS plots showing the selective elimination of cardiac total DCs (top) and DCs subsets (bottom) in mice injected twice with DT at a 24 h interval. (B) Percentage (mean 6 SD; n = 6 mice) of total DCs and DC subsets. (C and D) WT mice were injected i.p. with PBS (2) or with 2 mg of Flt3L (+) each day for nine consecutive days. (C) The expression of IRF4 and IRF8 was analyzed in cardiac DC subsets of mice injected with PBS (bottom) or with Flt3L (top). Representative FACS plots from at least three experiments. (D) Left: representative FACS plots of cardiac IRF4+ DCs, IRF8+ DCs, and DN cells. Right: numbers of IRF4+ DCs, IRF8+ DCs, and DN cells normalized to the control PBS group. Graph indicates the mean 6 SD; n = 10. (E) Representative FACS plots from at least three experiments showing the absolute numbers of cells expressing IRF4 and IRF8 in total CD11c+ MHC II+ cardiac DCs from WT and Zbtb-DTR chimeric mice treated with DT. (F and G) Morphology and phagocytic activity of cardiac CD103+ DCs, CD11b+ DCs, MFs, and small resting B cells. Cardiac cells from 10 WT hearts were pooled, incubated in serum-containing RPMI medium with 1-mm fluorescent yellow green microspheres for 1 h, and stained with markers as in Fig. 1A. Cells were sorted by flow cytometry based on cell type–specific markers, counterstained with DAPI (blue), and cytospined to assess morphology by fluorescence microscopy (F) and phagocytic activity by flow (Figure legend continues) The Journal of Immunology 1789 proportions of IRF8hi cells is mostly caused by the response of cells in the different parts of the normal heart were determined, CD11b2 CD1032 DN cells (Fig. 2D). Based on IRF4 and IRF8 normalized by weight, and expressed for each subset as the per- expression, all DC subsets were affected by DT in the Zbtb46- centage of these cells in a given section of the heart. Our flow DTR mice (Fig. 2E). These results suggest that the heart harbors cytometry analysis shows that CD45+ leukocytes, including MFs, CD1032 CD11b2 pre-cDCs, which are mostly committed to the cDC1s, and cDC2s more densely populate the atria and principally IRF8+ DC lineage. the RA (Fig. 3A, 3B, Supplemental Fig. 2C, 2D). Immunofluo- Although DCs share some phenotypic properties with MFs, they rescence microscopy confirmed the strong presence of CD11c+ constitute a morphologically distinct and functionally specialized cells in the RA of CD11c-YFP mice (Fig. 3C, 3D). We also ex- cellular subpopulation. Key characteristics that help distinguish amined whether DCs were present in the tricuspid and mitral these cell types are their morphology and phagocytic capability valves included in Seg1. We surgically separated Seg1, includ- (39). To further confirm the cDC nature of the CD103+CD11b2 ing tricuspid and mitral valves, into Seg1 only or valves and CD1032CD11b+ populations, we bathed cardiac cell sus- (Supplemental Fig. 2E). Although CD45+ leukocytes were re- pensions in serum-containing RPMI medium with 1-mm fluores- duced ∼2-fold in valves compared with Seg1 only (Supplemental cent beads. Then, cells were stained based on the phenotypic Fig. 2F), MFs as well as total DCs and the various subsets appear markers described above, sorted by flow cytometry, and analyzed to be equally distributed between Seg1 and the valves (Fig. 3E, 3F, by fluorescence microscopy. Fig. 2F demonstrates the distinct Supplemental Fig. 2F). Notably, cardiac CD11c+ cells in CD11c- morphology of MFs and their high phagocytic activity as com- YFP mice displayed DC morphology in the valve (Supplemental pared with CD103+ DCs, CD11b+ DCs, and the small resting Fig. 2G). Altogether, these results demonstrate that cardiac cDC

B cells. The specific accumulation of large numbers of phagocy- subsets and MFs show a similar distribution and localize more Downloaded from tosed fluorescent beads in MFs can be readily confirmed by flow densely in the RA. cytometry (Fig. 2G). Increased numbers of cDCs and Tregs in the As activated cDCs are highly migratory, we investigated the heart following MI behavior of cardiac cDCs in mice treated with poly I:C. DC maturation induced by an inflammatory stimulus in nonlymphoid We and others have shown that DC numbers increase under

tissues triggers the migration toward lymphoid organs (40). inflammatory conditions such as aging and atherosclerosis (15, http://www.jimmunol.org/ C57BL/6 mice received poly I:C i.v., and cardiac cDCs were 41). Aging is a risk factor for all cardiovascular diseases, most analyzed by flow cytometry after 24, 48, and 72 h (Fig. 2H). likely related to the development of a panoply of conditions in Whereas neutrophils were increased in the heart by the inflam- the elderly, such as hypertension (42). As we have recently 2/2 matory stimulus, total cDCs and all subsets rapidly decreased in shown in Ldlr mice that atherosclerosis induced by a numbers after poly I:C injection but gradually increased over time western-type diet correlated with an increase in the number of (Fig. 2I). Efficient systemic action of poly I:C was confirmed by DCs in the aorta (15), we hypothesized that this inflammatory CD86 upregulation on spleen DCs (Fig. 2J). condition, as well as aging, could modulate cDC numbers in the Next, we asked whether human cDCs could be identified in the heart. Interestingly, we found a positive correlation between age + + heart of hu-mice (15). Human myeloid DCs are classified on the and the number of cardiac cDCs (CD103 and CD11b ) by guest on September 28, 2021 basis of their expression of BDCA3 or BDCA1, and these sub- (Supplemental Fig. 3A). However, this was not the case for DN 2 2 2/2 sets are related to mouse cDC1s and cDC2s, respectively (10, CD103 CD11b DCs. Then, we investigated in Ldlr mice 39). Irradiated mice reconstituted with human stem cells were the impact of a high fat diet on the number of heart DCs. After + injected i.p. with PBS or with 2 mg of Flt3L each day for nine 10 wk, there was an increase in the number of CD11b DCs + consecutive days. Total DCs (HLA-DR+CD11c+), including both (Supplemental Fig. 3B). Interestingly, CD103 DC and DN 2 2 BDCA1+ and BDCA3+ subsets, were detected in hu-mice heart CD103 CD11b DC numbers remained unchanged in these (Supplemental Fig. 1F), and this pool greatly expanded in re- conditions, suggesting a specific role for cDC2s in inflammation sponse to Flt3L treatment (Supplemental Fig. 1G). IRF8 was of the heart. more expressed in BDCA3+ DCs and, conversely, IRF4 expres- Next, we investigated the impact of MI on immune cells sion was higher in BDCA1+ DCs (Supplemental Fig. 1H). numbers. We performed coronary artery ligation in wild type Both IRF4- and IRF8-expressing cells responded to Flt3L (WT) mice (43), and after 3 d, we analyzed by flow cytometry (Supplemental Fig. 1I). These results demonstrate the capacity the immune cell composition within the myocardium. In re- of human DC subsets, functionally related to those found in sponse to ischemic injury, it is believed that the myocardium mice, to colonize the heart. Altogether, our data provide evi- recruits a large number of leukocytes to protect injured or in- + dence for the presence of bona fide cardiac cDC1s and cDC2s in fected tissue (44). Although the number of heart CD45 in- mammals. creased ∼4-fold, granulocytes and total DCs increased ∼15- and 10-fold, respectively (Fig. 4A, 4B). Of the DC subsets, CD11b+ F Codistribution of M s and cDCs in the heart DCs appeared to be undergoing the most important change in We have refined the anatomical distribution of cardiac MFs and terms of absolute numbers (Fig. 4C). Importantly, DCs display cDCs by surgically separating the heart into segment 1 (Seg1) signs of activation following MI, as judged by the increase in with atria, segment 2 (Seg2), and segment 3 (Seg3). Seg1 with CD40 expression (Fig. 4D). MFswerealsoelevatedintheheart atria was further separated into right atrium (RA), left atrium, and 3 d after infarction (Fig. 4E). Interestingly, as opposed to gran- Seg1 (Supplemental Fig. 2A). Considering the small size of the ulocytes, DCs were reduced in spleen and skin-draining LNs atria (Supplemental Fig. 2B), absolute numbers of hematopoietic (Fig. 4F, 4G).

cytometry (G). (H) WT mice were injected i.v. with 25 mg of poly I:C, and heart cells were stained for granulocytes (top), total DCs (middle), and DC subsets (bottom). Absolute numbers of cells in each gate are shown. Data from three representative experiments were plotted in (I) to show the variations over time. (J) The biological activity of injected poly I:C was confirmed by CD86 upregulation on spleen DCs. 1790 cDCs IMPAIR RECOVERY AFTER MYOCARDIAL INFARCTION Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 3. Anatomical location of DCs under steady-state conditions in mouse heart. (A) Cells from various mouse heart sections (Supplemental Fig. 2A) were analyzed by flow cytometry, and the absolute numbers of total DCs (left panels) and DC subsets (right panels) are shown. (B) Absolute numbers were normalized relative to the weight (Supplemental Fig. 2B) of the different sections, and these numbers were used to compare the percentage (mean 6 SD; n = 7 mice) of a given cell population between heart sections. (C and D) Representative immunofluorescence sections (top, tile scan confocal images) of heart atria (C) and right and left ventricle (D) from CD11c-YFP+/+ mice under steady-state condition stained with anti-GFP Ab (white) and DAPI (gray). All experiments were performed on 12-mm frozen sections. (E) Valves were dissociated from Seg1 (Supplemental Fig. 2E), and cells were analyzed by flow cytometry. Absolute numbers for total DCs (upper panels) and DC subsets (bottom panels) are shown. (F) Percentages of total DCs and DC subsets in each section of the heart. Graphs indicate mean 6 SD; n =7.

The importance of Tregs in MI and tissue recovery after injury has been revealed (15). Our results show that the proportions of has been addressed in human and rodent studies (45). Recently, in T cells and Tregs increase in the heart and spleen 3 d after MI atherosclerosis, an important interplay between Tregs and APCs (Fig. 4H–J, Supplemental Fig. 3C, 3D). The Journal of Immunology 1791 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 4. Cardiac immune cell numbers increase following MI. C57BL/6 mice underwent cardiac surgery to permanently block blood flow down- stream of the left anterior descending coronary artery. Cells from untreated (control) mice or mice 3 d after MI were analyzed by flow cytometry. All relative cell numbers were normalized to the control group, and graphs indicate mean 6 SD; n = 9 mice. (A) Percentages and relative numbers of cardiac CD45+ cells. (B) Absolute numbers of total cardiac DCs (upper left plots) and granulocytes (lower left plots). Relative numbers of cardiac total DCs and granulocytes (right panels). (C) Absolute numbers of cardiac DC subsets (left panels). Relative numbers are shown in the right panels. (D) The expression of CD40 was analyzed in cardiac DC subsets of mice with or without MI. Data from at least three experiments. (E) Absolute numbers and relative numbers of cardiac MFs. (F) Percentages (left panels) and relative numbers (right panels) of splenic total DCs and granulocytes. n = 8 mice. (G) Percentages and relative numbers of total DCs in skin-draining LNs. (H) Gating strategy for total T cells, CD4+ T cells, and Tregs. CD4 and Foxp3 were detected following cell permeabilization. (I) Representative FACS plots and absolute numbers of gated cells (left panels) and relative numbers (right panels) of cardiac total T cells, CD4+ T cells. n = 8 mice. (J) Representative FACS plots and absolute numbers of gated cells (left panels) and relative numbers (right panels) of cardiac Tregs. n = 8 mice. 1792 cDCs IMPAIR RECOVERY AFTER MYOCARDIAL INFARCTION Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 5. Depletion of pDCs does not modulate cardiac functions after MI. (A) Heart pDCs from WT mice (black line) were further characterized phenotypically using a series of mAbs and their specific isotype controls (gray filled lines). pDCs were first gated according to the strategy depicted in Fig. 1A. Zbtb46 (Zbtb46+/+ [WT] and Zbtb46gfp/+) and CX3CR1 (CX3CR1+/+ [WT] and CX3CR1gfp/+) reporter mice were used for assessing the expression of these molecules. Other markers were surface stained except Clec9a, which was stained following permeabilization. Cardiac MFs were included as positive controls for the assessment of MertK expression. (B–D) WT and Flt3/Flk2/CD135-deficient mice were injected i.p. with PBS (2) or with 2 mgof Flt3L (+) each day for nine consecutive days. Bar graphs indicate mean 6 SD; n = 10. Cardiac pDC numbers were determined by flow cytometry. Representative FACS plots (left panels) and relative numbers normalized to the PBS groups (right panels) of cardiac pDCs in heart (B) and in spleen (C). (D) Cardiac pDCs were quantified from WT and Flt3 KO mice. Representative FACS plots from at least three experiments. (E and F) Cells from the various mouse heart sections (Supplemental Fig. 2A, 2E) were analyzed by flow cytometry. (E) pDCs were analyzed in the different heart sections, and absolute numbers are shown. (F) Absolute numbers were normalized relative to the weight of the different sections of the heart. Graphs indicate mean 6 SD; n =6 or 7. (G–I) Hu-mice were injected i.p. with 2 mg of Flt3L each day for nine consecutive days. (G) Left: representative FACS (Figure legend continues) The Journal of Immunology 1793

Depletion of pDCs does not affect muscular contractility injections of DT or PBS, and echocardiography was performed at following MI each time point according to the protocol depicted in Fig. 6A. We have recently shown that pDCs regulate inflammation in the Interestingly, depletion of cDCs by DT led to a significant aorta (15). These cells may play a key role in the heart as well. In improvement of LV fractional shortening after 2 wk (Fig. 6B, 6C, normal mice, cardiac pDCs were identified as PDCA1+ Ly6C+ Supplemental Fig. 4F, Supplemental Videos 1–4). This result CD11b2 cells according to the gating strategy shown in Fig. 1A prompted further characterization of the heart. We found that (46, 47). They express Clec9a, Siglec H, PDCA-1/BST2/CD317, the absence of cDCs reduced infarct size (Fig. 6D). More- and low levels of CX3CR1 (48) but lack or express low levels of over, the area of the border zone cardiomyocytes was signifi- MF, monocyte, and DC markers (Fig. 5A). We previously cantly decreased in mice that received DT, indicating adverse reported that total pDCs in the aorta and spleen were expanded by cardiomyocyte hypertrophy was improved with cDC depletion Flt3L (15). To confirm that the cardiac pDCs were genuine, we (Fig. 6E). examined the impact of Flt3L on their numbers (49). We show that The fact that cDCs impaired cardiac functions and remodeling Flt3L increased pDC numbers in both the heart and spleen of WT after MI suggested a role for these cells in the establishment of a mice but not in Flt3/Flk2/CD135-deficient (Flt3 KO) mice proinflammatory local environment. We analyzed the cardiac he- (Fig. 5B–D). The anatomical distribution of heart pDCs is similar matopoietic cell compartment in the heart of Zbtb46-DTR– to the one described above for cDCs, except that pDCs were found reconstituted mice injected with PBS or DT for 7 d after MI. to be equally concentrated in the two atria, and their relative Interestingly, in the subacute phase after MI (day 7) (Fig. 7A, 7B), proportions are lower in valves than in Seg1 (Fig. 5E, 5F). In the depletion of DCs led to a reduction in the number of total

+ Downloaded from addition, we could identify human pDCs in the heart of hu-mice leukocytes (CD45 cells) and MFs (Fig. 7C). These results show (Fig. 5G–I). These CD123+ BDCA2+ human cardiac pDCs ex- that cDCs are required for maintaining the innate leukocyte in- press IRF8 and were increased by Flt3L injections (Fig. 5G, 5H). filtrate. Then, we assessed if cDCs were required for infiltration of The presence of pDCs in the heart of hu-mice was confirmed by T cells into the myocardium. Notably, DT significantly decreased immunohistochemistry using an anti-LAMP5 (Bad-LAMP) mAb, total T cells as well as Tregs in the myocardium of Zbtb46-DTR which is specific for pDCs (50) (Fig. 5I). chimeric mice 7 d after MI. (Fig. 7D, 7E). Thus, our results in- Then, we characterized the impact of MI on cardiac pDC dicate that the myocardium contains both pDC and cDCs and that http://www.jimmunol.org/ numbers and the importance of these cells in muscular contractility after MI, only cDCs are functionally involved and play a key during recovery. We ligated the left anterior descending coronary pathological role. By specifically depleting cDCs, we demonstrate artery and analyzed pDCs after 3 d. These cells were increased ∼4- that the immune response, as measured by infiltration of MFs, fold in the injured heart but significantly decreased in spleen neutrophils, and multiple T cells subsets is blunted, correlating (Fig. 5J). To assess the importance of pDCs in the pathophysiol- with improved indices of cardiac structure and function. Indeed, ogy of MI, we made use of the BDCA2-DTR mice. We tested the we found that depletion of cDCs correlated with less IL-1b and impact of DT on cardiac pDCs in lethally irradiated C57BL/6 IFN-g proinflammatory cytokine mRNAs in the heart (Fig. 7F–H). mice reconstituted with WT or BDCA2-DTR BM. In the Interestingly, TGF-b mRNA was increased following depletion of by guest on September 28, 2021 BDCA2-DTR chimeric mice, DT treatment resulted in the selec- cDCs, suggesting that this cytokine may exert protective functions tive elimination of heart pDCs (Fig. 5K). Remarkably, mice if inflammation is controlled (51). without pDCs were not affected in left ventricle function as measured by fractional shortening for up to 3 wk after MI. Discussion (Fig. 5L, 5M, Supplemental Fig. 3E). Altogether, the results Cardiovascular diseases represent major global health issues that suggest that in response to ischemic myocardial injury, pDCs are expected to worsen in the coming decades. Although our un- accumulate within the myocardium but did not have a major derstanding of the contribution of MFs to this process has yielded functional role in cardiac function after MI. many important insights, the role of DC subsets has remained elusive (7). In a number of experimental settings, DCs were shown Depletion of cDCs improves cardiac function following MI to have a major impact on adaptive immune responses but also on We next evaluated the effect of Zbtb46+ DCs depletion on the inflammation per se. To shed light on the role of cardiac DCs in outcome of MI. Five weeks after reconstitution, Zbtb46-DTR the modulation of sterile inflammation, we examined the local and chimeric mice underwent MI surgery and DT, or control PBS distal quantitative/qualitative changes occurring in hematopoietic was administered (Supplemental Fig. 4A). We verified the effi- cells in response to MI. These issues are particularly relevant cient depletion of DCs by DT following MI (Supplemental Fig. given the recently demonstrated role of cDCs in priming autore- 4B). Total DCs and all subsets were already reduced 3 d post- active T cells following MI (52). infarction after a single DT injection, whereas MF, neutrophils, In this study, we identified two types of cardiac classical DCs, and monocyte numbers were not affected by DT (Supplemental known as cDC1s (CD103+CD24+XCR1+) and cDC2s (CD11b+ Fig. 4B–E). Mice received two additional weekly (weeks 6 and 7) CD172a+), which are poorly phagocytic as compared with MFs

plots of cardiac pDCs (top) and IRF8+ pDCs (bottom). Percentages are shown in black and cell numbers in red. Right: Expression of IRF8 in pDCs (black) compared with isotype control (gray). (H) Relative number of cardiac pDCs normalized to the PBS group. Graph indicates mean 6 SD; n =7.(I) Paraffin sections of heart from Flt3L-injected hu-mice were stained for LAMP5 (black) and counterstained with hematoxylin. Original magnification 3200. (J) C57BL/6 mice underwent cardiac surgery to permanently block blood flow downstream of the left anterior descending coronary artery. Cells from normal (control) mice or mice 3 d after MI were analyzed by flow cytometry. Representative FACS plots with percentages of gated cells and relative numbers of cardiac (left panels) and splenic (right panels) pDCs. Graphs indicate mean 6 SD; control n =7,MIn =9.(K) Selective elimination of cardiac pDCs from WT and BDCA2-DTR chimeric mice injected with DT twice at a 24 h interval. Percentages (black) and absolute numbers (gray) of pDCs are shown. (L) Schematic representation of the protocol used to assess heart damage following infarction. Five weeks after the transplantation of BDCA2-DTR BM into irradiated WT mice, MI surgery was performed, and mice were injected with PBS or DT for 3 wk. (M) LV systolic function quantified by fractional shortening in the left ventricle of BDCA2-DTR chimeric mice for 3 wk after MI. PBS n =5,DTn =4. 1794 cDCs IMPAIR RECOVERY AFTER MYOCARDIAL INFARCTION Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 6. Depletion of cDCs improves LV remodeling after MI. (A) Schematic representation of the protocol used to assess heart damage following infarction. (B–E) Five weeks after the transplantation of Zbtb46-DTR BM into irradiated WT mice, MI surgery was performed, and mice were injected with PBS or DT for 3 wk. (B) LV fractional shortening was measured by echocardiography 3 wk after MI. Graphs indicate mean 6 SD; n = 10 mice. (C) Representative FACS plots (left panels) and percentage (right panels; graph indicates mean 6 SD; n = 10) of total DCs from spleen of Zbtb46-DTR mice in (B). (D) Histomorphological changes of the heart at 3 wk after MI, as detected by Masson trichrome staining. Graphs indicate mean 6 SD; PBS n = 10, MI n =8.(E) Wheat germ agglutinin staining at 3 wk after MI. Quantitative analyses represent counting of multiple fields from three independent samples per group (∼50 cells per field assessed). Scale bar, 100 mm. Graphs indicate mean 6 SD; PBS n = 23, MI n = 19.

(53). Cardiac CD103+ DCs express Clec9a, DEC205, CD24, and conduction (55). If DCs are involved directly or indirectly in cardiac CD283 as well as IRF8 and Flt3/CD135, which are critical for conduction remains to be addressed. Altogether, our findings reveal their development but lack monocyte-specific markers such as the existence of bona fide cardiac cDCs with phenotypes that comply CD14, CD172a, CX3CR1, F4/80, Ly6C, CD11b, and CD115/M- with the recent criteria proposed for the identification of these cells in CSFR. In contrast, CD11b+ DCs express CX3CR1, F4/80, Ly6C, various tissues and across species (10). and CD14 as well as IRF4 and CD115/M-CSFR. However, they In addition to cDCs, the heart contains a population of pDCs that do not express XCR1, CD24, CD103, Clec9a, DEC205, CD24, increased in the infarcted heart. However, depletion of this DC CD283, and Flt3/CD135. Importantly, both types of cDCs express subset did not affect fractional shortening. The function of cardiac Zbtb46. This transcription factor is not found in pDCs, monocytes, pDCs may lie in their capacity to produce type I IFN and protect the or MFs, allowing for the identification and deletion of cDCs tissue against viral infections. We have recently demonstrated the in vivo (54). Moreover, we were able to detect HLA-DR+CD11c+ role of tolerogenic CCR9+ pDCs in producing IDO and regulating DCs, including both BDCA1+ and BDCA3+ subsets in the heart of Tregs in the aorta during atherosclerosis (15). As shown in this hu-mice. Similar to mouse cardiac DC subsets, human BDCA3+ DCs paper, however, pDC functions do not appear critical in the expressed IRF8, whereas BDCA1+ DCs expressed IRF4. These hu- pathophysiology of cardiac repair after MI. man DCs were also dependent on Flt3/CD115, more so for the IRF8hi As we focused on CD642 DCs, we cannot exclude the possi- cells. The density of cDCs subsets and MFs appears to culminate in bility that monocyte-derived DCs (moDCs) are somehow involved the RA. Such specificity was not observed for monocytes or gran- in cardiac inflammation. A great deal has been learned of these ulocytes. This finding is particularly interesting given the recently cells over the last decade, but their precise identification will re- described role of MFs in normal and aberrant atrioventricular quire a better knowledge of their origin and phenotype (56). The Journal of Immunology 1795 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 7. Depletion of cDCs correlates with reduced numbers of heart MFs and Tregs 7 d after MI. (A) Hearts were isolated from myocardial infarcted Zbtb46-DTR mice that received DT or control PBS for 7 d. (B) Left: selective elimination of cardiac total DCs (top) and DCs subsets (bottom) in Zbtb46- DTR mice injected with DT twice at a 72 h interval. Right: percentage (mean 6 SD; n = 6 mice) of total DCs and cDC subsets. (C) Left: heart cells were analyzed by flow cytometry, and the percentages of CD45+ cells (top) and absolute numbers of MFs (down) are shown. Right: relative numbers (mean 6 SD; n = 10) of cardiac CD45+ cells and MFs normalized to the control PBS group. (D and E) Representative FACS plots and absolute numbers of gated cells (left panels) and relative numbers (right panels; mean 6 SD; PBS n =4,MIn = 5) of cardiac total T cells, CD4+ T cells (D), and Tregs (E) 7 d after MI for mice reconstituted with Zbtb46-DTR BM and treated with PBS or DT. (F–H) mRNA expression levels of RGF-b (F), IL-1b (G), and INF-g (H) from heart of myocardial infarcted Zbtb46-DTR mice treated with DT or control PBS for 7 d measured by quantitative PCR. Relative mRNA expressions were normalized to b-actin. Graphs indicate mean 6 SD; n = 10.

Recent fate-mapping experiments using Zbtb46-DTR mice start expressing Zbtb46 upon differentiation from monocytes, as showed that monocytes generate GM-CSF–dependent moDCs and was seen for BM-derived DCs in vitro (28, 58, 59). If this is the that these monocytes precursors are distal from the DC lineage case, such cells would have been eliminated in experiments based (57). Although Zbtb46 expression appears to be restricted to cDCs on the use of the Zbtb46-DTR mice. Future studies should address (32), it was postulated that moDCs may lose Ly6c expression and the importance of the CD11b+ population heterogeneity. 1796 cDCs IMPAIR RECOVERY AFTER MYOCARDIAL INFARCTION

Besides the DC subsets described above, there was a prominent (72). Although our data showed that Tregs in the heart and spleen CD1032 CD11b2 DN DC population in the heart at steady state. of mice with MI were significantly increased, the depletion of Just like precursor DCs, these cardiac cells are Zbtb46+, and their cDCs correlated with a significant decrease in cardiac Treg number, principally the IRF8hi subset, dramatically increased in numbers. These results demonstrate the importance of cDCs in response to Flt3L. So far, such DN DCs have been described in shaping the immune landscape of the heart following MI. The fact mouse dermis and intestinal lymph, but they remain poorly that depletion of cDCs reduces infiltration of both proin- characterized (20, 60, 61). Our data suggest that the heart harbors flammatory cells and Tregs demonstrates that protection of the pre-cDCs showing lineage-imprinted programs based on their infarcted heart from further injury results from a delicate balance mutually exclusive IRF4 and IRF8 expression (17, 62). between various cell populations under the control of cDCs. This After MI, we found that the myocardium recruits large numbers of implies that interventions aimed at controlling cDCs numbers or leukocytes, including MFs and granulocytes, most likely with the functions has therapeutic potential (73). Indeed, modulating objective of protecting the injured or infected tissue and initiating proinflammatory signals transduction pathways or generating wound healing (44, 63). Notably, CD11b+ cDCs were massively tolerogenic DCs in various mixtures, including cytokines such as increased in heart and mediastinal LN of infarcted mice, whereas IL-37 and TNF-a, were shown to have a beneficial role in post-MI the number of these cells decreased in spleen and skin-draining remodeling (74–76). Future experiments should address more in LNs. We speculate that the cardiac insult resulted in the local depth the spatio-temporal relationship between DCs and other production of GM-CSF by endothelial cells and MFs, allowing the immune cells in health and disease. differentiation of cardiac precursors into cDC2s. Indeed, it was shown in rats that administration of GM-CSF following MI in- Acknowledgments Downloaded from creased DC infiltration and exacerbated LV remodeling (6). Such This work is dedicated to the memory of Dr. Cheolho Cheong. We thank cDC2s may secrete IL-23 to the detriment of IL-10 and lead to the Amanda Kasneci and Kathy Ann Forner for performing cardiac surgeries. polarization of T cells toward Th17 (64). Of note, 3 d after MI, We thank the Institut de Recherches Cliniques de Montre´al animal facility some DCs could be found in the ventricle and septum, whereas after for providing excellent support. 7 d, DCs also appeared in the left atrium (Supplemental Fig. 5).

Recently, the role of DCs in various pathologies, including Disclosures http://www.jimmunol.org/ postinfarction healing, was studied using genetic models such as T.K. is an employee and shareholder of Celldex Therapeutics. The other CD11c-DTR, CD11b-DTR, and LysM–Cre x iDTR (65). Although authors have no financial conflicts of interest. these mice allow the depletion of cells expressing these markers, the specificity toward cDCs has been questioned (25). To over- come these limitations, we used chimeric Zbtb46-DTR mice in References 1. Sanchis-Gomar, F., C. Perez-Quilis, R. Leischik, and A. Lucia. 2016. Epide- which the DTR is only expressed in mouse DCs (28). We con- miology of coronary heart disease and acute coronary syndrome. Ann. Transl. firmed that DT treatment successfully and selectively eliminated Med. 4: 256. cardiac, LN, and splenic CD11c+MHC II+ DCs in chimeric mice 2. Liu, J., H. Wang, and J. Li. 2016. Inflammation and inflammatory cells in myocardial infarction and reperfusion injury: a double-edged sword. Clin. Med. by guest on September 28, 2021 reconstituted with BM from Zbtb46-DTR mice. Importantly, these Insights Cardiol. 10: 79–84. DT-treated Zbtb46-DTR mice showed a dramatic reduction in 3. Nian, M., P. Lee, N. Khaper, and P. Liu. 2004. Inflammatory cytokines and postischemic injury size by preventing ventricular remodeling and postmyocardial infarction remodeling. Circ. Res. 94: 1543–1553. 4. Nahrendorf, M., M. J. Pittet, and F. K. Swirski. 2010. Monocytes: protagonists of improving cardiac function compared with the PBS-treated con- infarct inflammation and repair after myocardial infarction. Circulation 121: trol group. We next checked that any cell populations in the heart 2437–2445. 5. Korf-Klingebiel, M., M. R. Reboll, S. Klede, T. Brod, A. Pich, F. Polten, of WT/Zbtb46-DTR mice with MI-treated DT were changed. L. C. Napp, J. Bauersachs, A. Ganser, E. Brinkmann, et al. 2015. Myeloid- Surprisingly, the selective depletion of heart DCs in Zbtb46-DTR derived growth factor (C19orf10) mediates cardiac repair following myocardi- mice caused a significant decrease of MFs as well as CD45+ cells al infarction. [Published erratum appears in 2016 Nat. Med. 22: 446.] Nat. Med. 21: 140–149. at 7 d but not at 3 d after MI. This suggested that inflammatory 6. Naito, K., T. Anzai, Y. Sugano, Y. Maekawa, T. Kohno, T. Yoshikawa, responses after MI, which trigger accumulation of DCs in injured K. Matsuno, and S. Ogawa. 2008. Differential effects of GM-CSF and G-CSF on tissue, are involved in wound healing but can lead to detrimental infiltration of dendritic cells during early left ventricular remodeling after myocardial infarction. J. Immunol. 181: 5691–5701. effects, such as expansion of infarct size and interstitial fibrosis, 7. Epelman, S., P. P. Liu, and D. L. Mann. 2015. Role of innate and adaptive im- wall thinning, and lower cardiac function. The exact mechanisms mune mechanisms in cardiac injury and repair. Nat. Rev. Immunol. 15: 117–129. 8. Frangogiannis, N. G., and M. L. Entman. 2005. Chemokines in myocardial is- by which cDCs maintain inflammation and negatively impact chemia. Trends Cardiovasc. Med. 15: 163–169. cardiac function will require further investigation. In addition to 9. Epelman, S., K. J. Lavine, A. E. Beaudin, D. K. Sojka, J. A. Carrero, their T cell stimulatory role in draining LNs, cDCs could act lo- B. Calderon, T. Brija, E. L. Gautier, S. Ivanov, A. T. Satpathy, et al. 2014. Embryonic and adult-derived resident cardiac macrophages are maintained cally in the cardiac tissue. For example, cDCs could sustain the through distinct mechanisms at steady state and during inflammation. Immunity immune reaction by secreting some proinflammatory cytokines 40: 91–104. and chemokines in response to damage-associated molecular 10. Guilliams, M., C. A. Dutertre, C. L. Scott, N. McGovern, D. Sichien, S. Chakarov, S. Van Gassen, J. Chen, M. Poidinger, S. De Prijck, et al. 2016. patterns (66). These reactions, potentially of autoimmune nature, Unsupervised high-dimensional analysis aligns dendritic cells across tissues and would contribute to cardiomyocyte death (67). Also, as seen in the species. Immunity 45: 669–684. 11. Steinman, R. M. 2012. Decisions about dendritic cells: past, present, and future. allergic eye, cDC2s with increased aldehyde dehydrogenase ac- Annu. Rev. Immunol. 30: 1–22. tivity could directly activate fibroblasts through the retinoic acid 12. Murphy, T. L., G. E. Grajales-Reyes, X. Wu, R. Tussiwand, C. G. Brisen˜o, receptor and increase fibrosis (68). Secretion of IL-12 by myeloid A. Iwata, N. M. Kretzer, V. Durai, and K. M. Murphy. 2016. Transcriptional control of dendritic cell development. Annu. Rev. Immunol. 34: 93–119. DCs has also been shown to inhibit angiogenesis (69). This could 13. Merad, M., P. Sathe, J. Helft, J. Miller, and A. Mortha. 2013. The dendritic cell be due, in part, to polarization of MFs toward a less angiogenic lineage: ontogeny and function of dendritic cells and their subsets in the steady M1 phenotype (70, 71). state and the inflamed setting. Annu. Rev. Immunol. 31: 563–604. + + + 14. Sage, A. P., D. Murphy, P. Maffia, L. M. Masters, S. R. Sabir, L. L. Baker, The role of CD4 CD25 Foxp3 Tregs in the initial development H. Cambrook, A. J. Finigan, H. Ait-Oufella, G. Grassia, et al. 2014. MHC Class and progression of many cardiovascular diseases is pivotal (45). It II-restricted antigen presentation by plasmacytoid dendritic cells drives proa- therogenic T cell immunity. Circulation 130: 1363–1373. has previously been reported that the depletion of Treg cells be- 15. Yun, T. J., J. S. Lee, K. Machmach, D. Shim, J. Choi, Y. J. Wi, H. S. Jang, fore MI induction resulted in aggravated cardiac inflammation I. H. Jung, K. Kim, W. K. Yoon, et al. 2016. Indoleamine 2,3-dioxygenase- The Journal of Immunology 1797

expressing aortic plasmacytoid dendritic cells protect against atherosclerosis by 41. Liu, P., Y. R. Yu, J. A. Spencer, A. E. Johnson, C. T. Vallanat, A. M. Fong, induction of regulatory T cells. [Published erratum appears in 2016 Cell Metab. C. Patterson, and D. D. Patel. 2008. CX3CR1 deficiency impairs dendritic cell 24: 886.] Cell Metab. 23: 852–866. accumulation in arterial intima and reduces atherosclerotic burden. Arterioscler. 16. Lippens, C., F. V. Duraes, J. Dubrot, D. Brighouse, M. Lacroix, M. Irla, Thromb. Vasc. Biol. 28: 243–250. J. P. Aubry-Lachainaye, W. Reith, J. N. Mandl, and S. Hugues. 2016. IDO- 42. Vigen, R., T. M. Maddox, and L. A. Allen. 2012. Aging of the United States orchestrated crosstalk between pDCs and Tregs inhibits autoimmunity. J. population: impact on heart failure. Curr. Heart Fail. Rep. 9: 369–374. Autoimmun. 75: 39–49. 43. Gao, E., Y. H. Lei, X. Shang, Z. M. Huang, L. Zuo, M. Boucher, Q. Fan, 17. Schlitzer, A., V. Sivakamasundari, J. Chen, H. R. Sumatoh, J. Schreuder, J. Lum, J. K. Chuprun, X. L. Ma, and W. J. Koch. 2010. A novel and efficient model of B. Malleret, S. Zhang, A. Larbi, F. Zolezzi, et al. 2015. Identification of cDC1- coronary artery ligation and myocardial infarction in the mouse. Circ. Res. 107: and cDC2-committed DC progenitors reveals early lineage priming at the 1445–1453. common DC progenitor stage in the bone marrow. Nat. Immunol. 16: 718–728. 44. Luster, A. D., R. Alon, and U. H. von Andrian. 2005. Immune cell migration in 18. Worbs, T., S. I. Hammerschmidt, and R. Fo¨rster. 2017. Dendritic cell migration inflammation: present and future therapeutic targets. Nat. Immunol. 6: 1182– in health and disease. Nat. Rev. Immunol. 17: 30–48. 1190. 19. Hildner, K., B. T. Edelson, W. E. Purtha, M. Diamond, H. Matsushita, 45. Meng, X., J. Yang, M. Dong, K. Zhang, E. Tu, Q. Gao, W. Chen, C. Zhang, and M. Kohyama, B. Calderon, B. U. Schraml, E. R. Unanue, M. S. Diamond, et al. Y. Zhang. 2016. Regulatory T cells in cardiovascular diseases. Nat. Rev. Cardiol. 2008. Batf3 deficiency reveals a critical role for CD8alpha+ dendritic cells in 13: 167–179. cytotoxic T cell immunity. Science 322: 1097–1100. 46. Gautier, E. L., T. Shay, J. Miller, M. Greter, C. Jakubzick, S. Ivanov, J. Helft, 20. Ginhoux, F., K. Liu, J. Helft, M. Bogunovic, M. Greter, D. Hashimoto, J. Price, A. Chow, K. G. Elpek, S. Gordonov, et al; Immunological Genome Consortium. N. Yin, J. Bromberg, S. A. Lira, et al. 2009. The origin and development of 2012. Gene-expression profiles and transcriptional regulatory pathways that nonlymphoid tissue CD103+ DCs. J. Exp. Med. 206: 3115–3130. underlie the identity and diversity of mouse tissue macrophages. Nat. Immunol. 21. Geissmann, F., M. G. Manz, S. Jung, M. H. Sieweke, M. Merad, and K. Ley. 13: 1118–1128. 2010. Development of monocytes, macrophages, and dendritic cells. Science 47. Jakubzick, C., E. L. Gautier, S. L. Gibbings, D. K. Sojka, A. Schlitzer, 327: 656–661. T. E. Johnson, S. Ivanov, Q. Duan, S. Bala, T. Condon, et al. 2013. Minimal 22. Hashimoto, D., J. Miller, and M. Merad. 2011. Dendritic cell and macrophage differentiation of classical monocytes as they survey steady-state tissues and heterogeneity in vivo. Immunity 35: 323–335. transport antigen to lymph nodes. Immunity 39: 599–610. 23. Schlitzer, A., and F. Ginhoux. 2014. Organization of the mouse and human DC 48. Swiecki, M., and M. Colonna. 2015. The multifaceted biology of plasmacytoid Downloaded from network. Curr. Opin. Immunol. 26: 90–99. dendritic cells. Nat. Rev. Immunol. 15: 471–485. 24. Liu, K., G. D. Victora, T. A. Schwickert, P. Guermonprez, M. M. Meredith, 49. Bjo¨rck, P. 2001. Isolation and characterization of plasmacytoid dendritic cells K. Yao, F. F. Chu, G. J. Randolph, A. Y. Rudensky, and M. Nussenzweig. 2009. from Flt3 ligand and granulocyte-macrophage colony-stimulating factor-treated In vivo analysis of dendritic cell development and homeostasis. Science 324: mice. Blood 98: 3520–3526. 392–397. 50. Defays, A., A. David, A. de Gassart, F. De Angelis Rigotti, T. Wenger, 25. Choi, J. H., C. Cheong, D. B. Dandamudi, C. G. Park, A. Rodriguez, V. Camossetto, P. Brousset, T. Petrella, M. Dalod, E. Gatti, and P. Pierre. 2011. S. Mehandru, K. Velinzon, I. H. Jung, J. Y. Yoo, G. T. Oh, and R. M. Steinman. BAD-LAMP is a novel biomarker of nonactivated human plasmacytoid dendritic

2011. Flt3 signaling-dependent dendritic cells protect against atherosclerosis. cells. Blood 118: 609–617. http://www.jimmunol.org/ Immunity 35: 819–831. 51. Euler, G. 2015. Good and bad sides of TGFb-signaling in myocardial infarction. 26. Swiecki, M., S. Gilfillan, W. Vermi, Y. Wang, and M. Colonna. 2010. Plasma- Front. Physiol. 6: 66. cytoid dendritic cell ablation impacts early interferon responses and antiviral NK 52. Van der Borght, K., C. L. Scott, V. Nindl, A. Bouche´, L. Martens, D. Sichien, and CD8(+) T cell accrual. Immunity 33: 955–966. J. Van Moorleghem, M. Vanheerswynghels, S. De Prijck, Y. Saeys, et al. 2017. 27. Meredith, M. M., K. Liu, G. Darrasse-Jeze, A. O. Kamphorst, H. A. Schreiber, Myocardial infarction primes autoreactive T cells through activation of dendritic P. Guermonprez, J. Idoyaga, C. Cheong, K. H. Yao, R. E. Niec, and cells. Cell Rep. 18: 3005–3017. M. C. Nussenzweig. 2012. Expression of the zinc finger transcription factor zDC 53. Steinman, R. M., and Z. A. Cohn. 1974. Identification of a novel cell type in (Zbtb46, Btbd4) defines the classical dendritic cell lineage. J. Exp. Med. 209: peripheral lymphoid organs of mice. II. Functional properties in vitro. J. Exp. 1153–1165. Med. 139: 380–397. 28. Satpathy, A. T., W. Kc, J. C. Albring, B. T. Edelson, N. M. Kretzer, 54. Cybulsky, M. I., C. Cheong, and C. S. Robbins. 2016. Macrophages and den- D. Bhattacharya, T. L. Murphy, and K. M. Murphy. 2012. Zbtb46 expression dritic cells: partners in atherogenesis. Circ. Res. 118: 637–652.

distinguishes classical dendritic cells and their committed progenitors from other 55. Hulsmans, M., S. Clauss, L. Xiao, A. D. Aguirre, K. R. King, A. Hanley, by guest on September 28, 2021 immune lineages. J. Exp. Med. 209: 1135–1152. W. J. Hucker, E. M. Wulfers, G. Seemann, G. Courties, et al. 2017. Macrophages 29. Ishibashi, S., M. S. Brown, J. L. Goldstein, R. D. Gerard, R. E. Hammer, and facilitate electrical conduction in the heart. Cell. 169: 510–522.e20. J. Herz. 1993. Hypercholesterolemia in low density lipoprotein receptor 56. Mildner, A., and S. Jung. 2014. Development and function of dendritic cell knockout mice and its reversal by adenovirus-mediated gene delivery. J. Clin. subsets. Immunity 40: 642–656. Invest. 92: 883–893. 57. Menezes, S., D. Melandri, G. Anselmi, T. Perchet, J. Loschko, J. Dubrot, 30. Matthews, W., C. T. Jordan, G. W. Wiegand, D. Pardoll, and I. R. Lemischka. R. Patel, E. L. Gautier, S. Hugues, M. P. Longhi, et al. 2016. The heterogeneity 1991. A receptor tyrosine kinase specific to hematopoietic stem and progenitor of Ly6Chi monocytes controls their differentiation into iNOS+ macrophages or cell-enriched populations. Cell 65: 1143–1152. monocyte-derived dendritic cells. Immunity 45: 1205–1218. 31. Cheong, C., and J. H. Choi. 2012. Dendritic cells and regulatory T cells in 58. Brisen˜o, C. G., M. Haldar, N. M. Kretzer, X. Wu, D. J. Theisen, W. Kc, V. Durai, atherosclerosis. Mol. Cells 34: 341–347. G. E. Grajales-Reyes, A. Iwata, P. Bagadia, et al. 2016. Distinct transcriptional 32. Meredith, M. M., K. Liu, A. O. Kamphorst, J. Idoyaga, A. Yamane, programs control cross-priming in classical and monocyte-derived dendritic P. Guermonprez, S. Rihn, K. H. Yao, I. T. Silva, T. Y. Oliveira, et al. 2012. Zinc cells. Cell Rep. 15: 2462–2474. finger transcription factor zDC is a negative regulator required to prevent activation 59. Wu, X., C. G. Brisen˜o, V. Durai, J. C. Albring, M. Haldar, P. Bagadia, of classical dendritic cells in the steady state. J. Exp. Med. 209: 1583–1593. K. W. Kim, G. J. Randolph, T. L. Murphy, and K. M. Murphy. 2016. Mafb 33. Patel, B. B., A. Kasneci, A. M. Bolt, V. Di Lalla, M. R. Di Iorio, M. Raad, lineage tracing to distinguish macrophages from other immune lineages reveals K. K. Mann, and L. E. Chalifour. 2015. Chronic exposure to bisphenol a reduces dual identity of Langerhans cells. J. Exp. Med. 213: 2553–2565. successful cardiac remodeling after an experimental myocardial infarction in 60. Shklovskaya, E., B. Roediger, and B. Fazekas de St Groth. 2008. Epidermal and male C57bl/6n mice. Toxicol. Sci. 146: 101–115. dermal dendritic cells display differential activation and migratory behavior 34. Thibault, H., L. Gomez, E. Donal, G. Pontier, M. Scherrer-Crosbie, M. Ovize, while sharing the ability to stimulate CD4+ T cell proliferation in vivo. J. and G. Derumeaux. 2007. Acute myocardial infarction in mice: assessment of Immunol. 181: 418–430. transmurality by strain rate imaging. Am. J. Physiol. Heart Circ. Physiol. 293: 61. Cerovic, V., S. A. Houston, C. L. Scott, A. Aumeunier, U. Yrlid, A. M. Mowat, H496–H502. and S. W. Milling. 2013. Intestinal CD103(-) dendritic cells migrate in lymph 35. Tsapogas, P., C. J. Mooney, G. Brown, and A. Rolink. 2017. The cytokine Flt3- and prime effector T cells. Mucosal Immunol. 6: 104–113. ligand in normal and malignant hematopoiesis. Int. J. Mol. Sci. 18: 6. 62. Grajales-Reyes, G. E., A. Iwata, J. Albring, X. Wu, R. Tussiwand, W. Kc, 36. Maraskovsky, E., K. Brasel, M. Teepe, E. R. Roux, S. D. Lyman, K. Shortman, N. M. Kretzer, C. G. Brisen˜o, V. Durai, P. Bagadia, et al. 2015. Batf3 maintains and H. J. McKenna. 1996. Dramatic increase in the numbers of functionally autoactivation of Irf8 for commitment of a CD8a(+) conventional DC clono- mature dendritic cells in Flt3 ligand-treated mice: multiple dendritic cell sub- genic progenitor. Nat. Immunol. 16: 708–717. populations identified. J. Exp. Med. 184: 1953–1962. 63. Swirski, F. K., M. Nahrendorf, M. Etzrodt, M. Wildgruber, V. Cortez-Retamozo, 37. Waskow, C., K. Liu, G. Darrasse-Je`ze, P. Guermonprez, F. Ginhoux, M. Merad, P. Panizzi, J. L. Figueiredo, R. H. Kohler, A. Chudnovskiy, P. Waterman, et al. T. Shengelia, K. Yao, and M. Nussenzweig. 2008. The receptor tyrosine kinase 2009. Identification of splenic reservoir monocytes and their deployment to in- Flt3 is required for dendritic cell development in peripheral lymphoid tissues. flammatory sites. Science 325: 612–616. Nat. Immunol. 9: 676–683. 64. Schlitzer, A., N. McGovern, P. Teo, T. Zelante, K. Atarashi, D. Low, A. W. Ho, 38. Belz, G. T., and S. L. Nutt. 2012. Transcriptional programming of the dendritic P. See, A. Shin, P. S. Wasan, et al. 2013. IRF4 transcription factor-dependent cell network. [Published erratum appears in 2013 Nat. Rev. Immunol. 13: 149.] CD11b+ dendritic cells in human and mouse control mucosal IL-17 cytokine Nat. Rev. Immunol. 12: 101–113. responses. Immunity 38: 970–983. 39. Guilliams, M., F. Ginhoux, C. Jakubzick, S. H. Naik, N. Onai, B. U. Schraml, 65. Anzai, A., T. Anzai, S. Nagai, Y. Maekawa, K. Naito, H. Kaneko, Y. Sugano, E. Segura, R. Tussiwand, and S. Yona. 2014. Dendritic cells, monocytes and T. Takahashi, H. Abe, S. Mochizuki, et al. 2012. Regulatory role of dendritic macrophages: a unified nomenclature based on ontogeny. Nat. Rev. Immunol. 14: cells in postinfarction healing and left ventricular remodeling. Circulation 125: 571–578. 1234–1245. 40. Banchereau, J., and R. M. Steinman. 1998. Dendritic cells and the control of 66. Piccinini, A. M., and K. S. Midwood. 2010. DAMPening inflammation by immunity. Nature 392: 245–252. modulating TLR signalling. Mediators Inflamm. 2010: 672395. 1798 cDCs IMPAIR RECOVERY AFTER MYOCARDIAL INFARCTION

67. Eriksson, U., R. Ricci, L. Hunziker, M. O. Kurrer, G. Y. Oudit, T. H. Watts, after myocardial infarction by modulating monocyte/macrophage differentiation. I. Sonderegger, K. Bachmaier, M. Kopf, and J. M. Penninger. 2003. Dendritic Circ. Res. 115: 55–67. cell-induced autoimmune heart failure requires cooperation between adaptive 73. Takahashi, K., S. Fukushima, K. Yamahara, K. Yashiro, Y. Shintani, and innate immunity. [Published erratum appears in 2004 Nat. Med. 10: 105.] S. R. Coppen, H. K. Salem, S. W. Brouilette, M. H. Yacoub, and K. Suzuki. Nat. Med. 9: 1484–1490. 2008. Modulated inflammation by injection of high-mobility group box 1 re- 68. Ahadome, S. D., R. Mathew, N. J. Reyes, P. S. Mettu, S. W. Cousins, V. L. Calder, covers post-infarction chronically failing heart. Circulation 118(14, Suppl.): and D. R. Saban. 2016. Classical dendritic cells mediate fibrosis directly via the S106–S114. retinoic acid pathway in severe eye allergy. JCI Insight 1: e87012. 74. Zhu, R., H. Sun, K. Yu, Y. Zhong, H. Shi, Y. Wei, X. Su, W. Xu, Q. Luo, 69. Curiel, T. J., P. Cheng, P. Mottram, X. Alvarez, L. Moons, M. Evdemon-Hogan, F. Zhang, et al. 2016. Interleukin-37 and dendritic cells treated with Interleukin- S. Wei, L. Zou, I. Kryczek, G. Hoyle, et al. 2004. Dendritic cell subsets dif- 37 plus Troponin I ameliorate cardiac remodeling after myocardial infarction. J. ferentially regulate angiogenesis in human ovarian cancer. Cancer Res. 64: Am. Heart Assoc. 5: e004406. 5535–5538. 75. Choo, E. H., J. H. Lee, E. H. Park, H. E. Park, N. C. Jung, T. H. Kim, Y. S. Koh, 70. Jetten, N., S. Verbruggen, M. J. Gijbels, M. J. Post, M. P. De Winther, and E. Kim, K. B. Seung, C. Park, et al. 2017. Infarcted myocardium-primed den- M. M. Donners. 2014. Anti-inflammatory M2, but not pro-inflammatory M1 dritic cells improve remodeling and cardiac function after myocardial infarction macrophages promote angiogenesis in vivo. Angiogenesis 17: 109–118. by modulating the regulatory T cell and macrophage polarization. Circulation 71. Yu, X. L., B. T. Wu, T. T. Ma, Y. Lin, F. Cheong, H. Y. Xiong, C. L. Xie, 135: 1444–1457. C. Y. Liu, Q. Wang, Z. W. Li, and Z. G. Tu. 2016. Overexpression of IL-12 76. Maekawa, Y., N. Mizue, A. Chan, Y. Shi, Y. Liu, S. Dawood, M. Chen, reverses the phenotype and function of M2 macrophages to M1 macrophages. F. Dawood, G. de Couto, G. H. Li, et al. 2009. Survival and cardiac remodeling Int. J. Clin. Exp. Pathol. 9: 8963–8972. after myocardial infarction are critically dependent on the host innate immune 72. Weirather, J., U. D. Hofmann, N. Beyersdorf, G. C. Ramos, B. Vogel, A. Frey, interleukin-1 receptor-associated kinase-4 signaling: a regulator of bone marrow- G. Ertl, T. Kerkau, and S. Frantz. 2014. Foxp3+ CD4+ T cells improve healing derived dendritic cells. Circulation 120: 1401–1414. Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021