Existence of Conventional Dendritic Cells in Gallus gallus Revealed by Comparative Expression Profiling

This information is current as Thien-Phong Vu Manh, Hélène Marty, Pierre Sibille, Yves of September 24, 2021. Le Vern, Bernd Kaspers, Marc Dalod, Isabelle Schwartz-Cornil and Pascale Quéré J Immunol 2014; 192:4510-4517; Prepublished online 16 April 2014;

doi: 10.4049/jimmunol.1303405 Downloaded from http://www.jimmunol.org/content/192/10/4510

Supplementary http://www.jimmunol.org/content/suppl/2014/04/16/jimmunol.130340 Material 5.DCSupplemental http://www.jimmunol.org/ References This article cites 34 articles, 13 of which you can access for free at: http://www.jimmunol.org/content/192/10/4510.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 24, 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 © 2014 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Existence of Conventional Dendritic Cells in Gallus gallus Revealed by Comparative Profiling

Thien-Phong Vu Manh,*,†,‡,1 He´le`ne Marty,†,x,{,1 Pierre Sibille,‖ Yves Le Vern,x,{ Bernd Kaspers,# Marc Dalod,*,†,‡,2 Isabelle Schwartz-Cornil,‖,2 and Pascale Que´re´x,{,2

The existence of conventional dendritic cells (cDCs) has not yet been demonstrated outside mammals. In this article, we identified bona fide cDCs in chicken spleen. Comparative profiling of global and of immune response gene expression, morphology, and T cell activation properties show that cDCs and macrophages (MPs) exist as distinct mononuclear phagocytes in the chicken, resembling their human and mouse cell counterparts. With computational analysis, core gene expression signatures for cDCs, MPs, and T and B cells across the chicken, human, and mouse were established, which will facilitate the identification of these subsets in other vertebrates. Overall, this study, by extending the newly uncovered cDC and MP paradigm to the chicken, suggests that these two phagocyte lineages were already in place in the common ancestor of reptiles (including birds) and mammals in evolution. Downloaded from It opens avenues for the design of new vaccines and nutraceuticals that are mandatory for the sustained supply of poultry products in the expanding human population. The Journal of Immunology, 2014, 192: 4510–4517.

onventional dendritic cells (cDCs) are functionally char- cDCs derive from a clonogenic common DC progenitor in the acterized by their exquisite capacities to present Ags bone marrow and are ontogenetically distinct from other mono- C to naive T cells, having a key role in maintenance of nuclear phagocytes, such as macrophages (MPs), monocyte -de- http://www.jimmunol.org/ tolerance and induction of immune effectors against invading rived dendritic cells and yolk sac– and fetal liver–derived skin pathogens (1). cDCs constitute a unique immune cell lineage, as Langerhans cells (LCs) (1, 5–7). Compared with other mononu- recently revealed in mice by genetic cell fate mapping (2) and the clear phagocytes, cDCs are endowed with a much higher efficacy precursor–progeny relationship at the single-cell level (3) and in in patrolling virtually all peripheral tissues and in migrating to humans by comparative gene expression profiling (4). Mouse lymph nodes both at steady state and upon stimulation. Further- more, cDCs and the other types of mononuclear phagocytes are differentially involved in setting and tuning specific arms of im- *Centre d’Immunologie de Marseille-Luminy, Aix-Marseille University, UM2, 13288 Marseille Cedex 9, France; †INSERM, U1104, 13288 Marseille, France; munity, depending on tissue location, inflammatory response, and ‡Centre National de la Recherche Scientifique, Unite´ Mixte de Recherche 7280, + by guest on September 24, 2021 x pathogens (1). Finally, cDCs include two subsets, the CD8a and 13288 Marseille, France; Institut National de la Recherche Agronomique, Unite´ CD11b+ types in the mouse and their homologous BDCA3+ Mixte de Recherche 1282 Infectiologie et Sante´ Publique, 37380 Nouzilly, France; + {Universite´ Franc¸ois Rabelais de Tours, Unite´ Mixte de Recherche 1282, 37000 and BDCA1 types in humans, respectively, that display distinct ‖ Tours, France; Institut National de la Recherche Agronomique, Unite´ de Recherche gene expression programs, surface phenotypes, and functional 892 Virologie et Immunologie Mole´culaires, Domaine de Vilvert, 78352 Jouy-en- + Josas Cedex, France; and #Department of Veterinary Sciences, University of Munich, specialization (4, 8, 9). In particular, the mouse CD8a type, and, + 80539 Munich, Germany to some extent, the human BDCA3 type cDCs, are specialized in + 1T.-P.V.M. and H.M. contributed equally to this work. cross-presentation of endocytosed Ags to CD8 T cells (8, 10–15). 2M.D., I.S.-C., and P.Q. are senior authors. The cDCs have been described only in mammals. Nonmamma- Received for publication December 23, 2013. Accepted for publication March 14, lian vertebrates, such as fishes and birds, have the same basic 2014. principles of adaptive immunity as mammals, but they present This work was supported by institutional funding from the Agence Nationale de la many evolutionary particularities, with different repertoires of Recherche PhyloGenDC, ANR-09-BLAN-0073-02; the PhyloGenDC ANR grant and , cells, and lymphoid organs, including absence of lymph the European Research Council (T.-P.V.M.); and European Research Council Frame Program 2007-2013 Grant Agreement 281225 for the SystemsDendritic project nodes (16). In these species, cells with dendritic morphology that (M.D.). seem to be distinct from MPs have been described (16–18). In I.S.-C. orchestrated the research program; M.D., I.S.-C., and P.Q. designed experi- chicken in vitro, DC-like cells have been obtained upon differ- ments and directed research; I.S.-C. wrote the paper with input from M.D., T.-P.V.M., entiation of bone marrow cells in GM-CSF and IL-4 cultures (19). H.M., P.S., and P.Q.; H.M., P.S., Y.L.V., I.S.-C., and P.Q. performed experiments and analyzed data; T.-P.V.M. and M.D. carried out bioinformatics analyses; and B.K. In vivo, LCs have been identified in skin (20); in spleen and bursa, + provided reagents and array annotations. which is the primary B cell lymphoid organ in birds, CD83 cells + The microarray data presented in this article have been submitted to the Gene Ex- and CD205 cells have been described, the later corresponding to pression Omnibus (http://www.ncbi.nlm.nih.gov/geo/info/) under accession number both putative DCs and MPs (21, 22). A subset of lung phagocytes GSE55642. presenting higher endosomal pH compared with resident MPs has Address correspondence and reprint requests to Dr. Isabelle Schwartz-Cornil, Institut also been identified in chicken (23). Although birds seem equipped National de la Recherche Agronomique, Unite´ de Recherche 892 Virologie et Im- munologie Mole´culaires, Domaine de Vilvert, 78352 Jouy-en-Josas, France. E-mail with LCs and can develop DC-type cells from bone marrow ex vivo, address: [email protected] it is not known whether they dispose of the cDC lineage homolo- The online version of this article contains supplemental material. gous to that of mammals (23). In addition, whether the biological Abbreviations used in this article: cDC, conventional dendritic cell; GSEA, gene set differences between cDCs and MPs recently unraveled in mammals enrichment analysis; LC, Langerhans cell; MGG, May–Grunwald–Giemsa€ (stain); pertain to distant vertebrate species such as birds is not known. MP, macrophage; pDC, plasmacytoid DC; qPCR, qualitative PCR. Gene expression profiling has proved to be a very efficient an- Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 alytical approach to delineate homologies between mononuclear www.jimmunol.org/cgi/doi/10.4049/jimmunol.1303405 The Journal of Immunology 4511 phagocyte subsets of mice, sheep, and humans (4, 24, 25). In this Bioinformatics analyses article, a comparative gene expression analysis between chicken The bioinformatics analyses were adapted from previous reports (4, 24). (Gallus gallus), human and mouse immune cell subsets shows the Raw gene expression data were background corrected using the “normexp” existence in chicken of MPs and of bona fide cDCs that are ca- method and quantile normalized with the limma package through Bio- pable of T cell activation. These data suggest that the establish- conductor in the R statistical environment (version 2.15.0). Independent ment of the cDC lineage occurred before the common bird/mammal biological triplicates (B cells) and quadruplicates (cDCs, MPs, and T cells) were performed using cells isolated from four different animals. Quality ancestor, and is likely to apply to amniotes in general. control of the expression data was assessed by boxplots of raw expression data, density plots of normalized data, scatter plots, and calculation of the Materials and Methods Pearson correlation coefficients between arrays, using the Ringo package. Animals The microarray data have been assigned the Gene Expression Omnibus number GSE55642 (www.ncbi.nlm.nih.gov/geo/info/), and they are pub- Highly inbred white Leghorn chickens that originate from the GB1 Athens licly available. line and are homozygous for the B13 histocompatibility B complex were To statistically test whether mouse and human cell transcriptional fin- used for the cDCs, MPs, and T and B cell characterization. The PA12 gerprints were enriched in specific chicken cell types, we performed outbred line was used to isolate CD4+ T cells in the MLR experiments; the pairwise comparisons between cell types, using the gene set enrichment PA12 chickens present mainly the B21 MHC haplotype (97%), and a small analysis (GSEA) method from the Massachusetts Institute of Technology proportion of them present the B19 MHC haplotype (3%) (B. Bed’hom, (www.broad.mit.edu/gsea). personal). Chickens were 6.5–8.5 wk of age and were raised under specific The mouse and human cell-specific transcriptional fingerprints (Supplemental pathogen–free conditions (Plate-Forme d’Infectiologie Expe´rimentale– Dataset 2) correspond to genes that were found more highly expressed Institut National de la Recherche Agronomique, Nouzilly, France). In (.1.5 fold) in the cell population of interest compared with all other cell compliance with French law, the chickens were euthanized according to populations considered in the preselected compendium of arrays [see list Downloaded from protocols approved by the Animal Ethics Committee (Re´gion Centre, of public database array IDs taken into account in the human and the France). mouse compendia (Supplemental Dataset 1)]. The cut-off of 1.5-fold has been applied using the stringent “min/max” procedure calculating fold Spleen cell subset isolation change as (minimum expression among all replicates for all cell types selected/maximum expression among all other replicates for all other cell Spleen from four different exsanguinated chickens, indifferently male or types), as previously published (29). To perform GSEA on chicken ex- female, were used for cell isolation. After an incubation of 30 min at 37˚C in pression data, using fingerprints composed of human or mouse genes, we

RPMI 1640 plus collagenase D (400 U/ml; Roche Applied Science) and http://www.jimmunol.org/ used annotations provided by the Sigenae pipeline on the orthologs in human m DNase (15 g/ml; Sigma-Aldrich), spleen pieces were crushed on a steel or mouse of the chicken genes. sieve. Isolated splenocytes were washed and centrifuged on a 1.077 density Histopaque gradient (Sigma-Aldrich) to remove the nucleated erythro- Regulation of immunity-related gene expression in homologous subsets cytes. For B and T cell sorting, splenocytes were labeled with 2 mg/ml PE- across human, mouse, and chicken species. We established a list of 248 conjugated anti–Bu-1 AV20 mAb (26) and PE-conjugated anti-chicken immunity-related genes (sensors, cytokine, cytokine receptors, signaling CD3 CT-3 mAb, respectively (Southern Biotech) (27). For MP sorting, receptors) from which 116 genes 1) were present on the chicken array, 2) splenocytes were labeled with 2 mg/ml PE-conjugated KUL01 mAb (28) gave signals above background in at least one subset, and 3) were differ- and FITC-conjugated anti–MHC-II mAb (Southern Biotech), and they entially expressed ($2-fold) across all cell subsets (cDC, MP, and T and were sorted as KUL01+ MHC-II+ cells (Southern Biotech). For cDC can- B cells). Of these 116 chicken genes, 94 were found to have an orthologous didate cell sorting, splenocytes were first stained with the cell supernatant gene on both human and mouse gene chips (except CLEC2B, for which an of the 8F2 hybridoma directed to the putative chicken CD11c (19), fol- orthologous gene was present only on the human chip) as well as having by guest on September 24, 2021 lowed by 50 mg/ml Alexa 647–conjugated goat anti-mouse Fab (Jackson a differential expression of $2-fold across all murine and human cell ImmunoResearch), and subsequently reacted with a mixture of FITC- subsets. Whenever several probes were found for a given gene, the most conjugated mAb anti-chicken MHC-II, PE-conjugated KUL01, anti Bu-1, highly expressed probe across all cell subsets in each species was selected. and anti-CD3 mAbs. The MoFlo Legacy Cell Sorter (Beckman Coulter) Gene expression values of the three species were then cross normalized: was used for the sorting. The labeled cells were analyzed with the FlowJo Each of the three batches of arrays, corresponding to cell subsets from the version 10 software (TreeStar). three species, were scaled by setting the mean expression value to 0 and the variance to 1. RNA extraction and hybridization on microarrays Real-time PCR Total RNA from subsets was extracted using the PicoPure RNA Isolation Kit (Arcturus Life Technologies) and checked for quality with an Agilent For relative quantitation of gene expression in cellular subsets, RNA was 2100 Bioanalyzer using RNA 6000 Nano or Pico Kits (Agilent Technologies). reverse transcribed using random primers and SuperScript III Reverse All RNA samples had an RNA integrity number .9. Hybridizations were Transcriptase (Invitrogen). Real-time qualitative PCR (qPCR) was carried performed at the Platform Biopuces et Se´quenc¸age (Institut de Ge´ne´tique out with 10 mM primers in a final reaction volume of 15 ml of 2 iQ SYBR et de Biologie Mole´culaire et Cellulaire, 67400 Illkirch, France, www. Green Supermix (Bio-Rad). The primers used to amplify the chicken cDNA igbmc.fr/grandesstructures/cbi/). When insufficient total RNA amounts were designed with Primer-BLAST (National Center for Biotechnology for hydridization were obtained (,50 ng), the RNA from the sorted subsets Information), using publicly available GenBank sequences. The primers of distinct spleen pools was mixed. RNA labeling was performed using the are as follows: FLT3 forward: 59-CATTCGGACCCAGTACATGTTTAC-39, one-color Low Input Quick Amp Labeling Kit (Agilent Technologies), and reverse: 59-TGAGCCGTAGAAGAGCAGGTATAA-39; ZBTB46 forward: following the manufacturer’s recommendations. Each RNA sample (50 ng) 59-CAGGAACGTCATCGAGGTGAT-39, and reverse: 59-GCCTGGACG- was amplified and cyanin 3 labeled, and subsequently the cRNA was checked ATATCCGTCAT-39; BATF3 forward: 59-GGAGAGAGAAGAACCGAG- for quality on Nanodrop and the Agilent 2100 Bioanalyzer. Subsequently, TTGCT-39, and reverse: 59-CGTGAAGTTTGTCCGCTTTC-39;MAFB the cRNA (600 ng) was fragmented and used for hybridization on custom- forward: 59-AGGACCGGTTCTCGGATGAC-39, and reverse: 59-CCTCGG- designed Agilent chicken arrays. Our custom-designed array is based on AGGTGCCTGTTG-39;CD14forward:59-CATGCTTGGCAGTCTGCAAA- and includes the entire commercial chicken Agilent technologies array 39,andreverse:59-CAGGAGGACCTCAGGAACCA-39; CADM1 forward: [Gallus gallus oligo microarray (V2), 42 K]. About 1100 new probes were 59-GGAGCGTGGACCATGCA-39, and reverse: 59-CAAAGCATGGCAAA- designed with the eArray software from Agilent Technologies. The new TACAACCA-39; XCR1 forward: 59-CCTTCGGGTGGATTTTTGGT-39,and probes include gene probes derived from a list of candidates known to be reverse: 59-CGCTGTAGTAGCCAATGGAGAA-39; BCL11B forward: 59- selectively expressed in both mouse and human DC population subsets GCGATGCCAGTGTAAATGCC-39, and reverse: 59-CACATGGTCAGC- (4). When direct one-to-one orthologs could be identified between humans CTGTGTGA-39; PLCG1 forward: 59-CACTGCTTCGTGGTCCTCTA-39, and chickens (with genomicus V67.01, www.genomicus.biologie.ens.fr/), and reverse: 59-CATCCTCGGACGTGGCTTG-39; BEND5 forward: 59- the new probes were directly included in the chip. Otherwise, the hu- TCTCGAGAATGGAAAGAATGTTTGT-39, and reverse: 59-TGACTTT- man and murine candidates were blasted on the Gallus gallus transcript GGAGAGTCTCTGTGC-39; C5AR1 forward: 59-AACCTTCAACGA- database to select for potential chicken candidates that were used for CATCGGCG-39, and reverse: 59-AAGGAAAATGGCGGCGTAGA-39;CD5 the design of new probes. When no orthologous candidates were found, forward 59-GCGTTACCTCCACCTCTGTC-39, and reverse: 59-TTCGACT- the human and mouse probes were directly included in the chip (60 K TCTGGCAGACCAC-39; CD79B forward: 59-TACGTGTGCGACAGCAA- array). GAA-39, and reverse: 59-TTGCTGCGACTCATCACTCT-39; CIITA forward: 4512 cDC IDENTIFICATION IN CHICKEN

59-TGCCTGAGAAGTGGTACAGC-39, and reverse: 59-TTGCTTCCCT- MHC-II molecules. We used the well-characterized mAbs directed CCTGAGGTCT-39; CXCR5 forward: 59-CCCTGACCAAGCTAGAAG- to the CD3 and Bu-1 Ags to exclude T and B cells, respectively, CC-39, and reverse: 59-CCAATGGCCTCTGTCACCAT-39; FAM46C for- and the KUL01 to exclude MPs. After electronic exclusion gating ward: 59-AGCTGACCGTTTCTTTGGAGA-39, and reverse: 59-AGTTC- AGCACACTGCATGAC-39; INPP4B forward: 59-CACGGTCCATACC- of cells positive for CD3, Bu-1, or KUL01, we sorted the re- ACTGTCC-39, and reverse: 59-GCCCAGGAAGCTTCTTCCAA-39; KLHL14 maining cells in four populations based on their staining pattern forward: 59-CTGGCTGTGCAGTCCTTGAT-39, and reverse: 59-GATTT- with an anti–MHC-II mAb and the 8F2 mAb thought to target GTAGGCCCCCATGCT-39; PLEKHA5 forward: 59-CTGTGGGCAGAA- chicken CD11c, which was previously shown to label bone mar- CATCACGA-39, and reverse: 59-ACGTCTGACAACTGGTGCAT-39; TCF7 forward: 59-CCGACCTCAAGTCCTCGTTG-39, and reverse: 59-TGGTG- row–derived DCs (Fig. 1A). We sorted, in parallel, MP candidates + + + CTTCATACCTGCATCG-39; UBASH3A forward: 59-AGCTCACATCGC- as KUL01 MHC-II cells (designated as KUL01 ), as well as T GCAAAAAG-39, and reverse: 59-TTGCAGTGAGAGTGCAACCA-39; (CD3+) and B (Bu-1+) cells (Fig. 1B–D). In an exploratory ap- VPREB3 forward: 59-AGTGACTTCGGCATCACCTG-39,andreverse: proach, we tested these cell subset candidates for the expression 59-GCGCTCCGTGTTGTAGTAGA-39; CD83 forward: 59-CCCTGTGC- of three murine/human cDC-associated genes (FLT3, ZBTB46 AATGTTTGGAGC-39, and reverse: 59-CCAAGACACTGCCTGGTAGG-39; GAPDH forward: 59-GTCCTCTCTGGCAAAGTCCAAG-39, and reverse: alias BTBD4, ARGHAP22) (4), and two murine/human MP-associ- 5 9-CCACAACATACTCAGCACCTGC-39; IFNG forward 59-CTCCCGA- ated genes (MAFB, CD14) (4, 31), using qPCR. MAFB and CD14 TGAACGACTTGAG-39, and reverse 59-CTGAGACTGGCTCCTTTTCC-39. mRNA were found expressed at the highest level in KUL01+ cells PCR cycling conditions were 95˚C for 5 min, linked to 40 cycles of 95˚C for 10 s and 60˚C for 1 min. A melting curve was established using the Chromo4 Real-Time Detector (Bio-Rad). The primers were confirmed to give a single hit by primer-BLAST analysis on the whole Gallus Refseq

National Center for Biotechnology Information database, except in the Downloaded from case of BCL11B, ZBTB46, and BATF3, owing to the existence of variants of the same gene, and in the case of CADM1, XCR1 CD14, and MAFB, in which hits were found on several gene targets for one of the two primers, but with more than three mismatches. In all cases, the melting curve of the PCR product and migration on an agarose gel demonstrated that a single amplicon was detected, supporting the specificity of all the primer pairs used for all the targeted genes. Real-time qPCR data were collected by the Opticon Monitor Software http://www.jimmunol.org/ (Bio-Rad) and 22DCt calculations for the relative expression of the different genes (arbitrary units) were performed using GAPDH for normalization. All qPCR reactions showed .95% efficacy. The qPCR data per gene were normalized to maximal expression across subsets. In the case of IFN-g qPCR, quantitation of IFN-g and GAPDH transcript numbers was done using calibration plasmids.

MLR After cell sorting, equal numbers of chicken spleen MPs and cDCs (3–5 3 by guest on September 24, 2021 104 per well, depending on experiments) were distributed in flat-bottom 96-well plates. Flow cytometry–sorted CD4+ T responder cells from an outbred chicken with a different MHC (PA12 chicken line) were added for 24–48 h of culture at 40˚C in DMEM F12 (Fisher) supplemented with penicillin–streptomycin, nonessential amino acids (Fisher), 50 mM 2-mer- captoethanol, and 2% FCS (Lonza). The ratio of responder to stimulator cells was 10. Transcription of IFN-g was tested as a readout of the MLR, using qPCR (3 per point, respectively). Cell viability was tested using trypan blue exclusion and was .95%.

Regulation of immunity-related gene expression in homologous subsets across human, mouse, and chicken species We established a list of 116 immunity-related genes (sensors, cytokine, cytokine receptors, signaling receptors) (30), for which the expression in the cell subsets was normalized across species (Supplemental Fig. 2).

Statistical analyses Normalized IFN-g transcript values were compared between stimulation conditions, using the nonparametric Mann–Whitney U test.

Results FIGURE 1. Isolation and exploratory analyses of cDC and MP candi- Identification of MP and cDC candidates in chicken spleen dates from chicken spleen. (A) cDC candidates were sorted from chicken 2 2 2 We previously identified DC-like cells in chicken splenocytes splenocytes. FSChi CD3 BU-1 KUL01 cells were first electronically that were morphologically distinct from MPs and that were not selected to sort single and double positive cells for MHC-II and CD11c + labeled by KUL01, a mAb usually considered to mark MPs (18, (putative) expression. (B) B cells were sorted as Bu-1 cells. (C) T cells + D + 28). To gain further insight into the existence of distinct MP were sorted as CD3 cells. ( ) MP candidates were sorted as KUL01 MHC class-II+ cells. (E) MGG staining of cDC candidates (MHC-II+ and cDC phagocytes in chicken, we followed a flow cytometry sort- CD11c+). Original magnification 3100. (F) MGG staining of MP candi- ing strategy to isolate candidates from suspensions of inbred SPF dates (KUL01+). Original magnification 3100. (G) Exploratory qPCR Leghorn chicken splenocytes, using a combination of pertinent analysis of cDC- and MP-associated gene expression in the sorted subsets markers among the relatively few number of mAbs available in from two independent experiments (pool of four spleens). The gene ex- this species. We based our reasoning on the fact that in mammals, pression was normalized across subsets to maximal expression and is re- most cDCs are “lineage”-negative cells that coexpress CD11c and ported as a heat map based on red intensity. The Journal of Immunology 4513

(Fig. 1G). FLT3 and ZBTB46 mRNA was found expressed at the GSEA (35). In all comparisons, the chicken cDCs* were found highest level in MHC-II+ CD11c+ splenocytes, and at relatively enriched for the mouse and human cDCs and CD8a+/BDCA3+ low levels in MHC-II+ CD11c2 and MHC-II2 CD11c+ spleno- Genesets (q values , 0.05 in most cases), whereas the chicken cytes. As expected, comparatively low expression of these cDC- MPs* were found enriched for the mouse and human myeloid and MP-associated genes was detected in T and B cells. In May– Genesets (q values , 0.05) (Fig. 2). In the cDC* versus MP* Grunwald–Giemsa€ (MGG) staining (32), the KUL01+ cells resemble comparison, the cDC and CD8a+/BDCA3+ signatures were the typical MPs with a high cytoplasmic/nuclear ratio and vacuolated only ones to be significantly enriched in a conserved manner. cytoplasm, whereas MHC-II+ CD11c+ cells presented large den- Conversely, in the MP* versus cDC* comparison, the dominant drites (Fig. 1E, 1F). The KUL01+ MHC-II+ cells appeared to often signature was the myeloid one. As positive controls, we confirmed include KUL01hi MHC-IIint and KUL01int MHC-IIhi populations that chicken T and B cells were enriched for the respective mouse/ (Fig. 1D) that appeared indistinguishable in the exploratory cDC human T and B cell Genesets in all comparisons (q values , 0.05). and MP gene expression analysis and that presented a homoge- The human and mouse pDC Genesets were not found together neous morphology in the MGG staining (data not shown). Thus, enriched in most comparisons using our selected populations of the exploratory qPCR and morphological analyses reassured us in chicken splenocytes, consistent with the fact that our strategy was selecting for the next steps the whole KUL01+ and the MHC-II+ not designed to identify pDCs in chicken that would have required CD11c+ populations as the MP* and cDC* candidates, respectively use of different cell surface marker combinations. (*marks the candidate status in this article). Thus the gene expression profiles show that cDCs* and MPs* are distinct cell types in the chicken and that they strongly re- Transcriptome mapping of chicken cDC* and MP* semble mouse and human cDCs and myeloid cells, respectively. Downloaded from To align the chicken MP* and cDC* with mammalian counter- Furthermore, chicken cDCs* appear to include a significant pro- parts, we devised a scale-up of a global gene expression analytic portion of cells of the CD8a+/BDCA3+ cDC types. method that was previously demonstrated to be relevant for dis- tinguishing new myeloid human subsets (33, 34) and for charac- Establishment of core gene expression signatures for T, B, terizing sheep cDC subsets (24). The transcriptome of each cDC, and MP subsets across mice, humans, and birds

chicken subset—that is, B, T, MP*, and cDC* (independent tri- or We sought to identify core gene expression signatures for T, B, http://www.jimmunol.org/ quadruplicates)—was obtained from mRNA hybridization on cus- cDCs, and MPs that would allow distinguishing these subsets from tomized chicken gene chips (see Materials and Methods). From other immune cell types across chickens, humans, and mice, and publicly available expression data (Supplemental Dataset 1), we that would likely apply to amniotes. identified mouse and human fingerprints for T cells, B cells, We selected the mRNA transcripts that contribute most to the plasmacytoid DCs (pDCs), and CD8a+/BDCA3+ cDCs, which enrichment of the mouse T, B, cDC, CD8a+, and MP fingerprints in were established as the list of genes expressed at least 1.5-fold all the chicken subset pairwise comparisons and that are provided higher in the index cell population than in a large number of other as “leading edge” lists by the GSEA. We repeated the same pro- immune cell types (Supplemental Dataset 2). We also identified cedure with the human fingerprints. We identified as core sig- mouse and human “cDC” and “myeloid” fingerprints as genes natures the genes found in common in the mouse against chicken by guest on September 24, 2021 overexpressed in all cDCs compared with in other myeloid cells and in the human against chicken leading edges lists (Fig. 3). The (i.e., monocytes and/or MPs from many different tissues), and MP core signature across chicken, human, and mouse encom- reciprocally (Supplemental Dataset 2). We next tested whether the passes the TLR4, CTSB, and CTSD genes. The transspecies cDC selected mouse and human fingerprints were enriched between the core signature encompasses CIITA, FAM46C, KIT, ZBTB46, four chicken subsets by performing pairwise comparisons using FLT3, PLEKHA5 (cDC fingerprint) as well as the XCR1 and

FIGURE 2. Murine and human cDC and myeloid fingerprints are enriched in the candidate chicken cDC* and MP*, respectively. GSEA was performed using sets of genes corresponding to the transcriptional fingerprints of specific murine (left) and human (right) cell types generated from compendia of expression data from many leukocytes (see Supplemental Dataset 2). Pairwise comparisons between chicken cDCs*, MPs*, and B and T cells were performed to assess enrich- ments of the mouse and human fingerprints. Results are represented as dots: The red and blue colors of the dots correspond to the color at- tributed to the subset in which the Geneset was enriched (see boxed legend within the figure); the circle surface area is proportional to the absolute value of the normalized enrichment score (NES), which varies between ~1 (no enrichment) and ~5 (best enrichment possible). The color intensity of circles is indicative of the false-discovery rate (FDR) statistical q value. 4514 cDC IDENTIFICATION IN CHICKEN

partially overlapping (30). To identify immune response genes that are differentially regulated in cDCs and monocytes/MPs across humans, mice, and chickens, we analyzed the expression of a se- lection of immune response genes across the arrays used in this study (Supplemental Dataset 1; see Materials and Methods). A group of genes clustered together that appeared to be expressed at higher levels in monocytes/MPs and cDCs than in lymphoid cells in several comparisons (Supplemental Fig. 2, pink dendo- gram). Among these genes, some showed a higher expression in cDCs than in monocytes/MPs in all three species, that is, FLT3, XCR1, and TLR3 (Fig. 4, right panel). Several immunity mol- ecule mRNAs were overexpressed in chicken cDCs compared with MPs, for example, for CD86, ICOSLG, and CSFR2A, whereas it was only the case in one of the two mammals. Conversely, other genes showed a higher expression in monocytes/MPs than in cDCs in all three species: CD14, TLR2, TLR4, P2RY13, SIRPa, and CSFR1 (Fig. 4, left panel). Finally, inflammatory cytokine genes (IL6, IL1B) were markedly overexpressed in chicken MPs

compared with cDCs and far less clearly so in the corresponding Downloaded from mammalian subsets. The different array data originated from cells that were isolated from distinct tissue types with different markers for subsetting, which may hinder similar regulation of gene ex- pression in homologous immune subsets across species. The ex- pression of CD14, TLR2, and TLR4 was higher in monocytes/MPs

than in cDCs and lymphoid cells, demonstrating a conserved http://www.jimmunol.org/ specialization of MPs in the recognition of bacteria in all three species. + FIGURE 3. Establishment of a core signature of cDCs, MPs, and B and Chicken cDCs are efficient at CD4 T cell activation T cells across chicken, human, and mouse species. Bar chart representing In mammals, cDCs are potent APCs that can activate naive T the chicken gene chip normalized expression values of signature genes for cells with optimal efficacy. We thus tested whether chicken cDCs chicken cDCs (red), MPs (grey), B cells (blue), and T cells (green). Genes presented the capacity to activate allogeneic CD4+ T cells. In five were selected as 1) being part of both the mouse and human transcriptomic independent experiments, we observed that chicken spleen cDCs, fingerprints (T, MP, CD8a/BDCA3+-type cDC, cDC), and as 2) found enriched in the equivalent chicken cell type according to the GSEA “lead- without exogenous activation, were efficient at stimulating allo- by guest on September 24, 2021 + , ing edge” definition. Each bar corresponds to the average value of three geneic CD4 T cells for IFN-g (p 0.05), whereas splenic MPs to four replicates for the four different chicken cell types. Error bars rep- isolated in parallel were not (Fig. 5). Thus, as in mammals, cDCs resent the SD for the replicates. Normalized expression values were scaled are more potent in activating allogeneic T cells than are MPs. to the highest expression value for each gene. Discussion In this article, we show that MP and cDC subsets exist in the CADM1 genes (CD8a+/BDCA3+ cDC type fingerprint). Core chicken as two distinct immune cell lineages that display gene signatures for B cells and T cells across species were also estab- expression profiles sharing substantial homology with their mam- lished (B cells: CXCR5, SWAP70, CD79B, KLHL14, POU2AF1, malian counterparts. To our knowledge, this is the first time that EBF1, VPREB3, PAX5; T cells: LEF1, CD5, TCF7, UBASH3A, cDCs have been identified outside mammals, and it shows that the CD28, BCL11B). cDC/MP paradigm extended beyond mammals, to vertebrates that These lists represent the minimal core signatures that are re- diverged .300 million years ago from the common ancestor of stricted by the fact that several pertinent genes are probably miss- reptiles, birds, and mammals. Birds also present other phagocytic ing because of misannotated, nonfunctioning, or absent probes cells that mammals do not—including heterophils (the chicken on at least one of the three species arrays. To confirm and extend functional equivalent of mammalian neutrophils) and thrombocytes the transspecies cell subset core signatures, we performed com- (homologous in function to mammalian platelets, which are absent plementary qPCR analyses for a selection of genes of the minimal in the chicken)—for which their role as APCs is not established. core signature established above, and we added genes showing Comparative gene expression profiling allowed us to determine a conserved expression pattern between cDCs and MPs in the similarities between immune cell subsets across the chicken, hu- mouse and human for which the signals from the chicken microarray man, and mouse. Whereas this approach avoids biases of a priori remained in the noise, with no difference between the forward and hypotheses, it unravels similarities between subsets, but not exact reverse probes, suggesting that the forward probes were nonfunc- equivalences, for different reasons. First, the MHC-II+ CD11c+ tioning (Supplemental Fig. 1). This qPCR analysis showed that we subset that was sorted with our strategy, corresponding to ∼1% of can add ARGHAP22 and BEND5 to the cDC, C5AR1 to the MP, splenocytes, was clearly enriched for bona fide cDCs, but it is and PCLG1 to the T cell signatures across the three species. possible that non-cDCs, such as NK and innate lymphoid cells, were included. In addition, our selection process of human and Conserved regulation of immune response–related genes in murine cell subset fingerprints was stringent and eliminated many cDCs and MPs in the human, mouse, and chicken genes that were not found in common in the whole range of the In humans and mice, DC subsets and MPs express specific and datasets corresponding to a population type (MP, DC), owing to relatively conserved profiles of immune response genes, although tissue or species exception, absence of functional probes, or mis- The Journal of Immunology 4515 Downloaded from

FIGURE 4. Comparative expression of immune response–re- lated genes in cDCs, MPs, and T and B cells across chicken, human, and mouse species. Bar chart representing the relative expression (mean 6 SD) of a selection of monocyte/MP-specific http://www.jimmunol.org/ (top) or cDC-specific (bottom) immune response–related genes in chicken (left), mouse (middle), and human (right) cDC and monocyte/MP populations (cDC in red/orange and MP in black/ gray in the legend, respectively). These populations include, for chicken: MPs and cDCs; for human: monocyte (mono)–derived MPs, PBMC-derived MPs, nonclassical blood monocytes, clas- sical monocytes, BDCA3+ cDCs, and BDCA1+ cDCs; for mouse: peritoneal cavity MPs, lung MPs, nonclassical monocytes, clas- sical monocytes, splenic CD8a+ cDCs, cutaneous lymph node

CD8a+ cDCs, splenic CD11b+ cDCs, and cutaneous lymph node by guest on September 24, 2021 CD11b+ cDCs. Values were computed from two to five inde- pendent replicates for each cell population, except for human mono-derived MP (n = 1). Normalized gene expression values were scaled for each gene and species to the highest expression value across all cell types.

annotated probes. An example is the absence of CD14 and MAFB are highly expressed in chicken MPs, do not fall in the myeloid from the mouse myeloid signature because of their low expression core signature across the chicken, human, and mouse. To avoid in MPs of some tissues. Consequently, CD14 and MAFB, which this problem, the human and mouse fingerprints should have been 4516 cDC IDENTIFICATION IN CHICKEN

and B lymphocytes, with no difference between cDCs and MPs. In agreement with this finding, a newly generated anti-chicken CD205 mAb was found to label both putative DCs (CD83+)and MPs (KUL01+) in chicken spleen (22). Thus, in the chicken, CD205 may not discriminate cDCs from MPs. The core gene expression profiles we identified in this study allow proposal of a unifying molecular definition of T, B, MP, and cDC subsets across the chicken, human, and mouse, which should be valid across amniotes. These core signatures can be used to search for homologous subsets in other model and application species, including distant vertebrates such as fishes. Most genes of + FIGURE 5. Chicken cDCs are efficient at allogeneic CD4 T cell the transspecies MP and cDC core signatures are included in the stimulation. cDCs and MPs were sorted from pools of four chicken spleens core gene lists of MP and cDCs across mouse tissues identified in (inbred B13 MHC line), and they were plated at a 1:10 ratio with sorted an independent study by the Immgen group (31). Several of the allogeneic splenic CD4+ T cells (outbred PA12 chicken). After cocul- a+ CADM1 XCR1 ture, the RNA was extracted and subjected to qPCR for IFN transcript CD8 cDC-type genes, such as and , were found quantitation, using GAPDH for normalization. The relative IFN-g mRNA in the transspecies cDC core signature, strongly indicating that + copy numbers from five independent experiments (shown as distinct chicken are equipped in CD8a -type cDCs and that their cDC symbols) are reported with means and SEM. Significant statistical lineage is likely composed of the two subsets previously identified differences were given by the nonparametric Mann–Whitney U test. in mammals, CD8a+-type and CD11b+-type cDCs. Whereas sev- Downloaded from *p , 0.05. eral genes of the cDC signature have already been studied for their biological implication in cDC biology, nothing is known regarding PLEKHA5 and FAM46C. Similarly, the molecules of the T and derived from the same tissue (= spleen) and with similar gating B cell core signature are well known for their function, except for strategies as the one used for chicken, but such transcriptomic data KLHL14. As these genes are conserved in the gene expression are at present not available. Finally, the commercial chicken gene program of specific immune cell subsets across evolution, they http://www.jimmunol.org/ chips, although improved by us, were not as optimal as the com- should be studied for their role in the specific function or devel- mercial human and mouse ones, owing to misannotations, missing opment of these subsets. probes, missing known orthologs to mammals, and nonfunctioning Our results underline some degree of conservation of functions probes. These handicaps resulted in a reduced number of analyzed for cDCs and MPs across amniotes. Several immunity-related probes that probably led to an underestimation of the GSEA and to genes were similarly regulated between cDCs and MPs across incomplete core gene expression lists. Specifically, as mentioned the three species under study. The conserved selective expression before, no probes were identified on the chicken microarrays as of sensors that are considered more viral (TLR3) and bacterial being specific for PLCG1, C5AR1, BEND5, and ARGHAPP22, but oriented (TLR2, TLR4, CD14) in cDCs and MPs, respectively, we designed qRT-PCR for these genes to complement knowledge suggest that these two lineages are specialized in distinct pathogen by guest on September 24, 2021 on the core lists featuring genes with a conserved expression sensing in amniotes. Furthermore, CTSB and CTSD are members pattern between human, mouse, and chicken immune cells. Other of the MP core signature, further pointing to the strong lysosomal genes with a conserved expression pattern across cell types in activity in MPs for degradation purposes. In addition, the stronger humans and mice, for which no probes were identified on the capacity for CD4 T cell activation, as well as the higher expres- chicken microarrays but for which an ortholog exists in this sion of the CD86 and ICOSLG genes in chicken cDCs compared species according to Ensembl, included HLA-DOB and ETV3 with MPs, shows that in all species, cDCs developed as special- for the cDC fingerprint; TCF4 (E2-2) for the pDC fingerprint; ized in T cell activation. BANK1, CD22, FCRL1, and FCRL5 for the B cell fingerprint; Efficient strategies to prevent disease outbreak in chicken farms CD3G, CD6, and DGKA for the T cell fingerprint; and MRC1 is mandatory to supply the growing world demand for poultry (CD206), which is selectively expressed in monocyte-derived products that is expected to increase at a 2.5% rate per year until DCs and some MPs in both the mouse and human. In the 12th 2030. The poultry sector strongly relies on vaccination for main- Avian Immunology Research Group Conference (in 2012), the tenance of health and production status, with birds receiving .20 KUL01 mAb was reported to be specific to CD206 (36). Thus, vaccines during their short life (16). Our identification and mo- in consistency with what was observed in the mouse and human, lecular characterization of cDCs in chickens will help in designing CD206 seems selectively expressed in the monocyte/MP and not better vaccine strategies and in developing new immunomodula- in the cDC lineage in the chicken. No orthologs are yet reported tors as alternatives to antibiotics. in the chicken for CLEC9A for the signature of XCR1+ DC, CD19 and CD79A for the B cell signature, as well as for SELPLG Acknowledgments (CD162) and TNFSF18 (GITRL). For CD162, CD19, and CD79A, We thank Christelle Thibault-Carpentier from the Plate-forme Biopuces orthologs have been identified in birds other than the chicken, in (Strasbourg, France) for performing the microarray experiments (www- reptiles, and in fishes, strongly suggesting that orthologs exist in microarrays.u-strasbg.fr); Mickael Bourge (Imagif, Gif-sur-Yvette, France) the chicken but still await identification. For CLEC9A and TNFSF18, for the pilot flow cytometry sorting experiments; the Plate-Forme d’Infec- no orthologs have been identified outside mammals, strongly sug- tiologie Expe´rimentale, Nouzilly (France) for breeding and providing gesting that these genes do not exist in other vertebrates, including high standard SPF chickens; Bertrand Bed’hom (Institut National de la birds. In consistency with this, in the case of CLEC9A, we included Recherche Agronomique, France) for the genotyping of the MHC of the probes on the gene chip corresponding to the mouse and human PA12 chicken line; Pierre Boudinot for helpful and enlightening discus- gene, but they did not give any signal. LY75 (CD205) is selectively, sions; and Luc Jouneau for guidance on chicken microarray improvements. although not specifically, expressed on XCR1+ DCs in the human and mouse. Two corresponding probes were present on the chicken Disclosures array and gave an ∼2-fold higher signal in cDCs and MPs than in T The authors have no financial conflicts of interest. The Journal of Immunology 4517

References 18. Que´re´, P., J. Pierre, M. D. Hoang, E. Esnault, J. Domenech, P. Sibille, and I. Dimier-Poisson. 2013. Presence of dendritic cells in chicken spleen cell 1. Merad, M., P. Sathe, J. Helft, J. Miller, and A. Mortha. 2013. The dendritic cell preparations and their functional interaction with the parasite Toxoplasma gon- lineage: ontogeny and function of dendritic cells and their subsets in the steady dii. Vet. Immunol. Immunopathol. 153: 57–69. state and the inflamed setting. Annu. Rev. Immunol. 31: 563–604. 19. Wu, Z., L. Rothwell, J. R. Young, J. Kaufman, C. Butter, and P. Kaiser. 2010. 2. Schraml, B. U., J. van Blijswijk, S. Zelenay, P. G. Whitney, A. Filby, S. E. Acton, Generation and characterization of chicken bone marrow-derived dendritic cells. N. C. Rogers, N. Moncaut, J. J. Carvajal, and C. Reis e Sousa. 2013. Genetic Immunology 129: 133–145. tracing via DNGR-1 expression history defines dendritic cells as a hematopoietic 20. Igya´rto´, B. Z., E. Lacko´,I.Ola´h, and A. Magyar. 2006. Characterization of lineage. Cell 154: 843–858. chicken epidermal dendritic cells. Immunology 119: 278–288. 3. Naik, S. H., L. Perie´, E. Swart, C. Gerlach, N. van Rooij, R. J. de Boer, and 21. Hansell, C., X. W. Zhu, H. Brooks, M. Sheppard, S. Withanage, D. Maskell, and T. N. Schumacher. 2013. Diverse and heritable lineage imprinting of early I. McConnell. 2007. Unique features and distribution of the chicken CD83+ cell. haematopoietic progenitors. Nature 496: 229–232. J. Immunol. 179: 5117–5125. 4. Robbins, S. H., T. Walzer, D. Dembe´le´, C. Thibault, A. Defays, G. Bessou, 22. Staines, K., J. R. Young, and C. Butter. 2013. Expression of chicken DEC205 H. Xu, E. Vivier, M. Sellars, P. Pierre, et al. 2008. Novel insights into the reflects the unique structure and function of the avian immune system. PLoS relationships between dendritic cell subsets in human and mouse revealed by ONE 8: e51799. genome-wide expression profiling. Genome Biol. 9: R17. 23. de Geus, E. D., C. A. Jansen, and L. Vervelde. 2012. Uptake of particulate antigens 5. Satpathy, A. T., X. Wu, J. C. Albring, and K. M. Murphy. 2012. Re(de)fining the in a nonmammalian lung: phenotypic and functional characterization of avian re- dendritic cell lineage. Nat. Immunol. 13: 1145–1154. spiratory phagocytes using bacterial or viral antigens. J. Immunol. 188: 4516–4526. 6. Meredith, M. M., K. Liu, G. Darrasse-Jeze, A. O. Kamphorst, H. A. Schreiber, 24. Contreras, V., C. Urien, R. Guiton, Y. Alexandre, T. P. Vu Manh, T. Andrieu, P. Guermonprez, J. Idoyaga, C. Cheong, K. H. Yao, R. E. Niec, and K. Crozat, L. Jouneau, N. Bertho, M. Epardaud, et al. 2010. Existence of CD8a- M. C. Nussenzweig. 2012. Expression of the zinc finger transcription factor zDC like dendritic cells with a conserved functional specialization and a common (Zbtb46, Btbd4) defines the classical dendritic cell lineage. J. Exp. Med. 209: molecular signature in distant mammalian species. J. Immunol. 185: 3313–3325. 1153–1165. 25.Haniffa,M.,A.Shin,V.Bigley,N.McGovern,P.Teo,P.See,P.S.Wasan, 7. Hoeffel, G., Y. Wang, M. Greter, P. See, P. Teo, B. Malleret, M. Leboeuf, D. Low, X. N. Wang, F. Malinarich, B. Malleret, et al. 2012. Human tissues contain G. Oller, F. Almeida, et al. 2012. Adult Langerhans cells derive predominantly CD141hi cross-presenting dendritic cells with functional homology to from embryonic fetal liver monocytes with a minor contribution of yolk sac- mouse CD103+ nonlymphoid dendritic cells. Immunity 37: 60–73. Downloaded from derived macrophages. J. Exp. Med. 209: 1167–1181. 26. Rothwell, C. J., L. Vervelde, and T. F. Davison. 1996. Identification of chicken 8. Haniffa, M., M. Collin, and F. Ginhoux. 2013. Ontogeny and functional spe- Bu-1 alloantigens using the monoclonal antibody AV20. Vet. Immunol. Immu- cialization of dendritic cells in human and mouse. Adv. Immunol. 120: 1–49. nopathol. 55: 225–234. 9. Schlitzer, A., N. McGovern, P. Teo, T. Zelante, K. Atarashi, D. Low, A. W. Ho, 27. Bernot, A., and C. Auffray. 1991. Primary structure and ontogeny of an avian P. See, A. Shin, P. S. Wasan, et al. 2013. IRF4 transcription factor-dependent CD3 transcript. Proc. Natl. Acad. Sci. USA 88: 2550–2554. CD11b+ dendritic cells in human and mouse control mucosal IL-17 cytokine 28. Mast, J., B. M. Goddeeris, K. Peeters, F. Vandesande, and L. R. Berghman. 1998. responses. Immunity 38: 970–983. Characterisation of chicken monocytes, macrophages and interdigitating cells by

10. Jongbloed, S. L., A. J. Kassianos, K. J. McDonald, G. J. Clark, X. Ju, the monoclonal antibody KUL01. Vet. Immunol. Immunopathol. 61: 343–357. http://www.jimmunol.org/ C. E. Angel, C. J. Chen, P. R. Dunbar, R. B. Wadley, V. Jeet, et al. 2010. Human 29. Baranek, T., T. P. Manh, Y. Alexandre, M. A. Maqbool, J. Z. Cabeza, CD141+ (BDCA-3)+ dendritic cells (DCs) represent a unique myeloid DC E. Tomasello, K. Crozat, G. Bessou, N. Zucchini, S. H. Robbins, et al. 2012. subset that cross-presents necrotic cell antigens. J. Exp. Med. 207: 1247–1260. Differential responses of immune cells to type I interferon contribute to host 11. Poulin, L. F., M. Salio, E. Griessinger, F. Anjos-Afonso, L. Craciun, J. L. Chen, resistance to viral infection. Cell Host Microbe 12: 571–584. A. M. Keller, O. Joffre, S. Zelenay, E. Nye, et al. 2010. Characterization of 30. Crozat, K., E. Vivier, and M. Dalod. 2009. Crosstalk between components of the human DNGR-1+ BDCA3+ leukocytes as putative equivalents of mouse innate immune system: promoting anti-microbial defenses and avoiding immu- CD8alpha+ dendritic cells. J. Exp. Med. 207: 1261–1271. nopathologies. Immunol. Rev. 227: 129–149. 12. Bachem, A., S. Guttler,€ E. Hartung, F. Ebstein, M. Schaefer, A. Tannert, 31. Gautier, E. L., T. Shay, J. Miller, M. Greter, C. Jakubzick, S. Ivanov, J. Helft, A. Salama, K. Movassaghi, C. Opitz, H. W. Mages, et al. 2010. Superior antigen A. Chow, K. G. Elpek, S. Gordonov, et al; Immunological Genome Consortium. cross-presentation and XCR1 expression define human CD11c+CD141+ cells as 2012. Gene-expression profiles and transcriptional regulatory pathways that homologues of mouse CD8+ dendritic cells. J. Exp. Med. 207: 1273–1281. underlie the identity and diversity of mouse tissue macrophages. Nat. Immunol. 13. Crozat, K., R. Guiton, V. Contreras, V. Feuillet, C. A. Dutertre, E. Ventre, 13: 1118–1128. by guest on September 24, 2021 T. P. Vu Manh, T. Baranek, A. K. Storset, J. Marvel, et al. 2010. The XC che- 32. Russell, A. P., ed. 1997. Delmar’s Clinical Laboratory Manual Series: Hema- mokine receptor 1 is a conserved selective marker of mammalian cells homol- tology, 1st Ed. Delmar Publishers, New York, p. 53. ogous to mouse CD8alpha+ dendritic cells. J. Exp. Med. 207: 1283–1292. 33. Crozat, K., R. Guiton, M. Guilliams, S. Henri, T. Baranek, I. Schwartz-Cornil, 14. Mittag, D., A. I. Proietto, T. Loudovaris, S. I. Mannering, D. Vremec, B. Malissen, and M. Dalod. 2010. Comparative genomics as a tool to reveal K. Shortman, L. Wu, and L. C. Harrison. 2011. Human dendritic cell subsets functional equivalences between human and mouse dendritic cell subsets. from spleen and blood are similar in phenotype and function but modified by Immunol. Rev. 234: 177–198. donor health status. J. Immunol. 186: 6207–6217. 34. Segura, E., M. Touzot, A. Bohineust, A. Cappuccio, G. Chiocchia, A. Hosmalin, 15. Cohn, L., B. Chatterjee, F. Esselborn, A. Smed-So¨rensen, N. Nakamura, M. Dalod, V. Soumelis, and S. Amigorena. 2013. Human inflammatory dendritic C. Chalouni, B. C. Lee, R. Vandlen, T. Keler, P. Lauer, et al. 2013. Antigen cells induce Th17 cell differentiation. Immunity 38: 336–348. delivery to early endosomes eliminates the superiority of human blood BDCA3+ 35. Subramanian, A., H. Kuehn, J. Gould, P. Tamayo, and J. P. Mesirov. 2007. dendritic cells at cross presentation. J. Exp. Med. 210: 1049–1063. GSEA-P: a desktop application for Gene Set Enrichment Analysis. Bioinformatics 16. Wu, Z., and P. Kaiser. 2011. Antigen presenting cells in a non-mammalian model 23: 3251–3253. system, the chicken. Immunobiology 216: 1177–1183. 36. Staines, K., J. R. Young, L. Hunt, J. A. Hammond, and C. Butler. 2012. An 17. Lugo-Villarino, G., K. M. Balla, D. L. Stachura, K. Ban˜uelos, M. B. Werneck, expanded family in avians is suggestive of a diversified and D. Traver. 2010. Identification of dendritic antigen-presenting cells in the functional repertoire. In 12th avian immunology research group conference zebrafish. Proc. Natl. Acad. Sci. USA 107: 15850–15855. proceedings. Edinburgh, U.K., p. 37. Normalized expression values 0.0 0.2 0.4 0.6 0.8 1.0 1.2

PLCG1 * T CD5 MP

T B TCF7 cDC UBASH3A

BCL11B

C5AR1 * MP TLR4

CXCR5

B CD79B

VPREB3

XCR1

CADM1

BEND5 *

ARGHAP22 * cDC CIITA

ZBTB46

FLT3

PLEKHA5

Figure S1. qPCR analysis of the core gene expression signature of cDC, MP, B and T cells in the chicken cell subsets. RNA from cDC, MP, T and B cells of 2 disnct pools of 4 chicken spleen was subjected to qPCR detecon of the core gene expression signatures of immune cell subsets established in Fig. 3 and of transcripts from the mouse and human gene subset selected compendia that could not be detected on the array due to defecve probes, i.e. ARGHAP22, BEND5, C5AR1, and PLCG1 (labeled by a star on the figure). Data are represented as the mean and SD of relave gene expression levels normalized to GAPDH expression and the maximal expression across the cell types was set to 1 (independent experimental duplicates). B cell T cell cDC Monocyte/MP Chicken Human Mouse

Figure S2. Unsupervised hierarchical clustering of orthologous immune response genes across chicken, human and mouse reveals globally conserved clusters of lymphoïd- specific and myeloïd- specific genes. Heatmap of cross-normalized expression profiles for immune response genes present on all three species arrays and regulated at least 2 folds across all cell subsets, including chicken B cells (c_B), T cells (c_CD3), MP (c_MP) and cDC (c_cDC), human B cells (h_B), T cells (h_CD4_T and h_CD8_T), monocyte-derived MP (h_MoMP), peripheral blood mononucleated cell-derived MP (h_PBMC_MP), non-classical monocytes (h_non- classical_MO), classical monocytes (h_classical_MO), BDCA3+ cDC (h_BDCA3), BDCA1+ cDC (h_BDCA1), murine B cells (m_B), T cells (m_CD4_T and m_CD8_T), peritoneal cavity MP (m_PC_MPII-480HI), lung MP (m_LU_MP), non-classical monocytes (m_non-classical_MO), classical monocytes (m_classical_MO), splenic CD8α+ cDC (m_SP_DC1), subcutaneous lymph node CD8α+ cDC (m_LN_DC1), splenic CD11b+ cDC (m_SP_DC2), subcutaneous lymph node CD11b+ cDC (m_LN_DC2). The color scale indicates the expression levels in each cell subset normalized to the mean expression level across all cell types for each species, from blue and blue (lower expression) to yellow and red (higher expression). The figure was generated with the Gene-E soware (version 2.1.71) from the Broad Instute, using the average linkage as a clustering method and the one-pearson distance as a correlaon metric. (The pink box highlights the genes that are more highly expressed in MP and cDC than in B and T cells) DataSet1_mouse: Listing of the mouse gene chips data used to establish mouse cell subset transcriptional fingerprints database MOUSE arrays sample name cell type Tissue GEO GSM538258 DC1.SP CD8α+ cDC Spleen GEO GSM538259 DC1.SP CD8α+ cDC Spleen GEO GSM538260 DC1.SP CD8α+ cDC Spleen GEO GSM538261 DC1.SP CD8α+ cDC Spleen GEO GSM605827 DC1.SP CD8α+ cDC Spleen GEO GSM538255 DC1.LN CD8α+ cDC CLN GEO GSM538256 DC1.LN CD8α+ cDC CLN GEO GSM538257 DC1.LN CD8α+ cDC CLN GEO GSM538282 MF.LU macrophage lung GEO GSM538283 MF.LU macrophage lung GEO GSM538284 MF.LU macrophage lung GEO GSM605850 MF.II‐480HI.PC macrophage peritoneum GEO GSM605851 MF.II‐480HI.PC macrophage peritoneum GEO GSM605852 MF.II‐480HI.PC macrophage peritoneum GEO GSM605872 MO.6C+II‐.BL c monocyte Blood GEO GSM605873 MO.6C+II‐.BL c monocyte Blood GEO GSM605874 MO.6C+II‐.BL c monocyte Blood GEO GSM605884 MO.6C‐II‐.BL nc monocyte Blood GEO GSM605885 MO.6C‐II‐.BL nc monocyte Blood GEO GSM538201 B6SPLFO B lymphocyte Spleen GEO GSM538202 B6SPLFO B lymphocyte Spleen GEO GSM538203 B6SPLFO B lymphocyte Spleen GEO GSM854306 GN.BL neutrophils Blood GEO GSM854307 GN.BL neutrophils Blood GEO GSM854308 GN.BL neutrophils Blood GEO GSM538398 T.8Mem.Sp CD8 T cells Spleen GEO GSM538399 T.8Mem.Sp CD8 T cells Spleen GEO GSM538400 T.8Mem.Sp CD8 T cells Spleen GEO GSM538365 T.4Mem.Sp CD4 T cells Spleen GEO GSM538366 T.4Mem.Sp CD4 T cells Spleen GEO GSM538367 T.4Mem.Sp CD4 T cells Spleen GEO GSM538315 NK.SP NK cells Spleen GEO GSM538316 NK.SP NK cells Spleen GEO GSM538317 NK.SP NK cells Spleen GEO GSM605840 DC.PDC.8+.SP CD8+ pDC Spleen GEO GSM605841 DC.PDC.8+.SP CD8+ pDC Spleen GEO GSM605842 DC.PDC.8+.SP CD8+ pDC Spleen GEO GSM605843 DC.PDC.8‐.SP CD8‐ pDC Spleen GEO GSM605844 DC.PDC.8‐.SP CD8‐ pDC Spleen GEO GSM605845 DC.PDC.8‐.SP CD8‐ pDC Spleen GEO GSM605826 DC2.SP CD11b+ cDC Spleen GEO GSM538248 DC2.SP CD11b+ cDC Spleen GEO GSM538249 DC2.SP CD11b+ cDC Spleen GEO GSM538250 DC2.SP CD11b+ cDC Spleen GEO GSM538251 DC2.SP CD11b+ cDC Spleen GEO GSM538245 DC2.LN CD11b+ cDC CLN GEO GSM538246 DC2.LN CD11b+ cDC CLN GEO GSM538247 DC2.LN CD11b+ cDC CLN GEO GSM538280 DC.LC.SK Langerhans cells epidermis GEO GSM538281 DC.LC.SK Langerhans cells epidermis GEO GSM879263 BMDC MoDC in vitro derived GEO GSM879264 BMDC MoDC in vitro derived DataSet1_human: Listing of the human gene chips data used to establish human cell subset transcriptional fingerprints database HUMAN arrays cell type Tissue ArrayExpress E‐TABM‐34 MBA:P BDCA3+ cDCs Blood ArrayExpress E‐TABM‐34 MBA:Q BDCA3+ cDCs Blood ArrayExpress E‐TABM‐34 MBA:R BDCA3+ cDCs Blood ArrayExpress E‐TABM‐34 MBA:M BDCA1+ cDCs Blood ArrayExpress E‐TABM‐34 MBA:N BDCA1+ cDCs Blood ArrayExpress E‐TABM‐34 MBA:O BDCA1+ cDCs Blood ArrayExpress E‐TABM‐34 MBA:J pDCs Blood ArrayExpress E‐TABM‐34 MBA:K pDCs Blood ArrayExpress E‐TABM‐34 MBA:L pDCs Blood ArrayExpress E‐TABM‐34 MBA:S nc monocytes Blood ArrayExpress E‐TABM‐34 MBA:T nc monocytes Blood ArrayExpress E‐TABM‐34 MBA:U nc monocytes Blood footnote1 1_FRSharp_mono c monocytes Blood footnote1 2_FRSharp_mono c monocytes Blood footnote1 3_FRSharp_mono c monocytes Blood footnote1 1_FRSharp_CD4_T CD4 T cells Blood footnote1 2_FRSharp_CD4_T CD4 T cells Blood footnote1 3_FRSharp_CD4_T CD4 T cells Blood footnote1 1_FRSharp_CD8_T CD8 T cells Blood footnote1 2_FRSharp_CD8_T CD8 T cells Blood footnote1 3_FRSharp_CD8_T CD8 T cells Blood footnote1 1_FRSharp_B_Lympho B lymphocytes Blood footnote1 2_FRSharp_B_Lympho B lymphocytes Blood footnote1 3_FRSharp_B_Lympho B lymphocytes Blood footnote1 1_FRSharp_NK NK cells Blood footnote1 2_FRSharp_NK NK cells Blood footnote1 3_FRSharp_NK NK cells Blood footnote1 1_FRSharp_neu neutrophils Blood footnote1 2_FRSharp_neu neutrophils Blood footnote1 3_FRSharp_neu neutrophils Blood GEO GSM109787 macrophages in vitro derived GEO GSM109788 macrophages in vitro derived GEO GSM109789 macrophages in vitro derived GEO GSM181857 MoDC in vitro derived GEO GSM181931 MoDC in vitro derived GEO GSM181933 MoDC in vitro derived GEO GSM181971 MoDC in vitro derived GEO GSM181973 MoDC in vitro derived GEO GSM181978 MoDC in vitro derived GEO GSM579100 MoDC in vitro derived GEO GSM579101 MoDC in vitro derived GEO GSM579102 MoDC in vitro derived GEO GSM213500 macrophages in vitro derived GEO GSM38347 macrophages lung GEO GSM38349 macrophages lung GEO GSM38352 macrophages lung GEO GSM38354 macrophages lung GEO GSM38357 macrophages lung GEO GSM38359 macrophages lung GEO GSM38360 macrophages lung GEO GSM38366 macrophages lung GEO GSM38367 macrophages lung GEO GSM38369 macrophages lung GEO GSM38374 macrophages lung GEO GSM38376 macrophages lung GEO GSM38380 macrophages lung GEO GSM38382 macrophages lung GEO GSM579106 Langerhans cells epidermis GEO GSM579107 Langerhans cells epidermis GEO GSM579108 Langerhans cells epidermis GEO GSM176001 B lymphocytes Blood, cultured ex vivo GEO GSM176007 B lymphocytes Blood, cultured ex vivo GEO GSM176003 c monocytes Blood, cultured ex vivo GEO GSM176009 c monocytes Blood, cultured ex vivo GEO GSM175999 T cells Blood, cultured ex vivo GEO GSM176005 T cells Blood, cultured ex vivo 1 http://www‐microarrays.u‐strasbg.fr/files/datasetsE.php Dataset S2. Transcriptomic signatures of mouse cell types CDC_VS_MYELOÏD CD8α DC B T MYELOÏD_VS_CDC PDC ADAM23 A530099J19RIK CR2 ART2B PILRA SIGLECH H2‐EB1 XCR1 FCER2A GM13949 SEPX1 HAVCR1 H2‐AB1 GCET2 EBF1 THEMIS CLEC4E DUXBL H2‐AA TLR3 CD19 TCRA MGST1 DNTT DNASE1L3 CXCL9 FAIM3 CD3G FGD4 GRM8 KIT SNX22 PAX5 CD3E THBD GRIA3 APOL7C IFI205 IGK‐V28 CD6 GDA BST2 H2‐OA TLR11 MS4A1 CD3D SQRDL LRP8 CD74 FAM149A CD79A CAMK4 KLRA2 CD209D SLAMF7 HEPACAM2 GM5571 GM13969 PLOD3 PDZD4 CCL5 1700009J07RIK ANGPTL1 CD5 MPP1 TUBGCP5 KLRI1 LEPREL1 BLK GM8800 C5AR1 2210020M01RIK CIITA ECE1 IGHMAC38.205.12 NSG2 HGF CCR9 ANPEP TTC39A CD79B GM8740 DOK3 TEX2 H2‐EB2 IL12B BANK1 LAT PROS1 UPB1 ZFP366 GPR33 GM189 EHD3 FRY TCF4 H2‐OB CADM1 IGK DZIP1 HIP1 CD300C CCR7 BC037703 POU2AF1 UBASH3A SLC16A10 COX6A2 DPP4 ALMS1 LOC100047053 THY1 GSR GM10790 H2‐DMB2 HTR7 CD55 A130014H13RIK ABCC5 CHDH IL4I1 PPT1 LOC435333 IFT80 PTPLAD2 EAR14 P2RY10 BCL2L14 CNN3 BCL11B TCN2 LAG3 FLT3 CLEC1A IGKV4‐71 GM14085 FCGR3 GM12253 SERPINA3F FAM40B IGH‐6 CXCR6 MS4A6D PAQR5 ICOSL GCLC PIK3C2B LOC547323 CEBPB OBSCN KLRD1 PDIA5 LOC676175 GM13907 GSTM1 PIR TRAF1 AIF1 LOC100046973 CD28 ABCD2 KLK1 IL7R CXX1C IGHV1‐72 GM10673 LRP1 ARHGAP32 RGS12 NOTCH4 SCD1 PLCG1 PLEKHG1 TMEM229B ASB2 TMEM27 RALGPS2 TCRB‐J TOM1 RUNX2 TRIM7 FNDC7 LOC672291 TNFSF8 FCGR1 PNCK TSPAN33 ITGAE CACNA1I TRAV14D‐1 SEPP1 SLC41A2 SLCO5A1 TXNDC15 LOC100046496 DNAHC8 SMPDL3B RNF122 RAB30 FGD2 DENND5B GM16452 SMPDL3A HS3ST1 GPR68 FAM190A GM1077 CD27 CCL6 RAB33B GBP4 ITPRIPL1 CHST3 SLC35F2 KLF10 GM14207 ADAM11 WDFY4 IL5RA DENND2D DGKG CLCN5 HMGN3 FNBP1 IGK‐V19‐14 LEF1 ALDH3B1 MIR592 SLC4A8 B3GNT5 FAM78A PDLIM1 SRL ADAM19 GM1419 GM10890 F10 TMEM221 ZBTB46 D130062J21RIK CCDC64 CCDC125 GCOM1 AP1S3 IGK‐V1 TCRA‐V8 SH3BP5 LIFR FSCN1 ZFP318 FAAH TMEM38B GNE FAM46C GM1502 TCF7 CTSB SCRN1 CXX1C 2010007H06RIK A130082M07RIK BMX SERINC5 TMEM123 BEND5 GM13926 GRK5 SEMA4B FAM190A LOC674190 ITK WLS RUNDC2A ETV3 TNFRSF13C GM26 CTSD ZFP521 ROGDI IGK‐V19‐20 GM13959 C130050O18RIK LY6C1 GPR124 GM10883 DGKA CDS2 PLXDC1 SIGLECG IGHG SIDT1 PYGL CDS1 IL21R GM4964 KLK8 DUSP3 MAN2A2 SEMA7A PGAP1 LPHN1 PILRB2 BCR CACNB3 LOC382693 IL27RA TLR4 SRGAP3 MREG PXK ZAP70 FXYD5 KDR TMEM231 CD22 C920008G01RIK NLN B3GALNT1 FCHSD2 BRWD1 GM13893 TPPP3 LRRC16A SPIB FFAR1 G6PDX ANGPTL7 RELB CCR6 CYP2AB1 5730403B10RIK KCNIP3 GM10880 ARHGEF3 SLCO4A1 KMO IGL‐V1 TSPAN14 PACSIN1 BASP1 VPREB3 IRAK3 DIRC2 TLR11 DCLK2 NKIRAS2 DUSP9 H2‐M2 CDKL1 PRDX5 CXXC5 SLC14A1 GM16970 PGD FIGF GALNT12 LOC640614 PAG1 SPNS3 TSPAN3 FCRL1 NFIL3 IFI44 JAK2 DAF2 GRINA KDM1B TBC1D4 AY498738 ASPH RGP1 CXCL16 BACH2 AGPAT2 ZDHHC13 41153 GM1418 NRP1 STOML1 PLEKHA5 GM8760 NAIP2 PTPRS ICAM1 GM16848 ATG7 ZC3H12B VCAM1 SBK1 P4HA1 CALCOCO1 DEXI CC2D2B LDLR PTPRF PLXNC1 CXCR5 GZF1 PHF17 SLAMF8 KLHL14 HGSNAT GPR52 FRMD5 ELL3 DRAM2 A130050O07RIK FAM40B BC006779 RAB31 RABGAP1L MTMR4 4930420K17RIK ZSWIM6 ABCA5 SH3PXD2A PARP1 PLD1 FOXRED2 MYCL1 SH3BP5 TNFRSF21 CACNA1E RAB11FIP4 GM9861 MOCS1 TMEM163 BIRC2 SNN LRRC8D ZFP599 PFKFB3 MAPK11 ALDOA HMGN3 SYVN1 TMEM50A ZFP658 TNFRSF13B ATP6AP1 CLEC4G Dataset S2. Transcriptomic signatures of mouse cell types CDC_VS_MYELOÏD CD8α DC B T MYELOÏD_VS_CDC PDC SRPK3 LGMN MCTP2 TTPAL XBP1 NUCB2 CABLES2 LAMP2 AP1M2 PRKCE FNDC3B SLC29A3 UQCRB RHOQ FAM125A ITSN2 LRRC8B FADS2 PHXR2 TNFRSF1A LY6A GGA2 MEGF9 RPGRIP1 FCHSD2 CYBB ADAM11 CPM GSTM3 ATP2A1 GM7016 DGKA TOM1 SLC4A3 ZFP810 LOC641089 HMGCS2 IL12A LEFTY1 GM459 GM1965 SUSD1 GPR162 HCRTR2 ATP1B1 PHTF2 RHOBTB2 GM10759 ZC3H12C POU2F1 EPHB2 TEC SLC25A12 SNX25 IPO9 PLEKHA2 PHLPP2 EML4 GM6498 DMXL1 ATP13A2 GM10877 FGGY IGK‐V21‐2 LYNX1 FANCM CDH5 LOC100046275 OLFR164 AI324046 RBM15B BIN1 PGLS GM5574 CD209A FOXP1 OLFR166 GDF11 SLC39A14 SWAP70 SLA2 FCRL5 TXNDC5 D730005E14RIK PPIF 6430601O08RIK PGAM2 H2‐OB PTAR1 ETNK1 ZDHHC14 GM9900 RPS6KA1 LOC100047070 SGCB 9530009G21RIK DLL4 RFTN2 LAIR1 AFF3 PPM1E NUP160 LOC625360 SLC9A7 ARHGEF6 HIP1R LDOC1L ZFP882 KLHDC5 DONSON PLEKHM3 ACSF2 NOTCH3 N4BP2 SLC35E2 SNX8 GM10561 ARPC5L UBLCP1 V165‐D‐J‐C PECAM1 IGL‐V2 TRAF5 SMCHD1 DGKD GRPEL2 4930523C07RIK RMI1 BTK MED13 KCTD3 PTP4A3 SNX2 TFAM HVCN1 TRIM59 TXNDC16 PIKFYVE INTS4 LMBRD1 SERPINI1 ZMYM5 RASGRP3 TRIM7 MIR103‐2 Dataset S2. Transcriptomic signatures of human cell types CDC_VS_MYELOÏD BDCA3 DC B T MYELOÏD_VS_CDC PDC RP11‐126K1.6 RP3‐522D1.1 RP11‐421L21.3 CHRM3‐AS2 CTSS RP11‐305O6.3 ARL5A RP11‐31F15.1 RP11‐693J15.4 LINGO4 CYP1B1 SUZ12P1 MRPL9 RP11‐13L2.4 TBC1D27 RORC GNS CCS MAGEF1 PPM1H RP11‐160E2.11 LRRN3 RBM43 RP11‐214O1.2 FBL ST7 ANKRD36BP2 RTKN2 CARD6 MINA HLA‐DPB2 GCET2 IGKJ5 TRAT1 C5AR1 COBLL1 SNRPA CCDC74A IGKV1‐13 RP11‐941F15.1 LAMP1 PCBP1‐AS1 GRIP1 CDCA7 IGHV3‐7 NR3C2 C15ORF23 THUMPD2 NDUFV1 KIAA1958 IGHV3‐9 TNFRSF25 DYX1C1‐CCPG1 ERCC1 LEPREL1 CADM1 IGHV3‐20 MAL SCARB2 MDFIC RPL37 CAMK2D IGHV3‐21 TMEM116 OAS1 LEPREL1 SPNS3 OSBP2 IGHV4‐59 TRAV13‐2 SLC31A2 AJAP1 SCN9A ENPP1 IGHV4‐61 GZMK NPC1 CCDC50 SRSF8 XCR1 P2RX5 TRAC CTSL1 SCN9A OXCT1 FLT3 SLC38A11 ZNF204P MS4A7 KPTN VARS WDFY4 SCN3A RP11‐664D1.1 OSTM1 DPPA4 IMPA2 SNX3 IFT57 IL7R RASA1 CUX2 ADAM28 GPER IGLV1‐51 RP11‐23P13.6 HMOX1 TTTY15 ANKRD65 KIF16B IGLV1‐47 PASK ERVK13‐1DERL3 SKP2 MTERFD3 IGLV3‐25 INPP4B CLIP4 SLC2A11 PAFAH1B3 SEPT3 IGLV2‐14 CCDC64 CTSD C5ORF34 PDCD2L PTK2 IGLV3‐10 CD3G TMTC1 AP000350.10 DEPTOR LYRM4 IGLV2‐8 MIAT IFIT3 PAIP1 AFF3 FARS2 IGLL5 DOCK9 MAFB RP4‐647C14.2 SPINT2 CYP2E1 IGLC1 FAM84B LAIR1 WDR52 P2RY14 ZNF662 IGLC2 FAM171A1 GTF2E1 SIDT1 KCNK6 NET1 IGLC3 MORC2 GLUL SNHG14 MAP4K1 ENOX1 VPREB3 PIK3IP1 RNASEL SIVA1 DUSP2 KCND3 IGKC FAM102A NPL SERPINF1 UNC119B ZNF711 IGKV4‐1 TMEM204 SLC22A15 PLD4 IL18R1 CSRP1 POU2AF1 RASGRF2 CYBB AC006276.7 PPA1 NAV1 IGKV1‐17 CDR2 DUSP6 CADM4 CD72 PTPLB IGKV3‐20 CXCR6 RNF219 KIAA0226 ASPHD2 MFSD2B IGKV2‐28 LEF1 CTSB GAB1 ZNF487P FKBP1B SNX29P1 CD6 AZI2 SEC61A2 KLHL22 IDO1 IGKV1‐37 RP11‐18H21.1 SLC11A1 AC007254.3 CDH1 CD38 IGKV1‐39 CD5 SEPT10 TCL6 ZNF256 GPR126 IGKV1D‐39 C14ORF64 SOWAHC SCAMP5 GMDS C1ORF54 IGKV1D‐37 BCL11B C3AR1 SLC33A1 NAPSB BATF3 IGKV2D‐28 SCML4 LAMP2 UGCG TMEM109 PARP3 IGKV1D‐17 NELL2 TMEM106B RP11‐313P13.3 FLT3 KIT IGKV1D‐13 SIRPG CR1 RP11‐511P7.2 SUB1 C8ORF47 C15ORF57 THEMIS PSAP RP11‐313P13.4 SLC38A1 DBN1 BTNL9 LRIG1 CITED2 IL18R1 UBFD1 BEND5 ANKRD36B FLT3LG TRIQK SEPHS1 FBLN2 NEGR1 CD79A NOSIP CCPG1 PMS2P5 POLD2 OSBPL9 AC016745.2 ATP8B2 HCAR2 DSN1 CRIP3 KATNA1 METTL8 ASF1A HCAR3 AC138783.12 FAM60A ASB2 CD22 PLCG1 ASAH1 PMS2P10 FAM125B CLNK ZBTB32 ZNF101 HAL AC004878.3 CSNK1E FAM135A IGHA2 RP11‐5N23.2 ELL2 PMS2P2 IMPDH2 ACSL5 IGHG2 LY9 ATP6V1B2 PMS2P3 SH3BP1 CCDC74B IGHA1 UBASH3A RAB10 XXbac‐BPG300A18.12 PDXP POLA2 IGHG1 CD28 FBXO8 EGFL8 C22ORF32 CPNE3 IGHD RNF157 TLR4 DTX2P1 CST3 TLR3 IGHM PLEKHB1 ABCC3 DTX2P1‐UPK3BP1‐PMS2P11 PHF10 TMEM14A KIAA0125 B3GALT2 AQP9 RP11‐64P12.8 RP11‐72I8.1 CD24P2 CFH KIAA0232 PMS2P11 TYK2 IGHV4‐4 CD2 GPR137B PMS2P9 SLC4A3 RP11‐358M11.2 CFHR1 SASH1 SLC32A1 C1QBP AC012065.7 CAMK4 SLFN11 AC074183.4 CTNNBIP1 AC104699.1 FAM153B DOK3 TPM2 C5ORF20 TTN DGKA ZNF438 NLRP7 ITM2C XXbac‐B461K10.4 TRIM59 DMXL2 ATP13A2 TIFAB IGHGP HIVEP2 ACSL1 ZNF521 GS1‐465N13.1 AL928768.3 GPRASP1 SNX24 HIGD1AP14 AUTS2 LL22NC03‐80A10.6 INADL GIMAP8 CRYM DANCR GUSBP11 OXNAD1 GIMAP4 SMPD3 RP11‐738E22.2 AP000347.4 TRBC2 GIMAP5 IRF4 LYRM4 IGLL3P ACSL6 GBP2 FAM221B ACAP1 CD72 STMN3 TCF7L2 VIMP ARL4C IGHV5‐78 VSIG1 APH1B PLAC8 SSR1 ANKRD36 FAM153A KLHL9 ST6GALNAC4 CYP2E1 TMED8 GIMAP5 CKAP4 CIB2 PM20D2 KLHL14 TCF7 KCNJ2 MOXD1 RUFY3 CD24P4 IDH3A AP1G2 MS4A1 PPP1R14BP3 WDR41 QRSL1 RP11‐291L15.2 TOP1MT PRICKLE1 WASF1 HTR7 MEF2C KCNA5 ETV3 CPNE5 RP11‐164P12.4 NET1 WASIR2 RP11‐164P12.3 OFD1 PTPRK CYP46A1 RAB39B MACROD2 TCF4 CLIC2 CTD‐2576D5.4 RP11‐395I6.3 Dataset S2. Transcriptomic signatures of human cell types CDC_VS_MYELOÏD BDCA3 DC B T MYELOÏD_VS_CDC PDC NEK3 SLC6A16 GPR114 NFATC2 PAX5 PLCB4 PARM1 COL4A4 LAMP5 CNN2 COL4A3 TASP1 FAM46C NT5E RP11‐1280I22.1 AZI1 ZNF439 RP11‐20I23.8 GNG7 ZCCHC7 RP11‐248C1.2 NAV1 ABCB4 ACAD11 PTPLB SP140 NPHP3 TRIT1 WASIR1 MYB MYCL1 CNR1 CYBASC3 NREP CTA‐250D10.23 AC093642.3 IDO2 PKIG OBSL1 SEPT6 FCRL5 ENTPD3 SPECC1 FCRL2 AHI1 ZNF789 FCRL1 EPHB1 CHD1L AIM2 CEP128 EIF2D EML6 SNHG7 IFT20 PPAPDC1B ZFAT GPR126 FCRLA VIPR2 ITGB7 SYBU ASIP UVRAG RASSF6 KCNK10 HLA‐DOB EAF2 ITM2C XXbac‐BPG246D15.9 SLC15A2 AR GPR125 PCDH9 ZNF175 CTC‐524C5.2 RALGPS2 CYYR1 TRAF4 MICAL3 AC006978.6 ACSF2 BLK SLC12A3 HOXA9 DEFA4 SCARA5 RP1‐170O19.20 SWAP70 FAM129C HLA‐DPA1 CXCR5 STAMBPL1 BZW2 EBF1 KCTD19 GPR160 CR2 HIGD1A KIT OSBPL10 FZD3 CCNB1IP1 HIP1R LRRC36 BEND6 TEX9 PPM1K RAVER2 DOK7 N4BP2L1 TXNDC16 CSGALNACT1 UBE2J1 HDAC9 BANK1 C6ORF170 CIITA LCN6 KCNK17 PLEKHA5 LCN10 RBPMS APEX1 RAB30 LILRA4 PCNXL2 CNTNAP2 MYBL2 FKBP14 IL7 TOX2 ZBTB46 TNFRSF17 OGT BEND5 TPD52 CXCR3 NEGR1 MBD4 AEBP1 BZRAP1 SGCE SH3BGR NDRG2 STAP1 ZDHHC17 BIN1 CD19 SLC37A1 ZNF512B PEG10 TLR7 IGFBP7 SNX25 OFD1 EZH2 SNX29P2 RECQL5 ZEB1 GEN1 GPM6B C11ORF2 CD79B NFX1 TM7SF2 BEND4 CYSLTR1 ZBTB10 D87015.1 PAPLN N4BP2 MIR650 FCHSD2 MCOLN2 HSA‐MIR_5195 ABI2 REPIN1 PLXNA4 PON2 BCL11A DPH5 THSD1 SNORA26 MTHFD2L MIR4700 PHEX RASD1 DACH1 SRPX CLN8 PDIA5 SLITRK5 CA8 PPP1R14B CLEC4C SHD NREP PTPRS NUDT17 RIMS3 RBMS3 BSPRY ZNF789 NIPA1 BLNK MAPKAPK2 PFKFB2 Dataset S2. Transcriptomic signatures of human cell types CDC_VS_MYELOÏD BDCA3 DC B T MYELOÏD_VS_CDC PDC NOTCH4 WNT10A GPX7 RP11‐192H23.4 HHAT C7ORF31 POLB SGK494 VASH2 APLF NEK8 TRAF4 CNP SLC7A5 USP24 PLP2 C9ORF91 ABHD15 TP53I13 PARP10 TSPAN13 CARD11 SEL1L3 CLCN5 BEND6 MYL6B MAGED1 GRAMD1B MAGED4B GALNT2 SUZ12 MAGED4 TLR9 RP11‐330H6.5 KIAA1984 RAB15 ADC SOX4 SUPT3H SLC35F3 TARBP1 RUNX2 FAM81A ERO1LB AACS SSR4 CUEDC1 FLNB DUSP5 TMEM132B SLC30A5 SLC15A4 GPLD1 TTC39A RGS7 OPN3 ALDH5A1 PTGDS KTI12 FUBP1 PROC TNFRSF21 HNRNPA1 CD2AP KLHL13 EP400NL NFATC2IP COL24A1 SMC6 DCPS NGLY1 KIRREL3 PACSIN1 NUCB2 ZDHHC4 IRF7 SEMA5A EFHC1 C12ORF75 SLC35A3 SDK2 MIR4668 SNORA43