Ptk7-Deficient Mice Have Decreased Hematopoietic Stem Cell Pools As a Result of Deregulated Proliferation and Migration

Ptk7-Deficient Mice Have Decreased Hematopoietic Stem Cell Pools as a Result of Deregulated Proliferation and Migration

This information is current as Anne-Catherine Lhoumeau, Marie-Laure Arcangeli, Maria of September 24, 2021. De Grandis, Marilyn Giordano, Jean-Christophe Orsoni, Frédérique Lembo, Florence Bardin, Sylvie Marchetto, Michel Aurrand-Lions and Jean-Paul Borg J Immunol 2016; 196:4367-4377; Prepublished online 18

April 2016; Downloaded from doi: 10.4049/jimmunol.1500680 http://www.jimmunol.org/content/196/10/4367

Supplementary http://www.jimmunol.org/content/suppl/2016/04/16/jimmunol.150068 http://www.jimmunol.org/ Material 0.DCSupplemental References This article cites 55 articles, 24 of which you can access for free at: http://www.jimmunol.org/content/196/10/4367.full#ref-list-1

Why The JI? Submit online. by guest on September 24, 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 © 2016 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Ptk7-Deﬁcient Mice Have Decreased Hematopoietic Stem Cell Pools as a Result of Deregulated Proliferation and Migration

Anne-Catherine Lhoumeau,*,†,‡,x Marie-Laure Arcangeli,†,‡,x,{,1 Maria De Grandis,†,‡,x,{ Marilyn Giordano,*,†,‡,x Jean-Christophe Orsoni,*,†,‡,x Fre´de´rique Lembo,*,†,‡,x Florence Bardin,†,‡,x,{ Sylvie Marchetto,*,†,‡,x Michel Aurrand-Lions,†,‡,x,{ and Jean-Paul Borg*,†,‡,x

Hematopoietic stem cells (HSCs) located in adult bone marrow or fetal liver in mammals produce all cells from the blood system. At the top of the hierarchy are long-term HSCs endowed with lifelong self-renewal and differentiation properties. These features are controlled through key microenvironmental cues and regulatory pathways, such as Wnt signaling. We showed previously that PTK7,

a tyrosine kinase receptor involved in planar cell polarity, plays a role in epithelial Wnt signaling; however, its function in hema- Downloaded from topoiesis has remained unexplored. In this article, we show that PTK7 is expressed by hematopoietic stem and progenitor cells, with the highest level of protein expression found on HSCs. Taking advantage of a Ptk7-deficient mouse strain, we demonstrate that loss of Ptk7 leads to a diminished pool of HSCs but does not affect in vitro or in vivo hematopoietic cell differentiation. This is correlated with increased quiescence and reduced homing abilities of Ptk7-deficient hematopoietic stem and progenitor cells, unraveling novel and unexpected functions for planar cell polarity pathways in HSC fate. The Journal of Immunology, 2016, 196: 4367–4377. http://www.jimmunol.org/ ematopoiesis is a biological process consisting of the particularly in the liver, during its earliest phases. Long-term production of all blood cell types from hematopoietic HSCs are characterized by their capacity to provide lifelong H stem cells (HSCs) located in the bone marrow (BM) of reconstitution of all blood cell lineages after transplantation into adult mammals. Fetal hematopoiesis occurs in different sites, lethally irradiated recipients, whereas short-term HSCs do so for only 8–10 wk. In the mouse, long-term HSCs belong to the LSK cell com- *Centre de Recherche en Cance´rologie de Marseille, Polarite´ Cellulaire, Signalisa- 2/low + + tion Cellulaire et Cancer “Equipe labelliseé Ligue Contre le Cancer,” INSERM, partment, as defined by the Lin Sca-1 c-Kit phenotype, U1068, Marseille F-13009, France; †Institut Paoli-Calmettes, Marseille F-13009, irrespective of their fetal liver (FL) or adult BM origin. Additional ‡ x

France; CNRS, UMR7258, Marseille F-13009, France; Universite´ d’Aix- by guest on September 24, 2021 { markers, known as signaling lymphocytic activation molecules Marseille, Marseille F-13284, France; and Centre de Recherche en Cance´rologie de Marseille, Molećules d’Adhe´rence Jonctionnelles dans Host/Tumeur Interactions, (SLAMs), can be used to subdivide this compartment (1, 2). In- INSERM, U1068, Marseille F-13009, France deed, CD150+CD482 LSK cells contain ∼40% of HSCs with 1Current address: INSERM U967, Institut de recherche en radiobiologie cellulaire et long-term reconstitution potential in BM, as well as in FL (3). molećulaire – Commissariat a` l’eńergie atomique, Fontenay-aux-Roses F-92265, France. Hematopoiesis is a highly controlled multi-step process that relies on complex interaction networks involving cell surface re- ORCIDs: 0000-0001-6465-8723 (M.D.G.); 0000-0002-8361-3034 (M.A.-L.); 0000- 0001-8418-3382 (J.-P.B.). ceptors, growth factors, and adhesion molecules expressed by Received for publication March 23, 2015. Accepted for publication March 11, 2016. HSCs and their environment (4–8). Because HSCs generate mature This work was supported by Institut National du Cancer (INCa) (Projet Libre INCa hematopoietic cells, including immune cells, their replacement 2012-108), the Site de Recherche Inte´greé en Cance´rologie program (INCa- must be adjusted to homeostatic or stress conditions, such as in- Direction geńe´rale de l’offre de soins-INSERM 6038), and Canceropole Provence- fections, inflammation, or blood loss, and their expansion must be Alpes-Coˆte d’Azur (all to M.A.-L. and J.-P.B.). A.-C.L. and M.-L.A. were supported by la Fondation pour la Recherche Me´dicale. M.D.G. was supported by la Fondation controlled to avoid exhaustion. This control is made possible by de France. J.-P.B.’s laboratory is supported by INSERM, Institut Paoli-Calmettes, the coordinated regulation of quiescence, self-renewal, and dif- and La Ligue Nationale Contre le Cancer (“Equipe labelliseé”). J.-P.B. is a scholar at ferentiation through appropriate signals delivered by functional the Institut Universitaire de France. BM niches (9). HSCs are retained in these BM niches by cell A.-C.L. designed experiments, generated deficient mice, performed all experiments, and wrote the manuscript. M.D.G., M.G., F.L., F.B., and S.M. helped to perform surface molecules endowed with adhesive and/or signaling experiments. J.-C.O. generated deficient mice. M.-L.A., M.A.-L., and J.-P.B. de- functions expressed by HSCs. These receptors belong to different signed experiments and wrote the manuscript. protein families, including integrins (VLA-4) (10), Ig superfam- Address correspondence and reprint requests to Dr. Jean-Paul Borg, Centre de Re- ily adhesion molecules (6, 11), G protein–coupled receptors cherche en Cance´rologie de Marseille, Inserm, CNRS, Aix-Marseille Universite´, Institut Paoli-Calmettes, 27 bd Leı¨ Roure, BP 30059, 13273 Marseille cedex 09, (CXCR4) (12), and tyrosine kinase receptors (TIE2, c-Kit) (13– France. E-mail address: [email protected] 15), which interact with ligands present within the BM microen- The online version of this article contains supplemental material. vironment (16). The discovery of novel molecules implicated in Abbreviations used in this article: BM, bone marrow; CLP, common lymphoid pro- this multi-step program is of particular importance to grasping its genitor; DN, double negative; DP, double positive; E, embryonic day; EPO, erytho- complexity, to developing new strategies aimed at the regeneration poietin; ES, embryonic stem; FL, fetal liver; HSC, hematopoietic stem cell; HSPC, hematopoietic stem and progenitor cell; MPP, multipotent progenitor; MPP1, MPP of damaged hematopoietic tissues, and to understanding hemato- type 1; MPP2, MPP type 2; MPP3, MPP type 3; MPP4, MPP type 4; PCP, planar cell poietic diseases. polarity; RFI, ratio fluorescence intensity; SDF-1a, stromal-derived factor-1a; PTK7 is a planar cell polarity (PCP) receptor belonging to the Ig shRNA, short hairpin RNA; SLAM, signaling lymphocytic activation molecule. superfamily, and it plays important roles during development (17, Copyright Ó 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 18). PCP is controlled by a noncanonical Wnt pathway, organizes www.jimmunol.org/cgi/doi/10.4049/jimmunol.1500680 4368 ROLE OF THE POLARITY PROTEIN PTK7 IN MURINE HEMATOPOIESIS polarization of many epithelial tissues and organs within the filtered using a 70-mm cell strainer (BD Falcon), and lysis of RBCs was plane, and drives cell migration and cell intercalation of non- done using ACK lysis buffer (0.15 M NH4Cl, 10 mM KHCO3, 0.1 mM epithelial cells, such as mesenchymal cells (19, 20). Ptk7-deficient Na2EDTA; Life Technologies) for 2 min at room temperature. Cells were spun and resuspended in PBS–2% FCS, with or without 0.125 mM EDTA mouse embryos show all signs of typical PCP defects, such as for staining. For blood samples, RBC lysis was done after staining using severe neural tube and abdomen closure defects with secondary BD FACS Lysing buffer solution (BD Biosciences), following the manu- development abnormalities (21). PTK7 is a type I protein com- facturer’s instructions. posed of seven Ig loops in its extracellular region, a transmem- For flow cytometry analysis and cell sorting, cells were stained with mAbs purchased from eBioscience or BioLegend against the following brane domain, and an intracellular region with a tyrosine kinase molecules: B220 (clone 6B2), CD3 (145-2C11), CD4 (RM4-5), CD8 (53-6.7), domain (17, 18, 22, 23). Although PTK7 is catalytically inactive, CD11b/Mac-1 (M1/70), CD11c (N418), CD16/32 (93), CD19 (6D5), CD24 the cytoplasmic domain containing the tyrosine kinase domain is (M1/69), CD34 (RAM34), CD45 (30F11), CD45.1/Ly5.1 (A20), CD45.2/ mandatory for receptor function in vitro and in vivo (24–27). Both Ly5.2 (104), CD48 (HM48-1), CD49b (DX5), CD49d (R1-2), CD117/c-Kit extracellular and intracellular domains of PTK7 can be released (2B8 and/or ACK2), CD135 (A2F10), CD150 (TC15-12F12.2), F4/80 (BM8), Gr1 (RB6-8C5), Sca-1 (D7), and Ter-119. Human PTK7 Ab and glyco- from the plasma membrane following the sequential cleavage of the phorin A (11E4B-7-6) were purchased from Miltenyi Biotec and Beckman receptor by membrane type 1-matrix metalloproteinase/ADAM-17 Coulter, respectively. and g-secretase, interfering with its PCP and promigratory activ- All experiments were done by multicolor parametrics analysis using ities (24, 28, 29). FITC, PE, allophycocyanin, PE-Cy7, allophycocyanin-Cy7, PerCP-Cy5.5, Alexa Fluor 700, Alexa Fluor 594, Pacific Blue, PE-Cy5, and Q-dot 605 We recently found that PTK7 expression is not restricted to fluorochroms or spectral equivalents. Commercial Abs were either directly epithelial cells. Indeed, PTK7 is expressed at the cell surface of conjugated or biotinylated. For mouse PTK7 expression, polyclonal rabbit human hematopoietic progenitors and blast cells from patients with Ab raised against recombinant PTK7-Fc was used in the experiments. For Downloaded from acute myeloid leukemia. Overexpression of the PCP protein leads detection of PTK7, PE- or allophycocyanin-conjugated goat anti–rabbit to increased cell survival and cell migration of leukemic cells and (Jackson ImmunoResearch Laboratories) was used as a secondary Ab. Streptavidin conjugated with Alexa A594 (Invitrogen) or Pacific blue confers a poor prognosis to patients independently of other risk (eBioscience) was used to exclude Lin+ cells. Irrelevant isotype-matched factors (27). Similar conclusions were recently drawn from studies Abs were used as controls. Cells were stained for 20 min at 4˚C and rinsed done in chronic lymphocytic leukemia in which a set of PCP in PBS–2% FCS solution. Viability marker (vivid dye; Invitrogen) was added during the last staining in PBS solution. Stained cells were analyzed proteins is upregulated (30). http://www.jimmunol.org/ by an LSR II SORP (lasers of 405, 488, 561, and 633 nm; BD Biosciences) Although the functions of the canonical and noncanonical Wnt or sorted using a FACSAria III (BD Biosciences). LSK cell populations pathways have been studied extensively in hematopoiesis were purified by cell sorting using a FACSAria (.95% purity). Results (reviewed in Refs. 31–33), the role of PTK7 has never been ex- were analyzed using BD FACSDiva software version 6.1.2 (BD Biosci- plored in this process. In this study, we addressed the function of ences) or FlowJo software version 7.6.2 (TreeStar). PTK7 in physiological hematopoiesis using a mouse model defi- Homing experiments cient for ptk7 expression. Using flow cytometry, we first mapped the expression of PTK7 on HSCs and progenitors. Because PTK7- E14.5 embryos were sacrificed and immediately genotyped. FL cells from wild-type or gene-trap mice were suspended, counted, and stained with cell deficient mice die perinatally, we then studied the function of tracker Calcein, AM and Calcein Red-Orange (both from Invitrogen), re- PTK7 in fetal hematopoiesis and found fewer hematopoietic stem spectively, according to the manufacturer’s instructions. Cells were mixed by guest on September 24, 2021 and progenitor cells (HSPCs) in FL of Ptk7-deficient mice. Al- at a 1:1 ratio, and 4 3 106 cells were injected through the retro-orbital vein though HSC differentiation was not affected, we found a dramatic in previously irradiated (8 Gy) mice. Twelve hours after injection, mice decrease in the proliferation and migration properties of Ptk7- were sacrificed, and blood, BM, spleen, liver, and thymus samples were analyzed for fluorescent cells. deficient HSPCs that revealed novel unexpected functions for this PCP molecule. Transplantation assays For LSK sorting, FL cells (CD45.2+) were labeled for Lin+ cells with the following biotinylated primary rat Abs (lineage mix): CD3, CD4, CD8, Materials and Methods + Mice CD19, B220, DX5, TER-119, GR-1, CD11c. Lin cells were depleted by magnetic separation using anti-rat Ig-coated magnetic beads (DynaBeads; 2 C57BL/6 and C57BL/6-CD45.1 mice were purchased from JANVIER Life Technologies). Lin cells were also stained and sorted by FACS, 2 LABS and Charles River Laboratories, respectively. All mice used for trans- according to the phenotype Lin CD45.2+c-Kit+Sca-1+. CD45.1 mice were plantation experiments were 6–12 wk old. Lethal irradiation was done with an lethally irradiated and grafted with 1000 or 400 CD45.2+ LSK cells X-ray irradiator (X-RAD 160 X-Ray, PXi) using a single dose of 8 Gy. through the retro-orbital vein, along with 1 3 105 total BM recipient cells PTK7 gene-trap mice were obtained by injection of gene-trap ES to allow short-term reconstitution and mice survival. Recipient mice sur- XST087 (Bays Genomics) into embryonic day (E)3.5 blastocysts from vival was followed, and BM engraftment was examined by monitoring C57BL/6 mice. Chimeric mice were back-crossed with B6 mice for seven donor cell (CD45.2+ cells) and host cell (CD45.1+ cells) frequency in the generations to obtain a homogenous genetic background for experiments. blood. After lethal irradiation, mice were maintained on water containing All mice were housed and maintained under specific pathogen–free con- neomycin sulfate (1 g/500 ml) for 2 wk. ditions. E14.5 and E15.5 mice were used in all FL experiments. All animal experiments were performed in agreement with the French Guidelines for Serial transplantation assay Animal Handling. Lethally irradiated (8 Gy) mice (C57BL/6, CD45.1) received 3 3 106 total Genotyping FL cells (first transplantation) or BM total cells (subsequent transplantations). BM engraftment was monitored as described above. Genotyping of adult mice and embryos was performed by genomic PCR using the following pairs of primers: 59-ACGTCACTGGAGAGG- Competition assay AAGTCCGCAG-39 and 59-GATCGAAGCAGGTCCTGTGGTCCT-39 for wild- 6 Lethally irradiated (8 Gy) mice (C57BL/6, CD45.1) received 2 3 10 total 9 9 9 2/2 6 type allele and 5 -TTGTGAAGTGGAGGCTCCGGACC-3 and 5 -CGATCT- FL cells from CD45.2 PTK7 mutant embryos and 2 3 10 total FL 9 + + +/+ TCCTGAGGCCGATACTGTC-3 for gene-trap insertion allele. cells from CD45.1 CD45.2 PTK7 embryos, issues from crossing Flow cytometry and cell sorting CD57Bl/6, CD45.1 and CD57Bl/6, CD45.2 mice. Mouse plugs are obtained the same day. BM engraftment was examined by monitoring the For the analysis of HSPCs in FLs, FLs were dissected from E14.5 embryos, frequency of CD45.2+ cells from PTK7 mutant donors and CD45.1+ and a single-cell suspension was made in PBS–2% FCS. CD45.2+ cells from wild-type donors, excluding CD45.1+ cells from host For analysis of HSPCs in BM, thymus, or blood, cells obtained from adult cells, in the blood. After lethal irradiation, mice were maintained on water mice were suspended in PBS–2% FCS. For BM and thymus, cells were containing neomycin sulfate (1 g/500 ml) for 2 wk. The Journal of Immunology 4369

Cell cycle analysis and apoptosis analysis Ab generated against the extracellular region of PTK7 was used. FL progenitors were enriched as previously described and stained with c- This Ab recognizes mouse PTK7 endogenously expressed on Kit–allophycocyanin–eFluor 750, CD150-allophycocyanin, Sca-1–PerCP– endothelial and hematopoietic cells by FACS analysis in wild- Cy5.5, and CD48-PECy7. The biotin-conjugated lineage combination was type, but not ptk7-deficient, mice (Supplemental Fig. 1). Analy- revealed with Alexa Fluor 594–streptavidin (Invitrogen). For cell cycle sis of PTK7 expression in peripheral blood and in splenocytes analysis, after extracellular staining, cells were permeabilized, fixed, and revealed that none of the mature mouse leukocyte subsets (gran- stained with anti–Ki67-FITC mAb (BD Biosciences) and DAPI, as previously described (34). For apoptosis analysis, cells were stained with PE– ulocytes, monocytes, B cells, and T cells) expressed PTK7 at Annexin V (BD Biosciences) to evaluate Annexin V externalization and detectable levels (Fig. 1A, data not shown). We next analyzed DAPI (Sigma-Aldrich). PTK7 expression in murine BM hematopoietic progenitors using In vitro methylcellulose assays multiparametric flow cytometry and the gating strategies described in Fig. 1B. When LSK subsets were examined, PTK7 For methylcellulose colony assays, FL or BM cell suspensions were pre- expression was inversely correlated with differentiation stages, pared in IMDM supplemented with 2% FBS. A total of 3 3 104 FL cells or 1 3 105 BM cells was seeded in MethoCult M3434 containing recombi- with the highest expression on HSCs and multipotent progenitor nant mouse SCF (50 ng/ml), recombinant mouse IL-3 (10 ng/ml), (MPP) type 1 (MPP1; ratio fluorescence intensity [RFI] = 40) and recombinant human IL-6 (10 ng/ml), and erythopoietin (EPO) (U/ml) intermediate expression on MPP type 2 (MPP2; RFI = 20) and (STEMCELL Technologies, Grenoble, France) in 35-mm dishes, accord- MPP type 3 (MPP3) and MPP type 4 (MPP4) (RFI = 15) (Fig. ing to the manufacturer’s instructions. All assays were done in triplicate. 1C). More committed progenitors, including common myeloid Colonies (.30 cells) were counted after 13 d of incubation at 37˚C, 5% progenitors and common lymphoid progenitors (CLPs), expressed CO2, and humidified atmosphere. PTK7 at lower levels, whereas the protein was almost undetect- Downloaded from Adhesion assays able on the surface of granulocyte-macrophage progenitors and MS5 stromal cells were plated on 96-well flat bottom plates, at a con- megakaryocyte-erythroid progenitors (Fig. 1D). A similar phe- centration of 5 3 103 cells/well, 1 d before the assay. MS5 is a stromal cell notypic analysis was conducted on differentiating T cells isolated line that was generated by irradiating adherent cells in long-term BM from the thymus. T cell progenitors do not express CD4 and CD8 culture. Lin2 FL cells (5 3 105), obtained as previously described, were cocultured for 2 h with MS5 stromal cells. Plates were washed three times markers and are called double-negative (DN) cells before differ- with PBS, and adherent cells were removed by strong pipetting in PBS entiation into double-positive (DP) and single-positive cells. http://www.jimmunol.org/ added with 1.25 mM EDTA and 2% FCS. The phenotype of adherent cells Within the DN compartment, four thymic progenitor populations was determined using anti-CD117, Sca-1, CD150, CD48, and FITC- 2 can be identified according to CD44 and CD25 expression from lineage mix to exclude any residual mature cells contaminating the Lin the most immature (DN1) to the most mature (DN4) subsets. preparation. Absolute numbers were measured using the BD High + 2 Throughput Sampler coupled to the LSR II SORP (BD Biosciences). PTK7 is expressed by the CD44 CD25 (DN1) population, and expression is conserved during thymic differentiation within the Migration assays DN2, DN3, and DN4 subpopulations (Supplemental Fig. 2). PTK7 Lin2 FL cells were enriched as previously described and suspended in expression is also detected on DP thymocytes before becoming IMDM containing 2% BSA. About 2 3 105 cells in 75 ml were loaded in progressively lost at the single-positive stage. Altogether, these a5-mm-pore Transwell system (Corning Life Sciences). The lower phenotypic analyses reveal that PTK7 expression is inversely by guest on September 24, 2021 chamber was filled with serum-free medium supplemented with the chemoattracting murine stromal-derived factor-1a (SDF-1a, 100 ng/ml; correlated with hematopoietic differentiation, with the highest PeproTech). The same number of cells was plated in distinct wells to be expression found in HSCs. This suggests that PTK7 may play a used as control for migration. Plates were incubated for 4 h at 37˚C, 5% role at this early stage of hematopoietic differentiation. CO . All experiments were performed in duplicates. Cells were also 2 2 2 stained using anti-CD117, Sca-1, and streptavidin coupled to Pacific Ptk7 / embryos have a quantitative defect in HSCs Blue, as described above. Absolute numbers were measured using the BD High Throughput Sampler coupled to the LSR II SORP (BD Bio- We next questioned the function of PTK7 in hematopoiesis. To this sciences), and the migration ratio was defined as the number of migrating end, we generated a Ptk7-deficient mouse strain using a gene- cells in migration wells/control wells. trapped embryonic stem (ES) cell available through the Interna- TF1 cell differentiation tional Gene Trap Consortium. As expected from the previous description of Ptk7-deficient strains, we observed a perinatal le- Growth of the human erythroid/macrophage progenitor cell line TF1 is thality resulting from profound developmental defects (21). dependent on GM-CSF or IL-3, and it can be induced to differentiate in the 2/2 presence of growth factors. TF1 cells were grown in IMDM supplemented Ptk7 embryos harbor a severe phenotype with cranial neural with 5 ng/ml GM-CSF and supplemented with EPO (2 UI/ml) for erythroid tube closure defect, gastroschisis, and kidney and forelimb abnor- differentiation. TF1 cells were infected by viral particles containing short malities (Fig. 2A, 2B). Considering this dramatic phenotype, it was hairpin RNA (shRNA) against PTK7, as previously described (27). A impossible to study adult hematopoiesis in these mice; thus, we scramble nontargeting shRNA was used as a control. shRNA efficiency for decided to study FL hematopoiesis. In FL, as well as in BM, HSCs PTK7 downregulation was controlled by flow cytometry before and at the end of experiments. TFI cells were then differentiated in IMDM with are included within the LSK cell compartment. Using the same 2 IU/ml EPO and without GM-CSF for a total of 7 d, as previously de- gating strategy as in Fig. 1, we analyzed the consequences of scribed (35). Erythroblastic differentiation was evaluated by glycophorin PTK7 deficiency in the different LSK cell populations (Fig. 2C). A expression. We observed a 25% decrease in total cellularity at E14.5 in PTK7- Statistical analysis deficient FL compared with wild-type FL (Fig. 2D). However, the proportion of leukocytes (e.g., CD45+ cells) was equivalent be- Statistical significance was determined with a nonparametric Mann– Whitney U test using Prism Version 5 software (GraphPad). tween deficient and wild-type FL (Fig. 2E), suggesting that the decrease in FL cellularity in Ptk72/2 embryos is likely due to a global effect of ptk7 deficiency on FL size. We next examined the Results hematopoietic progenitor compartment. A decrease of ∼50% in PTK7 is expressed on HSCs and myeloid, lymphoid, and thymic LSK cell frequency was observed in Ptk7-deficient FL compared progenitors with control FL (1.33% Ptk7+/+ and Ptk7+/2 LSK cells versus As the first approach, we studied PTK7’s pattern of expression in 0.67% Ptk72/2 LSK cells) (Fig. 2F). Of the different LSK cell adult murine hematopoietic cells. To this end, a rabbit polyclonal populations, the most affected one was the HSC-MPP1 subset 4370 ROLE OF THE POLARITY PROTEIN PTK7 IN MURINE HEMATOPOIESIS Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 1. PTK7 expression in wild-type BM. (A) Expression of PTK7 in blood mature cells. Granulocytes are defined as Gr1+CD11b+ cells. Monocytes are defined as F4/80+ CD11b+CD115+ cells. Lymphocytes are gated using B220, CD19, CD3, CD4, and CD8. Mean fluorescence intensity ratios are indicated. (B) Strategy of BM progenitors gating. Dot plot showing the LSK+ gate within Lin2 BM cells (middle left panel). Dot plot showing early progenitor gating using SLAM markers (left panel). HSCs are defined as CD150+CD482 cells. More mature MPPs are defined as CD150+CD48+ (MPP2) and then CD1502CD48+ (MPP3/MPP4). Dot plot showing engaged progenitors within c-Kit+Sca-12Lin2 cells (middle right panel). Common myeloid progenitors are defined as CD16/32loCD34+. Granulocyte-macrophage progenitors are defined as CD34+CD16/32hi (far right panel). Megakaryocyte- erythroid progenitors are defined as CD16/322CD342. CLPs are defined as Lin2IL7Ra+B2202CD24loCD117lo.(C) Line graphs showing expression of PTK7 during differentiation. Open graphs: PTK7; shaded graphs: isotype control (preimmune rabbit antiserum). (D) Quantitative comparison of expression levels of different progenitor populations: normalization of PTK7 with CD117 mean of fluorescence on LSK subpopulations and progenitors (CD117+Sca-12 cells). All data are mean 6 SD. **p , 0.01 nonparametric Mann–Whitney U test. ns, not significant.

(Fig. 2C, 2G). No significant difference in the frequency of MPP2 HSC number in Ptk72/2 embryos compared with controls, and MPP3-MPP4 subsets was observed in Ptk7-deficient FL whereas there was an ∼50% decrease in MPP2 and MPP3/MPP4 compared with wild-type or heterozygous FL. When we examined cell numbers (Fig. 2I). the absolute number of the different subsets of hematopoietic To assess the frequency of functional HSCs in FL, we trans- progenitors, we noticed that all LSK subsets were affected by planted 1000 and 400 sorted LSK cells from Ptk7+/+ or Ptk72/2 FL PTK7 deficiency. Indeed, there was a 68% decrease in the LSK bearing the CD45.2 allele into two groups of lethally irradiated cell number in Ptk7-deficient embryos compared with control mice bearing the CD45.1 allele. As expected, all mice included in embryos (Fig. 2H). The most dramatic decrease was observed in the group injected with 1000 LSK cells survived. However, when the HSC compartment. Indeed, there was up to a 75% decrease in 400 cells were injected, Ptk7+/+ FL LSK cells were able to re- The Journal of Immunology 4371 Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 2. Phenotype of PTK7-deficient mouse and analysis of fetal liver at E14.5. (A) PTK7 mutants embryos showing severe PCP and developmental abnormalities. (B) Western blot of murine embryonic fibroblasts obtained from a litter of our PTK7 gene-trapped established mouse line. (C) Dot plots showing LSK gate within Lin2 FL cells and SLAM gating within LSK cells. (D) Quantification of total FL cells. Data are pooled from six litters: Ptk72/2, n = 12; Ptk7+/2, n = 19; Ptk7+/+, n = 12. (E) Percentage of CD45+ cells within FL. Data are pooled from eight litters: Ptk72/2, n = 15; Ptk7+/2, n = 26; Ptk7+/+, n =9.(F) Percentage of LSK cells within CD45+ FL cells. (G) Percentages of LSK subpopulations using SLAM markers. (H) Absolute number of LSK cells in E14.5 FL. (I) Absolute number of LSK subpopulations using SLAM markers. Data in (F)to(I) are pooled from 10 litters: Ptk72/2, n = 19; Ptk7+/2, n = 29; Ptk7+/+, n = 10. (J) Survival curves of CD45.1 C57BL/6 mice engrafted with 400 LSK cells sorted from CD45.2 FL, together with 1 3 105 congenic CD45.1 total BM cells. Black line: engraftment with Ptk7+/+ FL cells; dashed line: engraftment with Ptk72/2 FL cells. Data shown are a representative of three separate experiments (n = 6 mice/group). (K) Clonogenic capacity of total FL cells. A total of 3 3 104 FL cells was seeded in M3434 methylcellulose. Data shown are a representative from three separate experiments. Ptk72/2, n =6;Ptk7+/+, n = 4. All data are mean 6 SD. *p , 0.05, **p , 0.01, ***p , 0.001 nonparametric Mann–Whitney U test. ns, not significant.

constitute recipient mice between 2 and 12 wk after injection, We found that a greater proportion of HSCs and early progeni- 2/2 whereas all mice injected with Ptk7 LSK cells, except one, tors were in the G0 phase in Ptk7-deficient embryos compared died within 6 wk (Fig. 2J). We concluded that loss of PTK7 leads with control. Indeed, 56% of HSC-MPP1 cells were quiescent in to a defect in functional HSC frequency. In vitro clonogenicity Ptk7-deficient embryos versus 46% in control embryos (Fig. 3B). assays confirmed the defect in HSCs and early precursor fre- This significant increase in quiescent cells was maintained quency because we observed a 2-fold reduction in CFU starting throughout hematopoietic differentiation, with 46 and 39% and 4 from 3 3 10 total FL cells (Fig. 2K). 19 and 14% of the cells in the G0 phase at the MPP2 and MPP3- MPP4 stages, respectively (Fig. 3B). To exclude that these un- PTK7-deficient FL cells are more quiescent than wild-type FL balanced proportions of quiescent cells were not due to selective cells apoptosis of cycling Ptk7-deficient cells, we analyzed the per- To determine whether the absence of PTK7 could alter HSC centage of apoptotic cells in Ptk7-deficient and control FL cells. quiescence, we performed cell cycle phenotypic analysis of LSK We did not observe a difference in apoptosis as measured by FL cells using an Ab directed against Ki-67, a nuclear marker to Annexin V staining between wild-type and mutant mice in total measure cell cycle, and DAPI to evaluate the DNA content (Fig. 3A). FL cells and LSK cells (Fig. 3C). These results demonstrate that 4372 ROLE OF THE POLARITY PROTEIN PTK7 IN MURINE HEMATOPOIESIS

FIGURE 3. Cell cycle and apoptosis analysis of LSK FL cells. (A) Repre- sentative dot plots of DNA content (DAPI) plotted versus Ki-67 nuclear Ag staining on LSK cells. The cell cycle 2 phases were defined as G0 (Ki-67 and + 2n DNA), G1(Ki-67 and 2n DNA), and + S-G2-M (Ki-67 and DNA . 2n). (B) Percentage of LSK FL cells in cell cycle phases using SLAM markers. Data are pooled from three separate experiments, and embryos (Ptk72/2, n = 10; Ptk7+/+, n = 11) are from seven litters. (C) Per- centage of apoptotic cells, as determined by staining with Annexin V and DAPI Downloaded from (Annexin V+DAPI2). Results are representative of eight Ptk72/2 and six wild- type embryos from five litters. All data are mean 6 SD. *p , 0.05, **p , 0.01 nonparametric Mann–Whitney U test. ns, not significant. http://www.jimmunol.org/

PTK7 delivers proliferative signals to HSCs but does not affect Recipient CD45.1 mice were lethally irradiated and reconstituted HSC survival. with CD45.2 donor cells from wild-type and mutant mice. The by guest on September 24, 2021 first injection contained 3 3 106 total FL cells. After 12 wk, BM PTK7-deficient cells are able to reconstitute definitive cells were harvested, and 3 3 106 cells were injected into a second hematopoiesis in recipient mice lethally irradiated recipient. The same protocol was applied for To investigate the functional capacity of Ptk7-deficient HSCs, we the following transplantations. We monitored chimerism by flow generated chimeric mice using FL cell transplantation. CD45.1 cytometry analysis and obtained a very good reconstitution with C57BL/6 mice were lethally irradiated (8 Gy) and then recon- ∼95–100% of donor cells in BM for all transplantations (Fig. 4H). stituted with 3 3 106 total FL cells from Ptk7-deficient or wild- We also investigated erythroblastic reconstitution in mice and did type FL embryos (CD45.2). Reconstitution and chimerism were not find any differences with regard to hemoglobin level and RBC followed over time by monitoring the frequency of donor cells count (Supplemental Fig. 3A, 3B). In vitro differentiation using (CD45.2) and recipient cells (CD45.1) in the peripheral blood of TF1 cells, which are able to differentiate in response to EPO, did recipient mice (Fig. 4A). As expected, all recipient mice injected not reveal any difference in the presence or absence of PTK7 with control FL cells were reconstituted between 2 and 12 wk. The (Supplemental Fig. 3C, 3D). Thus, we concluded that PTK7 is not same results were obtained with PTK7-deficient FL cells, dem- required for hematopoietic differentiation. onstrating that, in the absence of any competition, Ptk7-deficient To gain further insights into PTK7 function in early hemato- HSCs are able to reconstitute all mature hematopoietic lineages of poiesis, we performed mixed chimera competition experiments. lethally irradiated recipients (Fig. 4B, 4C). To investigate whether CD45.1 C57BL/6 mice were lethally irradiated (8 Gy) and then Ptk7 deficiency could lead to more subtle hematopoietic defects, reconstituted with 2 3 106 total FL cells from Ptk7-deficient BM of chimeric mice was analyzed after 12 wk. No significant embryos (CD45.2) and 2 3 106 total FL cells from wild-type FL difference was found in the frequencies of mature and progenitor embryos (CD45.23CD45.1). Reconstitution and chimerism were cells in the BM (Fig. 4E, 4F). In agreement with reconstitution followed over time by monitoring the frequency of PTK72/2 cells experiments, in vitro differentiation assays showed that Ptk7-de- (CD45.2) and PTK7+/+ cells (CD45.2+CD45.1+) versus recipient ficient FL cells are able to give rise to all subtypes of CFU in cells (CD45.1) in the peripheral blood of recipient mice. We did the same proportion as control FL cells (Fig. 4G, left panel, not observe significant differences at days 30 and 60, but we ob- Supplemental Fig. 4). However, we observed fewer colonies served a dominance of PTK72/2 cells in the long-term (.120 d) generated by Ptk7-deficient HSPCs than by control cells (Fig. 4G, (Fig. 5A). Because maintenance of quiescence is a stem cell right panel). This is likely due to decreased cell cycle activity of property, we tested cell cycle activity of adult HSCs in chimeric Ptk7-deficient HSPCs, whereas the potential of differentiation is mice. Recipient CD45.1 mice were lethally irradiated and recon- not affected. Because quiescence of HSCs could be correlated stituted with CD45.2 donor cells from wild-type and mutant FL with loss of self-renewal (36), we then tested self-renewal ca- mice, as previously described. We performed cell cycle analysis pacities of Ptk7-deficient HSCs by serial transplantation assays. on hematopoietic progenitors at .6 mo posttransplantation, when The Journal of Immunology 4373 Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 4. Hematological capacity of reconstitution and differentiation. Lethally irradiated CD45.1 mice (8 Gy) were reconstituted with 3 3 106 CD45.2 total FL cells isolated from Ptk+/+ and Ptk2/2 embryos. (A) The percentage of each hematological population from the donor (CD45.2) was monitored by flow cytometry. Percentage of mature cells in recipient blood (B) and in spleen (C). Percentage of mature BM cells (D) and BM progenitors (E) in recipient mice. (F) Percentage of LSK cells and LSK subpopulations in recipient BM. Open bars depict mice reconstituted with Ptk2/2 FL. Filled bars depict mice reconstituted with Ptk+/+ FL (n = 6 mice/group). (G) In vitro differentiation capacity of FL cells. A total of 3 3 104 FL cells was seeded in methylcellulose. Quantification and characterization of CFU in culture (CFU-C) (left panel). Proportion of CFU (right panel). ***p , 0.001 nonparametric Mann–Whitney U test. (H) Serial transplantation. First transplantation (T1): lethally irradiated mice were injected with 3 3 106 of total E14.5 FL cells. The subsequent transplantations (T2–T5) consisted of injecting 3 3 106 total BM cells from previously reconstituted mice into irradiated C57BL/6 CD45.1 mice. Percentages of donor cells (CD45.2) and recipient cells (CD45.1) were monitored. Data represents chimerism obtained after definitive reconstitution of hematopoiesis 3 mo after irradiation. Data are mean 6 SD (n = 6 mice/group). ns, not significant. stable hematopoietic reconstitution is obtained. We confirmed that Cells were injected at a 1:1 ratio, and mice were sacrificed 12 h hematopoietic progenitors isolated from mice reconstituted with after injection. Analysis of homing was performed by flow PTK7-deficient cells are more quiescent than their wild-type cytometry (Fig. 6A–C). Although we observed comparable levels counterparts (Fig. 5B, 5C). Clonogenic experiments confirmed of residual circulating cells, regardless of the PTK7 status, in that fewer colonies are generated by Ptk7-deficient HSPCs than by peripheral blood, Ptk7-deficient cells had a very poor ability to control HSPCs from chimeric mice (Fig. 5D). These results sug- home to the BM, spleen, liver, and thymus compared with control gest that PTK7 is essential for quiescence, independent of the (Fig. 6D). Because homing defects can be explained by the loss of microenvironment. cell adhesion to BM stromal cells and/or by a defect in cell migration toward a chemoattractant, we tested these two hypotheses. Homing and migration defects of Ptk7-deficient cells Adhesion of Ptk7-deficient LSK cells on the murine MS5 stromal Because we previously described a promigratory function of PTK7 cell line was not significantly different from control LSK cells in leukemic cell lines and because increased HSC quiescence may (Fig. 7A). In contrast, we found a decrease in spontaneous transmi- not be the unique cause of the functional defect in Ptk72/2 HSCs gration of Ptk7-deficient FL cells compared with control cells in a (Fig. 2J), we decided to investigate the role of PTK7 in hemato- Boyden chamber assay (8% versus 5%, Fig. 7B). No effect was ob- poietic homing. FL cells from several Ptk72/2 or Ptk7+/+ embryos served with heterozygous FL cells. We also observed significant dif- were pooled and stained separately with two fluorescent trackers. ferences when the chemokine SDF-1a was used as a chemoattractive 4374 ROLE OF THE POLARITY PROTEIN PTK7 IN MURINE HEMATOPOIESIS Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 5. Competition experiments and cell cycle in adult mice. (A)Atotalof23 106 cells from Ptk7+/+ (CD45.13CD45.2) mice and 2 3 106 cells from Ptk72/2 (CD45.2) mice were injected into recipient (CD45.1) mice. Chimerism was followed in blood samples. Data are representative of four independent experiments (n = 6 mice/group). (B–D) Recipient CD45.1 mice were reconstituted with CD45.2 donor cells from wild-type and mutant FL mice. (B)Rep- resentative dot plots of DNA contents (DAPI) plotted versus Ki-67 nuclear Ag staining on LSK cells. (C) Percentage of LSK cells in cell cycle phases using SLAM markers. (D) Clonogenic capacity of total BM cells. A total of 1 3 105 BM cells was seeded in M3434 methylcellulose. Data are representative of two independent experiments (n = 6 mice/group). *p , 0.05, **p , 0.01, ***p , 0.001 nonparametric Mann–Whitney U test. ns, not significant. molecule. Indeed, although 12% of control FL cells migrated resistance to chemotherapeutic regimens (27). Recently, another toward SDF-1a, only 8% of PTK7-deficient FL cells were able to study revealed that PTK7 is also upregulated in T acute lympho- do so (Fig. 7C). Without chemoattractant, ,5% of LSK cells cytic leukemia (37). In humans, we and other investigators showed migrated across the Transwell, regardless of the PTK7 status that PTK7 expression is primarily restricted to normal early pro- (Fig. 7D). When SDF-1a was used as a chemoattractive molecule, genitors committed to myeloid and T lymphoid lineages. a 2-fold decrease in LSK transmigration was observed for Ptk7- In the current study, we show that PTK7 expression is similar deficient cells compared with control cells (20% versus 40%, between mouse and human systems. PTK7 is absent in mature Fig. 7E). blood cells, and its expression in BM hematopoietic progenitors is inversely correlated with differentiation. The highest expres- Discussion sion of PTK7 is found in the HSC-MPP1 compartment. PTK7 The discovery of novel molecular pathways implicated in HSC expression is also conserved between mice and humans when maintenance is of particular interest for developing new strategies the thymic compartment is considered. PTK7 is easily detectable aimed at the regeneration of damaged hematopoietic tissues and to at the cell surface of CLPs and immature DN and DP populations, understand hematopoietic diseases, such as leukemia. We previ- whereas its expression is lost in single-positive CD4 or CD8 lym- ously identified the cell polarity protein PTK7 as a novel tyrosine phocytes. These data correlate with the results from other groups who kinase receptor that is overexpressed in acute myeloid leukemia. demonstrated expression of PTK7 by recent thymic emigrant cells High expression of PTK7 is correlated with poor prognosis and and thymocytes (37, 38). The Journal of Immunology 4375 Downloaded from http://www.jimmunol.org/

FIGURE 6. Homing abilities of deﬁcient PTK7 cells. (A) A total of 2 3 106 FL cells from Ptk72/2 and Ptk7+/+ embryos was stained with cell tracker

Calcein, AM and Calcein Red-Orange, respectively, pooled in a 1:1 ratio, and injected into recipient mice. (B–D) Twelve hours after injection, mice were by guest on September 24, 2021 sacrificed, and biological samples were removed and analyzed to evaluate the homing abilities of each cell. (B) BM of uninjected mice was used as control for staining specificity. (C) Dot plots of stained cells removed from blood, BM, spleen, liver, and thymus. (D) Quantification and statistical analysis. Data were pooled from three independent experiments for all organs except for the spleen panel which shows a representative experiment (n = 6 mice/group). *p , 0.05, ***p , 0.001 nonparametric Mann–Whitney test. ns, not significant.

To study the role of PTK7 in hematopoiesis, we generated ligands (41, 42). In this way, Wnt5a antagonizes Wnt3a- Ptk7-deficient mice using available gene-trapped ES cells. Our mediated canonical signaling and increases the quiescence of Ptk7-deficient strain exhibited the expected PCP defects com- HSCs (43). This occurs through activation of the small Rho pared with another strain of mice issued from the same ES clone GTPase Cdc42, which increases HSC polarity (44, 45). Such a (21). Because these mutant mice cannot reach adulthood, func- function of PTK7 in the Wnt5a-mediated noncanonical pathway tions of PTK7 were investigated in FL hematopoietic cells. The would fit with the PCP function of PTK7 and with the altered major hematopoietic phenotype of Ptk7-deficient mice consists homing and migrating properties of Ptk7-deficient cells that of increased quiescence in HSPCs and shrinkage of the FL HSPC likely depend on small Rho GTPase activation. Some studies pool. Although these phenotypes may sound logical if one showed that PTK7 regulates the balance between the Wnt ca- considers that less proliferation produces fewer cells, they are nonical and noncanonical pathways by interacting with Dish- difficult to reconcile with the loss of functional reconstitution evelled (26, 46), ROR2 (42, 47), b-catenin (25), and LRP6 (39). observed when recipient mice are engrafted with limited num- We unsuccessfully tried to identify downstream pathways that bers of LSK cells (Fig. 2J). Indeed, PTK7 is implicated in the could be regulated by Ptk7 in HSCPs using a candidate gene/ canonical and noncanonical Wnt pathways (17, 39), which are protein approach. We also looked at gene expression differences thought to be the guardians of the balance between HSC self- between Ptk7-knockout and control hematopoietic cells using renewal and HSC differentiation (32). We showed recently that unbiased gene expression profiling. Although we identified dif- PTK7 plays a role in the canonical pathway downstream of ferences (data not shown), we could not find obvious gene Wnt3a, suggesting that deficiency in PTK7 may mirror defi- ontology enrichments or pathways that could be tested ex- ciency in Wnt3a (25). This would be consistent with results perimentally. Further studies are needed to unravel how Ptk7 showing that Wnt3a deficiency leads to a reduction in the functions in HSPCs. numbers of HSPCs in the FL but would not fit with the reduced In our model of FL hematopoietic cells, we confirmed the role of reconstitution capacity observed in secondary transplantation PTK7 in cell migration and found a new role for this protein in cell assays of Wnt3a-deficient cells, which is not observed with Ptk7- homing to hematopoietic organs and in the control of cell pro- deficient cells (40). Alternatively, it may well be that PTK7 liferation. This observation confirms that our previous data ob- controls the noncanonical pathway through interaction with Wnt tained from human leukemia samples (27) can be linked with 4376 ROLE OF THE POLARITY PROTEIN PTK7 IN MURINE HEMATOPOIESIS

FIGURE 7. Adhesion and migration assays. (A) Adhesion assay on MS5 stromal cells. Percentage of adherent FL cells on MS5 stromal BM cell line after 2 h of incubation. Total FL cells (left panel). LSK cells (right panel). Data are representative of three independent experiments. Ptk2/2, n =4; Ptk+/2, n =6;Ptk+/+, n =3.(B–E) Migration in Boyden chamber assay. Number of migrating cells were normalized to the number of input cells for each sample (duplicate wells). (B) Migration of Lin2 enriched FL cells without chemoattractant. (C) Migration of Lin2 enriched FL cells with SDF- Downloaded from 1a as chemoattracting reagent. (D) Migration of LSK FL cells without chemoattractant. (E) Migration of LSK FL cells with SDF-1a as chemoattracting reagent. **p , 0.01, ***p , 0.001 nonparametric Mann–Whitney test. ns, not significant. http://www.jimmunol.org/ by guest on September 24, 2021 already-described functions of PTK7 in cell migration and in Altogether, our results provide a novel unexpected function for metastatic dissemination. Indeed, we (27) and other investigators the PCP molecule PTK7 in the cross-talk between expansion of the (29, 48, 49) found that PTK7 expression on various cancer cells is HSPC compartment and HSPC migration. This opens a new avenue linked with higher metastasis potential. Recently, we also de- for the manipulation of HSC fate using agonists or antagonists of scribed a poor prognosis of PTK7 expression in colon cancer cells PTK7. linked with higher metastatic events (50). In addition, our data show for the first time, to our knowledge, that PTK7 is involved in Acknowledgments the regulation of the HSPC cycle. Lack of PTK7 expression in- We thank Patrick Gibier, Patrick Garzino, Annaelle Legrand, and Marie- creases quiescence and enhances the long-term repopulating po- Laure Thibult (L’animalerie et la plateforme de cytome´trie en flux, Centre tential after 6 mo, whereas a trend toward decreased chimerism is de Recherche en Cance´rologie de Marseille). observed after 1 or 2 mo. This could reflect the dual function of loss of PTK7 expression in decreasing short-term homing to the Disclosures BM and increasing long-term establishment of hematopoiesis. The authors have no financial conflicts of interest. Alternatively, it may be that a homing assay performed with total liver cells reveals a constitutive promigratory function of PTK7, in contrast to long-term engraftment reflecting BM microenviron- References mental signaling through PTK7 in HSCs. Similar complex mech- 1. Kiel, M. J., O. H. Yilmaz, T. Iwashita, O. H. Yilmaz, C. Terhorst, and anisms of cell cycle regulation and differential homing of mature S. J. Morrison. 2005. SLAM family receptors distinguish hematopoietic stem and progenitor cells and reveal endothelial niches for stem cells. Cell 121: 1109–1121. cells versus HSPCs homing to the BM were also attributed to 2. Oguro, H., L. Ding, and S. J. Morrison. 2013. SLAM family markers resolve CXCR4 (51–54). Whether PTK7 and CXCR4 signaling share more functionally distinct subpopulations of hematopoietic stem cells and multipotent progenitors. Cell Stem Cell 13: 102–116. in common remains to be addressed. In conclusion, PTK7 over- 3. Kim, I., S. He, O. H. Yilmaz, M. J. Kiel, and S. J. Morrison. 2006. Enhanced expression in immature blast cells of acute myeloid leukemia was purification of fetal liver hematopoietic stem cells using SLAM family receptors. associated with a bad prognosis, early relapse, and chemothera- Blood 108: 737–744. 4. Golan, K., Y. Vagima, A. Ludin, T. Itkin, S. Cohen-Gur, A. Kalinkovich, peutic resistance, and a recent study showed that PTK7 can be used O. Kollet, C. Kim, A. Schajnovitz, Y. Ovadya, et al. 2012. S1P promotes murine as a stem cell marker of colon cancer stem cells (27, 55). Taken progenitor cell egress and mobilization via S1P1-mediated ROS signaling and together, our data provide evidence that PTK7 is expressed in stem SDF-1 release. Blood 119: 2478–2488. 5. Lapidot, T., P. Goichberg, K. Lapid, A. Avigdor, and O. Kollet. 2007. The end- cells, including cancer stem cells, and that it has a role in stem cell osteum region keeps human leukemic stem cells alive. Cell Stem Cell 1: 483–484. biology. We think that these new data about PTK7’s involvement in 6. Arcangeli, M.-L., V. Frontera, F. Bardin, E. Obrados, S. Adams, C. Chabannon, C. Schiff, S. J. C. Mancini, R. H. Adams, and M. Aurrand-Lions. 2011. JAM-B the cell cycle are very promising for understanding the molecular regulates maintenance of hematopoietic stem cells in the bone marrow. Blood mechanisms of leukemia/cancer development. 118: 4609–4619. The Journal of Immunology 4377

7. Ooi, A. G., H. Karsunky, R. Majeti, S. Butz, D. Vestweber, T. Ishida, 33. Florian, M. C., K. J. Nattamai, K. Do¨rr, G. Marka, B. Uberle, V. Vas, C. Eckl, T. Quertermous, I. L. Weissman, and E. C. Forsberg. 2009. The adhesion mol- I. Andra¨, M. Schiemann, R. A. J. Oostendorp, et al. 2013. A canonical to non- ecule esam1 is a novel hematopoietic stem cell marker. Stem Cells 27: 653–661. canonical Wnt signalling switch in haematopoietic stem-cell ageing. Nature 503: 8. Winkler, I. G., V. Barbier, B. Nowlan, R. N. Jacobsen, C. E. Forristal, 392–396. J. T. Patton, J. L. Magnani, and J.-P. Le´vesque. 2012. Vascular niche E-selectin 34. Wilson, A., G. M. Oser, M. Jaworski, W. E. Blanco-Bose, E. Laurenti, regulates hematopoietic stem cell dormancy, self renewal and chemoresistance. C. Adolphe, M. A. Essers, H. R. Macdonald, and A. Trumpp. 2007. Dormant and Nat. Med. 18: 1651–1657. self-renewing hematopoietic stem cells and their niches. Ann. N. Y. Acad. Sci. 9. Morrison, S. J., and D. T. Scadden. 2014. The bone marrow niche for haema- 1106: 64–75. topoietic stem cells. Nature 505: 327–334. 35. Wang, F., J. Travins, B. DeLaBarre, V. Penard-Lacronique, S. Schalm, 10. Papayannopoulou, T., C. Craddock, B. Nakamoto, G. V. Priestley, and E. Hansen, K. Straley, A. Kernytsky, W. Liu, C. Gliser, et al. 2013. Targeted N. S. Wolf. 1995. The VLA4/VCAM-1 adhesion pathway defines contrasting inhibition of mutant IDH2 in leukemia cells induces cellular differentiation. mechanisms of lodgement of transplanted murine hemopoietic progenitors be- Science 340: 622–626. tween bone marrow and spleen. Proc. Natl. Acad. Sci. USA 92: 9647–9651. 36. Fleming, H. E., V. Janzen, C. Lo Celso, J. Guo, K. M. Leahy, H. M. Kronenberg, 11. Hidalgo, A., F. Sanz-Rodrı´guez, J. L. Rodrı´guez-Fernańdez, B. Albella, and D. T. Scadden. 2008. Wnt signaling in the niche enforces hematopoietic C. Blaya, N. Wright, C. Cabanãs, F. Pro´sper, J. C. Gutierrez-Ramos, and stem cell quiescence and is necessary to preserve self-renewal in vivo. Cell Stem J. Teixido´. 2001. Chemokine stromal cell-derived factor-1alpha modulates VLA- Cell 2: 274–283. 4 integrin-dependent adhesion to fibronectin and VCAM-1 on bone marrow 37. Jiang, G., M. Zhang, B. Yue, M. Yang, C. Carter, S. Z. Al-Quran, B. Li, and hematopoietic progenitor cells. Exp. Hematol. 29: 345–355. Y. Li. 2012. PTK7: a new biomarker for immunophenotypic characterization of 12. Ara, T., K. Tokoyoda, T. Sugiyama, T. Egawa, K. Kawabata, and T. Nagasawa. maturing T cells and T cell acute lymphoblastic leukemia. Leuk. Res. 36: 1347– 2003. Long-term hematopoietic stem cells require stromal cell-derived factor-1 1353. for colonizing bone marrow during ontogeny. Immunity 19: 257–267. 38. Haines, C. J., T. D. Giffon, L.-S. Lu, X. Lu, M. Tessier-Lavigne, D. T. Ross, and 13. Arai, F., A. Hirao, M. Ohmura, H. Sato, S. Matsuoka, K. Takubo, K. Ito, D. B. Lewis. 2009. Human CD4+ T cell recent thymic emigrants are identified G. Y. Koh, and T. Suda. 2004. Tie2/angiopoietin-1 signaling regulates hemato- by protein tyrosine kinase 7 and have reduced immune function. J. Exp. Med. poietic stem cell quiescence in the bone marrow niche. Cell 118: 149–161. 206: 275–285. 14. McCulloch, E. A., L. Siminovitch, J. E. Till, E. S. Russell, and S. E. Bernstein. 39. Bin-Nun, N., H. Lichtig, A. Malyarova, M. Levy, S. Elias, and D. Frank. 2014. 1965. The cellular basis of the genetically determined hemopoietic defect in PTK7 modulates Wnt signaling activity via LRP6. Development 141: 410–421. Downloaded from anemic mice of genotype Sl-Sld. Blood 26: 399–410. 40. Luis, T. C., B. A. Naber, W. E. Fibbe, J. J. van Dongen, and F. J. Staal. 2010. 15. Barker, J. E. 1994. Sl/Sld hematopoietic progenitors are deficient in situ. Exp. Wnt3a nonredundantly controls hematopoietic stem cell function and its deficiency Hematol. 22: 174–177. results in complete absence of canonical Wnt signaling. Blood 116: 496–497. 16. Wilson, A., and A. Trumpp. 2006. Bone-marrow haematopoietic-stem-cell 41. Peradziryi, H., N. A. Kaplan, M. Podleschny, X. Liu, P. Wehner, A. Borchers, niches. Nat. Rev. Immunol. 6: 93–106. and N. S. Tolwinski. 2011. PTK7/Otk interacts with Wnts and inhibits canonical 17. Lhoumeau, A.-C., F. Puppo, T. Pre´bet, L. Kodjabachian, and J.-P. Borg. 2011. Wnt signalling. EMBO J. 30: 3729–3740. PTK7: a cell polarity receptor with multiple facets. Cell Cycle 10: 1233–1236. 42. Martinez, S., P. Scerbo, M. Giordano, A. M. Daulat, A.-C. Lhoumeau, V. Thome´,

18. Peradziryi, H., N. S. Tolwinski, and A. Borchers. 2012. The many roles of PTK7: L. Kodjabachian, and J.-P. Borg. 2015. The PTK7 and ROR2 Protein Receptors http://www.jimmunol.org/ a versatile regulator of cell-cell communication. Arch. Biochem. Biophys. 524: Interact in the Vertebrate WNT/Planar Cell Polarity (PCP) Pathway. J. Biol. 71–76. Chem. 290: 30562–30572. 19. Angers, S., and R. T. Moon. 2009. Proximal events in Wnt signal transduction. 43. Nemeth, M. J., L. Topol, S. M. Anderson, Y. Yang, and D. M. Bodine. 2007. Nat. Rev. Mol. Cell Biol. 10: 468–477. Wnt5a inhibits canonical Wnt signaling in hematopoietic stem cells and en- 20. Sebbagh, M., and J.-P. Borg. 2014. Insight into planar cell polarity. Exp. Cell hances repopulation. Proc. Natl. Acad. Sci. USA 104: 15436–15441. Res. 328: 284–295. 44. Florian, M. C., K. Do¨rr, A. Niebel, D. Daria, H. Schrezenmeier, M. Rojewski, 21. Lu, X., A. G. Borchers, C. Jolicoeur, H. Rayburn, J. C. Baker, and M. Tessier- M.-D. Filippi, A. Hasenberg, M. Gunzer, K. Scharffetter-Kochanek, et al. 2012. Lavigne. 2004. PTK7/CCK-4 is a novel regulator of planar cell polarity in Cdc42 activity regulates hematopoietic stem cell aging and rejuvenation. Cell vertebrates. Nature 430: 93–98. Stem Cell 10: 520–530. 22. Jung, J. W., A. R. Ji, J. Lee, U. J. Kim, and S. T. Lee. 2002. Organization of the 45. Schlessinger, K., E. J. McManus, and A. Hall. 2007. Cdc42 and noncanonical human PTK7 gene encoding a receptor protein tyrosine kinase-like molecule and Wnt signal transduction pathways cooperate to promote cell polarity. J. Cell

alternative splicing of its mRNA. Biochim. Biophys. Acta 1579: 153–163. Biol. 178: 355–361. by guest on September 24, 2021 23. Mossie, K., B. Jallal, F. Alves, I. Sures, G. D. Plowman, and A. Ullrich. 1995. 46. Wehner, P., I. Shnitsar, H. Urlaub, and A. Borchers. 2011. RACK1 is a novel Colon carcinoma kinase-4 defines a new subclass of the receptor tyrosine kinase interaction partner of PTK7 that is required for neural tube closure. Development family. Oncogene 11: 2179–2184. 138: 1321–1327. 24. Na, H.-W., W.-S. Shin, A. Ludwig, and S.-T. Lee. 2012. The cytosolic domain of 47. Podleschny, M., A. Grund, H. Berger, E. Rollwitz, and A. Borchers. 2015. A protein-tyrosine kinase 7 (PTK7), generated from sequential cleavage by a PTK7/Ror2 Co-Receptor Complex Affects Xenopus Neural Crest Migration. disintegrin and metalloprotease 17 (ADAM17) and g-secretase, enhances cell PLoS One 10: e0145169. proliferation and migration in colon cancer cells. J. Biol. Chem. 287: 25001– 48. Shin, W.-S., J. Kwon, H. W. Lee, M. C. Kang, H.-W. Na, S.-T. Lee, and 25009. J. H. Park. 2013. Oncogenic role of protein tyrosine kinase 7 in esophageal 25. Puppo, F., V. Thome´, A.-C. Lhoumeau, M. Cibois, A. Gangar, F. Lembo, squamous cell carcinoma. Cancer Sci. 104: 1120–1126. E. Belotti, S. Marchetto, P. Lećine, T. Pre´bet, et al. 2011. Protein tyrosine kinase 49. Jin, J., H. S. Ryu, K. B. Lee, and J.-J. Jang. 2014. High expression of protein 7 has a conserved role in Wnt/b-catenin canonical signalling. EMBO Rep. 12: tyrosine kinase 7 significantly associates with invasiveness and poor prognosis in 43–49. intrahepatic cholangiocarcinoma. PLoS One 9: e90247. 26. Shnitsar, I., and A. Borchers. 2008. PTK7 recruits dsh to regulate neural crest 50. Lhoumeau, A.-C., S. Martinez, J.-M. Boher, G. Monges, R. Castellano, migration. Development 135: 4015–4024. A. Goubard, M. Doremus, F. Poizat, B. Lelong, C. de Chaisemartin, et al. 2015. 27. Prebet, T., A.-C. Lhoumeau, C. Arnoulet, A. Aulas, S. Marchetto, S. Audebert, Overexpression of the Promigratory and Prometastatic PTK7 Receptor Is As- F. Puppo, C. Chabannon, D. Sainty, M.-J. Santoni, et al. 2010. The cell polarity sociated with an Adverse Clinical Outcome in Colorectal Cancer. PLoS One 10: PTK7 receptor acts as a modulator of the chemotherapeutic response in acute e0123768. myeloid leukemia and impairs clinical outcome. Blood 116: 2315–2323. 51. Cashman, J., I. Clark-Lewis, A. Eaves, and C. Eaves. 2002. Stromal-derived 28. Golubkov, V. S., A. V. Chekanov, P. Cieplak, A. E. Aleshin, A. V. Chernov, factor 1 inhibits the cycling of very primitive human hematopoietic cells in vitro W. Zhu, I. A. Radichev, D. Zhang, P. D. Dong, and A. Y. Strongin. 2010. The and in NOD/SCID mice. Blood 99: 792–799. Wnt/planar cell polarity protein-tyrosine kinase-7 (PTK7) is a highly efficient 52. Broxmeyer, H. E., L. Kohli, C. H. Kim, Y. Lee, C. Mantel, S. Cooper, proteolytic target of membrane type-1 matrix metalloproteinase: implications in G. Hangoc, M. Shaheen, X. Li, and D. W. Clapp. 2003. Stromal cell-derived cancer and embryogenesis. J. Biol. Chem. 285: 35740–35749. factor-1/CXCL12 directly enhances survival/antiapoptosis of myeloid progenitor 29. Golubkov, V. S., N. L. Prigozhina, Y. Zhang, K. Stoletov, J. D. Lewis, cells through CXCR4 and G(alpha)i proteins and enhances engraftment of P. E. Schwartz, R. M. Hoffman, and A. Y. Strongin. 2014. Protein-tyrosine competitive, repopulating stem cells. J. Leukoc. Biol. 73: 630–638. pseudokinase 7 (PTK7) directs cancer cell motility and metastasis. J. Biol. 53. Foudi, A., P. Jarrier, Y. Zhang, M. Wittner, J.-F. Geay, Y. Lecluse, T. Nagasawa, Chem. 289: 24238–24249. W. Vainchenker, and F. Louache. 2006. Reduced retention of radioprotective 30. Kaucka´, M., K. Plevova´, S. Pavlova´, P. Janovska´, A. Mishra, J. Verner, hematopoietic cells within the bone marrow microenvironment in CXCR4-/- J. Procha´zkova´, P. Krejcı´, J. Kotaskova´, P. Ovesna´, et al. 2013. The planar cell chimeric mice. Blood 107: 2243–2251. polarity pathway drives pathogenesis of chronic lymphocytic leukemia by the 54. Nie, Y., Y.-C. Han, and Y.-R. Zou. 2008. CXCR4 is required for the quiescence regulation of B-lymphocyte migration. Cancer Res. 73: 1491–1501. of primitive hematopoietic cells. J. Exp. Med. 205: 777–783. 31. Kokolus, K., and M. J. Nemeth. 2010. Non-canonical Wnt signaling pathways in 55. Jung, P., C. Sommer, F. M. Barriga, S. J. Buczacki, X. Hernando-Momblona, hematopoiesis. Immunol. Res. 46: 155–164. M. Sevillano, M. Duran-Frigola, P. Aloy, M. Selbach, D. J. Winton, and 32. Malhotra, S., and P. W. Kincade. 2009. Wnt-related molecules and signaling E. Batlle. 2015. Isolation of Human Colon Stem Cells Using Surface Expression pathway equilibrium in hematopoiesis. Cell Stem Cell 4: 27–36. of PTK7. Stem Cell Rep. 5: 979–987. A. B.

PTK7 PTK7

Supplementary Figure S1: Validation of anti-PTK7 antibodies on MEF cells.

Expression of PTK7 on MEF cells using rabbit polyclonal antibody (A) or home made monoclonal antibody 1H6.6 (B).MEF Ptk7+/+: black line; MEF Ptk7-/-: dark grey; isotypic control: rabbit serum JPB3 or rat monoclonal 9B5 respectively, light gray. A. Gated on DN cells DN1 0.5% DN2 2% CD4 CD44

3.5% DN4 35% DN3 58%

CD8 CD25

B. DN1 DN2 DN3 DN4 DP SP CD4+CD8- SP CD4-CD8+

PTK7

Supplementary Figure S2: PTK7 expression in thymus.

(A) Strategy of gating.Left panel: Dot plot showing gate of double negative (DN) (CD4-CD8- thymus cells). Right panel: gating of thymic progenitors. Most immature progenitors are defined as DN1 (CD25-CD44+) and then successively differentiate in DN2 (CD25+CD44+), DN3 (CD25+CD44-) and DN4 (CD25-CD44+). DP: Double positive cells (CD4+CD8+). SP: Single positive cells(CD4+CD8- or CD4-CD8+). (B) Expression of PTK7 in thymus. Black line: PTK7, gray histogram: isotypic control. A. B. 15 ns 8 ns

6 10

4 5 2 Hemoglobin (g/dl)

0 Red Blood Cells (106/µl) 0 +/+ -/- +/+ -/-

C. D.

21%

PTK7 Glycophorin A

Supplementary Figure 3: Erythroblastic differentiation.

(A,B) Erythroblastic reconstitution in transplanted mice. (A) Hemoglobin level and (B) red blood cells numeration in reconstituted mice with Ptk7+/+ or Ptk7-/- fetal liver. Data are the mean ± s.d. N=6 mice per group. (C, D) Differentiation of TF1 cells. (C) Expression of PTK7 on TF1 cells infected by viral particles containing scramble or specific anti-PTK7 ShRNA (shC and Sh 431). (D) Expression of Glycophorine A on TF1 infected cells. TF1 ShScramble: black line, TF1 Sh431 : dark grey, TF1 ShC : light grey. On D panel, thick black line: untreated TF1 cells. A. B.

C. D.

** * **

Supplementary figure 4: Methylcellulose assays.

Differentiation of Ptk7-/- FL cells in methylcellulose medium in (A) CFU-M , (B) CFU-G, (C) CFU-E, (D) CFU-GM (*) and three CFU-G (**).