Abstract
Despite the lack of unanimous consent on hematopoietic cell development, the current paradigm is that early hematopoietic progenitors are exclusively embryonic and short-lived. Their progeny resides in the adult and maintains its homeostasis exclusively by self-renewing without de novo generation. Here, we show that hematopoietic progenitor cells (HPC), which passed through a CD4 positive stage (HPC-CD4) give rise to a fraction of γδT cells, B cells, and dendritic cells, a larger part of liver and lung macrophages, and megakaryocyte-erythroblast progenitors. These hematopoietic progenitor cells emerged during embryonic life, are observed but not generated in adults, where their graft allows the production of myeloid and lymphoid cells. Thus, we propose HPC-CD4 as a distinct embryonic HPC, which maintains a continuous generation of a fraction of hematopoietic cells through the lifespan.
Introduction
It is now accepted that the mammalian hematopoiesis is a multilayer process, which includes two consecutive waves of blood cell generations named primitive and definitive. During embryonic development, the earliest progenitor, with dual vascular and hematopoietic potential, the hemangioblast, differentiates into endothelial-like cell progenitor with hemogenic potential called hemogenic endothelium, which can further differentiate into hematopoietic progenitors (Kennedy et al., 1997; Lugus et al., 2009). The first blood cells are extraembryonic and detected in an anatomic structure called blood islands situated in the yolk sac (YS) (Moore and Metcalf, 1970; Palis and Yoder, 2001). In mice, their appearance occurs on days E7.0 – E7.25 with the generation of primitive (fetal-type) erythroid and macrophage progenitors (Palis et al., 1995; Palis et al., 1999; Wong et al., 1986). The next extra-embryonic progenitor, defined as erythro-myeloid progenitor (EMP), emerges in the YS around day E8.5, lacks defined lymphoid differentiation potential, and mainly gives rise to adult-type myeloid cells and erythrocytes before the functional circulatory system is established (Lux et al., 2008). Upon transplantation in immunocompromised adults, the EMPs do not have long-term reconstitution potential but can provide transient adult-like red blood cell reconstitution, as well as small numbers of myeloid cells and platelets (McGrath et al., 2015).
Besides, a wave of progenitors known as lymphoid-primed multipotent progenitors (LMPPs), with differentiation potential in B-1 cells and T cells are detected in the YS from E8.25 to E9.5. These progenitors emerge before the appearance of the definitive hematopoietic stem cells (HSCs), suggesting that the YS can generate HSC-independent lymphoid lineage cells (Böiers et al., 2013; Yoshimoto et al., 2011; Yoshimoto et al., 2012). The definitive hematopoietic progenitors arise from the intra-embryonic hemogenic endothelium, shortly after the emergence of the EMPs at E8.5. The definitive hematopoietic program is initiated at different sites within the developing vasculature of the embryo, mainly in the dorsal aorta within the aorta–gonad– mesonephros (AGM) region of the embryo at E10.5. The main feature of the definitive program is its potential to generate adult bone marrow HSCs that can be engrafted in adult animals and give rise to both lymphoid and myeloid progeny (de Bruijn et al., 2002; Medvinsky and Dzierzak, 1996).
The dogma that HSCs are the individual source of all hematopoietic lineages has been established by genetic models, single-cell analysis, and lineage tracing. Fate mapping analysis of precursors of definitive HSCs demonstrated that up to 95% of adult HSCs derive from these cells (Busch et al., 2015; Pei et al., 2017), suggesting that a minority might originate from earlier progenitors. Moreover, whether some HPCs established during early embryonic life (Samokhvalov et al., 2007) still play a role in the adult is a key question that remains elusive likely due to the lack of markers to distinguish them from HPC developed in late embryonic stages.
In this study, using the CD4 fate-mapping approach, we report the unexpected presence of a small fraction of myeloid, lymphoid populations, as well as megakaryocyte-erythroblast progenitors, which are CD4 negative but had expressed Cd4. We reveal HPCs that had expressed Cd4 (HPC–CD4) before day E7.5 and reside in the adult bone marrow (BM) as CD4neg cells. The graft of HPC-CD4 is sufficient to reconstitute myeloid, lymphoid populations, and megakaryocyte-erythroblast progenitors. Thus, a small number of hematopoietic cells emerge from precursors, which have activated Cd4 during the early embryonic life, and are maintained in adulthood, suggesting that CD4 fate-mapping can be used as a labeling strategy to distinguish in adult HPCs established during the early embryonic life.
Results
A fraction of thymic precursor cells expressed CD4
In both humans and mice, the CD4 expression on hematopoietic cells has been reported to be largely restricted to αβT lymphocytes and a fraction of dendritic cells (DC) (Sawada et al., 1994). Surprisingly, the analysis by flow cytometry of the thymus from 7 days-old CD4-Cre;ROSA-YFP mice, which expressed the yellow fluorescent (YFP) reporter system under the control of the Cd4 promoter (Sawada et al., 1994), revealed that a fraction (~30%) of γδT cells were YFPpos (Figure 1. A), whereas these cells are known not to pass through a CD4 expressing stage during their thymic development (Dudley et al., 1994; Taghon et al., 2006). Given that in the thymus the degree of definitive commitment to the γδT cell fate correlates with the down-regulation of CD24 surface expression (Haks et al., 2005; Kreslavsky et al., 2008; Lauritsen et al., 2009), the presence of around 15 % of CD24low cells among the YFPpos γδT cells (Figure 1. A, B), excluding that they could correspond to a no surviving intermediate stage of differentiation in the thymus. In the same vein, YFPpos γδT cells were observed in the spleen and corresponded to mature CD24low cells (Figure 1. A, B). Interestingly, the YFPpos γδT cell fraction was not exclusively found in young mice, but also observed in adult animals (Figure 1. C, D).
Given that in CD4-Cre;ROSA-YFP mice the expression of Cre has been depicted to begin at the double-negative (DN) (CD4negCD8neg) DN4 (CD44negCD25neg)/ double-positive (DP) (CD4posCD8pos) stage of the thymocyte development (Sawada et al., 1994), thus after γδT lymphocyte branching, which occurs at the DN2 stage (CD44posCD25pos) and terminates at the DN3 stage (CD44negCD25pos) (Petrie et al., 1992; Shortman et al., 1991), we hypothesized that the Cd4 activation in the YFPpos γδ thymocytes could have occurred during or before the γδT cell differentiation. To address this question, we analyzed the early DN1 (CD44posCD25pos) precursor stage of thymocytes and found that YFPpos cells composed ~3-10% of these cells (Figure 1. E). It was notable that the YFPpos DN1 cells were mainly c-Kitneg, a phenotype classically associated with early thymic progenitors DN1d (c-KitnegCD24pos) and DN1e (c-KitnegCD24neg) with poor survival ability but capable to give rise to cells other than T cells (Porritt et al., 2004) (Figure 1. E-G). Importantly, no YFPpos γδT cell was observed in CD4-Creneg;ROSA-YFP mice, excluding any unspecific leakiness of the YFP expression in the absence of Cd4-driven Cre recombinase (Figure 1. E). Thus, this first set of data suggests that the activation of the Cd4 promoter occurs transiently either in early thymic precursors or even earlier in the HSCs.
A small fraction of lymphoid and myeloid cells are YFPpos
To assess whether the activation of the Cd4 promoter occurs in earlier precursors than that of thymic precursors, we first sought the presence of YFPpos cells with different hematopoietic identities in CD4-Crepos;ROSA-YFP animals. Due to the expression of Cd4 in both αβT lymphocytes and some innate lymphoid cells (ILCs) (Mebius et al., 1997), we excluded mature T lymphocytes and ILCs by analyzing CD45pos CD90.2neg CD4neg cells. Strikingly, the YFP expression was observed in several hematopoietic cell types in young and adult mice (Figure 2. A and S1). We detected CD4neg YFPpos hematopoietic cells in the spleen (~5%,) the liver (~8%), and the lungs (~12%) (Figure 2. A, B). Interestingly, in contrast to their YFPneg counterparts, the YFPpos cells were mainly enriched for CD11cneg CD11bhi cells (Figure 2. C, D), which expressed the macrophage-associated marker F4/80 at their surface (Figure 2. E, H). Notably, the CD11cposCD11bneg fraction of the YFPpos cells observed in the lungs contained a subset under-represented among the YFPneg fraction which expressed both Ly6-G and/or F4/80 markers (Figure 2. C and S2), in agreement with an alveolar macrophage phenotype (Misharin et al., 2013). Interestingly, within the dendritic cell (DC) compartment, where the proportion of conventional myeloid DC (CD11chi CD11bdull) (cDCs) was under-represented within the YFPpos cells, the plasmacytoid DCs (pDCs) (CD11chi CD11bneg SiglecHpos B220pos) were largely enriched, particularly in the liver (5% vs 0.5%) (Figure 2. F, I). In addition to the pDCs, other lymphoid cells were present among the YFPpos cells, including a small fraction of B cells (CD19pos) (Figure 2. G, J). Moreover, brain analysis revealed some CD11bpos YFPpos microglia-like cells in CD4-Cre;ROSA-YFP (Figure 2. L) and (Figure S3. A, B, C) animals. Interestingly, CD45posCD11bposYFPpos cells with large size and granularity were detected in embryos on day E10.5, suggesting an early embryonic origin of the CD4 expression event (Figure S3. D, E, F). Hence, the presence of CD4negYFPpos differentiated myeloid and lymphoid cells in the different organs of CD4-Cre;ROSA-YFP animals (Figure 2. K) suggests either their long-term maintenance through self-renewing or renewing from a common hematopoietic precursor present throughout the lifespan, which has passed through CD4 expression stage.
A fraction of erythroid and platelet progenitors, as well as LSK cells is YFPpos
Next, we analyzed the presence of either YFPpos HPCs or a common YPFpos HSC which could give rise to the different aforedescribed YFPpos populations. Thus, the analysis of the HPC compartment in the bone marrow (BM) of CD4-Cre;ROSA-YFP animals revealed that among the CD45pos CD4neg Linneg cells ~3% were CD4neg YFPpos cells (Figure 3. A-B). Though the CD4neg YPFpos HPCs represented a small fraction of the HPCs, as their YFPneg counterparts, they were composed of a similar proportion of erythroid (TER119pos CD41neg) and megakaryocyte progenitors (Mkp) (TER119neg CD41pos) (Figure 3 C). Interestingly, we found that CD4neg YPFpos HPCs were enriched in erythroid progenitors (ErP) (TER119pos c-Kitpos CD105pos) (Figure 3. D, F). Notably, the analysis of the CD45pos CD4neg cells revealed (~0.3-1%) of (Linneg SCA-1pos c-Kitpos) YFPpos LSK cells (Figure 3. E). On average, per mouse, ~2 600 YFPpos MkP, ~150 YFPpos ErP, and 250-500 YFPpos LSK cells were found in the BM from the femurs and tibia, underlying the small size of this population (Figure 3F). Thus, altogether these data reveal that a tiny fraction of HPCs, including MkP ErP, and LSK precursors, passed through CD4 expressing stage but failed to maintain the CD4 expression.
Multilineage reconstitution potential of sorted adult BM CD4negYFPpos LSKs
To further confirm the hematopoietic progenitor status of the YFPpos BM LSK cells we observed in the CD4-CRE;Rosa-YFP mice, we assessed the ability of purified YFPpos LSK cells to reconstitute the hematopoietic compartments by grafting them into sublethally irradiated Rag2KO recipients (Figure 4. A). The analysis of the reconstituted animals revealed the presence of CD45pos CD4neg YFPpos cells in spleen ~13%, liver ~9%, and lymph nodes ~40% (Figure 4. B, C) of the reconstituted recipients. We detected the presence of both lymphoid and myeloid YFPpos cells (Figure 4. D-G). CD45pos YFPpos CD4neg cells were observed in the peripheral organs and were composed of cDCs (Figure 4. F, G), neutrophils, and macrophages (Figure 5. A, B), B cells, and T cells with variation from one recipient to another (Figure 5. C-F). As in CD4-Cre;ROSA-YFP mice, the BM of recipient mice contained a small fraction of YFPpos ErP (TER119pos cKitpos CD105pos), and MkP (CD41pos), as well as YFPpos LSK cells (Figure S4. A-F). Taken together this set of data confirm the presence of YFPpos CD4neg hematopoietic progenitors in adult mice, capable to differentiate into both lymphoid and myeloid cell types.
CD4 expression on LSK cells is an early embryonic life event
To further analyze the origin of the YFPpos HPCs, we took advantage of CD4-CreERT2 system inducing floxed allele excision in cells expressing Cd4, after tamoxifen injection, including in mouse embryos (Aghajani et al., 2012). The tamoxifen is known to be cleared out of the mouse body within three days upon its introduction, hence each injection covers the injection day, as well as, the next two following days (Senserrich et al., 2018). After tamoxifen injection, YFPpos cells were observed in the spleen in but in clear contrast with data exposed in Figure 3, we failed to find any YFPpos Mkp, Ep, or LSK cells, with a very low number of YFPpos BM cells, representing most probably resident T cells (Figure 6. A). Moreover, the transfer of sorted BM YFPnegCD45posLINneg hematopoietic progenitor cells from CD4-Crepos;ROSA-YFP mice into sublethally irradiated RAG2KO recipient mice failed to generate YFPpos progeny in the bone marrow (Figure. 6 B-E), suggesting that the CD4/YFP expression event does not occur in adult animals. Taken together, the aforementioned observations rule out the adult origin of YFPpos precursors cells and suggest that the activation of the CD4 promoter has occurred at the early embryonic stages. To address this question, pregnant CD4-CreERT2;ROSA-YFP mice were injected with tamoxifen from days E2.5 to E4.5, as well as, on days E7.0 and E10.5. The latter was previously reported to cover the time frame when the progenitors of the primitive and definitive hematopoietic waves first appear. The E1.5 to E4.5 injections did not lead to a viable progeny, most probably because of the described adverse effects of tamoxifen on the embryo implantation (Bloxham et al., 1977; Pugh and Sumano, 1982; Ved et al., 2019). Interestingly, tamoxifen injections on days E7.0 and E10.5 resulted in a viable progeny, but no YFPpos cells were detected in the analyzed mice (data not shown). Altogether these results suggest that the Cd4 expression event probably occurs before or on day E4.5 in progenitor precursor cell types, such as hemangioblast or HE.
Discussion
Mammalian hematopoiesis is a multilayered process, starting from early embryonic life and proceeding through the adult lifespan. Whether some hematopoietic precursors present during embryonic life remain present functional at the adult stage is a matter of debate likely due to the absence of makers specific for their labeling. In the present study, we showed the existence of CD4negYFPpos lymphoid and myeloid cells, as well as hematopoietic stem cells in the CD4-Cre;ROSA-YFP mouse model, which is extensively used to both, label and recombine floxed sequences in the CD4pos and CD8pos αβT cells. Other research groups have also demonstrated that in this model and its derivatives not only T cells are labeled by reporter gene expression or affected by Cre-mediated recombination. Efficient Cd4-driven Cre expression in alveolar macrophages, alveolar epithelial cells (Chen et al., 2019), and chondrocytes (Wehenkel et al., 2017) has been also reported. In our study, we initially detected thymic CD4negYFPpos γδT cells, as well as DN1 cells, known not to express the CD4 marker under physiological conditions, which prompted us to further investigate this interesting observation. The thymic CD4negYFPpos γδT cells were abundant in the early postnatal period and were similar to their YFPneg counterparts regarding the expression of different cell surface markers (data not shown). Besides, the presence of CD4negYFPpos lymphoid and myeloid cells in the peripheral organs, as well as erythroid and platelet progenitors in the bone marrow also argue for the existence of a common progenitor, which passes through the CD4 expression stage. It has been reported that human NK and B-cell, (Bernstein et al., 2006; Zhang and Henderson, 1994), as well as murine DCs (Vremec et al., 2000), can express the CD4 antigen. This raises the question if the YFPpos signal we detect in the CD4neg lymphoid and myeloid cell types is inherited from their hematopoietic precursors or is induced by transient expression of the Cd4 gene followed by its downregulation in already differentiated populations. Both of these scenarios are theoretically possible but we demonstrated the multilineage potential of the sorted CD4negYFPpos LSKs when injected into irradiated recipients. Before our study, the expression of the CD4 marker has been already detected by other groups in a fraction of murine hematopoietic stem cells with multilineage reconstitution abilities (Frederickson and Basch, 1989; Goetz et al., 2016; Ishida et al., 2002; Muench et al., 1997; Wineman et al., 1992; Wu et al., 1991), but information about their origin and embryonic precursors is still missing. To determine if CD4 expression in hematopoietic stem cells is an embryonic or adult life event, we used the tamoxifen-inducible CD4-CreERT2;ROSA-YFP mouse model, in which, YFP labeling occurs upon tamoxifen treatment, only in cells, in which the CD4 marker and CreERT2 are actively produced. As expected, the CD4 expressing cells in the spleen were YFPpos upon tamoxifen treatment. However, unlike the spleen, the percentage of YFPpos cells in the BM was very low and YFPpos LSKs were not detected, suggesting that the CD4negYFPpos LSKs have been labeled during their embryonic period, which prompted us to investigate the CD4 expression kinetics in the early embryonic life. To map the embryonic day on which CD4 expression is first initiated, we injected tamoxifen in timely pregnant mice on days E7.5 and E10.5, the moment when the progenitors of the primitive and the definitive hematopoietic waves respectively appear. Analysis of the progeny of these mice showed no presence of YFPpos cells, suggesting that CD4 expression might be an event that has happened earlier during embryogenesis. In addition to these days, we also injected tamoxifen in timely pregnant females from day E1.5 to E4.5. However, the adverse effects of tamoxifen on embryo implantation resulted in the absence of progeny. We were also unable to detect cell-surface CD4 expression upon the culture of murine embryonic stem cells to embryonic bodies from day 1 to day 5 in the absence of hematopoietic cytokines (data not shown). Thus, it remains to be elucidated what is the functional role of CD4 expression during the early stages of the murine embryonic hematopoiesis, the timing, and the cell type, in which this event initially occurs. It has been reported that the CD4 molecule directly interacts with Interleukin-16 (IL-16) (Cruikshank et al., 1991; Rand et al., 1991; Ryan et al., 1995), a unique cytokine with no sequence homology to other cytokines or chemokines, which is also produced by nonlymphoid cell types, such as epithelium (Bellini et al., 1993; Hessel et al., 1998; Laberge et al., 1997; Mashikian et al., 1998) and fibroblast (Franz et al., 1998). It is known that apart from its role as a chemoattractant for CD4+ T cells, when added to the basic HSC-expanding cytokine cocktail, IL-16 significantly enhanced the expansion of human CD34+ cord blood cells in vitro (Rofani et al., 2008), suggesting a potential role of this cytokine during the early stages of the embryonic hematopoiesis. Alternatively, in T cells, it has been reported that the CD4 molecule can trigger an intracellular signaling cascade upon MHC class II binding in the absence of T cell receptor-mediated stimulation by activating the cyclic AMP (cAMP) pathway and PKA (Zhou and König, 2003), thus modulating the function of a plethora of transcription factors (Sassone-Corsi, 2012). Besides, cAMP induction increases HSC-like cell frequencies during the HE to HSC transition, whereas its inhibition decreases both hemogenic and non-hemogenic endothelium formation and abrogates hematopoietic cell generation (Saxena et al., 2016). In line with this data, MHC class II antigen expression has been also reported on human nonlymphoid cell types, such as epithelial cells (Daar et al., 1984), suggesting that, CD4-mediated signaling driven by MHC class II molecules potentially expressed by the murine embryonic environment might have a functional role for the development of the early hematopoietic progenitors. Taken together, these observations led us to the hypothesis that Cd4 expression might occur during the appearance of the HE, which specifies the primitive hematopoietic wave. In agreement with this, human liver sinusoidal endothelial cells pass through the CD4 expressing stage between gestational weeks five and twelve of the pregnancy (Couvelard et al., 1996; Poisson et al., 2017), suggesting that this process might be conserved in embryonic endothelial cell populations in mice. The strongest evidence, which supports this hypothesis is the presence of CD4negYFPpos liver and alveolar macrophages, as well as microglia. Liver and lung macrophages have been reported to be generated exclusively by primitive yolk sac (YS) hematopoietic progenitors (Epelman et al., 2014; Ginhoux et al., 2010; Guilliams et al., 2013; Hoeffel et al., 2012; Schneider et al., 2014; Schulz et al., 2012), suggesting an early embryonic origin of the CD4negYFPpos lymphoid and myeloid progenitors. The brain microglia is exclusively generated by the E7.5 “early” EMPs (Hoeffel et al., 2015), arguing that the CD4 expression may label the HE, which generates this “early” EMP hematopoietic wave. In conclusion, our study demonstrates the existence of CD4negYFPpos HSCs in adult CD4-Cre;ROSA-YFP mice, which probably represent descendants of the “early” EMP wave. The current understanding for these EMPs is that they are a short-lived embryonic wave, with exclusive myeloid differentiation potential, which does not persist in the adult stage. Our data support a model, in which embryonic HPCs-CD4 give rise to “early” EMPs, the descendants of which can persist in adults under the form of hematopoietic stem cells with multilineage differentiation potential. The YFP labeling in these adult HSCs is not due to the CD4/YFP activation at the adult stage but more likely is a remnant of transient Cd4 expression during the early embryonic development before the appearance of the primitive and the definitive hematopoietic progenitor waves.
Materials and Methods
Mice
The CD4-Cre;ROSA-YFP mouse model contains CD4-Cre transgene, which is composed of CD4 enhancer, promoter, and silencer sequences driving the expression of a Cre recombinase gene specifically in CD4-expressing T cells (Sawada et al., 1994). When these transgenic mice are bred with mice containing a YFP transgene preceded by loxP-flanked stop sequence in the Gt(ROSA)26Sor locus (Srinivas et al., 2001), Cre-mediated recombination results in deletion of the floxed sequences in the Cre recombinase-expressing cells specifically labeling them with YFP signal. The CD4-CreERT2;ROSA-YFP mouse model has been previously described (Aghajani K et al. 2012). Rag2 KO mice were purchased from Charles River laboratories. Mice were maintained in a specific pathogen–free animal facility at the Cancer Research Center of Lyon (France) and handled in the accordance with the institutional guidelines, and the protocols were approved by Comité Régional d’Ethique pour l’Expérimentation Animale.
Cell Preparations and Flow Cytometry
Cell suspensions were prepared from the thymus, spleen, lymph nodes, lung, liver, and bone marrow. Red blood cells were lysed with 168 mM NH4Cl (Merck, #A9434-1KG). Lung and liver lymphocytes were separated from epithelial cells by 30% Percoll (GE Healthcare) diluted in 1x DPBS (Thermofisher, #14200075) by centrifugation (10 min x 1300g at 20°). Cells were washed with RPMI medium supplemented with 10% FBS (Thermofisher, #A3160402) and resuspended in FACS buffer (1x DPBS; 2% FBS; 2 mM EDTA) in 96 well plate (Corning, # CLS3897-100EA). The cells were incubated with Fc-block (20 min at 4°), washed and centrifuged (2 min x 500g at 4°), and stained with various combinations of monoclonal antibodies in FACS buffer (20 min at 4°). The cells were then fixed with 1% PFA (VWR International, #20910.330) (30 min at 4°) and washed with 1x DPBS before FACS. For intracellular staining, cells were first labeled with antibodies against cell surface antigens and processed as described above. The cells were next fixed with Fixation and Permeabilization Buffer kit (eBiosciences, #88-8824-00) according to the manufacturer protocol. Cells were then analyzed using BD LSRFortessa flow cytometer or purified by flow cytometric cell sorting on a BD FACSAriaII. For Hematopoietic Stem Cell (HSC) enrichment, the bone marrow suspensions were first depleted by biotinylated lineage cocktail antibodies and MojoSort Streptavidin Nanobeads (Biolegend, # 480016) according to the manufacturer protocol. The enriched cells were then prepared for cell sorting by staining with desired FACS antibodies, as well as streptavidin conjugate as a dump channel for the remaining lineage cells.
Flow cytometry Antibodies
BD Bioscience antibodies
Vδ 6.3/2 TCR-BV421 (clone 8F4H7B7); CD8a-BV510 (clone 53-6.7); CD24-BV605 (clone M1/69); Vδ 4 TCR-BV650 (clone M1/69); CD4-BV711 (clone GK1.5); Ly6G-BV510 (clone 1A8); NK1.1-BV605 (clone PK136); TCRγδ-PE (clone GL3); CD3e-PE (clone 145-2C11); NK1.1-PE (clone PK136); CD19-PE/CF594 (clone1D3); CD3e-PE-Cy7 (clone 145-2C11); TER119-PE-Cy7 (clone TER-119); CD11b-APC (clone M1/70); CD45-APC/Cy7 (clone 30-F11); CD19-PerCP/Cy5.5 (clone 1D3); Biotinylated: CD8a (clone 53-6.7); NK1.1 (clone PK136);
Biolegend antibodies
Fc-block (clone 93); CD8a-BV605 (clone 53-6.7); Ly6G-PE/Dazzle594 (clone 1A8); CD11c- eF450 (clone N418); CD90.2 (Thy-1.2)-BV570 (clone 30-H12); F4/80-BV650 (clone BM8); SiglecH-PE (clone 551); CD11b-PE (clone M1/70); TCRb-PE (clone H57-597); NK1.1-PE/Dazzle594 (clone PK136); CD41-PE/Dazzle594 (clone MWR Reg30); MHC class II (I-A/I-E)-APC/eF780 (clone M5/114.15.2); TCR Vγ1.1/Cr4-APC (clone 2.11); CD105-APC (clone MJ7/18); CD117-BV421 (clone 2B8); TCRb-AF700 (clone H57-597); Streptavidin-BV605 (405229); Biotinylated: CD11c (clone N418); CD11b (clone M1/70); CD45R(B220) (clone RA3-6B2); F4/80 (clone BM8); TER119 (clone TR119); CD4 (clone G415); TCRb (clone H57-597); CD19 (clone 6D5); Nkp46 (CD335) (clone 29A1.4);
eBioscience antibodies
CD8a-PE (clone 53-6.7); CD11c-PE (clone N418); Nkp46(CD335)-PE (clone 29A1.4); TCR Vγ2-PerCP-eFluor710 (clone UC3-10A6); MHC class II (I-A/I-E)-PerCP-eF710 (clone M5/114.15.2); Nkp46 (CD335)-PE-Cy7 (clone 29A1.4); CD45-AF700 (clone 30-F11); Ly6A-E(SCA-1)-AF700 (clone D7); CD45R(B220)-APC/eF780 (clone RA3-6B2); CD45R(B220)-PE-Cy7 (clone RA3-6B2); CD19-AF700 (clone 1D3); Streptavidin-PE (12-4317-87) Biotinylated: Ly6G (clone RB6-8C5); CD3e (clone 17A2); CD5 (clone 53-7.3); MHC class II (I-A/I-E) (clone M5/114.15.2); TCRγδ (clone UC7-13D5);
Radiation BM chimeras
Radiation BM chimeras were generated upon transfer of 1 × 102 sorted YFP+ HSCs cells into Rag2KO recipients, irradiated by sub lethal dose of 6 Gy
Immunofluorescence
For immunofluorescence microscopy, brain tissue was fixed for 48h in 4% PFA diluted in DPBS (Thermofisher, #14200075). Cryostat cut brain sections were fixed on positively charged adhesion slides (Thermo Scientific™ J3800AMNZ). The fixed tissue was permeabilized for 5 min at 25° with Acetone (Merck, #1070212511). The slides were then washed three times with 1x DPBS and blocked with DAKO solution (Agilent, #S302283-2) for 1 hour, before overnight incubation with the following antibodies: GFP-AF488 (1:300; Biolegend #BLE668208) and CD11b-APC (1:100; BD, #553312), diluted in the blocking solution. The slides were then washed three times with 1x DPBS and stained with DAPI (Merck, # 10236276001) for 20 min, diluted 1:1000 in 1x DPBS. The slides were then mounted with Fluoromount (Sigma, #F4680-25ML) and coverslips (Knittel Glass VD11515Y1A.01). The images were taken with Microscope Zeiss Axioimager (SIP 60549).
Statistics
The unpaired, two-tailed Student’s t-test was used for statistical confirmation of cell number comparisons.
Flow cytometry analysis
Flow Cytometry analysis was performed with FlowJo software.
Author Contributions
A.K.A.: Conceived and performed experiments, performed data analysis, wrote the manuscript.
J.C.M.: Conceived and performed experiments, performed data analysis, wrote the manuscript, provided funding.
Conflict of interest
The authors declare no competing interests.
Acknowledgments
We would like to thank Dr. Valerie Kouskoff (The University of Manchester) and Dr. Philippe Kastner (IGBMC, Strasbourg) for the helpful discussions. This work was supported by LabEx.