Abstract
Tbet-deficient mice have reduced NK cells in blood and spleen, but increased NK cells in bone marrow and lymph nodes, a phenotype that is thought to be due to a defect in S1PR5-mediated migration. Here, we revisit the role of Tbet in NK cell bone marrow egress. We definitively show that the accumulation of NK cells in the bone marrow of Tbet-deficient (Tbx21−/−) animals occurs because of a cell-intrinsic migration defect. We identify a profile of gene expression, co-ordinated by Tbet, which affects the localisation of NK cells in the bone marrow. The most underexpressed gene in the absence of Tbet was Cxcr6, and we confirmed that CXCR6 protein was also not expressed in Tbet-deficient NK cells. Cxcr6-expressing ILC progenitors and immature NK cells accumulate in the bone marrow of CXCR6-deficient mice. This suggests that CXCR6 is among the mediators of migration, controlled by Tbet, that work together to coordinate NK cell bone marrow egress. Understanding NK cell bone marrow egress will become increasingly important as we move into an era in which NK cell immunotherapies are being developed.
Introduction
Tbet was originally described as the key transcription factor directing Th1 lineage commitment [1]. More recently, it has become clear that Tbet also drives differentiation and memory cell generation in a number of lymphocyte lineages [2]. In particular, Tbet is required for the development and survival of ILC1 [3]. Natural killer (NK) cells are now recognised to form a separate lineage from “helper” ILC1 [4], and are less dependent on Tbet than ILC1 [5,6]. Nevertheless, Tbet-deficient mice do display some defects in NK cell number, maturation status and function [7,8].
Tbet-deficient mice have reduced NK cells in blood and spleen, but increased NK cells in the bone marrow and lymph nodes [7,8]. These observations led to the suggestion that Tbet is required for NK cells to leave the bone marrow or lymph nodes and enter the blood [8]. Tbet-deficient NK cells express lower levels of S1pr5 mRNA than their wild type counterparts, and S1pr5 knockout mice phenocopy Tbet-deficient animals, suggesting that Tbet may mediate bone marrow and lymph node egress by upregulating S1PR5 expression [8].
We were prompted to revisit the role of Tbet in bone marrow egress by the unexpected finding that a conditional knockout of Tbet in NK cells did not display the accumulation of NK cells in the bone marrow that has been reported in other Tbet-deficient strains of mice. Here, we report that in the absence of Tbet, NK cells display a cell-intrinsic defect in their ability to leave the bone marrow, and in their ability to differentiate to the final stage of NK cell development. Although S1pr5 mRNA is downregulated in the absence of Tbet, we find that the most downregulated mRNA in Tbet-deficient NK cells is Cxcr6 and that CXCR6-deficient mice also display an accumulation of NK cells in their bone marrow, suggestive of a defect in bone marrow egress. Therefore, we propose that CXCR6 is one of the migration molecules through which Tbet mediates NK cell bone marrow egress.
Results
Tbet-deficient NK cells accumulate in the bone marrow
NK cells have previously been reported to accumulate in the bone marrow of Tbet-deficient mice [7,8] which we confirm here in Tbx21−/− mice (Fig. 1a,b). Moreover, we show the accumulation predominantly affects CD11b+ mNK cells, while CD11b− iNK cell frequencies remain unaltered. However, in the bone marrow of conditional knockout mice in which Cre recombinase-mediated excision of Tbx21 (Tbet) is driven by the Ncr1 (NKp46) promoter (Ncr1iCre Tbx21fl/fl mice) we did not observe the accumulation of NK cells that we saw in Tbx21−/− mice (Fig. 1c). The most obvious explanation for this difference would be if the requirement for Tbet in NK cell egress from the bone marrow were not cell-intrinsic. Bone marrow chimeras of Tbet-deficient and -sufficient bone marrow convincingly showed that the requirement for Tbet in NK cells leaving the lymph nodes is cell-intrinsic, but for the bone marrow the results were less clear [8]. Therefore we created mixed bone marrow chimeras to determine whether the requirement for Tbet is cell-intrinsic (Fig. 1d,e). In the chimeras, we saw an increase in the ratio of Tbet-deficient:Tbet-sufficient cells within the mNK cell compartment of the bone marrow, but not within the iNK cell compartment. This mirrors our observations in Tbx21−/− mice, where an accumulation of NK cells is observed in the mNK, but not in the iNK compartment. These results therefore support the idea that the requirement for Tbet in NK egress from the bone marrow is cell-intrinsic.
We next considered the possibility that NK cells did not accumulate in the bone marrow of conditional knockout mice because in these mice Tbx21 is excised relatively late in NK development. In contrast to Tbx21−/− mice where Tbet is absent at all times, Tbet depletion in conditional knockout mice is dependent on Ncr1 expression. NKp46 (encoded by Ncr1) is not detectable at the protein level until the immature NK cell stage of development (Fig. 1f). Therefore we considered the possibility that transient Tbet expression early in development, before Ncr1-mediated Cre recombinase activity is switched on, could be sufficient to allow NK cell egress later in development. A similar situation occurs for the transcription factor Nfil3, which is required before NK-lineage commitment for later development into functional NK cells [9]. NK cells in conditional knockouts of Nfil3 under the Ncr1 promoter develop comparably to controls, whereas they fail to develop in mice globally deficient in Nfil3, because Nfil3 is required earlier in NK cell development than it is excised in the conditional knockout mice [10].
To address whether the requirement for Tbet, like that of Nfil3, occurs at a specific developmental stage, we examined bone marrow leukocytes from Ncr1iCre Tbx21fl/fl mice for Tbet expression at various stages of NK development. We found significant loss of Tbet in NK-lineage committed progenitors, rNKP, in both conditional knockout and knockout mice relative to their controls, and notably before Ncr1 is detectable at the protein level (Fig. 1g,h). Since the deletion of Tbet occurs earlier than might be expected in conditional knockout mice, this indicates that Ncr1 mRNA, and consequently Cre-recombinase, is being transcribed before the protein reaches a level that can be detected by flow cytometry. These results suggest that the differences in NK cell accumulation that we observe between the global and conditional knockouts are therefore unlikely to be caused by transient expression of Tbet.
NK cells are less able to leave the bone marrow in the absence of Tbet
Previous reports attribute the bone marrow accumulation of Tbet-deficient NK cells to a defect in migration, involving S1PR5, since S1pr5 mRNA is underexpressed in Tbet-deficient NK cells, S1pr5−/− mice phenocopy the accumulation of NK cells in the bone marrow of Tbet-deficient mice and in S1pr5−/− mice the accumulation is caused by a migration defect [8,11,12]. However, a defect in NK cell migration from the bone marrow in the absence of Tbet has not yet been formally shown. Therefore, we examined NK cell migration from the bone marrow in vivo. Recently-migrated cells translocate to the sinusoids before exiting the bone marrow, so sinusoidal leukocyte staining acts as an indicator of migration from the bone marrow. The procedure involves intravenous injection of fluorescently-labelled anti-CD45.2 monoclonal antibody, followed by a 2 minute incubation to allow the circulation to carry the antibody to the sinusoids and stain leukocytes in the blood and sinusoids while sparing parenchymal leukocytes [13].
As NK cells mature from CD11b− iNK to CD11b+CD27+ mNK1 and then to CD11b+CD27− mNK2 in wild type mice, their frequencies decrease in the parenchyma (Fig. 2c) and increase in the sinusoids (Fig. 2d), indicating migration out of the bone marrow. We did not observe this pattern in Tbx21−/− NK cells, whose frequency instead remained constant between iNK and mNK1 in both parenchyma and sinusoid, indicating a reduced ability to migrate from the parenchyma to the sinusoid. We also observed a decrease in the frequency of mNK2 in both anatomical compartments, consistent with a defect at the mNK1 to mNK2 transition that has previously been reported [14]. In the conditional knockout mouse compared to the control, we also observed a failure of developing NK cells to leave the parenchyma and enter the sinusoid, similar to that in the conventional knockout. Therefore, this more sensitive test of bone marrow egress reveals that conditional knockout NK cells, like their conventional knockout counterparts, do have a defect in leaving the BM, although this is not apparent from simple phenotyping.
Tbet controls the expression of cell migration mediators in bone marrow NK cells
Having confirmed that the accumulation in the bone marrow of Tbx21−/− mice is caused by a migration defect, we sought to define the mediators involved. We therefore sorted iNK and mNK1 cells from the bone marrow of wild type and Tbx21−/− mice, and mNK2 (which are not present in Tbx21−/−; Fig. 2a-d) from wild type mice and analyzed them by RNASeq. Raw RNASeq data and differentially expressed gene lists are available from the National Center for Biotechnology Information Gene Expression Omnibus under accession no. GSE122874. Consistent with previous reports [8], we found S1pr5 mRNA underexpressed by a factor of approximately seven-fold in both iNK (Fig. 3a,c) and mNK (Fig. 3b,c) cells in Tbx21−/− mice compared to wild type. More severely affected by the absence of Tbet, however, was Cxcr6, which was underexpressed by a factor of 125-fold in iNK cells (Fig. 3a,c). The expression of Kit, often considered a marker of immature NK cells was increased in the absence of Tbet (Fig 3.a,b,d) whereas the expression of genes associated with mature NK effector function, such as Ifng, Gzmb and Prf1, were reduced (Fig. 3a,b,e). Consistent with reports that Tbet promotes Zeb2 [14] and antagonizes Eomes [15] transcription, Zeb2 and Eomes expression were increased and decreased, respectively, in Tbx21−/− NK cells (Fig. 3a,b,f).
We next selected a number of genes identified as differing transcriptionally between wild type and Tbx21−/− NK cells to confirm their expression at the protein level by flow cytometry (Fig. 4). The genes were selected based on a significance cutoff of padj < 0.05 and a two-fold or more differential expression between wild type and Tbx21−/− iNK and mNK1 cells. Sell and Cd69 did not have a two-fold difference between KO and WT cells, but were included in the analysis because we observed differential transcription of these genes in both NK subsets.
The expression of a number of proteins associated with cell migration differed between wild type and knockout NK cells. In particular, CXCR6 was expressed at a lower level by knockout iNK and mNK1, compared to their wild type counterparts (Fig. 4g). Consistent with the RNASeq data, the most significant drop in expression was within the iNK cell compartment. CD69, which suppresses the expression of S1PR1 in order to retain immune cells in lymph nodes and tissues, is highly upregulated in both knockout NK subsets (Fig. 4d). CD62L (encoded by Sell), which is involved in leukocyte homing from the blood to tissues, also has elevated expression in knockout NK cells compared to wild type (Fig. 4c). We also observed a decrease in the expression of CD49a in knockout iNK (Fig. 4b). CD49a is associated with ILC1 tissue residency in mice [5], but it is not known if it has a role in bone marrow retention of cells.
S1pr5 and Ccr2 displayed relatively large differences at the RNA level (Fig. 3a,b), but we did not see any differences at the protein level (Fig. 4a,i). Neither did we see any differences in CX3CR1 expression (Fig. 4h).
CXCR6+ ILC progenitors and iNK accumulate in the bone marrow of CXCR6-deficient mice
We were particularly intrigued by the underexpression of CXCR6 in Tbx21−/− NK cells, since this was the most differentially expressed gene identified by our RNASeq screen, and because we have previously reported that ILC progenitors accumulate in the bone marrow of CXCR6-deficient mice [16]. If CXCR6 is one of the mediators through which Tbet controls NK cell bone marrow egress, we would expect to see an accumulation of NK cells in the bone marrow of CXCR6-deficient mice, similar to our previous observations of ILC progenitors.
To test this hypothesis, we examined NK cells in the bone marrow of Cxcr6gfp/gfp mice, in which both Cxcr6 genes are inactivated and replaced by a reporter cassette encoding GFP, or Cxcr6gfp/+ controls. We did not observe an accumulation among the general population of CLP, αLP1 or αLP2 in the bone marrow of these mice, although we did observe an accumulation specifically among the Cxcr6+ population of αLP2 (Fig. 5a). Examining NK cell progenitors, we observed a slight accumulation among the general population of rNKP and iNK, although this was not significant. However, we did observe a significant accumulation among the Cxcr6+ population of iNK (Fig. 5b). In contrast to the Tbet knockout, we did not detect an accumulation of cells among either of the mature NK cell subpopulations in the bone marrow of CXCR6-deficient mice, suggesting that the requirement for CXCR6 in bone marrow egress occurs relatively early in NK cell development.
Discussion
In this study, we revisited the role of Tbet in mediating NK cell egress from the bone marrow during development. We observed an accumulation of NK cells in the bone marrow of Tbx21−/− mice, similar to previous reports [7,8]. The accumulating NK cells in Tbet-deficient mice have previously been described as CD27hi and KLRG1lo [8], but no distinction was made between the two CD27+ subsets, CD27+CD11b− iNK and CD27+CD11b+ mNK1. We report that the accumulation occurs specifically within the mNK1 compartment, although this is unsurprising given a more recent report, which we confirm here, that there is a block at the mNK1 to mNK2 transition in the absence of Tbet [14]. We confirmed the idea that there is a cell-intrinsic requirement for Tbet in NK cell bone marrow egress, showing accumulation of Tbx21−/− NK cells in the bone marrow of mixed bone marrow chimeras more convincingly than in previous studies [8]. Potentially, this could be because the earlier studies were carried out using a Tbet-deficient strain called Duane, in which Tbet protein levels were reduced by about four-fold, whereas Tbx21−/− NK cells completely lack Tbet.
The accumulation of NK cells in the bone marrow of Tbet-deficient mice, together with the cells’ underexpression of S1pr5 mRNA and the defect in bone marrow egress displayed by S1pr5−/− NK cells led to the suggestion that Tbet mediates NK cell bone marrow egress via S1PR5 [8,11,12]. However, no study had formally addressed the question of whether the accumulation of NK cells in the absence of Tbet occurs due to a migration defect. We used in vivo staining of sinusoidal NK cells to show that Tbet-deficient iNK and mNK1 cells are less able to move from the parenchyma to the sinusoid of the bone marrow, supporting the idea that Tbet promotes NK cell bone marrow egress. The in vivo bone marrow egress staining revealed that Tbet knockout iNK cells and conditional knockout iNK and mNK1 cells all have a migration defect that was not apparent on simple phenotyping by flow cytometry suggesting that this test detects migration defects with greater sensitivity and should be preferred. We also noted a slight increase in the frequency of NK cells in the parenchyma of Tbx21fl/fl mice, compared to wild type, although the experiments were not designed such that these two could be statistically compared. This slight accumulation of parenchymal NK cells in the control animals may have contributed to the reduced difference between the conditional knockouts and controls on simple phenotypic examination, despite the defect in NK cell migration still being present.
After we confirmed that the NK cell accumulation in Tbet-deficient bone marrow was caused by defective migration, we explored which mediators were involved, identifying a transcriptional signature associated with Tbet expression in NK cells. We confirmed previous reports that S1pr5 is reduced in Tbet-deficient NK cells at the transcriptional level [8], but to our surprise we did not observe this at the protein level. We were also surprised not to see differences in CCR2 and CX3CR1 protein expression, since both of these differed at the transcript level and both CCR2-deficient and CX3CR1-deficient mice display an accumulation of NK cells in their bone marrow [17,18]. On the other hand, a number of molecules that are involved in cell migration did differ between wild type and Tbx21−/− NK cells, with increased expression of CD69 and CD62L, which would be expected to retain cells in the parenchyma, as we observed [19] and decreased expression of CD49a and CXCR6.
CXCR6 is already known to be important for ILC progenitors leaving the bone marrow [16] so we investigated the role of CXCR6 in NK cell bone marrow egress by examining the bone marrow of CXCR6-deficient mice. Consistent with a role for CXCR6 in NK cell bone marrow egress, we found an accumulation of iNK in the bone marrow of CXCR6-deficient mice, although this was only significant among CXCR6-expressing cells. This accumulation among iNK was consistent with our observation that CXCR6 expression was both highest and most differentially expressed within this population. We did not observe an accumulation among the mNK populations. Therefore, the NK cell accumulation we observed in the absence of CXCR6 was less pronounced than that which we observed in the absence of Tbet, and was only observed in early developmental stages. Taken together with the differential expression data, this suggests that CXCR6 is likely to be one of the molecules through which Tbet acts to promote NK cell bone marrow egress in progenitors and iNK cells, but that Tbet also controls the expression of other cell migration molecules, which promote egress at later stages of NK cell development.
Overall, we have shown that Tbet is required for NK cells to leave the bone marrow in a cell intrinsic manner. We further define a profile of gene expression coordinated by Tbet which promotes NK cell bone marrow egress. Among this group of chemokine receptors and proteins, which mediate localization of NK cells within the bone marrow, is CXCR6. Bone marrow egress of NK progenitors and iNK cells is somewhat defective in the absence of CXCR6, but the phenotype is less pronounced than in the absence of Tbet, suggesting that other targets of Tbet, such as S1pr5, are likely to be important in both early and later NK cell developmental stages. It will be interesting in the future to better define the roles of some of the other genes that we identified in NK cell bone marrow egress. As we move into an era where the cancer-fighting potential of NK cells is being harnessed to develop immunotherapies, it will become increasingly important to understand NK egress in order to maximize the full potential of those therapies [20].
Materials and Methods
Mice
B6.129S6-Tbx21<tm1Glm>/J (RRID IMSR_JAX:004648; “Tbx21−/−”) and B6.129-Tbx21tm2Srnr/J mice (RRID IMSR_JAX:022741; “Tbx21fl/fl”) were purchased from the Jackson Laboratory. B6(Cg)-Ncr1tm1.1(icre)Viv/Orl mice [21] (RRID MGI:5309017; “Ncr1iCre”) were acquired from the European Mutant Mouse Archive as frozen embryos and rederived at the Royal Free Hospital, London. Ncr1iCre mice were crossed onto Tbx21fl/fl to produce Ncr1iCre Tbx21fl/fl conditional knockouts and Ncr1WT Tbx21fl/fl littermate controls, as previously described [22]. Tbx21−/− mice were crossed onto C57BL/6J mice, bred at the Royal Free Hospital, and the resultant heterozygotes were crossed to produce Tbx21−/− homozygote knockouts and Tbx21+/+ homozygote wild type littermate controls. Mice were sacrificed between 6 and 12 weeks of age with direct cervical dislocation. Death was confirmed by cessation of circulation. The hind leg bones (tibia and femurs) and spleen were dissected to isolate leukocytes. Animal husbandry and experimental procedures were performed according to UK Home Office regulations and institute guidelines under project licence 70/8530.
Cxcr6gfp/+ and Cxcr6gfp/gfp mice [23] were bred in the animal facilities at Pasteur Institute, Paris. Mice were bred in accordance with Pasteur Institute guidelines, in compliances with European welfare regulations, and all animal studies were approved by Pasteur Institute Safety Committee in accordance with French and European guidelines. Mice were sacrificed between 8 and 12 weeks of age with carbon dioxide pump. Death was confirmed by cessation of circulation. The hind leg bones (tibia and femurs) were dissected to isolate leukocytes.
Bone marrow transplantation
Recipient mice were given myeloablative irradiation before syngeneic bone marrow transplantation at 10 weeks of age. Mice received a total of 11Gy lethal irradiation, 5.5 Gy each on day −2 and day 0. On day 0, mice were reconstituted 4 hrs post-irradiation with 5 million donor bone marrow cells in 200 μl Hank’s Balanced Salt Solution (HBSS; Lonza, distributed by VWR, Lutterworth, UK), administered by intravenous injection into the tail vein. C57BL/6J (“CD45.2”) wild type mice were used as recipient mice, which received a 1:1 mixture of donor bone marrow cells from CD45.1 WT and (CD45.2) Tbx21−/− mice to produce mixed chimeras. Bone marrow transplanted mice were sacrificed and analysed 8 weeks post-transplant.
Sinusoidal cell labelling
Mice received an intravenous injection of 1 μg anti-CD45.2 PE-eFluor 610 monoclonal antibody (eBioscience) in phosphate buffered saline (PBS; Life Technologies brand, Thermo Fisher Scientific, Waltham, MA, USA). Mice were sacrificed after two minutes.
Cell isolation
Bone marrow was flushed from dissected bones with supplemented RPMI 1640 medium (Life Technologies). Tissue clumps were pelleted at 500 xg, 4°C for 5 minutes, then mechanically disaggregated and filtered through a 70 μm cell strainer. RPMI 1640 medium was supplemented with 10% fetal calf serum, non-essential amino acids, sodium pyruvate, penicillin-streptomycin, 25 mM HEPES buffer and 50 μM 2-Mercaptoethanol (all from Life Technologies).
For cell sorting on a FACSAria (BD Biosciences, Oxford, UK), isolated bone marrow cells were ACK lysed by incubation in ACK lysing buffer (Life Technologies) for 5 minutes at room temperature to remove excess red blood cells and to enrich for leukocytes before cell sorting. The leukocytes were further enriched for NK cells by immunomagnetic depletion of lineage-associated cells (CD3, CD8α, CD19, Gr-1) using the relevant FITC-conjugated antibodies and anti-FITC MicroBeads (Miltenyi, Woking, UK). The sorting buffer consisted of PBS, 0.5% bovine serum albumin (BSA; Sigma Aldrich, Hammerhill, UK) and 2 mM EDTA (Sigma Aldrich).
Spleens were passed through a 40 μm cell strainer. Red blood cells were lysed by incubation in ACK lysing buffer for 5 minutes at room temperature.
Flow cytometry
The following anti-mouse antibodies were used: anti-CCR2-PE-Cyanine7 (clone SA203G11, Biolegend, London, UK), anti-CD3-FITC (17A2, Biolegend), anti-CD3-biotin (clone 145-2C11, Sony Biotechnology, Surrey, UK), anti-CD4-BV786 (clone RM4-5, BD Horizon), anti-CD8α-FITC (clone 53-6.7, Biolegend), anti-CD8α-biotin (clone 53-6, Sony), anti-CD11b-APC (clone M1/70, Biolegend), anti-CD11b-PE (clone M1/70, Biolegend), anti-CD11b-AlexaFluor700 (clone M1/70, BD Pharmingen), anti-CD19-FITC (clone 6D5, Biolegend), anti-CD19-PerCP-Cyanine5.5 (clone 6D5, Biolegend), anti-CD19-biotin (clone 6D5, Sony), anti-CD27-APC (clone LG.3A10, Biolegend), anti-CD27-APCeFluor780 (clone LG.7F9, eBioscience, San Diego, CA), anti-CD27-PE-Dazzle594 (clone LG.3A10, Biolegend), anti-CD27-APC-Cyanine7 (clone LG.7F9, eBio-science), anti-CD45-BV510 (clone 30-F11, Biolegend), anti-CD45-APC-Cyanine7 (clone 30-F11, BD Pharmingen), anti-CD45.2 PE-eFluor610 (clone 104, eBioscience), anti-CD45.1-PE-Cyanine7 (clone A20, Biolegend), anti-CD49a-BUV395 (clone Ha31/8, BD Bioscience), anti-CD49b-BV510 (clone HMα2, BD Optibuild), anti-CD62L-PerCP-Cyanine5.5 (clone MEL-14, Biolegend), anti-CD69-PerCP-Cyanine5.5 (clone H1.2F3, Biolegend), anti-CD117(cKit)-BV510 (clone 2B8, Biolegend), anti-CD117 (ckit)-APC-Cyanine7 (clone ACK2, Sony), anti- CD122 (IL-2Rβ)-eFluor450 (clone TM-β1, eBioscience), anti-CD122 (IL-2R-β)-APC (clone TM- β1, Biolegend), anti-CD127 (IL-7Rα)-PE (clone A7R34, eBioscience), anti-CD127 (IL-7Rα)-PE-Cyanine7 (clone A7R34, Sony), anti-CD135 (Flt3)-PerCP-eFluor710 (clone A2F10, eBioscience), anti-CD135 (Flt3)-PE (clone A2F10, Sony), anti- CD244.2- PE-Cyanine7 (clone m2B4 (B6)458.1, Biolegend), anti-CX3CR1-BV510 (clone SA011F11, Biolegend), anti-CXCR6-PE-Cyanine7 (clone SA051D1, Biolegend), anti-Gr-1-FITC (clone RB6-8C5, Biolegend), anti-Gr-1-APC (clone RB6-8C5, Biolegend), anti-GR1-biotin (clone RB6-8C5, BD Pharmingen), anti-integrin α4β7-BV421 (clone DATK32, BD Horizon), anti-NK1.1-FITC (clone PK136, Biolegend), anti-NK1.1-APC-eFluor780 (clone PK136, eBioscience), anti-NK1.1-BV650 (clone PK136, Sony), anti-NKp46-BV711 (clone 29A1.4, Biolegend), anti-NKp46-V450 (clone 29A1.4, BD Horizon), anti-Sca-1- AlexaFluor700 (clone D7, Biolegend), anti-Sca-1-BV510 (clone D7, Sony), anti-S1PR5-PE (clone 1196A, Bio-Techne, Abingdon, UK), anti-TCRαβ-biotin (clone H57 597, Sony), anti-TCRγδ-biotin (clone GL3, BD Pharmingen), anti-TER119-biotin (clone TER-119, Sony) and streptavidin-PE-Cyanine5 (BD Pharmingen) for surface antigens; and anti-Tbet-eFluor660 (clone eBio4B10, eBio-science) for intracellular Tbet staining.
In Figures 1-4, the lineage cocktail used to identify iNK, mNK1 and mNK2 consisted of anti-CD3-, CD8α-, CD19- and Gr-1-FITC (Lin1). The lineage cocktail used for the identification of LMPP, CLP, preNKP and rNKP in Figure 1 consisted of anti-CD3-, CD8α-, CD19-, Gr-1- and NK1.1-FITC. Lineage cocktail for Figure 5 consisted of anti-CD3-, CD8α-, CD19-, TCRαβ-, TCRγδ-, GR1- and TER119-biotin, stained with streptavidin-PE-Cyanine5. Dead cells were excluded using fixable viability dye eFluor 450 (eBioscience), fixable viability dye eFluor 455 ultraviolet (eBioscience), Propidium iodide (Sigma Aldrich) or DAPI (4’,6-Diamidino-2-Phenylindole, Dilactate; Biolegend). Intracellular staining was carried out using Human FoxP3 Buffer (BD Biosciences) according to the manufacturer’s instructions. Data were acquired on an LSRFortessa II (BD Biosciences) and analyzed using FlowJo (Tree Star, Ashland, OR). Cells were sorted on a FACSAria (BD Biosciences).
RNA sequencing
Total RNA was extracted from sorted iNK, mNK1 and mNK2 bone marrow NK cells using an RNeasy Micro kit (Qiagen, Manchester, UK). cDNA Libraries were prepared from 2ng of total RNA (RIN 3.2-10) using the Ovation Solo assay (Nu-GEN, AC Leek, The Netherlands) with 15 cycles of amplification. Libraries were assessed for correct size distribution on the Agilent 2200 TapeStation and quantified by Qubit DNA High Sensitivity assay (Thermo Fisher Scientific) before being pooled at an equimolar concentration. Sequencing was performed on the Illumina NextSeq 500, generating approximately 12million 75bp single end reads per sample. Differential expression analysis was carried out using SAR-Tools [24] filtering at padj < 0.05.
ACKNOWLEDGEMENTS
This work was funded by Royal Society/Wellcome Trust Sir Henry Dale Fellowship WT105677 (to VM). The authors declare no competing interests.