Nr2f6-deficient mice on the C57BL/6N background were bred in-house under specific pathogen-free conditions (44). The animals were maintained at room temperature with water and food provided ad libitum. The experiments were performed with 8-12 week-old mice, for each individual experiment, mice were age and sex-matched in a non-randomized manner, with a minimum of 3 mice per group. The ethics committee of the Medical University of Innsbruck and the Austrian Federal Ministry of Education Science and Research approved and authorized all animal procedures (GZ: 2021-0.406.862; GZ: 2021-0.406.863).

Adult mice were terminally anesthetized by intraperitoneal (i.p.) injection of 200 µL Ketasol (100 mg/mL, Livisto)/Xylasol (20 mg/mL, Livisto) solution. Bone marrow was harvested from the tibia and femur. Briefly, bones were isolated under sterile conditions; one end of the bone was cut open and placed into a 0.5 ml tube containing a hole in the bottom (18-21 G needle puncture), which was put into a 1.5 ml tube. BM was collected by centrifugation of the isolating device in a table-top centrifuge at 5,000 × g for 20 s (52). The supernatant was sucked off or collected for cytokine measurements. Spleens were homogenized through a 100 μm cell strainer in FACS Buffer (PBS (Sigma Aldrich-Aldrich, P5493-1L) supplemented with 3% FBS, 1% Penicillin/Streptomycin (Sigma-Aldrich, A2213)). Blood was collected via the femoral artery after sacrifice. Red blood cells were lysed in erythrocyte lysis buffer, as described previously (49). Subsequently, single-cell suspension was filtered through a 70µm cell strainer prior to further processing.

If the FC receptors CD16 and CD32 were not specifically targeted, the cells were incubated for 15 min with FcR block (anti-CD16/32; BD Biosciences) in FACS buffer. The cells were then incubated with fluorochrome-conjugated or biotinylated monoclonal antibodies for 30 min at 4°C in FACS buffer. After washing with PBS, we used the Fixable Viability Stain 780 (BD Biosciences, #565388) or Fixable Viability Stain 575V (BD Bioscience, #565694) for 10 min at 4°C to stain dead cells. If necessary, cells were incubated for 30 minutes with streptavidin-conjugated antibodies. The cells were then washed and resuspended in FACS buffer for subsequent flow cytometry analysis. Flow-cytometric analysis was conducted with BD LSRFortessa^™ (BD Biosciences, Germany) or BD FACSCanto II BD Biosciences, Germany), followed by data analysis with the FlowJo software (BD Biosciences, New Jersey, v10.8.1). The flow cytometric data underwent quality control procedures, and any irregularities were removed. Only data of sufficient quality were utilized in subsequent analyses. See Supplementary Table 1 for details of the markers used to identify stem and progenitor populations in the bone marrow and Supplementary Table 2 for the full list of antibodies used for flow cytometry.

12µl of blood was used to investigate blood parameters using a Vet ABC ™Hematology Analyzer.

According to the manufacturer’s protocol, CFU assays were conducted using the mouse methylcellulose complete media (R&D, HSC007). In brief, 2.5x10⁴ bone marrow cells in 1.1 ml of mouse methylcellulose complete media were seeded in 35 mm cell culture dishes (triplicates). Eight days post-seeding, dishes containing CFUs were screened (Olympus CKX53), and movies of the complete dish were made (Jenoptik Optical Systems GmbH, Gryphax, Version 2.2.0.1234). Movies were blinded and randomized and CFUs on the plates were counted manually.

GM-CSF, M-CSF, G-CSF, and IL-1β cytokine levels from mouse bone marrow supernatant (see organ harvest) were analyzed with LEGENDplex™ Kit (mix and match) according to the instruction manual.

Statistical analysis was performed using Prism 8.0 and 10.2.3. Unless otherwise indicated, experiments were repeated at least two times using a minimum of 3 mice per group. The Shapiro–Wilk test evaluated the normality of our data. When normally distributed, we performed statistical analysis with an unpaired Student’s t-test for samples with equal variance (F test) or an unpaired t-test with Welch’s correction for datasets with different variances. A Mann–Whitney U test was used if data were not normally distributed. Triplicates in the CFU assay were tested using a nested t-test. Results are shown as median ± IQR with whiskers from min. to max. Outliers were excluded using ROUT (Q=1%) in GraphPad Prism. A p-value < 0.05 was considered statistically significant. *p < 0.05; **p < 0.01, ***p < 0.001.

In mice, hematopoietic stem cells (HSCs) in the bone marrow are found in the live Lin⁻Sca-1⁺c-Kit⁺ (LSK⁺) fraction, which includes both the hematopoietic stem and progenitor cell (HSPC) and lymphoid-myeloid primed progenitor (LMPP) populations (5). The HSPC population can be further divided into long-term (LT) LT-HSCs (LSK⁺CD135⁻CD48⁻CD150⁺) and short-term (ST) ST-HSCs (LSK⁺CD135⁻CD48⁻CD150^-) (5, 53, 54). HSCs produce distinct parallel subsets of lineage-biased multipotent progenitors (MPPs), which coordinate myeloid and lymphoid lineage outputs in response to hematopoietic demands (5). MPP2s (LSK⁺CD135⁻CD48⁺CD150⁺) and MPP3s (LSK⁺CD135⁻CD48⁺CD150^-) are contained within HSPCs (LSK⁺CD135⁻), in contrast, the MPP4 (CD48⁺CD150^-) population is contained within the LMPP (LSK⁺CD135⁺) population (5).

Analysis of Immgen database data revealed Nr2f6 expression in various bone marrow stem cell populations ( Figure 1A ) (55). To explore the role of NR2F6 in adult mouse hematopoiesis, we isolated the bone marrow from the femurs and the tibias of age and sex-matched wild-type or germline Nr2f6-deficient mice and characterized progenitor populations utilizing flow cytometry ( Supplementary Figure 1A ). We determined the percentage of parent gate, the percentage of total (out of live, single cells), and the live cell numbers of each individual population within the bone marrow of both genotypes. The size and weight of the femur and tibia and the total bone marrow counts were not altered between the genotypes ( Supplementary Figure 1B ).

While the percentages of parent long-term (LT) and short-term (ST) hematopoietic stem cell (HSC) populations were unchanged, the percentages of total LT-HSCs and ST-HSCs were significantly higher in the bone marrow of Nr2f6-deficient mice when compared to wild-type controls ( Figures 1B–D , Supplementary Figure 1C ). Along this line, total cell numbers of ST-HSC were significantly increased, with a trend to an enhanced LT-HSC population in Nr2f6-deficient mice ( Figures 1C, D ).

Within the distinct parallel subsets of lineage-biased multipotent progenitors (MPP) derived from the HSC compartment, we did not observe differences in percentages of total and MPP2 cell numbers in the bone marrow of Nr2f6-deficient mice. However, the percentage of parent MPP2 population was reduced in Nr2f6-deficient mice compared to wild-type controls ( Figure 1E ). In contrast, we did not observe an alteration in the percentage of parent MPP3s, although the percentage of total MPP3s significantly increased compared to the wild-type ( Figure 1F ). Consequently, although not statistically significant, we also observed a trend to enhance total MPP3 cell numbers in the bone marrow of Nr2f6-deficient mice ( Figure 1F ). No differences between the genotypes were observed within the MPP4 population, defined in the fms-like tyrosine kinase 3 (FLT-3; CD135) positive LMPP compartment ( Figure 1G ). Along this line, MFI levels of CD48 and CD150 in all populations were similar between the genotypes ( Figure 1H , Supplementary Figure 1D ). To assess the proliferation capacity of these single stem cell populations, we performed Ki-67 staining. Our results demonstrated a significant reduction in proliferation in the Nr2f6-deficient LT and ST-HSC populations ( Supplementary Figures 1E, F ). A comparable trend, although not reaching statistical significance, was observed in MPP2, MPP3, and MPP4 populations ( Supplementary Figure 1F ). In parallel to Ichim et al. (42), we investigated Lin⁺CD11b⁺ cells within the BM of Nr2f6-deficient mice and observed an increase in the percentage of total but not cell numbers of the Lin⁺CD11b⁺ population ( Supplementary Figure 1G ).

The alterations observed in the MPP populations of Nr2f6-deficient mice prompted us to ask if the output among hematopoietic lineages was altered by loss of NR2F6. Therefore, we investigated potential changes in hematopoietic stem and progenitor cells (HSPC), the lymphoid-primed multipotent progenitor (LMPP), and Lin⁺Sca-1⁺c-Kit⁺ (LSK) populations in the bone marrow.

Notably, the percentages of parent and total HSPCs were significantly enhanced in Nr2f6-deficient mice, resulting in a trend to increased total HSPC cell numbers, albeit not statistically significant, compared to wild-type controls ( Figures 2A, B ).

In contrast, the percentage of the parent LMPP population was significantly reduced in Nr2f6-deficient mice, but the percentage of total and cell numbers were similar to wild-type, independent of CD135 expression (MFI) levels ( Figures 2A, C , Supplementary Figure 2A ).

The shift in percentages of total HSPCs populations, namely LT-HSCs, ST-HSCs and MPP3 cells, but not MPP2 cells, results in increased percentages of parent and total LSK⁺ cells in the bone marrow of Nr2f6-deficient mice ( Figures 2A, D ). The expression levels (MFI) of CD117 and Sca-1 within the LSK⁺ populations were comparable between the genotypes ( Figure 2E ).

We did not observe differences in the Lin^-CD127^- pre-gate ( Supplementary Figure 2B ). Additionally, the percentage of parent and total as well as total cell numbers of the Sca-1^-c-Kit⁺ (CD117⁺) population were similar between the genotypes ( Figures 2A, F ). Ki-67 staining revealed no significant differences in Ki-67 levels within HSCPC LMPP and LSK+ cells ( Supplementary Figure 2C ). Taken together, the absence of NR2F6 leads to an increase in the percentage of total LT and ST-HSCs and MMP3 cells, resulting in an increased percentage of total HSPCs and LSK⁺ cells ( Supplementary Figure 2D ).

Whereas the common lymphoid progenitor (CLP) population (CD135⁺CD127⁺) is derived from Lin⁻Sca-1⁺c-Kit^lo/int cell population, common myeloid progenitors (CMPs) are contained in the Lin^-CD127^-Sca-1^-c-Kit⁺CD115^- cell fraction and give rise to granulocyte-monocyte (macrophage) progenitors (GMP CD16/32⁺CD34^+/int) and megakaryocyte-erythroid progenitors (MEP CD16/32^-CD34^-) (1, 5, 56). Furthermore, CMPs give rise to the monocyte-dendritic cell progenitor (MDP) compartment (CD135⁺CD117⁺), which contribute together with GMPs to the common monocyte progenitors (cMoP) (CD135^-Ly6C⁺) that can differentiate into tissue-specific macrophages in case needed (17, 19, 26, 27, 57, 58).

NR2F6 has been shown to inhibit hematopoietic cell differentiation (42). Consequently, we examined the early hematopoietic common myeloid (CMP) and common lymphoid progenitors (CLP) in the bone marrow of wild-type and Nr2f6-deficient mice. CMPs, megakaryocyte-erythrocyte progenitors (MEP) and granulocyte-macrophage progenitors (GMP) express high levels of CD117 but not Sca-1 or CD115. The percentage of parent and total as well as total cell numbers of Sca-1^-CD117^highCD115^- cells were similar between the genotypes ( Supplementary Figure 2E ). Despite an increase in the percentage of parent CMP population in Nr2f6-deficient bone marrow, the percentage of total and total cell numbers of CMPs were unaltered compared to wild-type controls ( Figures 3A, B , Supplementary Figure 2F ). CD34 (MFI) expression within CMPs was similar between the genotypes ( Figure 3B ).

In contrast, GMPs exhibited a significantly reduced percentage of parent and total cells, with correspondingly decreased total cell numbers in the bone marrow of Nr2f6-deficient mice compared to wild-type controls ( Figures 3A, C ). While CD34 expression was unaltered, the MFI of CD16/32 was significantly reduced in the Nr2f6-deficient GMP progenitors, potentially contributing to the diminished GMP cell population ( Figures 3C–E ). Conversely, MEP populations were unaltered between the genotypes ( Figures 3A, F ).

We further investigated whether the CD34^high or CD34^lo-int GMP population contributed to the GMP decrease. Within the GMPs, the percentage of CD34^lo-int cells was elevated in Nr2f6-deficient compared to control mice ( Supplementary Figure 2G, H ). However, the percentage of total was similar between the genotypes, and cell counts were significantly reduced in the CD34^high populations in Nr2f6-deficient mice and CD34^lo-int GMPs showed a similar trend, suggesting that both cell populations contribute to the decrease in overall GMPs ( Supplementary Figures 2G, I ).

In contrast to expectations, CMPs from Nr2f6-deficient bone marrow showed lower proliferation than those from controls. A comparable trend, although not statistically significant, was observed in GMPs, while MEP from both genotypes showed similar Ki-67 staining ( Supplementary Figure 2J ).The percentage of parent and total, as well as the total cell numbers of CLPs in the bone marrow of Nr2f6-deficient mice, were similar to wild-type controls ( Figures 3G, H , Supplementary Figure 3A ). This gating strategy offered the opportunity to define the natural killer (NK) pre-pro NKa progenitor population (CD135⁺CD127^-) (59–61), which was unaltered in Nr2f6-deficient mice when compared to wild-type ( Figure 3G , Supplementary Figures 3A, B ). We also determined the MFI of CD127 and CD135 within the Nr2f6-deficient CLP population, which was not altered for CD127 but reduced for CD135 compared to wild-type ( Supplementary Figure 3C ).

To further specify the impact of NR2F6 on myeloid progenitor populations, we conducted a detailed analysis of the monocyte-dendritic cell progenitors (MDP) and common dendritic cell progenitor (CDP) populations via investigating CD117⁺ versus CD117^lo-int populations within the Lin⁻CD11b^-CD115⁺CD135⁺ cell population (26, 27), ( Supplementary Figures 3D, E ).

In Nr2f6-deficient mice, the percentages of parent, total, and total CD135⁺CD117⁺ MDP cell numbers were significantly enhanced ( Figures 4A, B ). In contrast, we observed no difference in the percentages of parent, total, or total cell numbers of CD135⁺CD117^lo-int CDPs in Nr2f6-deficient mice when compared to wild-type ( Figure 4C ). The MFI of CD117 and CD135 were similar within both genotypes ( Figure 4D ).

Analyzing the common monocyte progenitors (cMoP) within the Lin⁻CD11b^-CD115⁺ CD117⁺CD135^- cell population ( Supplementary Figures 3D–F ), we found no differences in percentages of parent or total cell numbers despite enhanced percentage of total cells in Nr2f6-deficient mice ( Figure 4E ). The same pattern was observed in the pre-gate within the CD135CD117⁺ population ( Supplementary Figure 3G ). Ki-67 staining in MDPs, CDPs, and cMoPs was comparable between genotypes ( Supplementary Figure 3H ).

MDPs are also able to generate common dendritic cell (DC) progenitors (CDP) (CD135⁺CD117^-), which produces pre-conventional (c)DCs and plasmacytoid (p)DCs (23, 27, 28).

To investigate potential alterations beyond the CDP population, we assessed pre-DCs with potential for conventional (c)DC differentiation (Lin⁻CD11b^-CD11c⁺MHC-II^-CD135⁺CD172a^- Ly6C⁺SiglecH⁺ & Ly6C^-SiglecH⁺) as well as pre-DCs committed to the cDC1 lineage (CD135⁺CD11c⁺CD117^int/CD226⁺ Ly6C⁻SiglecH⁻) and pre-DCs committed to the cDC2 lineage (Lin⁻CD11b^-CD11c⁺MHC-II^-CD135⁺CD172a^- Ly6C⁺SiglecH⁻) (23, 24, 27). ( Supplementary Figures 4A, B ). Analysis of Nr2f6-deficient BM revealed no alterations in percentages of parent, total and total cell numbers in the CD135⁺CD11c⁺ population, as well as the pre-cDC1 progenitors compared to wild-type controls ( Supplementary Figures 4B–D ). The Ly6C⁻SiglecH⁺ and Ly6C⁺SiglecH⁺ pre-DCs precursor populations were also unaltered between the genotypes ( Figures 4F–H ). However, the Nr2f6-deficient Ly6C⁺SiglecH⁻ population (pre-DCs committed to the cDC2 lineage) was significantly reduced in both percentages of parent and total as well as total cell numbers ( Figures 4F, I ). Despite these changes, the proliferative capacities of pre-DCs committed to either the cDC1 or cDC2 lineage were comparable between genotypes ( Supplementary Figure 4E ).

In conclusion, the loss of NR2F6 reduced the GMP but not the CMP or CLP progenitor population. In contrast, the MDP progenitors but not CDP and cMoP progenitor populations were increased in Nr2f6-deficient mice. In addition, the loss of NR2F6 reduced the pre-cDC2 but not the pre-cDC1 population ( Supplementary Figure 4F ).

To investigate potential changes in the bone marrow microenvironment, we measured cytokine concentrations of G-CSF, M-CSF, and IL-1β in the bone marrow supernatant. While there was a slight trend towards reduced G-CSF levels, no significant differences in M-CSF and IL-1β concentrations were observed between wild-type and Nr2f6-deficient mice ( Figure 5A ).

To determine how Nr2f6 affects hematopoiesis independently of the bone marrow environment, an in vitro colony-forming unit (CFU) assay was performed. The cells can differentiate into burst-forming units-erythroid (BFU-E), colony-forming unit-granulocytes (CFU-G), colony-forming unit-macrophages (CFU-M), colony-forming unit-granulocyte, macrophage (CFU-GM) and colony forming unit-granulocyte, erythrocyte, macrophage, megakaryocyte (CFU-GEMM). CFU-G are clonogenic progenitors of granulocytes that give rise to a homogeneous population of eosinophils, basophils, or neutrophils, while CFU-M are clonogenic progenitors of macrophages that give rise to a homogenous population of macrophages. CFU-GM arise from earlier progenitors that give rise to colonies containing a heterogeneous population of macrophages and granulocytes. CFU-GEMM originates from multi-lineage progenitors that give rise to erythroid, granulocytes, macrophages, and megakaryocyte lineage.

The bone marrow of Nr2f6-deficient mice exhibited an unexpected increase in the total number of colonies that formed ( Figure 5B ). However, the number of BFU-E, CFU-M and CFU-GM did not vary between the genotypes. The observed differences were attributed to a significant increase in CFU-GEMM, as well as a notable, although not statistically significant, increase in CFU-G ( Figure 5C ).

As a next step, we investigated how alterations in hematopoiesis and myelopoiesis influence the myeloid compartment in the bone marrow, blood, and spleen and of adult Nr2f6-deficient mice.

In the bone marrow, the percentages of parent and of total CD45⁺ CD11b⁺ myeloid cells, CD115⁺ monocytes, and both classical Ly6C^hi and non-classical Ly6C^mid-lo monocytes were comparable between genotypes ( Figures 5D–H , Supplementary Figure 5A ). The percent of parent Ly6G⁺ neutrophils was enhanced, but no alterations were observed within the CD45⁺ cell population ( Supplementary Figures 5A, B ). Notably, the Bst2⁺B220⁺ plasmacytoid dendritic cells (pDCs) percentage of parent and CD45⁺ in the bone marrow was reduced ( Figure 5I, J , Supplementary Figure 5A ).

In the blood, the proportions of granulocytes, eosinophils, lymphocytes, or red blood cells were similar between the genotypes using a Vet ABC ™Hematology Analyzer ( Figure 6A ).

In more detail, we investigated CD115⁺ monocytes, including classical Ly6C^hi and non-classical Ly6C^mid-lo monocytes, as well as Ly6G⁺ neutrophils and Bst2⁺B220⁺ pDCs in the blood of wild-type and Nr2f6-deficient mice ( Supplementary Figure 5C ). The percentage of total monocytes (CD115⁺) within the CD45⁺ cell population remained unaltered ( Figures 6B, C ), but the ratio of Ly6C^hi to Ly6C^mid-lo was significantly altered ( Supplementary Figure 5D ), resulting in enhanced non-classical monocytes within CD45⁺ cells when compared to wild-type ( Figure 6D ). We investigated CCR2 expression to detect inflammatory Ly6C^hiCCR2⁺ monocytes but could not detect this subset during steady-state in either genotype ( Figure 6B ). Analysis of Ly6G⁺ neutrophils and Bst2⁺B220⁺ pDCs revealed no differences in the percent of parent and of CD45⁺ cells between genotypes ( Figures 6E–H , Supplementary Figure 5D ).

In the spleen of Nr2f6-deficient mice, the percentages of total monocytes, as well as Ly6C^hi and Ly6C^mid-lo monocytes, were similar to those of the controls ( Figures 7A–D , Supplementary Figure 5E ). However, Ly6G⁺ neutrophils were reduced, while Bst2⁺B220⁺ pDCs were elevated compared to the controls ( Figures 7E–H , Supplementary Figure 5E ). Details on dendritic and macrophage populations in the spleen of Nr2f6-deficient mice are currently under revision elsewhere (Woelk-J et al. and Benz-J et al.).

In summary, Nr2f6-deficiency results in reduced pDCs in the bone marrow and enhanced non-classical monocytes in the blood. In the spleen, reduced neutrophils but increased pDC populations were detected in Nr2f6-deficient mice.