Abstract

   While lymphoid progenitors have demonstrated unexpected plasticity in
   vivo, their differentiation into myeloid cells under in vitro
   conditions has been largely dismissed as an artifact or biologically
   irrelevant. Consequently, the functional properties of these cells
   remain poorly characterized. In this study, we show that cultured
   common lymphoid progenitors (cCLPs) differentiate into CD11b⁺CD115⁺
   monocytic cells (cCLP-Ms) via IL-6-C/EBPβ signaling. Molecular and
   phenotypic analyses revealed that cCLP-Ms acquire essential features of
   myeloid cells, including innate immune sensor expression and phagocytic
   capacity, while retaining unique characteristics distinct from bone
   marrow-derived macrophages (BMDMs), such as reduced MHC class II
   expression and TNF-α production. Functionally, cCLP-Ms exhibit
   immunoregulatory properties, effectively suppressing IgE-mediated
   cutaneous allergic inflammation upon adoptive transfer. These findings
   highlight the plasticity of lymphoid progenitors and establish a robust
   platform for investigating the mechanisms underlying myeloid
   differentiation. This system deepens our understanding of hematopoietic
   cell lineage flexibility and offers a foundation for exploring
   therapeutic applications in immune regulation and inflammation.

   Subject terms: Myelopoiesis, Differentiation

Introduction

   Hematopoiesis has traditionally been conceptualized as a hierarchical
   process where hematopoietic stem cells (HSCs) differentiate into
   distinct blood cell lineages through progressively restricted
   progenitor states. In this classical model, HSCs give rise to
   multipotent progenitors (MPPs), which further differentiate into
   myeloid lineages including common myeloid progenitors (CMPs), and
   lymphoid lineages including common lymphoid progenitors (CLPs) through
   lymphoid-primed multipotent progenitors (LMPPs) [[34]1, [35]2].
   However, recent advances in single-cell analysis technologies have
   fundamentally challenged this paradigm, revealing that early
   hematopoiesis is better described as a continuous process where stem
   and progenitor cells gradually acquire lineage-specific characteristics
   [[36]2, [37]3]. This revised understanding emphasizes the heterogeneity
   and plasticity inherent in hematopoietic differentiation, suggesting
   that lineage commitment is more flexible and complex than previously
   appreciated.

   Monocytes and macrophages are essential components of the innate immune
   system, serving as critical sentinels in host defense against pathogens
   [[38]4]. These cells express various pattern recognition receptors
   (PRRs), including Toll-like receptors (TLRs), NOD-like receptors
   (NLRs), RIG-like helicase (RLHs), and components of the inflammasome,
   which enable them to recognize pathogen-associated molecular patterns
   (PAMPs) and danger-associated molecular patterns (DAMPs) [[39]5]. Upon
   activation of these innate sensors, monocytes and macrophages initiate
   inflammatory responses through the production of cytokines and
   chemokines [[40]4, [41]5]. Conventionally, monocytes and macrophages
   arise from myeloid progenitors through a well-defined pathway involving
   CMPs, granulocyte-macrophage progenitors (GMPs), and monocyte-dendritic
   cell progenitors (MDPs). The differentiation and function of these
   cells are regulated by key transcription factors such as PU.1, C/EBPα,
   and IRF8, while cytokines including M-CSF, GM-CSF, and IL-34 control
   their survival and functional maturation [[42]1, [43]4]. BMDMs,
   generated by culturing bone marrow cells with M-CSF, have served as a
   valuable model system for studying macrophage biology and function in
   numerous research applications [[44]6–[45]8].

   Accumulating evidence has revealed unexpected plasticity in lymphoid
   progenitor populations, challenging the traditional view of strict
   lineage restriction. While LMPPs are known to retain myeloid potential
   [[46]9], the extent of this plasticity in more committed lymphoid
   progenitors has been a subject of intensive investigation. CLPs,
   initially identified by Lineage-marker-negative (Lin^-)
   cKit^loSca-1^loIL-7Rα^hi and subsequently refined as
   Lin^-cKit^intSca-1^intIL-7Rα^+Flk2^+(CLP^f), lack any myeloid potential
   and equally give rise to all lymphocytes including T, B, and natural
   killer (NK) cells [[47]10, [48]11]. Notably, within re-defined
   Lin^-Flk2^+IL-7Rα^+ CLP subsets, CD27^+Ly6d^- all-lymphoid progenitors
   (ALPs) retain low levels of myeloid differentiation potential under
   physiological conditions, whereas more committed CD27^+Ly6d^+
   B-cell-biased lymphoid progenitors (BLPs) lack this capability
   [[49]12]. Furthermore, even committed lymphoid progenitors have shown
   the ability to adopt myeloid fates in vivo as well as in vitro
   [[50]13–[51]16], although the functional properties and physiological
   relevance of such cells remain largely unexplored.

   To facilitate detailed investigation of lymphoid progenitor plasticity,
   we previously established a novel primary culture system enabling the
   long-term expansion of murine Lin^-Flk2^+IL-7Rα^+CD27^+Ly6d^-
   uncommitted CLPs under defined serum-free conditions supplemented with
   specific cytokines, including Flt3 ligand (Flt3L) and IL-7, without any
   genetic manipulation. This system not only maintains the lymphoid
   differentiation potential of these cultured CLPs (cCLPs) but also
   provides several unique advantages for studying cellular plasticity and
   differentiation mechanisms. The use of cCLPs enables precise tracking
   of differentiation processes and allows for detailed molecular
   mechanistic studies that would be technically challenging with CLPs in
   vivo or ex vivo due to their rarity.

   Our previous studies revealed that these cCLPs can differentiate into
   myeloid cells with phagocytic activities in vitro. The present study
   aims to comprehensively characterize these cCLP-derived myeloid cells,
   focusing on their phenotypic features, molecular regulation for their
   differentiation, and functional properties.

Results

IL-6 is critical for the generation of CD11b^+CD115^+ monocytic cells from
cCLPs

   We previously demonstrated that cCLPs can induce the development of
   CD11b^+ myeloid cells with phagocytotic function when exposed to a
   complex cytokine mixture [[52]17]. To delineate the key factors
   responsible for driving myeloid lineage differentiation, we evaluated
   individual cytokines and their combinations using 3 weeks-cultured
   cCLPs freshly generated from Lin^-Flk2^+IL7Rα^+CD27^+Ly6d^- bone marrow
   cells. Strikingly, IL-6 emerged as the primary driver of CD11b⁺CD115⁺
   cell development, surpassing the effects of canonical myeloid factors
   such as M-CSF (Fig. [53]1a, b). In the presence of Flt3L (FL), which is
   essential for cCLP survival [[54]17], IL-6 induced the generation of
   CD11b⁺CD115⁺ cells, reaching approximately 12% of the population by day
   3. This population expanded further, achieving numbers over the initial
   cCLP input by day 6 (Fig. [55]1c). Consistent findings were also
   observed in cryopreserved cCLPs maintained for 2 months in culture
   (Supplementary Fig. [56]S1a). These cCLPs retained the differentiation
   potential to lymphoid cells such as B cells as previously reported
   (Supplementary Fig. [57]S1b) [[58]17]. The differentiation of these
   CD11b⁺ myeloid cells was substantially affected by the type of FBS
   used, whereas B cell differentiation was less influenced by FBS
   (Supplementary Fig. [59]S1c). The myeloid differentiation process was
   characterized by reciprocal regulation of key lineage markers:
   downregulation of IL-7Rα, essential for lymphoid identity, concurrent
   with upregulation of CD115, a defining marker of the
   monocyte/macrophage lineage. The addition of SCF enhanced this
   phenotypic conversion (Fig. [60]1d, e). Furthermore, supplementation
   with all-trans retinoic acid (ATRA) augmented both the frequency and
   absolute numbers of CD11b⁺CD115⁺ cells, suggesting a synergistic effect
   in promoting myeloid differentiation (Supplementary Fig. [61]S1d, e).
   Mechanistically, we found that cCLPs express both components of the
   IL-6 receptor complex—IL-6Rα and gp130—enabling direct response to IL-6
   stimulation. Time-course analysis revealed rapid STAT3 phosphorylation
   following IL-6 exposure, indicating active signal transduction through
   this pathway (Fig. [62]1f, g). These results suggest that IL-6
   signaling initiates the myeloid differentiation program by modulating
   the balance between lymphoid maintenance (IL-7Rα) and myeloid
   specification (CD115) pathways.

Fig. 1. IL-6 rather than M-CSF efficiently induces a CD11b^+CD115^+ myeloid
lineage from cCLPs.

   [63]Fig. 1
   [64]Open in a new tab

   a–c cCLPs generated from Lin^-Flk2^+IL7Rα^+CD27^+Ly6d^- cells in murine
   bone marrow were maintained in cCLP medium for 3 weeks prior to the
   following experiments. a Representative FACS profiles of CD11b and
   CD115 expression 3 days after culturing 3000 cCLPs with Flt3L (FL)
   alone (upper panel) or together with IL-6 (FL/IL-6) (lower panel). The
   percentage in the gate is shown. b, c Cell number (upper panel) and
   percentage (lower panel) of CD11b^+CD115^+ cells on day 3 after culture
   with the indicated cytokines (n = 3 each) or at the indicated time
   points (n = 4 each) with FL (open squares) or FL/IL-6 (closed squares).
   d Representative FACS profiles on IL-7Rα and CD115 expression on day 7
   in cCLPs cultured with the indicated cytokines. The percentages in the
   gate are shown. e Percentage of CD115^+IL-7Rα^- cells (n = 3 each) in
   (d). f Histograms of IL-6Rα (upper panel) and gp130 (lower panel)
   expression (red histograms) on cCLPs, with isotype control shown in
   gray. Mean fluorescence intensity (MFI) values are shown in each panel.
   g pSTAT3 expression in cCLPs at 0 (gray) or 15 min (red) after the
   addition of IL-6 (100 ng/ml). The dotted line represents the isotype
   control. h May-Grünwald Giemsa staining images of cCLPs (control, left
   panel) or CD11b^+CD115^+Ly6C^+CCR2^+ cells isolated by FACS 4 days
   after cCLP culture with FL/IL-6/SCF (middle panels). The right panels
   show a representative image of CD11b^+CD115^+Ly6G^-Ly6C^+CCR2^+
   monocytes isolated ex vivo (BMmono) or cultured macrophages (BMDMs)
   from the bone marrow of C57BL/6 mice. Scale bar = 10 μm. i cCLPs were
   cultured with 10 ng/mL of FL/IL-6/SCF for 4 days. A representative FACS
   profile of indicated molecules. The percentages in the gate and
   histograms are shown. Data are mean ± SD with statistical significance
   determined by one-way ANOVA (in b, e) or two-way ANOVA (in c). The
   p-values are represented as **, <0.01; ***, <0.001; ****, <0.0001.
   n.s., not significant. All data are representative of at least two
   independent experiments.

   The resulting CD11b⁺CD115⁺ cells exhibited classical
   monocyte/macrophage characteristics. Morphological analysis using
   May-Grünwald-Giemsa staining revealed cells with kidney-shaped nuclei
   and abundant cytoplasm, hallmark features of monocytes (Fig. [65]1h,
   upper panels). A subset of cells displayed even more extensive
   cytoplasmic development, resembling BMDMs (Fig. [66]1h, lower panels).
   Flow cytometric analysis confirmed a surface marker profile typical of
   monocytes: Ly6C⁺^/⁻Ly6G⁻CCR2⁺MHC class II⁻ (Fig. [67]1i). Furthermore,
   the in vivo transfer experiment showed that cCLPs gave rise to
   CD11b⁺CD115⁺ cells in LPS-treated mice, but not in PBS-treated mice,
   ruling out that the myeloid differentiation potential of cCLPs is due
   to specialized culture conditions. (Supplementary Fig. [68]S1f).
   Collectively, these findings demonstrate that IL-6 signaling can direct
   lymphoid progenitors toward a monocytic fate, generating cells that
   closely resemble conventional monocytes/macrophages in both phenotype
   and morphology.

Lymphoid progenitors show myeloid potential ex vivo under specific cytokine
milieu

   To determine whether the observed lineage plasticity represents a
   general property of lymphoid progenitors rather than a culture-induced
   phenomenon, we examined the myeloid potential of freshly isolated
   progenitor populations, uncommitted CLPs, and LMPPs from mouse bone
   marrow. When cultured with FL/IL-6/SCF, both primary populations
   generated CD11b⁺CD115⁺ cells within 4 days. These cells exhibited the
   characteristic CCR2⁺Ly6C⁺^/⁻ monocytic phenotype, mirroring our
   observations with cCLPs (Fig. [69]2a). Importantly, both progenitors
   generated greater number of CD11b^+CD115^+ cells under FL/IL-6/SCF
   conditions rather than with M-CSF alone (Fig. [70]2b). Limiting
   dilution analysis further demonstrated that LMPPs had a higher clonal
   potential for differentiation into CD11b^+CD115^+ cells than CLPs (Fig.
   [71]2c). Overall, these results suggest that primary lymphoid
   progenitors possess intrinsic myeloid potential that can be accessed
   under specific cytokine conditions.

Fig. 2. Lymphoid progenitors from murine bone marrow can differentiate to
myeloid lineages upon culture with FL/IL-6/SCF.

   [72]Fig. 2
   [73]Open in a new tab

   LMPPs and uncommitted CLPs were isolated from the bone marrow of
   C57BL/6 mice and cultured with 10 ng/mL of FL/IL-6/SCF or 10 ng/ml of
   M-CSF for 3 days. a A representative FACS profile of indicated
   molecules before (left panels) and after (middle and right panels)
   culture of 1000 LMPPs (upper panels) or CLPs (lower panels) with
   FL/IL-6/SCF. The percentages in the gate are shown. b The number (upper
   panel) and frequency (lower panel) of CD11b^+CD115^+ cells after
   culture (n = 3, each). c Limiting dilution assay. LMPPs (closed
   symbols) and CLPs (open symbols) serially diluted from 16 cells to 1
   cell by 2-fold are shown. Data are mean ± SD with statistical
   significance determined by two-way ANOVA with multiple comparisons. The
   p values are represented as ***, <0.001. n.s. not significant. All data
   are representative of at least two independent experiments.

Comprehensive gene expression profiling characterizes that cCLP-derived
monocytic cells are closely related to monocytes/macrophages

   To assess the transcriptional signatures of cCLP-derived myeloid cells,
   RNA-seq-based whole transcriptome analysis was performed to compare the
   gene expression profile with that of BMDMs from cell culture as well as
   of CD11b^+Ly6G^+ neutrophils (BMneu) or CD11b^+Ly6C^hi monocytes
   (BMmono) isolated ex vivo from bone marrow (Supplementary Fig.
   [74]2a–d). In a principal component analysis (PCA) of the whole
   transcriptome, comprising 15,074 expressed genes and accounting for
   approximately 70% of the variance across the PC1 and PC2 axes, the cell
   population derived from cCLPs a day after the induction of
   differentiation with FL/IL-6/SCF (D1) was located almost in the same
   position as cCLPs. In contrast, the population 3 days after the
   induction of differentiation (D3) was positioned closest to BMDMs and
   nearer to BMmono than to BMneu. These results suggest that cCLPs
   cultured in the presence of FL/IL-6/SCF acquire a myeloid gene
   expression profile closely aligned with the monocyte lineage (Fig.
   [75]3a). To place these findings in a broader hematopoietic context, we
   integrated our data with a comprehensive mouse hematopoiesis database
   ([76]GSE116177) [[77]18]. This expanded analysis confirmed our previous
   observation that unstimulated cCLPs closely resembled fresh CLPs
   [[78]17]. More importantly, it revealed that D3 cells clustered with
   cultured macrophages and bone marrow monocytes from both our
   preparation and the reference dataset, further validating their
   monocytic identity (Fig. [79]3b).

Fig. 3. cCLP-derived myeloid cells display gene expression signatures
resembling monocytes/macrophages.

   [80]Fig. 3
   [81]Open in a new tab

   a PCA of whole transcripts of 15,074 genes with CPM values greater than
   1 in either sample across all samples from cCLPs, and those culture
   with FL/IL-6/SCF for 1 day (D1) or 3 days before the isolation of
   CD11b^+CD115^+CCR2^+Ly6C^+Ly6G^- (D3), together with BMDMs, BMmono, and
   BMneu. n = 3, each. b PCA plots of whole transcripts in indicated cell
   populations merged with a publicly available database of murine
   hematopoietic cells ([82]GSE116177). c k-means clustering analysis of
   6500 genes with SD > 1 across all samples. The color bar on the left
   shows each cluster; A (n = 773), B (n = 502), C (n = 1387), D
   (n = 936), E (n = 1255), F (n = 259), G (n = 998), H (n = 390). d The
   expression levels of the indicated genes derived from RNA-seq data
   representative in Cluster D (upper panels) and Cluster G (lower
   panels). The values are shown as log[2] (CPM + 4). Data are mean ± SD
   with statistical significance determined by one-way ANOVA with multiple
   comparisons. The p values are represented as *, <0.05; **, <0.01; ***,
   <0.001; ****, <0.0001. n.s., not significant. Enrichment analysis of
   terms (e, h) and cell population (f, i) by Metascape and the heatmap
   analysis of gene expression related to macrophages (g, j) enriched in
   (f) or (i) of Cluster D or G, respectively. Color code values indicate
   the Z-score for each gene across samples showing the log[2] fold change
   for each gene.

   Unsupervised k-means clustering identified eight distinct gene
   expression patterns including particularly informative clusters
   associated with the differentiation process (Fig. [83]3c). Cluster D
   (936 genes) captured a macrophage-specific signature shared between D3
   cells and BMDMs. This included Spp1, encoding a secreted factor
   characteristic of macrophages [[84]19], and several genes associated
   with immunosuppressive, M2-like macrophages, such as the transcription
   factor Maf (also known as c-Maf) [[85]20] and the surface marker Mrc1
   (Fig.[86]3d). Cluster G (998 genes) represented a broader myeloid
   program shared among D3 cells, BMDMs, BMmono, and BMneu (Fig. [87]3c).
   This cluster encompassed key regulators of both lymphoid and myeloid
   differentiation, including the master transcription factor PU.1 (Spi1).
   It also contained numerous components of innate immune sensing
   pathways: inflammasome proteins (Nlrp3, Nlrc4), intracellular pattern
   recognition receptors (Nod2, Ddx58), and multiple Toll-like receptors
   (Tlr1, Tlr2, Tlr4, Tlr6, and Tlr13) (Fig. [88]3d and Supplementary Fig.
   [89]3). Further enrichment analysis revealed that genes related to
   ‘Lysosome’ and ‘Ion homeostasis’ were enriched in Cluster D (Fig.
   [90]3e), with PaGenBase enrichment analysis identifying 36 genes as
   macrophage-enriched, including P2ry2 and Pparg (Fig. [91]3f, g).
   Similarly, Cluster G genes were associated with ‘Innate immune
   response’ (Fig. [92]3h) and included 35 macrophage-enriched genes such
   as Irf7 and Mx1, critical for antiviral responses (Fig. [93]3i, j).
   Importantly, the majority of macrophage-related genes enriched in
   Cluster D and Cluster G were expressed at low levels in cCLPs and D1
   cells (Fig. [94]3g, j). Collectively, these results suggest that
   cCLP-derived myeloid cells induced by FL/IL-6/SCF over three days
   acquire a gene expression program characteristic of innate immune
   responses, including pathogen recognition sensors and lysosomal
   molecules involved in intracellular digestion. This gene expression
   signature, resembling that of monocytes/macrophages, indicates
   differentiation from the lymphoid to myeloid lineage at the
   transcriptional level. Therefore, we refer to these cells as cCLP-Ms
   (cCLP-derived monocytic cells).

Investigation of transcriptional regulators critical for cCLP-M induction
identifies C/EBPβ

   To identify the transcriptional regulators driving cCLP-M
   differentiation, we performed differential expression analysis
   comparing cCLPs with cells after one day of IL-6 stimulation (D1). This
   analysis revealed 216 upregulated and 177 downregulated genes (fold
   change >2 or <0.5, p < 0.05) in D1 cells (Fig. [95]4a). GO pathway
   enrichment analysis indicated that 87 of the upregulated genes were
   associated with the GO biological process “Cell differentiation”
   (Supplementary Tables [96]1, [97]2). Remarkably, 57 of these genes
   maintained elevated expression in fully differentiated cCLP-Ms (D3),
   representing a core set of differentiation-associated factors (Fig.
   [98]4b). Among these 57 candidate genes, we focused on the
   transcription factors Id1 and C/EBPβ (Fig. [99]4c), both of which have
   been previously implicated in IL-6-mediated responses [[100]21]. The
   expression levels of these genes were upregulated in D1 (Id1: ~8-fold,
   Cebpb: ~8-fold) and further increased in cCLP-Ms (D3) (Id1: ~32-fold,
   Cebpb: ~400-fold), reaching levels comparable to those observed in
   BMDMs, BMmono, and BMneu (Fig. [101]4d). To test the functional
   requirement for these factors, we employed CRISPR-Cas9-mediated gene
   editing to target Id1, Cebpb, or its relative Cebpa in cCLPs. Following
   clonal expansion (Supplementary Fig. [102]4a), we observed distinct
   effects on cell proliferation and survival. While control (mock) and
   Id1-targeted clones showed robust expansion with all clones exceeding
   100 cells, Cebpa-targeted clones showed reduced viability, with only
   89% (16/18 clones) achieving similar expansion. Cebpb-targeted clones
   were even more severely affected, with only 50% (9/18 clones) showing
   substantial growth (Supplementary Fig. [103]4b).

Fig. 4. Investigation of transcriptional regulators critical for cCLP-M
induction identifies C/EBPβ.

   [104]Fig. 4
   [105]Open in a new tab

   a Scatter plot of DEG (FDR < 0.05, >2-fold or <0.5- fold) between
   undifferentiated cCLPs and cCLPs after 24 h incubation with FL/SCF/IL-6
   (D1). b Venn diagram of genes up-regulated in D1 to cCLPs with
   enrichment analysis of “Cell Differentiation” by GO Biological process
   and those up-regulated in D3 to cCLPs. c Heatmap view of 57 genes
   overlapped in (b). Color code values indicate the Z-score for each gene
   across samples. d Gene expression of Cebpb (left panel) and Id1 (right
   panel) in indicated samples. The values are shown as log[2] (CPM + 4)
   from RNA-seq data. e Representative FACS plots of cCLP-Ms
   differentiated from single cCLP clones where indicated genes were
   targeted by Crispr/Cas9. The name of a representative clone targeted to
   each gene is shown (＃2–3, ＃2-17, #2-16). f Frequency of wells with more
   than 100 live cells in (e). g Frequency of wells with less than 40%
   CD11b^+CD115^+ cells. h Representative histogram of GFP expression
   (left panels), plots for CD11b and CD115 expression on GFP^- gate
   (middle panels), or GFP^+ gate (right panels) in cCLPs on day 4 after
   retroviral transduction with pMIG empty vector (upper panels) or pMIG/
   C/EBPβ (lower panels). i Percentages of CD11b^+CD115^+ cells in the
   indicated fractions in (h). Data are mean ± SD with statistical
   significance determined by one-way ANOVA (in d) or two-way ANOVA (in i)
   with multiple comparisons. The p values are represented as **, <0.01;
   ***, <0.001; ****, <0.0001. n.s. not significant.

   The impact on differentiation capacity was even more striking. When
   induced toward monocytic differentiation, control clones consistently
   generated substantial CD11b⁺CD115⁺ populations ( > 40% in all 18
   clones). Similarly, most Id1-targeted (14/16 clones) and Cebpa-targeted
   (19/21 clones) clones retained differentiation capacity. In contrast,
   only 22% (2/9 clones) of viable Cebpb-targeted clones produced
   significant CD11b⁺CD115⁺ populations (Fig. [106]4e, f), suggesting the
   critical contribution of C/EBPβ in both the differentiation and the
   survival for cCLP-Ms. To determine whether C/EBPβ was sufficient to
   drive monocytic differentiation, we performed gain-of-function
   experiments using retroviral expression. Strikingly, enforced
   expression of Cebpb in cCLPs induced CD11b⁺CD115⁺ populations even in
   the absence of IL-6 or other inducing cytokines (Fig.[107]4g, h). Taken
   together, these results indicate that C/EBPβ is a key regulator
   sufficient for cCLPs to differentiate into cCLP-Ms, implicating it as a
   pivotal factor in regulating lineage commitment within this system.

cCLP-Ms exhibit distinct functional properties and immunoregulatory capacity

   Having established the molecular basis of cCLP-M generation, we next
   investigated their functional capabilities. As our transcriptional
   analysis had revealed robust expression of innate immune sensors in
   cCLP-Ms, including Tlr4 (Fig. [108]3c), we first examined their
   response to TLR ligands. When stimulated with the TLR4 ligand LPS,
   cCLP-Ms produced significant amounts of IL-6 at levels comparable to
   BMDMs, but notably lower levels of TNF-α (Fig. [109]5a, b). Other TLR
   ligands such as R848 (TLR7) and ODN2395 (TLR9) also elicited these
   cytokine production from cCLP-Ms but not from cCLPs, despite comparable
   expression of Tlr7 and Tlr9 (Supplementary Fig. [110]5a–c).
   Interestingly, cCLP-Ms also produced IL-10 in response to R848 and
   ODN2395, but not to LPS, suggesting selective activation of
   anti-inflammatory pathways (Supplementary Fig. [111]5d). A striking
   feature of cCLP-Ms was their limited MHC class II expression, which
   remained low even after TLR stimulation, contrasting sharply with the
   robust upregulation observed in BMDMs and BMmono (Fig. [112]5c, d, and
   Supplementary Fig. [113]5e). These findings suggest that cCLP-Ms
   displays distinct potential on cytokine secretion and molecular
   expression in response to TLR ligands compared to BMDMs and BMmono.

Fig. 5. cCLP-Ms function in vitro and in vivo.

   [114]Fig. 5
   [115]Open in a new tab

   a–d The indicated cells were stimulated with (filled bars) or without
   (open bars) LPS (1 μg/mL) for 24 h. Culture supernatants and cells were
   harvested for ELISA and FACS, respectively. TNFα and IL-6 production,
   representative histograms for MHC class II (I-A/I-E) expression, and
   the percentage of I-A/I-E^+ cells are shown (n = 3, each). e
   Representative FACS plots and f the relative MFI of PD-L1 and PD-L2
   expression on cCLP-M three days after culture with M-CSF, LPS /IFN-γ,
   IL-4, or medium alone (control) (n = 3, each). g Experimental scheme
   for efferocytosis assay. Dex: Dexamethasone. h Representative FACS
   plots for CFSE and CD11b expression on cCLP-Ms. i The percentage of
   CFSE-positive cells among indicated populations, or within the CD11b^+
   subset for cCLP-Ms and those cultured with M-CSF or IL-4 (n = 4, each).
   Among each population, the groups treated with DMSO (control) or
   Latrunculin B (LatB) are shown in dark gray and light gray bars. j
   Experimental scheme for IgE-CAI in mice transferred with cCLP-Ms (k).
   Time course of ear swelling (Δ ear thickness) after the challenge with
   TNP-OVA (antigen) alone (in blue) or combined with the transfer of
   cCLPs (in red) or cCLP-Ms (in green) in anti-TNP IgE-sensitized mice
   (n = 3 each). Data are mean ± SD with statistical significance
   determined by one-way ANOVA (in f) or two-way ANOVA with multiple
   comparisons (in a, b, d, i) or tukey’s multiple comparisons test (in
   k). The p values are represented as *, <0.05; **, <0.01; ***, <0.001;
   ****, <0.0001. n.s., not significant. All data are representative of at
   least two independent experiments.

   To explore their functional plasticity, we examined whether cCLP-Ms
   could undergo classical M1/M2 polarization, a hallmark of macrophage
   versatility [[116]22]. Whereas stimulation with LPS and IFN-γ induced
   the upregulation of the M1 marker PD-L1, IL-4 stimulation led to
   increased expression of the M2 marker PD-L2 in cCLP-Ms [[117]23],
   indicating that cCLP-Ms have the plasticity to undergo M1/M2
   polarization (Fig. [118]5e, f). We next assessed their phagocytic
   capability through an efferocytosis assay, measuring the uptake of
   apoptotic thymocytes (Fig. [119]5g). Unlike cCLPs or OP9, cCLP-Ms
   efficiently bound CFSE-labeled apoptotic cells in an actin-dependent
   manner, as demonstrated by inhibition with Latrunculin B (Fig. [120]5h,
   i). This phagocytic capacity was further enhanced by M-CSF or IL-4
   treatment, indicating that cCLP-Ms can acquire advanced macrophage
   functions through additional maturation signals.

   A previous report showed that inflammatory monocytes play a critical
   role in ameliorating IgE-dependent cutaneous allergic inflammation
   (IgE-CAI) in mice by converting into anti-inflammatory M2-type
   macrophages in the skin. This was demonstrated through the adoptive
   transfer of bone marrow-derived CD115^+ monocytes from wild-type mice
   to the affected ear skin (Fig. [121]5j) [[122]24]. We tested the in
   vivo function of cCLP-Ms in this mouse model of allergic inflammation.
   Remarkably, intradermal transfer of cCLP-Ms, but not cCLPs, reduced ear
   swelling by approximately 50% compared to control conditions (Fig.
   [123]5k). Tracking cell trace violet (CTV)-labeled cCLP-Ms following
   transfer revealed the presence of CTV^+F4/80^+PD-L2^+ cells exclusively
   in the inflamed ear tissue (TNP-OVA + cCLP-M) induced by IgE-CAI
   (Supplementary Fig. [124]6). This observation aligns with previous
   reports, suggesting that the transferred cCLP-Ms function in vivo by
   differentiating into M2 macrophages within the inflammatory site,
   thereby contributing to the suppression of IgE-CAI.

Discussion

   In this study, we demonstrated that IL-6 stimulation drives the
   differentiation of CD11b⁺CD115⁺ monocytic cells (cCLP-Ms) from cCLPs.
   This differentiation was observed not only in cCLPs but also in primary
   lymphoid progenitors under similar in vitro conditions. Furthermore, in
   vivo transfer experiments revealed that cCLPs can give rise to
   CD11b⁺CD115⁺ cells following LPS administration, highlighting their
   latent myeloid potential under inflammatory conditions. These findings
   suggest that lymphoid progenitors exhibit lineage plasticity, shifting
   toward myeloid differentiation in response to inflammatory signals.

   The cCLP-Ms displayed unique phenotypic and functional attributes that
   distinguish them from BMDMs, conventional cultured macrophages derived
   from bone marrow progenitors. While acquiring innate immune
   characteristics such as innate sensor expression and phagocytic
   activity, cCLP-Ms maintained lower levels of MHC class II and TNF-α
   expression, suggesting they may constitute a distinct myeloid
   population with unique immunological functions. Moreover, these cells
   demonstrated remarkable plasticity in M1/M2 polarization and
   immunoregulatory potential, as evidenced by their ability to suppress
   IgE-mediated cutaneous allergic inflammation upon adoptive transfer.
   These findings highlight the utility of cCLP-Ms as a valuable system to
   unravel the molecular mechanisms underlying the differentiation and
   functional properties of lymphoid-derived monocytes/macrophages.

   At the molecular level, we identified C/EBPβ as a key transcription
   factor mediating the IL-6-induced differentiation of cCLP-Ms. Unlike
   conventional macrophage differentiation pathways regulated by key
   transcription factors such as PU.1, C/EBPα, and IRF8 [[125]1, [126]4,
   [127]25], the cCLP-M differentiation process appears to be
   predominantly C/EBPβ-dependent as it was up-regulated ~400-fold in
   cCLP-Ms compared to cCLP. Notably, overexpression of C/EBPβ alone was
   sufficient to induce CD11b⁺CD115⁺ differentiation from cCLPs in the
   absence of IL-6. These results align with previous reports linking IL-6
   to STAT3-mediated induction of C/EBPβ and its downstream regulation of
   Csf1r (CD115) transcription [[128]26–[129]28]. C/EBPβ is known to
   regulate specialized myeloid populations, such as Ly6C^lo non-classical
   monocytes and segregated-nucleus-containing atypical monocytes (SatM)
   [[130]28, [131]29]. It also contributes to the development of resident
   macrophages, such as large peritoneal macrophages (LPM), which exhibit
   high CD11b and F4/80 expression but low MHC class II levels [[132]30].
   The involvement of C/EBPβ in cCLP-M differentiation suggests potential
   molecular links with these specialized populations, warranting further
   investigation into their transcriptional programs.

   IL-6 stimulation markedly reduced IL-7Rα expression, a receptor
   essential for lymphoid progenitor survival and differentiation
   [[133]31, [134]32], shedding light on the mechanisms underlying
   lymphoid-to-myeloid transition. IL-7Rα expression is regulated by
   transcription factors such as PU.1, which plays a pivotal role in both
   lymphoid and myeloid differentiation [[135]33, [136]34]. High PU.1
   levels favor myeloid differentiation, while low levels promote lymphoid
   development [[137]35]. PU.1 is also required for the developmental
   progression of multipotent progenitors to common lymphoid progenitors
   [[138]36]. Notably, PU.1 has been reported to interact with C/EBPβ to
   regulate monocyte/macrophage gene expression in addition to its
   association with C/EBPα on lineage commitment to macrophages
   [[139]37–[140]39]. Therefore, IL-6 stimulation may deplete low-level
   PU.1 required for IL-7Rα expression, thereby driving myeloid
   differentiation, or that C/EBPβ directly or indirectly suppresses
   IL-7Rα expression. These molecular mechanisms warrant further
   investigation to elucidate their contributions to lineage plasticity.

   While our data present compelling evidence for IL-6–mediated lineage
   reprogramming, the conclusions are based primarily on in vitro culture
   systems. However, these findings raise the possibility that similar
   mechanisms could occur under pathological conditions characterized by
   chronic inflammation, such as infections, autoimmune diseases, or
   cancer. In particular, the role of IL-6 in promoting myeloid-biased
   differentiation at the expense of lymphoid output during inflammation
   has been well-documented [[141]21, [142]40, [143]41]. Recent studies
   have also provided evidence that macrophage-like cells (termed B-MF)
   are generated from B-cell precursors, a lymphoid progenitor downstream
   of CLPs in vivo in various murine tumor models and samples from humans
   with breast and ovarian cancers [[144]42]. Additionally, human
   hematopoietic stem cells have been shown to differentiate into myeloid
   cells through the IL-6–C/EBPβ axis, consistent with our findings
   [[145]43, [146]44]. Understanding the precise molecular pathways and
   contexts in which lymphoid-to-myeloid plasticity occurs could uncover
   novel therapeutic targets to modulate immune responses. Further
   investigations are required to establish whether IL-6 and C/EBPβ
   mediate myeloid differentiation from lymphoid progenitors under
   physiological or pathological conditions in vivo.

   In conclusion, our study establishes a system for investigating aspects
   of monocyte development and reveals unexpected properties of
   CLP-derived myeloid cells. This system provides valuable opportunities
   for studying fundamental aspects of monocyte biology and the molecular
   mechanisms underlying myeloid cell differentiation. Our findings also
   highlight the importance of cell plasticity in immune cell development.
   While CLPs are traditionally viewed as lymphoid-committed progenitors,
   their ability to generate functional myeloid cells under specific
   conditions suggests that lineage commitment may be more flexible than
   previously appreciated. Although the physiological relevance of this
   pathway remains to be fully clarified, our results provide new insights
   into the diversity and plasticity of myeloid cell development.

Materials and methods

Mice

   All experiments utilized C57BL/6 mice obtained from CLEA Japan (Tokyo,
   Japan). Animals were maintained under specific pathogen-free conditions
   in the animal facilities of Hiroshima University and Tokyo Medical and
   Dental University.

Reagents

   Recombinant mouse cytokines were obtained from BioLegend (San Diego,
   CA, USA): Flt3L (550706), IL-7 (577802), SCF (579702), IL-3 (575504),
   IL-6 (575702), TNF-α (575208), M-CSF (576404), GM-CSF (576302), IL-4
   (574304), and IFN-γ (575304). TLR ligands included LPS from Escherichia
   coli O111:B4 (SIGMA, St. Louis, MO, USA, L2630), R848 (Adipogen, San
   Diego, CA, USA, AG-CR1-3582-M005), and ODN2395 (Adipogen,
   IAX-200-007-C100).

Cell lines

   OP9 stromal cells were maintained in OP9 medium consisting of α-minimum
   essential medium (Thermo Scientific, Waltham, MA, USA, 11900-073)
   supplemented with 20% fetal bovine serum (FBS), 2.2 g/L NaHCO₃, and 1×
   penicillin-streptomycin (Nacalai Tesque, Kyoto, Japan). Cells were
   passaged every 3–4 days and plated at 1 × 10⁴ cells/cm² one day before
   use in cCLP cultures. Plat-E retrovirus packaging cells were maintained
   in standard Dulbecco’s Modified Eagle Medium (DMEM) culture medium
   [DMEM (Fujifilm, Tokyo, Japan, 043-30085), 10% FBS (Nichirei, Tokyo,
   Japan), 1% Penicillin/Streptomycin (Nacalai Tesque, 26253-84), and
   25 mM HEPES (Fujifilm, Tokyo, Japan, H3375)] [[147]45]. All cells were
   cultured in a humidified atmosphere of 5% CO[2] at 37 °C and were
   confirmed to be free from mycoplasma contamination.

Preparation and culture of cCLPs

   A serum-free basal culture medium, KIDMEM, was prepared according to
   the previously established protocols [[148]17]. cCLPs were generated
   from either freshly isolated bone marrow as previously described
   [[149]17] or derived from cryopreserved stocks (1–3 ×105 cells) that
   had been cultured in vitro for 3-10 weeks. Cells were cultured in
   plates pre-coated with OP9 cells in cCLP culture medium containing
   KIDMEM supplemented with 25 ng/mL Flt3L and 0.1 ng/mL IL-7. Under these
   conditions, 70-80% of cells maintained the Lin⁻IL-7Rα⁺CD27⁺Ly6D⁻
   phenotype characteristic of uncommitted progenitors. In some
   experiments, these undifferentiated cells were sorted for further
   analysis.

Generation of bone marrow-derived macrophages (BMDMs)

   BMDMs were generated by culturing bone marrow cells isolated from mouse
   femurs in 150-mm petri dishes (FALCON, NY, USA, 351058) in DMEM culture
   medium containing 20 ng/mL M-CSF. This standardized protocol
   consistently yields highly pure populations of functional macrophages.

In vitro differentiation from cCLPs to CD11b^+ myeloid cells

   cCLPs were cultured for 3–4 days on OP9 cells in complete RPMI medium
   [RPMI1640 (Fujifilm, 189-02145), 10% FBS, 1× sodium pyruvate (Nacalai
   Tesque, 06977-34), 1× non-essential amino acids (Nacalai Tesque,
   06344-56), 10 mM HEPES, 50 μM β-mercaptoethanol (Thermo Scientific,
   21985023), and 1× penicillin-streptomycin (Nacalai Tesque, 09367-34)].
   Cytokine supplementation included Flt3L (10 ng/mL), SCF (10 ng/mL),
   IL-3 (10 ng/mL), IL-6 (10 ng/mL), TNF-α (20 ng/mL), M-CSF (10 ng/mL),
   and GM-CSF (20 ng/mL) individually or in combination. To induce
   differentiation into cCLP-Ms, a cytokine mixture of Flt3L (10 ng/mL),
   SCF (10 ng/mL), and IL-6 (10 ng/mL) was applied. In some experiments,
   1 μM all-trans retinoic acid (ATRA, Nacalai Tesque, 6331-44) was added
   to enhance differentiation efficiency. Cultures were maintained at
   37 °C in a humidified atmosphere containing 5% CO₂, with regular
   monitoring of cell morphology and surface marker expression.

Ex vivo culture and limiting dilution assay of lymphoid progenitor cells

   Lin^- bone marrow cells from 7-week-old C57BL/6 mice were enriched by
   magnetic depletion with a lineage antibody cocktail (biotinylated CD3,
   CD19, B220, CD11b, CD11c, Gr-1, NK1.1, Ter-119), followed by BD IMag™
   Streptavidin Particles Plus (BD Biosciences, 557812). LMPPs
   (Lin⁻Sca1⁺c-Kit⁺Flk2⁺IL-7Rα⁻) and uncommitted CLPs
   (Lin⁻Flk2⁺IL-7Rα⁺CD27⁺Ly6D⁻) were isolated using FACSAria™ II or
   FACSAria™ Fusion (BD Biosciences) and cultured with 10 ng/mL each of
   Flt3L/IL-6/SCF or 10 ng/mL M-CSF. Cells were plated on OP9-pre-coated
   flat-bottom plates, with 1000 cells/well in 96-well plates for 3 days,
   or at serial twofold dilutions (16 to 1 cell per well) for 7 days for
   limiting dilution assay. Cells were then harvested for flow cytometric
   analysis. In the limiting dilution assay, wells containing more than 10
   CD11b⁺CD115⁺ cells were considered positive, and the frequency of
   myeloid potential for each population was determined by counting the
   number of negative wells and calculating frequencies based on Poisson
   distribution.

Flow cytometry

   Single-cell suspensions were prepared in FACS buffer PBS containing 1%
   FBS, 5 mM EDTA, and 0.2% sodium azide and blocked with anti-CD16/CD32
   (clone 2.4G2) before staining with the following
   fluorochrome-conjugated anti-mouse monoclonal antibodies: FITC-labeled
   Ly6G (1A8, #127605), Ly6C (1G7.G10, #128006), and I-A/I-E (M5/114.15.2,
   #107605) from BioLegend, and PD-L1 (MIH6, MCA2626F) from AbD Serotec;
   PE-labeled IL-6Rα (W18166A, #160405), Flk2 (#135305), Ly6d (49-H4,
   #138606), I-A/I-E (M5/114.15.2, #107607), and B220 (RA3-6B2, #103207)
   from BioLegend; PE/Cy7-labeled CD11b (M1/70, #101215), Ly6C (1G7.G10,
   #128018), and IL-7Rα (A7R34, #135014) from BioLegend; APC-labeled CD115
   (AFS98, #135509), gp130 (4H1B35, #149405), F4/80 (BM8, #123116), CD27
   (LG.3A10, #124211), and PD-L2 (TY25, #107206) from BioLegend; Alexa
   700-labeled Sca1 (D7, #108141) from BioLegend; APC/Cy7-labeled Ly6G
   (1A8, #127624), Ly6C (1G7.G10, #128025), I-A/I-E (M5/114.15.2,
   #107627), and B220 (RA3-6B2, #103224) from BioLegend; APCVio770-labeled
   Ly6d (49-H4, #130-115-315) from Miltenyi Biotec (Germany);
   BV421-labeled CD115 (AFS98, #135513), F4/80 (BM8, #123131), and c-Kit
   (2B8, #105827) from BioLegend; BV605-labeled CD19 (1D3, #115539) from
   BioLegend; BV650-labeled CD11b (M1/70, #101239) from BioLegend;
   PerCP/Cy5.5-labeled Streptavidin (#405214) from BioLegend; and
   PE-labeled CCR2, either Y15-488 from BD Biosciences (San Jose, CA, USA,
   #475301) or clone 475301 from R&D Systems (Minneapolis, MN,
   USA,FAB5538P). Dead cells were excluded using propidium iodide
   staining. Data were acquired on a CytoFLEX S flow cytometer (Beckman
   Coulter, Brea, CA, USA) with standardized voltage settings and analyzed
   using FlowJo software (BD Biosciences).

Intracellular staining for phosphorylated proteins

   For phospho-STAT3 analysis, 1 × 10⁵ cCLPs were stimulated with
   100 ng/mL IL-6 for defined time intervals (0–60 min) at 37 °C.
   Reactions were terminated by the addition of ice-cold MACS buffer.
   Cells were fixed using CytoFix™ fixation buffer (BD Biosciences),
   permeabilized with cold methanol, and stained with Alexa 647-conjugated
   anti-phospho-STAT3 (Ser727) antibody (BioLegend, 698913,
   clone#A16089B). Finally, cells were washed with FACS buffer and
   analyzed using a BD FACSCanto™ flow cytometer (BD Biosciences).

May-Grünwald Giemsa staining

   Cell morphology was assessed using May-Grünwald-Giemsa staining of
   cytocentrifuge preparations. Briefly, sorted populations were suspended
   at 1 × 10⁵ cells/mL and applied to glass slides using a Cytospin 3
   centrifuge (200 rpm, 2 min). After methanol fixation, slides were
   sequentially stained with May-Grünwald solution (Merck, Darmstadt,
   Germany, 1014240100) for 3 min and diluted Giemsa solution (Merck,
   1092040100) for 15 min. Images were subsequently acquired with a
   Keyence BZ-X800 microscope (Osaka, Japan).

Stimulation with toll-like receptor (TLR) ligands

   BMDMs were harvested on day 6 using ice-cold PBS/5 mM EDTA, verified
   for CD11b and F4/80 expression, and re-plated at 2 × 10⁴ cells/well in
   complete RPMI medium. cCLPs, cCLP-Ms, and BMmono were isolated using
   FACSAria™ II or FACSAria™ Fusion (BD Biosciences) and plated at 2 × 10⁴
   cells/well in 96-well plates in complete RPMI medium. Cells were
   stimulated with LPS (1 µg/mL), R848 (1 µM), or ODN2395 (1 µM). Culture
   supernatants were collected after 24 h for cytokine analysis. Cells
   were harvested simultaneously for flow cytometric assessment of
   activation markers.

ELISA

   Cytokine levels in culture supernatants were measured using
   standardized ELISA protocols. Mouse TNF-α, IL-6, and IL-10 were
   quantified using ELISA MAX™ Standard Sets (BioLegend) according to the
   manufacturer’s specifications. Wells were reacted with ELISA POD
   Substrate TMB Kit (Nacalai Tesque, 05298-80), and the reaction was
   stopped using 2 N H[2]SO[4]. Absorbance was measured at 450 nm using an
   iMark microplate reader (Bio-Rad Laboratories, Hercules, CA, USA).

Efferocytosis assay

   Apoptotic target cells were generated from freshly isolated mouse
   thymocytes (4 × 10⁷) treated with 2 μM dexamethasone (Sigma, D1159) for
   18 h. Apoptotic cells were labeled with 1 μM carboxyfluorescein
   succinimidyl ester (CFSE; BioLegend, 423801), washed extensively, and
   resuspended at 2 × 10⁷ cells/mL. cCLPs, cCLP-Ms, or their
   differentiated derivatives were plated at 2 × 10⁵ cells/mL in 96-well
   plates and cultured overnight. Where indicated, cells were pre-treated
   with 1 μM Latrunculin B (Calbiochem, 428020) for 30 min to inhibit
   actin polymerization. CFSE-labeled apoptotic cells were added at a 10:1
   ratio (apoptotic cells: phagocytes) and co-cultured for 3 h at 37 °C.
   Phagocytosis was quantified by flow cytometry, measuring the percentage
   of CD11b⁺ cells that had acquired CFSE fluorescence.

M1/M2 polarization

   cCLP-Ms were harvested, washed thoroughly, and cultured for 3 days at
   2 × 10^5/mL with rm M-CSF (10 ng/mL), LPS (100 ng/mL) + IFN-γ
   (10 ng/mL), or IL-4 (10 ng/mL). Polarization was assessed by flow
   cytometric analysis of characteristic markers including PD-L1 (M1) and
   PD-L2 (M2).

RNA-sequencing analysis

   Bone marrow cells from C57BL/6 mice (7–18 weeks old) were prepared to
   isolate CD11b⁺Ly6C⁻Ly6G⁺ neutrophils (BMneu) and
   CD11b⁺CD115⁺CCR2⁺Ly6C^hiLy6G⁻ monocytes (BMmono). Additionally, bone
   marrow cells were cultured with 20 ng/mL M-CSF for 4 days to generate
   CD11b⁺ cultured macrophages (BMDMs). cCLPs (passages 6–21, cultured in
   vitro for 1–2.5 months) and those stimulated with 10 ng/mL FL/IL-6/SCF
   for 1 day (D1) were collected for RNA extraction. Similarly, cCLPs
   cultured with 10 ng/mL FL/IL-6/SCF for 3 days (D3) were harvested to
   isolate CD11b⁺CD115⁺CCR2⁺Ly6C⁺Ly6G⁻ cells. Cell sorting was performed
   using FACSAria™ II or FACSAria™ Fusion (BD Biosciences).

   Total RNA was extracted using the RNeasy Micro Kit (QIAGEN, Germany,
   74004) according to the manufacturer’s instructions. Three biological
   replicates per group were subjected to RNA sequencing using the BGISEQ
   platform with a PE100 sequencing length (BGI). Library preparation and
   sequencing were performed by BGI using a DNBSEQ sequencer. Raw
   sequencing reads underwent quality control and adapter trimming using
   fastp (v0.21.0). Trimmed reads were mapped and quantified using RSEM
   (v1.3.1), with STAR (v2.7.10) as the aligner. Expression data were
   imported into iDEP.96 [[150]46], normalized using CPM (counts per
   million), and log-transformed with EdgeR [log₂(CPM + 4)] for principal
   component analysis (PCA) and k-means clustering analysis.

   Differentially expressed genes (DEGs) were identified using DESeq2
   (v1.34.0), applying thresholds of |log₂ fold-change (FC) | > 1.0 and
   adjusted pvalue < 0.05. These DEGs were visualized using heatmaps and
   scatter plots. For functional enrichment analysis, Metascape
   ([151]http://metascape.org) [[152]47] was used to perform Gene Ontology
   (GO) annotation analysis and tissue/cell type association analysis
   based on the PaGenbase database [[153]48].

CRISPR/Cas9-mediated gene editing

   Synthesized Alt-R® CRISPR-Cas9 crRNAs, Alt-R® CRISPR-Cas9 tracrRNA
   (#1072534), Alt-R® S.p. Cas9 Nuclease V3 (#1081059), and the
   electroporation enhancer (#1075916) were purchased from Integrated DNA
   Technologies (IDT; Coralville, IA, USA) and assembled into RNA
   ribonucleoprotein (RNP) complexes according to the manufacturer’s
   protocol. The crRNAs for Cebpa (5’-TAGAAGTCGGCCGACTCCAT-3’), Cebpb
   (5’-GAGGCTCACGTAACCGTAGT-3’), and Id1 (5’-GCATGTAATCGACTACATCA-3’) were
   validated by IDT. cCLPs (1 × 10⁵) were suspended in 80 µL of Opti-MEM,
   mixed with Cas9-RNP complexes, and electroporated using the NEPA21
   system (NEPAGENE, Chiba, Japan) at 275 V and 2.5 ms pulse. Following
   electroporation, cells were cultured in 24-well plates containing OP9
   cells in cCLP culture medium for 4 days to allow recovery. The cells
   were harvested and re-plated at 1 cell/well in a 96-well flat-bottom
   plate coated with OP9 in cCLP culture medium for clonal expansion over
   seven days. The resulting single-cell clones were validated through T7
   endonuclease I assay (New England Biolabs, Ipswich, MA, USA) and Sanger
   sequencing of the targeted regions, followed by functional assessment
   of differentiation capacity.

Retroviral vector construction and transduction

   Total RNAs were extracted from cells using TRI reagent (Molecular
   Research Center, Cincinnati, OH, USA, TR118), and cDNAs were
   synthesized using ReverTra Ace (Toyobo, Osaka, Japan, TRT-101) with
   oligo-dT[18] (Eurofins) and random primers (Invitrogen, 48190011). The
   long isoform of mouse Cebpb (LAP, transcript variant 1,
   [154]NM_001287738.1) was amplified from cCLP-M-derived cDNA using
   KOD-Plus-NEO polymerase (Toyobo, KOD-401) with specific primers
   (forward: 5ʹ-aggcgccggaattcgttaacctcgagatggaagtggccaacttctactac-3ʹ;
   reverse: 5ʹ-gaggggcggaattgatcccgctcgagctagcagtggcccgccgag-3ʹ). The
   amplified PCR product was inserted at the XhoI site of the
   pMSCV-IRES-GFP retroviral vector (pMIG) using NEBuilder HiFi DNA
   Assembly Master Mix (New England Biolabs, E2621S). The construct was
   verified by Sanger sequencing and transfected into Plat-E cells using
   PEI-MAX (Polyscience, Warrington, PA, USA, 24765-100). Viral
   supernatants were harvested 72 h post-transfection. For retroviral
   transduction, 1.2 × 10⁵ cCLPs were suspended in a mixture containing
   equal volumes (500 μL each) of viral supernatant and cCLP medium. Spin
   infection was performed at 1150 × g and 30 °C for 3 h. The infected
   cells were washed twice, and 1 × 10⁴ cells were cultured in 200 μL
   complete RPMI medium in 96-well plates pre-coated with OP9 for 4 days.
   Half of the cell suspension (100 μL) was collected for flow cytometric
   analysis.

Induction of IgE-induced chronic allergic inflammation (IgE-CAI)

   IgE-CAI was induced as previously described [[155]49] by intravenously
   administering 300 μg of anti-TNP-IgE to wild-type mice, followed by
   intradermal injection of 10 μg TNP-conjugated ovalbumin (TNP-OVA) into
   the right ear as an allergen. As a control, 10 μg of unconjugated
   ovalbumin (OVA) was administered to the left ear. The value of the
   differences in ear thickness (right—left) was calculated for the
   evaluation of inflammation. For cell transfer experiments, 1 × 10⁶
   cells were intradermally co-injected into the ear along with Flt3L
   (10 ng) simultaneously with the allergen. cCLPs and cCLP-Ms were
   isolated via IMag system with biotinylated anti-CD45 and anti-CD115,
   respectively, followed by BD IMag™ Streptavidin Particles Plus (BD
   Biosciences) before labeling with or without cell trace violet (CTV,
   Thermo Scicentific, [156]C34557) according to the manufacturer’s
   protocol.

Statistical analysis

   Cells and mice were randomly assigned to each treatment group and
   experiment. No statistical method was used to predetermine sample size.
   Sample size was chosen empirically based on our previous experience in
   the calculation of experimental variability. Data were analyzed with
   the GraphPad Prism software version 7.0. pvalues were calculated by
   one-way or two-way ANOVA. Statistical significance was determined with
   alpha <0.05 and presented as *, <0.05; **, <0.01; ***, <0.001; ****,
   <0.0001.

Supplementary information

   [157]Supplementary information^ (1.1MB, docx)
   [158]Supplementary Table 1^ (12.3KB, xlsx)
   [159]Supplementary Table 2^ (19.3KB, xlsx)

Acknowledgements