Lymphoid origin of intrinsically activated plasmacytoid dendritic cells in mice

Alessandra M. Araujo; Joseph D. Dekker; Kendra Garrison; Zhe Su; Catherine Rhee; Zicheng Hu; Bum-Kyu Lee; Daniel Osorio Hurtado; Jiwon Lee; Vishwanath R. Iyer; Lauren I. R. Ehrlich; George Georgiou; Gregory C. Ippolito; S. Stephen Yi; Haley O. Tucker

doi:10.7554/eLife.96394.1

eLife assessment

This useful study aims to add fresh insights into the sharing of lymphoid and CDP (common DC precursor) lineage origin of plasmacytoid dendritic cells (pDCs). The evidence for a small subset of pDCs sharing its origin with B cell progenitors and depending on BCL11a expression is solid, although further functional characterization of the reported pDC subset would significantly enhance the significance of the study. The study will be of relevance to cellular immunologists interested in the ontogeny of plasmacytoid dendritic cells.

https://doi.org/10.7554/eLife.96394.1.sa2

Significance of findings

useful: Findings that have focused importance and scope

landmark
fundamental
important
valuable
useful

Strength of evidence

solid: Methods, data and analyses broadly support the claims with only minor weaknesses

exceptional
compelling
convincing
solid
incomplete
inadequate

During the peer-review process the editor and reviewers write an eLife assessment that summarises the significance of the findings reported in the article (on a scale ranging from landmark to useful) and the strength of the evidence (on a scale ranging from exceptional to inadequate). Learn more about eLife assessments

Abstract

We identified a novel mouse plasmacytoid dendritic cell (pDC) lineage derived from the common lymphoid progenitors (CLPs) that is dependent on expression of Bcl11a. These CLP-derived pDCs, which we refer to as “B-pDCs”, have a unique gene expression profile that includes hallmark B cell genes, normally not expressed in conventional pDCs. Despite expressing most classical pDC markers such as SIGLEC-H and PDCA1, B-pDCs lack IFN-α secretion, exhibiting a distinct inflammatory profile. Functionally, B-pDCs induce T cell proliferation more robustly than canonical pDCs following Toll-like receptor 9 (TLR9) engagement. B-pDCs, along with another homogeneous subpopulation of myeloid derived pDCs, display elevated levels of the cell-surface receptor tyrosine kinase AXL, mirroring human AXL⁺ transitional DCs in function and transcriptional profile. Murine B-pDCs therefore represent a phenotypically and functionally distinct CLP-derived DC lineage specialized in T cell activation and previously not described in mice.

Introduction

Plasmacytoid dendritic cells (pDCs) specialize in the production of type I interferons (IFNs) and promote antiviral immune responses following engagement of pattern recognition receptors. They have been mainly implicated in the pathogenesis of autoimmune diseases that are characterized by a type I IFN signature (notably, IFN-α). Yet, pDCs have also been shown to be able to induce tolerogenic immune responses^5–8. Due to the clinical significance of pDCs, several studies aimed at mapping pDC lineage derivation have emerged recently. Yet, a complete understanding of pDC development and their existing subsets in mice is still lacking. The transcription factor 4 (TCF4) is required for pDC development and for lineage identity^9–11. TCF4 is a component of a multiprotein complex that includes both positive and negative regulators^{6, 12}. One of these components, the transcription factor BCL11a, which is also essential for B cell development^13–15, induces Tcf4 transcription and initiates a positive feedback loop with TCF4 to maintain pDC lineage commitment and function^14–16.

Unlike their conventional dendritic cell (cDC) counterparts, pDCs express transcriptional regulators and markers associated with B-lymphocyte development in addition to BCL11a (e.g. B220, SPIB)^{17, 18}. These features, along with the established generation of pDCs from myeloid restricted precursors^{58, 76}, have made it difficult to define pDC lineage affiliation^{11, 17, 19, 20}. This has led to the hypothesis that pDC subsets may have distinct origins derived from either the Common Lymphoid Progenitor (CLP) or the Common Myeloid Progenitor (CMP)^{18, 61}. Beyond the complex nature of their lineage, pDCs with different functional attributes (e.g. variable IFN-α expression levels) or different surface markers (e.g. CD19⁺ pDCs detected in tumor-draining lymph nodes) have also been identified^{17, 21–23}.

On par with the ever-increasing heterogeneity described within pDC populations, a novel AXL+ dendritic cell (DC) population with many pDC-like properties has recently been discovered in human blood^{1, 3, 4}. While AXL+ DCs express many canonical pDC markers (e.g. CD123, BDCA2/CD303), they also express CD2, the Ig-like lectins SIGLEC1 and SIGLEC6, as well as the activation marker CD81¹. In a separate study², a similar population of pDCs was shown to express high levels of CD5 and CD81 in mice and humans, two glycoproteins normally associated with the B cell receptor (BCR) signaling complex. The origin of AXL+ transitional DCs is currently unclear and the presence of a homologous population in mice remains muddled. Here, we report the identification of a lymphoid-derived pDC subset (B-pDCs) that shares inflammatory features with myeloid derived Axl+ pDC populations in mice. We also demonstrate an in vivo requirement for Bcl11a in the transcriptional specification of B-pDCs.

Results

We previously demonstrated that conditional deletion of Bcl11a in the hematopoietic stem cell (HSC) compartment mediated by Vav-1-Cre or by inducible Mx1-Cre recombinase results in complete abolishment of pDC development¹⁴. Spurred by previous speculation of pDC origin from the CLP^{17, 20}, we next selectively deleted floxed (^F) Bcl11a alleles in CD79a⁺ cells ^{24, 25}, as mediated by mb1-Cre in vivo. Expression of the mb1 gene (CD79a) initiates at the LY6D⁺ CLP stage, in B-cell-biased lymphoid progenitors (BLPs)²⁶, downstream of LY6D⁻ CLPs (Fig. S1). Bcl11a^F/Fmb1-Cre mice (cKO) and littermate controls were examined for pDC frequencies among nucleated cells in the bone marrow (BM). B220⁺ PDCA1⁺ pDCs (all found to be CD11c ^int) were consistently and significantly reduced by an average of ∼25% (24.8 ± 2.4%) in cKO mice relative to littermates (Fig. 1A-B). Loss of B cells (B220^hi PDCA1^-) served as a gauge of mb1-Cre deletion efficiency (Fig. 1A-B). Taken together, these data indicate that a significant proportion of pDCs are derived from BLPs or BLP-derived cells and are Bcl11a dependent.

Mb1-Cre deletion of *Bcl11a* identifies a CLP-derived subset of pDCs
(A) Representative FACS plots of pDC (gated as B220⁺ PDCA1⁺) and B cell (gated as B220⁺ PDCA1-) percentages in the bone marrow (BM) of *Bcl11a* ^F/F *mb1*-Cre mice (cKO) and littermate controls. (B) Quantification of pDC and B cell populations in the BM of *Bcl11a* ^F/F *mb1*-Cre mice (cKO) and littermate controls as cells/100,000 cells. (C-D) Flow cytometric analysis of BM and spleens of recipient mice 8 weeks post BM transplantation. BM was transferred from either BCL11A -sufficient reporter control mice (*mb1*-Cre-YFP) or BCL11A-deficient cKO mice (*Bcl11a* ^F/F *mb1*-Cre-YFP) into lethally irradiated C57BL/6J recipients. (E) B cell numbers after BM transplantation in BM and spleens of recipient mice. (F) pDC numbers after BM transplantation in BM and spleens of recipient mice. (G) Comparison of YFP⁺ pDC percentages in the spleen and BM of recipient mice post BM transplantation. Student’s t-test was used for all statistical comparisons. Error bars=mean±s.d. The results are representative of two experiments each containing 3-4 mice per group.

To expand these in vivo observations, and to confirm that this defect is intrinsic to hematological progenitor cells, we transferred BM from either BCL11A -sufficient reporter control mice (mb1-Cre-YFP) or BCL11A-deficient cKO mice (Bcl11a^F/Fmb1-Cre-YFP) into lethally irradiated wild type C57BL/6J recipients. After 8 weeks, <10% of B cells (B220⁺ PDCA1^-) in the spleens of mb1-Cre-YFP recipients were YFP⁻, confirming elimination of recipient hematopoiesis (Fig. 1D-E). As expected, Bcl11a^F/Fmb1-Cre-YFP BM resulted in significantly reduced B cell and pDC cellularity compared to mb1-Cre-YFP controls, whereas BCL11A-sufficient (YFP⁻) pDCs persisted (Fig. 1C, 1E-F). Approximately 1/3 of pDCs in the spleen of wild-type chimeras were YFP⁺ compared to only 1/5 in BM (Fig. 1G). This increased fraction of YFP⁺ pDCs in the spleen suggests that BLP-derived pDCs preferentially home to that organ. While the residual YFP⁺ pDCs in the BM might be an indication that Bcl11a is not shut-off at the most precise developmentally-critical point within the B cell lineage for complete CD79a+ pDC ablation, other hematopoietic lineages were capable of development in normal numbers and contained a paucity of YFP⁺ cells (Fig. S2). Of note, BCL11A-deficient progenitors yielded higher splenic T cell chimerism at the expense of B cells and pDCs, but less than 2% of T cells were YFP⁺ (S2, and not shown). This indicated that mb1-cre expression occurs subsequent to T-B lineage divergence, consonant with previous observations^{26, 68, 69} and our mb1-cre progenitor analysis in which YFP⁺ cells are confined to the BLP derived compartment (Fig. S1). Because of their exclusive lymphoid derivation post T-B bifurcation, we were prompted hereafter to refer to this pDC lineage as “B-pDC”.

Relative to CD79a^- pDCs, resting B-pDCs express higher levels of MHC Class II (Fig 2A-B) suggesting that they may be primed for immediate response to pro-inflammatory signals. To test this hypothesis, we delivered TLR9 ligand (CpG:ODN) into mb1-Cre-YFP mice via tail vein injection, and splenic pDCs were phenotyped via flow cytometry 24 hours later. While both pDC and B-pDC compartments expanded relative to controls (Fig. 2C-D), the YFP⁺ B-pDC fraction increased almost 2-fold above that of the YFP⁻ pDC fraction in numbers (Fig. 2E). Additionally, fold increase of B-pDCs coexpressing CD83 and CD86 upon activation was also significantly higher than that of YFP^- pDCs (Fig. 2F-G). These results suggested that relative to YFP^- pDCs, B-pDCs are intrinsically activated and primed for rapid expansion upon TLR9 engagement. To confirm the functional phenotype of B-pDCs, we tested their ability to secrete cytokines known to be elicited by pDCs after Toll-like receptor (TLR) engagement. Specifically, we tested each pDC lineage for the production of IFN-α or IL-12p40 when activated by TLR9-bound CpG oligonucleotides. We sorted B-pDCs and pDCs, engaged TLR9 with CpG:ODNs for 24 hours and collected supernatant for cytokine specific ELISAs. IFN-α production was almost negligible in B-pDC (p<0.0001, Fig. 2H), yet IL-12p40 production was significantly augmented over pDC (p<0.0001, Fig. 2H). To test their ability to expand T lymphocytes in culture, sorted B-pDCs and pDCs were incubated with CpG:ODNs and co-cultured with freshly isolated CFSE-labeled lymphocytes. After 6 days, co-cultures were stained for CD3 and CFSE-negative cell percentages were recorded (Fig. 2I-J). Altogether, our results showed B-pDCs were significantly better at expanding T cells in co-cultures than conventional pDCs (p<0.001, Fig. 2I-J) upon TLR9 activation.

CLP-derived B-pDC are licensed for T cell activation
(A-B) Comparison of MHCII MFI in YFP⁺ and YFP^- pDCs from *Mb1*-Cre-YFP reporter mice via flow cytometric analysis. ***p=0.0005 (C-D) *Mb1*-Cre-YFP mice were injected with 50 ug/ml (100 ul) of CpG:ODN or control GpC:ODN and analyzed via flow cytometry for splenic pDC numbers/100,000 cells (***p<0.0001). (E) Flow cytometry quantification of YFP^- pDC and B-pDCs (YFP⁺) fold change in cell numbers upon CpG:ODN in vivo challenge (**p=0.002142 for YFP negative pDC expansion; ***p= 0.00023 for B-pDC expansion, and *p=0.0151 for CpG stimulated B-pDC and YFP^- pDC number comparison). (F-G) Fold change in CD86⁺CD89⁺ cell numbers for each pDC population in mice restimulated with CpG:ODN; (B-pDCs, ***p<0.0001; YFP^- pDCs, **p=0.026). Difference in CD86⁺CD89⁺ cell numbers between stimulated B-pDCs and YFP^- pDCs was also significant, **p=0.0034 (H) In vitro TLR9 engagement of B-pDC or pDC for ELISA against IFN-α orIL-12p40. (****p<0.0001 for both cytokines) (I) T cells were magnetically isolated from wildtype C57BL/6J mouse splenocytes using MACs columns, labelled with CSFE, and then cultured (2.5×10⁴/well) alone or with CpG:ODN activated B-pDCs or pDCs (5×10³/well) for 6 d. (J) The percentage of CFSE-negative CD3⁺ T cells in co-cultures were significantly higher in B-pDC compared to pDC (****p<0.0001). Student’s t-test was used for all statistical comparisons. Error bars=mean±s.d. The results are representative of three independent experiments each containing at least four mice per group.

To further elucidate their phenotype at the genetic level, we performed RNA-seq analyses of purified B-pDCs and compared them to classical, myeloid-derived bulk pDCs. Expression of hallmark pDC genes was confirmed by RT-PCR of shared pDC markers (Fig. S3A). While the overall gene expression patterns were highly similar across the two subsets, ∼1% of transcripts (∼220 genes) differed significantly (q value< 0.05) (Fig. S3B). Among the top overexpressed genes in B-pDCs were Lyz1, Ccr3, Cd86, Id2, Axl, Siglec1, and Cd81 (Fig. S3C). Differentially expressed transcripts generated Gene Ontology (GO)²⁷ or Panther²⁸ terms including “immune response”, “inflammatory response”, “cell activation”, and “regulation of immune response” (p=1.36×10^-17, 2.15×10^-12, 3.67×10^-11, and 1.45×10^-10, respectively (Fig.S3C). Gene set enrichment analysis (GSEA) revealed elevation of each of these same GO/Panther terms within the B-pDC subset as compared to a pDC-related GSEA control dataset that showed no enrichment. Next, we compared all genes expressed by both pDC populations to one another and to published²⁹ RNA-seq of BM-derived mouse pre-B cells (B220⁺ IgM^- Kit^- CD25⁺)—a post-CLP B cell progenitor (pre-B) population. As shown in Figure S3D, pDCs and B-pDCs expression levels were strongly correlated with one another relative to early B cells (R² values = 0.8959, 0.4145 and 0.404, respectively) (Fig.S3D). Collectively, these data support our contention that B-pDCs are functionally distinct from classical pDCs and specialize in inflammatory responses, antigen presentation, and T cell activation.

AXL⁺ transitional DCs (tDCs or AS-DCs)^{1, 57} have been identified in humans by the high expression levels of the genes AXL, CD81, CD86, LYZ2, C1QA, CD2, and C1QB relative to classical pDCs ^1–4. They also exhibit many classical pDC features, having previously been described phenotypically as a “continuum” between pDC and cDC2 populations ². In order to confirm AXL⁺ pDCs are also present in mice and corroborate our RNA-seq findings, we phenotyped B-pDCs through imaging flow cytometry. First, we again verified that our primary gating scheme effectively encompassed bona-fide pDCs using imaging flow cytometry. As expected, B220⁺ PDCA1^- cells were >95% CD19⁺ SIGLEC-H^- (B cells). Most importantly,>95% of B220⁺ PDCA1 ^hi cells were SIGLEC-H ⁺ and CD19^-, ruling out any pDC population contamination (Fig. S.4A-B). At the protein level, we were able to stain mouse BM cells with anti-AXL antibodies without issues and confirmed that while ∼60% of B-pDCs were positive for AXL, only ∼10% of YFP negative pDCs expressed this protein (Fig. S4A, C-E). Finally, imaging of BM cells showed that B-pDCs resembled classical secretory pDCs morphologically, presenting a large nucleus and defined plasma-like features, in contrast to B cells (Fig. S4A). This preliminary phenotypic analysis led us to speculate that similarly to human AXL⁺ transitional DCs, B-pDC might also prioritize their secretory capacity for copious transcription and secretion of alternative immune system modulators. Thus, we performed high resolution 10x single cell RNA-seq analyses of magnetically sorted PDCA1⁺ bone marrow cells (96% purity as determined by flow cytometry) and compared the B-pDC transcriptional profile to that of B cells and other pDCs. Principal component-based clustering and UMAP visualization of sorted PDCA⁺ cells yielded 22 discrete clusters. To identify cell populations in an unbiased manner, we uploaded the top 1000 differentially expressed genes (DEGs) from each cluster into CIPR (Cluster Identity Predictor)⁶⁶, which uses the reference database ImmGen (mouse) to calculate cell identity scores (S. Table 1). Seurat clusters were named according to the consensus of the top 5 cell ID hits generated by CIPR (Fig. 3A). We confirmed CIPR assignments by evaluating cluster DEGs and expression of markers associated with specific myeloid cell populations (Fig. S5). PDCA1⁺ cells comprised 9 distinct pDC populations (including B-pDCs, which were unanimously classified as pDCs by CIPR identity scores despite its unique Cd79a expression). Of note, pDC8 had one of the lowest ID scores among consensus pDCs (S. Table 1) and yielded CIPR identity hits for both pDCs and common myeloid progenitors “CMPs” within its top 5 ID hits (not shown). This suggested that this single cluster (1.7% of total cells) represented a mixed population of pDCs and differentiating CMPs expressing high levels of Hbb family genes (hemoglobin associated genes) under the chosen default Seurat clustering resolution settings. Because of pDC8 mixed identity, we excluded this cluster from our further single cell transcriptional analysis. Besides pDCs, a small B cell cluster (2.75% of all cells analyzed) expressing low levels of Bst2 (PDCA1 gene), as well as distinct subpopulations of macrophages (Mac1, 2, and 3), monocytes (Mono1, 2 and 3), and granulocytes (Granul1, 2, and 3) were also identified. Two stem-progenitor populations (Prog1 and 2) were present in our cell pool (1.12% and 2.86% of cells analyzed, respectively) (Fig. 3A-B). The Prog2 population expressed high levels of genes associated with pDCs, including Siglech, Bst2, Flt3, TCF4, and Irf8 and had a “CDC” (common dendritic cell progenitor) signature according to its CIPR reference ID (Fig. S5, S. Table 2). In addition, the majority of cells in the Prog2 cluster were Csf1r^-, Il7r⁺, Siglech⁺, Ly6d⁺ (Fig. S5, S. Table 2), resembling the pro-pDC myeloid precursor population described by Rodriguez et al, as well as Dress and colleagues^{67, 70}. In contrast, Prog1, which clustered near B cells and was identified as “MLP” (multi-lymphoid progenitors entering the CLP stage), expressed several pre/pro B cell genes, including Jchain, Iglc1, Iglc2, Igha, Igkc (S. Table 2). Direct DEG comparison between Prog1 and Prog2 clusters confirmed Prog1 cells exhibit a more “Pre/Pro B cell-like” developmental profile and low expression of markers that are “pDC biased” relative to Prog2 (S. Table 2). After clustering and profiling PDCA1⁺ cells, we subsetted the data in order to compare B-pDCs gene expression to only B cell and pDC clusters using Wilcoxon rank sum tests. We plotted several genes commonly associated with B cells, and pDCs. Except for Cd79a (Mb1) and Cd79b, B-pDCs showed negligible expression of mature B cell associated genes such as Pax5, Cd19 and Cd22 (Fig. 3C). Expression of canonical pDC genes was present but reduced in B-pDCs relative to classical pDC subsets (Fig. 3D). Our analysis highlights the B-pDC cluster uniqueness among pDC subsets and further supported its lineage divergence from mature B cells.

Single cell RNA-seq analysis of mouse PDCA1⁺ bone marrow cells
(A) UMAP generated by Seurat clustering analysis. Cluster identities were assigned using the top 5 consensus CIPR identity scores. (B) Heatmap of the top 5 DEGs from each cluster. After Seurat clustering, cell reads were subsetted to include only cells classified as B cells, pDCs, and B-pDCs. DEGS were generated for the new data subset and markers associated with (C) mature B cells (D) classical pDCs were plotted as violin plots.

In order to temporally examine lymphoid transcriptional commitment to B cell or B-pDC lineages, we inferred developmental trajectories for B and B-pDC single cells. Pseudotime is a measure of how much progress an individual cell has made through a process such as differentiation^{59, 60}. In other words, cells that belong to a single cluster can be segregated across various transcriptional “states” based on number of reads per cell of a particular set of developmental genes. Although we cannot say the pre/pro B cell-like Prog1 cluster represents an immediately adjacent common progenitor for B cells and B-pDCs, we confirmed B-pDCs are not a direct precursor for B cells based on their single cell developmental trajectory. Cells in the Prog1 cluster were chosen as state “0” based on their abundant expression of lymphoid progenitor associated genes (Fig. 4A). Analysis of uniquely expressed genes showed Prog1 downregulated several multipotency markers such as Stmn1, Gata2, and Ctla2a, while B cells increased expression of transcription factors needed for B cell commitment (e.g., Pax5, Ebf1, and Ms4a1) (Fig. 4B and 4C). Although B-pDCs did not express detectable levels of most B cell commitment markers evaluated, they did retain expression of early B cell receptor associated genes, including Ly6d, Ighm, Igkc, Iglc2, and J Chain (Fig. S6A). Because these early B cell related markers are also expressed in plasma cells, we performed flow cytometric analysis of B220⁺ PDCA1⁺: CD79a⁺ cells. We discovered that B-pDCs express negligible levels of TACI (a transcription factor that silences the B-cell developmental program ⁶²), and that expression of CD138 and IgG1 was also negligible (Fig. S6B, and not shown), thus ruling out any major plasma cell overlap at the protein level. Lineage analysis also revealed that B-pDCs enhanced expression of the pDC associated markers Siglech, Tcf4, Bst2, and also of Xbp1 (a marker expressed at high levels in secretory cells). Committed B cells in turn downregulated these markers as they progressed through pseudotime (Fig. 4D). In summary, lineage analyses revealed that as B-pDCs develop, they retain expression of both early B cell receptor genes and pDC associated genes, diverging from B cells through continued suppression of transcription factors required for terminal B cell differentiation.

Single cell trajectory analysis of Prog1, B cells, and B-pDC clusters
(A) Single cells belonging to clusters Prog1, B cells, and B-pDCs were ordered and plotted as a function of pseudotime based on uniquely expressed markers using the unsurpevised Monocle dpFeature. Cluster Prog1 was classified as the root of the trajectory given its high expression of pre-pro B cell and stem cell associated markers. DEGs of interest were plotted as a function of pseudotime in Prog1 (B), B cells (C), or pDCs (D) using Monocle. Branch Y_112 represents the expression kinetic trend of the B-pDC cluster (green), while branch Y_83 represents the expression kinetic trend in expression of B cells according to branched expression analysis modeling, or BEAM.

In order to determine if murine B-pDCs are homologous with AXL+ human DCs, we subsetted pDCs clusters and plotted their relative expression of AXL+ DC associated genes. We identified B-pDCs and pDC5 as “noncanonical pDCs” due to their distinguishing high expression of Axl (Fig. 5A). Accordingly, Axl negative pDCs are referred hereafter as “classical pDCs”. We then analyzed the top DEGs shared between B-pDCs and pDC5s using the function “FindConservedMarkers”. Lyz2, the complement genes C1qa, C1qb, C1qc, and Cd5l were the top conserved markers for these two pDC subsets (Fig. 5A). Of note, Cd5l, a macrophage associated secreted protein, has been shown to play a major role in initiating inflammation and maintaining humoral responses. ^71–73 Unexpectedly, the number of reads for myeloid markers previously used to define human AXL+ DCs via CyTOF or RNA-seq^{1, 57, 67, 69, 70} was negligible in murine pDCs and weakly correlated with expression of Axl itself in mice (Fig. S7). Cd81 expression, however, was highly expressed in B-pDCs (Fig.S7A). Of note, Cd81 and C1qb have been described as hallmark genes of a discrete mouse pDC population that does not produce type I IFN and yet mediates important immune functions previously attributed to all pDCs^{2, 37}. Altogether, our analysis revealed that constitutive expression of Axl and innate activation markers are present in more than one transcriptionally distinct murine pDC profile, stressing the functional heterogeneity and plasticity of murine pDCs.

Identification of Axl⁺ non-canonical pDCs in mice
(A) Violin plots depicting relative expression of conserved “non canonical *AXL*⁺ DCs” associated markers in mouse pDC clusters. (B) BM scRNA-seq data was subsetted to include only pDCs and Prog2 subsets. We used Monocle3 for partition-based learning of cell developmental trajectories. Louvain partitions were generated from single cell reads and their developmental trajectory was inferred using SimplePPT (top right quadrant). Pseudotime across clusters is shown in the (lower bottom quadrant).

Noncanonical AXL+ DCs in humans have been shown to be derived from a myeloid pro-pDC progenitor^{67, 69, 70}. In mice, Feng and colleagues used clonal lineage to show that pDCs and cDCs originate from a common Flt3-dependent pathway of differentiation originating from transient CX3CR1⁺ progenitors. Since the authors state that their finding does “not completely rule out the common origin of pDCs and B cells, which may bifurcate earlier in hematopoiesis”, we set out to corroborate the hematopoietic origins of pDC5 and B-pDCs. We first calculated pDC5’s “community connectedness” ⁷⁴ to other pDC clusters and plotted their developmental trajectories originating from the pro-pDC progenitor cluster Prog2 using Monocle3⁷⁵. Partitioning Prog2 and pDC single cells into Louvain communities yielded 5 supergroups, with the largest group encompassing most pDC clusters, including pDC5. The other groups consisted of non pDCs (Prog2 cells), B-pDCs, and a few single cells (>20 cells) genotypically distinct from their original pDC clusters (Fig. 5B). Ultimately, this statistical grouping confirmed that B-pDCs were not developmentally linked to “classical” AXL negative pDCs or Axl+ pDC5 cells. Accordingly, B-pDCs did not integrate into the predicted lineage trajectory originating from Prog2, whereas the Axl+ pDC5 population did (Fig. 5B). Both B-pDCs and pDC5s exhibited a similar late developmental pseudotime, while clusters pDC6 and pDC4 were the first to develop from Prog2. Collectively, our data establish that murine non canonical pDCs (AXL+) consist of at least two major pDC subpopulations, which develop from lymphoid and myeloid progenitors, respectively.

Discussion

Dendritic cell–subset biology, development, and the ensuing nomenclature have long been unclear and everchanging. Here, we provide definitive evidence in support of the long-suspected “lymphoid past” of pDCs by establishing their ability to arise in vivo from CLP derived progenitors with B/pDC bipotential lineage capacity^{17–19, 43, 44}. Our data shows that the murine pDC compartment is bipartite, being comprised of B-pDCs—diverted from the CLP post T-B bifurcation—as well as myeloid-derived classical pDCs. Unlike most myeloid derived pDCs, B-pDCs constitutively express high levels of multiple innate activation markers, including C1qa, C1qb, Lyz2, and Cd81. Functionally, they expand more readily after TLR9 engagement than classical pDCs and excel at activating T cells in culture. While further functional definition awaits discovery, our work provides a framework for the identification and segregation of the B-pDC lineage (comprising almost 1/5 of the total pDC compartment) from other myeloid-derived pDC subpopulations.

Primarily, our observations support the hypothesis that DC functionality derives primarily from ontogeny rather than from tissue environment ⁵⁰, exemplified by evolution of a specialized pDC lineage from a lymphoid progenitor. In the case of the B-pDCs, our data suggest such a cell may deviate from B cell commitment after Ly6d expression, thought to be the earliest marker of B cell specification²⁶.

Most noticeably, our newly described B-pDC population expresses high levels of the tyrosine kinase AXL and specializes in T cell activation through displaying high levels of MHC II and costimulatory markers. Of note, ablation of AXL in mice has previously been shown to increase expression of type I IFN while impairing IL-1β production and T cell activation during viral infections. ^{63, 64} Our related findings that AXL+ B-pDCs excel at T cell priming but exhibit reduced IFNα expression might suggest these DCs play an important role advancing T cell expansion, while curbing innate immune responses that can result in autoimmune damage. In addition to describing a novel B cell-like pDC population, we profiled murine pDCs, and established that murine AXL⁺ pDCs can develop in parallel from either myeloid or lymphoid progenitors, with lymphoid derived B-pDCs retaining expression of additional early B cell genes that are not expressed in myeloid derived pDCs. Intriguingly, ChIP-seq analysis done by our lab of BCL11A target binding in multiple human cell lines suggests that an evolutionarily conserved transcriptional hierarchy might distinguish AXL⁺ pDCs and B cells in humans as well (unpublished data). The demarcation of lymphoid derived B-pDCs as one of the AXL⁺ pDC populations found in mice may help clarify the perceived plasticity of the pDC compartment in normal and disease contexts⁵¹, as well as provide a new cell for targeted study within the context of autoimmune disease, cancer, and infection models.

Supplemental Figure Legends

Progenitor population analysis for mb1-cre driven YFP expression.
We analyzed YFP expression in BM progenitors to determine when and where mb1-Cre is active (representative plots shown). (A) Lineage negative cells were analyzed to examine LSK Hematopoietic progenitor (Lin^-, Sca-1⁺, c-Kit⁺), MPP (LSK, Flt3^int), and LMPP (LSK, Flt3^hi), populations. Virtually all cells were YFP^-. (B) Myeloid Dendritic Progenitors (MDP; Lin^-, IL7Rα^-, Flt3⁺, c-kit^hi, Sca1^-) and (C) Common Dendritic Cell Progenitors (CDP; Lin-, IL7Ra-, Flt3⁺, c-kitlo, Sca1-, CD115⁺) were also negative for YFP expression, while (D) Common Lymphoid Progenitors (CLP; Lin^-, IL7Ra⁺, Flt3⁺, c-kit^lo, Sca1^lo) contained YFP⁺ cells. Error bars=mean±s.d. The results are representative of three independent experiments each containing at least three mice per group.

Cell distributions in the spleen of adoptively transferred recipient mice.
Both *Bcl11a* ^F/F *mb1*-Cre⁺ and *Mb1*-Cre-YFP adoptively transferred recipient mice reconstituted other splenic cell types in normal numbers, including BM macrophages (CD11b⁺ F4/80⁺), granulocytes (Gr-1⁺ CD11b⁺) and splenic cDCs (CD11c⁺ CD11b⁺ B220^-). Total T cells (CD3⁺ B220^-) as well as CD4⁺ and CD8⁺ subsets were significantly increased in number in proportion to B/pDC cell loss. Student’s t-test was used for all statistical comparisons. The results are representative of two independent experiments each containing 3-4 mice per experimental group. Error bars=mean±s.d

Transcriptional analysis identifies two populations of pDC in mice: myeloid-derived classical pDC and CLP-derived B-pDC.
(A) Bone marrow pDC (B220⁺ PDCA1⁺ CD11c^int CD11b^-) were sorted based on expression of YFP. Four mice were pooled for each isolated RNA sample, for a total of three pDC and three B-pDC groups from 12 mice. RT-PCR was used to confirm that *Bcl11a, Bst2, SpiC, and Id2* expression match RNA-seq trends between pDC and B- pDC (B) RNA-seq was performed for gene expression analysis of pDC vs B-pDC and 220/23,946 genes (∼1%, left) were significantly differentially expressed (q value < 0.05, right heatmap, Log2 expression difference displayed). (C) Gene Set Enrichment Analysis (GSEA). Normalized enrichment score (NES) and false discovery rate q-values(FDR);FDR≤.25isconsideredsignificant⁵⁶. GO Term or Panther-derived pathways identified by DAVID analysis of ∼220 differentially expressed genes between the pDC and B- pDC subsets^{28, 29}. Among the top B-pDC DEGs were *Lyz1, Ccr3, Cd86, Cd83, Id2, Axl, Siglec1, and Cd81.* (D) Scatter plot comparisons of all genes with Reads Per Kilobase of transcript, per Million mapped reads (RPKM) >1. Correlations of pDC vs B-pDC (R2 value = 0.8959), B cell vs. pDC R2 value = 0.4145) and B cell vs. B-pDC (R2 value = 0.404) are indicated.

Imaging flow cytometry as a tool to quantify AXL expression in pDC populations.
(A-B) Representative Imaging flow cytometry of manuscript gating scheme used to define B cells and pDCs and histograms for SIGLEC-H, AXL and CD19 in B cells and pDCs (C) Representative Imaging flow cytometry plots of AXL⁺ pDCs (here alternatively gated as SIGLEC-H⁺ PDCA1⁺) in *Mb1*-Cre-YFP reporter mice bone marrow cells (n=5). (D) Complete statistics table generated with IDEAS for Image Stream analysis including non-stained wildtype controls. (E) Comparison of AXL expression in YFP^- and YFP⁺ (B-pDCs) (p<0.0001, unpaired t tests). The results are representative of two independent experiments each containing at least 3 mice per group.

Confirmation of unbiasedly identified cell cluster identities.
Seurat generated violin plots of genes associated with distinct cell populations including mature B cells, pDCs, granulocytes, pre-pDCs, NK cells, monocytes, macrophages, and myeloid progenitor cells. Plot identities were initially determined by running the package CIPR (Cluster Identity Predictor).

Expression of early B cell receptor genes in B-pDCs.
(A) Monocle was used to plot select early B cell genes as function of branched pseudotime for Prog1, B Cells, and B-pDC populations. (B) Plasma cell contamination in *Mb1*⁺ pDCs (B220⁺ PDCA1⁺) was ruled out by flow cytometric analysis of TACI and CD138 in BM of 6 C57BL/6J mice.

Transcriptional heterogeneity of *Axl⁺* murine pDC populations.
(A) Dot plot depicting expression prevalence of select “*AXL⁺* DC” associated markers from literature select and pDC DEGs for pDC subsets. (B) Correlation between expression of AXL+ DC associated markers and *Axl* in non-canonical mouse pDC subsets.

Top CIPR Assigned Identity Scores for Seurat Cluster DEGs.
CIPR IDs were generated using the top 1000 DEGS from each cluster. The top 5 consensus CIPR ID generated was used to rename Seurat clusters.

DEG comparison between progenitor clusters.
The top 50 differentially expressed features of Prog1 in relation to Prog2 were calculated using Seurat using non-parametric Wilcoxon rank sum test. In blue are upregulated genes and in red are downregulated genes.

Methods

Mice

Generation of Bcl11a cKO Mice (C57BL/6J background) was performed as described¹⁴. To generate conditional knockouts for the Mb1 gene, specifically, Bcl11a ^F/F mice were crossed to Mb1-Cre⁺ Rosa26-YFP⁺ reporter mice. Mb1-Cre deleter and Rosa26-YFP reporter strains (C57BL/6J genetic background) were obtained from Jackson labs (catalog #’s 020505 and 006148, respectively), bred, and genotyped according to the vendor protocol and their validated primers). All housing, husbandry, and experimental procedures with mice were approved by the Institutional Animal Care and Use Committees at the University of Texas at Austin. 6-week-old gender matched male or female mice were used for all experiments described in this manuscript. For each experiment, at least 4 samples per experimental group were used, with each experiment repeated at least 3 times (except for bone marrow reconstitutions and imaging flow cytometry experiments, which were repeated twice).

Tissue processing

Mouse femurs were cut at the extremities and a 25G needle was inserted in the bones to flush out bone marrow cells with 10 ml of cold PBS onto a 70 uM strainer in a 50 mL tube. Cells were washed from the strainers into the tubes with an additional 10 mL of PBS. Tubes were spun at 300G for 5 minutes and supernatant was decanted. 2 mL of ACK red blood cell lysis buffer was used to resuspend cells, which were incubated for 2 minutes and 20 seconds. Samples were diluted with 20 mL of cold PBS, washed once, re-filtered and resuspended in 200 uL of FACS buffer for subsequential flow cytometry staining. Spleens were mashed onto pre- wet 70 uM filters in a 50 mL tube with syringe plungers and 20 mL of cold PBS was added through the strainers into the tube. Cells were spun at 300G for 5 minutes and supernatant was decanted. 2 mL of ACK red blood cell lysis buffer was used to resuspend cells, which were incubated for 2 minutes and 20 seconds. Samples were diluted with 20 mL of cold PBS, washed once, re-filtered and resuspended in 200 uL of FACS buffer for subsequential flow cytometry staining.

Flow cytometry

Single cell suspensions (1×10⁶ cells) were resuspended in 200 uL of FACS buffer (5mM EDTA and 1% Heat Inactivated Fetal Calf Serum in D-PBS) and incubated for 15 minutes with surface antibody cocktail (each antibody at a concentration of 1.25 micrograms/ml in FACS buffer) in the dark at 4C. Cells were then washed 3 times with FACS buffer and ran live on a BD LSRII Fortessa. Single color controls for panel compensation were made using AbC Total Antibody Compensation Bead kit (ThermoFisher Scientific), by staining each control with one of the panel antibodies and using these controls to compensate the flow cytometer experiment prior to running the samples. A non-stained sample of cells was included in experimental runs and used to adjust for autofluorescence and any nonspecific staining. FMO single cell controls were also included in some of the experimental runs (cells stained with all antibodies in the panel minus one), for compensation quality control and to detect positive MFI thresholds for each fluorophore using cells. Sorting was performed on a FACS Aria (BD Biosciences). All analysis was done using FlowJo (Tree Star) software. Single-cell suspensions were stained with the following antigen-specific monoclonal antibodies (clones in parenthesis): anti CD45R/B220-BV605 (RA3-6B2), CD19-Alexa Fluor 700 (6D5), CD86-APC-Cy7 (GL-1), I- A/I-E-Pacific Blue (M5/114.15.2), IL7R- BV421 (A7R34), cKit-PE-Cy7 (ACK2), CD11b-PerCP-Cy5.5 (M1/70), CD3e-PerCP-Cy5.5 (500A2), Gr1-PerCP-Cy5.5 (RB6-8C5), CD4-PerCP-Cy5.5 (RM4-5), CD8-PerCpCy5.5(53-6.7), NK1.1-PerCp-Cy5.5 (PK136), Sca1-BV711 (D7), CD135-APC (A2F10), CD150-BV605 (TC15-12F12.2), CD79a-FITC (24C2.5) (Invitrogen), Anti-CD11c-PerCP-Cy5.5 (N418) (eBioscience), Anti-AXL-APC (175128)(R&D Systems), Anti-CD23 BV510 (B3B4) (Biolegend), Anti-CD267-BV421 (8F10) (BD Biosciences), Anti-CD138-APC (281-2) (Biolegend), Anti-CD83-PeCy7 (Michel-19), and Anti-PDCA1-PE or Biotin (JF05-1C2.4.1) (Miltenyi Biotech), in D-PBS/2% (vol/vol) FBS FACS buffer. In some experiments, viability stain Ghost Dye Red 780 (Fisher Scientific) was used to stain cells in PBS prior to antibody staining according to the manufacturer ‘s protocol. Flow cytometry data was analyzed with FlowJo, while Imaging flow cytometry was performed on ImageStream X and analyzed on IDEAS® 6.0 (Amnis), by pre-filtering out of focus cells and performing automatic software guided compensation with single color controls.

Bone Marrow Transplantation

Bone marrow from 6-week-old Bcl11a^F/F/mb1-Cre ⁺ cKO or Mb-1-Cre-YFP reporter mice (n=3 per group) were collected from femurs and 5×10^6 bone marrow cells were transferred via retro-orbital injection into recipient immunocompetent C57BL/6J mice lethally irradiated with two doses of 450 rad 1 hour apart. Mice were kept on an antibiotic diet for 3 weeks to allow for immune reconstitution. Eight weeks post transfer, mice were sacrificed and cells collected from BM and spleen to investigate reconstitution of cellular subsets.

RNA isolation and RNA-seq

For sorting prior to RNA collection, BM from 12 mice was prepared from femurs at 6 weeks of age, combined into three groups (4 mice/group). Total RNA was extracted from FACS sorted Bcl11a cKO BM pDCs (approximately 100K YFP⁺ pDCs and 400K YFP^- pDCs per pooled sample) using TRIzol reagent (Invitrogen). Taq polymerase (New England Biolabs) and a Perkin-Elmer 2700 thermocycler were used to amplify transcripts for the following mouse genes: Bcl11a (F: 5’-GTGGATAAGCCGCCTTCCCCTT-3’, R: 5’-GGGGACTTCCGTGTTCACTTTC-3’), Bst2 (F:5’-AGGCAAACTCCTGCAACCTG-3’, R: 5’-ACCTGCACTGTGCTAGAAGTC-3’), SpiC (F: 5’-ATCCTCACGTCAGAGGCAAC-3’, R: 5’-TGTACGGATTGGTGGAAGCC-3’, Id2 (F: 5’-GGACATCAGCATCCTGTCCTTGC-3’ R: 5’-GTGTTCTCCTGGTGAAATGGCTGA-3’,and β-actin (F:5’AGGTGTGATGGTGGGAAT-3’, R: 5’-GGTGTAAAACGCAGCTCAGT-3’) and probe RNA sequencing potential. For bulk RNA-seq, twelve mice were pooled into three YFP⁺ groups, and total RNA was extracted as above. Oligo-dT-primed cDNA was prepared using SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen). Poly(A) mRNA was enriched using the NEBNext magnetic isolation module (E7490) and samples underwent DNAse treatment. cDNA was prepared using the Ultra Low kit from Clontech (Mountain View, CA). Libraries were prepared at the according to manufacturer’s instructions for the NEBNext Ultra II Direction RNA kit (NEB, product number E7760). The resulting libraries tagged with unique dual indices were checked for size and quality using the Agilent High Sensitivity DNA Kit (Agilent). Library concentrations were measured using the KAPA SYBR Fast qPCR kit and loaded for sequencing on the Illumina NovaSeq 6000 instrument (paired-end 2X150, or single-end, 100 cycles; 30×10^6 reads/sample). Data were analyzed using a high-throughput next-generation sequencing analysis pipeline: FASTQ files were aligned to the mouse genome (mm9, NCBI Build 37) using TopHat2⁵³. Gene expression profiles for the individual samples were calculated with Cufflinks54 as RPKM values. YFP⁺ pDC samples were normalized to each control YFP-pDC sample. GO terms identified as significantly different: GO:0006955, GO:0009611, GO:0048584, GO:0006954, GO:0002684, GO:0001775, GO:0009986. Panther terms: BP00148, BP00155, BP00102, BP00120. Gene Set Enrichment Analysis (GSEA) using ordered gene expression levels of Common Lymphoid Progenitor-derived B-pDC and Common Myeloid Progenitor-derived pDC were significantly enriched in both GO terms and Panther-derived gene sets. A randomly selected control GSEA curated set, GSE7831: UNSTIM_VS_ INFLUENZA_STIM_ PDC_4H_DN (defined as genes down-regulated in untreated pDC versus influenza virus infected pDC 55 showed insignificant enrichment. Normalized enrichment score (NES) and false discovery rate q-values (FDR); FDR ≤ 0.25 is considered significant⁵⁶.

TLR9 engagement

Mb1-cre-YFP reporter mice were injected via tail veins with 100ul of PBS containing 50ug of CpG:ODN (n=4) or GpC:ODN control (n=4) (CpG-B no.1826, TCCAT GACG TTCCT GACGTT; control non-CpG-B no.2138, TCCATGAGCTTCCTGAGCTT, Invivogen, USA). At 24 hours, mice were sacrificed and bone marrow and spleens were collected for pDC phenotyping.

Cell co-cultures

Mb1-YFP⁺ B-pDC (B220⁺ PDCA1⁺YFP⁺) and Mb1-YFP^- pDCs (B220⁺PDCA1⁺YFP^-) subsets were sorted on a FACS Aria (BD Biosciences) by and cultured in 96-well round bottom plates (n=3 wells in triplicates, repeated twice) with RPMI medium 1640 containing 10% (vol/vol) FBS, 2 mM L-glutamine, 100 units/mL penicillin and streptomycin, 1 mM sodium pyruvate, and 10 mM Hepes, and with or without 5 μg/ml class A CpGs (ODN 1585, InvivoGen) for 24 hours. pDCs were then washed three times to remove residual CpG. Preactivated pDCs (5 × 10³) were cultured with lymphocytes (2.5 × 10⁴) magnetically sorted frommousespleen singlecell preparations using MACs columns and the EasySep™ Mouse TCell Isolation Kit (StemCell Technologies Cat# 19851) for a total of 6d with 20 units/mL of IL-2. CSFE labelling of lymphocytes (for which sorting purity was determined to be 97% via flow cytometry) prior to coculturing experiment was done using the CellTrace™ CFSE Cell Proliferation Kit (ThermoFisher Scientific; Cat# C34554) accordingly the manufacturer’s protocol.

Enzyme-linked immunosorbent assays (ELISA)

Sorted B-pDC and pDCs (5 × 10³ sorted cells per well, n= 3 per group, done in duplicate wells for each condition) were stimulated with 5 uM CpG-A (ODN 1585, Invivogen) or CpG-C (ODN 2395, Invivogen); or left untreated for 24 hs. ELISAs for IFN-α (Invitrogen) or IL-12p40 (BioLegend), respectively, was performed in 100 uL of cell free supernatants according to manufacturer’s instructions. Data was acquired using a Bio-Tek model EL311 automated microplate reader measuring absorbance at 450 nm.

Statistical analysis

All flow cytometry or ELISA data were analyzed with Prism (Version 8; GraphPad, La Jolla, CA) using Student’s t-test. All bar graphs show means and standard deviation (SD) and are representative of repeated experiments. These data were considered statistically significant when p-values were <0.05.

10x Single cell RNA-seq analysis

PDCA1⁺ C57BL/6J bone marrow cells from 4 pooled 6 week old wildtype C57BL/6J mice were magnetically sorted using MACS columns into culture medium, washed once with PBS ⁺ 0.04% BSA, and re-suspended in 32 μl PBS ⁺ 0.04% BSA. Single cell suspensions (50,000) were processed in the University of Texas Genomic Sequencing and Analysis Facility (GSAF). Cell suspensions were loaded on the Chromium Controller (10X Genomics) and processed for cDNA library generation following the manufacturer’s instructions for the Chromium NextGEM Single Cell 3’ Reagent Kit v3.1 (10X Genomics). The resulting libraries were examined for size and quality using the Bioanalyzer High Sensitivity DNA Kit (Agilent) and their concentrations were measured using the KAPA SYBR Fast qPCR kit (Roche). Samples were sequenced on the NovaSeq 6000 instrument (paired end, read 1: 28 cycles, read 2: 90 cycles) with a targeted depth of 71,130 reads/cell. Cellranger (v3.0.2) was used to demultiplex samples, to map data to the mouse transcriptome (mm10) and quantify genes. The gene counts matrix was read into Seurat (v3.2.1). Cells with unique feature counts over 3,500 or less than 200, and >3% mitochondrial read counts were removed from the analysis. The data was normalized and transformed using scTransform. Cells were clustered based on the top 1000 variable features, 20 PCs, and a resolution of 0.6 (Seurat’s graph-based clustering approach), and then visualized using UMAP. Analysis of the normalized and filtered single cell gene expression data (median of 16,181 genes across 13,838 single-cell transcriptomes) was used for the final expression table and downstream analysis. Wilcoxon rank sum test was used to test for differential expression among clusters and identify gene markers (logfc.threshold = - 1, min.pct= 0.25). Gene markers and other genes of interest were visualized using violin plots, dot plots, and heatmap plots. Unsupervised ordering of PDCA1⁺ cells was done with the Seurat integrated results as input to build a tree-like differentiation trajectory using the DDRTree algorithm of the R package Monocle v2 ⁵⁹. Select DEGs of B-pDCs, B Cells, and Prog1 populations were set as ordering genes for each trajectory, with the root state set as Prog1 classified cells. Expression of genes as a function of pseudotime was plotted with the plot_genes_branched_pseudotime Monocle function. We also used Monocle3 to generate partition-based learning of single cell developmental trajectories with SimplePPT and plot UMAP based pseudotime figures with initial reclustering parameters num_dim=60, resolution=1e-5.

Data Deposition

RNA-seq: GSE105827; ChIP-seq: GSE99019; 10x ScRNA-seq: GSE225768. Accession numbers to previously published data sets: GSE52868 (pre-B RNA-seq); GSE55043 (CAL1 ChIP-seq); ENCODE BCL11A ChIP-seq in GM1287838.

Author contributions

J.D.D. and G.C.I. designed research; A.A., J.D.D., K.G., C.R., Z.H., B-K.L., and G.C.I. performed research; J.D.D., J.L. V.R.I, L.I.E., S.Y., G.G., A.A., and G.C.I., contributed new reagents/analytic tools; A.A., J.D.D., C.R., B.-K.L., S.Z., D.O.H., and G.C.I., analyzed data and A.A. and J.D.D. wrote the paper. The authors declare no conflict of interest.

Acknowledgements

We thank June V. Harriss for expert assistance in the generation of Bcl11a conditional knockout mice, and Chhaya Das and Maya Ghosh for help in ChIP experiments and cell culture. The CAL1 cell line was kindly provided by Drs. Takahiro Maeda and Boris Reizis. Library preparation and Illumina ChIP- and RNA-seq were performed at the NGS core of the MD Anderson Cancer Center. Single Cell RNA-seq was performed at the University of Texas Genomic Sequencing and Analysis Facility with the help of Holly Stevenson and Dhivya Arasappan. The Lymphoma Research Foundation Fellowship 300463 (to J.D.D.); NIH grant F32CA110624 and Owens Medical Research Foundation grant (to G.C.I.); NIH grant R35GM133658 (to S.Y.); NIH grant R01AI104870 (to L.I.R.E.); NIH grant R01CA130075 (to V.R.I.); NIH Grant R01CA31534; Cancer Prevention Research Institute of Texas (CPRIT) Grant RP120348 to the MD Anderson NGS core and CPRIT RP120459 and the Marie Betzner Morrow Centennial Endowment (to H.O.T.) provided support for this work.

References

1
1. Villani A. C.
2. et al.
2017Single-cell RNA-seq reveals new types of human blood dendritic cells, monocytes, and progenitorsScience 356https://doi.org/10.1126/science.aah4573
2
1. Zhang H.
2. et al.
2017A distinct subset of plasmacytoid dendritic cells induces activation and differentiation of B and T lymphocytesProceedings of the National Academy of Sciences of the United States of America 114:1988–1993https://doi.org/10.1073/pnas.1610630114
3
1. Alcantara-Hernandez M.
2. et al.
2017High-Dimensional Phenotypic Mapping of Human Dendritic Cells Reveals Interindividual Variation and Tissue SpecializationImmunity 47:1037–1050https://doi.org/10.1016/j.immuni.2017.11.001
4
1. Matsui T.
2. et al.
2009CD2 distinguishes two subsets of human plasmacytoid dendritic cells with distinct phenotype and functionsJ Immunol 182:6815–6823https://doi.org/10.4049/jimmunol.0802008
5
1. Di Domizio J.
2. Cao W
2013Fueling Autoimmunity: Type I Interferon in Autoimmune DiseasesExpert review of clinical immunology 9https://doi.org/10.1586/eci.1512.1106
6
1. Swiecki M.
2. Colonna M
2015The multifaceted biology of plasmacytoid dendritic cellsNat Rev Immunol 15:471–485https://doi.org/10.1038/nri3865
7
1. Sisirak V.
2. et al.
2014Genetic evidence for the role of plasmacytoid dendritic cells in systemic lupus erythematosusJ Exp Med 211:1969–1976https://doi.org/10.1084/jem.20132522
8
1. Ronnblom L.
2. Pascual V
2008The innate immune system in SLE: type I interferons and dendritic cellsLupus 17:394–399https://doi.org/10.1177/0961203308090020
9
1. Cisse B.
2. et al.
2008Transcription factor E2-2 is an essential and specific regulator of plasmacytoid dendritic cell developmentCell 135:37–48
10
1. Ghosh H. S.
2. Cisse B.
3. Bunin A.
4. Lewis K. L.
5. Reizis B
2010Continuous expression of the transcription factor e2-2 maintains the cell fate of mature plasmacytoid dendritic cellsImmunity 33:905–916https://doi.org/10.1016/j.immuni.2010.11.023
11
1. Reizis B.
2. Bunin A.
3. Ghosh H. S.
4. Lewis K. L.
5. Sisirak V
2011Plasmacytoid dendritic cells: recent progress and open questionsAnnu Rev Immunol 29:163–183https://doi.org/10.1146/annurev-immunol-031210-101345
12
1. Grajkowska L. T.
2. et al.
2017Isoform-Specific Expression and Feedback Regulation of E Protein TCF4 Control Dendritic Cell Lineage SpecificationImmunity 46:65–77https://doi.org/10.1016/j.immuni.2016.11.006
13
1. Liu P.
2. et al.
2003BCL11A is essential for normal lymphoid developmentNat Immunol 4:525–532
14
1. Ippolito G. C.
2. et al.
2014Dendritic cell fate is determined by BCL11AProceedings of the National Academy of Sciences of the United States of America 111:E998–1006https://doi.org/10.1073/pnas.1319228111
15
1. Yu Y.
2. et al.
2012BCL11A is essential for lymphoid development and negatively regulates p53J Exp Med 209:2467–2483https://doi.org/10.1084/jem.20121846
16
1. Wu X.
2. et al.
2013BCL11A controls Flt3 expression in early hematopoietic progenitors and is required for pDC development in vivoPLoS One 8https://doi.org/10.1371/journal.pone.0064800
17
1. Pelayo R.
2. et al.
2005Derivation of 2 categories of plasmacytoid dendritic cells in murine bone marrowBlood 105:4407–4415https://doi.org/10.1182/blood-2004-07-2529
18
1. Sathe P.
2. Vremec D.
3. Wu L.
4. Corcoran L.
5. Shortman K
2013Convergent differentiation: myeloid and lymphoid pathways to murine plasmacytoid dendritic cellsBlood 121:11–19https://doi.org/10.1182/blood-2012-02-413336
19
1. Shigematsu H.
2. et al.
2004Plasmacytoid dendritic cells activate lymphoid-specific genetic programs irrespective of their cellular originImmunity 21:43–53https://doi.org/10.1016/j.immuni.2004.06.011
20
1. Wang Y. H.
2. Liu Y. J
2004Mysterious origin of plasmacytoid dendritic cell precursorsImmunity 21:1–2https://doi.org/10.1016/j.immuni.2004.07.003
21
1. Yang G. X.
2. et al.
2005Plasmacytoid dendritic cells of different origins have distinct characteristics and function: studies of lymphoid progenitors versus myeloid progenitorsJ Immunol 175:7281–7287
22
1. Mellor A. L.
2. et al.
2005Cutting edge: CpG oligonucleotides induce splenic CD19+ dendritic cells to acquire potent indoleamine 2,3-dioxygenase-dependent T cell regulatory functions via IFN Type 1 signalingJ Immunol 175:5601–5605
23
1. Munn D. H.
2. et al.
2004Expression of indoleamine 2,3-dioxygenase by plasmacytoid dendritic cells in tumor-draining lymph nodesJ Clin Invest 114:280–290https://doi.org/10.1172/jci21583
24
1. Hobeika E.
2. et al.
2006Testing gene function early in the B cell lineage in mb1-cre miceProceedings of the National Academy of Sciences of the United States of America 103:13789–13794https://doi.org/10.1073/pnas.0605944103
25
1. Sakaguchi N.
2. Kashiwamura S.
3. Kimoto M.
4. Thalmann P.
5. Melchers F
1988B lymphocyte lineage-restricted expression of mb-1, a gene with CD3-like structural propertiesThe EMBO journal 7:3457–3464
26
1. Inlay M. A.
2. et al.
2009Ly6d marks the earliest stage of B-cell specification and identifies the branchpoint between B-cell and T-cell developmentGenes & Development 23:2376–2381https://doi.org/10.1101/gad.1836009
27
1. Huang D. W.
2. Sherman B. T.
3. Lempicki R. A
2009Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene listsNucleic Acids Res 37:1–13https://doi.org/10.1093/nar/gkn923
28
1. Thomas P. D.
2. et al.
2003PANTHER: a library of protein families and subfamilies indexed by functionGenome research 13:2129–2141https://doi.org/10.1101/gr.772403
29
1. Liu G. J.
2. et al.
2014Pax5 loss imposes a reversible differentiation block in B-progenitor acute lymphoblastic leukemiaGenes Dev 28:1337–1350https://doi.org/10.1101/gad.240416.114
30
1. Greenwald R. J.
2. Freeman G. J.
3. Sharpe A. H
2005The B7 family revisitedAnnual Rev Immunol 23:515–548https://doi.org/10.1146/annurev.immunol
31
1. Seth S.
2. et al.
2011CCR7 Essentially Contributes to the Homing of Plasmacytoid Dendritic Cells to Lymph Nodes under Steady-State As Well As Inflammatory ConditionsThe Journal of Immunology 186:3364–3372https://doi.org/10.4049/jimmunol.1002598
32
1. Riol-Blanco L.
2. et al.
2005The Chemokine Receptor CCR7 Activates in Dendritic Cells Two Signaling Modules That Independently Regulate Chemotaxis and Migratory SpeedThe Journal of Immunology 174:4070–4080https://doi.org/10.4049/jimmunol.174.7.4070
33
1. Penna G.
2. Vulcano M.
3. Sozzani S.
4. Adorini L
2002Differential migration behavior and chemokine production by myeloid and plasmacytoid dendritic cellsHuman immunology 63:1164–1171
34
1. Cavanagh L. L.
2. Von Andrian U. H
2002Travellers in many guises: The origins and destinations of dendritic cellsImmunol Cell Biol 80:448–462
35
1. Shay T.
2. Kang J
2013Immunological Genome project and systems immunologyTrends in immunology 34:602–609https://doi.org/10.1016/j.it.2013.03.004
36
1. Liu Y.-J
2005IPC: Professional Type 1 Interferon-Producing Cells and Plasmacytoid Dendritic Cell PrecursorsAnnual Review of Immunology 23:275–306https://doi.org/10.1146/annurev.immunol.23.021704.115633
37
1. Lood C.
2. et al.
2009C1q inhibits immune complex-induced interferon-alpha production in plasmacytoid dendritic cells: a novel link between C1q deficiency and systemic lupus erythematosus pathogenesisArthritis and rheumatism 60:3081–3090https://doi.org/10.1002/art.24852
38
1. ENCODE
2004The ENCODE (ENCyclopedia Of DNA Elements) ProjectScience 306:636–640https://doi.org/10.1126/science.1105136
39
1. Karrich J. J.
2. et al.
2012Karrich, J. J. et al. The transcription factor Spi-B regulates human plasmacytoid dendritic cell survival through direct induction of the antiapoptotic gene BCL2-A1. Vol. 119(2012).The transcription factor Spi-B regulates human plasmacytoid dendritic cell survival through direct induction of the antiapoptotic gene BCL2-A1 119
40
1. Carmona-Sáez P.
2. et al.
2017Metagene projection characterizes GEN2.2 and CAL-1 as relevant human plasmacytoid dendritic cell models. Bioinformatics (OxfordEngland 33:3691–3695https://doi.org/10.1093/bioinformatics/btx502
41
1. Maeda T.
2. et al.
2005A novel plasmacytoid dendritic cell line, CAL-1, established from a patient with blastic natural killer cell lymphomaInternational journal of hematology 81:148–154
42
1. Ceribelli M.
2. et al.
2016A Druggable TCF4- and BRD4-Dependent Transcriptional Network Sustains Malignancy in Blastic Plasmacytoid Dendritic Cell NeoplasmCancer Cell 30:764–778https://doi.org/10.1016/j.ccell.2016.10.002
43
1. Bjorck P.
2. Kincade P. W
1998CD19+ pro-B cells can give rise to dendritic cells in vitroJ Immunol 161:5795–5799
44
1. Izon D.
2. et al.
2001A common pathway for dendritic cell and early B cell developmentJ Immunol 167:1387–1392
45
1. Ghosh H. S.
2. et al.
2014ETO family protein Mtg16 regulates the balance of dendritic cell subsets by repressing Id2J Exp Med 211:1623–1635https://doi.org/10.1084/jem.20132121
46
1. Sasaki I.
2. et al.
2012Spi-B is critical for plasmacytoid dendritic cell function and developmentBlood 120:4733–4743https://doi.org/10.1182/blood-2012-06-36527
47
1. Zhu X.
2. Schweitzer B. L.
3. Romer E. J.
4. Sulentic C. E.
5. DeKoter R. P
2008Transgenic expression of Spi-C impairs B-cell development and function by affecting genes associated with BCR signalingEur J Immunol 38:2587–2599https://doi.org/10.1002/eji.200838323
48
1. Li S. K.
2. Solomon L. A.
3. Fulkerson P. C.
4. DeKoter R. P
2015Identification of a negative regulatory role for spi-C in the murine B cell lineageJ Immunol 194:3798–3807https://doi.org/10.4049/jimmunol.1402432
49
1. Braig F.
2. et al.
2017Resistance to anti-CD19/CD3 BiTE in acute lymphoblastic leukemia may be mediated by disrupted CD19 membrane traffickingBlood 129:100–104https://doi.org/10.1182/blood-2016-05-718395
50
1. Heidkamp G. F.
2. et al.
2016Human lymphoid organ dendritic cell identity is predominantly dictated by ontogeny, not tissue microenvironmentScience Immunology 1https://doi.org/10.1126/sciimmunol.aai7677
51
1. Li S.
2. Wu J.
3. Zhu S.
4. Liu Y. J.
5. Chen J
2017Disease-Associated Plasmacytoid Dendritic CellsFront Immunol 8https://doi.org/10.3389/fimmu.2017.01268
52
1. Affandi A. J.
2. Carvalheiro T.
3. Radstake T.
4. Marut W
2017Dendritic cells in systemic sclerosis: Advances from human and mice studiesImmunol Lett https://doi.org/10.1016/j.imlet.2017.11.003
53
1. Kim D.
2. et al.
2013TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusionsGenome Biol 14https://doi.org/10.1186/gb-2013-14-4-r36
54
1. Trapnell C.
2. et al.
2013Differential analysis of gene regulation at transcript resolution with RNA-seqNature biotechnology 31:46–53https://doi.org/10.1038/nbt.2450
55
1. Iparraguirre A.
2. et al.
2008Two distinct activation states of plasmacytoid dendritic cells induced by influenza virus and CpG 1826 oligonucleotideJournal of leukocyte biology 83:610–620https://doi.org/10.1189/jlb.0807511
56
1. Mootha V. K.
2. et al.
2003PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetesNat Genet 34:267–273https://doi.org/10.1038/ng1180
57
1. Leylek R.
2. et al.
2019Integrated cross-species analysis identifies a conserved transitional dendritic cell populationCell Reports 29:3736–3750https://doi.org/10.1016/j.celrep.2019.11.042
58
1. Feng J.
2. et al.
2022Clonal lineage tracing reveals shared origin of conventional and plasmacytoid dendritic cellsImmunity 3:405–422https://doi.org/10.1016/j.immuni.2022.01.016
59.
1. Qiu X.
2. et. al.
2017Reversed graph embedding resolves complex single-cell trajectories:979–982https://doi.org/10.1038/nmeth.4402
60
1. Cao J
2. Spielmann M
3. Qiu X
4. Huang X
5. Ibrahim DM
6. Hill AJ
7. Zhang F
8. Mundlos S
9. Christiansen L
10. Steemers FJ
11. Trapnell C
12. Shendure J
2019The single-cell transcriptional landscape of mammalian organogenesisNature 566https://doi.org/10.1038/s41586-019-0969-x
61
1. Nikolic T
2. Dingjan GM
3. Leenen PJ
4. Hendriks RW
2002A subfraction of B220+ cells in murine bone marrow and spleen does not belong to the B cell lineage but has dendritic cell characteristicsEur J Immunol 32:686–92https://doi.org/10.1002/1521-4141(200203)32:3<686::AID-IMMU686>3.0.CO;2-I
62
1. Shapiro-Shelef M.
2. Calame K
2005Regulation of plasma-cell developmentNat Rev Immunol 5:230–242https://doi.org/10.1038/nri1572
63
1. Edward T
2. Schmid Iris K
3. Pang Eugenio A Carrera Silva Lidia Bosurgi Jonathan J Miner Michael S Diamond Akiko Iwasaki Carla V Rothlin
2016AXL receptor tyrosine kinase is required for T cell priming and antiviral immunity. eLife 5https://doi.org/10.7554/eLife.12414
64
1. Zhang D.
2. et al.
2022Axl Mediates Resistance to Respiratory Syncytial Virus Infection Independent of Cell AttachmentAm J Respir Cell Mol Biol 67:227–240https://doi.org/10.1165/rcmb.2021-0362OC
65
1. Hao Y
2. Hao S
3. Andersen-Nissen E III WMM
4. Zheng S
5. Butler A
6. Lee MJ
7. Wilk AJ
8. Darby C
9. Zagar M
10. Hoffman P
11. Stoeckius M
12. Papalexi E
13. Mimitou EP
14. Jain J
15. Srivastava A
16. Stuart T
17. Fleming LB
18. Yeung B
19. Rogers AJ
20. McElrath JM
21. Blish CA
22. Gottardo R
23. Smibert P
24. Satija R
2021Integrated analysis of multimodal single-cell dataCell https://doi.org/10.1016/j.cell.2021.04.048
66.
1. Ekiz H.A.
2. Conley C.J.
3. Stephens W.Z.
4. O’Connell R.M
2020CIPR: a web-based R/shiny app and R package to annotate cell clusters in single cell RNA sequencing experimentsBMC Bioinformatics 21https://doi.org/10.1186/s12859-020-3538-2
67.
1. Dress RJ
2. Dutertre CA
3. Giladi A
4. Schlitzer A
5. Low I
6. Shadan NB
7. Tay A
8. Lum J
9. Kairi MFBM
10. Hwang YY
11. Becht E
12. Cheng Y
13. Chevrier M
14. Larbi A
15. Newell EW
16. Amit I
17. Chen J
18. Ginhoux F.
2019Plasmacytoid dendritic cells develop from Ly6D+ lymphoid progenitors distinct from the myeloid lineageNat Immunol 20:852–864https://doi.org/10.1038/s41590-019-0420-3
68.
1. Nagaharu K
2. Kojima Y
3. Hirose H
4. Minoura K
5. Hinohara K
6. Minami H
7. Kageyama Y
8. Sugimoto Y
9. Masuya M
10. Nii S
11. Seki M
12. Suzuki Y
13. Tawara I
14. Shimamura T
15. Katayama N
16. Nishikawa H
17. Ohishi K
2022A bifurcation concept for B-lymphoid/plasmacytoid dendritic cells with largely fluctuating transcriptome dynamicsCell Rep 40https://doi.org/10.1016/j.celrep.2022.111260
69.
1. Herman JS Sagar
2. Grün D
2018FateID infers cell fate bias in multipotent progenitors from single-cell RNA-seq dataNat Methods 15:379–386https://doi.org/10.1038/nmeth.4662
70.
1. Rodrigues P.F.
2. Alberti-Servera L.
3. Eremin A.
4. et al.
2018Distinct progenitor lineages contribute to the heterogeneity of plasmacytoid dendritic cellsNat Immunol 19:711–722https://doi.org/10.1038/s41590-018-0136-9
71.
1. Sanjurjo L
2. Amézaga N
3. Aran G
4. Naranjo-Gómez M
5. Arias L
6. Armengol C
7. Borràs FE
8. Sarrias MR
2015The human CD5L/AIM-CD36 axis: A novel autophagy inducer in macrophages that modulates inflammatory responsesAutophagy 11:487–502https://doi.org/10.1080/15548627.2015.1017183
72.
1. Martinez VG
2. Escoda-Ferran C
3. Tadeu Simões I
4. Arai S
5. Orta Mascaró M
6. Carreras E
7. Martínez-Florensa M
8. Yelamos J
9. Miyazaki T
10. Lozano F
2014The macrophage soluble receptor AIM/Api6/CD5L displays a broad pathogen recognition spectrum and is involved in early response to microbial aggressionCell Mol Immunol 11:343–54https://doi.org/10.1038/cmi.2014.12
73.
1. Oskam Nienke
2. Boer Maurits den
3. Lukassen Marie V.
4. Heer Pleuni Ooijevaar-de
5. Veth Tim S.
6. van Mierlo Gerard
7. Lai Szu-Hsueh
8. Derksen Ninotska I.L.
9. Yin Victor C.
10. Streutker Marij
11. Franc Vojtech
12. Siborova Marta
13. Damen Mirjam
14. Kos Dorien
15. Barendregt Arjan
16. Bondt Albert
17. van Goudoever Johannes B.
18. Haas Carla J.C.
19. Aerts Piet C.
20. Muts Remy M.
21. Rooijakkers Suzan H.M.
22. Vidarsson Gestur
23. Rispens Theo
24. Heck Albert J.R.
CD5L is a canonical component of circulatory IgMhttps://doi.org/10.1101/2023.05.27.542462
74.
1. Wolf F.A.
2. Hamey F.K.
3. Plass M.
4. et al.
2019PAGA: graph abstraction reconciles clustering with trajectory inference through a topology preserving map of single cellsGenome Biol 20https://doi.org/10.1186/s13059-019-1663-x
75.
1. Trapnell C.
2. et. al.
2014The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cellsNat. Biotechnol :381–386https://doi.org/10.1038/nbt.2859
76.
1. Harman Benjamin C.
2. Miller Juli P.
3. Nikbakht Neda
4. Gerstein Rachel
5. Allman David
2006Mouse plasmacytoid dendritic cells derive exclusively from estrogen-resistant myeloid progenitorsBlood 108:878–885https://doi.org/10.1182/blood-2005-11-4545

Article and author information

Author information

Alessandra M. Araujo
Department of Chemical Engineering, The University of Texas at Austin, Austin TX, USA
Joseph D. Dekker
Department of Chemical Engineering, The University of Texas at Austin, Austin TX, USA
ORCID iD: 0000-0002-2068-3529
- Corresponding Authors: Haley O. Tucker, Department of Molecular Biosciences. The University of Texas at Austin, 1 University Station A5000, Austin TX 78712, USA. Phone: (512) 475-7706; Fax (512) 475-7707; E-mail: haleytucker@austin.utexas.edu, Joseph D. Dekker, Department of Molecular Biosciences. The University of Texas at Austin, 1 University Station A5000, Austin TX 78712, USA. Phone: (505)4637855; E-mail: joedekker@gmail.com, S. Stephen Yi, Livestrong Cancer Institutes, and Department of Biomedical Engineering, The University of Texas at Austin, Austin TX 78712, USA. Phone: (512) 495-5245; E-mail: stephen.yi@austin.utexas.edu;
Kendra Garrison
Department of Chemical Engineering, The University of Texas at Austin, Austin TX, USA
Zhe Su
Department of Biomedical Engineering, and Livestrong Cancer Institutes, The University of Texas at Austin, Austin TX, USA
Catherine Rhee
Department of Molecular Biosciences, The University of Texas at Austin, Austin TX, USA
Zicheng Hu
Department of Molecular Biosciences, The University of Texas at Austin, Austin TX, USA
Bum-Kyu Lee
Department of Molecular Biosciences, The University of Texas at Austin, Austin TX, USA
Daniel Osorio Hurtado
Department of Biomedical Engineering, and Livestrong Cancer Institutes, The University of Texas at Austin, Austin TX, USA
ORCID iD: 0000-0003-4424-8422
Jiwon Lee
Department of Chemical Engineering, The University of Texas at Austin, Austin TX, USA
Vishwanath R. Iyer
Department of Molecular Biosciences, The University of Texas at Austin, Austin TX, USA
Lauren I. R. Ehrlich
Department of Molecular Biosciences, The University of Texas at Austin, Austin TX, USA
George Georgiou
Department of Chemical Engineering, The University of Texas at Austin, Austin TX, USA, Department of Molecular Biosciences, The University of Texas at Austin, Austin TX, USA
Gregory C. Ippolito
Department of Molecular Biosciences, The University of Texas at Austin, Austin TX, USA
S. Stephen Yi
Department of Biomedical Engineering, and Livestrong Cancer Institutes, The University of Texas at Austin, Austin TX, USA
- Corresponding Authors: Haley O. Tucker, Department of Molecular Biosciences. The University of Texas at Austin, 1 University Station A5000, Austin TX 78712, USA. Phone: (512) 475-7706; Fax (512) 475-7707; E-mail: haleytucker@austin.utexas.edu, Joseph D. Dekker, Department of Molecular Biosciences. The University of Texas at Austin, 1 University Station A5000, Austin TX 78712, USA. Phone: (505)4637855; E-mail: joedekker@gmail.com, S. Stephen Yi, Livestrong Cancer Institutes, and Department of Biomedical Engineering, The University of Texas at Austin, Austin TX 78712, USA. Phone: (512) 495-5245; E-mail: stephen.yi@austin.utexas.edu;
Haley O. Tucker
Department of Molecular Biosciences, The University of Texas at Austin, Austin TX, USA
- Corresponding Authors: Haley O. Tucker, Department of Molecular Biosciences. The University of Texas at Austin, 1 University Station A5000, Austin TX 78712, USA. Phone: (512) 475-7706; Fax (512) 475-7707; E-mail: haleytucker@austin.utexas.edu, Joseph D. Dekker, Department of Molecular Biosciences. The University of Texas at Austin, 1 University Station A5000, Austin TX 78712, USA. Phone: (505)4637855; E-mail: joedekker@gmail.com, S. Stephen Yi, Livestrong Cancer Institutes, and Department of Biomedical Engineering, The University of Texas at Austin, Austin TX 78712, USA. Phone: (512) 495-5245; E-mail: stephen.yi@austin.utexas.edu;

Version history

Preprint posted: January 21, 2024
Sent for peer review: January 31, 2024
Reviewed Preprint version 1: May 15, 2024
Reviewed Preprint version 2: August 8, 2024
Version of Record published: September 13, 2024

Copyright

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Not revised: This Reviewed Preprint includes the authors’ original preprint (without revision), an eLife assessment, and public reviews.

Reviewing Editor
Dipyaman Ganguly
Indian Institute of Chemical Biology, Kolkata, India
Senior Editor
Satyajit Rath
Indian Institute of Science Education and Research (IISER), Pune, India

Reviewer #1 (Public Review):

Summary:

Plasmacytoid dendritic cells (pDCs) represent a specialized subset of dendritic cells (DCs) known for their role in producing type I interferons (IFN-I) in response to viral infections. It was believed that pDCs originated from common DC progenitors (CDP). However, recent studies by Rodrigues et al. (Nature Immunology, 2018) and Dress et al. (Nature Immunology, 2019) have challenged this perspective, proposing that pDCs predominantly develop from lymphoid progenitors expressing IL-7R and Ly6D. A minor subset of pDCs arising from CDP has also been identified as functionally distinct, exhibiting reduced IFN-I production but a strong capability to activate T-cell responses. On the other hand, clonal lineage tracing experiments, as recently reported by Feng et al. (Immunity, 2022), have demonstrated a shared origin between pDCs and conventional DCs (cDCs), suggesting a contribution of common DC precursors to the pDC lineage.

In this context, Araujo et al. investigated the heterogeneity of pDCs in terms of both development and function. Their findings revealed that approximately 20% of pDCs originate from lymphoid progenitors common to B cells. Using Mb1-Cre x Bcl11a floxed mice, the authors demonstrated that the development of this subset of pDCs, referred to as "B-pDCs," relied on the transcription factor BCL11a. Functionally, B-pDCs exhibited a diminished capacity to produce IFN-I in response to TLR9 agonists but secreted more IL-12 compared to conventional pDCs. Moreover, B-pDCs, either spontaneously or upon activation, exhibited increased expression of activation markers (CD80/CD86/MHC-II) and a heightened ability to activate T-cell responses in vitro compared to conventional pDCs. Finally, Araujo et al. characterized these B-pDCs at the transcriptomic level using bulk and single-cell RNA sequencing, revealing them as a unique subset of pDCs expressing certain B cell markers such as Mb1, as well as specific markers (Axl) associated with cells recently described as transitional DCs.

Thus, in contrast to previous findings, this study posits that a small proportion of pDCs derive from B cell-committed lymphoid progenitors, and this subset of B-pDCs exhibits distinct functional characteristics, being less specialized in IFN-I production but rather in T cell activation.

Strengths:

Previously, the same research group delineated the significance of BCL11a as a critical transcription factor in pDC development (Ippolito et al., PNAS, 2014). This study elucidates the precise stage during hematopoiesis at which BCL11a expression becomes essential for the emergence of a distinct subset of pDCs, substantiated by robust genetic evidence in vivo. Furthermore, it underscores the shared developmental origin between pDCs and B cells, reinforcing prior research in the field that suggests a lymphoid origin of pDCs. Finally, this work attributes specific functional properties to pDCs originating from these lymphoid progenitors shared with B cells, emphasizing the early imprinting of functional heterogeneity during their development.

Weaknesses:

The authors delineate a subset of pDCs dependent on the BCL11a transcription factor, originating from lymphoid progenitors, and compare it to conventional pDCs, which they suggest differentiate from common DC progenitors of myeloid origin. However, this interpretation lacks support from the authors' data. Their single-cell RNA sequencing data identifies cells corresponding to progenitors (Prog2), from which the majority of pDCs, termed conventional pDCs, likely originate. This progenitor cell population expresses Il7r, Siglech, and Ly6D, but not Csfr1. The authors describe this progenitor as resembling a "pro-pDC myeloid precursor," yet these cells align more closely with lymphoid (Il7r+) progenitors described by Rodrigues et al. (Nature Immunology, 2018) and Dress et al. (Nature Immunology, 2019). Furthermore, analysis of their Mb1 reporter mice reveals that only a fraction of common lymphoid progenitors (CLP) express YFP, giving rise to a fraction of YFP+ pDCs. However, this does not exclude the possibility that YFP- CLP could also give rise to pDCs. The authors could address this caveat by attempting to differentiate pDCs from both YFP+ and YFP- CLPs in vitro in the presence of FLT3L. Additionally, transfer experiments using these lymphoid progenitors could be conducted in vivo to assess their differentiation potential in competitive settings.

Using their Mb1-reporter mice, the authors demonstrate that YFP pDCs originating from lymphoid progenitors are functionally distinct from conventional pDCs, mostly in vitro, but their in vivo relevance remains unknown. It is crucial to investigate how Bcl11a conditional deficiency in Mb1-expressing cells affects the anti-viral immune response, for example, using the M-CoV infection model as described by Sulczewski et al. in Nature Immunology, 2023. Particularly, the authors suggest that their B-pDCs act as antigen-presenting cells involved in T-cell activation compared to conventional pDCs. However, these findings contrast with those of Rodrigues et al., who have shown that pDCs of myeloid origin are more effective than pDCs of lymphoid origin in activating T-cell responses. The authors should discuss these discrepancies in greater detail. It is also notable that B-PDCs acquire the expression of ID2 (Figure S3A), commonly a marker of conventional/myeloid DCs. The authors could analyze in more detail the acquisition of specific myeloid features (CD11c, CX3CR1) by this B-PDCs subset and discuss how the expression of ID2 may impair classical pDC features, as ID2 is a repressor of E2-2, a master regulator of pDC fate.

Finally, through the analysis of their single-cell RNA sequencing data, the authors show that the subset of B-pDCs they identified expresses Axl, confirmed at the protein level. Given this specific expression profile, the authors suggest that B-pDCs are related to a previously described subset of transitional DCs, which were reported to share a common developmental path with pDCs, (Sulczewski et al. in Nature Immunology, 2023). While intriguing, this observation requires further phenotypic and functional characterization to substantiate this claim.

https://doi.org/10.7554/eLife.96394.1.sa1

Reviewer #2 (Public Review):

Summary:

The origin of plasmatoid dendritic cells and their subclasses continues to be a debated field, akin to any immune cell field that is determined through the expression of surface markers (relative to clear subclass separation based on functional biology and experimentation). In this context, in this manuscript by Araujo et al, the authors attempt to demonstrate that a subtype of pDCs comes from lymphoid origin due to the presence of some B cell gene expression markers. They nomenclature these cells as B-pDCs. Strikingly, pDCs function via expression of IFNa where as B-pDCs do not express IFNa - thereby raising the question of what are their physiological or pathophysiological properties. B-pDCs also express AXL, a marker not seen in mouse pDCs but observed in human pDCs. Overall, using a combination of gene expression profiling of immune cells isolated from mice via RNA-seq and single-cell profiling the authors propose that B-pDCs are a novel subtype of pDCs in mice that were not previously identified and characterized.

Weaknesses:

My two points of discussion about this manuscript are as follows.

(1) How new are these observations that pDCs could also originate from common lymphoid progenitors. This fact has been previously outlined by many laboratories including Shigematsu et al, Immunity 2004. These studies in the manuscript can be considered new based on the single-cell profiling presented, only if the further characterization of the isolated B-pDCs is performed at the functional biology level. Overlapping gene expression profiles are often seen in developing immune cell types- especially when only evaluated at the RNA expression level- and can lead to cell type complexity (and identification of new cell types) that are not biologically and functionally relevant.

(2) The authors hardly perform any experiments to interrogate the function of these B-pDCs. The discussion on this topic can be enhanced. Ideally, some biological experiments would confirm that B-pDCs are important.

https://doi.org/10.7554/eLife.96394.1.sa0

Significance of findings

Strength of evidence

Abstract

Introduction

Results

Mb1-Cre deletion of Bcl11a identifies a CLP-derived subset of pDCs

CLP-derived B-pDC are licensed for T cell activation

Single cell RNA-seq analysis of mouse PDCA1+ bone marrow cells

Single cell trajectory analysis of Prog1, B cells, and B-pDC clusters

Identification of Axl+ non-canonical pDCs in mice

Discussion

Supplemental Figure Legends

Progenitor population analysis for mb1-cre driven YFP expression.

Cell distributions in the spleen of adoptively transferred recipient mice.

Transcriptional analysis identifies two populations of pDC in mice: myeloid-derived classical pDC and CLP-derived B-pDC.

Imaging flow cytometry as a tool to quantify AXL expression in pDC populations.

Confirmation of unbiasedly identified cell cluster identities.

Expression of early B cell receptor genes in B-pDCs.

Transcriptional heterogeneity of Axl+ murine pDC populations.

Top CIPR Assigned Identity Scores for Seurat Cluster DEGs.

DEG comparison between progenitor clusters.

Methods

Mice

Tissue processing

Flow cytometry

Bone Marrow Transplantation

RNA isolation and RNA-seq

TLR9 engagement

Cell co-cultures

Enzyme-linked immunosorbent assays (ELISA)

Statistical analysis

10x Single cell RNA-seq analysis

Data Deposition

Author contributions

Acknowledgements

References

Article and author information

Author information

Alessandra M. Araujo

Joseph D. Dekker

Kendra Garrison

Zhe Su

Catherine Rhee

Zicheng Hu

Bum-Kyu Lee

Daniel Osorio Hurtado

Jiwon Lee

Vishwanath R. Iyer

Lauren I. R. Ehrlich

George Georgiou

Gregory C. Ippolito

S. Stephen Yi

Haley O. Tucker

Version history

Copyright

Peer review process

Editors

Single cell RNA-seq analysis of mouse PDCA1⁺ bone marrow cells

Identification of Axl⁺ non-canonical pDCs in mice

Transcriptional heterogeneity of Axl⁺ murine pDC populations.