記住我
Research ArticleHematology
Open Access | 
10.1172/jci.insight.183889
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by Booth, C. in: JCI | PubMed | Google Scholar
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by Bouyssou, J. in: JCI | PubMed | Google Scholar
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by Togami, K. in: JCI | PubMed | Google Scholar
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by Armand, O. in: JCI | PubMed | Google Scholar
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by
Rivas, H.
in:
JCI
|
PubMed
|
Google Scholar
|
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by Yan, K. in: JCI | PubMed | Google Scholar
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by Rice, S. in: JCI | PubMed | Google Scholar
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by Cheng, S. in: JCI | PubMed | Google Scholar
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by Lachtara, E. in: JCI | PubMed | Google Scholar
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by
Bourquin, J.
in:
JCI
|
PubMed
|
Google Scholar
|
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by
Kentsis, A.
in:
JCI
|
PubMed
|
Google Scholar
|
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by Rheinbay, E. in: JCI | PubMed | Google Scholar
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by
DeCaprio, J.
in:
JCI
|
PubMed
|
Google Scholar
|
1Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, USA.
2Program in Virology, Graduate School of Arts and Sciences, Harvard University, Cambridge, Massachusetts, USA.
3Tow Center for Developmental Oncology, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
4Molecular Pharmacology Program, Sloan Kettering Institute, New York, New York, USA.
5Departments of Pediatrics, Pharmacology, and Physiology & Biophysics, Weill Medical College of Cornell University, New York, New York, USA.
6Krantz Family Center for Cancer Research, Boston, Massachusetts, USA
7Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
8Division of Oncology, Children’s Research Center, University Children’s Hospital, Zurich, Switzerland.
Address correspondence to: Andrew A. Lane, Dana-Farber Cancer Institute, 450 Brookline Ave., Mayer 413, Boston, Massachusetts, 02215, USA. Phone: 617.632.4589; Email: andrew_lane@dfci.harvard.edu.
Find articles by
Lane, A.
in:
JCI
|
PubMed
|
Google Scholar
|
Published November 5, 2024 - More info
Published in Volume 9, Issue 24 on December 20, 2024
AbstractMYB fusions are recurrently found in select cancers, including blastic plasmacytoid DC neoplasm (BPDCN), an acute leukemia with poor prognosis. They are markedly enriched in BPDCN compared with other blood cancers and, in some patients, are the only obvious somatic mutation detected. This suggests that they may alone be sufficient to drive DC transformation. MYB fusions are hypothesized to alter the normal transcription factor activity of MYB, but, mechanistically, how they promote leukemogenesis is poorly understood. Using CUT&RUN chromatin profiling, we found that, in BPDCN leukemogenesis, MYB switches from being a regulator of DC lineage genes to aberrantly regulating G2/M cell cycle control genes. MYB fusions found in patients with BPDCN increased the magnitude of DNA binding at these locations, and this was linked to BPDCN-associated gene expression changes. Furthermore, expression of MYB fusions in vivo impaired DC differentiation and induced transformation to generate a mouse model of myeloid-dendritic acute leukemia. Therapeutically, we present evidence that all-trans retinoic acid (ATRA) may cause loss of MYB protein and cell death in BPDCN.
Graphical Abstract
Introduction
Rearrangements of the hematopoietic transcription factor MYB are recurrently found in 20% of patients with blastic plasmacytoid DC neoplasm (BPDCN), a rare leukemia arising from cells of the plasmacytoid DC (pDC) lineage (1–4). These rearrangements result in fusion of the MYB N-terminus to the C-terminus of one of a number of partner proteins, including PLEKHO1, DCPS, and ZFAT (2). The MYB::ZFAT fusion is out of frame, resulting in expression of truncated MYB with no partner protein. Loss of the C-terminus of MYB is thought to increase its activity via a number of proposed mechanisms (2, 5, 6).
MYB fusions have been identified in a select number of solid tumors, including adenoid cystic carcinoma (ACC) (7, 8) and angiocentric glioma (9). While expression of WT MYB is a dependency in many leukemias (10–13) and is sometimes associated with translocations that bring an active promoter into proximity of the MYB locus (14, 15), MYB fusions are very rare in non-BPDCN leukemias (4). Interestingly, in children, MYB fusions occur in up to 50% of patients and appear to require few or possibly no cooperating mutations to induce BPDCN (2). This suggests that, in some contexts, MYB fusions may be sufficient to induce leukemic transformation of pDC lineage cells. How this occurs, and why other hematopoietic lineages are not similarly affected, is unknown.
Globally, the mutational landscape of BPDCN shows similarities to both myeloid and lymphoid malignancies (4), which may reflect the lympho-myeloid ontogeny of pDCs (16, 17). Myeloid-type mutations in epigenetic regulators (such as TET2, ASXL1, and EZH2) and splicing factors (such as ZRSR2 and SRSF2) are highly prevalent (18–20) and may arise from preleukemic clones (21). Other common BPDCN mutations, such as in the transcription factors IKZF1 and ETV6 (22, 23) and cell cycle regulator CDKN2A (22, 24–26), are normally associated with lymphoid malignancies. Disruption of the G1/S checkpoint via mutations in CDKN2A and/or RB1 appears to be a hallmark of BPDCN (4, 24, 27). The CDKN2A gene encodes 2 proteins, p16INK4A and p14ARF (p19ARF in mouse), which together act to halt cell cycle progression in G1 phase through regulation of cyclin-dependent kinases and p53 (28, 29).
All-trans retinoic acid (ATRA) is a well-established treatment for acute promyelocytic leukemia (APL), which is induced by the fusion protein PML::RARA (30). In ACC, ATRA has been shown to cause downregulation of MYB expression, leading to reduced viability (31). Use of ATRA to downregulate MYB could, therefore, be a potential therapeutic approach for MYB-dependent leukemias, including BPDCN.
ResultsMYB aberrantly regulates G2/M cell cycle genes in BPDCN. To investigate the role of MYB in BPDCN, we performed CUT&RUN for MYB and the marker of active chromatin H3K27 acetylation (H3K27ac) in BPDCN cells and in normal pDCs. We used CUT&RUN rather than ChIP-Seq because of its lower cell input requirement (100,000 cells or fewer) (32) and the fact that normal pDCs are difficult to harvest from peripheral blood in sufficient quantities for ChIP-Seq. We observed similar chromatin profiles in primary patient BPDCN cells and in patient-derived xenografts (PDXs) from patients with BPDCN (Figure 1A and Supplemental Figure 1, A and B; supplemental material available online with this article; https://doi.org/10.1172/jci.insight.183889DS1). This facilitated more robust mechanistic analyses, as PDXs are a sustainable source of larger quantities of primary leukemia cells that phenocopy BPDCN cells harvested directly from patients (20, 33).
Figure 1MYB aberrantly regulates G2/M cell cycle genes in BPDCN. (A) MYB, H3K27ac, and IgG CUT&RUN tracks at the BCL2 and CLEC4C (BDCA2) loci in the indicated cell types. (B) Top 5 ranked motifs enriched in MYB CUT&RUN peaks in normal pDCs and BPDCN PDX cells, and percent of peaks containing each motif. (C) Top 10 ranked Reactome gene sets overlapping with genes showing promoter binding of MYB in the indicated samples. (D) (Left) GSEA comparing normal pDCs and BPDCN PDX cells for genes showing differential MYB binding at promoter regions in BPDCN PDX AL03 relative to normal pDCs (n = 283 genes increased and n = 181 genes decreased MYB binding). (Right) Top 3 ranked motifs enriched in differentially bound MYB sites mapping to promoters, and percent of sites containing each motif. (E) MYB and IgG CUT&RUN tracks at the CDC20 locus in the indicated cell types.
In normal pDCs, BPDCN PDXs, and the BPDCN cell line CAL1, we observed MYB binding and H3K27ac at pDC-lineage genes such as CLEC4C (BDCA2), as expected (Figure 1A). In contrast, at chromatin sites associated with oncogenes, such as at a BCL2 3′ enhancer, we observed increased MYB occupancy and H3K27ac in BPDCN cells compared with normal pDCs (Figure 1A), linking MYB to increased BCL2 expression in BPDCN (34).
We generated consensus MYB CUT&RUN peak sets for each sample type: normal pDCs (n = 16,018), BPDCN PDX (n = 17,806), and CAL1 (n = 17,055) (Supplemental Figure 1C and Supplemental Table 1). In all 3 cell types, MYB binding sites were strongly enriched for the DNA binding motif of TCF4, a critical transcriptional regulator of the pDC lineage (35, 36) (Figure 1B and Supplemental Figure 1D). In agreement with this, MYB binding sites in CAL1 cells showed strong overlap with TCF4 binding sites in CAL1 cells from a previously published ChIP-Seq dataset (36) (Supplemental Figure 1E). Canonical MYB DNA binding motifs were relatively weakly enriched compared with TCF4, SPI1, RUNX, and IRF motifs. This may indicate cooperative DNA binding with other transcription factors at lineage-related MYB binding sites in these cells.
To explore which genes were regulated by MYB in BPDCN and normal pDCs, we narrowed our analysis to binding sites located at gene promoters. Genes with promoters bound by MYB in CAL1, BPDCN PDX, and normal pDCs (n = 353) were enriched in Reactome gene sets related to the pDC lineage, such as “Innate Immune System” and “Viral Infection Pathways” (Figure 1C). Conversely, genes with promoters selectively bound by MYB in CAL1 and BPDCN PDX but not normal pDCs (n = 380) were enriched in the “Cell Cycle” and “Cell Cycle, Mitotic” gene sets, suggesting that MYB plays a role in direct regulation of cell cycle genes in malignant BPDCN, particularly those active in the G2/M checkpoint (37).
We then performed genome-wide DiffBind analysis to determine all regions significantly differentially bound by MYB in BPDCN relative to normal pDCs (Supplemental Table 2). We used 2 PDX samples derived from separate patients with BPDCN, one of which (AL05) carried a MYB::ZFAT fusion (Supplemental Figure 1F) and compared each to normal pDCs from 3 healthy donors. In both PDX samples, sites that were significantly more strongly bound by MYB in BPDCN compared with normal pDCs and located at promoter regions showed increased expression in BPDCN relative to normal pDCs (Figure 1D and Supplemental Figure 1G) (20). These included cell cycle regulators such as CDC20 and CCP110 (Figure 1E and Supplemental Figure 1H) and were enriched for MYB binding motifs (Figure 1D and Supplemental Figure 1G). Conversely, sites significantly more strongly bound in pDCs compared with BPDCN showed decreased expression in BPDCN relative to normal pDCs and were enriched for TCF4 motifs (Figure 1D and Supplemental Figure 1G). These findings were true for both BPDCN PDX samples and implicate MYB binding at cell cycle gene promoters containing MYB binding motifs in BPDCN transformation.
CUT&RUN enables high-resolution localization of transcription factor binding sites using footprint analysis (38). We performed this analysis for MYB binding in CAL1 cells and, for each of the DNA motifs identified at MYB binding sites in CAL1 (Supplemental Figure 1D), identified CUT&RUN footprints containing these motifs. Compared with MYB CUT&RUN footprints containing TCF4, RUNX, IRF, or SPI1 motifs, footprints containing MYB motifs were more likely to be located at promoter regions (Supplemental Figure 1I), suggesting a role for these binding sites in direct control of gene expression. Indeed, genes with MYB CUT&RUN footprints containing MYB motifs at their promoters were enriched for increased expression in BPDCN PDX cells relative to normal pDCs (Supplemental Figure 1I and Supplemental Table 3) (20). Conversely, genes with MYB CUT&RUN footprints containing TCF4, RUNX, IRF, or SPI1 motifs at their promoters were enriched for increased expression in normal pDCs. The 500 highest-confidence footprints containing TCF4 motifs and mapping to promoters were enriched at pDC lineage-associated genes, while the 324 footprints containing MYB motifs and mapping to promoters were enriched at cell cycle genes (Supplemental Figure 1J).
Together, these findings implicate MYB in regulation of pDC lineage genes in normal pDCs via cooperative DNA binding with other transcription factors, particularly TCF4. Conversely, in BPDCN, MYB is a direct regulator of cell cycle genes which contain MYB DNA binding motifs at their promoters. We hypothesized that one of the key functions of MYB in driving BPDCN is by direct binding to DNA at cell cycle gene promoters via its DNA-binding domain. This aberrant binding could occur in BPDCNs with or without MYB fusions, since the N-terminal DNA-binding domain is retained in BPDCN MYB fusions (2, 3). Aberrant MYB binding could cause dysregulated expression of these genes, leading to loss of cell cycle control and contributing to BPDCN pathogenesis, since cycling is normally inhibited in dendritic progenitors undergoing terminal pDC differentiation (17).
BPDCN MYB fusions cause increased binding to G2/M cell cycle genes. Since MYB fusions are the only detectable mutations in some patients with BPDCN (2), we wondered whether MYB fusions are sufficient to cause increased binding of MYB at G2/M cell cycle genes. To first investigate this in a cellular context independent of the pDC lineage and using endogenous MYB expression levels, we knocked in V5-tagged MYB constructs to the endogenous MYB locus of the acute myeloid leukemia (AML) cell line K562 (Figure 2, A and B). This allows direct comparison of chromatin binding between WT MYB (MYB-WT) and BPDCN-associated MYB fusions. CUT&RUN for V5 revealed a strong relative enrichment of truncated MYB (MYB-TR) and MYB::PLEKHO1 binding compared with MYB-WT binding at the transcription start sites of G2/M cell cycle regulators such as CDC20 (Figure 2C) and CDCA3 (Supplemental Figure 2A). These binding sites contained MYB DNA binding motifs (Figure 2C and Supplemental Figure 2A).
Figure 2BPDCN MYB fusions cause increased binding to G2/M cell cycle genes. (A) Schematic showing expression of MYB-V5–knock-in constructs from the endogenous MYB locus in K562 cells. (B) Western blots showing expression of MYB-V5–knock-in constructs in K562 cells. (C) V5 and IgG CUT&RUN tracks at the CDC20 locus in MYB-V5–knock-in K562 cells of the indicated genotypes, for 2 different V5 antibodies. (D) Top 3 ranked motifs enriched in MYB CUT&RUN peaks in K562 nontransfected control (NTC) cells, and percent of peaks containing each motif. (E) Top 10 ranked Reactome gene sets overlapping with genes showing differentially increased MYB CUT&RUN binding in MYB-TR and MYB::PLEKHO1 relative to NTC and MYB-WT K562 cells at promoter regions. (F) Top 3 motifs enriched in MYB CUT&RUN peaks showing differentially increased binding in MYB-TR and MYB::PLEKHO1 relative to NTC and MYB-WT K562 cells, and percent of peaks containing each motif. (G) GSEA comparing (left) normal pDCs and primary BPDCN cells or (right) normal pDCs and BPDCN PDX cells for genes showing differentially increased MYB CUT&RUN binding at promoter regions in MYB-TR and MYB::PLEKHO1 relative to NTC and MYB-WT K562 cells (n = 180 genes).
We next performed CUT&RUN using a MYB antibody that recognizes the N-terminus (retained in BPDCN MYB fusions), detecting both knocked-in MYB-V5 and endogenous MYB (Supplemental Figure 2B). MYB-bound sites were enriched for GATA motifs (Figure 2D), consistent with GATA factors nucleating transcriptional regulatory complexes in K562 cells (39). We performed genome-wide analysis to identify sites of differential MYB binding in MYB-TR and MYB::PLEKHO1–knock-in cells relative to MYB-WT–knock-in cells and nontransfected control (NTC) K562 cells (Supplemental Table 2). This revealed enriched binding of MYB-TR and MYB::PLEKHO1 at transcription start sites of G2/M cell cycle regulators including CCP110 and CENPM (Figure 2E and Supplemental Figure 2C). In contrast, non–cell cycle genes expressed in K562 cells such as ERMAP and CCRL2 did not show differential binding (Supplemental Figure 2C). Differentially bound regions were enriched for MYB motifs and not GATA motifs (Figure 2F), suggesting importance of the N-terminal domain of MYB-TR and MYB::PLEKHO1 in binding to DNA at these sites. These differentially bound genes were strongly enriched for expression in primary BPDCN or BPDCN PDX cells relative to normal pDCs (Figure 2G) (20).
These data, using endogenous MYB expression levels (Figure 2B), demonstrate that BPDCN MYB fusions increase the intrinsic binding of MYB to DNA specifically at G2/M cell cycle genes that are strongly expressed in BPDCN. Given that MYB fusions may independently cause BPDCN in some patients (2), we wondered whether this increased DNA binding is sufficient to induce leukemia in vivo.
BPDCN MYB fusions impair differentiation in dendritic progenitor cells. To address this, we generated Hoxb8-FL immortalized hematopoietic progenitor cells as previously described (40), from mice constitutively expressing Cas9 and GFP (41). Hoxb8-FL cells proliferate indefinitely in the presence of estradiol and FLT3 ligand (FLT3L), and undergo myeloid, lymphoid, and dendritic differentiation upon withdrawal of estradiol in vitro or transplantation in vivo (40). This system allows efficient lentiviral sgRNA KO or cDNA expression in nontransformed hematopoietic progenitor cells, followed by interrogation of differentiation potential using in vitro and in vivo assays (42, 43). Using Hoxb8-FL cells also allowed us to explore whether MYB or BPDCN MYB fusions could confer self-renewal specifically in hematopoietic progenitor cells, rather than targeting already self-renewing hematopoietic stem cells as would be the case using lentiviral transduction of primary bone marrow.
To first investigate the effect of BPDCN MYB fusions on DC differentiation, we coexpressed a fluorescent dTom reporter with V5-tagged MYB-WT, MYB-TR, or MYB::PLEKHO1 in Hoxb8-FL cells using lentiviral transduction (Figure 3A and Supplemental Figure 3A). Empty vector–transduced (EV-transduced) cells, expressing only the dTom reporter, were used as controls. Since deletion of the G1/S cell cycle regulator CDKN2A is a common event in BPDCN (22, 24–26), we performed these experiments using either Cdkn2a-WT or Cdkn2a-KO Hoxb8-FL cells (Supplemental Figure 3, B and C).
Figure 3BPDCN MYB fusions impair differentiation in dendritic progenitor cells. (A) Schematic showing coexpression of V5-MYB constructs with a dTom reporter in Hoxb8-FL cells. (B) Experimental setup for Hoxb8-FL in vitro differentiation assays. (C) Representative flow cytometry plots at day 7 of in vitro differentiation of Hoxb8-FL cells of the indicated genotypes. Histogram on right shows B220 expression in empty vector cDCs and pDCs. (D) Number of dTom+ cells of the indicated cell types as a proportion of total dTom+ cells (n = 3 for each genotype). CD11b–CD11c–B220– undifferentiated cells, CD11b+CD11c+B220– cDCs, and CD11b–CD11c+B220+ pDCs. (E) Experimental setup for Hoxb8-FL in vivo differentiation assays. (F) Number of GFP+dTom+ bone marrow cells of the indicated cell types as a proportion of total GFP+dTom+ bone marrow cells (n = 3 for each genotype). CD19–CD11c–CD11b–B220– undifferentiated cells, CD11c–CD19–CD11b+ myeloid cells, CD11c–CD19+CD11b–B220+ B cells, and CD11c+CD19– DCs. (G) Number of GFP+dTom+ spleen cells of the indicated cell types as a proportion of total GFP+dTom+CD11c+ spleen cells, normalized to empty vector (n = 3 for each genotype). CD11c+CD19–CD11b+B220– cDCs and CD11c+CD19–CD11b–B220+BST2+SIGLEC-H+ pDCs. (D, F, and G) Data represent mean ± SEM. Significance determined by 1-way ANOVA with Tukey correction for multiple comparisons. *P < 0.05, **P < 0.01, ***P < 0.001.
We induced DC differentiation by withdrawal of estradiol and culture with FLT3L for 7 days (40) (Figure 3B). At day 7, all EV-transduced dTom+ cells showed surface expression of B220, CD11b, and/or CD11c, indicating conventional DC (cDC; CD11b+CD11c+B220–) or pDC (CD11b–CD11c+B220+) differentiation (Figure 3, C and D). A large proportion (50%–65%) of MYB-WT–, MYB-TR–, and MYB::PLEKHO1-transduced dTom+ cells, however, remained undifferentiated (CD11b–CD11c–B220–). This indicates impaired DC differentiation induced by exogenous expression of MYB. KO of Cdkn2a caused slightly increased numbers of pDCs (Figure 3D) but, otherwise, had little effect in this short-term differentiation assay.
We next interrogated in vivo differentiation potential by analyzing bone marrow and spleen of recipient mice 7 days after i.v. transplantation of Cdkn2a-KO Hoxb8-FL cells (Figure 3E). EV-transduced GFP+dTom+ cells mainly gave rise to myeloid cells (CD11b+CD11c–) and B cells (CD19+B220+) but also DCs (CD11c+) (Supplemental Figure 3D). Though rare, Hoxb8-FL–derived GFP+dTom+ pDCs (CD11b–CD11c+B220+SIGLEC-H+BST2+) were detectable in recipient bone marrow and spleen (Supplemental Figure 3E). Compared with EV, expression of all forms of MYB caused increased myeloid cells and reduced B cells and DCs as a proportion of total GFP+dTom+ cells in the bone marrow (Figure 3F) and spleen (Suppleme
留言 (0)