The long road
Hilgendorf, Susan
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Hilgendorf, S. (2018). The long road: The autophagic network and TP53/ASXL1 aberrations in hematopoietic malignancies. Rijksuniversiteit Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
in human hematopoietic and progenitor cells
Susan Hilgendorf1, Hendrik Folkerts1, Jan Jacob Schuringa1, and Edo Vellenga1
1 Department of Experimental Hematology, Cancer Research Center Groningen, University Medical Center Groningen, University of Groningen, The Netherlands
ABSTRACT
ASXL1 is frequently mutated in myelodysplastic syndrome and other hematological
malignancies. It has been reported that a loss of ASXL1 leads to a reduction of H3K27me3
via the polycomb repressive complex 2. To determine the role of ASXL1 loss in normal
hematopoietic stem and progenitor cells, cord blood CD34+ cells were transduced with
independent shRNAi lentiviral vectors against ASXL1 and cultured under myeloid and
erythroid permissive conditions. Knockdown of ASXL1 led to a significant reduction
in stem cell frequency and a reduced cell expansion along the myeloid lineage. Cell expansion along the erythroid lineage was also significantly reduced and accompanied by an increase in apoptosis of erythroid progenitor cells throughout differentiation and by an accumulation of cells in the G0/G1 phase. Bone marrow stromal cells supported
the growth of immature erythroid cells but did not alter the adverse phenotype of ASXL1
knockdown. Chromatin Immunoprecipitation revealed no loss of H3K27me3 in myeloid
progenitor cells but demonstrated a loss of H3K27me3 on the HOXA and also the p21 locus
in erythroid progenitors. We conclude that ASXL1 is essential for erythroid development
and differentiation and that the aberrant differentiation is, at least in part, facilitated via the polycomb repressive complex.
2
INTRODUCTION
Myelodysplastic syndrome (MDS) is a stem cell disorder characterized by a differentiation defect in one or more hematopoietic lineages in association with ineffective hematopoiesis1. A
paradox seen in MDS is that patients have peripheral cytopenias, while commonly displaying a hypercellular bone marrow2. In low-risk MDS this is attributed to an increased susceptibility
to programmed cell death, which appears to be less pronounced in high-risk MDS patients3–5.
Several mutations have recently been identified that contribute to the MDS phenotype. In
the low-risk patients, mutations are commonly found in TET2 and SF3B6,7, while mutations
in DNMT3a, TP53, and IDH1 are often present in high-risk MDS patients8–11. Mutations
of ASXL1 are found in low-and high-risk patients and are associated with an unfavorable
prognosis12. However, mutations in the ASXL1 gene are not restricted to MDS but have
also been demonstrated in acute myeloid leukemia (AML), myelofibrosis, and chronic
myelomonocytic leukemia (CMML). Mutations of ASXL1 may lead to truncation of the
protein and thereby to loss of its chromatin interacting and modifying domain. In addition,
ASXL1 is not the only gene of the ASXL family that can be affected. A study by Micol et al.
demonstrated that mutations in ASXL2 is predominantly observed in AML patients carrying the t(8, 21) translocation13.
One of the main functions attributed to ASXL1 is the stabilization and/or recruitment of the polycomb repressive complex 2 (PRC2) to certain loci of histone H3-lysine 27, leading
to trimethylation and repression of these loci14. Functional studies in mice have shown that
loss of ASXL1 function results in embryonic lethality, while heterozygous mice develop
an MDS-like phenotype after a long latency15–17. Abdel-Wahab et al. discovered that the
posterior HOXA cluster could be affected by down modulation of ASXL1 and HOXA genes
have been linked to malignant transformation in mice14,18,19. However, it is unclear how loss
of ASXL1 affects the PRC2 pathway in human hematopoietic stem and progenitor cells (HSPC). Additionally, the consequences of loss-of-function of ASXL1 in these cells are not well defined. Therefore, we used an RNAi approach to study the consequences in HSPC with
several independent shRNAs in vitro. The obtained results demonstrate that loss of ASXL1
leads to reduced erythroid differentiation and progenitor development due to increased
apoptosis and an increased accumulation of cells in the G0/G1 phase. Knockdown of ASXL1
coincides with a loss of H3K27 trimethylation and increased gene expression of targeted genes in the erythroid lineage but not in the myeloid lineage. Stem cell frequencies and
METHODS AND MATERIALS
Primary cell isolation
CD34+ cord blood cells (CB) were isolated after informed consent in accordance with
the Declaration of Helsinki from the Obstretics departments at the Martini Hospital and
University Medical Center Groningen, Groningen, The Netherlands. CB CD34+ were purified
with the AutoMacs (Miltenyi Biotec, Amsterdam, The Netherlands).
Cell culture
CB CD34+ cells were grown under myeloid liquid conditions (IMDM supplemented with 20%
FCS (Sigma, Zwijndrecht, The Netherlands)), interleukin 3 (IL-3; Gist-Brocades, Delft, The Netherlands), and stem cell factor (SCF, Amgen,Thousand Oaks, USA). Cells were cultured under erythroid permissive conditions in DMEM (Westburg, Leusden, The Netherlands) supplemented with 12% FCS, 10 mg/mL bovine serum albumin, 1% pen-strep, 1.9 mM sodium bicarbonate, 1 µM dexamethasone, 1 µM b-estradiol, 0.1 mM 2-mercaptoethanol, 0.3 mg/mL rHu Holo-Transferrin (Sigma, city country), 5 U/mL rHu EPO, 20 ng/mL SCF, 40 ng/mL rHu IGF-1 (Sigma).
Long-term cultures on stroma
Long-term cultures were performed as previously described20. CB CD34+ cells were expanded
on MS5 stromal cells in long-term culture (LTC) medium (Gartner’s) (αMEM supplemented with 12.5% heat-inactivated FCS, 12.5% heat-inactivated horse serum, pen-strep, 57.2 μM β-mercaptoethanol (Sigma) and 1μM hydrocortisone (Sigma)), with or without cytokines
(SCF (20ng/mL)) and EPO (2U/mL)). Cultures were grown at 37°C and 5% CO2
LTC-IC and CFC assays
LTC-initiating cell (LTC-IC) in limiting dilutions and colony forming cell (CFC) assays were performed as previously described20. LTC-IC assays was performed by plating CB CD34+ cells
in limiting dilutions in the range of 6 to 1458 cells per well on MS5 stromal cells in 96-well plates in Gartner’s medium. A demi of the medium of the cultures was conducted on a weekly basis. After 5 weeks, the medium was removed and replaced with methylcellulose (H4230, Stem Cell Technologies, city country) supplemented with 20ng/mL IL-3, interleukin-6, SCF, recombinant Human granulocyte-colony stimulating factor (G-CSF), Flt-3L, 10ng/ mL granulocyte macrophage (GM)-CSF, and 1U/mL Epo. After an additional 2 weeks,
wells were scored negative or positive for colony forming cells (CFC). For CFC assays, GFP+
CD34+ cells were plated in duplicate in 1 mL methylcellulose. Colonies were scored within
2
Lentiviral transductions
The lentiviral short hairpin RNA vectors of human ASXL1 and EZH2 were generous gifts
from prof.dr. Giovanni. Morrone (University of Catanzaro Magna Græcia, Catanzaro, Italy).
ASXL1 hairpins target the following sequences (1) GCTATGTCACAGGACAGTAAT
and (2) CCAGGAGAATCAGTGCGTATA. EZH2 targets the following sequence
TATGATGGTTAACGGTGATCA. The ASXL1 and EZH2 hairpins were cloned from a
puro vector into the pLKO.1 lentiviral vector containing either GFP or mCherry by cutting the restrictions sites with MUNI and SACII. A pLKO.1 GFP or mCherry vector containing a scrambled (SCR) short hairpin was used as controls. Lentiviral particles were produced by using the Fugene transfection system (Promega, Madison, WI, USA) together with glycoprotein envelope plasmid VSV-G, the packaging construct PAX2, and construct of interest. 293T cells were then transiently transfected. Stable transduction of cell lines or CB
CD34+ cells was performed and transduction efficiencies were measured by
fluorescence-activated cell sorter (FACS) analysis. Knockdown was verified by real-time polymerase chain reaction and Western blot.
Flow cytometry analysis
The following antibodies were used: allophycocyanin (APC)-conjugated anti-CD34 (581, BD, Alphen a/d Rijn, The Netherlands), phycoerythrin (PE), anti-CD14 (HCD14, Biolegend, Alphen a/d Rijn, The Netherlands), pacific blue-conjugated anti-CD15 (W6D3, Biolegend), brilliant violet anti-CD71 (M-A712, BD), R-PE anti-Glycophorin-A (JC159, Dako, Heverlee, Belgium). All cells were blocked with anti-human FcR block (Stem Cell Technologies). Cells were then incubated with antibodies for 30 minutes at 4°C. All FACS analyses were performed on a FACS LSR-II, cells were sorted on a MoFlo-XDP or -Astrios (DakoCytomation, Carpinteria, CA, USA) and data were analysed using FlowJo vX.0.7.
Real time polymerase chain reaction
Total RNA was isolated using the RNeasy kit from QIAGEN (Venlo, The Netherlands) in accordance with the manufacturer’s recommendations. RNA was reverse transcribed with iScript reverse transcriptase (Biorad). Using the iQ SYBR Green supermix (Biorad), cDNA was real-time amplified with the CFX connect Real-Time System (Biorad) thermocycler. For primers and primer sequences, see Supplementary Table S I.
Western blot
The following primary antibodies were used for Western blotting: ASXL1 (Santa Cruz H-105X), Akt (Cell Signaling 9272S), Caspase 3 (Cell Signaling 9662), H3K27me3 (Millipore 07-449), p21 (Abcam ab16767), PIM1 (Abcam ab117518), BNIP3L (Abcam ab8399). Sorted cells were boiled in Laemmli sample buffer for up to 10 min and separated
on 7.5%-12% SDS-polyacrylamide gels. Proteins were transferred to PVDF membrane (Millipore, Etten Leur, The Netherlands) by semidry electroblotting. Membranes were blocked in Odyssey blocking buffer (Westburg, Leusden, The Netherlands) and incubated with primary primary antibodies. Secondary antibodies were labeled with Alexa680 or IRDye800 (Invitrogen, Breda, The Netherlands) and used to detect binding of primary antibodies. Subsequently, membranes were scanned using an Odyssey infrared scanner (Li-Cor Biosciences, Lincolns, NE, USA).
Annexin V staining
Cells were washed with calcium buffer (10 mM HEPES, 140 mM NaCL, 2.5 mM CaCl2) and Annexin-V-APC antibody (IQ products IQ-120F/A) was added for 20 minutes at 4°C. Cells were washed with calcium buffer and analysed by FACS LSR-II.
Chromatin immunoprecipitation
ChIP experiments for H3K27me3 were carried out as previously described21. Briefly, cells
were transduced with shASXL1, sorted for GFP+ and cross-linked in 1% formaldehyde, then either snap-frozen or ChIP was performed directly.
Antibodies used for ChIP include anti-H3K27me3 (Millipore, 07-449). For ChIP polymerase chain reaction was conducted using the iQ SYBR Green supermix. For primers and primer sequences, see Supplementary Table S II.
Statistical Analysis
Student t-test from GrapPad Prism was used to analyze data. The frequency of stem cells was calculated using L-Calc software (Stem Cell Technologies). Data were expressed as means±s.d. and *p<0.05, **p<0.01, ***p<0.001 were considered significant.
RESULTS
Downregulation of ASXL1 leads to reduced stem cell and progenitor frequencies
CB CD34+ cells were transduced with control vectors (shSCR) or two independent shRNA
vectors to knockdown ASXL1 (shASXL1 #1 and shASXL1 #2). Transduction efficiencies
for both hairpins were above 30%, resulting in a reduction in ASXL1 expression at the
mRNA as well as at the protein level (Fig 1A). In order to evaluate the consequences of
ASXL1 knockdown on erythroid and myeloid progenitors, shSCR and shASXL1 #1 and
2
revealed significant reductions in BFU-Es (burst-forming unit-erythroid) and CFU-GMs(colony-forming unit-granulocytes/-macrophages) (Fig 1B). If not stated otherwise, further
experiments were conducted using hairpin #1. shASXL1 transduced CB CD34+ were then
grown under myeloid permissive conditions and followed for 15 days. At day 12, cells with
shASXL1 revealed a two-fold growth disadvantage (Fig 1C). Throughout growth, shASXL1
cells displayed a trend towards a lower percentage and cell number of CD14+ cells while the
CD15 population remained largely unchanged (Fig 1D and S1A). In addition, a significant
reduction in CFC frequencies throughout the culture period was observed (Fig 1E).
Rel. ASXL1 expression shSCR shASXL1 #1shASXL1 #2 ASXL1 Akt-t A weeks Day 12 shSCR shASXL1 #1shASXL1 #2 0.0 0.5 1.0 1.5 BFU-E CFU-GM B shASXL1 #1 Colonies / 500 cells shSCR 0 20 40 60 80 shASXL1 #2 shSCR * days D CD14+ (%) 5 8 12 0 10 20 30 40 * shSCRshASXL1 CD15+ (%) 5 8 12 0 10 20 30 40 days E CFU-GM / 10000 cells Day 8 0 20 40 60 shASXL1 shSCR *** 0 10 20 30 40 *** shASXL1 shSCR ** *** Figure 1 Day 14 CFU-GM / 10000 cells G 0 10 20 30 * shASXL1 shSCR 1% 10% 100% 0 500 1000 1500 Plated Cells Negative wells (%) shSCR shASXL1 1 in 1165 1 in 221 H shSCRshASXl1 0.000 0.002 0.004 0.006 0.008 LTC-IC frequency (%) *** ns ns ns ns ns days 0 5 10 15 0.1 1 10 100 1000 10000 shSCR shASXL1 Cumulative cell count myeloid liquid culture
C ns 0 20 40 60 80 100 *** ns *
Cumulative cell count MS5 co-culture F 0 1 2 3 4 5 0.1 0.2 0.3 0.4 0.6 1 1.6 shSCR Replica 3 shASXL1 Replica 3 shSCR Replica 1 shASXL1 Replica 1 shSCR Replica 2 shASXL1 Replica 2
Figure 1. ASXL1 loss leads to reduced stem cell and myeloid progenitor frequencies. (A) Knockdown of ASXL1 on mRNA (N=3) and
protein levels. (B) CFC analysis of CB cells expressing SCR, ASXL1 #1 or ASXL1 #2 (N=4). (C) Cumulative cell count of CB cells in myeloid
+ +
liquid culture (N=3). (D) CD14 and CD15 cells from myeloid liquid cultures at different time points (N=2). (E) CFC analysis of cells from myeloid liquid cultures at different time points (N=2). (F) Cumulative cell count of CB cell cultured on MS5 stroma (N=3) (G) CFC analysis of CB cells from suspension (N=3). (H) Limiting-dilution LTC-IC assay was performed (Representative of one experiment).
Error bars represent standard deviation; *p<0.05;**p<0.01; ***p<0.001
Figure 1. ASXL1 loss leads to reduced stem cell and myeloid progenitor frequencies. (A) Knockdown
of ASXL1 on mRNA (N=3) and protein levels. (B) CFC analysis of CB cells expressing SCR, ASXL1 #1
or ASXL1 #2 (N=4). (C) Cumulative cell count of CB cells in myeloid + liquid culture (N=3). (D) CD14+
and CD15+ cells from myeloid liquid cultures at different time points (N=2). (E) CFC analysis of cells
from myeloid liquid cultures at different time points (N=2). (F) Cumulative cell count of CB cell cultured
on MS5 stroma (N=3) (G) CFC analysis of CB cells from suspension (N=3). (H) Limiting-dilution
LTC-IC assay was performed (Representative of one experiment).
These findings demonstrate that shASXL1 affects myeloid development. Since MDS cells of low-risk patients are strongly dependent on the microenvironment, we evaluated the
consequences of shASXL1 in CB CD34+ cells in co-cultures on MS5 bone marrow stromal
cells in 3 independent experiments (Fig 1F). Downmodulation of ASXL1 impaired
long-term expansion during a 5-weeks co-culture period in comparison to shSCR CD34+ cells.
CFU-GM frequencies declined significantly over time (day 14, p<0.01) indicating that the MS5 stromal cells did not prevent reduction of progenitor frequencies upon knockdown
of ASXL1 (Fig 1G). To determine whether long-term culture-initiating cell (LTC-IC)
frequencies were also affected, limiting dilution LTC-IC assays were performed with CD34+
GFP+-sorted cells in 2 independent experiments. Data in Figure 1H reveals a significant
reduction in stem cell frequency of shASXL1 CD34+ cells versus shSCR CD34+ cells (1 in
1165 vs 1 in 221, respectively; p<0.0001, Fig S1B).
Together, this data demonstrates that reduction of ASXL1 impairs the maintenance of stem-
and progenitor cells.
Depletion of ASXL1 results in a proliferative disadvantage of erythroid progenitors in association with increased apoptosis and cell cycle arrest
To investigate the effects of ASXL1 downregulation in more detail on the erythroid
differentiation program, shSCR or shASXL1-transduced CB CD34+ cells were grown in
suspension in the presence of EPO and SCF. Erythroid expansion was markedly affected by
ASXL1 down regulation (Fig 2A) with similar findings for the second hairpin (Fig S2A).
In order to determine whether an increase in apoptosis would underlie the reduced output, AnnexinV staining was performed. Indeed, the reduction in cell expansion was, at least in part, due to an increase in apoptosis. At day 7, a significant overall increase in Annexin V
staining (p<0.02) was observed (Fig 2B, S2B), accompanied with reduced erythroid
differentiation and erythroid progenitor development in ASXL1 down modulated cells
as compared to controls (Fig 2C, S2C). Control shSCR CD34+ cells followed a normal
pattern of in vitro erythroid differentiation over time, in which cells gained CD71mid and
then CD71bright marker expression, followed by CD71bright/GlycophorinA (GPA+) double
expression. ASXL1 depleted cells, however, revealed a significantly increased percentage of
the CD71mid compartment (p<0.01), while the CD71bright compartment was significantly
reduced (p<0.006). Additionally, fewer cells became CD71bright/GPA+ over time upon
ASXL1 knockdown compared to controls (Fig S2D). Moreover, a significant increase of
Annexin V positivity among progenitor-like cells (CD71bright and CD71bright/GPA+, p<0.02
and p<0.003 respectively) was observed. The apoptotic phenotype in the ASXL1 knockdown
cells was limited to CD71bright and CD71 bright/GPA+ cells. Subsequently the whole CD71
positive population was studied for changes in gene expression (Fig 2D). In accordance
2
knockdown compared to control cells. Additionally, several apoptotic, cell cycle genes, andgenes involved in erythroid differentiation were investigated due to observed phenotype.
CDKN1A (referred to as p21), a cell cycle regulator, was significantly upregulated, while BNIP3L, a gene involved in erythroid development and differentiation, and PIM1, an
anti-apoptotic gene and proposed regulator of HSCs, showed significant reduced expression levels. A reduction and an increase in protein levels for BNIP3L and p21 could be observed, respectively (Fig S2E-F) and similar results were obtained for the second hairpin. Several other
well-known pro- and anti-apoptotic genes were not affected by ASXL1 knockdown (Fig S2G).
A 3 8 13 0.1 1 10 100 1000 shSCR shASXL1
Cumulative cell count erythroid liquid medium
** ** ** days B Annexin V+ (%) * shSCRshASXL1 0 5 10 15 20 shSCR shASXL1 0 0 GPA CD71 CD71bright/GPA 0.6% 103 104 105 0 103 104 105 0 103 104
105CD71bright 60.4% CD71bright/GPA 6.15% CD71bright 16.4%
CD71mid
29.8% CD71mid 78.5%
Relative expression
ASXL1 HOXA9 BNIP3L PIM1 p21
shASXL1 shSCR D 0 1 2 3 4 5 * ** * *** * BFU-E /10000 cells E F Cell cycle (%) shSCRshASXL1 0 50 100 G0/G1 phase S phase G2 phase 0 50 100 * shSCRshASXL1 C Figure 2 ns Erythroid differentiation (%) Annexin V+ (%) CD71mid CD71bright CD71bright/ GPA+ 0 50 100 * ** * shSCRshASXL1 0 20 40 60 80 shSCRshASXL1 ** shSCRshASXL1shSCRshASXL1 ns ns ns ns
Figure 2. Erythroid progenitors are compromised upon ASXL1 knockdown. (A) Cumulative cell count of CB cells cultured under
+
erythroid permissive conditions (N=3). (B) Total percentage of Annexin V cells at day 7 in erythroid liquid cultures (N=3). (C)
+
Percentage of cells in stages of erythroid differentiation and their percentage of Annexin V and FACS plot on day 7 of erythroid liquid cultures (N=3). (D) Gene expression normalized to NACA and RPS11 at day 7 of erythroid liquid cultures (N=2). (E) Cell cycle analysis at day 7 of erythroid liquid cultures (N=2). (F) CFC analysis of cells from erythroid liquid cultures at day 7 (N=3). Error bars represent standard deviation; *p<0.05; **p<0.01; ***p<0.001
Figure 2. Erythroid progenitors are compromised upon ASXL1 knockdown. (A) Cumulative cell
count of CB cells cultured under + erythroid permissive conditions (N=3). (B) Total percentage of
Annexin V+ cells at day 7 in erythroid liquid cultures (N=3). (C) Percentage of cells in stages of erythroid
differentiation and their percentage of Annexin V+ and FACS plot on day 7 of erythroid liquid cultures
(N=3). (D) Gene expression normalized to NACA and RPS11 at day 7 of erythroid liquid cultures (N=2). (E) Cell cycle analysis at day 7 of erythroid liquid cultures (N=2). (F) CFC analysis of cells from erythroid
liquid cultures at day 7 (N=3).
Cell cycle analysis demonstrated accumulation of cells in G0/G1 (87.7% ± 4.9) upon ASXL1
knockdown compared to shSCR (81.6% ± 3.6) with a reduced percentage of cells in S-phase and G2 (11.9% ± 0.8 vs. 6.4% ± 4.4 respectively) as compared to control cells (8.9% ± 1.7 vs.
3.4% ± 3.2 respectively) (Fig 2E). These changes were associated with reduced BFU-E output
over time (p<0.01) (Fig 2F). Taken together, these data indicate that upon knockdown of
ASXL1 apoptosis is a key contributor to the observed phenotypes in erythroid lineage, which
specifically affects the progenitor cells
Bone marrow stromal cells do not rescue adverse erythroid proliferation caused by ASXL1 depletion in vitro
To define whether the phenotype observed in Figure 2 could be reversed in the presence of
bone marrow stromal cells, co-cultures on MS5 were performed in the presence of EPO and
SCF (Fig 3A). As observed with suspension cells, ASXL1 depletion led to a profound decrease
in cell expansion (day 13, p<0.001). On day 13, reduced cell growth was accompanied by a significant increase of Annexin V+ cells (Fig 3B, p<0.01) but no changes on other time points
(Fig S3A). Opposite of what we observed in erythroid suspension cultures, the CD71mid
compartment revealed an increase in total cell numbers over time in ASXL1 downmodulated
cells (Fig 3C) suggesting that the stromal layer can support these more immature erythroid
cells. However, the survival of differentiating cells with ASXL1 knockdown was still impaired,
similar to erythroid cells cultured in suspension (Fig 3D). ASXL1 down modulation led
to a greater percentage of Annexin V+ cells among erythroid progenitors than in controls
(CD71bright/GPA+ p<0.01). Taken together, these data indicate that upon knockdown of
ASXL1 apoptosis is a key contributor to the observed phenotypes in EPO stimulated
co-cultures, which specifically affects erythroid progenitor-like cells.
Knockdown of ASXL1 leads to loss of H3K27me3 during erythroid but not myeloid progenitor development
To investigate whether the observed phenotypes could be attributed to changes in PRC-2
mediated epigenetic marks, we transduced CB CD34+ cells and cultured them for 7 days either
in myeloid or erythroid liquid suspension cultures in two independent experiments. GFP+
cells were then sorted and immediately cross-linked. Targeted ChIP PCR was conducted on
the HOXA9 promoter region and HOXA11 gene body, as the HOXA genes have previously
been shown to be affected by ASXL1 depletion14,17. In addition, ChIP for the p21 promoter
region was performed based on its upregulation on gene expression (Fig 2D) and protein
level (Fig S2D). In the ChIP of the myeloid liquid cultures, no reduction of H3K27me3 could
be observed (Fig 4A) whereby percentage of input for the HOXA9 cluster was generally
low. Moreover, gene expression for HOXA9 was not altered (Fig S4A). Downmodulation of ASXL1 in erythroid cells, however, revealed a significant reduction of H3K27me3 on the
2
HOXA9 locus of more than 80% for hairpin #1 and more than 65% for hairpin #2 (p<0.05both) (Fig 4B). On the HOXA11 locus, the loss of H3K27me3 was more than 55% and
41% for (p<0.02) hairpin 1 and 2, respectively. The observed increased p21 gene expression was accompanied by a reduction in H3K27me3 of more than 65% for hairpin #1 and more than 40% for hairpin #2 (p<0.05). However, the global trimethylation levels of H3K27me3
did not appear to be reduced upon ASXL1 knockdown (Fig S2D), suggesting that a loss in
H3K27me3 is locus-specific. 3 8 13 0.1 1 10 100 1000 10000 shSCR shASXL1 days
Cumulative cell count MS5 co-culture +EPO,SCF
** * Annexin V+ (%) B A shSCRshASXL1 0 5 10 15 C
Cell numbers of erythroid differentiating cells 101
102 103 104 105 106 107 108 109 1 shSCR shASXL1+++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ Day 7 10 13 7 10 13 7 10 13 7 10 13 ** ** ** * D Erythroid differentiation (%) CD71mid CD71bright/ GPA+ GPA+ CD71bright 0 50 100 * shSCRshASXL1 Annexin V+ (%) 0 5 10 15 * CD71bright/ GPA+ CD71mid GPA+
shSCRshASXL1 shSCRshASXL1 shSCRshASXL1
CD71mid CD71bright/ GPA+ GPA+ CD71bright Figure 3 * ns ns ns ns ns ns ns ns ns ns ns ns
Figure 3. ASXL1 knockdown cells compromised on MS5 stromal cells. (A) Cumulative cell count of CB cell cultured on stroma
+
supplemented with EPO and SCF (N=3). (B) Percentage of Annexin V cells at day 13 of suspension culture (N=3). (C) Total numbers of
+
cells in stages of erythroid differentiation (N=3). (D) Percentage of erythroid differentiating cells and their Annexin V percentage at day 13 (N=3).
Error bars represent standard deviation; *p<0.05; **p<0.01
Figure 3. ASXL1 knockdown cells compromised on MS5 stromal cells. (A) Cumulative cell count of
CB cell cultured on stroma + supplemented with EPO and SCF (N=3). (B) Percentage of Annexin V+ cells
at day 13 of suspension culture (N=3). (C) Total numbers of cells in stages of erythroid differentiation
(N=3). (D) Percentage of erythroid differentiating cells and their Annexin V+ percentage at day 13 (N=3).
Error bars represent standard deviation; *p<0.05; **p<0.01
To determine whether additional components of the PRC2 complex affected the erythroid
lineage in a comparable manner, CB CD34+ were transduced by shEZH2 vectors which
resulted in 71% reduced expression, while EZH1 expression remained largely unaffected (Fig
4C, S4B). Similarly to loss of ASXL1, EZH2 knockdown led to impaired BFU-E and
CFU-GM colony formation as compared to control cells (Fig 4C). Additionally, cell expansion was
significantly reduced and accompanied by increased apoptosis on multiple time points (Fig
than in control cells (Fig 4F). As noted before, throughout erythroid maturation an increase
in Annexin V+ could be observed. Additionally, total cell numbers decreased or remained low
over time in the EZH2 knockdown cells compared to controls (Fig 4G).
HOXA9 HOXA11 H3K27me3 Ig % of Input A B 0 2 4 6 shSCR shASXL1 #1shASXL1 #2 0 10 20 30 40 50 60 70 80 shSCR shASXL1 #1shASXL1 #2 0 5 10 15 20 25 30 35 * * shSCR shASXL1 #1shASXL1 #2 HOXA9 % of Input 0 10 20 30 40 50 60 70 * HOXA11 0 5 10 15 20 25 * p21 shSCR
shASXL1 #1shASXL1 #2 shSCR shASXL1 #1shASXL1 #2 Figure 4
ns
ns nsns
C
days Cumulative cell count erythroid liquid medium
Annexin V+ (%) D E 3 8 13 0.1 1 10 100 1000 shSCR shEZH2 * * shEZH2 Colonies / 500 cells BFU-E CFU-GM shSCR 0 50 100 *** **
Rel. EZH2 expression0.0 0.5 1.0 1.5 shSCR shEZH2 0 5 10 15 20 25 30 shEZH2 shSCR 7 13 days ** ns ns
Cell numbers of erythroid differentiating cells G 1 101 102 103 104 105 106 107 ** ** ** ** shSCR shEZH2 + + + + + + + + + + + + Day 7 13 7 13 7 13 CD71mid CD71bright CD71bright/ GPA+ ns ns Annexin V+ (%) CD71mid CD71bright CD71bright/ GPA+ Erythroid differentiation (%) 0 50 100 F shSCR shEZH2 shSCR shEZH2 0 10 20 30 40 50 60 70
shSCR shEZH2 shSCR shEZH2 shSCR shEZH2 shSCR shEZH2 shSCR shEZH2 shSCR shEZH2 ** Day 7 13 7 13 7 13 ns ns ns ns ns ns ns ns ns ns ns
Figure 4. Knockdown of ASXL1 leads to loss of H3K27me3. (A) H3K27me3 ChIP analysis of myeloid progenitor cells on HOXA9 promoter and
HOXA11 gene (N=2). (B) H3K27me3 ChIP analysis of erythroid progenitor cells on p21 and HOXA9 promoter and HOXA11 on gene (N=2). (C) Relative EZH2 expression compared to RPS11 (N=2). CFC analysis of cells of CB cells expressing SCR and EZH2 (N=3). (D) Cumulative cell count of
+
CB cell cultured under erythroid permissive conditions (N=3). (E) Total percentage of Annexin V cells (N=2). (F) Percentage of cells in stages of
+
erythroid differentiation and their percentage of Annexin V (N=2). (G) Total numbers of cells in stages of erythroid differentiation (N=2). Error bars represent standard deviation; *p<0.05; **p<0.01
Figure 4. Knockdown of ASXL1 leads to loss of H3K27me3. (A) H3K27me3 ChIP analysis of
myeloid progenitor cells on HOXA9 promoter and HOXA11 gene (N=2). (B) H3K27me3 ChIP analysis
of erythroid progenitor cells on p21 and HOXA9 promoter and HOXA11 on gene (N=2). (C) Relative
EZH2 expression compared to RPS11 (N=2). CFC analysis of cells of CB cells expressing SCR and EZH2 (N=3). (D) Cumulative cell count of CB cell cultured under erythroid permissive conditions (N=3). (E)
Total percentage of Annexin V+ cells (N=2). (F) Percentage of cells in stages of + erythroid differentiation
and their percentage of Annexin V (N=2). (G) Total numbers of cells in stages of erythroid differentiation
(N=2).
2
DISCUSSION
ASXL1 has recently been identified as an important gene involved in malignant transformation.
Mutations in ASXL1 have been found in MDS but also in AML, CMML and myelofibrosis and are in general associated with an unfavorable prognosis22–25.
ASXL1 is a proposed partner of the PRC2 complex and as such involved in epigenetic gene regulation14. In the present study, we analyzed the consequences of loss of ASXL1 in normal
HSPC and observed that stem cells and specifically progenitor cells are strongly affected in their expansion, while modestly affecting the differentiation program. Surprisingly, our
data show that loss of ASXL1 does not immediately contribute to transformation but rather
leads to loss of cells due to apoptosis. It is likely that a subset of cells that adapt to ASXL1
knockdown may participate in the process of transformation. Our in vitro assays are apparently
not sufficiently long-term to allow hematopoietic clones to outgrow with the unfavorable ASXL1 mutation. In fact, mice with an ASXL1 mutation reveal an MDS phenotype only after serial transplantation15. This suggests that additional epi- or non-epigenetic alterations might
be necessary for malignant transformation. Population-based studies revealed recently also
age-related mutations including ASXL1 in healthy individuals that are associated with clonal
hematopoiesis26,27. Only individuals with several mutations are at greater risk to develop
myeloid malignancies. Data from patient studies indicate also that additional mutations such as U2AF1, EZH2, or NRAS/KRAS co-occur with ASXL1 mutations, suggesting that ASXL1 might need cooperating hits for malignant transformation. In addition, the microenvironment of the patient might have been adapted and be permissive for ongoing transformation.
It appeared that, upon knockdown of ASXL1, particularly the erythroid lineage was affected.
The strong response in the erythroid lineage might be related to the relative high expression of ASXL1 in these cells28. It was shown that in purified hematopoietic cell populations ASXL1
is higher expressed in erythroid progenitors as compared to more immature hematopoietic cells or myeloid progenitors.
Loss of ASXL1 function triggered an apoptotic response and culturing these cells on stromal
layer did not give them a selective advantage, in contrast to MDS erythroid progenitors,
which have a strong benefit form their own micro-environment3. Gene expression studies of
shASXL1 cells in erythroid liquid cultures demonstrated that the increased apoptosis could
be linked to p21 in conjunction with reduced expression of BNIP3L, and PIM1 that are acting
together in the altered differentiation program and increased apoptosis. The set of affected genes were in general not in line with the results of the study of Wahab et al14. Their data set
revealed changes in several apoptosis related genes, which were, for the most part, unchanged in our study. The difference might be related to differences in experimental set-up between both studies.
It has been proposed that ASXL1 affects the epigenetic machinery, which modifies chromatin
in concert with the PRC2 complex14. PRC2-ASXL1 activity may lead to an increase of
H3K27 trimethylation and therefore, a loss of ASXL1 presumably leads to a reduction
of this silencing mark. Integration of gene expression and ChIP data following ASXL1
downmodulation in erythroid progenitor cells identified HOXA9 and p21 as targets with
reduced H3K27me3. Previous studies have suggested that HOXA9 upregulation coincides
with malignant transformation but in the present experimental set-up with normal CB
CD34+ cells, transformation was not observed possibly due to the activation of cell death
pathways18,19. In addition, the changes in trimethylation were in particular observed in the
erythroid lineage. Trimethylation levels within the myeloid liquid cultures were not affected,
possibly due to the already low trimethylation levels of the HOXA9 locus. A loss of ASXL1
may not lead to early noticeable changes in cell proliferation and differentiation among the myeloid lineage and changes on epigenetic level may take place at later time points.
Knockdown of EZH2, an alternative member of the PRC2 complex, revealed similar
phenotypes in the erythroid lineage supporting the concept that ASXL1 conveys its function via the PRC2 complex in erythroid development. Although ASXL1 mutations are considered as a loss of function, which can be mimicked by an RNAi approach, a recent study demonstrated that the ASXL1 truncated protein may act as a gain-of-function in the context
of the ASXL1-BAP1 complex29. More future studies need to be conducted to clarify if the
truncated ASXL1 protein is indeed functional.
Based on our results, we conclude that ASXL1 is necessary for proper function of the
erythroid differentiation and, to a lesser extent, for the myeloid compartment. In addition, loss of ASXL1 negatively affects the frequency of stem cells in long-term cultures. Elucidating
the role of ASXL1 in hematopoietic stem cells further may contribute to improved treatment
2
REFERENCES
1. Jaiswal S, Ebert B. MDS Is a Stem Cell Disorder After All. Cancer Cell 2014;25713–714.
2. Kerbauy DB, Deeg HJ. Apoptosis and anti-apoptotic mechanisms in the progression of MDS. Exp Hematol 2007;35(11):1739–1746. 3. Folkerts H, Hazenberg CLE,
Houwerzijl EJ, et al. Erythroid progenitors from patients with low-risk myelodysplastic syndromes are dependent on the surrounding micro environment for their survival. Exp Hematol 2015;43(3):215–222.e2. 4. Houwerzijl EJ, Pol H-W, Blom
NR, Van der Want JJL, de Wolf Jt, Vellenga E. Erythroid precursors from patients with low-risk myelodysplasia demonstrate ultrastructural features of enhanced autophagy of mitochondria. Leukemia 2009;23(5):886–891. 5. Houwerzijl EJ, Heuvel FA, Blom NR, Want JJ, Mulder AB, Vellenga E. Sinusoidal endothelial cells are damaged and display enhanced autophagy in myelodysplastic syndromes. Br J Haematol 2013;161(3):443–446. 6. Kosmider O, Gelsi-Boyer V,
Cheok M, et al. TET2 mutation is an independent favorable prognostic factor in myelodysplastic syndromes (MDSs). Blood 2009;114(15):3285–3291. 7. Malcovati L, Karimi M,
Papaemmanuil E, et al. SF3B1 mutation identifies a distinct subset of myelodysplastic syndrome with ring sideroblasts. Blood 2015;126(2):233–241. 8. Patnaik MM, Hanson CA,
Hodnefield JM, et al. Differential prognostic effect of IDH1 versus IDH2 mutations in myelodysplastic syndromes: a Mayo Clinic Study of 277 patients. Leukemia;26101–105. 9. Sallman DA, Komrokji R,
Vaupel C, et al. Impact of TP53
mutation variant allele frequency on phenotype and outcomes in myelodysplastic syndromes. Leukemia 2016;30(3):666–673. 10. Thol F, Weissinger EM, Krauter J,
et al. IDH1 mutations in patients with myelodysplastic syndromes are associated with an unfavorable prognosis. Haematologica 2010;95(10):1668–1674. 11. Walter MJ, Ding L, Shen D, et al.
Recurrent DNMT3A mutations in patients with myelodysplastic syndromes. Leukemia 2011;25(7):1153–1158. 12. Bejar R, Stevenson KE, Caughey
BA, et al. Validation of a Prognostic Model and the Impact of Mutations in Patients With Lower-Risk Myelodysplastic Syndromes. J Clin Oncol 2012;30(27):3376–3382. 13. Micol J-B, Duployez N, Boissel N,
et al. Frequent ASXL2 mutations in acute myeloid leukemia patients with t (8; 21)/RUNX1-RUNX1T1 chromosomal translocations. Blood 2014;124(9):1445–1449. 14. Abdel-Wahab O, Adli M, LaFave
LM, et al. ASXL1 Mutations Promote Myeloid Transformation through Loss of PRC2-Mediated Gene Repression. Cancer Cell 2012;22(2):180–193. 15. Abdel-Wahab O, Gao J, Adli M,
et al. Deletion of Asxl1 results in myelodysplasia and severe developmental defects in vivo. J Exp Med 2013;210(12):2641–2659. 16. Inoue D, Kitaura J, Togami K, et al. Myelodysplastic syndromes are induced by histone methylation-altering ASXL1 mutations. J Clin Invest 2013;123(11):4627–4640. 17. Wang J, Li Z, He Y, et al. Loss of
Asxl1 leads to myelodysplastic syndrome-like disease in mice. Blood 2014;123(4):541–553. 18. Kroon E, Thorsteinsdottir
U, Mayotte N, Nakamura T,
Sauvageau G. NUP98–HOXA9 expression in hemopoietic stem cells induces chronic and acute myeloid leukemias in mice. EMBO J 2001;20(3):350–361.
19. Ayton PM, Cleary ML. Transformation of myeloid progenitors by MLL oncoproteins is dependent on Hoxa7 and Hoxa9. Genes Dev 2003;17(18):2298– 2307.
20. Schepers H, van Gosliga D, Wierenga ATJ, Eggen BJL, Schuringa JJ, Vellenga E. STAT5 is required for long-term maintenance of normal and leukemic human stem/progenitor cells. Blood 2007;110(8):2880–2888. 21. Frank SR, Schroeder M, Fernandez
P, Taubert S, Amati B. Binding of c-Myc to chromatin mediates mitogen-induced acetylation of histone H4 and gene activation. Genes Dev 2001;15(16):2069– 2082.
22. Thol F, Friesen I, Damm F, et al. Prognostic Significance of ASXL1 Mutations in Patients With Myelodysplastic Syndromes. J Clin Oncol 2011;29(18):2499–2506. 23. Schnittger S, Eder C, Jeromin S, et al. ASXL1 exon 12 mutations are frequent in AML with intermediate risk karyotype and are independently associated with an adverse outcome. Leukemia 2013;27(1):82–91. 24. Cui Y, Tong H, Du X, et al.
Impact of TET2, SRSF2, ASXL1 and SETBP1 mutations on survival of patients with chronic myelomonocytic leukemia. Exp Hematol Oncol;4(1):. 25. Vannucchi AM, Lasho TL,
Guglielmelli P, et al. Mutations and prognosis in primary myelofibrosis. Leukemia 2013;27(9):1861–1869. 26. Jaiswal S, Fontanillas P, Flannick
J, et al. Age-Related Clonal Hematopoiesis Associated with
Adverse Outcomes. N Engl J Med 2014;371(26):2488–2498. 27. Xie M, Lu C, Wang J, et al.
Age-related mutations associated with clonal hematopoietic expansion and malignancies. Nat Med 2014;20(12):1472–1478. 28. Novershtern N, Subramanian
A, Lawton LN, et al. Densely Interconnected Transcriptional Circuits Control Cell States in Human Hematopoiesis. Cell 2011;144(2):296–309. 29. Balasubramani A, Larjo A, Bassein
JA, et al. Cancer-associated ASXL1 mutations may act as gain-of-function mutations of the ASXL1– BAP1 complex. Nat Commun 2015;67307.
2
SUPPLEMENTARY FIGURES & TABLES
A
shSCR shASXL1 Cell Number CD15+ Cell Number CD14+ 5 8 12 1 10 100 1000 10000 100000 1000000 5 8 12 1 10 100 1000 10000 100000 1000000B
LTC-IC frequency (%) 0.000 0.002 0.004 0.006 shASXL1 shSCR *** 1% 10% 100% 0 500 1000 1500 Plated cellsNegative wells (%) shSCRshASXL1
1 in 892
1 in 287
Supplemental Figure 1
ns ns ns ns ns ns
S1. ASXL1 loss leads to reduced stem cell
+ +
frequency. (A) Total cell number CD14 and CD15
cells from myeloid liquid cultures at different time points (N=2). (B) Limiting-dilution LTC-IC assays were performed (Representative of one experiment).
Error bars represent standard deviation; ***p<0.001
S1. ASXL1 loss leads to reduced stem cell frequency. (A) Total cell number CD14+ and CD15+ cells
from myeloid liquid cultures at different time points (N=2). (B) Limiting-dilution LTC-IC assays were
performed (Representative of one experiment). Error bars represent standard deviation; ***p<0.001
Supplemental Figure 2 C E D Asxl1 Akt-t Bnip3L p21 H3K27me3 shASXL1 #1 shSCR shASXL1 #2
GATA2 BNIP3 DAPK1 BAX p53 PUMA BCL2
Relative expression 0.0 0.5 1.0 1.5 2.0 2.5 shASXL1 shSCR B
Cell numbers of erythroid differentiating cells 1 101 102 103 104 105 106 107 Day 7 10 13 7 10 13 7 10 13 * * * ** * shASXL1 shSCR + + ++ ++ ++ ++ ++ ++ ++ ++ CD71mid CD71bright CD71bright/ GPA+ 3 8 13 0.1 1 10 100 1000 10000 shSCR shASXL1 #2 days
Cumulative cell count erythroid liquid medium
ns 7 10 0 5 10 15 20 25 Annexin V+ (%) * ns days shASXL1 #2 shSCR Erythroid differentiation (%) Annexin V+ (%) 0 50 100 shSCR shASXL1 shSCRshASXL1 ** * *** *** * Day10 13 0 20 40 60 80
shSCRshASXL1 shSCRshASXL1shSCRshASXL1 shSCRshASXL1shSCRshASXL1 shSCRshASXL1
Day10 13 10 13 10 13 CD71mid CD71bright CD71bright/ GPA+ *** ** ns ns ns ns ns ns ns ns ns Annexin V+ (%) A 10 13 0 5 10 15 20 25 30 shASXL1 shSCR days * ns 0 1 2 3 4 5 Relative expression
ASXL1 HOXA9 BNIP3L PIM1 p21
* * * *** **
shASXL1 #2 shSCR
F G ns ns ns ns ns ns ns
S2. Erythroid progenitors are compromised upon ASXL1 downmodulation. (A) Cumulative cell count of CB cells cultured under
+
erythroid permissive conditions for hairpin #2 (N=3). (B) Total percentage of Annexin V cells for both hairpins in erythroid liquid
+
cultures (N=3). (C) Percentage of cells in stages of erythroid differentiation and their percentage of Annexin V at day 10 and 13 (N=3). (D) Total numbers of cells in stages of erythroid differentiation (N=3). (E) Cells sorted at day 7 for western Blot analysis of Asxl1, Bnip3L, p21, and H3K27me3. Akt-t functions as loading control. (F-G) Hairpin #1 and #2: Cells sorted at day 7 for gene expression analysis, repectively (N=2). Relative expression of target genes normalized against NACA and RPS11.
Error bars represent standard deviation *p<0.05; **p<0.01; ***p<0.0001
S2. Erythroid progenitors are compromised upon ASXL1 downmodulation. (A) Cumulative
cell count of CB cells cultured under erythroid permissive conditions for hairpin #2 (N=3). (B) Total
percentage of Annexin V+ cells for both hairpins in erythroid liquid cultures (N=3). (C) Percentage of
cells in stages of erythroid differentiation and their percentage of Annexin V+ at day 10 and 13 (N=3). (D)
Total numbers of cells in stages of erythroid differentiation (N=3). (E) Cells sorted at day 7 for western
Blot analysis of Asxl1, Bnip3L, p21, and H3K27me3. Akt-t functions as loading control. (F-G) Hairpin #1
and #2: Cells sorted at day 7 for gene expression analysis, repectively (N=2). Relative expression of target genes normalized against NACA and RPS11.
2
3 8 13 0.1 1 10 100 1000 10000 shSCR shASXL1 daysCumulative cell count MS5 co-culture +EPO,SCF
** * Annexin V+ (%) B A shSCRshASXL1 0 5 10 15 C
Cell numbers of erythroid differentiating cells 101 102 103 104 105 106 107 108 109 1 shSCR shASXL1+++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ Day 7 10 13 7 10 13 7 10 13 7 10 13 ** ** ** * D Erythroid differentiation (%) CD71mid CD71bright/ GPA+ GPA+ CD71bright 0 50 100 * shSCRshASXL1 Annexin V+ (%) 0 5 10 15 * CD71bright/ GPA+ CD71mid GPA+
shSCRshASXL1 shSCRshASXL1 shSCRshASXL1
CD71mid CD71bright/ GPA+ GPA+ CD71bright Figure 3 * ns ns ns ns ns ns ns ns ns ns ns ns
Figure 3. ASXL1 knockdown cells compromised on MS5 stromal cells. (A) Cumulative cell count of CB cell cultured on stroma
+
supplemented with EPO and SCF (N=3). (B) Percentage of Annexin V cells at day 13 of suspension culture (N=3). (C) Total numbers of
+
cells in stages of erythroid differentiation (N=3). (D) Percentage of erythroid differentiating cells and their Annexin V percentage at day 13 (N=3).
Error bars represent standard deviation; *p<0.05; **p<0.01
S3. Erythroid progenitors are compromised when cultured on stroma. (A) Total percentage
of Annexin V+ suspension cells at day 7 and 10 (N=3). (B) Percentage of cells in stages of erythroid
differentiation and their percentage of Annexin V+ (N=3).
HOXA9 HOXA11 H3K27me3 Ig % of Input A B 0 2 4 6 shSCR shASXL1 #1shASXL1 #2 0 10 20 30 40 50 60 70 80 shSCR shASXL1 #1shASXL1 #2 0 5 10 15 20 25 30 35 * * shSCR shASXL1 #1shASXL1 #2 HOXA9 % of Input 0 10 20 30 40 50 60 70 * HOXA11 0 5 10 15 20 25 * p21 shSCR
shASXL1 #1shASXL1 #2 shSCR shASXL1 #1shASXL1 #2 Figure 4
ns
ns nsns
C
days Cumulative cell count erythroid liquid medium
Annexin V+ (%) D E 3 8 13 0.1 1 10 100 1000 shSCR shEZH2 * * shEZH2 Colonies / 500 cells BFU-E CFU-GM shSCR 0 50 100 *** **
Rel. EZH2 expression0.0 0.5 1.0 1.5 shSCR shEZH2 0 5 10 15 20 25 30 shEZH2 shSCR 7 13 days ** ns ns
Cell numbers of erythroid differentiating cells G 1 101 102 103 104 105 106 107 ** ** ** ** shSCR shEZH2 + + + + + + + + + + + + Day 7 13 7 13 7 13 CD71mid CD71bright CD71bright/ GPA+ ns ns Annexin V+ (%) CD71mid CD71bright CD71bright/ GPA+ Erythroid differentiation (%) 0 50 100 F shSCR shEZH2 shSCR shEZH2 0 10 20 30 40 50 60 70
shSCR shEZH2 shSCR shEZH2 shSCR shEZH2 shSCR shEZH2 shSCR shEZH2 shSCR shEZH2 ** Day 7 13 7 13 7 13 ns ns ns ns ns ns ns ns ns ns ns
Figure 4. Knockdown of ASXL1 leads to loss of H3K27me3. (A) H3K27me3 ChIP analysis of myeloid progenitor cells on HOXA9 promoter and
HOXA11 gene (N=2). (B) H3K27me3 ChIP analysis of erythroid progenitor cells on p21 and HOXA9 promoter and HOXA11 on gene (N=2). (C) Relative EZH2 expression compared to RPS11 (N=2). CFC analysis of cells of CB cells expressing SCR and EZH2 (N=3). (D) Cumulative cell count of
+
CB cell cultured under erythroid permissive conditions (N=3). (E) Total percentage of Annexin V cells (N=2). (F) Percentage of cells in stages of
+
erythroid differentiation and their percentage of Annexin V (N=2). (G) Total numbers of cells in stages of erythroid differentiation (N=2). Error bars represent standard deviation; *p<0.05; **p<0.01
S4. Slight changes in EZH1 expression. (A) Gene expression normalized to NACA and RPS11 at day
7 of myeloid liquid cultures (N=2). (B) Relative EZH1 expression upon EZH2 knockdown normalized
against NACA and RPS11 (N=2).
2
Table S I: List of qPCR primer sequences in this studyGene Forward primers (5’ to 3’) Reverse primers (5’ to 3’)
ASXL1 GGTCAAATGAAGCGCAACAGAG ACGGAGGTTGGTGTTGACAAG
BAX CCAGCAAACTGGTGCTCAAG GGAGGCTTGAGGAGTCTCAC
BCL-2 AACATCGCCCTGTGGATGAC GGCCGTACAGTTCCACAAAG
BNIP3 AGCTCACAGTCTGAGGAAGATG GCGCTTCGGGTGTTTAAAGAGG
BNIP3L ACACGTACCATCCTCATC GATCTGCCCATCTTCTTG
CDKN1A CGACTGTGATGCGCTAATGG CGTTTTCGACCCTGAGAG
DAPK1 GGTCTTGAGGCAGATATG GTAGTTGACAGCGGATAC
EZH2 CAGTTCGTGCCCTTGTGTGATA GGAAAGCGGTTTTGACACTCTG
GATA1 ACACTGTGGCGGAGAAATGC AGATGCCTTGCGGTTTCGAG
GATA2 AGCAAGGCTCGTTCCTGTTC GTCGGTTCTGCCCATTCATC
HOXA9 TGCAGTTTCATAATTTCCGTCG ACGTAGTAGTTGCCCAGGGCC
NACA GCCCTGCTTCAGATACTTAC GAGACAGCTTCACCTTGAAC
P53 GAGATGTTCCGAGAGCTGAATGAGGC TCTTGAACATGAGTTTTTTATGGCGGGAGG
PIM1 TCAAACACGTGGAGAAGG TAATGACGCCGGAGAAAC
PUMA GACCTCAACGCACAGTACG GGCAGGAGTCCCATGATGAG
RPS11 AAGATGGCGGACATTCAGAC AGCTTCTCCTTGCCAGTTTC
Table S II: List of ChIP primer sequences used in this study
Gene Forward primers (5’ to 3’) Reverse primers (5’ to 3’)
HOXA9 AGTCAGTCAGGGACAAAGTG CCGGCCTTATGGCATTAAAC
HOXA11 AGAGCCCATAGCTGAGGAG ATGCAGTCGGAGCGTTAAAG
